Bilingual Education in South America 9781853598203

Citation preview

Bilingual Education in South America

BILINGUAL EDUCATION AND BILINGUALISM Series Editors: Professor Colin Baker, University of Wales, Bangor, Wales, Great Britain and Professor Nancy H. Hornberger, University of Pennsylvania, Philadelphia, USA Recent Books in the Series Cross-linguistic Influence in Third Language Acquisition J. Cenoz, B. Hufeisen and U. Jessner (eds) Learners’ Experiences of Immersion Education: Case Studies of French and Chinese Michèle de Courcy Language Minority Students in the Mainstream Classroom (2nd edn) Angela L. Carrasquillo and Vivian Rodríguez World English: A Study of its Development Janina Brutt-Griffler Power, Prestige and Bilingualism: International Perspectives on Elite Bilingual Education Anne-Marie de Mejía Identity and the English Language Learner Elaine Mellen Day Language and Literacy Teaching for Indigenous Education: A Bilingual Approach Norbert Francis and Jon Reyhner The Native Speaker: Myth and Reality Alan Davies Language Socialization in Bilingual and Multilingual Societies Robert Bayley and Sandra R. Schecter (eds) Language Rights and the Law in the United States: Finding our Voices Sandra Del Valle Continua of Biliteracy: An Ecological Framework for Educational Policy, Research, and Practice in Multilingual Settings Nancy H. Hornberger (ed.) Languages in America: A Pluralist View (2nd Edition) Susan J. Dicker Trilingualism in Family, School and Community Charlotte Hoffmann and Jehannes Ytsma (eds) Multilingual Classroom Ecologies Angela Creese and Peter Martin (eds) Negotiation of Identities in Multilingual Contexts Aneta Pavlenko and Adrian Blackledge (eds) Beyond the Beginnings: Literacy Interventions for Upper Elementary English Language Learners Angela Carrasquillo, Stephen B. Kucer and Ruth Abrams Bilingualism and Language Pedagogy Janina Brutt-Griffler and Manka Varghese (eds) Language Learning and Teacher Education: A Sociocultural Approach Margaret R. Hawkins (ed.) The English Vernacular Divide: Postcolonial Language Politics and Practice Vaidehi Ramanathan

For more details of these or any other of our publications, please contact: Multilingual Matters, Frankfurt Lodge, Clevedon Hall, Victoria Road, Clevedon, BS21 7HH, England http://www.multilingual-matters.com

BILINGUAL EDUCATION AND BILINGUALISM 50 Series Editors: Colin Baker and Nancy H. Hornberger

Bilingual Education in South America Edited by

Anne-Marie de Mejía

MULTILINGUAL MATTERS LTD Clevedon • Buffalo • Toronto

Library of Congress Cataloging in Publication Data Bilingual Education in South America/Edited by Anne-Marie de Mejía. Bilingual Education and Bilingualism: 50 Includes bibliographical references. 1. Education, Bilingual–South America–Cross-cultural studies. 2. Bilingualism–South America–Cross-cultural studies. I. De Mejía, Anne-Marie. II. Series. LC3735.S68B55 2004 370.117'098–dc22 2004026149 British Library Cataloguing in Publication Data A catalogue entry for this book is available from the British Library. ISBN 1-85359-819-4 (hbk) Multilingual Matters Ltd UK: Frankfurt Lodge, Clevedon Hall, Victoria Road, Clevedon BS21 7HH. USA: UTP, 2250 Military Road, Tonawanda, NY 14150, USA. Canada: UTP, 5201 Dufferin Street, North York, Ontario M3H 5T8, Canada. Copyright © 2005 Anne-Marie de Mejía and the authors of individual chapters. The contents of this book also appear in the journal Bilingual Education and Bilingualism, Vol. 7, No. 5. All rights reserved. No part of this work may be reproduced in any form or by any means without permission in writing from the publisher. Printed and bound in Great Britain by the Cromwell Press Ltd.

Contents Anne-Marie de Mejía: Introduction Kendall King: Language Policy and Local Planning in South America: New Directions for Enrichment Bilingual Education in the Andes

vii 1

María Elena García: Rethinking Bilingual Education in Peru: Intercultural Politics, State Policy and Indigenous Rights

15

Carlos Skliar and Ronice Muller Quadros: Bilingual Deaf Education in the South of Brazil

35

Anne-Marie de Mejía: Bilingual Education in Colombia: Towards an Integrated Perspective

48

Cristina Banfi and Raymond Day: The Evolution of Bilingual Schools in Argentina

65

Susan Spezzini: English Immersion in Paraguay: Individual and Sociocultural Dimensions of Language Learning and Use

79

JoEllen M. Simpson: A Look at Early Childhood Writing in English and Spanish in a Bilingual School in Ecuador

99

Claudia Lucía Ordóñez: EFL and Native Spanish in Elite Bilingual Schools in Colombia: A First Look at Bilingual Adolescent Frog Stories

116

Introduction Anne-Marie de Mejía Universidad del Valle, Cali, Colombia

The development of bilingual education in South America can be traced back to the arrival of the Spanish and Portuguese colonisers in the 15th and 16th centuries, when Catholic missionaries began their evangelisation of the indigeneous peoples using local vernaculars, as well as Latin, Spanish and Portuguese. An official policy of compulsory Castilianisation in much of the area under Spanish control meant that in the 17th century, children of the wealthy elites attended schools where they developed literacy in the dominant colonial language. In addition, there were attempts to institutionalise the use of Spanish as the language of education in the Amerindian communities (see García, and de Mejía, this volume). More recently, however, in the 19th and 20th centuries, there have been moves to revalue bilingual education, both as a means of providing students with access to wider educational, cultural and business opportunities, and also as a way of reaffirming cultural identity. Traditionally in South America, as elsewhere, debate on bilingual education has been conducted in two separate arenas. On the one hand, there is the work carried out in majority contexts, generally involving international languages, such as English, French, Spanish and Portuguese. On the other, there are ethnic minority, community-based bilingual projects aimed at maintaining and enriching the use of native Amerindian languages, such as Quechua, Kaxinawá and Ika, or minority languages, such as the different national Sign Languages of the South American Deaf communities, as well as providing access to the dominant languages used in the different societies. In this publication, studies on bilingual education carried out in both majority and minority language contexts have been included in order to provide the reader with an integrated perspective on the issues raised in relation to bilingualism and bilingual education in the sub-continent, and to try to bridge the divide between the different traditions. This Special Issue is thus devoted to examining aspects of bilingual education and bilingualism in six South American nations: three Andean countries (Peru, Ecuador, and Colombia) and three ´Southern Cone´ nations (Brazil, Argentina and Paraguay). The languages discussed include Quechua (Quichua) and Guaraní, as well as Brazilian Sign Language, Spanish, Portuguese and English. The contributors bring a considerable depth of expertise to bear on these topics. Some of them (King, García, Skliar, Muller Quadros, de Mejía, Banfi and Day) have adopted wide-ranging linguistic and historical survey perspectives, where they discuss general tendencies across different national settings (King) or within individual countries. These enable the reader to gain a sense of some of the main issues at stake in the field of bilingual education in this part of the world. Others (Spezzini, Simpson and Ordoñez) provide more specific, field-based visions, focused on student attainment in both oracy and literacy in different bilingual contexts. All the articles included in this Special Issue were written specifically for this volume. The first paper, by Kendall King, sets the tone for the whole collection. The vii Introduction

viii

Bilingual Education in South America

author argues for the adoption of an enrichment model of bilingual education as a way forward for both minority and majority programmes in the Andean region. She incorporates insights derived from a particular research context, the Indian community of Saraguro in Ecuador, as evidence that localised language planning decisions, such as developing programmes to teach Quichua as a second language to community children, can be effective in equalising opportunities for groups which have been traditionally marginalised, as well as helping to maintain linguistic and cultural diversity across the whole region. The focus on indigenous bilingual education is continued in Maria Elena García’s paper on the development of bilingual intercultural education in Peru. The author situates her discussion on linguistic and cultural issues within a historical perspective, starting from the arrival of the Spanish in the 16th century, and traces the development of experimental bilingual projects in the country from the 1940s onwards. She then goes on to discuss the high profile given to Intercultural Bilingual Education since the 1990s, making reference both to the differing discourses of indigenous activist groups, who are interested in highlighting the particular needs of the indigenous communities as culturally and ethnically autonomous groups, and also addressing governmental concerns about the inclusion of these communities within the Peruvian nation. The third paper in this volume discusses a different minority context, the situation of Deaf bilingual education in the South of Brazil. The authors analyse recent changes in the representation of both Deafness and Deaf people in the country, which move away from oralist conceptions towards a recognition of the importance of a bilingual focus, involving Brazilian Sign Language and Brazilian Portuguese, in the construction of Deaf identity. They argue that Deaf adults have an essential role to play in this process and cite evidence of initiatives designed to involve the different actors, both Deaf and hearing, in the promotion and conciousness-raising of Deaf Culture and Education in the country. The next paper is focused on bilingual education in Colombia and makes a case for the adoption of an integrated approach towards the two differing traditions of bilingual provision in the country – Ethnoeducation programmes for ethnic minority groups, and bilingual programmes for learners of majority languages. After a brief historical survey of developments in both traditions, the author discusses in more detail the situation of majority bilingual education, considering some of the factors which have influenced the successful spread of this type of programme in Colombia in recent years. She ends by indicating areas of convergence between the two traditions as ways forward for the future. A second survey-type paper, this time focused on bilingual schools in Argentina, discusses the development of this type of educational provision from its beginnings in the early 19th century. After a consideration of similarities and differences within a wide range of colegios bilingües currently available, Banfi and Day chart their origins from Heritage or Community schools founded for the needs of particular immigrant communities, to Dual Language type programmes catering mainly for monolingual Spanish speakers. The authors conclude that in spite of a perception that Argentine bilingual schools form a homogeneous group, they are, in fact, highly diversified, and have not, as yet, been studied systematically. The last three papers in this collection are based on research studies carried

Introduction

ix

out in bilingual schools in Paraguay, Ecuador and Colombia. In the first of these, Susan Spezzini analyses processes and patterns of language use by 12th grade students in a dual language (English–Spanish), American overseas school in Asunción, Paraguay. She comes to the conclusion that the linguistic ecology of immersion schooling represented in this context is based on complex interrrelationships and interdependencies, where affective factors, learning opportunities provided by the specific language learning context, as well as the students’ own investment in their learning contribute to successful bilingual development. The author argues, however, that the diglossic nature of immersion classrooms may hinder student acquisition of a range of different registers in their two languages. The following paper is based on research carried out on narratives in Spanish and English written by first grade students in a bilingual school in Quito, Ecuador. Based on a T-unit analysis, the author examines differing writing styles and evidence of their transfer from the foreign to the native language in these productions. She concludes that while there is evidence that the children had similar syntactic ability in both languages, the types of errors observed in the two languages were very different. Moreover, the ‘elaborate style’ of writing characteristic of native Spanish speakers was absent in the children’s narratives. The author recommends the carrying out of longitudinal comparative studies on students’ biliterate development in other South American bilingual schools. The final article in the Special Issue is also concerned with bilingual (English–Spanish) narratives, this time in oral mode. Claudia Ordoñez looks at the oral narrative proficiency of a group of 15-year-old adolescents at a bilingual school in Colombia, in order to ascertain the range and level of variability in their productions. She also compares these bilingual narratives to monolingual productions by comparable students in English and in Spanish. In a potentially worrying conclusion, the author shows that while the bilingual narratives exhibit a similar level of variability to monolingual stories, they also provide evidence of underdevelopment and limitation in narrative complexity in both languages. The author thus calls for more research to be carried out to discover whether this is a result of the introduction of bilingual education programmes at an early age in the country. Although some of the contributors express certain reservations about aspects of the bilingual programmes they have researched in general, it may be said that the papers included in this publication focus on an additive and/or enrichment view of bilingualism and assess contributions from sepcific programmes within this general viewpoint. Thus, individual weaknesses or programme deficiencies alluded to should be taken as offering posibilities for further investigation and future improvement.

Language Policy and Local Planning in South America: New Directions for Enrichment Bilingual Education in the Andes Kendall King Georgetown University, Washington, DC, USA This paper discusses bilingual education model types in South America with a special focus on the Andean region, and examines the recent language planning decisions by one Ecuadorian indigenous group to formally instruct Quichua as a second language in community schools. Specifically I argue that this type of localised planning – which promotes an enrichment model of bilingual education and use of students’ heritage languages – merits greater attention from researchers and language planners as a potential avenue for equalising opportunity and sustaining linguistic diversity. The paper is divided into four sections. First, I briefly overview these local language planning decisions and their larger political context. Next, I draw comparisons between these decisions and recent shifts in US language planning and policy, and outline the shared challenges of what have been termed ‘heritage language initiatives’ across the Americas. Lastly, I discuss the connections between these policy shifts and recent developments within the field of language policy and politics, and argue that this new form of bilingual education and local language planning represents the best hope for small language survival in the face of globalisation. Keywords: language policy, heritage languages, bilingual education, Ecuador, language revitalisation, Quichua

As the papers in this special issue suggest, bilingual education in South America encompasses a wide range of programme and model types targeting vastly different populations and distinct linguistic and pedagogical goals. Although only a minority of students in any South American country are enroled in bilingual education programmes, there are a wide variety of such programmes. These range from elite programmes in the British School in Bogot
a, Colombia or the Colegio Menor San Francisco in Quito, Ecuador, both of which offer instruction primarily in English for elementary and high school students (British School, 2002; Lannak, 1999), to bilingual literacy projects for rural women in Peru which aim ‘to contribute to intercultural and indigenous literacy in Quechua and to teach Spanish with a second language methodology’ (Chirinos, 1997: 251), to the 2000 primary schools and 40 high schools which are part of Ecuador’s national system of bilingual education and make varying use of indigenous languages, despite lacking the fiscal resources of regular monolingual (Spanish-language) schools (Haboud, 2004). In distinguishing different types of bilingual education in South America, it is helpful to refer to Hornberger’s (1991) typology of bilingual education

1

2

Bilingual Education in South America

models and programme types. Whereas models are defined in terms of their linguistic, cultural, and societal goals (e.g. transitional, maintenance or enrichment), programme types are determined by their characteristics relating to student populations, teachers and programme structure. Applying Hornberger’s typology to the South America context, and to the Andean region in particular, we observe that although there are multiple bilingual education programmes, most bilingual education models are of two basic types: (1) enrichment models which promote language development and additive bilingualism for the elite; and (2) transitional (or nominally maintenance) models for indigenous sectors which, in the long term, promote subtractive bilingualism. The enrichment models consist of programmes aimed at students who are monolingual speakers of their country’s official language (Spanish), and which instruct English (or less commonly, French or another European language) as a foreign language, in some cases using it as a medium of instruction. These programmes tend to be designed for – and in many instances are only available to – students from upper-middle- and upper-class backgrounds, and are created as a means to enrich the educational and social opportunities available to these students. While English is the official language of only one of South America’s 12 countries (Guyana), it is viewed as a language of high social status and economic advancement, and thus this type of school potentially facilitates a version of elite closure, wherein competence in English simultaneously signals elite status and is accessible only to elites (cf. MyersScotton, 1993). Enrichment programmes such as these, which promote an additive type of bilingualism, have often been labelled ‘elitist’ or ‘elite’ bilingual education (cf. de Mejia, 2002; Hornberger, 1991). Transitional models consist of programmes aimed at students who are dominant or monolingual in an indigenous language, and who need to acquire Spanish as a second language. These programmes are designed as vehicles for providing meaningful instruction and literacy training in students’ first language prior to (or concomitant with) their transition to Spanish. In contrast to the first class of programmes, these students tend to come from societal groups that have long been economically and socially marginalised within the national context. Advocates of such programmes argue that use of the students’ first language provides them with a greater chance of engaging with the school curriculum, developing literacy skills, and eventually participating on more equal terms in the wider national society (PEBI, n.d.). However, some critics of these so-called ‘maintenance’ programmes point out that despite the fact that students’ first languages are used in the early grades, in reality the programmes tend to be oriented overwhelmingly towards transition to Spanish, and hence to promote a subtractive form of bilingualism. These two types of schools participate in distinct ‘imagined communities’ (Norton, 2000), and envision different sorts of futures for their graduates. Students, parents, and school staff of enrichment model schools generally participate in the same ‘imagined community’, sharing the hope that students will master academic content, become fluent in a high status international language such as English or French, move in international circles and become members of the national elite. In the transitional model schools, in contrast,

Enrichment Bilingual Education in the Andes

3

conflicting ideological systems may be at work. Teachers and school administrators working with indigenous minority language groups tend to emphasise mastery of Spanish and transition to the dominant language and national culture (cf. DINEIB, 1994). Students of these programmes may share these goals or may actively resist them (Norton, 2000). Although bilingual education – broadly defined – exists in many regions across South America, the divergent models which serve elite and non-elite students result in unequal educational opportunities, potentially contributing to the increasing inequality which has been documented in the Andean nations (CIDA, 2002; Galbraith & Kum, 2002). For instance, while the former programmes allow students to access greater social=cultural capital and opportunities, the latter offer restricted opportunities for participants to develop their first language and hence to reach their full academic, intellectual, and personal potential (cf. SkutnabbKangas, 2000). Highlighting the need and possibilities for expanded and alternative enrichment approaches, the present paper describes a relatively new type of bilingual education in South America, one which combines the enrichment, additive goals of the first model, and the speakers and languages which tend to be associated with the second. This new type of bilingual education is clearly a descendent of the enrichment model; however, the programme is unusual in that it aims to achieve additive bilingualism not among Spanish-speaking elites as has traditionally been the case, but among indigenous students through instruction of their heritage tongue as a second language. The present paper examines the recent language planning decisions by one Ecuadorian indigenous group to instruct Quichua as a second language in formal and informal contexts to community children. Specifically I argue that this type of localised planning – which was supported by recent shifts in national policy and ideology – presents one viable avenue towards maintaining linguistic diversity across the region, decreasing social inequalities between social groups, and equalising opportunities for traditionally marginalised groups. The discussion is divided into four sections. First, I briefly overview the local language planning decisions supporting this new programme and their larger political context. Next, I draw comparisons between these decisions and recent shifts in US language planning and policy, and outline the shared challenges of what have been termed ‘heritage language initiatives’ across the Americas. Lastly, I discuss the connections between these policy shifts and recent developments within the field of language policy and politics and argue that this new form of bilingual education and local language planning represents the best hope for small language survival in the face of globalisation.

Language Planning in Saraguro, Ecuador The Saraguros are an indigenous group in Ecuador who number roughly 22,000; they primarily reside in approximately 60 rural communities scattered around the town of Saraguro in the southern Andean highlands.1 The Saraguros are members of the larger Quichua nation in the country, but are locally and nationally identifiable by their distinct hair and clothing styles. In the

4

Bilingual Education in South America

second half of the last century, Saraguros shifted from Quichua dominance to Spanish dominance (King, 2000a).2 Over the last two decades, an increasing number of Saraguros have found this loss troubling and have taken action to reverse this trend. Local efforts to reverse language shift gained momentum in 1989, when the indigenous Saraguro schools achieved greater independence from the national department of education. In Saraguro and elsewhere in the country, national political shifts resulted in the establishment of local indigenous directorates to administer schools in regions where more than half of the students were indigenous (see King, 2000a). In the mid-1990s, individual Saraguro communities and the indigenous education directorate took decisions concerning the instruction of Quichua in the schools. In brief, parents, teachers and school administrators agreed that it was deeply problematic that Saraguro students were nearly Spanish monolingual. They also agreed that the schools should attempt to correct this imbalance through the instruction of Quichua as a second language in grades one to six and through the promotion of more positive attitudes towards the language. In addition, parents were to be encouraged to use the language with their children. In the years that followed, many of the community primary schools developed instructional programmes to teach Quichua as a second language and to incorporate use of the language into everyday school life. While no immersion-type programmes were developed, many of the schools did at least symbolically incorporate Quichua into the language of school life, provide basic vocabulary instruction, and teach Quichua songs and games at least a few times each week (see King, 2000a, 2003, for further details). While these efforts may appear modest, and indeed, seem unlikely on their own to reverse language shift towards Spanish (cf. Fishman, 1991), it is important to view them within their historical and sociocultural context. Indeed, 50 years ago, efforts to instruct Quichua as a second language would have been unfathomable, as the dominant language ideologies across the region have long held that Quichua (and all other indigenous languages) were inferior to Spanish and inadequate for academic purposes (cf. Dorian, 1998). Indeed, these ideologies have not completely disappeared; for instance, it is still possible to hear older, non-indigenous residents of the region comment that Quichua is related to ‘animal language’ or based on ‘animal sounds’ (King, personal experience, 1994–1995). In this light, efforts to organise a Quichua bilingual programme oriented towards language renewal rather than language shift represent seismic shifts in both linguistic attitudes and language policy in Saraguro. Indeed, although the Saraguros’ language planning decisions were largely made by community members, they must also be seen as part of a larger ideological and political shift taking place in the region. As Hornberger suggests, in South America and elsewhere, multilingual language policies which recognise ethnic and linguistic pluralism as resources for nation building are increasingly in evidence (2002). These policies, many of which envision implementation through bilingual intercultural education, open up new worlds of possibility for oppressed indigenous and immigrant languages and their speakers, transforming former homogenising and assimilationist

Enrichment Bilingual Education in the Andes

5

policy discourse into discourses about diversity and emancipation (2002: 29–30). Along similar, although more pragmatic, lines Luykx (2000) argues that the reforms which have taken place across the Andean nations have been ‘cut from the same cloth’, inasmuch as they can each be understood as direct responses to pressure from indigenous organisations and from international donors concerned with promoting more democratic and inclusive educational systems. In Ecuador, such reforms have included the officialisation of bilingual education in 1980; the constitutional recognition of the right to use indigenous languages for schooling indigenous students in 1983; the establishment of a national directorate (Direcci
on Nacional de Educaci
on Indı´gena Intercultural Bilingu¨e) for the administration of bilingual intercultural education in 1989; and the inclusion of indigenous professionals in leadership positions within that directorate (see King, 2000a; von Gleich, 1994). Thus, the Saraguro language planning decisions, while reflecting the goals and initiatives of Saraguros, were made possible – both in theoretical and practical terms – by shifts in the larger national and international context.

Educational Planning for Heritage Languages Across the Americas The expansion or enrichment of heritage language bilingual education is of course not limited to Ecuador. Policy-makers, researchers, and language activists in Europe and other parts of the Americas, for instance, have also recently drawn attention to the need for pedagogically sound heritage language programmes (cf. Huss, 1999; Huss et al., 2003; King & Hornberger, 2004; Kreeft Peyton et al., 2001). For instance, in Bolivia similar designs have been undertaken at the national level: the 1994 National Education Reform sought to introduce each of Bolivia’s indigenous languages, starting with Quechua, Aymara and Guaranı´, as both subjects and media of instruction in Bolivia’s primary schools in indigenous language-speaking areas (Hornberger & L
opez, 1998, in Hornberger & Coronel-Molina, 2004). One of the emphases of this important and sweeping reform effort is the adoption of a resource-and-enrichment orientation towards indigenous languages and their speakers. In the US context, in turn, heritage language programmes potentially include either immigrant or indigenous languages; however, as in South America, they tend to differ from ‘traditional’ bilingual education model types in that they are neither geared toward elites (as in enrichment model types) or aimed at moving students from their native tongue towards the national language (as in transitional model types). US-based researchers maintain that heritage languages are worth preserving not only because they serve as a rich emotional, communicative and cultural resource for the group itself, but because ‘they can help solve a growing national problem – America’s need for citizens who are proficient in languages other than English’ (Kreeft Peyton et al., 2001: 9). The demand for heritage language education has grown exponentially in recent years (Kreeft Peyton et al., 2001); as an example of their current popularity, Chinese language programmes in the states now reach nearly 83,000 students in more than 600 schools (Compton, 2001).

6

Bilingual Education in South America

Significantly, heritage language programmes have continued to grow in popularity, size, and number despite the less than favourable overall national climate in the US for bilingual education. This climate was darkened considerably by Proposition 227 of 1998, which all but banned bilingual instruction in California, as well as by a similar initiative that passed in 2000 in the state of Arizona and the ‘No Child Left Behind Act’ (NCLB) of 2001 at the federal level. Title III of the NCLB Act replaces what for 34 years has been known as the ‘Bilingual Education Act’ (Title VII), placing greater emphasis on English language proficiency, national standards, and regular testing. While the NCLB Act does not explicitly limit the quantity or type of native language instruction or use, the language of the Act makes clear that rapid transition to English is top priority. As Crawford points out, this is a ‘180-degree’ reversal in language policy: ‘whereas the 1994 version of the Bilingual Education Act included among its goals ‘‘developing the English skills. . .and to the extent possible, the native-language skills’’ of LEP students, the English Language Acquisition Act (Title III) stresses skills in English only’ (2002: 1).3 Despite these moves away from the use of students’ first languages in education and the increasing stress on English language acquisition in US education policy, at the same time there is also greater recognition that heritage languages need to be cultivated and treated as resources. In the US, this is evidenced by the growth of interest in heritage language research and pedagogy, and exemplified by the Center for Applied Linguistics and the National Foreign Language Center’s ‘Heritage Language Initiative’. This project aims to ‘overcome the neglect’ of heritage languages in the US through research, publications, and events such as the ‘First National Conference on Heritage Languages in America’ in 1999 and a second national conference in October of 2002 (CAL, 2002; Kreeft Peyton et al., 2001). The initiative has established research priorities and holds summer institutes for heritage language teachers. One issue frequently highlighted through these efforts is that although heritage language and enrichment education programmes have gained greater attention and visibility in recent years in both North and South American contexts, numerous shared challenges to effective implementation remain.

Three Shared Challenges Analysis of the Saraguro case and others described in the literature suggest that there are shared obstacles in implementing enrichment model, heritage language bilingual education programmes. As made clear by the research in this area, the challenges are numerous and substantial (cf. Huss et al., 2003; Kreeft Peyton et al., 2001); the discussion here is only intended to highlight those which may be the most significant and require the most effort and planning to overcome (see also CAL, 2000). These include: (1) the development of appropriate pedagogical approaches and academic materials; (2) moving beyond controversy concerning the dialect or variety to be used in instruction; and (3) addressing conflicting language ideologies within the community and gaining widespread, committed support from parents, students, teachers and other community members.

Enrichment Bilingual Education in the Andes

7

Turning first to the issue of pedagogy and materials, it is clear that planners face special challenges in this arena. As Compton (2001) notes, ‘improving the quality and scope of the curricula and materials used in heritage language schools and classes has been of great concern to heritage language educators’ (2001: 153). One major challenge is reinventing instruction so that it moves beyond traditional, teacher-centred foreign language pedagogy. In Saraguro schools, for instance, it is common for teachers to rely on traditional pedagogical approaches, with lessons that overwhelmingly emphasise memorisation of numbers, colours, and basic objects. As an example, in one Saraguro school which was regularly visited, of the 17 formal Quichua lessons observed over the 1994–1995 school year, 11 consisted of vocabulary basics, such as parts of the body, numbers, and names of animals (King, 2000a). Furthermore, because there was no set curriculum, there was little to no increase in the difficulty of the vocabulary words of lessons over time, as might have been accomplished, for example, by gradually moving from the names of the animals, to how they look, what they eat, and what they contribute to the community. Indeed, in that same school, examination of the vocabulary taught to third and fifth grade students on the same day by different teachers revealed that overlap was roughly 50%, suggesting that not only did the practised vocabulary remain the same from week to week, but also varied little from year to year. More generally, the lessons did not demand that students use the language in any communicatively realistic way, nor did they push students to develop conversational skills (King, 2000a). This seems particularly unfortunate given the fact that at certain Saraguro schools, students arrived with strong comprehension skills and varied speaking abilities, and the skills which students had already mastered outside of school – for example, listening comprehension – were the primary ones developed at school. Although teachers seemed to be aware of the linguistic needs of the students (as one teacher noted, the students ‘already know some [Quichua], it just needs to be perfected’), their attempts to expand students’ competence, to have students begin to actually speak the language, and to move towards ‘perfection’, were, in fact, few and far between. Saraguro schools are far from unique in falling short of their academic goals, as many of these characteristics are shared by other school-based heritage language programmes across the Americas. For example, Steven Greymorning, a teacher of Arapaho in the US state of Wyoming, describes the Arapaho classes as ‘one dimensional, focusing almost exclusively on numbers, colours, and a variety of vocabulary words including food items and animals’ (1999: 11). He furthermore notes that students had few opportunities to use Arapaho in meaningful ways, observing that ‘although students from grade two to six could recite a translation of the Pledge of Allegiance, these students did not have any comprehension of what they were actually saying’ (1999: 11). Likewise, Michael Krauss, one of the most noted experts on endangered languages, observes of heritage language teaching programmes in the US that: quite often we see the bilingual Native language teacher of parental or grandparental generation, trained and teaching in school her or his ancestral language in the classroom, writing it with chalk on a black

8

Bilingual Education in South America

board, teaching colours, numbers, names of animals, maybe even verbs or sentence structure, to children who receive an hour a week or an hour a day of the language in that way, for a few years on and off. (1998: 15) Experience thus suggests that new pedagogical approaches are required, ones that not only make use of the best of our knowledge from the areas of second language acquisition and foreign language pedagogy, but also take into account the special needs of learners, both emotional (e.g. connecting the language with students’ heritage and recognising the potentially awkward and painful process of recuperating it) and linguistic (e.g. having teachers both recognise and build on students’ dormant or latent knowledge of the language). Addressing these challenges first and foremost requires an investment in human resources. While finding trained minority language teachers is often an obstacle (cf. Lopez, 1990), this issue is particularly acute in cases where the threatened heritage language is, by definition, not widely spoken. A second shared challenge to heritage language initiatives is to adequately address the sticky issues surrounding the selection of an instructional variety when there are multiple dialects or varieties of the threatened heritage language. The example of Quichua in Ecuador is again illustrative. The process of standardising Quichua was formally initiated in 1981 when representatives of the different Ecuadorian varieties of Quichua agreed upon a unified variety (Montaluisa, 1980; von Gleich, 1994). Quichua language planners, most of whom were indigenous political and education leaders, made decisions in two areas: a unified writing system, consisting of 20 consonants and three vowels; and lexical modernisation and purification, primarily through replacement of Spanish loan words with neologisms. These decisions were codified in subsequent dictionaries and grammars (CONAIE, 1990; MEC, 1982) and constituted a major step towards the standardisation of Ecuadorian Quichua, known as Quichua Unificado. Aims included facilitating the development of Quichua materials, and contributing to the maintenance and even revitalisation of the language. Despite the new written standard, it was expected that the regional varieties would continue to exist in their spoken forms (CONAIE, 1990). In practice, however, these goals have proved elusive. For instance, King (1999, 2000a) reports that as Quichua materials have been introduced into Saraguro schools, two distinct Quichua varieties have emerged. The Quichua pedagogical materials promote the nationally standardised variety (Quichua Unificado); this stands in contrast to what is commonly referred to as Quichua Aute´ntico, spoken by elderly and rural dwellers. Because children and young adults studying Quichua Unificado have not mastered the phonological system or the lexicon of the local variety, they learn not only to read, but also to speak Quichua Unificado. While the varieties are mutually intelligible to most, there are clashes and gaps in communication between the older and younger Quichua speakers; such tensions undermine Quichua use and exacerbate generational and social divisions. Again, it bears noting that this is a challenge shared by many other heritage language initiatives; often, a common thread across these controversies is the notion of ‘authenticity’, and in particular, which variety authentically

Enrichment Bilingual Education in the Andes

9

represents a particular community’s identity (Fishman, 1999). For instance, in the Hawaiian context, the existence of multiple varieties, competing standards, and few remaining native speakers have made it possible for individuals ‘with the greatest economic and political means to establish positions of authority [. . .] to define authenticity and [to] use that definition to promote their versions of language to be more authentic and therefore superior to others’ (Wong, 1999: 94–115). Further, ‘because the version that is promoted always happens to coincide with the promoter’s own version, the dominance of the promoter is perpetuated along with the authority to define his or her own superiority’ (1999: 94–115). Contestations and competitions of this sort result not only in hierarchies of status which potentially demean certain languages and their speakers in the classroom, but also the potential rejection of variety and the heritage language programme altogether. Addressing this sensitive issue, Schwartz recommends that teachers ‘acknowledge and use the dialect varieties of the students; validate their knowledge of the language by asking them to share and explain idiomatic expressions or vocabulary’ (2001: 239). While such practices require only minor pedagogical adjustments, they have the potential to go a long way toward diffusing negative attitudes and creating a respectful classroom environment where all varieties are equally valued. However, other experts have questioned the need for using a standard altogether. Luykx (2000), for example, challenges the assumption that standardisation of the written form is essential to the language’s elaboration and expanded use (2000). Examining the case of Quechua, she maintains that in practice the emphasis on standardisation tends to shift resources away from those domains which are Quechua’s stronghold, toward those where its disadvantage relative to Spanish is greatest; to prioritise concerns of linguists over those of most Quechua speakers; and to link dialectal variation with status and class hierarchies. The result is that standardisation tends to undermine meaningful use of indigenous languages in communities and schools. Although standardisation has long been viewed as an important means to improving status and prestige (Hornberger & King, 1996; Stewart, 1968), it seems appropriate to question whether it is always essential to the establishment of successful heritage language programmes. A third common challenge to all heritage language initiatives is addressing complicated, potentially conflicting language ideologies within the community and garnering widespread local support. Resistance to use of the heritage language in schools can be found on many fronts. For instance, some community members may feel that it is inappropriate to teach a language which has traditionally been transmitted in private, home contexts; or that such efforts are a waste of time and that energy is more wisely invested in more practical endeavours; or they may believe that it is a case of ‘too little, too late’. In Saraguro, conflicting language ideologies have undermined school efforts in multiple ways. For instance, as detailed in King (2000b), within Saraguro communities there is tension between the ‘positive’ language attitudes and explicit discourse concerning Quichua and the less obvious, implicit ideologies that devalue the language and reflect its low status within the wider society.

10

Bilingual Education in South America

These are evident, for instance, in parents’ stated preference to learn Spanish first and best, often hand-in-hand with the vague hope or idea that children will learn Quichua ‘later’ or ‘when they are older’. There are also differences within the communities concerning who or which bodies are primarily responsible for transmission and maintenance of the language, with teachers holding parents responsible for children’s lack of Quichua knowledge, and parents believing that the language will be ‘safe’ in the hands of the school (King, 2000a). Similar sorts of tensions are played out in varied ways across different communities. Martinez (2000), for instance, details the ways in which efforts to introduce Pueblo language and culture into the local schools in the US state of New Mexico have been challenged. Although the heritage language=enrichment programmes were designed with the linguistic and cultural needs of Pueblo students in mind, because they were funded with New Mexico state funds, they were required to be open to both Pueblo and non-Pueblo students. As Martinez recounts, ‘when the Tribal Council, the governing body of the Pueblo community, learned of the school district’s intentions, they reacted angrily, declaring that no outside entity or person had the right to use their language for any purpose without the council’s participation and official consent’ (2000: 211–213). The conflicting linguistic ideologies relating to ownership of the language and appropriate contexts for use further strained school–community relations, and hampered sorely needed language renewal efforts. While it is not uncommon for communities and schools to share different visions, this situation is exacerbated in minority language contexts where the group has generally been marginalised and subject to discrimination of various sorts, where the community’s very survival has entailed the development of resistance strategies, and where sensitivity to issues of identity, authenticity, and status are heightened (cf. Dorian, 1994; Fishman, 1999). It is critically important that language and education planners – regardless of their status in the community – learn about and take into account the complicated local ideologies which surround heritage languages, language learning, and hence heritage language programmes.

Summing Up While many challenges remain to be overcome in the development of successful heritage language programmes for indigenous populations, such programmes merit our attention and energy not only because they provide enrichment opportunities for traditionally marginalised groups, affording access to additive bilingualism to those from non-elite sectors, and potentially helping to stem the worldwide trend towards language loss, but also because they are in demand and desired by many indigenous groups. Indeed, while there are multiple indigenous perspectives on the appropriate role of indigenous languages and cultures in schools (cf. Aikman, 1999), many of the enrichment bilingual education or heritage language programmes which are presently in operation have been created and designed by the populations they serve or in response to the demands of those populations. There is some

Enrichment Bilingual Education in the Andes

11

evidence that such attention and energy has been forthcoming from some sectors; for instance within the field of language planning and policy there is a clear trend towards focusing on the needs of speakers of minority and threatened languages. Along these lines, Ricento (2000) has argued that research in the area of language policy and planning is best divided into three historical phases. In the first of these phases, language planning work was perceived as politically neutral, and as a technical, problem-solving exercise; goals frequently focused on achieving unification, modernisation, and efficiency within newly formed nationstates (2000: 198–199). Work within the second phase, in contrast, began to question the feasibility and the neutrality of these goals, as modernisation policies in the developing world failed and notions such as the ‘native speaker’ and ‘diglossia’ were critically scrutinised; this phase is generally characterised by a growing awareness of the negative effects – and inherent limitations – of planning theory and models, and a realisation that sociolinguistic constructs such as diglossia, bilingualism, and multilingualism were conceptually complex and ideologically laden and could not easily fit into existing descriptive taxonomies. (2000: 202) The third and most recent phase, in turn, builds on these insights, but is also informed by critical theory and a language ecology perspective, and concerned more specifically with the role of ideology in language policy, the maintenance of threatened languages and linguistic diversity, and support for linguistic human rights, each of which is in evidence in heritage language education. Given the multiple challenges to small languages presented by the ever-increasing forces of globalisation, this type of localised planning, supported by national policy and ideology, seems essential to the survival of linguistic diversity. While it is widely recognised that school-based efforts are far from a panacea to language loss – and indeed may well shift energy and attention away from the domains of intergenerational transmission (Fishman, 1991) – it is also the case that school efforts such as these hold great potential. In this vein, McCarty has argued that ‘while school-based action alone will not ‘‘save’’ threatened indigenous languages, schools and their personnel must be prominent in efforts to maintain and revitalise those languages’ (1998: 28). To ignore the potential of enrichment bilingual education models is to toss aside ‘enormous language maintenance resources’ (McCarty, 1998: 28). This paper calls greater attention to the need to continue to work in this direction, and specifically, to move beyond the traditional dichotomy of Andean bilingual education types: enrichment (typically English-oriented) programmes promoting additive bilingualism for elite students on the one hand, and transitional (Spanish-oriented) programmes promoting subtractive bilingualism for indigenous groups on the other. Although enrichment programmes for non-elites in South America have begun to take root, as highlighted here, more attention, research, and resources need to be allocated towards this alternative vision of bilingual education. Such a shift would not only provide multiple advantages to the student participants, but also affords one potential avenue for maintenance of the region’s linguistic and cultural diversity.

12

Bilingual Education in South America

Correspondence Any correspondence should be directed to Dr Kendall King, Department of Linguistics, Georgetown University, Intercultural Center 458, Washington DC 20057, USA ([email protected]). Notes 1. Much of this discussion is based on one year of intensive ethnographic fieldwork conducted in Saraguro, Ecuador in 1994–1995. See King (2000a) for detailed discussion of Saraguro context and research methods. 2. Quechua is the term used to refer to the varieties spoken in Peru, Bolivia, and parts of northern Chile; it is also the cover term for all varieties of the language. Quichua is used exclusively to refer to the Ecuadorian varieties of the language. The difference in terms stems from the differing phonological evolution of the language in Ecuador, as compared to the other countries. In the former case, the uvular stop =q= has been lost, and with it the lowering of the =i= vowel to =e= in proximity to the =q=. 3. Another example of this new climate: as part of the NCLB Act, the National Clearinghouse for Bilingual Education (NCBE) was renamed the National Clearinghouse for English Language Acquisition and Language Instruction Educational Programs (now known as NCELA); correspondingly, the Office of Bilingual Education and Minority Language Affairs (OBEMLA) has become the Office of English Language Acquisition, Language Enhancement and Academic Achievement for Limited-English-Proficient Students (OELALEAALEPS).

References Aikman, S. (1999) Intercultural Education and Literacy: An Ethnographic Study of Indigenous Knowledge and Learning in the Peruvian Amazon. Philadelphia: John Benjamins. British School (2002) Official homepage. On WWW at: http:==www.gimnasio-britanico. edu.co= Canadian International Development Agency (CIDA) (2002) Canadian Cooperation in South America 1999 Edition: Introduction. On WWW at: http:==www.acdi-cida. gc.ca. Center for Applied Linguistics (CAL) (2000) Heritage Language Research Priorities Conference Report, UCLA, September 21–23, 2000. On WWW at: http:==www.cal. org=heritage=priorities.html Center for Applied Linguistics (CAL) (2002) Heritage Languages Initiative. On WWW at: http:==www.cal.org=heritage Chirinos, A. (1997) An experience of indigenous literacy in Peru. In N. Hornberger (ed.) Indigenous Literacies in the Americas: Language Planning from the Bottom Up (pp. 251– 266). New York: Mouton de Gruyter. Compton, C.J. (2001) Heritage language communities and schools: Challenges and recommendations. In J. Kreeft Peyton, D.A. Ranard and S. McGinnis (eds) Heritage Languages in America: Preserving a National Resource (pp. 145–166). Washington, DC: CAL=ERIC=Delta Systems Inc. ˜ ucanchic Confederaci
on de Nacionalidades Indı´genas del Ecuador (CONAIE) (1990) N Shimi 1. Quito: MEC, DINEIIB, CONAIE. Crawford (2002) The Bilingual Education Act (1968–2002). Language Policy Research Unit, Arizona State University. On WWW at: http:==www.asu.edu=educ=epsl= lpru.htm Direcci
on Nacional de Educaci
on Intercultural Bilingu¨e (DINEIB) (1994) Modelo de educaci
on intercultural bilingu¨e. Pueblos Indı´genas y Educaci
on 29–30, 5–142. Dorian, N. (1994) Purism vs. compromise in language revitalization and language revival. Language in Society 23, 479–494.

Enrichment Bilingual Education in the Andes

13

Dorian, N. (1998) Western language ideologies and small-language prospects. In L.A. Grenoble and L.J. Whaley (eds) Endangered Languages: Current Issues and Future Prospects (pp. 3–21). Cambridge: Cambridge University Press. Fishman. J.A. (1991) Reversing Language Shift. Clevedon: Multilingual Matters. Fishman, J.A. (1999) Comments and reflections. Anthropology and Education Quarterly 30 (1), 116–124. Galbraith, J.K. and Kum, H. (2002) Inequality and globalization: Judging the data. Presentation at the World Bank. June 18, 2002. On WWW at: http:==utip.gov.utexas.edu Greymorning, S. (1999) Running the gauntlet of an indigenous language program. In J. Reyner, G. Cantoni, R.N. St. Clair and E.P. Yazzie (eds) Revitalizing Indigenous Languages (pp. 6–16). Flagstaff, AZ: Center for Excellence in Education, Northern Arizona University. Haboud, M. (2004) Quichua language vitality: An Ecuadorian perspective. International Journal of the Sociology of Language (Special issue on Quechua sociolinguistics edited by K.A. King and N.H. Hornberger) 167, 69–81. Hornberger, N.H. (1991) Extending enrichment bilingual education: Revisiting typologies and redirecting policy. In O. Garcia (ed.) Bilingual Education: Foccushrift in Honor of Joshua A. Fishman (Volume 1) (pp. 215–234). Philadelphia: John Benjamins. Hornberger, N.H. (2002) Multilingual language policies and the continua of biliteracy: An ecological approach. Language Policy 1 (1), 27–51. Hornberger, N.H. and Coronel-Molina, S. (2004) Quechua language shift, maintenance and revitalization in the Andes: The case for language planning. International Journal of the Sociology of Language (Special issue on Quechua sociolinguistics edited by K.A. King and N.H. Hornberger) 167, 9–67. Hornberger, N.H. and King, K.A. (1996) Language revitalization in the Andes: Can the schools reverse language shift? Journal of Multilingual and Multicultural Development 17 (6), 427–441. Hornberger, N.H. and L
opez, L.E. (1998) Policy, possibility and paradox: Indigenous multilingualism and education in Peru and Bolivia. In J. Cenoz and F. Genesee (eds) Beyond Bilingualism: Multilingualism and Multilingual Education (pp. 206–242). Clevedon: Multilingual Matters. Huss, L. (1999) Reversing language shift in the far north: Linguistic revitalization in northern Scandinavia and Finland. Studia Uralica Upsaliensia 31. Uppsala, Sweden. Huss, L., Camilleri, A. and King, K.A. (eds) (2003) Transcending Monolingualism: Linguistic Revitalisation in Education. Lisse, The Netherlands: Swets & Zeitlinger. King, K.A. (1999) Inspecting the unexpected: Language corpus and status shifts as aspects of language revitalization. Language Planning and Language Problems 23 (2), 109–132. King, K.A. (2000a) Language Revitalisation Processes and Prospects: Quichua in the Ecuadorian Andes. Clevedon: Multilingual Matters. King, K.A. (2000b) Language ideologies and heritage language education. International Journal of Bilingualism and Bilingual Education 3 (2), 167–184. King, K.A. (2003) Language pedagogy and language revitalization: Lessons from the Andes and beyond. In L. Huss, A. Camilleri and K.A. King (eds) Transcending Monolingualism: Linguistic Revitalisation in Education. Lisse, The Netherlands: Swets & Zeitlinger. King, K.A. and Hornberger, N.H. (eds) (2004) International Journal of the Sociology of Language (Special issue on Quechua sociolinguistics edited by K.A. King and N.H. Hornberger) 167. Krauss, M. (1998) The condition of Native North American languages: The need for realistic assessment and action. International Journal of the Sociology of Language 132, 9–21. Kreeft Peyton, J., Ranard, D.A. and McGinnis, S. (eds) (2001) Heritage Languages in America: Preserving a National Resource. Washington, DC: CAL=ERIC=Delta Systems Inc.

14

Bilingual Education in South America

Lannak, J. (1999) The founding of Colegio Menor San Francisco de Quito. Journal of Education 181 (1), 17–37. Lopez, L.E. (1990) Development of human resources in and for bilingual education in Latin America. Prospects 20 (3), 311–319. Luykx, A. (2000) Diversity in the new world order: State language policies and the internationalization of Quechua. Paper presented at the 2nd Spencer Early Career Institute in Anthropology and Education: Globalization and Education. Chicago, IL. Martinez, R.B. (2000) Languages and tribal sovereignty: Whose language is it anyway? Theory into Practice 39 (4), 211–219. McCarty, T.L. (1998) Schooling, resistance, and American Indian languages. International Journal of the Sociology of Language 132, 27–41. Mejia, A.M. de (2002) Power, Prestige, and Bilingualism: International Perspectives on Elite Bilingual Education. Clevedon: Multilingual Matters. ˜ ucanchic Shimiyuc-Panca. Ministerio de Educacion y Cultura (MEC) (1982) Caimi N Quito: Ministerio de Educaci
on y Cultura, Pontificia Universidad Cat
olica del Ecuador. Montaluisa, L.O. (1980) El vocabulario general de la lengua Quichua para el Ecuador. Revista de la Universidad Cat
olica 25, 99–119. Myers-Scotton, C. (1993) Elite closure as a powerful language strategy: The African case. International Journal of the Sociology of Language 103, 149–163. Norton, B. (2000) Non-participation, imagined communities, and the language classroom. Plenary presented at Second Language Research Forum, Madison, Wisconsin. Proyecto de Educaci
on Bilingu¨e Intercultural (PEBI) (no date). Project informational brochure. Ricento, T. (2000) Historical and theoretical perspectives in language policy and planning. Journal of Sociolinguistics 4 (2), 196–213. Schwartz, A.M. (2001) Preparing teachers to work with heritage language learners. In J. Kreeft Peyton, D.A. Ranard and S. McGinnis (eds) Heritage Languages in America: Preserving a National Resource (pp. 229–252). Washington, DC: CAL=ERIC=Delta Systems Inc. Skutnabb-Kangas, T. (2000) Linguistic Genocide in Education – or Worldwide Diversity and Human Rights? Mahwah, New Jersey. Stewart, W. (1968) A sociolinguistic typology for describing national multilingualism. In J.A. Fishman (ed.) Readings in the Sociology of Language (pp. 531–545). The Hague: Mouton. von Gleich, U. (1994) Language spread policy: The case of Quechua in the Andean republics of Bolivia, Ecuador, and Peru. International Journal of the Sociology of Language 107, 77–113. Wong, L. (1999) Authenticity and the revitalization of Hawaiian. Anthropology and Education Quarterly 30 (1), 95–114.

Rethinking Bilingual Education in Peru: Intercultural Politics, State Policy and Indigenous Rights Marı´a Elena Garcı´a Sarah Lawrence College, New York, USA This paper explores recent changes in Peruvian national education policy and the effects these have had on indigenous populations. Situating Peruvian education reforms within a context of international multicultural development, the paper traces the history of reforms as implemented by national and international actors in varying degrees and combinations. Against this historical backdrop, the paper focuses on changes in the 1990s that in theory promote bilingual intercultural education on a nationwide scale for all Peruvian citizens, but in practice are concentrated in rural indigenous areas. In particular, the paper examines some gaps between intercultural education rhetoric and implementation. I conclude with a discussion of the uneven achievements and unintended consequences of bilingual education programmes and provide some suggestions for policy-makers in the region. Keywords: intercultural education, Quechua, Peru, indigenous, Andes, language policy

Introduction On 28 July, 2001, Alejandro Toledo, Peru’s then newly elected president, addressed the nation from the sacred ruins of Machu Picchu, sang in Quechua, danced to huaynos, and promised to work toward indigenous rights. Throughout his campaign, Toledo and his supporters had made much of his indigenous ancestry, his background of poverty, and of his Belgian, Quechuaspeaking wife, Eliane Karp. Since his inauguration, Karp has been the person to emerge as a staunch supporter of indigenous cultural, social, and political rights.1 An important aspect of Karp’s claims for the defence of indigenous rights has been her support for the implementation of bilingual education in indigenous schools, or as it has been labelled in Peru, of ‘bilingual intercultural education’. Yet, about a decade earlier, progressive intellectuals in the country had already begun forcefully advocating indigenous rights by demanding the institutionalisation of a national curriculum that reflected the ethnic, cultural, and linguistic diversity of Peru. This paper focuses on the educational changes that took place during the 1990s, and the effects of reform on indigenous populations. First, I provide a brief historical background to situate language policy in Peru in a broader context, and then turn to the politics of contemporary education reform. Finally, I discuss some gaps between intercultural education theory and practice, and provide some suggestions for policy-makers in the region.2

15

16

Bilingual Education in South America

Peruvian Language Policy in Historical Perspective The Spaniards killed our Inca because they were very greedy. The Incas did not know paper or writing. When God wanted to provide them with paper, they rejected it because they did not send messages on paper but in vicun~a3 threads. . .. These threads were used as books but the Spaniards did not want them to exist so they gave the Inca a paper saying, ‘This paper talks’. The Inca responded, ‘I don’t see that it talks, this is nonsense, you want to fool me’. And he threw the paper to the floor. How could the paper talk if the Inca didn’t know how to read? This is how our Inca provoked his own death at the hands of the Spaniards. Since then the Inca-king has disappeared. . .. The Spaniards killed everybody. But now I say, what will the Spaniards say when our Inca returns? (Testimony from Gregorio Condori Mamani, cited in Valderrama & Escalante, 1977: 50, my translation) Before the encounter with the Spanish, the Incas had established Quechua as the official language of their Empire (Cerr
on-Palomino, 1989; Mannheim, 1991; Torero, 1983). While it was compulsory for local nobility, civil servants, administrators and traders, the use of particular languages or dialects of other ethnic groups in the region was not prohibited. In fact, bilingualism in the on-Palomino, 1989; von Tawantinsuyu4 was a general societal practice (Cerr
Gleich, 1994).5 As a consequence of the establishment of colonialism, Spanish replaced Quechua as the official language. However, the influence of Quechua not only continued, but it expanded beyond what had been the Inca Empire because of Spanish language policy.6 Shortly after their arrival in the Andes, the Spaniards had recognised the potential of Quechua as a lengua general (lingua franca) for administrative purposes, and particularly for proselytisation (Mannheim, 1991; von Gleich, 1994). Exploiting its role as general language, Spaniards promoted the use of Quechua as a vehicle of ‘linguistic homogenisation’ (Mannheim, 1991) that would facilitate the consolidation of colonial control, especially in areas where linguistic fragmentation was most significant (Cerr
on-Palomino, 1989; Heath & Laprade, 1982; Mar-Molinero, 1995). Thus, as a tool used for the establishment of colonial power, Quechua prevailed. However, it prevailed as the language of the colonised peoples, a condition which placed it unequivocally below Spanish. Significantly, Quechua was subordinate to Spanish not only because Spanish was the language of the new rulers, but because it was an oral language. As illustrated by Condori Mamani’s account of Atahualpa’s capture at the beginning of this section, Spanish also represented the imposition of writing as a source of political power and domination. Bruce Mannheim notes three beliefs about the relationship between language, culture and society which were widespread among the Spanish colonists: (1) speaking the same language forges bonds among speakers; (2) a language can be imposed on a defeated population by the right of conquest; and (3) language preserves cultural identity (Mannheim, 1991: 68). From the 16th century onward, there was a clear recognition of the central role of language in the maintenance of indigenous cultural identity. Although at first colonial language policy favoured the maintenance of Quechua, primarily because it

Rethinking Bilingual Education in Peru

17

was believed that maintaining the indigenous language would facilitate the transmission of Christianity, by the mid-17th century Castilianisation had become ‘fundamental to successful Spanish control of the Andes’ (Mannheim, 1984: 296). The most drastic change in policy toward Castilianisation, however, occurred in 1770, during the reign of Charles III, when it was explicitly declared that Castilianisation should be compulsory and should be used to suppress indigenous languages and by extension, indigenous culture. Despite royal decrees, however, administrators continued to communicate with Quechua speakers in Quechua. The upper classes particularly preferred to learn Quechua rather than allow the Indians under their control to learn Spanish, so as to maintain their privileged social and economic position over them. Counter movements to Spanish attempts at eradicating Quechua did exist, and these included the ‘self-consciously nationalistic cultivation of Southern Peruvian Quechua’ (Mannheim, 1984: 299) among Cuzco elites, who had been trained in bilingual schools and therefore had knowledge of Spanish. With policy changes toward compulsory Castilianisation, rural schools were created where Indians were supposed to study and learn Spanish. However, these were largely ineffectual, especially since teaching Spanish to Indians ran counter to the interests of land owners who preferred to maintain an uneducated labour force. Yet, as implied by the mention of bilingual Cuzque~ no elites, indigenous nobles and indigenous mestizos (such as the indigenous chronicler Garcilaso de la Vega) did attend schools where they learned how to read and write in the dominant language. These indigenous elites were also responsible for the beginning of literary production in Quechua. This was a significant development in the history of this indigenous language, and, following Benedict Anderson’s thoughts on the significance of the origins of national consciousness (Anderson, 1991), a significant step toward the emergence and manifestation of nationalist sentiment among bilingual elites. One of these bilingual mestizos, a merchant by the name of Jose´ Gabriel Condorcanqui (Tu´pac Amaru II), claimed direct descent from the last Inca murdered by the Spanish – Tupac Amaru I – and led a revolt against the Spanish administration which, although brutally crushed less than a year after its inception in 1780, is often cited as the precursor to Peruvian independence (O’phelan Godoy, 1985). Because of Tupac Amaru’s proclamation of the return of indigenous governance, colonial authorities associated the cultivation of Quechua cultural practices, including Quechua drama and other literature, with political nationalism and revolution. After the rebellion failed in 1781, Quechua language, cultural practices, Quechua theatre and other literary expressions were explicitly banned, a prohibition which remained in effect for almost 200 years. In the late 1700s, ‘language was the source and pillar of cultural memory in a situation in which what was wanted was forgetting’ (Mannheim, 1984: 298). As is evident from the large number of Quechua speakers throughout the Americas today,7 attempts to eradicate Quechua failed. However, Spanish colonisers were successful in relegating it to its current status as an undervalued language, even if it was used (as were other indigenous languages) during Creole independence movements as a symbol of liberation (Cerr
onPalomino, 1989).8

18

Bilingual Education in South America

Addressing the use of language during the wars of Independence in early 19th century Latin America, Benedict Anderson notes that as the language of the former coloniser, Spanish did not play a significant role in the process of national liberation, as it could not be considered a symbol of difference from Spain: Whether we think of Brazil, the USA, or the former colonies of Spain, language was not an element that differentiated them from their respective imperial metropoles. All, including the USA, were Creole states, formed and led by people who shared a common language and common descent with those against whom they fought. Indeed, it is fair to say that language was never even an issue in these early struggles for national liberation. (Anderson, 1991: 47) Quechua (and other Amerindian languages), on the other hand, was used by liberators such as San Martı´n to establish symbolic links to American territory (Cerr
on-Palomino, 1989: 22) that marked a significant difference with Spain. While Spanish may not have necessarily been ‘an issue’ in Latin American independence struggles, it most certainly played a role – in its written form – in the ensuing process of Latin American nation-building. While he does note that Spanish in Latin America did not operate in the same way as European linguistic nationalism, Anderson also states: [T]he independence movements in the Americas became, as soon as they were printed about, ‘concepts’, ‘models’ and indeed ‘blueprints’. In ‘reality’, Bolı´var’s fears of Negro insurrections and San Martı´n’s summoning of his indigenes to Peruvianness jostled one another chaotically. But printed words washed away the former almost at once, so that, if recalled at all, it appeared an inconsequential anomaly. Out of the American welter came these imagined realities: nation-states; republican institutions, common citizenship, popular sovereignty, national flags, and anthems, etc. (1991: 81) Certainly by the 20th century, language had become a factor in creating national identity. As the language of the ruling classes, Spanish became (and remains) the national and=or official language of the majority of Latin American countries. In Peru as in other countries, efforts to create a unified nation emphasised the Castilianisation of indigenous populations, and pointed toward the gradual extinction of indigenous languages. As we will see below, with a few exceptions, such as General Velasco’s officialisation of Quechua in 1975, Quechua has been subordinated to Spanish as the official language in the various constitutions from the time of independence (1821) up to the present day.

Bilingual Education and the State: Politics and Policies Before 1993 Although there have been several experimental bilingual education projects throughout the highland and lowland regions of Peru since the 1940s, most have been isolated experiences, initiated by individuals and organisations independent of the Peruvian state, and financed primarily with international funding (Pozzi-Escot, 1981).

Rethinking Bilingual Education in Peru

19

The first attempts at bilingual education in Peru came from the Summer Institute of Linguistics (SIL) or the Wycliffe Bible Translators, a religious organisation that operates worldwide, translating the Bible into indigenous languages and providing literacy training for indigenous groups as a step toward their conversion to Christianity. Until the early 1990s, the SIL was almost entirely responsible for bilingual education in the Peruvian Amazon region (de Vries, 1990). Other bilingual education projects throughout both lowland and highland areas have been experimental, temporary,9 and transitional (teaching in the indigenous language only as a way to facilitate learning in Spanish), and most have been funded by international organisations (e.g. United States Agency for International Development, German Institute for Technological Cooperation).10 The most recent and the most often cited project is the Experimental Project of Bilingual Education in Puno (PEEB-P), a programme that ran from the late 1970s until 1988. Funded in large part by the German Institute for Technical Cooperation (GTZ), this was the only project designed to work towards the maintenance of the native language, as well as the development of Spanish as a second language. Individuals implementing the Puno project faced many of the same problems which challenge education activists today, the principal one being the rejection of the bilingual education proposal by Quechua communities. In the case of the PEEB-P, the project was declared a failure after 11 years, although its legacy among activists as the most successful attempt at bilingual education in the highlands is routinely evoked (Hornberger, 1987). In part, this may be due to the fact that many of those who were associated with the PEEBP are now involved with the implementation of bilingual intercultural education in the highlands and other regions. A few of the leading bilingual education advocates today, in fact, had previously worked together in Puno. This project is best known for the pedagogical materials and texts it produced, still often used today.11 While, as stated above, most bilingual education projects were supported more by international funds, rather than by government commitment, there are two notable exceptions; first, the emphasis that indigenista intellectuals like Luis Valc
arcel and Jose´ Marı´a Arguedas – both men who worked at the Ministry of Education – placed on Quechua and bilingualism in the 1940s, and secondly, the Education Reform implemented by General Juan Velasco Alvarado in the 1970s. Indigenismo and bilingual education Indigenismo, a liberal, urban-based movement that emphasised the liberation and ‘uplifting’ of the Indian, emerged in the early 1900s among middle- and upper-class intellectuals. While it began as a literary and cultural movement portraying romantic visions of Andean peasants and Indians through novels, poetry, and paintings, indigenista ideology also penetrated state policies and national politics. While mestizaje became the preferred nation-building project of Latin American politicians and intellectuals throughout most of the region (Larson, 1995; Stepan, 1991; Vasconcelos, 1925), in Peru it never became an official nation-building ‘racial project’ (De la Cadena, 2000). Although it was favoured by some intellectuals from Lima (Belau´nde, 1965 [1942]; Garcı´a,

20

Bilingual Education in South America

1973 [1930]), mestizaje was forcefully rejected by Andean indigenistas who espoused sharp anti-mestizo rhetoric. Most indigenista intellectuals, however, advocated the political and economic emancipation of indigenous communities, although it is important to note that indigenista visions excluded Amazonian ‘savages’. Significantly, ‘progressive’ indigenista ideas informed the Constitution of 1920, a document that for the first time gave legal recognition to indigenous communities. During the 1920s, a series of laws and new institutions opened up channels for indigenous demands for education, land, and suffrage. In 1921, President Leguı´a created the Bureau of Indian Affairs and hosted an Indian congress from which emerged the Comite´ Pro-Derecho Indı´gena Tawantinsuyo, the first national indigenista organisation that included self-proclaimed indigenous leaders and intellectuals and called for indigenous literacy and citizenship (De la Cadena, 2000; Mallon, 1998). Jose´ Marı´a Arguedas, an Andean novelist and anthropologist, was perhaps the most impassioned advocate for the cultural autonomy of indigenous peoples. His defence of the Quechua language specifically was reflected in his literary work, and he became known for his extensive use of Quechua along with Spanish in his writing, including poetry, short stories and novels (Escobar, 1984). Like other indigenistas, Arguedas highlighted the oppression of indigenous peoples and presented somewhat utopic accounts of indigenous communal life. Unlike many of them, however, he also wrote about the challenges of migration and modernity, and labeled himself a ‘modern Quechua man’. During the 1940s indigenistas sought to move beyond purely historical accounts of indigenous life, and began working closely with Peruvian and foreign ethnologists conducting research into contemporary life in indigenous communities. Indigenismo, then, became infused with a social scientific air, and gradually, the implementation of any state programme targeting indigenous populations required consultation with and agreement from ethnologists working in the designated areas (Valc
arcel, 1981: 368–369). In 1946, the Peruvian Indigenista Institute was founded, and the Ministries of Education, and of Public Works also began to promote programmes highlighting the ‘cultural recovery’ of indigenous language, art, culture and religion. Also at this time, Arguedas worked as folklore curator in the Ministry of Education. Working under the Cuzque~ no indigenista Luis Valc
arcel (who was Minister of Education and Arguedas’ history professor), Arguedas invoked the importance of teaching indigenous highland peoples in Quechua. His experience as a primary school teacher in one of the most desolate regions of Cuzco provided him with a clear understanding of the adverse conditions and difficulties faced by teachers working in indigenous areas. By 1945, the Ministry of Education began implementing bilingual education in indigenous schools (Contreras, 1996). Through the Ministry, Arguedas also advocated teacher-training in bilingual pedagogy. Velasco’s education reform General Velasco’s ‘Revolutionary Government of the Armed Forces’ (1968– 1975) emerged in reaction to social unrest and the rise of peasant and guerrilla

Rethinking Bilingual Education in Peru

21

movements in the highlands in the 1950s and ’60s. Declaring the necessity of transforming Peru’s basic economic and social structures as the only way to counter instability and a future insurgency, and proclaiming an antiimperialistic, anti-oligarchic and fervently nationalistic ideology, Velasco launched a series of social reforms aimed at ameliorating the conditions of peasants and indigenous peoples.12 These reforms fundamentally challenged the existing power dynamics throughout the country. An important step in government efforts to incorporate highland peoples into national spaces came with the development of educational reform and with state-controlled media. Explicitly challenging the national educational system formerly in place, Velasco stated in 1971 that: The educational reform of the revolution is aimed at creating an educational system that: satisfies the necessities of the entire nation; that will reach the great masses of [indigenous] peasants, always exploited and always deliberately kept in ignorance; that will create a new consciousness among all Peruvians of the basic problems of our country; and that will contribute to forging a new type of man within a new social morality. (Velasco, 1972: 63) To achieve these stated goals, the Inca Plan – the official declaration of the government’s objectives – stated that education should be modified to fit Peruvian social ‘reality’ (Zimmerman Zavala, 1974). To this end, Velasco launched three major initiatives with respect to education. First, the Education Reform of 1972 sought to extend increasing control over educational policies and resources to all Peruvians, though it targeted primarily indigenous communities. Second, the National Policy of Bilingual Education, also in 1972, advocated the implementation of bilingual education in all highland, lowland and coastal areas where languages other than Spanish were spoken. This was a significant step, since the use of native languages for teaching had been prohibited since the 1780s. Finally, a law making Quechua a national language co-equal with Spanish, was passed in 1975. The law stated that after April 1976, the teaching of Quechua would be obligatory at all educational levels. Further, all legal proceedings involving monolingual Quechua speakers would have to be conducted in Quechua (Escobar et al., 1975: 61–63). This law, more than all others, emphasised the cultural and ethnic diversity of Peruvian society. Importantly, it highlighted the two areas where language had been previously used as a mechanism of domination over indigenous speakers: in the courts and in the schools (Escobar et al., 1975; Turino, 1991: 274). It was also the one that evoked the strongest reaction by middle- and upper-class society in Lima against Velasco and his government. Placing the Quechua language on an equal footing with Spanish was understood by these groups as symbolically placing Quechua and other indigenous people on the same level as those of European background, and as representative of the limitations the Velasco regime posed to development and ‘progress’. By legislating Quechua as a national language equal to Spanish, Peru became the first Latin American country to officialise an indigenous language. By placing Quechua next to the dominant language, however, it also exacerbated prejudices against anyone perceived to be from the Andean highlands.

22

Bilingual Education in South America

This was one of the factors leading to the replacement of Velasco with Morales Bermudez, another military leader, in 1975.13

Contemporary Education Reform: Constructing Intercultural Citizenship With the change of president (1975) and of constitution (1979), the law making Quechua an official language was changed to include Quechua not as an official national language, but rather as ‘a language of official use in the areas and in the way that the Law mandates’. However, the law that would mandate where and how Quechua could be considered an ‘official’ language was never passed (Pozzi-Escot, 1981; Rojas, 1982). Known as ‘Phase Two’ of the Revolutionary Government, this and other changes were a systematic dismantling of many of the reforms of the Velasco era, including the emphasis that had been placed on rural and indigenous education. The 1980s were also a difficult time with regards to indigenous development policies. Sendero Luminoso, a Maoist insurgency group that began its armed struggle in the highlands in 1980, immersed Peru in a bloody civil war for over a decade.14 Fighting between insurgents and government forces, and nationwide political and economic chaos devastated the entire country. Indigenous peoples in the highland and in the lowland regions, however, were the most affected by this war. Trapped between two armies, indigenous peoples, and anyone advocating indigenous rights, bore the brunt of political violence.15 Despite the political climate, however, there were some developments in language and education policies aimed at indigenous peoples. In 1985, for example, the Quechua and Aymara alphabets were officialised, and in 1987, after approximately 10 years of inactivity, President Alan Garcı´a (1985–1990) reinstated the National Office (Direcci
on General) of Bilingual Education. With the election of Alberto Fujimori as president of Peru in 1990, the country entered a new political scenario, one that combined neoliberalism and authoritarian rule. Military control of almost two thirds of the national territory and the ‘auto-golpe’ (or self-coup) of April 1992, meant the loss of civil and political rights. A turning point in this adverse political climate came in September 1992 with the capture of Abimael Guzm
an, the head of Sendero Luminoso. With the war officially over, by 1993 Peru began a slow transition toward democratic rule. In particular, the government enacted a new constitution designed, among other things, to ‘recognise and protect the ethnic and cultural plurality of the nation’ by guaranteeing the right of all people to use their own language before the state (Article 2). The constitution also highlights ‘the state’s obligation to promote intercultural and bilingual education, depending on the characteristics of each region’ (Article 17). However, characteristic of the contradictions of this period of Peruvian history, at the same time as Fujimori approved progressive changes to the constitution, he simultaneously dissolved the National Office for Bilingual Education, citing lack of funds as his reason for doing so. Due to international pressure from various organisations, and because of protests by intercultural activists throughout the country, the Office was reestablished, but only as a Unit (Unidad) within the National Office of Elementary Education (UNEBI) in 1996. Even then, most of its funding was cut off

Rethinking Bilingual Education in Peru

23

during the last months of 1996 and the first months of 1997. With the reestablishment of the UNEBI in the Ministry of Education in Lima in 1996, the minister of education invited Juan Carlos Godenzzi, an Andean linguist and one of the leading intellectuals advocating Intercultural Bilingual Education, to move from Cuzco (where he was working as programme director in an Andean research institute) to Lima and act as its chief. Godenzzi moved to Lima to head the UNEBI in May 1997 and remained in that position until his recent departure in September 2002. According to Godenzzi, between 1997 and 2001, the UNEBI (made up of only five staff members) produced 94 bilingual teaching manuals (in Aymara, five regional Quechua variations and in various Amazonian languages) (Godenzzi, 2001). With the help of several NGOs, universities and research institutes, the UNEBI’s staff also facilitated bilingual training for over 10,000 teachers working in bilingual areas, and provided enough materials so that each and every classroom in the areas where bilingual intercultural education was implemented could have its own bilingual library (Godenzzi, 2001: 6). Since the election of Alejandro Toledo as president of Peru, there has been increased emphasis on issues to do with indigenous rights. The UNEBI, for example, increased the number of its staff to 16, and was renamed the National Division of Intercultural Bilingual Education (DINEBI).16 No longer a unit within a different Division in the Ministry, it now has more autonomy over policy decisions, and thus is in a position to have greater impact on legislation affecting bilingual education and language policy. Significantly, one of the actions taken by the DINEBI almost immediately after its reconstitution, was to establish a 15-member National Consulting Committee on Intercultural Bilingual Education, of which nine are designated as indigenous professionals (Godenzzi, 2001: 4). The Committee held several national-level gatherings to consult with intellectuals, NGO workers, teacher-trainers and bilingual teachers, and then drafted a report on the National Politics of Languages and Cultures in Education. Based on this report, the DINEBI prepared a time-line (Plan Estrate´gico, 2001–2005) for the future development of bilingual intercultural education in the country. These steps are part of the effort to institutionalise intercultural bilingual education as part of state – and not only government – policy.17 Despite apparent support for multicultural agendas, there is much scepticism about government commitment to these cultural policies. However, the presence and financial support of international actors, such as NGOs and the World Bank, have helped institutionalise (albeit slowly) a multicultural development agenda throughout Latin America. It is within this broader context that the implementation of Intercultural Bilingual Education – as it is understood in Peru today – remains one of the primary demands of indigenous rights advocates. Even so, a large part of the international movement for indigenous rights requires involving the state as an active agent in the implementation of social programmes benefiting indigenous groups. In Peru, much of the struggle over the institutionalisation and implementation of alternative indigenous education, such as insuring that constitutional clauses guaranteeing multicultural rights are respected, has taken place at state level. Intercultural activists work closely with indigenous organisations throughout

24

Bilingual Education in South America

Latin America and with international funding agencies, and they make an important contribution to the development of international agendas, within which intercultural discourse must neatly fit. Yet, activists are also aggressive about maintaining and radicalising the changes they have already achieved in state education policy. In 2001 and 2002, an increasing number of indigenous leaders and advocates in Peru embraced intercultural education as an integral part of the struggle for both cultural and political rights. In fact, discussion of bilingual education and ‘interculturality’ is now indicative of broader concerns over democracy, selfdetermination, citizenship and social justice. After months of research on education and cultural activism in Cuzco, it is clear that bilingual intercultural education is an important focus-point for indigenous activism, However, intercultural discourse about democracy, respect for diversity and indigenous rights notwithstanding, there is still a wide gap between intercultural rhetoric and its implementation in indigenous communities. In the next section, I examine some of the tensions created by the difference between theory and practice, and then highlight some of the challenges and possibilities resulting from intercultural bilingual education in Peru.

Intercultural Politics, Education, and Indigenous Rights: Challenges and Possibilities Although bilingual education in Latin America has usually been linked to national integration policies that meant the eventual assimilation of indigenous groups into what is considered dominant mestizo society, advocates of intercultural bilingual education in the region are explicit about distinguishing their efforts at reform from previous attempts at assimilation through education. Rather than utilising indigenous languages only as a way to ease the transition into learning Spanish (transitional bilingual education) as has frequently been the case with bilingual programmes, bilingual intercultural education in Peru today promotes the maintenance of indigenous languages alongside the teaching of a second language (Spanish), as well as the acceptance and respect of cultural difference. Many advocates of educational reform in Peru and in other Latin American countries promote this kind of education as crucial to the cultural and ethnic autonomy of indigenous peoples. Moreover, all the NGO workers with whom I spoke in Cuzco cited similar benefits for intercultural education. One programme coordinator best summed up the many comments I collected about bilingual education during my research in Peru: By teaching children in and about their native language and by providing them with the opportunity to learn the dominant language as a second language – second to their own – children will learn to value their culture and language, and thus be able to raise their self-esteem, defend themselves from discrimination and abuse, and chart their own course through history within the dominant society. (Education Program Coordinator, Cuzco) But while autonomy and self-determination are the goals that some intercultural activists and indigenous leaders hope to achieve through intercultural

Rethinking Bilingual Education in Peru

25

education, official government documents provide a different interpretation of the benefits of a policy of bilingual education. In the first information pamphlet distributed by the UNEBI, the authors state that Peru’s ‘multicultural condition. . .challenges [Peruvian] education to promote interculturality as the fundamental principle leading to national unity’ (UNEBI, 1998). Intercultural education, then, is considered an educational process which – taking the country’s cultural diversity as backdrop – teaches cultural dialogue and promotes respectful interaction among individuals and groups of different cultural backgrounds in order to achieve national unity. A crucial part of the practical implementation of this definition is that, according to the state, Intercultural Bilingual Education reflects a ‘diversification’ of the existing national curriculum and methodology. Following guidelines from the Minister of Education, the DINEBI must be explicit about the fact that intercultural education is not intended to be separate from or parallel to the national curriculum. In other words, there are no special curricula. Instead, the national primary school curriculum has been modified or diversified to include methodologies for second language acquisition (to be implemented only in areas with indigenous populations) and culturally sensitive wording and illustrations. These policies promote the idea that indigenous peoples are not separate, but rather are a part of the nation, and thus, national education should be modified to include the existing cultural and linguistic differences within the country’s borders. After much debate between activists and government representatives over the definition and the goals of bilingual intercultural education, there seems to be a consensus that the significance of this kind of education lies in the fact that it is both sensitive to the particular needs of indigenous populations in the country as culturally and ethnically autonomous groups, as well as essential to the inclusion of these populations within the nation as Peruvian citizens. Interestingly, discussions among state representatives about what intercultural education means in Peru is often linked to a negative assessment of the Ecuadorian indigenous education proposal. One employee from the Rural Education office in the Ministry of Education, for example, during an interview stated that ‘the kind of parallel education being promoted in Ecuador is dangerous because it separates communities. It creates tension between ethnic groups and it then becomes one-sided, not intercultural like the name says’. By ‘parallel’ and ‘one-sided’, this man was referring to the exclusive focus of Ecuadorian education activists on indigenous education. During a similar conversation with some of the UNEBI’s staff, one man stated: I think the difference between us and the case of Ecuador is one of option. [Ecuadorians have] created a proposal parallel to the national proposal, and they have achieved administrative autonomy, their own curriculum, their own schools and teacher-training programs. What is dangerous about this, is that this kind of educational proposal can create a second-rate education (una educaci
on de segundo orden) which can lead to parental rejection of bilingual education centers. (1999)18 When I noted that many parents of Peruvian Quechua children also reject bilingual education in their schools, the response among those in the office

26

Bilingual Education in South America

was unanimous: Peruvian indigenous parents just did not know any better. This response – and the paternalistic assumptions it makes about parents of children affected by the implementation of bilingual education – is at the heart of the tensions between indigenous rights activists (from both the state and NGOs) and many indigenous parents. For indigenous Peruvians, becoming literate is not only equated with learning how to read; it is equated with learning how to read in Spanish, and disassociated almost entirely from indigenous languages. While the historical and social conditions that have merged to establish Spanish in Peru as the dominant language remain in place, activists are actively trying to change this by arguing that the development of indigenous languages – alongside the acquisition of Spanish – does not have to detract from the socioeconomic advancement and empowerment of indigenous peoples. In fact, not only does learning and developing indigenous languages not detract from the advancement of indigenous peoples in Peruvian society, but it can actually foster their democratic inclusion within the nation. Between 1996 and 1999, I spent approximately two years in Cuzco, Peru trying to understand the ideology behind Intercultural Bilingual Education in the country, and exploring the diverse dimensions of its implementation. I spoke with NGO programme directors, national and regional state representatives, teacher-trainers, highland teachers, indigenous parents, aunts and uncles and indigenous children throughout the department of Cuzco. What I found, as is often the case with bilingual education, was a powerful rejection, by indigenous parents and some highland teachers, of intercultural education in practice. The association between indigenous language, indigenous identity and socioeconomic status was very strong for parents who preferred concrete results (seeing their children speaking and reading in Spanish) to abstract promises about social and economic rights. I should also note that most of the indigenous parents with whom I spoke were sympathetic to the ideals behind education reform, and had in some cases been open to trying out ‘the new education’. However, after several years, problems over lack of resources and confusion over teaching methods (among other issues), parents demanded results, and yet were told to be patient. A mother summed up the feelings of many parents about this at a communal gathering: ‘I don’t have time to be patient. And my children don’t have time to waste on education that teaches them nothing but what they already learn at home talking to us’.19 She was referring to the main complaint of parents who say that bilingual education is another way to keep indigenous peoples from learning Spanish, and thus from being able to position themselves – on their own terms – within Peruvian society. Drawing from a reality that usually only allows indigenous children an elementary education (and only four or five years at that), indigenous parents feel that spending two years learning in the child’s native tongue – as bilingual education methodology dictates – before tackling Spanish, is just too long. Other concerns revolved around contradictions inherent in non-indigenous activists and teachers promoting indigenous language learning, while sending their own children to foreign language institutes, or to non-bilingual schools in the nearest city. Moreover, group discussions about interculturality,

Rethinking Bilingual Education in Peru

27

respect for cultural difference, and democratic (‘horizontal’) behaviour, rarely translated into changes to educational spaces and interactions. Teachers complain that teacher-training sessions are not long enough for them to grasp the nuances of new teaching methods, and that while they are expected to implement these novel ideas in creative ways, they are not provided with the resources necessary to put them into practice. Elsewhere, I have examined the contradictions of intercultural discourse, and the unintended consequences of intercultural activism on highland indigenous communities (Garcı´a, 2003). In those discussions I have focused on the strategies used by indigenous parents and leaders to bypass the limitations that they claim activists are placing on their children’s education. The challenges posed by indigenous peoples to Intercultural Bilingual Education and to intercultural activists are in effect a form of indigenous self-determination. Moreover, these challenges have led to consequential responses from the part of activists (such as promoting the training of indigenous Peruvians in transnational institutes for indigenous leaders), and to some modifications of state education policy (like the shift in emphasis toward second-language acquisition). In one of her earlier examinations of a bilingual education project in the Southern Andes, Nancy Hornberger explores what she called ‘bilingual education success, but policy failure’ (Hornberger, 1987: 205). She concludes that the failure of this particular project was especially due to the problems of implementation, of a rift between national language policy and the project’s goals, and to the low status of Quechua in Peruvian society. However, she ends her discussion stating that, ‘In every case, what is needed for successful language maintenance is. . .autonomy of the speech community in deciding about use of languages in their schools and a societal context in which. . .incenincentives exist for [their] use. . .’ (1987: 224). Hornberger’s conclusions in the 1980s are still relevant today. For example, the implementation of bilingual intercultural education in Amazonian regions appears to have received a much more positive response from indigenous communities than education reform in the highlands. This response stems from increased participation of community members in determining how changes in education will be made. Leaders from several Amazonian indigenous organisations, NGO personnel, and state representatives paint a distinctly different picture of the implementation of bilingual education programmes in the lowland regions. Unlike in the highlands, teachers in lowland community schools are indigenous and have grown up in the community. Moreover, community members, especially parents, participate in workshops that focus on intercultural discourse and pedagogical innovations. In addition, according to Amazonian indigenous leaders, community members work closely with activists in determining the best way to implement education reform in community schools, and indigenous parents in the Amazon have more of a say in what is being taught, and in how the teaching takes place (Perez, 2002). A thorough examination of the differences between highland and lowland experiences with Intercultural Bilingual Education would be an important place to begin to document, analyse, and confront both the setbacks and

28

Bilingual Education in South America

advances in intercultural education. It seems to me that the principal challenge to Bilingual Intercultural Education is the fact that indigenous parents have an entirely different vision than intercultural activists about the meaning of language rights, indigenous rights and the right to education. ‘Even if in theory, bilingual education is better for our children’, commented a Quechua community leader, ‘in our everyday reality, and for our future, it is disastrous’. If intercultural activists and policy makers are serious in their defence of intercultural education as a means of alleviating social injustice for indigenous peoples, and creating a nation more respectful of difference, indigenous concerns about education should not be dismissed, but rather incorporated into new agendas, both in theory and in practice.

Concluding Thoughts In August 2002, during the Fifth Latin American Conference of Intercultural Bilingual Education (held in Lima), the participants drafted a document titled ‘Multilingual Reality and Intercultural Challenge: Citizenship, Politics, and Education’. The first part of the document is worth citing at length: Democracy and citizenship in Latin America are aspirations for us all: however, they are still more of a promise than a reality, especially for indigenous peoples. Our societies have not stopped discriminating against certain categories of individuals and peoples (personas y pueblos); and they continue to allow intolerance, inequality, and authoritarianism. Broad sectors of the population have no rights; other sectors do not respect the rights of others. Our societies are fragmented and still suffer modes of domination and exclusion. It is because of this that indigenous peoples are more vulnerable to injustice, corruption, confrontation, and poverty. . .. Despite education reform,. . .there still have not been adequate responses. . .that might allow for a more visible role for indigenous leaders in their own development. The education offered to most [indigenous peoples], particularly that offered to indigenous women and girls, is devoid of quality, and of linguistic, cultural, and pedagogical relevance. (Declaraci
on de Lima, 2002: 1.1, 1.5, my translation) For indigenous rights activists, education, and particularly Intercultural Bilingual Education, holds the promise of a more equitable, diverse, and respectful society (Godenzzi, 1996; L
opez, 1996). Its proper implementation is a pledge to eradicating poverty in indigenous communities, while simultaneously promoting indigenous autonomy, cultural pride, and demanding social, cultural, economic and political rights. However, as is evident from the critical tone of the above citation, there is still much work to be done. Along with the striking advances of indigenous movements in many Latin American countries, proponents of Intercultural Bilingual Education can claim many victories. Constitutional reforms that recognise and legitimise linguistic, cultural and ethnic diversity, the creation and maintenance of institutes and programmes for indigenous students, and the development of educational materials in dozens of indigenous languages are only a few examples of the kind of progress made by Bilingual Intercultural Education advocates. Yet the inconsistencies found between rhetoric and implementation still need to

Rethinking Bilingual Education in Peru

29

be addressed. For instance, the DINEBI boasts that every classroom in indigenous schools where bilingual intercultural education is implemented has its own library. But during my visits to dozens of bilingual schools in the highlands, only one had a ‘library’, although teachers were still waiting for the library materials they had been promised. Similarly, while over 10,000 teachers have been ‘trained’ in bilingual education methodology, there is little emphasis on evaluation of the training (or the trainers), on follow-ups with individual teachers, or on feedback from teachers about the positive and negative aspects of training sessions. In other words, there seems to be a more immediate concern with quantity than with the quality of education and training received. Intercultural Bilingual Education is a bold initiative, but as currently implemented it leaves many gaps. First, there is the difference between the rhetoric of inclusion and the under-funded and paternalistic practices that still tend toward exclusion. Second, there is the gulf between the ideological authors of new policies and the teachers and trainers charged with the more difficult task of implementing those ideas. Finally, there is the lamentable distance between those who make policies and those who must live with the effects of these reforms. Similar gaps can surely be found in most democracies but the true test of democracy lies in closing them. If governments, activists, teachers and students succeed in bridging the distance between the promise of bilingual intercultural education and its current state, Peruvian democracy will be greatly in their debt. Correspondence Any correspondence should be directed to Dr Marı´a Elena Garcı´a, Sarah Lawrence College, Department of Anthropology, 1 Mead Way, Bronxville, New York 10708, USA ([email protected]). Notes 1. Her support for indigenous rights, however, is controversial. Some see it as imbued with dangerous paternalism. Her romanticised notions of indigenous identity and politics are increasingly questioned and challenged by both new and old indigenous organisations and by indigenous rights advocates and intellectuals in the country. See Garcı´a and Lucero (2004) for a more in-depth discussion about indigenous politics in Peru and of Karp’s problematic role as indigenous rights´ spokesperson. 2. The evidence in this article draws from 17 months of multi-sited ethnographic research in highland Peru and forms part of a larger project on indigenous activism and multicultural development in the Peruvian highlands (Garcia, forthcoming). Following standard ethnographic practice, sources were assured anonymity and are therefore not named. During this time, I worked closely with NGOs working on indigenous and rural education and with personnel from what was then the National Unit for Bilingual Intercultural Education (UNEBI). I conducted dozens of interviews with rural teachers, indigenous parents, NGO practitioners, state representatives and indigenous rights activists, including Peruvian linguists, anthropologists and intellectuals. I also spent days at a time with teachers, observed teacher-training workshops and was present during parent–school sessions. Additionally, for several weeks I lived in a monoligual Quechua indigenous community in one of the most isolated regions of the department of Cuzco. During this time, (and during the many times that I returned to this community), I spent

30

3. 4. 5.

6.

7.

8. 9. 10.

11. 12.

Bilingual Education in South America countless hours talking with parents about education policies, as well as with children about their school, language and education. Unless otherwise stated, all translations are my own. Vicu~ nas are animals from the same camelid family as llamas and alpacas. The name ‘Tawantinsuyu’ reflects the union of the four suyus or corners of the Inca Empire. During what has been labelled the Inca-expansion period (1430–1532), a period of intensive military expansion, the Incas executed a policy of language spread that massively diffused Quechua throughout their empire. The Incas subjugated over 40 major linguistic groups in about one century, and the empire was estimated at 6 million people (Rowe, 1946). The Incas’ policy of language spread was executed in order to maintain control over conquered territories, although the implementation of this linguistic policy has been described as ‘tolerant and by no means assimilationist’ (von Gleich, 1994: 83, see also Cerr
on-Palomino, 1989). Long before the Inca-expansion period, a few other languages (namely Aru, Aymara, Puquina, and Mochica), along with Quechua, were also spread as general languages in an effort to minimise the communication problems caused by the multitude of local languages and dialects that existed in the Andes at that time. However, by 1430, Quechua had already expanded at the expense of these other languages. While Spanish, as the language of conquest, became official with the siege of Cuzco in 1533, Quechua was still actively promoted for over 200 years. It was only after the indigenous rebellions of the late 1700s (and especially the Tu´pac Amaru Rebellion of 1780) that the Quechua language and other indigenous ‘markers’ such as religion were banned. According to the 1981 census, 22% of Peru’s total population over the age of five (14,570,881) speak Quechua (Instituto Nacional de Estadı´stica, 1984: 164). Of these, 35% (1,113,410) are monolingual (INE, 1984). In Cuzco, according to the 1993 census, 85.8% of the department’s rural population (460,000) speak Quechua as their first language, though in certain provinces of Cuzco, more than 95% of the population speak Quechua (INEI, 1994: 74–76). I should also note that 60.2% of Cuzco’s urban population (414,000) speak Spanish as their first language (INEI, 1994). Creoles or ‘criollos’ were descendants of Europeans who were born in the Americas. The Bilingual and Intercultural Programme of the Alto Napo is the exception. First established in 1975, it continues with some success in the present. Some of the programmes most cited by researchers of bilingual education and linguistics are: (1) the Bilingual Literacy Programme in Quinua, Ayacucho (1964–1970), developed with collaboration from the SIL and the Institutes for Applied Linguistics Research of San Marcos University and the University of Hua~ iga, 1985); (2) the Programme for Basic manga (see Burns, 1968, 1971; and Zu´n Bilingual Education in Cuzco in 1974 with collaboration from USAID, Cornell University and the Peruvian National Institute for Educational Research and Development (INIDE) (see de Vries, 1990); (3) the Experimental Project of Bilingual Education in Puno (1978–1988) (PEEB-P) (see below); and (4) the Bilingual and Intercultural Programme of the Alto Napo (1975–present), funded primarily by the Peruvian government in its beginning stages, and receiving aid from the GTZ by the 1990s (see San Rom
an, 1984; and Fern
andez, 1985). While the Peruvian government permitted the development and implementation of all of these programmes, and allowed the collaboration of regional universities and national education agencies with the groups interested in promoting bilingual education, it provided little, if any, funding toward their maintenance. For more on the PEEB-P, see Hornberger, 1988, 1989, 1990; Jung & L
opez, 1988; L
opez, 1988; and Jung (no date). Among these, Velasco’s radical agrarian reform – most often cited as a turning point in Peruvian history – was officially initiated throughout the highlands on 24 June, 1969, the national ‘day of the Indian’. On this day, Velasco announced the massive and forceful handover of large estates by landowners to their former serfs and employees. Moreover, he prohibited the use of the term indio, replacing it

Rethinking Bilingual Education in Peru

13.

14. 15.

16.

17.

18.

19.

31

with the label of ‘peasant’ (campesino). While this move created the category of highland ‘peasant communities’ it was not until 1974 that Velasco legally recognised and organised Amazonian peoples as ‘native communities’, further reinscribing the distinction between highland and lowland peoples. In an effort to integrate highland peasants into the national economy, Velasco’s plan included the redistribution of all haciendas by 1975. Although the idea was simple – distribute land among those who work it – the reform’s implementation was largely unsuccessful. While it effectively ended landowners’ control over peasant workers, it also greatly exacerbated tensions between ethnic groups, particularly between highland peasants and lowland ‘natives’ on the one hand, and the criollo middle- and upper-classes on the other. In Peru, as in other places in Latin America, regional identities can be highly racialised. While the coast is perceived to be white and European (even despite massive migration from the highlands into Lima), the Andes and Amazon are seen as Indian places. See Stern, 1998, for more on Sendero Luminoso and Peru’s ‘dirty war’. Because Sendero originated in the highlands, indigenous peoples were often presumed by government forces to be subversives or Sendero sympathisers. Thousands of indigenous peasants were tortured, disappeared, or executed, sometimes only because they could not speak Spanish (Poole & Re´nique, 1992). However, by the mid- to late-1980s, indigenous peoples had organised into civil defence patrol units, sometimes creating alliances with the military. Many argue that it was indigenous resistance to Sendero that led to the movement’s demise (Degregori, 1996; Starn 1998). The DINEBI was baptised as such on 5 April, 2001, under Valentı´n Paniagua’s transitional administration. Paniagua, President of the Congress when political scandals led then President Fujimori to flee the country, was instituted as President of a transitional government in Peru while new elections were held. The distinction between state policy (polı´tica de estado) and government policy (polı´tica de gobierno) is important, as national policies remain fixed regardless of future changes in government personnel and ideology, while governmental policies come and go with specific political parties and their representatives. Most likely, what the Peruvian state found dangerous about the Ecuadorian proposal was its association with indigenous political autonomy. By the time the UNEBI was just beginning to operate in Peru (1996), the Ecuadorian indigenous movement had already acquired tremendous political leverage. A recent manifestation of the national political visibility Ecuadorian indigenous leaders have achieved, came in January 2000 when they allied with military forces to oust the Ecuadorian president, Jamil Mahuad. While the Peruvian state was clearly not pleased with what was happening to their northern neighbour, activists were ecstatic, and cited this event as one demonstrating the potential benefits of bilingual intercultural education. The link between bilingual education and indigenous politics was unsettling to government agents. One source of tension between proponents of education reform and indigenous parents is the difference in ideas about language maintenance. While a strong motivation for bilingual intercultural education advocates is indigenous language revival and maintenance, indigenous peoples do not believe that learning Spanish will in any way detract from their ability – or their children’s – to speak, understand, and (if literate) read in their native language.

References Anderson, B. (1991) Imagined Communities. New York: Verso. Belau´nde, V. (1965) [1942] Peruanidad. Lima: Ed. Librerı´a Studium. Burns, D. (1968) Bilingual education in the Andes of Peru. In J. Fishman, C. Ferguson and J. Das Gupta (eds) Language Problems of Developing Nations (pp. 403–414). New York: John Wiley and Sons.

32

Bilingual Education in South America

Burns, D. (1971) Cinco An~os de Educaci
on Bilingu¨e en los Andes del Peru: 1965–1970. Informe final. Lima: Instituto Lingu¨´ıstico de Verano. Cerr
on-Palomino, R. (1989) Language policy in Peru: A historical overview. International Journal of the Sociology of Language 77, 11–33. Contreras, C. (1996) Sobre los Orı´genes de la Explosi
on Demogr
afica en el Peru´. Documento de trabajo no. 61. Lima: IEP. De la Cadena, M. (2000) Indigenous Mestizos: The Politics of Race and Culture in Cuzco, 1919–1991. Durham: Duke University Press. Declaraci
on de Lima (2002) Servindi: Servicio de Informaci
on Indı´gena, No 13. Degregori, C.I. (ed.) (1996) Las Rondas Campesinas y la Derrota de Sendero Luminoso. Lima: IEP. de Vries, L. (1990) Polı´tica lingu¨´ıstica y educaci
on bilingu¨e en los Paı´ses Andinos relativas a los Quechua-Hablantes. In L.E. L
opez and R. Moya (eds) Pueblos Indios, Estados y Educaci
on (pp. 27–44). Lima: GTZ. Escobar, A. (1984) Arguedas o la Utopı´a de la Lengua. Lima: IEP. Escobar, A., Matos Mar, J. and Alberti, G. (1975) Peru´ Paı´s Bilingu¨e? Lima: Instituto de Estudios Peruanos. Fern
andez, M. (1985) Programa de Educaci
on Bilingu¨e e Intercultural del Alto Napo-Peru. Seminario Subregional, 13–19 noviembre, Lima. Garcı´a, J.U. (1973) [1930]. El Nuevo Indio. Lima: Universo. Garcı´a, M.E. (2003) The politics of community: Education, indigenous rights, and ethnic mobilization in Peru. Latin American Perspectives 30 (1), 70–95. Garcı´a, M.E. (forthcoming) Making Indigenous Citizens: Identities, Education, and Multicultural Development in Peru. Stanford: Stanford University Press. Garcı´a, M.E. and Lucero, J.A. (2004) ‘Un Paı´s Sin Indı´genas’?: Rethinking indigenous politics in Peru. In N. Postero and L. Zamox (eds) The Struggle for Indian Rights in Latin America. Brighton: Sussex Academic Press. Godenzzi, J.C. (1996) Introducci
on=Construyendo la convivencia y el entendimiento: Educaci
on e interculturalidad en Ame´rica Latina. In J.C. Godenzi (ed.) Educaci
on e Interculturalidad en los Andes y la Amazonı´a (pp. 23–82). Cuzco: Centro de Estudios Regionales Andinos Bartolome´ de las Casas. Godenzzi, J.C. (2001) Polı´tica de lenguas y culturas en la educaci
on: El caso del Peru. Paper presented at the 23rd Meeting of the Latin American Studies Association. Washington, DC. Heath, S.B. and Laprade R. (1982) Castilian colonization and indigenous languages: The cases of Quechua and Aymara. In R. Cooper (ed.) Language Spread (pp. 118–147). Bloomington: Indiana University Press. Hornberger, N. (1987) Bilingual education success, policy failure. Language in Society 16, 205–226. Hornberger, N. (1988) Bilingual Education and Language Maintenance: A Southern Peruvian Quechua Case. Holland: Foris. Hornberger, N. (1989) Haku Yachaywasiman: La Educaci
on Bilingu¨e y el Futuro del Quechua en Puno. Lima: Programa de Educaci
on Bilingu¨e de Puno. Hornberger, N. (1990) E´xitos y desfases en la educaci
on bilingu¨e en Puno y la polı´tica lingu¨´ıstica Peruana. In L.E. L
opez and R. Moya (eds) Pueblos Indios, Estados y Educaci
on (pp. 379–410). Lima: GTZ. INEI (Instituto Nacional de Estadı´stica e Inform
atica) (1994) Departamento de Cusco. Perfil sociodemogr
afico. Censos Nacionales 1993, IX de poblaci
on y IV de vivienda. Colecci
on An
alisis Censal No. 8. Lima. Instituto Nacional de Estadı´stica (1984) Censos nacionales: VIII de poblaci
on, III de vivienda, 12 de julio de 1981: Resultados definitivos, de las variables investigadas por muestreo, Volume B, Book 1. Lima, Peru: Instituto Nacional de Estadı´stica. Jung, I. (n.d.) Conflicto Cultural y Educaci
on: El Proyecto de Educaci
on Bilingu¨e – Puno, Peru. Quito: Abya Yala. Jung, I. and L
opez, L.E. (1988) Las Lenguas en la Educaci
on Bilingu¨e: El Caso de Puno. Lima: Sociedad Alemana de Cooperaci
on Te´cnica.

Rethinking Bilingual Education in Peru

33

Larson, B. (1995) Andean communities, political cultures, and markets: The changing contours of a field. In B. Larson and O. Harris (eds) Ethnicity, Markets, and Migration in the Andes: At the Crossroads of History and Anthropology (pp. 5–54). Durham: Duke University Press. L
opez, L.E. (1988) Balance y perspectivas de la educaci
on bilingu¨e en Puno. In L.E. L
opez (ed.) Pesquisas en Lingu¨ı´stica Andina (pp. 79–108). Lima: Universidad Nacional del Altiplano. L
opez, L.E. (1996) No m
as danzas de ratones grises: Sobre interculturalidad, democracia y educaci
on. In J.C. Godenzzi (ed.) Educaci
on e Interculturalidad en los Andes y la Amazonı´a (pp. 23–82). Cuzco: Centro de Estudios Regionales Andinos Bartolome´ de las Casas. Mallon, F. (1998) Chronicle of a path foretold? In S. Stern (ed.) Shining and Other Paths (pp. 84–117). Durham: Duke University Press. Mannheim, B. (1984) Una Naci
on Acorralada. Southern Peruvian Quechua language planning and politics in historical perspective. Language in Society 13, 291–309. Mannheim, B. (1991) The Language of the Inka Since the European Invasion. Austin: University of Texas Press. Mar-Molinero, C. (1995) Language policies in multi-ethnic latin America and the role of education and literacy programmes in the construction of national identity. International Journal of Educational Development 15 (3), 209–219. O’phelan Godoy, S. (1985) Rebellions and Revolts in Eighteenth Century Peru and Upper Peru. Cologne. Pe´rez, W. (2002) Interview. Poole, D. and Re´nique, G. (1992) Peru, Time of Fear. London: Latin American Bureau. Pozzi-Escot, I. (1981) La educaci
on bilingu¨e en el marco legal de la reforma educative Peruana. In Acerca de la Historia y el Universo Aymara (pp. 113–123). Lima: CIED. Rojas, I. (1982) En torno a la oficializaci
on de las lenguas Quechua y Aymara. In R. Cerr
on-Palomino (ed.) Aula Quechua. Lima: Signo Universitario. Rowe, J. (1946) Inca culture at the time of the Spanish conquest. In J. Steward (ed.) Handbook of South American Indians (Vol. 2). Washington: Bureau of American Ethnology. San Rom
an, G. (1984) Programa de educaci
on bilingu¨e e intercultural del Alto Napo. Shupihui, 30. Starn, O. (1998) Villagers at arms: War and counterrevolution in the Central-South Andes. In S. Stern (ed.) Shining and Other Paths: War and Society in Peru, 1980–1995 (pp. 224–257). Durham: Duke University Press. Stern, S. (ed.) (1998) Shining and Other Paths: War and Society in Peru, 1980–1995. Durham: Duke University Press. Stepan, N.L. (1991) The Hour of Eugenics: Race, Gender and Nation in Latin America. Ithaca: Cornell University Press. Torero, A. (1983) La familia lingu¨´ıstica Quechua. In B. Pottier (ed.) Ame´rica Latina en sus Lenguas Indı´genas (pp. 61–92). Caracas: Monte Avila. Turino, T. (1991) The state and Andean musical production in Peru. In G. Urban and J. Sherzer (eds) Nation-States and Indians in Latin America (pp. 259–285). Austin: University of Texas Press. UNEBI (1998) Boletı´n Informativo, 3. Lima. Valc
arcel, L. (1981) Memorias. Lima: IEP. Valderrama, R. and Escalante C. (1977) Autobiografı´a de Gregorio Condorı´ Mamani. Cusco: Centro Bartolome´ de las Casas. Vasconcelos, J. (1925) La Raza C
osmica: Misi
on de la Raza Ibero-Americana. Notas de Viaje por la Ame´rica del Sur. Barcelona: Agencia Mundial de Librerı´a. Reprinted in Obras Completas, No. 2: 906–942. Me´xico: Librerors Mexicanos Unidos. Velasco Alvarado, J. (1972) La Voz de la Revoluci
on: Discursos del Presidente de la Repu´blica, General de Divisi
on, Juan Velasco Alvarado, 1970–1972, Tomo II. Lima: Ediciones Participaci
on.

34

Bilingual Education in South America

von Gleich, U. (1994) Language spread policy: The case of Quechua in the Andean Republics of Bolivia, Ecuador, and Peru. International Journal of the Sociology of Language 107, 77–113. Zimmermann Zavala, A. (1974) El Plan Inca, Objetivo: Revoluci
on Peruana. Lima: Empresa Editora del Diario Oficial ‘El Peruano’. ~ iga, M. (1985) La Educaci
Zu´n on Bilingu¨e Quechua-Castellano en Ayacucho: un Programa Experimental de la UNMSM, Documento de Trabajo No. 52, Lima.

Bilingual Deaf Education in the South of Brazil Carlos Skliar Universidade Federal do Rio Grande do Sul, Brazil

Ronice Muller Quadros Universidade Federal de Santa Caterina, Brazil This paper presents an analysis of Bilingual Deaf Education in Brazil, with particular reference to the South of the country. This subject is presented in context, and takes into account research carried out over the last 5–7 years. We consider the spread of bilingual=bicultural models, changes in the representation of Deaf people and Deafness and finally, the epistemological inversion of Deaf and Hearing ‘problems’ as present in the discussion related to Deaf Education in Brazil. The analyses of the experience in Brazil are not simple. In fact, the complexity is related to the different possible readings that ‘Bilingual Deaf Education’ can have, such as for instance, methodological, linguistic and psycholinguistic interpretations, all of which are considered in the present paper. In addition, we describe certain bilingual experiences that we have been engaged in, together with other Deaf researchers and Deaf teachers over the last decade. Keywords: bilingualism, deafness, Deaf Education, Brazil

Introduction Over the last 30 years, many changes have taken place in Deaf Education in Brazil, relating to issues such as ideological conceptions, educational organisation, Deaf empowerment, Deaf people in Deaf Education teams, the status of Brazilian Sign Language, and Portuguese as a second language, among others. Skliar (2001) selects three specific contributions of these changes that he maintains need to be considered in detail: (1) (2) (3)

the spread of a bilingual=bicultural model applied to Deaf Education; changes in the representations of Deafness and Deaf people; the epistemological inversion of what the Deaf ‘problem’ is in relation to hearing ‘problems’ in Deaf Education.

These three contributions can be seen as part of the bilingual movement in Brazil, considering that bilingualism was and is related to policies, in the sense of action, selected to guide decisions about Deaf Education. Thus, this is far more than just a type of educational proposal for Deaf people.

Bilingualism in Deaf Education in Brazil Deaf Bilingual Education can be defined as a complex phenomenon, since it reflects policies, power and knowledge (in Foucault’s (1980) sense). It is not just the presence of two languages, in our case, Brazilian Sign Language and

35

36

Bilingual Education in South America

Brazilian Portuguese, but involves ambiguity and relativity of truth. There is ambiguity, since the concept of Deaf Bilingual Education can be defined in different ways, often revealing antagonisms; and there is relativity of truth (including the minimal definition – two languages), because Deaf Bilingual Education presupposes a negation of more classical conceptions in the history of Deaf Education. In Brazil three different readings of the bilingual=bicultural model as applied to Deaf Education may be postulated: first, a methodological reading; second, a linguistic reading; and third, a psycholinguistic reading. The first reading involves applying the model as an academic system that has come to replace Total Communication as opposed to Oral Education, without reviewing the rules and the curriculum, or the Deaf person in the educational process. On the other hand, the linguistic and psycholinguistic views of Deaf Bilingual Education are concerned with the acquisition of language and the lexical, semantic and syntactic relations of the different modalities of the languages involved. This much-needed and meaningful discussion is crucial for Deaf Education, although it can also blur other important points such as, academic programmes, literacy, the relation between education=work, Deaf empowerment, the education of new professionals, mechanisms of exclusion=inclusion, and power relations. It is interesting to note that all these readings have kept, and still keep, a focus on the Deaf child and do not consider teenagers and adults. In fact, bilingual education should only be the starting point of Deaf education, since it is the beginning of policies about Deaf identity, Deaf and hearing knowledge and power, Deaf resistance movements, ideologies, hegemonic discourses, school roles and public policies. We cannot ignore all the aspects involved in Deaf Bilingual Education. This is well-known with respect to spoken languages in various countries in which there are bilingual=multilingual contexts (Collier & Thomas, 1989; Cummins, 1992, 1996; Cummins & Swain, 1986; Grosjean, 1982; Veltman, 1988). The implicit aims of Bilingual Education in the world have to be considered. Aspects such as assimilating individuals or groups into mainstream society, aiming at socialising people into full participation in the community, and unifying a multilingual community, endeavouring to bring unity to a multi-ethnic, or multinational linguistically diverse policy, are examples of what can be considered as the basis of bilingualism. In this sense, the political has a double value: first, as a historical, cultural and social construction of perspectives on Deafness which are based on discourses, and as power and knowledge relations in the process. Then, we can find Deaf policies with language, identity and body pressures (Davis, 1997). This can be understood as ‘hearing’ practices (Skliar, 1997, 1998) in which Deaf people are talked to by hearing people, being forced to be talked to and seen as hearing beings; in other words, being colonised (Quadros & Perlin, 2003). Therefore, a bilingual focus should encompass more than merely educational proposals. There needs to be an investigation into the power mechanisms of the relations inside and outside the Deaf school and Deaf education in general. In this way, bilingual education is a proposal that is related to human

Bilingual Deaf Education in the South of Brazil

37

rights. Sign language, in this view, is the language of Deaf people and the language of the school. It is the language in which the Deaf child will formulate hypotheses about the world, criticise, talk about emotions and discuss issues, as his or her right. The next point considered by Skliar (2001), refers to the changes in the representations of Deafness and Deaf people observed over the last 30 years in Brazil. These ideas are opposed to the oralist conceptions in vogue at the beginning of the ’80s, but this is not a simple opposition. The conceptualisation of certain bilingual proposals in Brazil still reflect intrinsic oralist policies. This is the present picture in Brazil. The discourses and practices are like networks with asymmetric power and knowledge relations about Deaf people and Deafness. The meanings and symbolic systems produce representations about Deafness and about Deaf bilingual education that are based on traditional conceptions, which use sign language as a tool for the dissemination of the official culture and language. Skliar (2001), therefore, proposes to define Deafness according to the following four dimensions: as political difference; as visual experience; as multiple identities; and as located in the discourse of ‘handicap’. In the first dimension, Deafness as political difference, it is very common to find different ways to identify the Deaf by using euphemisms such as ‘special needs’, ‘people with special needs’, ‘diversity’, ‘difference’. These terms seem to be examples of meanings that share various similarities. Deaf people are defined by meanings based on normality being invented, reinvented and produced by ‘hearing’ people. This happens because the norm is implicit and invisible; and, as a consequence of this invisibility, it is considered non-existent. We have found a preference for the term ‘diversity’ in Brazilian official documents. This reflects a traditional strategy that blurs the meaning of cultural differences. The ambiguity of the term ‘diversity’ leads, at best, to the acceptance of some degree of pluralism related to the ideal norm. However, all these terms are only ways to minimise conceptions that are still reproducing old frontiers of exclusion. We define Deafness as political difference, not only to replace all the adjectives used up to now, but as a clear option based on several analyses (for instance, those carried out by Bhabha, 1994 and McLaren, 1995). In this sense, difference is politically, historically and socially constructed; differences are always differences, and differences exist even if there is no authorisation, acceptance, respect, or permission from normality. With respect to the second dimension, understanding Deafness as difference, this implies recognising visual experiences that involve much more than cognitive and linguistic abilities. This leads to all kinds of cultural and community manifestations, including, of course, sign language. For instance, we use the term ‘The Deaf Way’ to refer to other people with name signs based on visual features; the use of visual metaphors about aesthetic information; the production of visual humour; the resistance movement with visual manifestation; the expression of the meaning of time in a visual way in space, visual mechanisms and didactics, and visual literature. The Deaf Way of understanding and producing knowledge is not the main point of educational discussions. However, it should occupy a space in the proposals and design of

38

Bilingual Education in South America

Deaf education, since it plays a crucial role in communication, didactics, curriculum and intellectual processes. The third dimension, defining Deafness as multiple Deaf identities, forces us to consider the ‘Deaf person’ not as a person with a unique and complete identity. Over the last few years, we have detected many deep transformations in Deaf identity, reflecting policies, cultural and social movements, and changes in the economy. All these seem to interfere clearly or implicitly in stereotypes of Deafness, especially in Deaf identities that change from time to time, according to different cultures, geographic space, historical times, change in agreement with different people etc. (Moura, 1999; Perlin, 1998). Thus, the Deaf community should be understood not only in its plural forms, but also in its fragments of identity. In this sense, we are moving towards understanding Deafness from a political perspective, that is, focusing on established power relations. The ‘movement of identity’ takes place in the meeting of Deaf–Deaf, that is, when Deaf people meet other Deaf people, and where space is organised in a different way. Also, this movement happens in discursive environments. Then, Deaf people will start to talk by themselves, to judge, to remember, to recognise and to perceive in different ways and, in so doing, establish new identities. As mentioned by Wrigley (1996), Deaf people invent Deafness by themselves. The last way to define Deafness is as located in the discourse of ‘handicap’. When talking about handicaps, people locate the discussion in a Special Education context that is considered to be a part (or sub area) of Education. Deafness is analysed as a small part of the huge problem that surrounds Education in general. This small part is related to taking care of the people who are not as able as ‘normal people’ are, since these people are handicapped with unfortunate families and sacrificial professionals – the specialists. Special Education manifests discontinuity in its theoretical discourses and is related much more to charity, aid and medical practices, and, in doing so, reproduces exclusion based on the binary relation of inclusion=exclusion. Over the last 10 years, the dominant ideology of Deaf Education seems to be dissatisfied with this view, in the light of the Deaf Way. The handicap definition is not the way to think about Deaf people; instead the epistemological way is considered the best to capture Deaf identities. Even when we are completely against the idea of Special Education, we have to consider that this definition still has a very strong influence on us. Therefore, the impossibility of passing planning proposals, which see the Deaf person as handicapped, and not as a person with visual experiences, as defined earlier, needs to be discussed. This means that if educational professionals continue with their discourses, trying to maintain the notion of Special Education, it is impossible to think of a Deaf Way. After defining Deafness, the last aspect mentioned by Skliar (2001), is epistemological inversion: the Deaf ‘problem’ is, in fact, hearing ‘problems’ in Deaf Education. This involves consideration of the problems that the hearing people suffer in their social, communicative and linguistic interactions with Deaf people. These are problems derived from the invention of the Deaf alter. For instance, from the cultural point of view, there is the problem related to the perception of who Deaf people are. Hearing

Bilingual Deaf Education in the South of Brazil

39

specialists or teachers have problems when planning their classes to understand ‘the other’, – the Deaf person. In other words, instead of continuing to try to understand hearing impairment, they need to understand the (political) meaning of hearing normality. Instead of thinking that sign language is a problem, they need to analyse the hearing discourse that represents this language as a problem. Instead of thinking that Deaf people are handicapped, they need to understand that the Deaf live a visual experience in the world. Following this order of ideas, we can also invert the problem of Bilingual Education. Before trying to decide whether Deaf people are bilingual, whether sign language is their first language, whether there is something called ‘Deaf culture’, whether there should be divisions between children and Deaf adults, whether the school should be special or regular, whether Deaf people should or should not be teachers, whether they are different from hearing people, etc, we should ask: what kind of problems do we, as hearing people, have when we think of Bilingual Education? Which mechanisms have we, as hearing people, invented to understand Deaf bilingualism? What are our representations of Deafness and Deaf people within and beyond educational practices and discourses? And, finally, what are the power and knowledge relations that we maintain or negate in our relations as hearing people with Deaf people? The bilingual experience in the South of Brazil: Educational work from the viewpoint of the Deaf adult, and the establishment of a network of Deaf Education We have been thinking and writing since 1996 about the existence of certain variables in Deaf Bilingual Education in the South of Brazil, incorporating Deaf views to make academic projects more complex and significant (Perlin, 1998, 2000; Quadros, 1997a, 1997b, 2000; Skliar, 1997, 1998, 2000, 2001). These variables follow from the pedagogic pathways in which Deaf people are inserted, established both from a historical and forward-looking bilingual proposal. They are the following: (1)

(2) (3) (4) (5) (6)

the reconstruction of the problems that govern bilingual education for Deaf people by inverting the ‘hearing’ logic of Deaf education in which the problem would be exactly the Deaf themselves, and heading towards another multi-dimensional analysis of the educational process; the circulation of the meanings and representations around Deafness and Deaf people in specific educational contexts; the participation of Deaf adults in the planning, development and evaluation of bilingual policies; the continuity of the educational project; the revision of school architectures and ideologies; and, finally, the structure and sequence of pedagogical goals.

This set of variables has opened new doors for Deaf Bilingual Education in the South of Brazil. In the first place, the transformation of the view that

40

Bilingual Education in South America

a bilingual education is not only an academic project, but is intimately associated with the politics of linguistics, identity, culture and differences. In the second place, and perhaps one of the most important aspects, bilingual education supposes two pathways for Deaf people. One is a pedagogical pathway with which the Deaf child is engaged until he=she becomes an adult. The second is a pathway that goes from the Deaf adult to the Deaf child. What does it mean to go from the Deaf adult to the Deaf child? It means that we cannot ignore the systematic qualification of Deaf adults as educational professionals and the many generations of Deaf adults who were not able to radically change their historical condition of illiteracy, low academic results and low quality of life: no jobs, sub-jobs, etc. Also, it means that the school should be preparing for the time in which the Deaf child and Deaf adult can meet each other, and that the school should be considering the qualifications of Deaf people to play a role in Deaf Education. Since 1996, in the South of Brazil, our focus has been specifically on this issue. Nowadays, community, undergraduate and graduate courses and projects are being organised and created to prepare Deaf people to be teachers of Brazilian Sign Language, and specific areas, such as literacy of Deaf teenagers and Deaf adults, arts and cultural manifestations. The main aim of all these initiatives is no other than to significantly transform the situation of the present generations of Deaf people in Brazil, in order to lead to autonomy of the Deaf in the political and education arenas; in other words, in order to facilitate Deaf empowerment. Almost all the Deaf people engaged in the process described here, are from Rio Grande do Sul and Santa Catarina (the two states in the South of Brazil), but some come from the countryside and other large cities. To offer a picture of Brazilian reality in terms of Deaf Education, we would describe the situation in the following terms. Brazil has a political policy of public (state) education in mainstream schools. The control of this position cuts across the regional administration that puts forward different positions across Brazil. This means that there are different understandings of what the ‘inclusion’ of Deaf people in mainstream schools really means. In Rio Grande do Sul, there are about ten Deaf Schools that make the difference in terms of Deaf Education. These schools have had systematic discussions about what Deaf people want in terms of education. Some of them have had support from the universities in the understanding of how to implement Deaf Education in a ‘Deaf way’. In Santa Catarina, there is no Deaf school in the entire state. However, there are a group of 32 Deaf people studying Deaf Education on an undergraduate course in a public university – Universidade Estadual de Santa Catarina, UDESC – in order to be able to teach Deaf children. The Deaf students are aware of the state context and they have been discussing the future of Deaf Education in their state and in Brazil as a whole. These future Deaf teachers, together with teachers or future teachers from Rio Grande do Sul, have been meeting to analyse the situation and to make proposals about Deaf Education in systematic forums. There is also a plan developed by these Deaf groups which presents the priorities of Deaf Education

Bilingual Deaf Education in the South of Brazil

41

(Documento do l Semin
ario Nacional de Surdos, 2001) that is partially translated as follows:

Ist National Deaf Seminar – Caxias do Sul=Rio Grande Do Sul – 27–29 September, 2001 – The Deaf View on Educational Practices Group 3 – Deaf education Infant Education (1) How to prepare the Deaf child in the first years of schooling. (2) What is the curriculum of Deaf children in their first years at school? (3) To offer access to Brazilian Sign Language to Deaf children of hearing families. Primary Education (1) (2) (3)

What are the best practices of primary education? To develop curriculum proposals for Deaf Education in Sign Language. To find out about literacy processes, the reading and writing of Portuguese for Deaf people.

Secondary Education (1)

To find out about Education and Work.

Adult Education (1)

To make Deaf Adults literate.

Undergraduate Education (1)

To identify didactic strategies and specific dynamics of classes for the Deaf.

Group 4 – Deaf education boundaries (1) To find out about the laws and their implications in Deaf Education. (2) To investigate education and the politics of social inclusion and exclusion. (3) To have a critical view on the relation between Deaf Education and Special Education. (4) To find out about political projects and institutional projects related to Deafness. (5) To consider conceptual models of Deafness and Deaf people: the clinical, the anthropological and the difference model. (6) To distinguish the discourse and the practice of ‘hearing impairment’ in Education. (7) To find out how ‘disability discourse’ functions in Education. (8) To discover traditional curriculums in Deaf Education: practices and discourses. (9) To provide an introduction to the Critical Theory of the Curriculum. (10) To study curriculum, ideology, language, power, culture and politics. (11) To identify strategies of the Pedagogy of Difference.

42

Bilingual Education in South America

These priorities reflect resistance movements from Deaf groups in the country. They have also discussed sign language acquisition, sign language teaching, technology, signwriting, family intervention and mental and physical health. In Deaf Rights forums in Santa Catarina, Deaf people have proposed certain guidelines for Deaf Education that are partially translated as follows (Diretrizes e Ac¸o˜es para a Educac¸a˜o de Surdos em Santa Catarina, 2000). Reconstruction of the politics of Deaf Education in Santa Catarina (1)

(2)

The creation of a Reference Centre for Deaf Education in teaching, research and extension to the community: recognition of Visual and Spatial Deaf Experience; production of research about cultures, identities, differences, visual methodologies, linguistics, education; to give support for instruction of the Deaf and hearing professionals; to give support for sign language interpreters; to teach courses open to the community in general about Deaf Education, Deaf Culture and Identities, Sign Language. Starting groups of Deaf students in central regular schools at all educational levels with Deaf teachers, bilingual teachers and sign language interpreters, with a visual-based structure and methodology (Pedagogy of Difference).

There is also a National Deaf Association (Federac¸a˜o Nacional de Educac¸a˜o e Integrac¸a˜o de Surdos – FENEIS) that has been fighting for more than 10 years for the recognition of sign language. Last year, a federal law (Lei 10.436, 24=04=2002) recognised Brazilian Sign Language as an official language in the country. This has been reflected directly in Deaf Education. Since then, it seems more appropriate to consider Deaf Education within the context of groups that have different languages from Brazilian Portuguese in Education in general, instead of relating it to the context of Special Education. To offer a quantitative dimension of the research projects that have been carried out recently across the south of the country, we would say that currently there are about 100 Deaf students studying on different undergraduate courses, four Deaf students working on their Masters theses and three working on their doctoral dissertations in Education, about 95 Deaf teenagers and Deaf adults involved in popular education programmes, in the process of becoming literate, and 40 Deaf teachers of Brazilian Sign Language, among others. Also, there are specific projects being planned to guarantee Deaf people becoming qualified as professionals, particularly in areas of Education. These include an undergraduate major in Brazilian Sign Language and Brazilian Portuguese, a graduate major in Deaf Education and another in the translation and interpretion of Brazilian Sign Language and Brazilian Portuguese. These fields of study are being organised, thanks to the work of several organisations such as: the Federac¸a˜o Nacional de Educac¸a˜o e Integrac¸a˜o de Surdos – FENEIS, the Universidade Federal de Santa Catarina – UFSC, the Universidade Federal do Rio Grande do Sul – UFRGS, and the Centro Federal de Formac¸a˜o Tecnol
ogica – CEFET=SC. Moreover, as mentioned above, an undergraduate major in Deaf Education is already being offered, in the Universidade Estadual de Santa Catarina – UDESC –with 32 Deaf students, and the training of qualified sign language

Bilingual Deaf Education in the South of Brazil

43

interpreters and Deaf sign language teachers financed by the Ministe´rio de Educac¸a˜o e Cultura (the highest federal educational organisation in Brazil) together with the organisations previously mentioned. In addition, there are two research groups in Deaf Studies, one from Universidade Federal do Rio Grande do Sul, and the other from Universidade Federal of Santa Catarina, which are carrying out different investigations. Some of them are briefly described below: (1)

(2)

(3)

(4)

Policies of the relationships between Deaf Education, Cultural Studies and Deaf Studies, Deaf identity, Deaf culture, the definition of a Deaf person, and Deaf community organisation are some topics that have been studied by Deaf and hearing students in the context of the logical inversion process (Miranda, 2001; Perlin, 2000; Quadros & Perlin, 2003). Deaf families are a complex topic. Cenzi (2002) presents a study identifying the representations of hearing parents, of sign language, and identities by their Deaf children. Infant Deaf Education, Teenager Deaf Education, Adult Deaf Education. There are different groups from Deaf schools analysing their practices and rethinking how to implement Deaf Education in each specific context. The discussions include Deaf claims, the political situation, teacher qualifications and the pedagogic and philosophical plans of the school. The main questions are: Who are the Deaf adults who have studied in our schools? Who do we want to educate for what? What do we, Deaf and hearing people, think about Deaf Education? It is interesting that the main points from Deaf claims about Deaf Education are related to access to information with quality in their language and Deaf teachers. The schools participating in this process are starting to realise that they are not good schools for Deaf people and, so, they are beginning to make plans and try to work out ways of changing the situation. These depend on each individual case, rather than being general proposals. For instance, in one of the Deaf public schools in Rio Grande do Sul, the group decided that they should have Deaf teachers participating in Deaf Education. They did not have any Deaf teachers with formal qualifications at that time, so they started looking for someone who might be a good Education Professional. They found a Deaf man and he quit his original job to assume a position as a kind of teacher in a special situation, with the mayor´s agreement. At the same time, he started to study, in order to be able to regularise his position at the school. Also, in the meantime, the town administration created a position for Deaf teachers with a specific qualification to work in Deaf schools. After about 8 years, he and three more Deaf teachers had normal teaching posts. Nowadays, there is a real possibility of being able to offer more posts for Deaf teachers in the future. Distance Deaf Education is a new area in Brazil. There is a huge project planning education at undergraduate and graduate levels through e-learning. This project is justified since the professionals who could become teachers in the programmes are spread over the country (and Brazil is a big country). This also makes sense, since Deaf students from

44

(5)

(6)

(7)

(8)

(9)

Bilingual Education in South America

different regions need to have access to education, which they currently do not have. The training and qualification of Deaf professionals is a concern of the Deaf people who have already been educated. They want to keep studying and get qualified. There are some Deaf professionals who have started graduate courses, especially, in the field of Education. Translation and interpretation of Brazilian Sign Language and Brazilian Portuguese is very important in the process. Deaf people are aware of the importance of having qualified interpreters. Because of this, some programmes to begin the training of interpreters have been started and currently there is a project to offer a programme for interpreters both at undergraduate and graduate major level. The latest programme to begin, the training of qualified sign language interpreters who already work as interpreters without any specific qualification, was offered to 60 people in the Ministe´rio de Educac¸a˜o e Cultura. One or two of these are from each state in the country. We have 27 states with completely different realities in terms of education in general, in terms of Deaf organisations, and in terms of sign language recognition. Although it was eclectic, at the same time, it represented a starting point for sign language interpreter qualification in the country. Brazilian Sign Language as part of the definition of the Deaf person has been a subject of recent research (Miranda, 2001; Perlin, 2000; Quadros, 1997a; Quadros & Perlin, 2003; Skliar & Quadros, 2000; Souza, 1998). It is already clear to people working in related areas that sign language is very important for the constitution of Deaf identities and cultures, but it is still unclear what to do with this difference in Education from the public and political point of view. To recognise difference is not difficult, but to plan what to do with difference is much more complex. Teaching Brazilian Sign Language as a first language, teaching Brazilian Sign Language as a second language, Brazilian Sign Language acquisition, Brazilian Sign Language grammar are some of the special issues that Deaf people are interested in investigating. They have begun to realise that each of these aspects requires specific research. They are also aware of specific methodologies organised in a Deaf way based on visual experiences. Specific technological areas in Deaf education have begun to be considered relevant in Deaf education. There are specific products related to technology which can be used by students in and out of the class, such as, software with lessons planned based on the visual experience and in Brazilian Sign Language (Napoli & Ramirez, 2002). Also, Stumpf (2002) is developing technology with signwriting, together with sign language and conceptual representations.

We would also like to mention a special project involving Deaf people from inland areas in Rio Grande do Sul. The project is mainly aimed at providing Deaf experiences for Deaf people from the countryside, who have not been organised as a Deaf community. Some of these Deaf people have never been in contact with other Deaf people and, therefore, they do not know how to

Bilingual Deaf Education in the South of Brazil

45

sign. This project was started in 1999 with very impressive results. In small cities, Deaf organisations started to be created and to connect with other Deaf organisations, constituting a network of different Deaf experiences. Moreover, they started to become literate as part of the educational programme. Miranda (2001) presented an analysis of this process showing the importance of the Deaf meeting with their peers and how sign language is the way to make all this possible. Nowadays, this project is growing and includes new inland areas as a State Government project in Rio Grande do Sul, together with the Universidade Federal de Santa Caterina (UFRGS). Martins (2003) analysed the building of subjectivity through sign language in these Deaf people, who had never signed before, when they met Deaf signers. He found that their lives changed after starting to learn sign language, in that they acquired sign language, constructed a Deaf identity, recognised themselves as students, and constructed discourses. Having access to language meant also having access to the world, becoming a significant actor in this play. The past started to be reviewed, interpreted and told by them. Something similar is happening in Santa Catarina, although at regional levels. There is a project preparing Deaf people from different areas of the state to become leaders. The aim is to give these Deaf leaders guidelines to organise the Deaf community in each area of the state. This is already happening in some places more than in others. This articulation also involves educational proposals, political issues and the creation of Deaf organisations. At the same time, we think that Deaf Education depends on the establishment of a network with all the people involved in this process, such as interpreters of Brazilian Sign Language, public power, public and private universities, Deaf and hearing teachers, Deaf organisations and researchers. Therefore, a discussion forum was created and opened to the community in two places: at UFRGS, in Porto Alegre, Rio Grande do Sul, and at CEFET=SC, in Florian
opolis. Since 1998, these two forums have been discussing different themes: the Deaf at university, intellectual Deaf people, teaching Deaf teenagers in the middle school, laws to officially recognise sign language in different areas of the country, the political organisation of the Deaf community and the participation of Deaf people in decisions about Deaf education. These forums have had an impact on growing Deaf political participation in society and they have made a difference to actions in Deaf Education. Deaf people are now participating actively in the decisions about their claims and rights.

Final Comments We are sure that Deaf Education involves all these paths. It does not involve only discussion on language, but this subject is complex and involves a rich interplay of many, often subtle, details, as we have tried to show here. The present is an ambiguous time, with changes in the meaning of the representations of Deafness, Deaf people, Deaf Education. It is the time of the epistemological inversion of what the Deaf ‘problem’ is to hearing ‘problems’ in Deaf Education. Deaf claims have started to be recognised by some of the hearing professionals and hearing state governors. They have started to realise the complexity of Deaf Education. Deaf leaders are

46

Bilingual Education in South America

continually fighting to show what they want, what they believe and what they are doing. Deaf people are engaged in the process; they are political people with fluctuating identities acting in the play of life. In presenting the South of Brazil experience, we feel that bilingualism in Deaf Education has to be the starting point to make Deaf people’s rights possible, in order for them to live as true citizens. Correspondence Any correspondence should be directed to Dr Carlos Skliar, Universidade Federal do Rio Grande do Sul, Brazil ([email protected]). References Bhabba, H. (1994) The Location of Culture. London: Routledge. Cenzi, L.B. (2002) Representac¸o˜es acerca da surdez de ma˜es ouvintes de crianc¸as surdas no contexto da escola. Dissertac¸a˜o de Mestrado. Universidade Federal do Rio Grande do Sul. Porto Alegre. Collier, V.P. and Thomas, W.P. (1989) How quickly can immigrants become proficient in school English? Journal of Educational Issues of Language Minority Students 5, 26–39. Cummins, J. (1992) Bilingual education and English immersion: The Ramı´rez report in theoretical perspective. Bilingual Research Journal 16, 91–104. Cummins, J. (1996) Language proficiency, bilingualism and academic achievement. In P.A. Ricard-Amato (ed.) Making It Happen: Interaction in The Second Language Classroom. From Theory to Practice. Cummins, J. and Swain, M. (1986) Bilingualism in Education. Applied Linguistics and Language Study. Longman. Davis, L. (1997) Enforcing Normalcy. Disability, Deafness and the Body. New York: Verso. Documento do l Semin
ario Nacional de Surdos (2001) Caxias do Sul, Rio Grande do Sul. Foucault, M. (1980) Di
alogos Sobre el Poder. Madrid: Alianza Editorial. Grosjean, F. (1982) Life with Two Languages: An Introduction to Bilingualism. Cambridge, MA: Harvard University Press. Martins, R.V. (2003) Lı´ngua de sinais e construc¸a˜o de narrativas. Tese de Doutorado em Psicologia. Pontifı´cia Universidade Cat
olica do RS, Porto Alegre. McLaren, P. (1995) White terror and oppositional agency. Towards a critical multiculturalism. In P. McLaren (ed.) Critical Pedagogy and Predatory Culture. Oppositional Politics in a Postmodern Era (pp. 117–144). London: Routledge. Miranda, W. (2001) Comunidade dos Surdos: Olhares sobre os contatos culturais. Dissertac¸a˜o de Mestrado. Universidade Federal do Rio Grande do Sul, Porto Alegre. Moura, M.C. de (1999) O Surdo: Caminhos Para uma Nova Identidade. Rio de Janeiro: Revinter. Napoli, M. and Ramirez, A. (2002) Elaborac¸a˜o de um sistema de ensino para surdos que sistematiza o ensino da lı´ngua portuguesa partindo de um perspectiva com a LIBRAS. Revista Ponto de Vista 4. Perlin, G. (1998) Identidades surdas. In C. Skliar and A. Carlos (eds) Surdez: um Olhar Sobre as Diferenc¸as Porto Alegre: Editora Mediac¸a˜o. Perlin, G. (2000) Identidade surda e currı´culo. In C.B. Feitosa Lacerda and M.C. Rafael de G
oes (eds) Surdez – Processos Educativos e Subjetividade Sa˜o Paulo: Lovise. Quadros, R.M. (1997a) Educac¸a˜o de Surdos: A Aquisic¸a˜o da Linguagem. Porto Alegre: Artes Me´dicas. Quadros, R.M. (1997b) Aquisic¸a˜o de L1 e L2: o contexto da pessoa surda. Anais do Semin
ario: Desafios e Possibilidades na Educac¸a˜o Bilingu¨e para Surdos. 21–23 July, 1997. Quadros, R.M. (2000) Alfabetizac¸a˜o e o ensino da lı´ngua de sinais. Textura, Canoas 3, 53–62. Quadros, R.M. de and Perlin, G. (2003) O ouvinte: o outro do outro surdo. Paper presented at the III Semin
ario de Estudos Interculturais. UFSC: Florian
opolis.

Bilingual Deaf Education in the South of Brazil

47

Skliar, C. (ed.) (1997) Educac¸a˜o & Exclusa˜o – Abordagens S
ocio-Antropol
ogicas em Educac¸a˜o Especial. Porto Alegre: Editora Mediac¸a˜o. Skliar, C. (ed.) (1998) A Surdez: um Olhar Sobre as Diferenc¸as. Porto Alegre: Editora Mediac¸a˜o. Skliar, C. (2000) Estudos surdos e estudos culturais em educac¸a˜o. In C. Broglia Feitosa Lacerda and M.C. Rafael de G
oes (eds) Surdez – Processos Educativos e Subjetividade. Sa˜o Paulo: Lovise. Skliar, C. (2001) Perspectivas polı´tica e pedag
ogicas da educac¸a˜o bilı´ngu¨e para surdos. In S. Silva and M. Vizim (eds) Educac¸a˜o Especial: Mu´ltiplas Leituras e Diferentes Significados (pp. 85–110). Sa˜o Paulo: Mercado Aberto. Skliar, C. and Quadros, R.M. (2000) Invertendo epistemologicamente o problema da inclusa˜o: os ouvintes no mundo dos surdos. Estilos da Clı´nica, Sa˜o Paulo V (9), 32–51. Souza, R.M. (1998) Que Palavra Te Falta? Sa˜o Paulo: Editora Martins Fontes. Stumpf, M. (2002) Transcric¸o˜es de lı´ngua de sinais brasileira em signwriting. In Letramento e Minorias. A.C.B. Lodi, K.M.P. Harrison, S.R.L. de Campos and O. Teske. (org.) Editora Mediac¸a˜o. Porto Alegre. 62–70. Veltman, C. (1988) The Future of the Spanish Language in the United States. Washington, DC: Hispanic Policy Development Project. Wrigley, O. (1996) The Politics of Deafness. Washington: Gallaudet University Press.

Bilingual Education in Colombia: Towards an Integrated Perspective Anne-Marie de Mejı´a Universidad del Valle, Cali, Colombia The purpose of this paper is to argue that the traditional division between bilingual education programmes offered to speakers of majority languages and those available to minority language speakers in Colombia should be reconsidered within a wider, integrated vision of bilingual provision. Initially, developments will be situated in relation to the sociolinguistic context of language use in the country. This will be followed by a brief historical account of how and why bilingual education started in ethnic minority contexts in relation to present developments, with a consideration of both advances and certain difficulties which need to be resolved if these programmes are to become more effective in the future. The main focus of the paper is on the situation of bilingual programmes for majority language speakers, particularly learners of English, related to both a historical perspective and current situation and effectiveness. Finally, possible areas of convergence between the two traditions will be indicated as ways forward for the future. Keywords: bilingual education, majority language contexts, minority language contexts, Colombia, policy, classroom practice

Introduction In Colombia, as in many other South American nations (see King, this volume), there exists a traditional divide between policy, practice and research into bilingualism and bilingual education programmes for majority language speakers, and modalities offered for minority language speakers. Thus, teachers and researchers who work in schools offering bilingual programmes in international languages, such as English, French and German, normally have little contact with researchers and practitioners who are concerned with bilingual education programmes in the Amerindian and Afro-Caribbean communities. Indeed, bilingual education programmes for the ethnic minorities are officially referred to as programmes of Etnoeducaci
on (Ethnoeducation) rather than Bilingual Education.1 This separation leads to a necessarily limited view of the progress of bilingualism and bilingual education within the country as a whole, and means that linguistic and pedagogical insights and perceptions from each tradition are not available to inform future general developments in the field. The visions of the communities that support bilingualism in these two traditions are also very different. Bilingual education in majority language contexts is associated with foreign language teaching and, as such, is connected with input from foreign-based organisations, such as the British Council, Goethe Institut, and L’Alliance Franc¸aise. The families who send their children to bilingual schools come from the Colombian middle- and upper-middleclasses, the international community, and those working for multinational

48

An Integrated Perspective in Colombia

49

organisations, particularly in the capital, Bogot
a. In contrast, those who work with the ethnic minority communities are generally Colombian anthropologists, ethnographers and more recently ethnolinguists, due to the influence of the Colombian Centre for Studies in Indigenous Languages (CCELA). The families whose children study in Ethnoeducation programmes are generally of peasant origin, and come from isolated rural communities, often caught between the demands of the left-wing guerrilla groups and the right-wing paramilitary organisations. Furthermore, in Colombia it may be noted that while bilingualism in internationally prestigious languages, such as Spanish–English, Spanish–French, and Spanish–German is considered worthy of investment of considerable sums of money, as it provides access to a highly ‘visible’, socially accepted form of bilingualism, leading to the possibility of employment in the global marketplace, bilingualism in minority Amerindian or Creole languages leads, in most cases, to an ‘invisible’ form of bilingualism in which the native language is undervalued and associated with underdevelopment, poverty and backwardness (de Mejı´a, 1998). In spite of these obvious differences, these two traditions have one thing in common: both of them may be seen as minority phenomena, if considered within mainstream Colombian education. As Iriarte (1997: 73) observes: Except for a few exceptions – pilot experiences of bilingual intercultural education in Indian regions and some foreign schools in the main cities in the country – primary and secondary education in Colombia has developed based on Spanish . . . as the first language of 93% of the Colombian population. Therefore, it is not possible to talk about bilingualism in Colombia as a generalised issue.2 Bearing in mind the above, and considering these two traditions as part of the same continuum, in this paper, after situating developments in relation to the sociolinguistic context of language use in the country, I will discuss the historical origins of these two different groupings. I will then go on to relate them to present-day developments in the area, noting both advances and certain difficulties which need to be resolved if bilingual programmes are to become more effective in the future. I will also attempt to indicate some areas of convergence between the two types of bilingual education provision as an illustration of the importance of considering the development of bilingual and multilingual provision within Colombia as an integrated phenomenon which goes beyond the separatist tendencies characteristic of present-day discussions.

Sociolinguistic Context of Language Use in Colombia From a sociolinguistic point of view, minority language groups in Colombia generally speak Spanish as a second language and have a minority community language as mother tongue, either a native Amerindian language or an English or Spanish-based Creole. Majority language speakers of Spanish as a first language are usually interested in becoming bilingual in an international language, such as English, French, German, Italian and Hebrew (see de Mejı´a, 2002).

50

Bilingual Education in South America

A great boost was given to the status and use of the Amerindian and AfroColombian (Creole) languages in the Colombian Constitution of 1991, where it was officially recognised, for the first time, that Colombia is a multi-ethnic and pluricultural nation (Article 7) and that the languages of the minority communities would be co-official with Spanish in the areas where these were spoken. Furthermore, the new constitution recognised bilingual education as the form of education to be implemented in these communities. Although official recognition was hailed by the communities as a great step forward, this did not immediately change ingrained attitudes towards these minority languages, particularly the Creole languages, which are often considered examples of badly-spoken, or ‘broken’ English or Spanish. However, as noted by Dieck (1998) there is recent evidence of change in this area, in that members of the Palenquero (Spanish-based Creole) community and the English Creole speakers of the Caribbean islands of San Andre´s and Old Providence are beginning to revalue their languages as part of their cultural heritage which needs to be preserved. In the case of majority languages, English is the foreign language which enjoys the highest status in the country, particularly in the domains of education, business and tourism (Zuluaga, 1996). North American English is generally the most favoured variety due to the ‘overwhelming attraction of the USA by dint of historical connections, family and teacher connection, proximity and of sheer glamour image’ (British Council, 1989: 10). Except for its use as a means of communication in small expatriate communities, mainly found in the capital, Santafe´ de Bogot
a, neither English, nor any of the other foreign languages referred to above, is generally used as a means of communication within the country.

Bilingual Education in Minority Language Contexts Amerindian communities After the Spanish conquest of Colombia in the 16th century and continuing into the 19th century, the education of the indigenous ethnic groups in Colombia was the responsibility of Catholic missionaries, whose objective, as expressed in the signing of a treaty with the State in 1888 was ‘the colonisation and christianisation of the whole of the (country’s) periphery’ (Jime´nez, 1998: 37). The priority of the religious orders at this time was the evangelisation of the Indian tribes according to Roman Catholic doctrine, as well as spreading the use of Spanish as the language of education (castellanizaci
on). In the 1930s, a system of religious boarding schools using Spanish as the language of teaching and learning, was established for selected Indian students from different ethnic communities. As a consequence, Spanish continued to be consolidated as the main language of education and prestige within the Indian communities. The year 1978 marked an important change of emphasis in educational provision for the Indian communities in the country. For the first time, the specificities of the type of education considered appropriate for their needs was officially recognised (Decree 1142) and it was seen as important to design curricula which took into account ‘the educational experiences of the communities

An Integrated Perspective in Colombia

51

themselves, within a respect for their cultural heritage . . . and the selection of teachers who should be bilingual’ (Jime´nez, 1998: 38). The Education Law of 1994 followed this breakthrough by establishing a policy of Ethnoeducation for the minority communities in Colombia, characterising this as ‘a permanent social process of reflection and collective construction, by means of which the Indian communities would strengthen their autonomy within an intercultural framework’ (Trillos, 1998: 73). In spite of these important advances in legislation, the current situation of Indian education in the country is still difficult. Some of the problems noted in the development of Ethnoeducation programmes with the 66 different Indian language groups derive from linguistic sources, while others are a result of educational difficulties. We will examine a few of these in more detail below. A common problem encountered by linguists and anthropologists working with the communities is that many of the teachers whose experience of primary education derives from what they were taught by the missionaries, have no experience of ‘intercultural pedagogy’ (Trillos, 1998: 337) and do not know what exactly is meant by the concept of ‘bilingual education’. Thus, while Ethnoeducation programmes continue to be officially promoted as ‘a good idea’, in many cases they ‘lack an active commitment to put them into practice; being conceptualised at national level in limited fashion and not being integrated into the school curriculum, where supposedly they should be the foundation, because they have no practical value’ (Trillos, 1998: 332). Another frequent difficulty for teachers and pupils is that there is an almost complete lack of written materials in native Indian languages, which can be used as classroom resources. Initiatives to translate standard textbooks written in Spanish into Indian vernaculars, such as Cams
a (a language spoken in the Southern Department of Putumayo), have not always respected the phonological particularities of the native languages in their concern to develop a common alphabet for both Spanish and the Indian languages (Jamioy, 1998). The heterogeneous levels of linguistic proficiency of children in the same classroom is another common source of difficulty for both practitioners and students. This has been documented by Bomba (2000) in his study of language use in Nasa community 5th Grade classrooms in the Southeastern Department of Cauca, in relation to Indian and Mestizo (Mixed Race) students, where Mestizo children generally have a more proficient use of Spanish than Indian children. However, the same researcher notes the progress of Nasa Indian students after working with a bilingual Indian teacher who gave explanations in their first language, of classroom activities previously carried out in Spanish. After the first four months of this type of bilingual intervention the Indian children began to smile and participate in Spanish in the class, as well as talking openly in their native language (Nasa Yuwe). Furthermore, in the breaks (recess) there was evidence of the Mestizo children asking how to say things in Nasa Yuwe, and imitating their Indian classmates in what they said. The lack of appropriate preservice and inservice teacher education courses, and the fact that many Indian teachers have not kept up with or have rejected their community traditions has led to teacher resistance to the cultural and linguistic demands involved in Ethnoeducation programmes. Thus, although there are many primary schools which are officially referred to as bilingual,

52

Bilingual Education in South America

intercultural schools, many of these teach the official pensum almost completely in Spanish, without designing alternative curricular proposals (Jamioy, 1998). In addition, according to Montes (1998: 270) many Indian teachers ‘feel insecure with regard to the community, who may well reproach them for their low level of knowledge of their culture in a demanding ethnoeducation situation’. In an attempt to help solve some of these difficulties, various Colombian Universities, particularly those in the South of the country, such as Universidad de la Amazonı´a (in Florencia, Caquet
a) and Universidad del Cauca (in Popay
an, Cauca), have designed distance learning programmes aimed at training Indian teachers to work in community bilingual education initiatives. The teacher training programme at the Universidad de la Amazonı´a, for example, is currently working with 36 students who come mainly from seven different Indian communities and who have had experience working in bicultural community schools. Afro-Caribbean communities Palenque de San Basilio The education of the Afro-Caribbean inhabitants of Palenque de San Basilio in the northern state of Bolı´var, descendants of runaway African slaves, who in the 17th century constructed fortified settlements or palenques from which they resisted attempts by the Spanish authorities to recapture them, has traditionally been carried out in Spanish. This ignores the fact that community members generally speak Palenquero as their first language, which has the distinction of being ‘the only Spanish-based Creole language which has survived in the Caribbean’ (Dieck, 1998: 324). However, the fact that the language was only recognised as such in 1970, is one reason for its long exclusion from the educational domain. Another is its traditional low status, both within the community and in the surrounding neighbourhood which has led to teachers repressing the use of what they considered ‘badly-spoken Spanish’ (Dieck, 1998: 329). However, since 1986, there has been a move by the regional authorities to develop educational provision based on the needs and cultural characteristics of the community, through initiatives such as the Black Communities Ethnoeducation Programme ‘Education for Identity’ which has produced the first reader in Palenquero for initial literacy purposes (Gobernaci
on del Departamento de Bolivar, 2001). Since the mid-80s Palenquero has been used in the schools in the locality and has had a positive impact on both students and parents, leading to a revaluation by the community due to its legitimisation as a language for education (Pe´rez, forthcoming). Furthermore, the local authorities are currently backing an ambitious project aimed at a corpus-based, linguistic analysis of the language with the aim of producing a bilingual Palenquero–Spanish dictionary, a Palenquero grammar, and an anthology, as well as pedagogical material. The island of San Andre´s This Caribbean island has a long history of multilingualism, since its original colonisation by a group of English Puritan settlers in 1631, whose tobacco and

An Integrated Perspective in Colombia

53

sugar plantations were worked by African slaves. After the Spanish conquered the island in 1786, the English-speaking native inhabitants were allowed to remain, provided they swore allegiance to the Spanish Crown, converted to Catholicism and communicated in Spanish. While there was nominal acquiescence to these demands, in reality language and religion became means of resistance to Spanish domination (O’Flynn de Chaves, 1998). The multilingual reality of language use is reflected in the education system. The majority of schools cater for Spanish speaking ‘Continentals’ and use Spanish as the medium for teaching and learning. However, seven primary schools and three secondary schools are officially recognised as bilingual institutions, where teaching and learning is carried out in Standard Caribbean English and in Spanish. The English-based Creole used by the majority of the native Islander population has not been considered, until recently, a language appropriate to be used in Education and has been relegated to informal, family domains. However, in 1998 the Christian University of San Andre´s, under the auspices of the Baptist churches of the Southern United States, together with representatives of the Summer Institute of Linguistics,3 designed a pilot project for the production of pedagogical material in ‘Islander English’ (Creole) to be used in three Baptist primary schools on the Island at the level of preschool and Grade 1. The idea was to begin literacy in Creole and then change to reading and writing in Standard English in Grades 2 and 3. Literacy in Spanish would begin in Grades 3 and 4, ending up with approximately 50% of teaching and learning in English and 50% in Spanish in Grade 5. Preliminary results indicate that both teachers and students feel more motivated working in Creole on familiar topics, which are easily understood by pupils. Teachers have been encouraged to produce other materials using Creole and feel proud of ‘being able to write in their family and community language, using their cultural knowledge and feelings’ (Bowie & Dittmann, forthcoming). The project has also helped them to become more consciously aware of separating their use of Creole from Standard English in the classroom. The problems noted so far in the project are connected with the lack of real financial support from the local Educational Authorities, as well as the need to familiarise both parents and members of the local government with the aims and methodology of the project. There is also need to provide better physical conditions and equipment for the participating schools.

Bilingual Education Provision in Majority Language Contexts Historical developments In similar fashion to the situation in the Indian communities, educational provision for the descendants of the Spanish settlers in Colombia was in the hands of the Catholic missionaries who followed in the wake of the conquistadores (conquerors). At this time, the languages taught in these schools were mainly Greek, Latin and Spanish. Later, after independence from Spain in 1810, the ruling class, based mainly in Santafe´ de Bogot
a, Popay
an and Cartagena de Indias, sent their children to study in France and England. They, in turn, brought back books which they had translated into Spanish, thus

54

Bilingual Education in South America

paving the way for the teaching and learning of these languages in Colombia (Zuluaga, 1996). In more recent times, after the Second World War in 1945, as in many other South American countries, English became the most important foreign language in Colombia, due to economic expansion, social, political and economic influence and the technological development of the United States (Zuluaga, 1996). It was taught at secondary school level, alternating with the use of French. In 1979, after a visit by the Colombian president to France, a decree was issued, making English compulsory for Grades 6 and 7 and French mandatory for Grades 10 and 11, with a free choice of either English or French in Grades 8 and 9. In practice, most schools chose to teach English for four years and French for two, with an intensity of three hours per week at all levels, except the final two years, when foreign languages were taught for two hours. As a report compiled by the British Council (1989) revealed: The Colombian Ministry of National Education has no firm foreign language policy for the secondary school curriculum . . . concerning the place of English and French, with decisions being made as a result of political pressures rather than educational considerations. (British Council, 1989: 7) More recently, with the General Education Law (1994) foreign languages were introduced for the first time at primary school level, generally from Third Grade onwards. It was stated that at this level attention should be focused on ‘The acquisition of elements of conversation and reading in at least one foreign language’ (Article 21, m). According to Rey de Castro and Garcı´a (1997: 5), ‘The new law gives clear signs of official recognition of the importance of English to support: (i) the development of the Colombian economy; (ii) the education systems to enhance Colombian opportunities in the era of globalisation’. Due to these developments, in recent years, the profile of language teaching, especially English Language teaching, has become more prominent in the country and there have been several initiatives at the level of the Ministry of Education aimed at raising standards in this area. One of the most recent was the ‘National New Technology and Bilingual Programme’ (1996–7) which promoted the setting up of 1500 computer multimedia classrooms connected to the Internet, throughout the country, to help students learn to use scientific and technological information in English. The programme also provided a three-month stay for 3200 selected teachers in the United States, in order that they might become familiarised with the use of information technology in language teaching. Present situation Bilingual education in Colombia is associated principally with private bilingual schools set up to cater for the middle- and upper-middle-classes (de Mejı´a, 2002). These are found mainly in urban areas, particularly in the cities of Santafe´ de Bogot
a, Medellin, Cali, Cartagena and Barranquilla and have increased greatly in demand over the last decade. There are around 40–50 bilingual schools currently in existence in the country, most of them

An Integrated Perspective in Colombia

55

providing English–Spanish bilingualism. The longest established institutions were founded in the 1910s and 1920s in order to provide the sons and daughters of the representatives of multinational companies stationed in Colombia and members of the expatriate communities with access to suitable bilingual and bicultural programmes. Since then, this type of educational provision has been extended to cater for Colombian nationals and today most of the students in bilingual schools come from monolingual Colombian families who wish to do postgraduate study abroad (de Mejı´a, 1994). According to preliminary results from this study, the majority of parents surveyed wanted a bilingual education for their children to enable them to study abroad at university level and to have better job opportunities when they returned. Although in the country there is no consensus as to what is precisely meant by the term ‘bilingualism’, most people take it to mean ‘proficiency in the use of the (foreign) language’ (Rey de Castro & Garcı´a, 1997: 5). A similar lack of precision can be seen in the designation of what is meant by the term ‘bilingual schools’. While most of the long-established bilingual schools use both Spanish and a foreign language as media of instruction in their programmes, there are many more recent bilingual educational institutions which, in fact, provide an intensive foreign language programme (usually ranging from between 8–20 hours per week), rather than offering bilingual content-based teaching and learning. According to a recent survey (de Mejı´a & Tejada, 2001) present-day bilingual schools can be divided roughly into two groups. The first group consists of those schools which have a strong foreign connection, such as The German School, with branches in Bogot
a, Medellin, Cali and Barranquilla, and the Colombo Brit
anico School in Cali. This type of school has close contacts with foreign governments and often receives direct financial support, or the appointment of foreign teachers to work in the schools. The headteachers are usually foreign nationals and many of the materials and books used are imported from abroad. Students often have the opportunity for direct contact with the foreign country through exchanges or supervised visits organised by the schools, and international exams like the German Sprachdiplom or the International Baccalaureate are offered, as well as the Colombian High School Diploma. The second group of bilingual schools are national institutions which aim at a high level of student proficiency in at least one foreign language, usually English, in addition to the first language, Spanish. Most of these establishments were founded by individuals or small groups of people, generally Colombians (Arau´jo & Corominas, 1996). Some examples of these schools are The Montessori School in Medellin and Los Nogales School in Bogot
a. Some of these schools may be classified as bilingual institutions, in the sense that they have a high degree of contact with the foreign language, foreign teachers and use two languages as vehicles of teaching and learning, yet they do not class themselves openly as such, because they wish to emphasise their role as educators of Colombian citizens. The headteachers are generally Colombian and are conscious of the importance of social and ethical values in education to counterbalance what they see as a strong tendency towards social disintegration in the country.

56

Bilingual Education in South America

English–Spanish Bilingual Schools Programmes and participants As can be seen from the above discussion, there is great variety in the programmes and practices of the different private bilingual schools in the country, which makes generalisation difficult. I will, therefore, take one example of a bilingual (English–Spanish) school in Cali, which may be considered fairly typical in relation to language and subject distribution at different moments in the curriculum, and try to characterise certain design features of this programme. Most of the information reported here comes from two studies: Cosh (1998) and de Mejı´a and Tejada (2001). The bilingual school in question is divided into three sections: pre-primary (children aged 4–6 years old); primary (Grades 1–5, children aged 7–11 years old); and secondary (Grades 6–11, students aged 12–17 years old). The bilingual programme begins in pre-primary and continues through to 11th Grade, where the students are encouraged to present the International Baccalaureate. Thus, the total length is 14 years. The programme may be considered an enrichment modality (Hornberger, 1991) in which the student’s first language is valued and supported at the same time as the foreign language is developed. Figure 1 gives an idea of the longitudinal distribution of both English and Spanish throughout the curriculum.

Figure 1 Bilingual education-immersion programme in Cali school

An Integrated Perspective in Colombia

57

Pre-primary In pre-primary there are three levels: A, B, C. In level A the language distribution is approximately 30% English: 70% Spanish; in level B there is approximately 50% English and 50% Spanish; and in level C there is 80% English and 20% Spanish. Thus, throughout pre-primary there is a gradual reversal of the language balance from Spanish towards English. The areas of the curriculum where Spanish is used at this level are: Music, Library, Physical Education, Games and Assembly. The other subjects are given in English. Primary In the primary school from Grades 1–4 there is an approximate language distribution of 50% English and 50% Spanish. In Grade 5 this changes to an approximate ratio of 40% English: 60% Spanish. At this level, Spanish Language, Social Studies, Religion, Art, Music and Physical Education are given in Spanish, while English Language, Mathematics and Science are taught in English. Secondary In secondary school there is a notable change. From Grades 6–11 there is an approximate language distribution of 25% English to 75% Spanish. Thus, it can be seen that whereas the first four years of primary school are focused on developing proficiency in both languages, for the rest of their school career, the students will have more contact with Spanish than with English. At secondary level, the subject division is as follows: Spanish Language, Social Studies, Art=Music, Philosophy=Religion, Physical Education, two thirds of the Mathematics programme, and approximately one third of the Natural Science programme (Biology, Chemistry and Physics) is taught in Spanish, while English Language, one third of the Mathematics Programme and two thirds of the Natural Science programme are taught in English. Over the entire school programme the approximate contact ratio between the two languages is Spanish 9825 hours: English 5262 hours. It may be considered as a type of partial immersion programme of a progressive nature, albeit with a variation in the relative ratios of language distribution proposed in the Canadian partial immersion model, which according to Baker (1996) favours a 50% balance at all grade levels. Many of the theoretical principles underlying programme design, especially at pre-primary level, are based on the work of Dodson (1985), who gave a seminar at the school in 1990. Dodson maintains that all individuals who are in the process of acquiring a foreign or second language, have a preferred language and another language that they use in specific domains. Proficiency in the foreign language may be increased by means of a bilingual communication between the preferred language and the other language if there is primary emphasis on the message rather than on the form. While the initial emphasis is on reception rather than production, by the third year of pre-primary the children are encouraged to develop suitable output in English as well as understanding what is said to them. Another key tenet of the bilingual programme described above is that the teaching and learning of English does not exclude the development of the children’s first

58

Bilingual Education in South America

language. New concepts at pre-primary level are introduced in the first language and then reinforced and consolidated in the foreign language (Cosh, 1998). Classroom language use at this level is characterised by extensive codeswitching, though generally it has been noted that English is used as the language of pedagogy, for classroom management and control, while Spanish is used for the introduction of academic concepts, for initial literacy, and as the affective language of contact between Colombian teachers and children (de Mejı´a, 1998). At primary level the emphasis is on providing the children with sufficient contact with the foreign language within the school, taking into account that English is not generally used as a means of communication in Colombian society. For this reason there is a general balance of 50% in the two languages used, and a separation approach is advocated. In secondary school, the development of English is conceived as a spiral throughout the curriculum. A separation approach to language use is advocated. The International Baccalaureate is a strong influence on the curriculum in the last two years (10th and 11th Grade), particularly in helping students to become more analytical and more autonomous in their learning. At present, Colombian Labour Law (Article 74C1) specifies that a maximum of 20% of teachers in any school may be foreigners brought into the country for the purpose of teaching in the school; the rest must be local. This ruling may be modified, by applying for permission to bring in extra teachers up to a limit of approximately 30% of the total staff. However, expatriate teachers living and working in Colombia are classed as local residents and therefore the numbers of foreign nationals working at any one time may be much higher than the 20% stipulated by the law. As a result of this situation, there are three different categories of teachers who work in most bilingual schools. First, there is the privileged class of foreign teachers contracted abroad, usually in USA or Britain, in the case of English–Spanish bilingual schools. Then there are foreign residents in Colombia, and finally there are a large number of Colombian nationals. There is a corresponding language continuum which goes from monolingual foreign language speakers, through bilingual speakers, to monolingual Spanish speakers. This differential in language proficiency is reflected in the areas the teachers work in. Generally foreign staff and national bilingual teachers teach Foreign Languages, Science, Maths and Economics in the foreign language, while Colombian monolingual staff teach Spanish language and literature, Physical Education, Art, Music and Religion in the first language. Although many schools see foreign language monolingualism among the staff as an advantage, in that the teachers will not be tempted to use Spanish in their classes, this can lead to difficulties of communication, especially with children who are in the initial stages of becoming bilingual. In general, school administrators value foreign language skills more highly in their staff than knowledge of the first language, and this is often reflected in differential rates of pay. According to the General Law of Education (1994), all Colombian staff need to have a recognised qualification in education (Article 198). Foreign staff, however, may be hired if they are qualified in areas other than education (Arti-

An Integrated Perspective in Colombia

59

cle 199). Many of the latter have training and experience in the teaching of English and other subjects to first language speakers. Relatively few, however, have been trained to deal with foreign language speakers of English, and while there is some in-service provision given, this is generally insufficient. In the case of language teachers, some bilingual schools require proof of foreign language proficiency of potential bilingual Colombian staff, such as the Michigan Test or the TOEFL, but this depends on the individual establishment. Few schools seem interested in what teachers know about the theory and practice of bilingual education, according to reports from graduates from the Postgraduate Diploma Programme in Bilingual Education at Universidad del Valle (1997). The focus of most bilingual schools is on the development of foreign language proficiency in the students and this is assumed to depend largely upon the level of the teachers’ foreign language proficiency. In work carried out as part of a recent ethnographic, participatory research programme in Colombian private schools which are in the process of becoming bilingual (de Mejı´a & Tejada, 2001), several important factors which have an influence on the spread of bilingual education in majority languages in the country have become apparent, as well as various difficulties. These will be summarised below. One fundamental factor influencing the success of bilingual educational programmes in Colombia is the potentially high reward of foreign language proficiency in terms of prestigious positions in business, science and technology, political institutions and academia. Due to the recent economic opening up of the country in response to globalising and internationalising tendencies, career advancement is dependent to a large degree on English language proficiency, and bilingual education is seen as the key to foreign language development. Thus, prestigious or ‘elite’ bilingualism has a very high profile among the Colombian middle and upper classes and there is increasing demand for bilingual programmes (especially English–Spanish provision). Parental support for bilingual education is thus very strong and attending a bilingual school is considered to be high status. Paradoxically, much of the success of the private bilingual schools in Colombia is that there is, as yet, no national bilingual education policy. While this situation leads to potential negative consequences with regard to consistency and co-ordination of bilingual provision at national level, it also allows schools the freedom to adapt features of established models to their own needs and philosophies and thus to cater more appropriately to the wishes of their students and parents. As noted by John Whitehead, British Council English Language Officer in Bogot
a in 1991: the differences between a private bilingual school in Bogot
a (the capital) where English is taught at pre-Kinder level and a state school in Villavicencio (a provincial city in the underdeveloped southern Department of Meta) . . . is huge. (personal communication) This flexibility is very much in accordance with current educational thinking in the country, as reflected in the General Education Law of 1994 which gave educational communities the autonomy to develop their own Institutional Educational Projects, taking into account Ministry guidelines.

60

Bilingual Education in South America

However, according to the results of a recent survey carried out on a selection of bilingual schools in the three largest cities in the country (de Mejı´a & Tejada, 2001), there is a need in each institution for a coherent school bilingual policy, which is incorporated into the Institutional Educational Project, where language distribution throughout the curriculum is justified both on academic and contextual grounds, and where the treatment of cultural aspects, from a multicultural, bicultural or intercultural perspective is contemplated. As yet, very few bilingual institutions in Colombia have policy documents of this kind. In the project referred to above (de Mejı´a & Tejada, 2001), it was noted that many schools see the idea of biculturalism as a threat to school philosophy and national identity. Certain binational institutions are seen as the only possible cultural model, and these are rejected by many, as leading to acculturation at best, and anomie and identity crises at worst. There is, thus, a great need for more knowledge about alternative linguistic and cultural models which could be adapted to the specific needs of bilingual schools in Colombia. In this respect, and considering the experience of many bilingual schools which contract foreign staff, it is important that schools which intend to become bilingual consider the implications of hiring foreign staff, whose presence, because of superior rates of pay and conditions, may cause division in the institution. While foreign expertise is often highly valued by both parents and school administrators as a sign of school status, foreign teachers are, in general, a transient population, who often do not have time to identify with the institution or its wider aims. The financial burden involved in hiring foreign staff often means that there are not sufficient financial resources for the professional and language development of the Colombian teachers. Furthermore, this dependence on foreign expertise has the disadvantage of potentially perpetuating a mentality of underdevelopment, in that foreign staff are often considered the principal purveyors of new ideas and methodologies. A further key condition for the development of successful and effective bilingual education programmes in Colombia, which has been recognised more clearly in the state sector than in the private domain, is the need for appropriate teacher training and development in this area. The traditional divide between language teachers and subject specialists needs to be bridged, either by team teaching arrangements or by qualifying bilingual teachers who are subject specialists and therefore able to teach content areas in two or more languages and who have an understanding of the basic tenets of bilingual education. As yet, bilingual teacher training and development programmes are in their infancy in Colombia, although the Postgraduate Diploma in Bilingual Education run by Universidad del Valle in Cali provides in-service training for teachers who are working in bilingual contexts.

Conclusions Bearing in mind the above, in this final section I will attempt to indicate some areas of convergence between the two traditional visions of bilingualism

An Integrated Perspective in Colombia

61

and bilingual education, which may act as reference for future policy initiatives aimed at the integration of these two areas. First of all, it is important to note that questions of the maintenance and loss of cultural identity are a preoccupation of both groups. In the case of the Indian communities, this concern has led to a reconsideration of the traditional function of schooling, which has tended to be centred on ‘the insertion of the new generations within the dominant culture, without taking into account the uses and the customs of the culture of origin’ (Trillos, 1998: 328). During the last 10 years, the Indian communities and their organisations, the National Indian Organisation of Colombia (ONIC) and the Indian Regional Council of Cauca (CRIC) among others, have supported moves to preserve traditional values and cultural models ‘which have remained hidden under the Europeanising influence which arrived with the Spanish’ (Trillos, 1998: 327). The spread of Ethnoeducation programmes, based on the concept of a multicultural and plurilingual nation, aim at providing students with competence in five main areas: biculturalism, or the ability to act appropriately both in the national society and in their own community; bilingualism, or proficiency in more than one language; knowledge of the main values of both cultures; positive attitudes to different linguistic and cultural groups; and equality of opportunity for children from minority communities (Trillos, 1998). Those working in bilingual education programmes for majority language speakers have traditionally ignored the implications of cultural contact, assuming that as students generally come from the dominant group there is no difficulty in this respect. However, as a recent article demonstrates (de Mejı´a & Tejada, 2002), several teachers in a school which was in the process of becoming bilingual ‘felt threatened by what they perceived as a loss of cultural identity. They feared that in the development of the project, the national culture . . . would be displaced or substituted by Anglo-Saxon culture’ (de Mejı´a & Tejada, 2002: 111). It has also been reported by teachers in binational schools that due to the immense prestige of US culture among middle and upper-class Colombians, some students in these type of schools have developed signs of cultural disorientation and anomie. A second area of convergence has to do with the status and development of the first language in both types of bilingual programmes. In majority as well as minority language contexts, there has been traditional emphasis on the importance of the development of the second or the foreign language, while the maintenance and development of the first language has been considered of secondary importance. In majority situations, the view is that as Spanish is the dominant language of the society, the children will ‘pick it up naturally’, without needing to spend time developing language skills in their first language when they might, more profitably be studying through the medium of the foreign language. In ethnic minority situations, as we have indicated, community languages have traditionally been categorised as low-prestige ‘dialects’, of little use and value for education (Landaburu, 1998: 308). In both cases, there has been a recent revaluation of the importance of the first language in the process of becoming bilingual, particularly in the initial stages of language learning. In the case of majority bilingual provision, teacher investigators at pre-school level in Cali (Arau´jo & Corominas, 1996) have

62

Bilingual Education in South America

recommended the parallel development of both languages, so that children are helped to associate advances in the foreign language with previous experience in their first language. In minority programmes, the increasing tendency towards constructing written systems for oral languages has had a positive effect on changing perceptions towards the status of oral vernaculars (Landaburu, 1998). Thirdly, official emphasis in the two traditions of the importance of incorporating specific concerns of the educational communities involved within bilingual education programmes is coherent with the flexible policies advocated by the Ministry of Education. It is recognised that each institution should adapt the type of bilingual programme offered, to the particular context of implementation, bearing in mind, in Baetens Beardsmore’s words, that it is important not to adopt: any single model, no matter how well-tried, without the necessary modifications to specific local circumstances . . . merely because the research background has proved (its) effectiveness in the context for which (it was) developed. (1995: 140) However, as has been discussed above, there are few concrete guidelines about how to operationalise principles in classroom practice. Thus, teachers in ethnic minority contexts often feel insecure about how to teach from an intercultural standpoint, while those who work on majority language provision are often equally unsure of what is entailed by a programme of bilingual education. Consequently, those who work in both minority and majority bilingual education are increasingly aware that the provision of appropriate pre-service and in-service teacher development courses is vital to the success of these programmes. In minority situations, staff training and development needs to help teachers to learn to ‘integrate their cultures within the curriculum as well as maintaining the majority culture, though not from an ethnocentric position . . . according to the situation and real needs of the communities’ (Trillos, 1998: 335). In majority situations, as has been noted, teachers need preparation that will help them understand the principles of bilingual and bicultural or intercultural development so that they may be able to cope with the challenges of educating individuals who can function appropriately in both their languages and cultures. Finally, it can be seen that as bilingual education programmes for majority and minority language speakers are concerned with language and cultural contact at regional, national, or international level, it is important to develop positive attitudes towards difference and diversity among those involved in these programmes. Increased contact and opportunities for sharing the results of research projects and pedagogical initiatives may help to avoid the danger of fragmentation and compartmentalisation which have characterised the development of bilingualism both in ethnic communities and in elite bilingual schools, and lead to greater understanding of common points of interest.4 As Genoveva Iriarte (1997: 79) aptly observes, ‘Educar en la diversidad es educar para la tolerancia’ (To educate from the standpoint of diversity is to educate towards tolerance).

An Integrated Perspective in Colombia

63

Acknowledgements The author would like to thank Pablo Mejı´a, Rocı´o Nieves, JoEllen Simpson and Harvey Tejada for their comments on earlier drafts of this article. An earlier version was presented as a paper in the ‘Third International Symposium of Bilingualism’ at the University of the West of England, Bristol, 18–20 April, 2001. Correspondence Any correspondence should be directed to Dr Anne-Marie de Mejı´a, c=o Escuela de Ciencias del Lenguaje, Universidad del Valle, A A 2188, Cali, Colombia ([email protected]). Notes 1. 2. 3. 4.

For more detail about this notion see Section Three. Author’s translation of this and subsequent quotations in Spanish in the original. Recently this organisation changed its name to ‘the Wycliffe Bible Translators’. Universidad del Valle in Cali has gone some way to providing a forum for researchers and teachers working in both traditions through the organisation of three symposia on bilingualism and bilingual education in recent years.

References Arau´jo, M.C. and Corominas, Y. (1996) Procesos de adquisici
on del Ingle´s como segunda lengua en ni~ nos de 5 a 6 a~ nos, de colegios bilingu¨es de la ciudad de Cali. Unpublished MA thesis, Cali: Universidad del Valle. Baetens Beardsmore, H. (1995) European models of bilingual education. In O. Garcı´a and C. Baker (eds) Policy and Practice in Bilingual Education. Extending the Foundations. Clevedon: Multilingual Matters. Baker, C. (1996) Foundations of Bilingual Education and Bilingualism (2nd edn). Clevedon: Multilingual Matters. Bomba, J. D. (2000) Aspectos del uso del Nasa Yuwe y del Castellano en una escuela oficial en el Resguardo de San Lorenzo de Caldono, Cauca. Unpublished monograph, Cali: Universidad del Valle. Bowie, P. and Dittmann, M. (forthcoming) El proyecto piloto trilingu¨e de la Universidad Cristiana de San Andre´s en las Escuela Emmanuel Bautista y Bautista Central en la Isla de San Andre´s. Proceedings of the Third National Symposium on Bilingualism and Bilingual Education. Cali: Universidad del Valle. British Council (1989) A Survey of English Language Teaching and Learning in Colombia: A Guide to the Market. London: English Language Promotion Unit. Cosh, E. (1998) Estudio de caso: An
alisis descriptivo de un programa educativo de inmersi
on en un colegio bilingu¨e de Cali. Unpublished report. Dieck, M. (1998) Criollistica Afrocolombiana. In L.A. Maya (ed.) Geografı´a Humana de Colombia: Los Afrocolombianos. Bogot
a: Instituto Colombiano de Cultura Hisp
anica. Dodson, C.J. (ed.) (1985) Bilingual Education: Evaluation, Assesssment and Methodology. Cardiff: University of Wales Press. Gobernaci
on del Departamento de Bolı´var (2001) Tesoros de la Lengua Palenquera. www.bolivar.gov.co=educacion=tercer milenio tesoros palenquera.htm Hornberger, N.H. (1991) Extending enrichment bilingual education: revisiting typologies and redirecting policy. In O. Garcı´a (ed.) Bilingual Education: Focusschrift in Honor of Joshua A. Fishman (Volume 1). Amsterdam=Philadelphia: John Benjamins. Iriarte, G. (1997) Bilingu¨ismo y sociolingu¨´ıstica. Bilingu¨ismo, Funci
on Cognoscitivo y Educaci
on. Bogot
a: Fondo de Publicaciones del Gimnasio Moderno.

64

Bilingual Education in South America

Jamioy, J. N. (1998) Proceso de educaci
on en el pueblo Cams
a. In M. Trillos (ed.) Educaci
on End
ogena Frente a Educaci
on Formal. Bogot
a, Universidad de los Andes: Centro Colombiano de Estudios de Lenguas Aborı´genes. Jime´nez, N. (1998) Etnoeducaci
on: Polı´tica oficial para la educaci
on en comunidades indı´genas. In M. Trillos (ed.) Educaci
on End
ogena Frente a Educaci
on Formal. Bogot
a, Universidad de los Andes: Centro Colombiano de Estudios de Lenguas Aborı´genes. Landaburu, J. (1998) La situaci
on de las lenguas indı´genas de Colombia: Proleg
omenos para una polı´tica lingu¨´ıstica viable. In M. Trillos (ed.) Educaci
on End
ogena Frente a Educaci
on Formal. Bogot
a, Universidad de los Andes: Centro Colombiano de Estudios de Lenguas Aborı´genes. Mejı´a, A.M. de (1994) Bilingual teaching-learning events in early immersion classes; A case study in Cali, Colombia. Unpublished PhD thesis, Lancaster University. Mejı´a, A.M. de (1998) Educaci
on bilingu¨e en Colombia en contextos lingu¨´ısticos mayoritarios: Hacı´a una caracterizaci
on del campo. Lenguaje 26, 1–12. Mejı´a, A.M. de (2002) Power, Prestige and Bilingualism. International Perspectives on Elite Bilingual Education. Clevedon: Multilingual Matters. Mejı´a, A.M. de and Tejada, H. (2001) Informe: Construcci
on de modalidades educativas bilingu¨es en colegios monolingu¨es de Cali. Gimnasio La Colina. Unpublished research report, Universidad del Valle. Mejı´a, A.M. de and Tejada, H. (2002) Empowerment y cambio en la construcci
on de currı´culos bilingu¨es (ingles – espa~ nol) en colegios monolingu¨es. In A.M. de Mejı´a and R. Nieves Oviedo (eds) Nuevos Caminos en Educaci
on Bilingu¨e en Colombia. Cali: Universidad del Valle. Montes, M.E. (1998) Las acciones lingu¨´ısticas sobre los pueblos indı´genas en Colombia. Algunas reflexiones preliminares. In M. Trillos (ed.) Educaci
on End
ogena Frente a Educaci
on Formal. Bogot
a, Universidad de los Andes: Centro Colombiano de Estudios de Lenguas Aborı´genes. O’Flynn de Chaves, C. (1998) El criollo Sanandresano. In Maya, L.A. (ed.) Geografı´a Humana de Colombia: Los Afrocolombianos. Bogot
a: Instituto Colombiano de Cultura Hisp
anica. Pe´rez, J.P. (forthcoming) Informe preliminar sobre bilingu¨ismo en San Basilio de Palenque. Proceedings of the Third National Symposium on Bilingualism and Bilingual Education. Cali: Universidad del Valle. Rey de Castro, R. and Garcı´a, D. (1997) Landmark Review of the Use of Teaching and Learning of English in Latin America. Colombia. London: British Council. Trillos, M. (1998) Conclusiones. In M. Trillos (ed.) Educaci
on End
ogena Frente a Educaci
on Formal. Bogot
a, Universidad de los Andes: Centro Colombiano de Estudios de Lenguas Aborı´genes. Zuluaga, O. (1996) La Ensen~anza de Lenguas Extranjeras en Colombia en 500 An~os. Popay
an: Taller Editorial, Unicauca.

The Evolution of Bilingual Schools in Argentina Cristina Banfi

Universidad Nacional de Co
rdoba and ESSARP, Argentina

Raymond Day Universidad de Belgrano, Argentina This paper sets out to provide a preliminary descriptive account of an important subsector of the Argentine education system: those schools locally known as ‘bilingual schools’ or ‘colegios bilingu¨es’. As the authors will show, the label ‘bilingual’ has, at times, been applied rather loosely to a number of institutions. For current purposes, the authors limit this description to those schools that aim to teach through the medium of two languages. These schools may be characterised both in terms of the shared features that set them apart from other schools in Argentina, and in terms of the distinctions among them, which make this a diversified sub-sector. The authors also argue that these schools have been transformed since they were first founded in the 19th century, evolving from what were originally Heritage Schools to Dual Language Schools and, ultimately, to a new form of Bilingual school that the authors label the ‘Global Language School’. Finally, the paper surveys some recent trends within this sector and suggests avenues for further research. Keywords: Argentina, heritage languages, bilingual schools, schools’ history, immigration, curriculum

Introduction This paper sets out to provide an initial description of an important sector of the Argentine education system; those institutions locally known as colegios bilingu¨es, i.e. bilingual schools. To be able to chart the evolution of bilingual schools we will firstly attempt to characterise and classify them in relation to existing definitions. At first sight, we can say that what differentiates these schools from others is their teaching of content through two languages. However, this does not mean to say that this is a homogenous group of schools, so we present certain similarities and differences across the Argentine bilingual school sector. We suggest that the evolution of Argentine bilingual schools is closely tied to the historical development of the country. We argue that this involves the change from what were originally Heritage Schools, to Dual Language Schools, and currently to a form of bilingual school we categorise as the ‘Global Language School’.

What is a Colegio Bilingu¨e? Applying a broad definition of bilingual education, e.g. ‘bilingual education involves using two languages in instruction’ (Garcı´a, 1997: 405), it is plausible to conclude that a significant number of Argentine schools fall into this

65

66

Bilingual Education in South America

category. However, this is not how the label ‘bilingual school’ is actually used in contemporary Argentine society.1 The popular perception stems from a long-standing tradition that has its origins in the private education sector, and that was, at first at least, initiated around immigrant communities. For the purposes of this paper, we will limit our discussion to those schools that teach content through the medium of two languages (Spanish and English, French, German or Italian) and cover areas of the Argentine curriculum and a foreign or international curriculum (for a summary of schools, see Table 1). For the purposes of constructing Table 1, we have applied the criteria above rather strictly, thus limiting the total number of schools under consideration. However, it should be pointed out that the label ‘bilingual school’ is often used by schools as a marketing tool even in cases that, under closer scrutiny, can more accurately be described as schools with intensive foreign language programmes, i.e. schools with programmes that exceed the Argentine national curriculum requirements of foreign language tuition. Although these distinctions are sometimes subtle and schools may (and do) switch from an intensive foreign language programme to a bilingual programme, it is important to note that not everything that is called ‘bilingual’ is in fact bilingual in the sense intended in this paper. In the case of the schools under scrutiny, we refer to institutions that, in general terms, match the criteria proposed by Johnson and Swain (1997) in their study of Immersion Education. In these cases, the L2 is used as a medium of instruction: the immersion curriculum parallels the local L1 curriculum; overt support exists for the L1; the programme aims for additive bilingualism; exposure to the L2 is largely confined to the classroom; students enter with similar and limited levels of L2 proficiency; (most) teachers are bilingual to some degree; and the classroom culture is that of the local L1 community. Even if the focus of our attention is restricted to those schools that comply with these criteria, the sector under study comprises over 150 schools that cater for thousands of children and adolescents. However, in spite of the size of this sector, there is surprisingly little published description or discussion Table 1 Argentine bilingual schools Language

Number of schools (approx.)

Location

Oldest school still in existence

English

100

Buenos Aires and other provinces

St Andrew’s Scots School 1838

No

French (in some)

Italian3

6

Buenos Aires

Colegio Tomas Devoto 1937

Yes

English

German

5

Buenos Aires

Goethe Schule 1897

Yes

English

French

2

Buenos Aires

Colegio Franco-Argentino 1945

Yes

English

Formal Other links with language(s) government taught of associated country2

The Evolution of Bilingual Schools in Argentina

67

regarding how these institutions originated and function. This obscurity can be ascribed to a number of factors, such as: the historical origin of these institutions; the fact that they stem from private initiatives; the relationship established with the public sector (e.g. state schools, universities, education authorities); and the curricular=extra-curricular division of their activities and staffing policies.

Similarities and Differences among Argentine Bilingual Schools A classification of bilingual education programmes, such as the one provided by Baker and Prys Jones (1998: 470), when applied to bilingual schools in Argentina, reveals a broad adhesion to a model of enrichment with the expected language outcome of bilingualism, biliteracy and biculturalism. This constitutes a strong form of bilingual education which aims to produce students who are proficient in two languages, biliterate in those two languages and familiar with both cultures involved. Traditionally, these schools have run parallel curricula, i.e. the Argentine and foreign curricula, to different extents. Typically (until a few years ago, when a trend of integration was initiated), the foreign curriculum was delivered as extra-curricular to the national Argentine requirements and, as such, was outside the supervision of the Argentine education authorities. However, it is the foreign curriculum, delivered in the foreign language, that is perceived by society as the most distinctive feature of the schools and this provides their competitive edge. All the schools under discussion are private, fee-paying institutions. Although some of them have close links with the government of the associated country, this only applies to a small minority of schools. These links often include the provision of teaching and managerial staff from those countries, as well as subsidies of different kinds. Another small group of schools also has foreign staff, not through links with a particular government, but by means of private contracts.4 The very nature of the curricular offering of all these schools makes bilingual staff indispensable, whether the members of staff are themselves bilingual, split into two groups of monolinguals, or some combination thereof. Another characteristic of bilingual schools in Argentina that is directly related to their sources of funding is the availability of material resources such as up-to-date technology (e.g. computers, science labs), and bibliographic (particularly imported) material. This creates a very real and, especially in the context of an economic crisis and currency devaluation, widening gap with state-funded schools. In spite of these similarities across schools, the bilingual school sector is by no means homogeneous and the distinctions cut across many of the categories discussed above. In some respects these differences are presented as polarities whereas others can be better understood as a continuum. Immediate distinctions can be made across schools when analysing some simple numerical facts. Firstly, they differ in terms of age. The oldest surviving bilingual school, St Andrew’s Scots School, was founded in 1838 (Escuela Escocesa San Andre´s, 1988), whereas many other bilingual schools were founded in the 1990s. There are also some schools which were founded some decades ago, and have only recently become bilingual. When, and in what circumstances, the schools were founded, has a direct influence on the traditions the schools draw on, and is

68

Bilingual Education in South America

also linked to another numerical factor: the size of the school. The student population may range from one hundred to two thousand. When founded, the schools often start out offering one or two levels (often kindergarten and primary) and build up from this foundation to cover the whole range of options, i.e. kindergarten, primary, secondary and even tertiary (teacher training). Furthermore, even though all these schools belong to the private sector and are fee-paying, there is great variation in fees across schools and, to a lesser extent, within schools across school years. These variations may be in the range of 300% across schools.5 This range of variation, in turn, has implications as regards the socioeconomic level of the student population because, although scholarships exist, they are few in number. In this respect, these bilingual schools can be said to constitute instances of elite bilingualism (see Banfi & Day, 2001; de Mejı´a, 2002). The schools also vary as regards their legal status. Some schools have the status of foundations or non-profit-making civil associations, i.e. charities, and are run by a board of trustees. Others are owned by an individual or a group of individuals in partnership. Others are legally-constituted companies or Sociedades de Responsabilidad Limitada (Limited Companies). There are confessional and lay institutions within this sector, but the latter are in the majority. As private educational institutions they are all regulated by the relevant section of the Ministry of Education, or equivalent, of their jurisdiction, i.e. the Provinces and City of Buenos Aires (see Federal Law of Education – Ley Federal de Educacio
n N
24195=93). Academically, the schools may vary in terms of the curricular options they offer. As regards the Spanish language curriculum, the schools are governed by the Argentine national curriculum requirements emanating from the Federal Law of Education, the Contenidos B
asicos Comunes produced by the Ministry of Education (1995a, 1995b), and curricular directives from their Provincial governmental jurisdictions. Regarding the curriculum delivered in the foreign language there are different options available and each school makes choices that constitute their institutional offering. In this respect, one cannot fail to mention the impact of international certification offered by examination bodies and the different associated governments. The trend among bilingual schools is to attempt to offer a varied and wide-ranging curriculum in the foreign language across school years. It should be pointed out that the larger and older bilingual schools usually provide more options. As mentioned above, another distinction across schools is the presence or lack of foreign contract staff. With few exceptions, the schools that hire foreign teachers either have foreign governmental support and=or more ample financial resources. It should be pointed out that teacher training in Argentina, although very progressive in the training of foreign language teachers,6 has not provided training options which could have, as an outcome, teachers specifically trained for teaching in bilingual schools. This did not pose major problems when the foreign language curriculum was delivered in what were officially considered extracurricular spaces. However, it does have considerable impact on the outcome of curriculum integration as we will see below. A possible explanation for the fact that little if any specific teacher training exists may be found in the fact that foreign language teacher training colleges

The Evolution of Bilingual Schools in Argentina

69

are overwhelmingly found in the public sector, whereas bilingual schools are in the private sector. The lack of specific teacher training is not a situation unique to Argentina, on the contrary, it is a pervasive problem as Baker (1995: 203) points out. Different permutations can thus be found when analysing the background of teachers teaching in bilingual schools. An ‘ideal’ profile might be a teacher who is qualified to teach in the language of instruction, e.g. he=she is a teacher of English, as well as in the academic discipline being taught, e.g. Physics. Individuals with a foreign language teacher’s degree who are graduates of teacher training colleges may be found teaching a number of different subjects, such as English Language, English Literature, History, Geography, etc. to children in the primary and secondary school. In some cases these choices are related to subjects that were, to some extent, covered in their training; in others, linked to other (partial) studies that the teachers may have undertaken, or some in-service training specifically provided by the schools (for example, see courses provided by ESSARP – The English-Speaking Scholastic Association of the River Plate7). There are also cases of schools that hire their own secondary school alumni to teach in the early years of primary schools; another option is hiring an alumnus who has gone on to study Law or Veterinary science, to teach History or Biology respectively. A component that is usually missing from the formal training of all these individuals is specific reflection on bilingualism and bilingual education and the teaching of content through the medium of a foreign language.8

The Evolution of Bilingual Schools: From Heritage to Dual to Global Argentine bilingual schools have undoubtedly changed over time, largely in response to changing social and economic circumstances, both locally and internationally. Here we present an outline of this evolution. The paradigmatic case is perhaps that of English–Spanish bilingual schools. We present the trends applying mainly to English–Spanish bilingual schools and leave open for future research the question of whether this analysis can be extended to the others. We propose that English–Spanish bilingual schools were initially founded as what may be labelled Heritage (or Community) schools in the sense proposed by Cummins (1995). As we have mentioned, the oldest surviving school is St Andrew’s Scots School (f. 1838), but there are reports of other schools that were founded at this time which did not survive (see Graham-Yooll, 2000: 133, 140, 178, 189). There are also official reports that allude to equivalent Italian and German schools (see Puiggr
os, 1996; Solari, 1972). This situation parallels what happened in the United States throughout the 19th century, when schools were organised by different ethnolinguistic groups (Perlmann, 1990). In Argentina, a considerable number of schools were founded around the turn of the century9 coinciding with the largest waves of European immigration arriving in the country, and with specific projects, such as the building of the railways, conducted by foreign commercial enterprise (Graham-Yooll, 2000: 268). These Heritage schools aimed at providing an education for the children of immigrants, in some cases for the immigrants themselves, and in others, those who were Argentine-born but had grown up within the immigrant community where the language spoken was not Spanish. The situation is parallel

70

Bilingual Education in South America

to that discussed in Garcı´a (1997) in the US context. These children grew up speaking English (or one of the other immigrant languages) in the home and were taught Spanish at school. Testimony to this situation is the fact that at St Andrew’s Scots School, tuition was initially conducted solely in English, and only in 1841 was the teaching of Spanish introduced (Escuela Escocesa San Andre´s, 1988: 36). The schools were set up by the community of immigrants, often linked to the community church, in response to an interest in providing their children with an education equivalent to the one they would have received ‘back home’, and often with the medium to long-term goal of returning home, in mind. In other cases (such as Northlands School and St Hilda’s College), the founders of the school were English governesses who had travelled to Argentina to teach the children of a wealthy family. Once the children had grown up, they remained in the country and were helped by their employers to found their schools.10 If one compares the number of immigrants from different countries with the number of bilingual schools corresponding to each community, we find little correlation, with certain languages over-represented, e.g. German; and others, virtually non-existent, e.g. Polish. This can be clearly seen if we compare the Italian and English communities. The number of Italian bilingual schools is negligible in relation to the fact that 39% of foreigners in Argentina in 1914 were of Italian origin, compared to only 1% who were of British origin. Currently there are only six Italian schools and 100 English–Spanish bilingual schools (for details of number of immigrants in Argentina in 1914 see Solberg, 1970: 38). In this respect, we could consider the situation as a reflection of a hierarchy of languages: certain languages have had a much better chance of achieving the status required to have bilingual programmes, whereas others have not. (See Verma et al., 1995 on immigrant languages in the UK; and Helot & Young, 2002 for the case of France.) As the flow of immigration began to slow down in the course of the first half of the 20th century, the number of Argentine bilingual schools did not decrease, as might have been expected. On the contrary, the number of schools (particularly English–Spanish bilingual schools) increased. Their student population changed significantly, however. The situation reported at St Andrew’s Scots School is a case in point: in 1939, out of 310 pupils enroled, 21 were born in the British Empire, 66 were born in Argentina to two British parents, 68 were born in Argentina to one British parent and the remaining 155 were other nationalities, including Argentine (out of these, two-thirds were English-speaking, with British grandparents). As these figures reveal, the school had started incorporating non-English-speaking children and, at the same time, was experiencing loosening ties with the close-knit, homogeneous founding community, altered by intermarriage and generational separation with the original immigrants (Escuela Escocesa San Andre´s, 1988: 76). As they evolved, Argentine bilingual schools had to adapt to the reality of this changing student population and situation. In some respect, this stage can be described in terms of Dual Language Programmes (see Lindholm-Leary, 2001) in the sense that there was a combination of language minority and language majority children in different proportions at different stages, but gradually tending towards an increase in majority language children.

The Evolution of Bilingual Schools in Argentina

71

Complementing this demographic change affecting Argentine bilingual schools during the first half of the 20th century was the impact of the growth and development of the national education system. Garcı´a (1997: 408) discusses a similar development in the US. Law 1420 was passed in 1884 and is a landmark in the process of organising the Argentine education system, which gathered momentum around the turn of the century. The development of the state’s educational policies implied a greater degree of supervision which was felt in bilingual schools, and the number of hours of tuition in Spanish had to be increased. At this stage, the pattern of the foreign and national curricula run side by side was set. The official supervision and review were concerned with the Argentine Spanish-language curriculum only. This situation encouraged the development of parallel Foreign and Spanish language structures within the schools, particularly at managerial levels, since the Heads and other school authorities had to comply with national requirements in terms of qualifications, requirements that did not apply to the curriculum delivered in the foreign language. In contrast, the delivery of the foreign language curriculum allowed for considerable flexibility in terms of hiring foreign teachers or individuals with different qualifications to teach the non-Spanish curriculum. As time passed, the situation that saw the schools shift from a Heritage to a Dual-language model was intensified by the decline of European immigration and the increase in the number of bilingual schools. Thus, by the end of the 20th century, most bilingual schools catered for children whose home language was Spanish and who learnt the second language at school from scratch. At present, only a few schools have a small proportion (5% at most) of foreign children who do not speak Spanish. In general, these children are members of internationally mobile families and their parents usually have short (e.g. threeyear) temporary contracts in Argentina. In some of these cases, the language of the home is neither language of the bilingual school, but a third language (e.g. French, Dutch, German). There is evidence to suggest that English is the language of choice of parents for their children’s education, even if it is not the language of the home, because it is regarded as more easily available as a language of instruction in possible future destinations as well as a language of opportunity for the future development of those children (Banfi & Day, 2001). The number of internationally mobile families increased substantially during much of the 1990s, a period of rapid growth for Argentina as trade opened up and large-scale privatisations led to significant increases in foreign investment, foreign ownership of Argentine companies and the transfer of foreign personnel to Argentina on temporary contracts. This trend decreased sharply in 2001 and reversed in 2002 as the Argentine economy contracted. A further trend that can be observed in the evolution of schools can be found in the adaptability shown. For example, the geographic distribution of schools has been greatly influenced by their following the movement of the student population. Already, as early as 1944, this concern is evident when St Andrew’s Scots School created a committee to study the best location where a high concentration of English-speaking people were to be found and to organise the relocation of the school to this area. This study resulted in a move

72

Bilingual Education in South America

of the school to Olivos, a northern suburb of the city of Buenos Aires, and led in turn to a pattern of similar moves by other schools (Escuela Escocesa San Andre´s, 1988: 86). A related phenomenon is what we might label ‘branching out’. Probably one of the first schools to pursue this policy was St George’s College. As reported in St George’s College (1998: 95), when the school was founded in 1898 it was located in a southern suburb of Buenos Aires called Quilmes. This was related to the student population who came from that area, since a sizeable number of English-speaking immigrants had settled there in connection with the development of the railway (Graham-Yooll, 2000: 268). Nearly a century later the school board decided to establish a branch, St George’s North, in Polvorines, a northern suburb of Buenos Aires, closely linked to a property development where the families of potential pupils had moved and were in the process of moving. The development of private estates or ‘gated communities’ in the suburbs of Buenos Aires throughout the 1990 s led to a parallel expansion of branches of existing and prominent bilingual schools to service the new and prosperous suburban populations. The presence, or planned presence, of the school also enhanced the value of the property market in the development.11 Another way in which schools have adapted has involved moving from single-sex to co-educational provision. Although a few retain their singlesex admissions policy, some of the most traditional single-sex schools have felt the need to adapt to changing demands.12

Bilingual Schools Now At this stage in their evolution, we might characterise Argentine bilingual schools as going through a ‘global language’ period or as ‘Global Language Schools’ (Banfi, 2001), meaning that many of the key features that characterise Argentine bilingual schools now may be connected with the processes of globalisation that have developed over the last decade. Whereas in the Heritage Schools stage, cultural roots were clearly identifiable and connected to particular traditions, these links were weakened as the schools evolved and the student population and context changed. This did not mean that the links were severed. On the contrary, they became even more central and were greatly facilitated by increasing international travel and developments in communications technology.13 The change was in the purposes and diversity of the links that imprinted an international outlook on the educational offering of the schools. The schools now aim to provide a broad educational programme that prepares children to take part in the decision-making stratum of a globalised world (see for example, Belle-Isle, 1986; Gellar, 1981). This international perspective has ramifications for the curricular decisions made by the schools. A further aspect of this development is the impact of English as a global language (Graddol, 1997). On the one hand, this partly explains the proliferation of English–Spanish bilingual schools as the perceptions of the importance of English spread across the society. On the other hand, we can see the impact of English on the other bilingual schools where English is included in the curriculum and presented as an important complementary aspect of school’s programme.

The Evolution of Bilingual Schools in Argentina

73

A key development at this stage is the greater move towards the integration of the local and foreign or international curricula, formally sanctioned in official resolutions, such as those of the City (Resoluci
on 2008, 1999) and Province of Buenos Aires (Resoluci
on 7102, 1996). These instruments make it possible to avoid some degree of duplication and overlapping present in the parallel curricular situation, when, for example, subjects like Mathematics and Chemistry were taught in both Spanish and the second language. The authorisation to integrate and the concomitant recognition of bilingual status make it possible to reduce, to some extent, what used to be a long school day. Another side-effect of integration is the possibility of reducing teaching staff numbers, always bearing in mind that this integration imposes official requirements on integrated subjects, thus posing new recruitment challenges. Complementing these developments is the increase in the range of curricular offerings bilingual schools now give their pupils. This is particularly evident in the older and larger bilingual schools which are able to offer choices of subjects that permit a degree of specialisation, especially in the last years of secondary school. Often these choices are conditioned by another development that has increased in relevance: foreign certification. Almost without exception, bilingual schools in Argentina offer some kind of foreign certification, particularly in terms of school-leaving certification. This foreign certification is awarded by international examination bodies (e.g. UCLES – the University of Cambridge Local Examinations Syndicate, and the IBO – the International Baccalaureate Association) or foreign governments (e.g. Abitur in German schools, Diploma di Maturita
Scientifica in Italian institutions). These certifications cover language ability accreditation and, more significantly in the case of bilingual schools, subject-specific accreditation at different levels. Two main bodies provide subject-specific accreditation. UCLES has been providing examination services in Argentina since the beginning of the 20th century,14 specifically it is Cambridge International Examinations (CIE)15 within UCLES that administers examinations of the kind we are concerned with here. Apart from Language and Literature papers, CIE examinations are written in English. There are two main levels at which they are taken; the International General Certificate of Secondary Education and the Advanced International Certificate of Education (IGCSE and AICE). Currently there are around 150 schools in Argentina that offer IGCSE certification and around 35 that offer AICE.16 Another body that offers subject-specific foreign certification is the IBO.17 In 2002 there were 45 schools in Argentina offering the IB Diploma.18 Not all these are bilingual schools, as some institutions offer the Diploma in Spanish or in French (the other two official IB languages).19 Part and parcel of the process of globalisation is the expansion and accessibility of information technology. This development has had a considerable impact on education in general. In the case of Argentine bilingual schools, the availability of resources to purchase technology and the schools’ relative autonomy have implied a significant change as regards the use of technology. This use applies both at the organisational level (communication of management with staff, of schools with parents, information management within the school, etc.) and in the realm of pedagogical activities. The ready

74

Bilingual Education in South America

availability of technology has made it possible to aim for a cross-curricular application of IT. To summarise, we can say that, as a general principle, Argentine bilingual schools aim to provide a broad educational experience with the individual aim of bilingualism=biculturalism and the concomitant societal aim of pluralism and enrichment. In this respect they constitute strong forms of bilingual education.

Conclusion This paper has attempted to provide a preliminary characterisation of what are popularly known as colegios bilingu¨es, i.e. bilingual schools, in Argentina. As we have shown, this sector presents significant similarities and, at the same time, a pattern of diversification across schools. The main shared feature is the teaching of content through the medium of the foreign languages, with features that resemble those proposed for Immersion Education by Johnson and Swain (1997). We have also argued that Argentine bilingual schools constitute a form of e´lite bilingualism as proposed in the definition given by de Mejı´a (2002). We have proposed a characterisation in terms of stages of evolution of the schools that are mainly related to the demographic changes in the student population and general societal changes. These stages have borrowed from models of bilingual education programmes proposed for other contexts, i.e. Heritage Schools and Dual Language Schools, and a new label has been proposed to describe schools as they are now: ‘Global Language Schools’. These stages are linked to the fact that Argentine bilingual schools have changed considerably over time. Originally, they were schools established by the immigrant communities which had maintained very close links to them (board members, investors etc.), but over the years they have seen a considerable weakening in their connections with these communities. Paralleling this process has been the changing nature of the immigrant or foreign communities, especially in their size. The schools have responded to these demographic changes by incorporating children from outside the original immigrant community, for example, and by changing the languages offered as the student population has altered; from initially Spanish as a Foreign Language to a foreign language, such as English or French as a Foreign Language. To this extent the schools have become more ‘Argentine’ and have repeatedly demonstrated their great ability to change and adapt according to demographic and curricular trends, and cultural changes in the community they serve. Connected also with this process of constant adaptation are idiosyncracies in the development of the school curriculum. A central or unified way of organising the curriculum is often absent. It is influenced by multiple factors. Certain areas of the curriculum are defined by the jurisdictional prescription (i.e. the national and=or provincial educational authorities). Other parts are defined by what one, or more, international body prescribes, as is the case with the influence of international examining bodies. Yet other parts are defined by ad hoc decisions made by Heads, Coordinators and Teachers. As staff in these positions fluctuate, these curriculum areas may be unstable and subject to constantly changing priorities. Always present is the reality that, as private sector

The Evolution of Bilingual Schools in Argentina

75

competing institutions, the decisions Argentine bilingual schools make about staffing, resources and the curriculum are also linked to the schools’ efforts to market themselves. As a consequence of the influences and changes referred to above we conclude that the bilingual school sector in Argentina is richly varied. Even though the Colegios Bilingu¨es are perceived as being a homogeneous group based upon key similarities, they are in fact highly diversified (in size, in curriculum, in age and traditions, in staffing, in resources, and in connections with an associated country. We also suggest that self-analysis on the part of Argentine bilingual schools, as institutions, is uncommon. Their role as institutions has not been examined systematically or analysed. Members of these institutions themselves (be they teachers, administrators, or members of governing bodies) do not reflect upon themselves and their activity in a structured fashion, and, even though there is a variety of informative and promotional published material, there is a notable absence of analytical publications about the schools. Therefore, there is considerable scope for further research in this sector in the light of increasing curricular complexity and diversity and the need for teacher development programmes that deal with the specific needs of teachers working within the field. Acknowledgements This article has benefitted from discussion with audiences at the 46th Annual Conference of the International Linguistics Association at New York University in 2001 and the Third International Symposium on Bilingualism at the University of the West of England in Bristol in 2001. We would also like to thank two anonymous reviewers for their very helpful comments and suggestions. And last but certainly not least, our thanks to the editor of this volume for her commitment to this innovative project. Needless to say, any errors still contained in this paper are our sole responsibility. Correspondence Any correspondence should be directed to Dr Cristina Banfi, ESSARP Centre, Esmeralda 672 piso 7, C1007ABF, Buenos Aires, Argentina (cbanfi@ essarp.org.ar). Notes 1. In fact, in the layman’s application of the term, some programmes that could be seen as falling into this category, such as programmes for Indigenous Languages or bilingual education for the Deaf, are not really included. 2. We employ the term ‘associated country’ to refer to the country that a bilingual school may consider its linguistic parent (in terms of the foreign language the school uses as a medium of instruction). This country may, for example, have been the place of origin of the immigrant community that founded the school. Alternatively it may be a country that the school considers itself most culturally, as well as linguistically, bound to outside Argentina. 3. In a recent update of the Italian Consulate website in Buenos Aires (www. consitalia-bsas.org.ar=espanol=scolastiche.htm) only four Italian schools are listed catering for 2225 pupils and with a staff of 309 teachers. These four schools have ‘ministerial’ teachers (22 of them in four schools), i.e. teachers employed by the Italian

76

4.

5.

6. 7. 8.

9. 10. 11. 12.

13. 14. 15. 16. 17. 18.

19.

Bilingual Education in South America government to teach at these schools. The total number of Italian schools is slightly higher if we take into account those schools without these direct diplomatic links. This trend, along with others in relation to bilingual schools, has been considerably affected by the economic crisis in Argentina, particularly as a consequence of the devaluation of the Argentine peso in 2002, which has made it virtually impossible for schools to be able to offer internationally competitive salaries. For some indicators of yearly fees of all private schools in the City and Province of Buenos Aires, see reports prepared by the Argentine Ministry of Economy. www.minproduccion.gov.ar=secdef=basehome=precios.htm, access 12=09=02. Note that this list includes non-bilingual schools. As early as 1904, an institution was founded specifically to train foreign language teachers; this institution is the Instituto de Ense~ nanza Superior en Lenguas Vivas ‘Juan Ram
on Ferna˜ndez’, (Vanbiesem de Burbridge & Bruno de Zamacona, 1982). www.essarp.org.ar, access 14=09=02. Some attempts have been made to address this problem within ESSARP courses, for example: ‘Understanding Bilingualism: from home to school and society’ course taught by Christine Helot in 2002. www.essarp.org.ar=course2002.html, access 14=09=02. ‘Introduction to Bilingualism and Bilingual Education’ course taught by Cristina Banfi in 2002. www.essarp.org.ar=course2002.html, access 14=09=02. ‘Bilingual Education in Primary Schools in Alsace, France’ course taught by Christine Helot in 2001. www.essarp.org.ar=course2001.html, access 14=09=02. ‘An Overview of Immersion Education’ course taught by Myriam Met in 2001. www.essarp.org.ar=course2001.html, access 14=09=02. For example. Barker College (f. 1897), St George’s College (f. 1898), Villa Devoto (f. 1908), Goethe Schule (f. 1897). See Graham-Yooll (2002) on itinerant teachers as well. Other examples can be found, such as the Northlands School extension to Nordelta: a new development north of Tigre (to the North of Buenos Aires) emulates similar moves by schools like St Catherine’s-Moorlands. Cardinal Newman College still admits boys only and Michael Ham Memorial College continues to admit girls only. Conversely, Northlands School, a traditional girls’ school is in the process of becoming co-ed, having opened its primary school to boys in 1999. The international links of the schools have specific concrete events and programmes. These include examples of exchange programmes of staff and students, sports and cultural trips. St Andrew’s Scots School reports its preparation of pupils for Cambridge Local Examinations as early as 1909 (Escuela Escocesa San Andre´s, 1988: 62). See www.cie.org.uk for a general overview of certification offerings. Litty Mora. Argentina Representative for Cambridge International Examinations, Personal Communication. See www.ibo.org for more information on curricular offerings. The IBO website shows 55 Argentine schools offering their programme but some schools appear more than once if they offer more than one of the three IBO programmes (the Primary Years Programme, the Middle Years Programme, and=or the IB Diploma). When we control for this factor we find that there are approximately 45 IB diploma schools in Argentina. For a broader overview of IB penetration into South America throughout the 1980s and 1990s see Fox (1998: 237). Although the IB listing provides the languages of instruction used in each school there are some inaccuracies.

References Baker, C. (1995) A Parents’ and Teachers’ Guide to Bilingualism. Bilingual Education and Bilingualism. Clevedon, Multilingual Matters. Baker, C. and Prys Jones, S. (1998) Encyclopedia of Bilingualism and Bilingual Education. Clevedon: Multilingual Matters.

The Evolution of Bilingual Schools in Argentina

77

Banfi, C. (2001) From heritage to global languages: An historical overview of bilingual education in Argentina. Paper presented at the 46th Annual Conference, International Linguistic Association. Banfi, C. and Day, R. (2001) The Argentine chicken and the English egg: Bilingual schools and the itinerant elite. Paper presented at the 3rd International Symposium on Bilingualism, University of the West of England, Bristol. Belle-Isle, R. (1986) Learning for a new humanism. International Schools Journal 11, 27–30. Consejo General de Cultura y Educaci
on, Direcci
on general de Cultura y Educaci
on, Provincia de Buenos Aires. Resoluci
on 7102, 25th November 1996. Cummins, J. (1995) Heritage language teaching in Canadian schools. In: O. Garcı´a and C. Baker (eds) Policy and Practice in Bilingual Education: Extending the Foundations (pp. 134–138). Clevedon: Multilingual Matters. Escuela Escocesa San Andre´s (1988) Un Siglo y Medio Despue´s: Escuela Escocesa San Andre´s 1838–1988. Olivos. Fox, E. (1998) The role of the International Baccalaureate in educational transformation: Chile as a case study. In M. Hayden and J. Thompson (eds) International Education: Principles and Practice (pp. 235–241) London: Kogan Page. Garcı´a, O. (1997) Bilingual education In F. Coulmans (ed.) The Handbook of Sociolinguistics (pp. 405–420). Oxford: Blackwell. Gellar, C. (1981) International education: Some thoughts on what it is and what it might be. International Schools Journal 1, 21–26. Graddol, D. (1997) The Future of English. London: The British Council. Graham-Yooll, A. (2000) La Colonia Olvidada. Buenos Aires: Emece´. Helot, C. and Young, A. (2002) Bilingualism and language education in French primary schools: Why and how should migrant languages be valued? International Journal of Bilingual Education and Bilingualism 5 (2), 96–112. Johnson, R.K. and Swain, M. (1997) Immersion Education: International Perspectives. Cambridge: Cambridge University Press. Lindholm-Leary, K. (2001) Dual Language Education. Clevedon: Multilingual Matters. Mejı´a, A.M. de (2002) Power, Prestige and Bilingualism: International Perspectives on Elite Bilingual Education. Clevedon: Multilingual Matters. Ministerio de Cultura y Educaci
on; Consejo Federal de Cultura y Educaci
on (1995a) Contenidos B
asicos Comunes para el Nivel Inicial. Buenos Aires: Ministerio de Cultura y Educaci
on. www.mcye.gov.ar=consejo=documentos=index.html, access 15=09=02. Ministerio de Cultura y Educaci
on; Consejo Federal de Cultura y Educaci
on (1995b) Contenidos B
asicos Comunes para la Educaci
on General B
asica. Buenos Aires: Ministerio de Cultura y Educaci
on. www.mcye.gov.ar=consejo=documentos=index.html, access 15=09=02. Ministerio de Cultura y Educaci
on; Consejo Federal de Cultura y Educaci
on (1997) Contenidos B
asicos Comunes para la Formaci
on Docente de Grado. Buenos Aires: Ministerio de Cultura y Educaci
on de la Naci
on. www.mcye.gov.ar=consejo=documentos= index.html, access 15=09=02. Perlmann, J. (1990) Historical legacies: 1840–1920. The Annals of the American Academy of Political and Social Science 506, 27–37. Puiggr
os, A. (1996) Sujetos, Disciplina y Currı´culum: En l os Orı´genes del Sistema Educativo Argentino (1885–1916) (3rd edn). Buenos Aires: Galerna. Repu´blica Argentina (1993) Ley Federal de Educaci
on No 24195=93. www.me. gov.ar=leyfederal, access 14=09=02. Secretarı´a de Educaci
on, Gobierno de la Ciudad de Buenos Aires. Resoluci
on 2008, 16th November 1999. Solari, M.H. (1972) Historia de la Educaci
on Argentina. Buenos Aires: Paidos. Solberg, C. (1970) Immigration and Nationalism: Argentina and Chile, 1890–1914. Austin: Published for the Institute of Latin American Studies by the University of Texas Press. St George’s College (1998) The Centenary Book. Buenos Aires.

78

Bilingual Education in South America

Vanbiesem de Burbridge, M. and Bruno de Zamacona, M. (1982) El Lenguas Vivas: Una Tradici
on Educativa. Buenos Aires: Instituto Nacional Superior del Profesorado en Lenguas Vivas ‘Juan Ram
on Fern
andez’. Verma, M.K., Corrigan, K.P. and Firth, S. (eds) (1995) Working with Bilingual Children. Clevedon: Multilingual Matters.

English Immersion in Paraguay: Individual and Sociocultural Dimensions of Language Learning and Use Susan Spezzini University of Alabama at Birmingham, AL, USA This investigation of English immersion learners explored processes of learning English, patterns of language use, and levels of comprehensibility among 34 predominantly Spanish-speaking 12th graders at the American School of Asuncion in Asuncion, Paraguay. The analysis of data from language learning histories, group interviews, perceived comprehensibility ratings, and questionnaires about language use led to the identification of linguistic dimensions affecting comprehensibility and of individual and sociocultural dimensions related to language learning and use. When judging comprehensibility, the native-speaker raters were influenced mainly by prosodic elements and overall fluency. Higher comprehensibility corresponded to girls as well as to students with some schooling in English language contexts, aspects of integrative motivation and more frequent L2 use. Findings were congruent with studies in immersion education, second language acquisition, sociolinguistics and social psychology. Keywords: comprehensibility, variability in language learning and use, immersion education, English immersion, gender, identity

Introduction This paper draws from a large study that explored processes of learning English, patterns of language use and levels of comprehensibility among English immersion learners at the American School of Asuncion (ASA), a dual language American overseas school located in Asuncion, Paraguay (Spezzini, 2002). Guided by the bilingual voice of 34 predominantly Spanish-speaking 12th grade students, of whom 76% were Paraguayan mostly of high socioeconomic status (SES), this explanatory account of variability identified linguistic dimensions affecting comprehensibility and described individual and sociocultural dimensions related to language learning and use. Qualitative and quantitative data were collected from language learning histories, group interviews, perceived comprehensibility ratings and questionnaires on selfreported language use and self-assessed language proficiency. Theoretical insights came from the fields of immersion education, second language acquisition, sociolinguistics and social psychology. In the final analysis, this study was conducted to provide information for establishing language learning objectives, improving content-based L2 instruction, and providing ongoing staff development. Such information should prove useful not only for ASA and other American overseas schools (AOSs)1 with mainly non-native English-speaking populations, but also for similar elite bilingual schools, especially in Latin America.

79

80

Bilingual Education in South America

Background Information ASA is one of over 500 AOSs worldwide conducted mainly in English (Rabbitt, 1994). Started by English-speaking expatriates for diplomatic, business and missionary purposes, many AOSs now have a majority of host country nationals (HCNs) and third country nationals (TCNs), whose main goal is to become proficient in English (Baker, 1995; Sears, 1998). Even if the term International or American appears in a school’s name, this ‘does not necessarily connote that its student body is either international or American’ (Rabbitt, 1994: 10). In fact, most AOSs in Latin America have significantly high populations of HCNs, who remain together for 14 years – from preschool through high school graduation (ISS, 1994; Rojas, 1999a). Because of their almost exclusive HCN student bodies, many AOSs in Latin America qualify as immersion schools, where language learners ‘develop English proficiency through a rigorous academic program designed for English mother-tongue students’ (Rojas, 1999b: 2). Although AOSs provide a unique context for successful language learning, especially among young learners, several schools struggle with language policies and practices, staff untrained in immersion education, myths about bilingualism and comparisons between emerging bilinguals and English monolinguals (Rojas, 1999a). Yet, in spite of these linguistic concerns, most AOS studies have addressed administrative or cultural issues, especially as related to third-culture students. Although some studies have analysed language programmes, from the organisation of ESL instruction (Orr & Conlan, 1982) to a review of language issues in diverse settings (Carder, 1995), few have examined language learning much less from the students’ perspective. Most AOS research on language learning has been conducted in connection with the European Council of International Schools (see www.ecis.org). Of these studies, Carder (1993) questioned whether students are being schooled without a firm foundation in either L1 or L2 and recommended improving bilingual education by placing experienced EFL teachers in pivotal positions. In the Latin American context, McKay suggested that perhaps ‘the economic rewards for knowing English are great enough to provide an important source of instrumental motivation’ (1992: 26), as illustrated by how parents register children at birth for coveted positions at elite English=Spanish schools. McKay’s description of institutional varieties within diverse EFL settings, including the AOS context, was supported by findings from two language studies at ASA (Boxer, 1999; Rojas, 1999b). In these studies, the students’ English was described as impressively high in cognitive academic language proficiency but low in basic interpersonal communication skills, with the opposite being true of their Spanish. According to Boxer, ‘Increasingly as they go higher in grade level, students appear to fossilize in some type of non-native pronunciation that is neither typical of Paraguayan English nor akin to native-like use’ (1999: 4). In addition to this need for language learning studies at AOSs, a need also exists for more research about language learners and L2 variability in all contexts. Met and Galloway (1996) call for additional research about intralearner variability, which they assert will inform other language learning issues. In

English Immersion in Paraguay

81

his call for more studies about language learning variability, Ellis (1994) wonders whether ‘learners who regularly converge towards target-language norms in their interactions with native speakers ultimately achieve high levels of L2 proficiency’ (1994: 236). He proposes naturalistic methodology for examining the dynamic interactions between learners’ individual differences and their learning experiences. Focusing directly on immersion education, Swain and Johnson (1997) question whether the students’ non-target-like proficiency is perhaps an inevitable outcome. De Courcy (2002), on the other hand, follows a qualitative approach in order to examine learners’ experiences and learning processes during late immersion. The current study at ASA responds to these calls for research in the following areas: language learning in the AOS context, L2 variability and immersion learners. This study classifies all ASA students as language learners, ranging from beginners in L2 to balanced bilinguals developing increasingly more complex degrees of bilingualism. Language proficiency is viewed as an ultimate level of attainment within a range of multicompetence (Cook, 1994; Saunders, 1988) and successful language learning as the learner’s ability for using language in a communicatively competent manner via culturally appropriate discourse patterns (Baker, 1995; Rutherford, 1987; Ur, 1998).

Methodology This study was designed to understand a societal concern – the spoken English of ASA students. Instead of seeking evidence as an end in itself or determining the effectiveness of a given intervention, this study examined questions deemed important by a given society – the ASA community. Although this concern was also of importance to immersion education and related fields, it arose primarily from within the context itself thereby classifying the study as applied research (Patton, 1990). Furthermore, because of its inherent focus on searching and examining as opposed to positing and proving, this study was aimed at generating rather than testing hypotheses (Brause & Mayher, 1994). Setting ASA is located in Asuncion, the capital city of Paraguay.2 Founded in 1956, ASA is a private nonsectarian school with approximately 600 students from pre-kindergarten through 12th grade.3 Although ASA serves dependents from the United States Embassy, as well as from other embassies and multinational firms, most students are the children of Paraguay’s political, economic, professional and cultural elite. These families live near ASA in high SES neighbourhoods and can afford to pay the highest tuition in Paraguay. ASA’s student body consists of approximately 13% dependents of US citizen(s), 60% HCNs and 27% TCNs (ASA, 2001b). English-speaking transfer students are admitted at all grade levels on a space-available basis. NonEnglish speaking students can only seek admission at age four into pre-kinder, but, as of 1992, their admission is contingent upon having first attended an English-medium nursery school. In spite of the school’s international aspect, over 90% of ASA students are native or near-native speakers of Spanish.

82

Bilingual Education in South America

This language situation is due not only to the overwhelming majority of Paraguayans among ASA’s student body, but also to the fact that many US students are dual nationals (American=Paraguayan), and most TCNs are permanent residents from neighbouring countries or immigrants from Asia. ASA is composed of an elementary school (pre-kinder – Grade 5), middle school (Grades 6–8), and high school (Grades 9–12). Class size is limited to 23 students. The school day runs from 8:00 a.m. to 3:30 p.m. and the school year from early August to late June. ASA’s college preparatory curriculum is accredited by the Southern Association of Colleges and Schools and by Paraguay’s Ministry of Education. About 95% of ASA graduates receive a Paraguayan diploma in addition to their US diploma. All subjects are taught in English except for Spanish language arts, Hispanic social studies, and Guarani,4 as well as high school electives in French and Portuguese. English-medium subjects are taught by US-certified teachers, and other subjects are taught by teachers certified in Paraguay. Although music, art and physical education are now in English, they were taught mainly in Spanish to the students in this study. The distinction between second and foreign language is unclear at ASA, resulting in a hybrid ESL=EFL situation (Green & Oxford, 1995). Since almost all students (82% in this study) share the same home language and since Spanish is also the main social language in both school and community, ASA’s social context fits the EFL definition. However, with respect to its academic context, ASA resembles ESL: All professionals who teach in English are native speakers of English, and all textbooks for content instruction are the same as those used in mainstream US classrooms. ASA students experience additive, elite, sequential, childhood bilingualism (Baker, 1995; Grosjean, 1999; Paulston, 1999). Although the US and Paraguayan programmes did not become merged into one integrated curriculum until 1996, bilingualism and biculturalism have been officially recognised at ASA since at least its 1980 statutes. According to ASA’s Student – Parent Handbook, ‘All students share the goal of becoming bilingual and do so through a bilingual integrated program’ (ASA, 2002, para. 3). The school’s faculty handbook provides guidelines for language use: ‘Be a good language model for students. If you teach in English, speak to your students in English outside of the classroom as well’ (ASA, 2001a: 9). Additional guidelines are provided to ASA’s secondary staff: ‘Teachers must require that the language of instruction be the language of communication for all purposes in the specific learning environment. Teachers must emphasize with students the importance of practice in oral communication at all times’ (2001a: 13). With respect to the different types of immersion education, ASA can be categorised as foreign language immersion within an international context (Lyster, 1999). All of Swain and Johnson’s (1997) immersion descriptors apply to ASA except for two: many ASA teachers are not bilingual in the school’s target languages and ASA’s curriculum does not parallel Paraguay’s but rather that of the US. Inasmuch as ASA is a programmatic variation of immersion education, it is similar to but yet different from several of the prototypical immersion models. For the majority of ASA students, their schooling experience most closely

English Immersion in Paraguay

83

resembles that of the Canadian model – early total immersion in L2 (Baker, 1993). However, ASA differs from Canadian immersion with respect to the introduction of L1.5 At ASA, Spanish instruction begins in kindergarten, possibly to counterbalance often minimal L1 literacy activities at home, a necessary condition for introducing literacy exclusively through L2. Not only is there a relatively undeveloped children’s literature tradition in Paraguay, but a caregiver role is often assumed by domestic employees, primarily Guarani speakers with limited schooling. Some ASA students experience double immersion, middle=late immersion, and=or two-way immersion. Double English=Spanish immersion is experienced by TCNs who have other home languages (mainly Chinese, Korean and Portuguese). ASA transfer students often experience middle or late Spanish immersion. Two-way immersion corresponds to the school’s kinder and 1st grade classrooms that have occasional monolingual English students and to the Spanish-medium classes at the upper grade levels with mainstreamed SSL students. The greatest difference between ASA and the classical twoway immersion model resides in the highly unequal number of native speakers for each target language. Participants Based on purposeful sampling (Patton, 1990), all students in both 12th grade English classes were targeted as potential participants. The other study participants included the 12th grade English teacher and the researcher as participant=observer. The students, their parents and the teacher provided written consent. Their identities remained concealed, and strict confidentiality was maintained. At the time of this study, the teacher participant was teaching 11th and 12th grade English as well as 12th grade philosophy. He was a US citizen, had advanced degrees in languages, and had taught for about 30 years, of which three were at ASA. He was interested in the school’s language situation and agreed to integrate research-related activities into his lessons. Based on Eckert’s (1989) research of high school students, this study also selected students in their final year of schooling. The 12th graders were about to embark into the real world and had assumedly reached a certain level of maturity for reflecting on language. At the start of this study, the students’ mean age was 18 years 2 months. Of these 34 students (16 girls and 18 boys), 74% were Paraguayan, and 82% had Spanish as their first language. For analysis purposes, these 34 students were divided into the following three groups: the K group for those who had entered ASA in pre-kinder; the PT group for Paraguayan transfer students; and the NT group for other nationalities, of whom all were transfer students. The K group consisted of 14 students (nine female and five male). All were Paraguayan, had Spanish as their home language, and had been schooled at ASA from pre-kinder through 12th grade. The PT group consisted of 11 Paraguayans (two female and nine male) who had transferred to ASA between first and tenth grade. Prior to entering ASA, eight of these PT students had attended other AOSs for periods ranging from 2–8 years – five students in Paraguay and three elsewhere (Japan, Egypt, and

84

Bilingual Education in South America

Italy=Syria=Honduras). Another three had attended stateside schools for periods ranging from three months to two years. Spanish was the home language for all 11 PT students. As their first schooling language, five reported English and six Spanish. The NT group contained nine students (five female and four male) of various nationalities: two US, three Chinese, and one each from Brazil, Bolivia, Nicaragua, and Korea. These NT students had transferred to ASA between first and tenth grade from the following settings: four from stateside schools, two from another AOS in Paraguay, two from bilingual schools in Ecuador and Chile=Argentina, and one from a Spanish-medium school complemented by five years of private EFL tutoring. They had been schooled in these other contexts for periods ranging from 1–8 years. The home languages of these NT students were: one English, three Spanish, one Portuguese, three Chinese and one Korean. Their first schooling languages were: four English, three Chinese, one Spanish, and one Korean. Procedure A preliminary 10-day data collection period began on 23 February 1999, at the start of ASA’s second semester and one week following its 2-month summer vacation. The most intensive data collection period ran from 24 May to 17 June, ending two weeks before graduation. Such timing allowed students to think about these issues upon starting their last semester in high school and then to reflect more deeply just before graduating. The collection of speech samples was determined by the dates for the school’s annually recurring senior speeches. Follow-up data collection took place one year later in June 2000 following a school assembly that featured recent graduates. Multiple sources provided qualitative data. In their language learning histories, students responded to one of the following topics: How I Learned English, My Language Learning Autobiography, Why Teenagers Speak the Way We Do, What Makes our ASA Talk so Unique. In both sections, two class discussions were taped: the first one as a pre-writing activity in preparation for the language learning histories and the second as a type of member check with feedback from the preliminary data analysis. There were 11 focus group interviews: two in March, eight in June, and one the following year. Each of these interviews had from four to five students, lasted about 20 minutes, and was taped and transcribed. Quantitative data were collected from a questionnaire of self-reported language use and self-assessed language proficiency (see Baker & Jones, 1998: 87; Spezzini & Oxford, 1998) and from two comprehensibility rating tasks, which were designed with input from Murray Munro. The students’ spoken English was quantified via a comprehensibility score based on how native-speaker raters perceived ease of interpretation (Munro & Derwing, 1995, 2001). In creating the comprehensibility rating tasks, taped samples were selected from the students’ senior speeches, an annually recurring ASA event.6 All statistical analyses were performed on SPSS1 9.0 for Windows (SPSS, 1999). Descriptive statistics included mean, standard deviation, range and skewness. Inter-rater reliability was established on the Guttman, parallel and split scales that produced an average intraclass correlation coefficient of

English Immersion in Paraguay

85

0.955 (95% confidence and df ¼ 33).7 Because of threats to internal validity, the scores of four transfer students (one girl and three boys) were eliminated from these statistical analyses. In an effort to discover potential bases for perceiving comprehensibility as easy or impossible to understand, a close linguistic analysis was conducted of six speech extracts used in the rating tasks. Because of the specific concern with the spoken English of students schooled exclusively at ASA, the selected samples were the three K students with the highest comprehensibility and the three with the lowest. Transcription notations were adapted from Schiffrin (1994) and also from Gass et al. (1999), Leech and Svartvik (1994), Pennington (1996) and Stockwell and Bowen (1965).

Analysis of Results By analysing the results of this study about language learning and use in ASA’s immersion setting, various individual and sociocultural dimensions were identified. The most salient findings are discussed here with respect to the following areas: learning English, using language in different contexts, variability in L2 comprehensibility and linguistic features influencing comprehensibility. Learning English Students provided perspectives on language learning at ASA and in the following contexts: other AOSs in Paraguay and abroad, stateside schools, elite bilingual schools in other Latin American countries, language institutes, private tutors and EFL classes at Spanish-medium schools. The K students generally recalled pleasant memories about early language learning. Those who recalled initial difficulties tended to have lower comprehensibility. Since such students often mentioned the problem of English pronunciation not corresponding to spelling, this could indicate low tolerance of ambiguity as well as L2 pronunciation being based on input from reading. Fear was openly expressed only by transfer students who had found themselves in foreign countries with the sudden need to learn a new language. Several students repeatedly mentioned friends as having been important for language learning. They also felt their L2 learning was supported by having begun at an early age and by having attended a high quality institution. Many students wanted proof of their ability to interact with native English speakers outside of ASA. Three had attended summer college in the US for high school juniors and another a semester at a stateside high school. Their ASA classmates seemed interested in the US slang they brought back. Yet, although these students viewed practice as the key to learning to speak fluently, they were adamant that not much could be done to make them speak more English at ASA. Many felt that in order to improve speaking, they needed ‘to be away from the people that they always talk to in Spanish’. Some students expressed views of language learning as a long process that started in the past and continues to the future. Almost all of them overwhelmingly expressed appreciation for family support in their having learned English and also in having maintained a home language (Chinese, Korean, Portuguese). They seemed aware that attitudes and motivation are

86

Bilingual Education in South America

indispensable to successful language learning: ‘Some people don’t care’, ‘It depends on you – if you want to’, and ‘I realised that if I wanted to learn, I could’. Nonetheless, many still viewed English primarily for instrumental purposes – schooling and career. Aspects of integrative motivation were usually only identifiable among transfer students who had lived in Englishspeaking environments. Several students seemed concerned about having poor pronunciation, getting stuck when talking, being confused when translating, and not having technical L1 vocabulary. Yet, in spite of these difficulties, all of them recognised the advantages of having learned two languages: ‘It is true that second language speakers are not as good as first language speakers, but we have the advantage of speaking two or more languages’. Not only did these students express their own theories about language learning, but they openly recognised their bilingualism as a product of their schooling: ‘My education made me who I am today, a bilingual person’. Using language in different contexts The students in this study provided perspectives on using language in different contexts – social, individual, and academic. They also self-assessed their language proficiency with respect to the four language skills – listening, speaking, reading and writing. Regardless of home language, 88% of students reported using only Spanish with peers, and 12% both Spanish and English (see Table 1). This finding indicates that L1 was the vernacular among ASA students. In their essays and interviews, many students also mentioned using only Spanish with classmates even during class: ‘We never speak English between us in the classroom, just when the teacher comes around’. Initially, this persistent use of Spanish seems to contradict studies that showed L2 use during structured activities in immersion classrooms (Cloud et al., 2000; Lyster, 1999; Tarone & Swain, 1995). However, just as these studies documented a drop in social L2 use as students progressed from lower to upper elementary Table 1 Summary of social, individual, and academic-related language use Self-reported language use

English

Spanish

Span.=Eng.

Sp.=Home Home

–

88%

12%

With siblings

–

82%

6%

3%

9%

With parents

3%

79%

2%

1%

15%

3%

78%

7%

10%

2%

Social situations With peers

–

–

Individual situations Expressing emotions Dreaming (9% blank)

3%

64%

12%

3%

9%

Thinking

6%

60%

22%

6%

6%

Reading for pleasure

41%

21%

29%

3%

6%

Counting and doing maths

65%

17%

9%

–

9%

Academic-related situations

English Immersion in Paraguay

87

grades (LaVan, 2001), the current study suggests an additional drop in L2 use by ASA secondary students during structured classroom activities. A slightly different pattern emerged regarding students’ self-reported language use for individual situations. When expressing emotions, the students reported Spanish at 78% as compared to English or English= Spanish at 10%. Students of other home languages (Chinese, Korean, Portuguese) often reported using their L1, either alone or with Spanish, especially for strong emotions (anger=frustration, joy=happiness). For thinking and dreaming, the exclusive use of Spanish dropped to about 60%, whereas English rose (alone or in combination with Spanish) to 15% for dreaming and 28% for thinking (which was perhaps associated with academic endeavours). Although ASA students still continued to use mainly Spanish (L1) for individual situations, they used more English (L2) than they had in social situations. The language pattern for academic-related situations was different. When reading for pleasure, 41% reported English, 27% English=Spanish, and 21% Spanish. This lack of L1 recreational reading seems related to the students’ self-assessment of low L1 academic skills. For counting and doing maths, 65% reported English, 6% English=Spanish, and only 17% Spanish. This high L2 use can be expected since 75% of the Spanish L1 students had never studied maths in Spanish. Almost all of the students perceived their speaking skills as high in Spanish and low in English. Many also self-assessed writing in L2 as higher than in L1. As native Spanish speakers, they were probably more critical of their L1 skills. Furthermore, not only had their L2 use been almost exclusively academic, but during their 14 years of schooling, these students had also written much more in English than in Spanish. Several students also claimed that it was easier to write English than to speak it. By having time to think and revise, students probably felt more in control of L2 writing as opposed to L2 speaking. These results suggest that ASA students had been using English primarily as a written medium. Some students also expressed the need to improve their Spanish: ‘We need to focus also on Spanish, not just on English. Our Spanish skills are very weak too’. This belief of weak L1 academic skills corroborated with findings from earlier studies both at ASA (Boxer, 1999; Rojas, 1999b) as well as at an elite bilingual school in Colombia (C.L. Ordo~ nez, personal communication, 29 October 2001). Such findings contrast with those from US=Canadian immersion studies where L1 academic skills met and even surpassed those of nonimmersion students (Lyster, 1999; Swain & Johnson, 1997; Thomas & Collier, 1997). Perhaps the home environments for supporting L1 literacy were significantly different from those in the US=Canadian immersion studies. Several students described their language use as an almost conscious effort to create a special identity: ‘ASA speech is unique – special, it’s ours’. According to Rebecca: The students at the American School of Asuncion have their own language. When we speak English, we speak ASA English, and when we speak Spanish, we speak ASA Spanish. Basically, we always are speaking Spanglish. . .. ASA talk is also unique because of the Guarani words that ASA students use.

88

Bilingual Education in South America

This is described by Alba as follows: ‘We create our own language, breaking up words and coming up with new ones, in a way that not even our parents would understand’. Aida explains: ‘I think that we hopefully are different, because I would not like to hear that every single person speaks the same way. What I am trying to say is that people become identified from others by their way of talking’. These ASA students demonstrated an awareness of the symbolic value of language and its importance for social=group identity, a phenomenon that had also been observed by Boxer (1994: 4): There are issues of identity coming into play here, reflected in the unique pronunciation patterns of the ASA students. Ethnic varieties have been known to be actively maintained or developed to express distinctive ethnic identity of users. This maintenance=development of marked features expresses assertions of solidarity towards their social reference group. No doubt at ASA, students’ pronunciation patterns reflect their desire to emphasise their distinct social identity. Some transfer students even perceived a loss of L2 fluency because of diminished L2 use at ASA when compared to former schools where English was the students’ social language. Most ASA students realised that they use English differently than do native speakers: ‘We do not talk at all as the people do in the US; we do not use slang, for example’. Nonetheless, students seemed satisfied and pleased with their bilingualism: ‘It’s surprising how one language can change the whole course of a person’s life’, and ‘Having English as my second language has shaped the course of my life in a very profound way of which I still have to see the results’. Variability in L2 comprehensibility In spite of apparent homogeneity for L2 learning at ASA, especially among the 14 students who had been schooled exclusively in this immersion setting, variability existed in their L2 output and perceived comprehensibility. Gender and former schooling were the two most significant variables affecting the students’ oral L2 output. Based on the comprehensibility ratings, which ranged from a low of 1 (impossible to understand) to a high of 7 (extremely easy to understand), the 16 girls had more highly developed L2 oral skills (M ¼ 5.81) than the 18 boys (M ¼ 4.86). This finding corroborates earlier studies showing girls outperforming boys in verbal ability, albeit mostly documented in L1 (Pennington, 1996; Swann, 1999). It also responds to Oxford’s call for more research on gender and L2 development: ‘Gender differences in L2 learning might be larger and more significant than teachers realise’ (1999: 559). Transfer students (PT & NT) exhibited higher comprehensibility than K students, who had been schooled exclusively at ASA (see Table 2). These results were not surprising since many transfer students had attended schools where English was the students’ social language. It confirms other studies that showed limited oral L2 development when restricted to immersion education (Cloud et al., 2000; LaVan, 2001) and greater development when access is

89

English Immersion in Paraguay Table 2 Comprehensibility means for gender and schooling Students

n

Scores1

Minimum

Maximum

m

sd

Skewness

9

216

2

7

5.69

0.52

0.12

2

48

3

7

6.04

0.24

–5

Girls K2 PT

3

NT4

4

96

2

7

5.98

0.92

0.07

All

15

360

2

7

5.81

0.61

0.21

5

120

2

6

4.23

0.45

0.30

Boys K2 3

6

144

2

7

5.17

0.54

0.34

NT4

4

96

2

7

5.19

0.63

1.49

All

15

360

2

7

4.86

0.68

0.22

Total

30

720

2

7

5.33

0.79

0.03

PT

Note Scale ranges from 1 (impossible to understand) to 7 (extremely easy to understand) 1 Number of scores provided by 12 raters on two listening tasks. 2K ¼ entry in pre-kinder (all Paraguayans). 3PT ¼ Paraguayans who transferred to ASA. 4NT ¼ other nationalities who transferred to ASA. 5Skewness cannot be computed due to only two items in the PT group

provided to additional practice, especially L2 use outside the school setting (Lyster, 1999; Mougeon & Rehner, 2001; Stein, 1999). As with other types of language learners, prolonged interaction with native-speaking peers seemed necessary to reach greater proficiency (Saville-Troike et al., 1984; Strong, 1983). Table 2 shows that gender exerted a stronger influence on L2 development among those students schooled exclusively at ASA than on those who had transferred to ASA. Gender totally divided the 14 K students, with all nine girls in this group being more comprehensible than their five male classmates with whom they had been schooled since kindergarten. With respect to former schooling, the comprehensibility of transfer students was positively influenced by a family’s move either to the US or to a country with an AOS where English was the students’ social language. Care should be taken not to interpret these ratings as indicative of unsuccessful language learning. To the contrary, the mean score of 5.3 (on a scale of 1 to 7) indicates that most students were perceived as comprehensible, with 85% as relatively easy to understand and 15% as difficult to understand. Based on the findings from US=Canadian immersion studies in which most students did not reach near-native L2 proficiency in the productive skills (Genesee, 1985; Swain, 1985; Swain & Johnson, 1997; Tarone & Swain, 1995; Tedick & de Gortari, 1998), it would be unrealistic to expect students schooled exclusively at ASA to attain native-like proficiency and high comprehensibility in speaking. Motivation also seemed to influence students’ use of English beyond the classroom. With respect to the students’ reported language use, variability existed in the use of English and Spanish within social domains (with peers and siblings), individual domains (thinking and expressing emotions), and

90

Bilingual Education in South America

academic-related domains (reading for pleasure and doing maths). Such variability in L2 use seemed related to attitudes about English, which in turn appeared to influence comprehensibility. Such attitudes were identified by questions dealing with ease=difficulty of language skills, enjoyment when studying English, and overall importance of learning English. Negative attitudes towards English often paralleled low comprehensibility especially for those boys who seemed to have felt pressured from the prohibited use of Spanish at a former school, from extensive language tutoring for passing an admission exam, or from academic tutoring for maintaining student status at ASA. Although 14 students had been schooled exclusively at ASA and all others had been there for many years, variability in L2 comprehensibility indicated that each student’s experience with learning and using English was a unique case. When categorised for gender, higher comprehensibility was related to more opportunities for L2 use, both socially and academically. Such experiences were either circumstantial, that of having lived in another country, or else motivational, that of using English as much as possible in Paraguay. As with studies about oral L2 support (Kofler et al., 2001), the students who had actually used English with greater frequency demonstrated higher levels of comprehensibility. Linguistic features that influenced perceived comprehensibility As in other immersion studies, ASA students exhibited non-native features, mainly phonological but also grammatical (Lyster, 1999). When judging comprehensibility, native-speaker raters were influenced primarily by prosodic dimensions and overall fluency. In spite of almost identical language learning opportunities for all 14 K students, several linguistic differences existed between the three with highest comprehensibility and the three with lowest. Native speaker listeners rated comprehensibility as high for speech samples containing the following phonological dimensions: high pitch in almost all tone units, more rising tones associated with the tonic syllable than falling and level tones, nuclear stress placed correctly in about 70% of tone units, medial stress as well as strong stress in most tone units, one strong or medial stress for about every two syllables approximating stress-timed rhythm, vowel reduction to schwa in about 30% of unstressed syllables, and about 70% of high vowels approximated to the targeted full form in stressed syllables. These same listeners rated comprehensibility as low for speech samples that contained: almost total lack of high pitch, strong predominance of falling and level tones associated with the tonic syllable, nuclear stress misplaced in about 76% of tone units, no medial stress, stress on every third or fourth syllable resembling syllable-timed rhythm, vowel reduction to schwa in only 7% of the unstressed syllables, and only 37% of high vowels approximated to the targeted full form in stressed syllables. Although difficulties also existed with other vowels as well as with consonants, especially in the aspiration= unaspiration of voiceless stops and the unexpected devoicing=loss of syllable-final non-resonants, such difficulties were observed among all students and did not readily distinguish between those of high and low

English Immersion in Paraguay

91

ratings. The strong contrast in prosodic features between high and low comprehensibility was congruent with other research findings (Coe, 1987; Munro, in press; Munro & Derwing, 1995; Pennington, 1996; Pickering, 2001). Unexpected prosodic features influenced raters to a greater degree than did phonetic errors. With respect to non-phonological dimensions, the native-speaker listeners rated comprehensibility as high for speech samples that contained: a fluency rate per 60 seconds of 159–192 words of which 95–103 were nouns, verbs, adjectives and adverbs; and from 17–18 errors related to word choice, discourse interruptions and word usage, which averaged 13–15% of the actual word count. These listeners rated comprehensibility as low for speech samples that contained: a fluency rate per 60 seconds of only 116–148 words of which 59–88 were nouns, verbs, adjectives and adverbs; and from 33–38 errors related to word choice, discourse interruptions and word usage, which averaged 28–40% of the actual word count. In comparison with their less comprehensible classmates, the highly comprehensible students had spoken 33% more words within a 60-second period, used 44% more content words, and averaged 47% fewer lexical=morphosyntactic errors and discourse interruptions. Although not as overwhelmingly conclusive as the phonological analysis, these findings were also congruent with earlier studies that had indicated how immersion students often fall behind in linguistic structures and appropriate use register (Genesee, 1985; Mougeon & Rehner, 2001; Swain, 1985; Swain & Johnson, 1997). The phonological features of the highly comprehensible K students (all girls) could represent a type of convergent accommodation (Giles & Byrne, 1982; Giles et al., 1973) to the school’s overt prestige norm (Trudgill, 1972, 2000), i.e. a conscious or unconscious desire to model their speech patterns on the standard English favoured by their US teachers (Mougeon & Rehner, 2001). Likewise, the phonological features of K students with low comprehensibility (all boys) could have resulted from divergent accommodation (Giles & Byrne, 1982; Giles et al., 1973) to the covert prestige norm of ASA boys (Labov, 1966; Trudgill, 1972, 2000), i.e. a conscious or unconscious desire to speak similarly to peers in their own reference group. These findings are congruent with earlier studies describing how ‘males generally orient their speech more to localised norms than do females’ (Pennington, 1996: 16) and how the divergent pronunciation between males and females is ‘a phonological marker of identity in that group’ (1966: 5). Morgan (1997) describes this relationship between identity and intonation as a dynamic process of language learning in educational settings. Due to reduced systematic complexity in speech samples of low comprehensibility, especially with respect to tone, stress and vowels, it seems that these ‘speakers of English as a second language may have little motivation to introduce or to maintain complex variation and subtle phonological features in the language for intricate social identification’ (Pennington, 1996: 13). The preferred role models for many language learners at ASA seems to be peers from their ingroup. Students’ language modelling in ASA’s immersion setting did not necessarily focus on the native-speaking teacher and occasional classmate but on non-native speaker norms. Most of the K girls adequately

92

Bilingual Education in South America

approximated the English modelled by teachers, whereas the K boys were influenced by ASA’s institutional variety (Eckert, 1989; Rickford, 1996).

Implications This study provides implications for practice, research and theory. Although specifically targeted for AOS settings with high HCN populations, these implications can be adapted for other immersion settings, such as elite bilingual schools in Latin America. Based on the students’ expertise in having become bilingual, the study’s applied research focus elicited their input for helping future ASA students with spoken English. These students suggested the following: teacher support in helping students improve pronunciation, more oral activities in the classroom, an obligatory speech class in high school, exchanges for high school students and additional support and=or motivation via oral exams. Above all, ASA students emphasised the need to provide more opportunities for practising English. Another implication from this study is increased staff development in AOS immersion settings in order to help teachers better understand theories of language learning=use and practices for maximising learning via contentbased instruction. Implications for research and theory are related to: comprehensible input versus comprehensible output among immersion students; the relationship between gender and L2 productive skills; and the L2 development of HCNs in AOS immersion settings vis-a`-vis student attitudes=motivation and language learning beliefs.

Conclusion This paper has offered an explanatory account of variability in language learning and use among English immersion learners at ASA. The students’ bilingual voice served in identifying individual and sociocultural dimensions within a spectrum of complex interrelationships and interdependencies, analogous to the characteristics of a rainbow. Metaphorically speaking, this rainbow represents the linguistic ecology of immersion schooling in which all conditions must be exactly right to ensure optimal L2 development towards increasingly higher levels of proficiency. This rainbow also symbolises the overlapping of various theoretical frameworks for interpreting the findings of language learning and use among ASA’s English immersion learners. To that end, the current study suggests that findings from studies in other immersion settings can be extended to the AOS immersion context. Similar to other immersion learners, ASA students demonstrated a relatively high functional L2 proficiency but not necessarily near-native (Lyster, 1999). As implied by Swain (1985) and Stein (1999), variability in the oral comprehensibility of ASA’s immersion students indicates that Krashen’s theory of comprehensible input (Krashen & Terrell, 1983) is not sufficient for explaining language learning in immersion education settings. Although Cummins’ (1979) theory of BICS=CALP was useful for differentiating proficiency types, the current study shows that an immersion AOS setting provides learners with more opportunities for developing academic language proficiency as opposed to oral communication skills.

English Immersion in Paraguay

93

SLA’s broader theoretical framework also provided insights into this study with gender as a determining factor (Ellis, 1994). The variability in L2 learning between girls, approaching a more standard norm, and boys, following a nonnative form, is best explained via both covert prestige (Labov, 1966; Trudgill, 1972) and the Accommodation Theory (Giles & Byrne, 1982; Giles et al., 1973). The students’ age was also instrumental in modelling speech on that of peers rather than on the teacher (Ellis, 1994; Labov, 1970). Attitudes and motivation played an important role, both positively and negatively (Ehrman & Do¨rnyei, 1998; Gardner, 1979). Since most students seemed to exhibit instrumental motivation, the lack of integrative motivation could have hampered progress towards greater proficiency (Oxford & Shearin, 1994). Anxiety and low tolerance for ambiguity also seemed to negatively influence comprehensibility (Ellis, 1994). This interpretation indicates how several aspects from SLA’s theoretical framework also affect language learning=use in ASA’s immersion setting. These findings also suggest that variability in language learning, comprehensibility, and language use may be directly influenced by the opportunities provided to the learner from within the specific language learning context (Norton & Toohey, 2001) as well as an individual’s own investment (Norton Peirce, 1995). Such was demonstrated by the higher comprehensibility of students who had attended schools where English was the students’ social language. ASA immersion students were ‘bound up not only in what they did individually but also in the possibilities . . . offered them’ (Norton & Toohey, 2001: 318). These findings also highlight the important relationship between friends and language learning. The more highly comprehensible students had actually used English with greater frequency, due either to circumstances of being in other countries or to motivational factors for maximising their use of English in Paraguay. This fits with Firth and Wagner’s implication (1997) that language learners be called language users. The social and group identity theories (Giles & Byrne, 1982; Tajfel, 1982), as well as other identity-based theories from the field of social psychology, also provided a theoretical background for explaining L2 variability in ASA students. Several students even alluded to their special ASA talk. Many immersion students, especially boys, seemed influenced by their reference group’s symbolic use of language for social identity (Eckert, 1989; McGroarty, 1996; Rickford, 1996). Successive generations of ASA students might even be reproducing a non-native variety, which in turn finds theoretical basis in interlanguage (Selinker, 1972, 1992), institutional varieties (McKay, 1992), and new World Englishes (Kachru & Nelson, 1996). Theoretical approaches from interactional sociolinguistics (Schiffrin, 1996) and from the framework of linguistic and sociocultural ecology (Davies, 1999; Erickson, 1996; Schecter & Bayley, 1997) were also useful in understanding ASA’s language learning context. Immersion classrooms are diglossic speech communities with each language reserved for certain communicative purposes. Within ASA’s linguistic ecology, students use mainly English for academic matters and Spanish for conversational interaction. Consequently, students do not develop an L2 vernacular and are restricted to using L2 solely as a formal register. Such interactional patterns provide students little

94

Bilingual Education in South America

opportunity for L2 use. This lack of comprehensible output limits the development of L2 productive skills. These insights from the fields of immersion education, second language acquisition, sociolinguistics and social psychology have provided a theoretical background with which to better understand variability in language learning and use among English immersion learners at ASA. By exploring processes of learning English, patterns of language use, and levels of comprehensibility, this study described individual and sociocultural dimensions within this English immersion setting. Such findings now provide the basis for future research at other elite bilingual schools. Correspondence Any correspondence should be directed to Dr Susan Spezzini, University of Alabama at Birmingham, 1530 3rd Ave South, Education Building 257, Birmingham AL 35294–1250, USA ([email protected]). Notes 1. AOSs are recognised by the Office of American Overseas Schools at the United States Department of State for the schooling of its dependents. International School is another term with a somewhat broader definition. Today there are approximately 850 International Schools worldwide (Baker & Jones, 1998). 2. Paraguay, about the size of California, is a landlocked agrarian country bordered by Bolivia to the north, Brazil to the east, and Argentina to the west and south. About a quarter of Paraguay’s approximately 5.8 million inhabitants live in and around Asuncion. In spite of the country’s nationwide Spanish=Guarani bilingualism, usually only Spanish is spoken in the homes of high SES families. 3. For more information about ASA, visit its electronic homepage: www.asa.edu.py 4. Spanish language learners take Spanish as a second language (SSL) and host country culture instead of Spanish language arts, Hispanic social studies and Guarani. 5. ASA kindergartners receive Spanish instruction for 30 minutes daily, which increases to 90 minutes as of first grade. Previously, Spanish had been introduced at the beginning of first grade (1996–1999), in the middle of first grade (1980s– 1995), and in second grade (1960s–1970s). 6. As part of their English class, students delivered 15-minute formal speeches with assumed identities representing countries they had created during their year-long government project. 7. According to Munro, several studies have confirmed the validity and reliability of rating tasks (in press). Munro and Derwing assessed overall interrater reliability using a method that ‘revealed moderate interrater agreement, with Pearson (r) values of 0.71 for comprehensibility’ (2001: 459).

References American School of Asuncion (2001a) Faculty Handbook. Asuncion, Paraguay. American School of Asuncion (2001b) Self-study Report for Reaccreditation. Asuncion, Paraguay. American School of Asuncion (2002) Student-parent Handbook. Retrieved January 4, 2002, from http:==www.asa.edu.py=pages=geninf hndbk geninfo eng.html Baker, C. (1993) Foundations of Bilingual Education and Bilingualism. Clevedon: Multilingual Matters. Baker, C. (1995) A Parents’ and Teachers’ Guide to Bilingualism. Clevedon: Multilingual Matters. Baker, C. and Jones, S. P. (1998) Encyclopedia of Bilingualism and Bilingual Education. Clevedon: Multilingual Matters.

English Immersion in Paraguay

95

Boxer, D. (1999) Report on Language Use at ASA. Asuncion, Paraguay: American School of Asuncion. Brause, R.S. and Mayher, J.S. (1994) Search and Re-search: What the Inquiring Teacher Needs to Know. London: Falmer Press. Carder, M. (1993) Are we creating biliterate bilinguals? International Schools Journal, 26, 19–27. Abstract [electronic version] retrieved April 15, 2001, from International Schools Journal: Index and Abstracts, 1981–2000. Petersfield, UK: European Council of International Schools (ECIS) Publications. Carder, M. (1995) Language(s) in international education: A review of language issues in international schools. In T. Skutnabb-Kangas (ed.) Multilingualism for All (pp. 113–158). Lisse, Netherlands: Swets and Zeitlinger B.V. Cloud, N., Genesee, F. and Hamayan, E. (2000) Dual Language Instruction: A Handbook for Enriched Education. Boston: Heinle & Heinle. Coe, N. (1987) Speakers of Spanish and Catalan. In M. Swan and B. Smith (eds) Learner English: A Teacher’s Guide to Interference and Other Problems (pp. 72–89). Cambridge: Cambridge University Press. Cook, V. (1994) Universal grammar and the learning and teaching of second languages. In T. Odlin (ed.) Perspectives on Pedagogical Grammar (pp. 25–48). Cambridge: Cambridge University Press. Cummins, J. (1979) Linguistic interdependence and the educational development of bilingual children. Review of Educational Research 49, 222–251. Davies, C.E. (1999) Linguistic ecology and the construction of sociocultural identity: Discourse communities of a southern American university. Paper presented at the Southeast Regional TESOL Conference, November 3–5, Birmingham, Alabama. De Courcy, M. (2002) Learners’ Experiences of Immersion Education. Clevedon: Multilingual Matters. Eckert, P. (1989) Jocks and Burnouts: Social Categories and Identity in the High School. New York: Teachers College Press. Ehrman, M. E. and Do¨rnyei, Z. (1998) Interpersonal Dynamics in Second Language Acquisition. Thousand Oaks, CA: Sage. Ellis, R. (1994) The Study of Second Language Acquisition. Oxford: Oxford University Press. Erickson, F. (1996) Ethnographic microanalysis. In S.L. McKay and N.H. Hornberger (eds) Sociolinguistics and Language Teaching (pp. 283–306). Cambridge: Cambridge University Press. Firth, A. and Wagner, J. (1997) On discourse, communication, and (some) fundamental concepts in SLA research. Modern Language Journal 81, 285–300. Gardner, R.C. (1979) Social psychological aspects of second language acquisition. In H. Giles and R.N. St Clair (eds) Language and Social Psychology (pp. 193–220). Baltimore: University Park Press. Gass, S., Sorace, A. and Selinker, L. (1999) Second Language Learning Data Analysis (2nd edn). Mahwah, NJ: Erlbaum. Genesee, F. (1985) Second language learning through immersion: A review of US programs. Review of Educational Research 55, 541–561. Giles, H. and Byrne, J.L. (1982) An intergroup approach to second language acquisition. Journal of Multilingual and Multicultural Development 3, 17–40. Giles, H., Taylor, D.M. and Bourhis, R. (1973) Towards a theory of interpersonal accommodation through language: Some Canadian data. Language in Society 2, 177–192. Green, J.M. and Oxford, R.L. (1995) A closer look at learning strategies, L2 proficiency, and gender. TESOL Quarterly 29, 263–297. Grosjean, F. (1999) Bilingualism: Individual. In B. Spolsky (ed.) Concise Encyclopedia of Educational Linguistics (pp. 284–290). Amsterdam: Elsevier. International Schools Services (ISS), Inc. (1994) The Teaching Overseas Information Handbook. Princeton, NJ.

96

Bilingual Education in South America

Kachru, B.B. and Nelson, C.L. (1996) World Englishes. In S.L. McKay and N.H. Hornberger (eds) Sociolinguistics and Language Teaching (pp. 71–102). Cambridge: Cambridge University Press. Kofler, A., Wegerif, R. and LeVoi, M. (2001) Oracy and the educational achievement of pupils with English as an additional language: The impact of bringing ‘talking partners’ into Bradford schools. International Journal of Bilingual Education and Bilingualism 4 (6), 403–419. Krashen, S.D. and Terrell, T.D. (1983) The Natural Approach: Language Acquisition in the Classroom. Hayward, CA: Alemany Press. Labov, W. (1966) Hypercorrection by the lower middle class as a factor in linguistic change: Discussion. In W. Bright (ed.) Sociolinguistics: Proceedings of the UCLA Sociolinguistics Conference, 1964 (pp. 84–113). The Hague: Mouton. Labov, W. (1970) The Study of Nonstandard English. Urbana, IL: National Council of Teachers of English. LaVan, C. (2001) Help! They’re using too much English! The problem of L1 vs. L2 in the immersion classroom. ACIE Newsletter: The Bridge from Research to Practice 4 (2), 1–4. Leech, G. and Svartvik, J. (1994) A Communicative Grammar of English (2nd edn). New York: Longman. Lyster, R. (1999) Immersion. In B. Spolsky (ed.) Concise Encyclopedia of Educational Linguistics (pp. 626–632). Amsterdam: Elsevier. McGroarty, M. (1996) Language attitudes, motivation and standards. In S. L. McKay and N. H. Hornberger (eds) Sociolinguistics and Language Teaching (pp. 3–46). Cambridge, UK: Cambridge University Press. McKay, S.L. (1992) Teaching English Overseas: An Introduction. Oxford: Oxford University Press. Met, M. and Galloway, V. (1996) Research in foreign language curriculum. In P. Jackson (ed.) Handbook of Research on Curriculum (pp. 852–890). New York: Simon & Schuster Macmillan. Morgan, B. (1997) Identity and intonation: Linking dynamic processes in an ESL classroom. TESOL Quarterly 31, 431–450. Mougeon, R. and Rehner, K. (2001) Acquisition of sociolinguistic variants by French immersion students: The case of restrictive expressions, and more. Modern Language Journal 85, 398–415. Munro, M.J. (in press) The perception of foreign-accented speech. In R.H. Smyth (ed.) Festschrift for Bruce L. Derwing. Amsterdam: John Benjamins. Munro, M.J. and Derwing, T.M. (1995) Foreign accent, comprehensibility, and intelligibility in the speech of second language learners. Language Learning 45, 73–97. Munro, M.J. and Derwing, T.M. (2001) Modeling perceptions of the accentedness and comprehensibility of L2 speech: The role of speaking rate. Studies in Second Language Acquisition 23, 451–468. Norton Peirce, B. (1995) Social identity, investment, and language learning. TESOL Quarterly 29, 9–31. Norton, B. and Toohey, K. (2001) Changing perspectives on good language learners. TESOL Quarterly 35, 307–322. Orr, P.G. and Conlan, A.C. (1982) The Organization of English as a Second Language in the American-sponsored Overseas Schools. Tuscaloosa, AL: International Education Associates. Oxford, R.L. (1999) Second language learning: Individual differences. In B. Spolsky (ed.) Concise Encyclopedia of Educational Linguistics (pp. 552–560). Amsterdam: Elsevier. Oxford, R.L. and Shearin, J. (1994) Language learning motivation: Expanding the theoretical framework. Modern Language Journal 78, 12–28. Patton, M.Q. (1990) Qualitative Evaluation and Research Methods. Newbury Park, CA: Sage. Paulston, C.B. (1999) Multilingualism. In B. Spolsky (ed.) Concise Encyclopedia of Educational Linguistics (pp. 60–63). Amsterdam: Elsevier. Pennington, M.C. (1996) Phonology in English Language Teaching. London: Longman.

English Immersion in Paraguay

97

Pickering, L. (2001) The role of tone choice in improving ITA communication in the classroom. TESOL Quarterly 35, 233–256. Rabbitt, M. (1994) The Teaching Overseas Information Handbook. Princeton, NJ: International Schools Services. Rickford, J.R. (1996) Regional and social variation. In S.L. McKay and N.H. Hornberger (eds) Sociolinguistics and Language Teaching (pp. 151–194). Cambridge: Cambridge University Press. Rojas, V.P. (1999a) International schools: The challenges of teaching languages overseas. Learning Languages 4 (2), 16–20. Rojas, V.P. (1999b) Linguistic Audit Report. Asuncion, Paraguay: American School of Asuncion. Rutherford, W.E. (1987) Second Language Grammar: Learning and Teaching. London: Longman. Saunders, G. (1988) Bilingual Children: From Birth to Teens. Clevedon: Multilingual Matters. Saville-Troike, M., McClure, E. and Fritz, M. (1984) Communicative tactics in children’s second language acquisition. In F.R. Eckman, L.H. Bell and D. Nelson (eds) Universals of Second Language Acquisition (pp. 60–71). Rowley, MA: Newbury House. Schecter, S.R. and Bayley, R. (1997) Language socialization practices and cultural identity: Case studies of Mexican-descent families in California and Texas. TESOL Quarterly 31, 513–542. Schiffrin, D. (1994) Approaches to Discourse. Malden, MA: Blackwell. Schiffrin, D. (1996) Interactional sociolinguistics. In S.L. McKay and N.H. Hornberger (eds) Sociolinguistics and Language Teaching (pp. 307–328). Cambridge: Cambridge University Press. Sears, C. (1998) Second Language Children in Mainstream Classrooms: A Handbook for Teachers in International Schools. Clevedon: Multilingual Matters. Selinker, L. (1972) Interlanguage. International Review of Applied Linguistics and Language Teaching 10, 209–231. Selinker, L. (1992) Rediscovering Interlanguage. London: Longman. Spezzini, S. (2002) Students’ bilingual voice at an American overseas school: Individual and sociocultural dimensions in the process of learning English and in patterns of language use. Unpublished doctoral dissertation, The University of Alabama, Tuscaloosa. Spezzini, S. and Oxford, R. (1998) Perspectives of pre-service foreign language teachers. System 26 (1), 65–76. SPSS (1999) SPSS1 Base 9.0 Applications Guide. Chicago. Stein, M. (1999) Developing oral proficiency in the immersion classroom [Electronic version]. ACIE Newsletter: The Bridge from Research to Practice 2 (3), 1–8. Stockwell, R.P. and Bowen, J.D. (1965) The Sounds of English and Spanish. Chicago: The University of Chicago Press. Strong, M. (1983) Social styles and the second language acquisition of Spanish-speaking kindergartners. TESOL Quarterly 17, 241–258. 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 (pp. 235–256). Rowley, MA: Newbury House. Swain, M. and Johnson, R.K. (1997) Immersion education: A category within bilingual education. In R.K. Johnson and M. Swain (eds) Immersion Education: International Perspectives (pp. 1–16). New York: Cambridge University Press. Swann, J. (1999) Gender and language. In B. Spolsky (ed.) Concise Encyclopedia of Educational Linguistics (pp. 200–202). Amsterdam: Elsevier. Tajfel, H. (1982) Instrumentality, identity and social comparisons. In H. Tajfel (ed.) Social Identity and Intergroup Relations (pp. 483–507). Cambridge, UK: Cambridge University Press. Tarone, E. and Swain, M. (1995) A sociolinguistic perspective on second language use in immersion classrooms. Modern Language Journal 79, 166–178.

98

Bilingual Education in South America

Tedick, D. and de Gortari, B. (1998) Research on error correction and implications for classroom teaching [Electronic version]. ACIE Newsletter: The Bridge from Research to Practice 1 (3), 1–6. Thomas, W.P. and Collier, V. (1997) School effectiveness for language minority students [electronic version]. Washington, DC: Disseminated by National Clearinghouse for Bilingual Education, the George Washington University, Center for the Study of Language and Education: US Department of Education, Office of Educational Research and Improvement, Educational Resources Information Center. Trudgill, P. (1972) Sex, covert prestige and linguistic change in the urban British English of Norwich. Language in Society 1, 179–195. Trudgill, P. (2000) Sociolinguistics: An Introduction to Language and Society (4th edn). London: Penguin Books. Ur, P. (1998) Grammar Practice Activities. Cambridge: Cambridge University Press.

A Look at Early Childhood Writing in English and Spanish in a Bilingual School in Ecuador JoEllen M. Simpson Formerly at Universidad del Valle, Cali, Colombia The present study examines 20 writing samples in English and Spanish selected from portfolios of first grade children at a bilingual school in Ecuador. Based on earlier findings that paragraphs composed in English and Spanish by children, adolescents and adults are different (Lux & Grabe, 1991; Monta~ no-Harmon, 1991; Reid, 1990; Reppen & Grabe, 1993; Simpson, 2000), an analysis was conducted of 20 written samples from first grade children, focusing on the physical structure and the topical structure of those writings. The physical characteristics of the paragraphs included the number of T-units, words, errors, error types, sentence types and connectors. Results of this quantitative analysis do not reflect findings from earlier studies describing English–Spanish differences, which may be due to the very young age of these writers or the influence of English instruction. The topical structure analysis, an analysis of coherence derived by examining the internal topical structure of each paragraph as reflected by the repetition of key words and phrases, provides insights into the organisational patterns used by these children. The results of the topical structure analysis show that these children use similar amounts of topical repetition in the two languages. Additionally, this analysis reveals developmental stages in narrative strategies. Keywords: early child bilingualism, contrastive rhetoric

Introduction Over the past 30 years, linguistic research on writing, initiated by Kaplan’s 1966 article about cultural thought patterns, has emphasised that writing is done distinctly by people from different cultures. Kaplan’s article has served as the starting point for many studies on what has been termed contrastive rhetoric. In order to discover more about differences of written texts from two languages, the present study is from within the field of contrastive text linguistics (as described by Connor, 1996) and is a comparative study of short narratives written by children in Spanish and English. Most of the research on Spanish contrastive rhetoric (see below) has been conducted with high school or university students and adults, and these studies have shown that there are specific styles for writing in Spanish and English writing, seen in children as young as 5th grade. It is the aim of this study to examine the written production of young children (first graders) in a bilingual school in Ecuador to see if the differences described by earlier studies can be seen at this level as well.

Spanish and English Contrastive Rhetoric Research in contrastive rhetoric in Spanish and English has resulted in a basic description of how these two languages are different. Analysis of written

99

100

Bilingual Education in South America

texts in Spanish and English show similar tendencies from primary school writers to professional academic writers. In English, for example, there is a long history of teaching people how to write, and this ‘direct’ style has come to be associated with English-speaking cultures. English writers have been described as ‘much less interactional than those of other groups (Spanish)’ (Lux & Grabe, 1991: 150), and they ‘appear to prefer a more reduced sentence style, as well as a more informationally oriented . . . style’ (1991: 151). Monta~ no-Harmon (1991) describes English writing as containing ‘logical relationships between one idea and the subsequent idea’ (1991: 421), frequently marked by the use of organisational words such as ‘first’, ‘second’, and ‘third’. The early work on contrastive rhetoric of Spanish and English consisted of unpublished dissertations (Santiago, 1970; Santana-Seda, 1970, cited by Connor, 1996). In the recent studies in Spanish–English contrastive rhetoric, the main focus has been on quantitative analyses of syntactic and lexical markers, with an eye towards enumerating physical differences. For example, Reid’s dissertation (cited in Reid, 1990) looked at syntactic features of English, Arabic, Chinese and Spanish university level students writing in English by studying essays from the Test of Written English from the TOEFL exam. Among other things, she found that writers of Spanish wrote significantly longer sentences in English, and she noted that the Spanish writers in her study used the same ‘elaborate style’ that had been reported by the earlier unpublished dissertations. This ‘elaborate style’ has been taken up by researchers who have followed Reid as one of the defining characteristics of Latin American Spanish (for example, Lux & Grabe, 1991; Monta~ no-Harmon, 1991; Reppen & Grabe, 1993). Elaborate style is usually defined in terms of the use of long sentences with many additive clauses as well as what Monta~ no-Harmon refers to as ‘flowery, poetic language’ (1991: 423). Simpson (2000) analysed academic articles written in English and Spanish by highly educated academic professionals and published in refereed journals. The analysis revealed that native Spanish-speaking writers writing in Spanish had a similar preference for the ‘elaborate style’ with long sentences and long clauses, and the English writers used shorter sentences typical of Written English, reflecting earlier studies. Additionally, the English authors analysed in the study also tended to use more repetition of key words and phrases as an internal coherence device within paragraphs, while the Spanish texts had a number of different topical subjects within one paragraph, without much repetition of key words. Monta~ no-Harmon (1991) reported on the writing of ninth graders in Mexico and the United States. She found that in a quantitative analysis of syntactic characteristics of the essays, the Mexican students writing in Spanish wrote longer essays with longer, but fewer, sentences. The Spanish writers also had more run-on sentences and fewer simple sentences than the Anglo students writing in English. She noted that the ESL students and MexicanAmerican students reflected the Mexican Spanish writing styles more closely than the Anglo-American styles exhibited in her study. Lux and Grabe (1991) reported on Spanish–English writing contrasts in the writing of university students (based on Lux’s 1991 dissertation research). They studied Ecuadorian students writing in Spanish and English (EFL) as

Early Childhood Writing in Ecuador

101

well as Anglo-Americans writing in English and Spanish (Spanish as a FL). They found that the Spanish texts had longer sentences, among other characteristics, and that Ecuadorian Spanish speakers used the same elaborate style described by earlier researchers. They concluded that Ecuadorian Spanish writers prefer ‘an elaborated sentence style in their writing, and they appear to prefer a more abstract informational presentation’ (1991: 151). Reppen and Grabe (1993), in a comprehensive review of these earlier studies, note that Spanish writing has been described as being ‘more ornate and formal’ (1993: 117) than English. They were interested in children’s writing in order to locate the source of the ‘elaborate style’ of native Spanishspeaking adolescents and young adults tested in earlier studies. The authors collected essays from three groups of fifth-grade students, some native English-speaking and some native Spanish-speaking. The students’ essays were analysed for a number of syntactic structures, and the authors found that these Spanish-speaking L1 fifth graders (writing in English) have characteristics similar to those of older Spanish-speaking subjects from the other studies, and they suggest that this style of writing must be culturally influenced because the children in the study were quite young and had not received much formal training in writing in Spanish. But what about writers who are six or seven years old? The studies reported here contain analyses of writing samples produced by grade school, high school, university and adult writers, and they all show a number of tendencies in the two languages across ages and writing styles. Reppen and Grabe (1993) were looking for a source for the elaborate style used by native Spanish speakers, and they discovered that fifth grade students wrote with similar characteristics seen in the studies of more mature writers. The current study looks at the writing of first grade students to see if they reflect the same tendencies. This study also examines the possibility of reverse transfer of writing style – from the foreign language to the native language – due to the fact that the policy at this bilingual school is to emphasise reading and writing in English while instruction of written skills in their native language is limited to one hour a day.

Skill Transfer and Early Bilingualism It is generally accepted that for learners of English as a second or foreign language, there is transfer of skills, abilities, and patterns from the L1 in to the L2, especially in writing (see for example, Bell, 1995; Eisterhold, 1990; Friedlander, 1990; Hinkel, 1994; Krapels, 1990: just to name a few). However, these authors are generally talking about older children and teenagers, where it is assumed that there has been some literacy training in the first language. These authors do not discuss what happens with very young children learning to write in a foreign language before they learn to write in their first language. Hinkel (1994: 355) states that although many specialists have come to recognize that NNSs rely on their knowledge of L1 rhetorical paradigms, it has not been established with certainty whether NNSs who have received extensive L2 training

102

Bilingual Education in South America

and have achieved a relatively high L2 proficiency can effectively bridge the gap between L1 and L2 writing conventions. If this is the case for students who have received ‘extensive L2 training’, what about children with virtually no L1 training in writing? Will their L2 influence the way they write their L1? Will their L1 be strong enough to not be influenced by ‘reverse’ transfer? Eisterhold (1990: 99) suggests that there are three components in the transfer of literacy skills: (1) (2) (3)

The cognitive processes involved in writing and reading; The structural components that underlie writing and reading; and The mechanism that allows the processes and structures to transfer, either across languages or across modalities.

In other words, this means that for transfer of skills to be effective, the child has to have enough cognitive development in general to learn about writing and reading. The child also must be made aware of certain conventions of writing (structural components) in either or both languages. And finally, the child must be made aware of the strategies necessary to transfer this background knowledge about literacy from one language to another. But in order to accomplish this, proficiency in both languages is definitely a factor. Eisterhold claims that ‘a threshold level is indicated before the transfer mechanism can be employed successfully’ (1990: 99). However, Urzua (1987) claims that even though children have not received literacy instruction in their first language, they will create their own hypotheses about the L2. Hudelson (1984) explains that students are able to read and write in the foreign language even though they do not have full control over all oral aspects of the language. In this way, children with little training will still be able to make an attempt to write based on these hypotheses. But even if this is possible, Edelsky (1982) recommends ‘a solid basis’ in the first language literacy before instruction is begun in the second language. The children in the present study, being so young and having such limited literacy experience (primarily in the L2), may have to rely heavily upon these ad hoc hypotheses in order to write in both languages. What possible effects may this have on these young writers?

The Study The school The school chosen for this study is a private bilingual school in Quito, Ecuador. This school caters to children from upper-middle- and upper-class families and covers preschool through high school and has approximately 1200 students. The school was started in 1995 as an alternative to traditional bilingual education. The directors believed that the arts should be integrated into all aspects of learning, and whole language was accepted as the model for instruction. In 2001 there was a change in direction because the students were not performing as well as desired on evaluations, and the whole language model was tempered in favour of a reading-based and balanced literacy model for elementary students with positive results so far.

Early Childhood Writing in Ecuador

103

The school emphasises early immersion in English, a foreign language for the majority of students (a small number of children come from English-speaking or bilingual homes). In preschool, there is a focus on oral language development along with pre-reading and pre-writing activities. By first grade, the children are producing short texts in both languages, and this development of basic writing continues throughout elementary school to 5th grade. Instruction in writing is primarily in English. Spanish as a language of instruction in primary school is used mainly in Spanish class and art, music, and physical education classes, although the school makes an effort to find bilingual arts teachers. The subjects The writing samples were taken from the year-end comprehensive portfolios of students who were in first grade in the 2001–2002 school year. The total number of first graders during that academic period was about 80 and the portfolios were randomly chosen. The average age of students in this grade is six years old. Selection of corpus From among the portfolios, 10 samples of writing in both English and Spanish were randomly chosen. The samples in English were written in response to a primary-school-wide writing prompt. The primary coordinators were looking for a way to help teachers learn to use grading rubrics, so all of the children from 1st to 5th grade were asked to write a short story titled ‘If I were a fly’. Ten samples of this essay were randomly selected from the first graders’ portfolios. The Spanish samples were generated after a week-long holiday, and all of the first grade teachers asked their students to write a narrative about what they had done during that week. Ten samples of this essay were also randomly selected from the 1st graders’ files. Methods for analysis The samples written by these young students do not follow accepted standards in terms of punctuation and capitalisation. In order to have a standard unit for comparing the texts, they were divided into T-units, defined as ‘one main clause with all the subordinate clauses attached to it’ (Hunt, 1965: 20). Scholfield justifies using T-units as the basis of analysis of child speech or writing by explaining that ‘they actually may be easier to spot in recorded spoken material, or writing with poor punctuation, where there are no full stops or capital letters’ (1995: 116). Gaies (1980: 55) recommends looking at error-free T-units as ‘a more valid measure of growth in a second language’ as well as the ratio of error-free T-units to the total number of T-units produced by a subject. There has been certain discussion about what an error-free T-unit is, according to Gaies, ranging from ‘perfect in all respects, including spelling and punctuation, for it to be counted as error-free’ to ‘free of all morphological and syntactic errors’ (1980: 55). For the purposes of this study, a more restrictive definition was used as the measure of error-free due to the fact that syntactically the T-units were quite simple with few resulting errors, and it was revealing to look at all manner of errors that the students made in both languages. Capitalisation and punctuation errors were not counted because the subjects did not follow conventions of either sentence punctuation or capitalisation.

104

Bilingual Education in South America

Each narrative was divided into T-units, which were then counted, as were the number of words and number of errors. The errors were classified by type. Then each T-unit was analysed for syntactic complexity and use of sentence connectors. In order to compare topical development of these narratives with earlier studies, a topical structure analysis was conducted. Lautamatti’s (1987) topical structure analysis (TSA) was designed to identify the internal structure of paragraphs by plotting the repetition of key words and phrases. TSA is conducted by identifying sentence elements and plotting topical subjects onto a table to see relationships in the development of the thematic structure of the paragraph. After the topical subject is identified in each sentence, it is plotted onto a graph, and a physical representation of the thematic development can be visualised. Lautamatti (1987) suggests three types of thematic progression in her presentation of TSA: parallel progression (two consecutive clauses with the same topical subject); extended parallel progression (a topical subject that occurs in two clauses that are not consecutive); and sequential progression (the rheme element of a clause becoming the theme element of the consecutive clause). Simpson (2000) identifies a fourth progression: extended sequential progression (the rheme element of a clause becoming the theme element of a non-consecutive clause). The following example is given to show how the topical structure analysis reveals the internal coherence of the narratives by means of repetition of key words. This particular example shows several incidences of parallel progression and extended parallel progression, which are the most common in the data collected from the children in this study. The boldface words are the topical subjects of each T-unit. 1

6

Yo me cede a cque a dormir con mi a wela y mi a guelo me compro bombas de havon y los compro tamien me compro una pelota ~ a~ y huvamos con mi n na bombas de havon ~ a~ y tambien huvando futbol con mi n na

7 8 9 10 11 12 13 14 15

pero en S
abado fuimos a la sienda bimos un sapo y mi tio le boto a el sapo tomo awa el sapo entonses mi papi ceria sabar a sapo y stuimos a buscar palos y mi tia busco un palo busco un palo grande y mi papi le sarbo un sapo.

2 3 4 5

(I stayed to sleep with my grandma) (And my grandpa bought me soap bubbles) (and he bought them) (Also he bought me a ball) (And I played soap bubbles with my little sister) (And I also played football with my little sister) (But on Saturday we went to the farm) (We saw a toad) (And my uncle threw away the toad) (The toad drank water) (So my daddy wanted to save the toad) (And we went to look for sticks) (And my aunt looked for a stick) (She looked for a big stick) (And my daddy saved a toad)

With this analysis it is possible to see how the child creates a narrative by describing activities and repeating key terms as the story progresses. Lines

105

Early Childhood Writing in Ecuador

2–3 have a parallel progression with the repetition of the topical subject bombas de jabon and then los, and most of the second half of the narrative uses parallel progression with the repetition of el sapo and then un palo. The two instances of extended parallel progression are the repetition of bombas de jabon in line 5 and el sapo in line 15.

Results and Discussion Tables 1 and 2 give a summary of basic details of the narratives from the two languages. Starting with the total number of words, it is clear that the children wrote much longer narratives in Spanish (714 words) than in English (528 words). Although the school’s formal instruction focuses on writing in English and the children have much more practice writing in English, the children are apparently more fluent in their native language. Reflecting this confidence in Spanish and the longer narratives, the number of T-units in Spanish (average 10.7) is greater than in English (average 8.0). It is interesting to note, however, that the number of words per T-unit is virtually the same in the two languages (6.93 in Spanish, 6.94 in English). This suggests that these children have a similar syntactic ability in both languages. However, this does not reflect earlier studies which have described Spanish writing as having longer sentences on average than English. In looking at the errors, it can be seen that there are almost twice as many errors in Spanish (average 29.2 per essay) than in English (average 16.6 per essay). However, since the essays were longer in Spanish than in English, it is more revealing to look at error-free T-units and number of errors per Tunit. There were twice as many correct T-units in English as in Spanish. The ratio of error-free T-units to total T-units in Spanish was 7 out of 107 (6%), while in English it was 15 out of 80 (18%). But looking at the average

Table 1 Spanish data Text No. of No. of No. of words No. of error-free No. of No. of errors identification words T-units per T-unit T-units errors per T-unit A

48

6

8

1

16

B

70

8

8.75

2

9

1.125

2

C

47

8

5.87

1

13

1.625

D

111

20

5.55

0

74

3.7

E

30

12

2.5

3

17

1.4

F

120

11

10.9

0

27

2.45

G

92

15

6.13

0

34

2.26

H

98

12

8.25

0

44

3.66

I

45

6

7.5

0

23

3.83

5.88

0

35

7

292

J Total Average

53

9

714

107

71.4

10.7

69.3 6.93

29.2

3.88 25.9 2.59

106

Bilingual Education in South America

Table 2 English data Text No. of No of No. of words No of error- No. of No. of errors identification words T-units per T-unit free T-units errors per T-unit A

55

8

6.87

0

27

B C

39

5

7.8

2

9

1.8

59

12

4.9

7

6

0.5

D

29

4

7.25

0

17

4.25

E

27

5

5.4

0

14

2.8

F

84

11

7.6

4

12

1.09

G

66

7

9.4

0

26

3.7

H

49

9

5.4

0

22

2.4

I

49

5

9.8

0

9

1.8

J

71

14

5.07

2

24

1.7

528

80

69.49

15

166

23.4

Total Average

52.8

8.0

6.94

16.6

3.375

2.34

number of errors per T-unit, the numbers for the two languages are very similar. This reveals that although there were more T-units without errors in English, the rest of the T-units had more errors than the average Spanish T-unit. This reflects the training in English (the children were able to correctly write a number of simple sentences in English), but also shows guessing strategies to write unknown words in both languages. In other words, there was better control of familiar words and sentences in English, but a higher number of errors when confronted with unfamiliar sentences, while in Spanish, the students had a more consistent number of errors in all sentences. To understand the errors better, Tables 3 and 4 show the distribution of error types across the two languages. As can be seen, while spelling is the most prevalent error in both languages, the other errors are quite different. There is much more variety of errors in English, including invented sound spellings, such as animols, warol (world), and Mayami; misspellings of words that they have used and=or read in school, such as broder (brother) and frinds (friends); and Spanish-influenced spellings, for example vi (bee) and tri (tree). Spelling errors in Spanish are similar, with some sound spellings, such as univercidat (universidad) and pstel (pastel); some misspellings of words that should be known, for example ben (bien); and some English-type spelling, such as wevo con aros (huevo con arroz). There are also a number of errors in words with silent letters in Spanish, such as ora (hora), uevo (huevo) and acienda (hacienda). Ammon (1985) suggests that invented spellings can be a strategy used by young children when faced with the rather daunting task of writing in a relatively unknown language. While they may have the tools to spell correctly, the demands of writing a coherent narrative may short-circuit the children’s ability to spell. As Ammon points out, ‘if a writer’s knowledge of

107

Early Childhood Writing in Ecuador Table 3 Error types in Spanish Text identification

Spelling

Word spacing

Tildes

A

10

2

4

B

4

2

3

C

5

1

4

D

20

8

24

E

12

1

4

F

18

G

18

2

12

H

29

3

12

I

15

3

2

J

15

1

6

146

23

76

Total

Omitted word

Reversed letters

Other

1 22

5

4 2 1 2

2 9

4

35

8

mechanics . . . does not lend itself to automaticity in performance, he may be forced into trade-offs between attention to mechanics and attention to organization at higher levels of discourse structure’ (1985: 77). This appears to be reflected in the data here. In Spanish, the children struggle with the use of accents (in most cases simply not using them) and with word spacing, frequently dividing words that should not be divided. Typical examples come from sample G in Spanish, a wela and a guelo for abuela and abuelo. In English, this is not much of a problem. Edelsky (1982) notes that it is not ‘self evident’ to children where to segment language into distinct words, and this observation is reflected in this data. Where there is more emphasis in instruction (English), there are few errors, but where they are instructed less (Spanish), there is creative use of segmentation. There are instances of reversed letters (usually d for b, some p for q), but this seems to be limited to individual students. Sample D in Spanish has the most errors, particularly with the repetition of the word despue´s with the d reversed (bespues). There is much more tendency in English to omit words than in Spanish; the one instance in Spanish is of a missing verb, which seems to be simply a slip rather than a true error (Yo a una mun~eca del cuerpo humano. The verb after the subject should be visite´ or fui). The omitted words in English are articles (2), subjects (2), prepositions (2), and (1), and do (1). The omitted subjects may result from transfer from Spanish, which does not require a subject to be named. The other errors are representative of developmental errors. For example, in narrative A in English the author has omitted several words. A typical example is I am fli in the box where the indefinite article is missing before fly. The errors that occur most frequently in English but not in Spanish are subject-verb agreement errors (total 6). For example, in narrative F the child writes If I be a fly, which shows that the child has yet to master counterfactual con-

4

23

121

I

J

Total

16

H

9

E

7

14

D

20

4

C

G

7

B

F

17

Spelling

A

Text identification

5

4

1

Word spacing

Table 4 Error types in English

8

1

1

1

5

Omitted word

3

2

1

Reversed letters

6

2

2

2

S-V agreement

5

3

1

1

1

Spanish

2

1

1

Irregular plural

3

1

1

1

Verb tense

2

1

1

Extra words

2

2

Word choice

7

1

1

1

2

1

1

Other

108 Bilingual Education in South America

109

Early Childhood Writing in Ecuador

ditional (although the teacher wrote the phrase ‘If I were a fly’ on the board). There are also irregular plurals (2) and verb tense errors (3), which appear to be developmental errors (for example, from narrative E, the mans is an example of irregular plural). The other miscellaneous errors are not repeated, except for code switching, or the use of individual words in Spanish. This is a developmental strategy to compensate for a limited vocabulary. In summary, an analysis of the types of errors in both languages shows that spelling errors are by far the most frequent error type. The other errors made by these subjects tend to be different in the two languages, showing both transfer of skills and developmental errors in English. Tables 5 and 6 show the types and frequencies of sentence connectors used by the subjects. As can be seen, the students use a much greater variety of sentence connectors in Spanish than in English.

Table 5 Sentence connectors in Spanish A

y

tambie´n

2

3

despue´s

primero

14

1

pero

entonces

1

1

Y despue´s

de alli

B C

6

D

3

E

1

F

9

G

7

H

1

1

I

4

1

J

2

2

1 1

2 4

Table 6 Sentence connectors in English A

and

For

3

1

but

B C D

1

E F

9

G

2

H

1

I J

4

110

Bilingual Education in South America

Tables 7 and 8 show sentence type usage. There are certain similarities in sentence types used by the subjects in the two languages, for example, there is very little use of sentences with S-V-O-PP (2% in Spanish, 3% in English) and PP-S-V-O (7% in Spanish, 3% in English). The sentence types favoured by the students in Spanish are S-V-O (30%) and S-V-PP (35%). While these are popular in the English data as well (23% and 20% respectively), students are more prone to use a simpler sentence structure, S-V, with slightly more frequency (25%). In the Spanish data, only 10% of the T-units reflect this structure. There is also a striking difference in the use of S-V-C sentences. Table 7 Sentence types in Spanish SVPP

SVOPP

SVCPP

PP(S)VO

A

SVO 1

2

1

1

1

B

2

4

C

1

1

2

D

6

4

6

E

9

F

SV

Other

1 1

1

2 1

2

2

G

5

1

1

6

H

3

4

1

1

1

1

I

1

3

J

2

5

2

1

1

Total

27

9

31

2

1

6

12

%

30

10

35

2

1

7

13

SVC

PP(S)VO

Other

4

2

1

Table 8 Sentence types in English SVO

SV

SVPP

SVOPP

A B

1

1

C

4

2

1 4

1

1

1

D

1

E

1

1

F

2

4

G

2

1

1

1

2

H

1

4

1

1

2

3

3

5

Total

15

16

13

2

9

2

7

%

23

25

20

3

14

3

11

1

1 1

I J

1 1

111

Early Childhood Writing in Ecuador

There is a much greater preference for these in English (14%) than in Spanish (1%). While there are some differences, the similarities are enough to say that these children have comparable syntactic ability in the two languages. The preference for S-V and S-V-C sentences in English may reflect the reading programme implemented at the school. These sentence types may be imitations of the stories the children have read. This aspect needs to be explored further to make a definitive statement. The analysis of the topical progression of the T-units in the narratives in this study shows that these young children do use a fair amount of repetition of key terms as a coherence device. Table 9 shows the thematic progression of the narratives as revealed by the topical structure analysis. As can be seen, the use of extended parallel progression (EPP) and sequential progression (SP) is quite similar in the two languages, with the occurrence in Spanish (EPP 10, SP 3) only marginally higher than in English (EPP 9, SP 2). On the other hand, the incidence of parallel progression (PP) is much greater in English (15) than in Spanish (10). There were no instances of extended sequential progression (ESP) in either language. This is probably due to the complexity of this structure. While the numbers are interesting, they are not very revealing about strategies used by the children to make coherent narratives. A look at some of the samples shows tendencies in the development of writing of these young learners. Table 10 gives a review of the basic strategies for organisation used by each student in both sets of data. There is heavy emphasis on simply listing sentences with no coherence devices beyond simple sentence connectors, while some subjects used a certain amount of repetition of topical subjects, giving the impression of internal coherence. Some more advanced writers were able to create narratives that were more coherent, repeating key terms as they developed their narratives.

Table 9 Topical structure analysis of narratives in English and Spanish Spanish PP A

English EPP

1

B C

SP 1

1

PP A

4

B

1

C

D

4

D

E

3

E

F

1

G

6

H I

2

1 3

G

1

1

H

2

4

1

2

4

15

9

1

I J

10

1

F

1

SP

1

2

J Total

EPP

10

3

Total

2

112

Bilingual Education in South America Table 10 Narrative strategies Spanish

English

A

Narrative then list

A

Narrative

B

List

B

Narrative

C

List

C

List

D

List then narrative

D

List

E

List with repetition

E

Narrative

F

Narrative

F

Narrative

G

Narrative

G

Narrative then list

H

List

H

List

I

Narrative

I

List

J

List

J

List with repetition

An example of a simple list narrative in Spanish is sample E. Pime vui al cotopatsi Vi tele Juge pelata Y mei un gol Juge nintendo Bistrec Bi las nuves Jui a la univercidat Vi tele Juge pelota Vi peliculo Vi tele

(First I went to Cotopaxi) (I watched television) (I played ball) (I made a goal) (I played Nintendo) (I saw Shrek) (I saw the clouds) (I went to the university) (I watched television) (I played ball) (I saw a movie) (I watched television)

As can be seen, there is a lot of repetition of ideas, which leads to a relatively high incidence of topical progression although the reality is not a coherent narrative. An example from English (sample I) shows a typical listing of activities without any topical progression. If I am a fly I If I am a fly I If I am a fly I If I am a fly I Mi name will

will go to the will go to the will go to the will go to the be Wilian.

semeteri parc garden a tri (tree)

This example shows how the children can simply repeat a single sentence structure and change the topical emphasis in each, creating an understandable text, but which lacks any real narrative qualities. These first two examples show a very early stage in narrative writing. On the other hand, there were children in both English and Spanish who were a little more advanced in narrative writing and were able to construct real narratives with use of different types of topical progression to show inter-

113

Early Childhood Writing in Ecuador

nal coherence. Some of these narratives appear as independent stories, while others are combined with lists as in the first two examples. This suggests different levels of development and reliance on earlier strategies to complete a writing assignment. The first example, from English (sample F), shows a relatively coherent narrative about being a fly. If I be a fly I eat nectar with all the trees And I fly up and up And I have many flowers I can have more friends so we can have more nectar And we se more more flowers And we can help more with friends so we can go to the home of nectar And there are the babas fly And their are racing And the babaas fly helping And if one peple come we sting the peple And the peple cry.

(and there are the baby flies) (and they are racing)

And this next example (D) from Spanish shows a combination of strategies, starting with a list of activities and then getting into a more coherent narrative about this child’s visit to a chocolate factory. (This is also an example of a child who has difficulties with reversing letters.) Yo besayuna despues me fui al ari-querto bespues me baje bel avion bespues me fui a la casa be my tia bespues salimos a comrar el besayuno bespues juge con mi pera bespues cocinamos wevo con aros bespee comi y me lav los bientes y me acoste ma~ nana sali a conos er las fabricas en te by com asian el Chocolate Primero coja n asucar bespues le metan en una macina granbe besbes ponian cosasen la ma cina bespues ponian el papel y nos bieron bespues me comi el Chocolate bespues tube pue sair a pasi arle a mi pera con mi papi bespues comi en mi casa bespues me fui a la escuela

(Yo desayuno – I eat bre¸ akfast) (Then I went to the airport) (Then I got off the airplane) (Then I went to my aunt’s house) Then we went out to eat breakfast) (Then I played with my dog) (Then we cooked eggs with rice) (Then I ate) (And I brushed my teeth) (And I slept) (Tomorrow I went out to see the factory on TV how they make chocolate) (First they take sugar) (Then they put it in a big machine) (Then they put things in the machine) (Then they put the paper) (And they saw us) (Then I ate chocolate) (Then I had to go out to walk my dog with my daddy) (Then I ate at home) (Then I went to school)

This sample shows a mixture of developmental narration strategies. It begins as a simple list of what the child did during the vacation. Then when the child reaches a point in the narration that is more interesting, there is a switch from simple listing to true narration of a series of events. The child also makes extensive use of sentence connectors, as in an oral narration, but the internal coherence, by repetition of key terms, is what makes this short text a real narrative.

114

Bilingual Education in South America

Conclusions From the data presented here, the first grade children in this study have a similar syntactic ability in both of their languages. Although they wrote longer narratives in Spanish than in English, the number of words per T-unit was almost the same for the two languages, suggesting similar competency in both. There are also data to back up this claim in the analysis of sentence types. While the children did not use as much variety in Spanish (they mostly used S-V-O and S-V-PP) and there was slightly more variety in English with the more use of S-V and S-V-C, the basic sentence types were used in both languages. In terms of errors, there were more correct T-units in English (twice as many), but at the same time more errors in other T-units made total number of errors similar in both languages. This suggests that the students had learned a number of simple sentences in English and were able to use them in their own writing, but when venturing into unknown territory in English, they made many more mistakes than in Spanish, which showed a consistent level of errors across all T-units. Additionally, except for spelling errors, the errors in the two languages were quite different, suggesting that the students have had different experiences writing in each one. The analysis of sentence connectors shows greater variety and number in Spanish than in English, which seems to reflect their oral competence in Spanish. Earlier studies described an elaborate style in Spanish that does not exist in English that is seen in the length and complexity of sentences, but this is not seen in these examples. This is probably because the children are very young and are just learning to write. Also, the intensive instruction in English writing probably has an influence on how they write. It would be necessary to see if older children in this school continue to write without the elaborate style noticed in native Spanish speakers in the other studies. In terms of the topical structure analysis, it has been shown that the children use similar amounts of sequential progression and extended parallel progression in both languages but more parallel progression in English. We can also see the development of narrative strategies in the children, going from simple lists with no internal progression in the form of repetition of key terms, through a combination of lists with repetition and true narratives with more topical progression, to true narratives with sophisticated use of repetition of key terms to provide internal coherence. To know more about how these children develop as writers in both languages, it would be interesting to follow these same children throughout their primary education at this institution to see how their writing develops in both languages. Alternatively, writing samples by children from the other grades could be analysed to see the development by age and grade level. Additionally, it would be interesting to compare this writing to that produced by children in other bilingual schools in Ecuador and throughout South America and to compare it to writing from native English-speaking children of the same age in bilingual programmes in English-speaking countries.

Early Childhood Writing in Ecuador

115

Correspondence Any correspondence should be directed to Dr JoEllen M. Simpson, c=o Escuela de Ciencias del Lenguaje, Universidad del Valle, AA 2188, Cali, Colombia ([email protected]). References Ammon, P. (1985) Helping children learn to write in English as a second language: Some observations and some hypotheses. In S.W. Freedman (ed.) The Acquisition of Written Language: Response and Revision (pp. 65–84). Norwood, NJ: Ablex. Bell, J. S. (1995) The relationship between L1 and L2 literacy: Some complicating factors. TESOL Quarterly 29 (4), 687–704. Connor, U. (1996) Contrastive Rhetoric: Cross-cultural Aspects of Second-language Writing. Cambridge: Cambridge University Press. Edelsky, C. (1982) Writing in a bilingual program: The relation of L1 and L2 texts. TESOL Quarterly 16 (2), 211–228. Eisterhold, J.C. (1990) Reading-writing connections: Toward a description for second language. In B. Kroll (ed.) Second Language Writing: Research Insights for the Classroom (pp. 88–101). Cambridge: Cambridge University Press. Friedlander, A. (1990) Composing in English: Effects of a first language on writing in English as a second language. In B. Kroll (ed.) Second Language Writing: Research Insights for the Classroom (pp. 109–125). Cambridge: Cambridge University Press. Gaies, S.J. (1980) T-unit analysis in second language research: Applications, problems and limitations. TESOL Quarterly 14 (1), 53–60. Hinkel, E. (1994) Native and nonnative speakers’ pragmatic interpretations of English texts. TESOL Quarterly 28 (2), 353–376. Hudelson, S. (1984) Kan yu ret and rayt en ingles: Children become literate in English as a second language. TESOL Quarterly 18 (2), 221–238. Hunt, K.W. (1965) Grammatical Structures Written at Three Grade Levels. NCTE Research Report No. 3. Champaign, IL: National Council for Teachers of English. Kaplan, R.B. (1966) Cultural thought patterns. Language Learning 16 (1), 1–20. Krapels, A.R. (1990) An overview of second language writing process research. In B. Kroll (ed.) Second Language Writing: Research Insights for the Classroom (pp. 37–56). Cambridge: Cambridge University Press. Lautamatti, L. (1987) Observations on the development of the topic in simplified discourse. In U. Connor and R.B. Kaplan (eds) Writing across Languages: Analysis of L2 texts (pp. 87–113). Reading, MA: Addison-Wesley. (Appeared originally in N.E. Enkvist and V. Kohonen (eds) 1978. Text Linguistics, Cognitive Learning and Language Teaching (pp. 71–104). Publications de l’association Finlandaise de Linguistique Appliquee (AFinLA), no. 22. Helsinki: Akateeminen kirjakauppa.) Lux, P. and Grabe, W. (1991) Multivariate approaches to contrastive rhetoric. Lenguas Modernas 18, 133–160. Monta~ no-Harmon, M.R. (1991) Discourse features of written Mexican Spanish: Current research in contrastive rhetoric and its implications. Hispania 74, 417–425. Reid, J. (1990) Responding to different topic types: A quantitative analysis from a contrastive rhetoric perspective. In B. Kroll (ed.) Second Language Writing: Research Insights for the Classroom (pp. 191–210). Cambridge: Cambridge University Press. Reppen, R. and Grabe, W. (1993) Spanish transfer effects in the writing of elementary school students. Lenguas Modernas 20, 113–128. Scholfield, P. (1995) Quantifying Language. Clevedon: Multilingual Matters. Simpson, J. M. (2000) Topical structure analysis of academic paragraphs in English and Spanish. Journal of Second Language Writing 9 (3), 293–309. Urzua, C. (1987) ‘You stopped too soon’: Second language children composing and revising. TESOL Quarterly 21 (2), 279–304.

EFL and Native Spanish in Elite Bilingual Schools in Colombia: A First Look at Bilingual Adolescent Frog Stories Claudia Lucı´a Ordo
n~ez


, Colombia Universidad de los Andes, Santafe´ de Bogota In Spanish-monolingual Colombia, social pressures push for access to early 50=50% Spanish–English medium instruction from the age of five. Parents and school administrators consider this the best way to achieve bilingualism. This article takes a first general look at the effects of this type of bilingual education on the Spanish and English oral narrative proficiency of 15-year-old adolescents. Data consist of 72 frog stories told following a picture-book (Mayer, 1969): 18 Spanish and 18 English stories from 15-year-olds with 10 years of bilingual education in a Colombian bilingual school constitute the main sample. The other 36 stories, 18 from 15-year-olds in Colombian monolingual schools and 18 from comparable English-monolingual adolescents from a high school in the Boston area, were used to compare the bilingual stories to monolingual productions. The range and variability of the stories in the bilingual group are discussed, as they are compared to the monolingual stories. General similarities and differences between Spanish and English monolingual and bilingual narratives are analysed. Bilingual stories in both languages show evidence of underdevelopment in relation to monolingual stories, as they are sparse in several linguistic variables that show narrative proficiency (i.e. description of events, evaluative language, logical connections). Qualitative discourse analysis is used to describe exemplary productions. Keywords: bilingualism, elite bilingualism, bilingual acquisition, bilingual schools

Introduction In cases in which education in two or more languages is considered as a desirable option by any sector of a community for any particular reason, it is often a subject of controversy and animated discussion among diverse interest groups. Research on multilingual education in different contexts (i.e. August & Hakuta, 1997; Carey, 1997; Cummins, 1995; Martin-Jones, 2000; Met & Lorenz, 1997; Snow, 1990) suggests arguments for and against, and different possible ways of implementation. The clearest conclusion in the midst of ongoing debates is that the outcomes of multilingual education are a function of the sociolinguistic circumstances surrounding the use of one or more languages within a given language community. Diverse forms of bilingual education have been put into practice and studied as a means of achieving different socially-valued goals. Thus, bilingual education has served to raise the status of minority native languages and to provide support for their maintenance, as in the case, for instance, of countries with indigenous languages in Latin America. Bilingual education has also been used to help minority students achieve the academic level of mainstream students, for example in communities with large immigrant populations.

116

Bilingual Adolescent Frog Stories

117

Finally, in some monolingual countries where knowledge of a foreign language is a socially desirable educational outcome, bilingual education is regarded as an efficient way of acquiring foreign languages. In such settings, there may be little difference in status between the native and foreign languages; a type of situation which is likely to lead to ‘additive bilingualism’ (Cummins, 1976; Lambert, 1974). This is the case in question in the present article. The acquisition of a second language, especially English, is very desirable as an educational outcome in Colombia, as it is all over the world today. In an increasingly globalised world, bilingualism broadens students’ access to advanced education, scientific information and professional opportunities. Thus, there are social and political pressures in Colombia for wide access to early English partial immersion, as the presumed best way to achieve bilingualism. Up to now, early bilingual education has been an urban, middle-class, private school phenomenon. Many Colombian private schools have been created, following or slightly modifying the pedagogical model of the first International Schools established in the country to educate the children of foreigners working there. The result is a partial immersion model, offering early Spanish=English-medium education with approximately 50%=50% time of instruction in each language. In this kind of bilingual model, students use English from kindergarten on (from four or five years of age), mostly for academic purposes in the areas of Science, Mathematics, and (English) Language Arts and mainly within the school environment. On the other hand, they use Spanish for areas such as (Spanish) Language Arts, Social Studies, Religion and the Arts. Extracurricular activities may be conducted in either of the two languages, according to the availability of English-speaking teachers. Language Arts are always offered in both languages to ensure bilingual literacy instruction and to provide ample opportunities for contact with the literature in each language. In most Colombian bilingual schools, literacy instruction is first provided in Spanish and only extended to English in or after the first grade. In this kind of programme, children get to speak English apparently fluently and perform a variety of academic tasks in English. For this reason, parental demand for bilingual education is constantly increasing in Colombia, from the youngest possible age. At present, the model appears widely admired. Furthermore, awareness of the practical advantages of mastering a second language is generalised, and there has been serious interest on the part of policy makers to find ways to provide access to early bilingual education in the public (state) sector. Considerable research has been conducted on the consequences of bilingual education in different situations, mostly focusing on outcomes related to academic achievement, cognitive development, or social=political phenomena related to the use of the native language and its maintenance or attrition. There have been far fewer studies which specifically focus on the quality of language development resulting from bilingual education programmes in different sociolinguistic contexts. Also, in linguistic research conducted in diverse sociolinguistic situations, attention has been mostly on the development of the second or foreign language.

118

Bilingual Education in South America

In the study reported here, I explored the actual linguistic consequences of early partial immersion education in a mostly monolingual environment such as Colombia. I analysed the oral Spanish and English proficiency of Colombian adolescents who have acquired English in a situation of additive bilingualism, as a result of 10 years of bilingual partial immersion education. I assessed proficiency quantitatively and qualitatively by comparing oral narrative productions of these bilingual adolescents in Spanish and English with those of monolingual Colombians and adolescents from the United States. The present article specifically reports on the range and variability in the stories of the bilingual adolescents and on the general differences I encountered between them and monolingual productions.

Methods Subjects I chose 15-year-olds as subjects for the present study, because they are supposed to have practically completed their native language acquisition process (Applebee, 1978; Crowhurst, 1979; Scott, 1988). In addition, these bilinguals have experienced 10 years of bilingual academic partial immersion, so they may be expected to have acquired similar skills in their first and second languages. I studied the narrative skills of a sample of 54 adolescents. Eighteen 15-yearold boys and girls in a private, co-educational Colombian bilingual school, eighteen 15-year-old boys and girls in a Colombian monolingual private school, and eighteen 15-year-old high school boys and girls from the Boston area, all volunteers, produced stories for the study. The Colombian adolescents belong to middle-class families, with universityeducated parents, often at Masters level or beyond. Many parents, especially fathers, hold managerial professional positions. Most mothers also work. I chose the 36 Colombian participants from specific schools, on the basis of descriptive information provided by school administrators about the population they serve. Many volunteers filled out a screening questionnaire on selection criteria, which included items such as not having changed schools since kindergarten and not having special access to English outside school. I also chose the 18 American students from volunteers who filled out a questionnaire on parental occupations and educational levels. Although many aspects of their schooling differ, students in both countries were from socioeconomic strata that could be similarly characterised in their respective contexts. Data I used the wordless picture-book Frog, Where Are You, by Mercer Mayer (1969), to elicit oral narratives. The adolescents first looked through the book carefully, and then told me the story depicted. They continued looking at the book while narrating, but I stayed at a distance, so it was clear that I was not able to see the pictures. Each bilingual subject produced two stories, one in each language. All the bilingual speakers chose Spanish to tell the story first and performed the task in English after a lapse of at least 60 minutes. I considered this choice an indication of the bilingual subjects’ preferred language.

119

Bilingual Adolescent Frog Stories Table 1 Sample Monolingual Stories

Bilingual Stories

Stories from Spanish-monolingual adolescents(18)

Spanish stories from bilingual adolescents(18)

Stories from English-monolingual adolescents(18)

English stories from bilingual adolescents(18)

The use of the primary language first probably increased confidence in performance in the second language and set a cognitive and linguistic standard that was then aimed at in the second language, ensuring the use of each language to the best possible level. The natural challenge implicit in the use of the second language eliminated the danger of the second task being a mere translation of the first. A total of 72 stories were collected for the study as shown in Table 1.

Transcription and coding The stories were transcribed in their entirety. They were then divided into clauses, as defined by Berman and Slobin (1994: 660).1 Verbs which indicate mental states (e.g. querer encontrar, decidir salir ¼ want to find, decide to go out) were considered clauses separate from their accompanying infinitival verbs. These in turn were coded as clauses containing predicted events. If several unified predicates had the same subject, they were separated into different clauses. The coding system has a functional orientation, i.e. it gives precedence to meaning over grammar. Grammatical form is considered important only as it serves to express different meanings (Berman, 1993; Berman & Slobin, 1994; Widdowson, 1978). I coded specific functional and formal aspects of the stories which have been distinguished as indicators of language proficiency in both English and Spanish narratives in previous frog story research. These are: (1) the length of the stories, measured in number of clauses; (2) the number of events comprising the plot; (3) the amount of orientation and (4) evaluation intertwined in the basic sequence of events; (5) the degree of cohesion in the stories, marked by the different aspects listed by Halliday and Hasan (1976); (6) the number and variety of inter-clause connectors used; and (7) the richness in the use of different verbal tenses and aspects to relate events in time. In addition to the seven variables indicating oral proficiency, two aspects indicating lack of proficiency were coded in the present work: formal and reference errors. I assessed reliability of the coding system with a second coder, using six randomly selected Spanish stories, three from the monolinguals and three from the bilinguals, and three randomly selected English stories produced by the bilingual adolescents (20% of the Colombian data). I used Cohen’s kappa (Bakeman & Gottman, 1986) to estimate the corrected-for-chance agreement between raters and obtained values of 72% for narrative devices, 83% for division into clauses and inter-clausal connection, and 96% for tense and aspect.

120

Bilingual Education in South America

In order to establish external validity for the variables chosen to measure Spanish and English oral proficiency, two native speakers of Spanish and two native speakers of English, experienced in narrative analysis for research on language development, assessed holistically the 72 stories produced by monolingual and bilingual adolescents. The raters individually sorted the stories into four levels of proficiency and assigned them scores from 1 to 4. Divergent ratings between each pair of experts were mathematically averaged, finally producing seven possible levels of proficiency among the different groups of stories. Data analysis For the analysis presented in this article I move gradually from the comparison of frequencies of elements in some important variables for each group of narratives into comparative qualitative discourse analysis to characterise basic similarities and differences between the oral proficiency of the bilingual adolescents in Spanish and English. I use the holistic scoring by experts to choose representative narratives to analyse.

Adolescent Narrative Competence: What Is Known We have a partial picture of what monolingual proficiency in the frog story task looks like from research by Bamberg and Damrad-Frye (1991), Sebasti
an and Slobin (1994) and Berman and Slobin (1994) with narratives from nine-year-olds and adults in English and Spanish. This research serves to establish what we can expect 15-year-olds to be able to do in the same task in both languages. Bamberg and Damrad-Frye (1991) use the frog story to analyse the skills of nine-year-olds and adults in producing evaluative comments in native English. In turn, Sebasti
an and Slobin (1994) and Berman and Slobin (1994) discuss skills at the same ages in expressing time perspectives, movement in space, and connections between ideas in frog stories produced in native English and Spanish. All these researchers also report mean lengths of the nineyear-olds’ and adults’ frog stories in clauses. Bamberg and Damrad-Frye (1991) show growth in English from age nine to adulthood in the mean frequencies of four specific types of evaluative devices (frames of mind, hedges, characters’ reported speech, negatives and clauses), from 7.92 to 33.67 per story.2 According to these researchers, adults generally produce significantly more references to frames of mind and hedges expressing non-commitment to the truth value of ideas (e.g. looks like, kind of, probably). Bamberg and Damrad-Frye (1991) report an average length of 46.08 clauses in the stories produced by nine-year-olds. Then they show this average length growing into 79.25 clauses in adult English stories. In turn, based on their own frog story research and on research by Sebasti
an and Slobin (1994), Berman and Slobin (1994) produced descriptions of the skills of proficient narrators in English and Spanish which show that good frog stories in both languages share some characteristics in varying degrees. Proficient story tellers, for example, use a specific anchor tense and introduce multiple variations in time from the anchor tense by indicating anteriority, durativity and simultaneity in different ways. Furthermore, proficient frog

Bilingual Adolescent Frog Stories

121

stories in both languages contain ideas richly connected in temporal, causal and concessive relations and in relations of subordination, allowing for the packaging of events in blocks rather than the narration of sequences of individual events. Berman and Slobin (1994) report an average length of 45.0 clauses in the English stories and 50.8 in the Spanish stories of nine-year-olds. They show these lengths growing into 75.3 clauses in adult English stories and 91.3 in adult Spanish stories.

Results Variability in all groups of stories The expert judges who assessed the stories produced by the 15-year-olds reported having used two of the characteristics of proficient frog story telling discussed by Bamberg and Damrad-Frye (1991) and Berman and Slobin (1994): the amount of information from the pictures included in the stories (related to degree of event elaboration and, thus, to length) and the amount of evaluation. In addition, the raters also assessed the organisation of the stories and their coherence. For organisation they looked for clear introduction, conflict and resolution or for episodes; for coherence they evaluated the relative ease with which it was possible to follow a narrative path and to understand who the agents of the different actions were. The results of these ratings give an initial idea of a very wide range of skills represented in my sample of frog stories (see Figure 1). Expert raters individually grouped the stories produced by 15-year-olds in four scoring groups (which ended up being seven levels after different ratings for the same story were averaged). Most of the stories (73%) received scores between 2 and 3.5, a range that includes at least four levels of narrative proficiency. Ten out of the 72 stories (14%) were rated with the maximum score of 4 and nine with 1 or 1.5 (13%). Of these totals per rating, the two groups of monolingual stories contained more stories rated 3 and higher, than stories rated 2.5 and lower. The English monolingual stories received ratings only equal to or above 2 and had the highest number of highly rated stories (14). In turn, each group of bilingual stories contained more stories rated 2.5 and lower than stories rated highly. The Spanish bilingual stories had the most low-rated ones (12) and the fewest 4s (1). Specific sources of this wide variability may be identified in the variables discussed by Bamberg and Damrad-Frye (1991) and Sebasti
an and Slobin (1994), since these roughly or completely correspond to some of the variables that both the expert raters and I analysed. In similar fashion to these researchers, I divided the stories into clauses to determine their length, a criterion also indirectly used by my expert raters. Amount of evaluation was also key to all evaluators. In fact, I included all four evaluation types studied by Bamberg and Damrad-Frye (1991) in my count of evaluative devices. Furthermore, following these researchers, I quantified appropriate changes in tense and aspect from the anchor tense of the stories. And finally, I also analysed frequencies of temporal, causal and other logical connections. Descriptive statistics of these variables for the four groups of stories in my sample are presented in Table 2.

122

Bilingual Education in South America

Figure 1 Number of stories per holistic rating in each group Table 2 Descriptive statistics for some variables in monolingual and bilingual Spanish and English stories (n ¼ 18 in each group) Variable

Type of story

Range

Mean

Std. Dev.

Clauses

Bil. Spanish Mon. Spanish Bil. English Mon. English

19–64 31–97 21–67 30–94

40.50 53.67 37.83 50.00

13.86 16.55 14.12 16.11

Evaluation

Bil. Spanish Mon. Spanish Bil. English Mon. English

23–105 20–131 18–105 16–105

44.83 67.39 36.50 46.11

20.19 31.87 20.27 21.02

Time

Bil. Spanish Mon. Spanish Bil. English Mon. English

3–31 8–42 1–23 2–24

14.11 21.33 10.06 12.00

7.72 10.76 6.01 6.83

Connections

Bil. Spanish Mon. Spanish Bil. English Mon. English

12–39 13–60 8–34 7–40

21.22 28.00 19.56 16.61

8.31 13.04 7.71 8.76

The table shows that the four variables previously analysed by Bamberg and Damrad-Frye (1991) and Sebasti
an and Slobin (1994) are also key dimensions of variability among the stories produced by 15-year-olds. These variables display wide ranges of values and high standard deviations in all monolingual and bilingual versions. This is especially true of the evaluation variable. Among the four groups of stories, the ones that contain the highest means and the widest variability in all four variables are monolingual Spanish narratives. Monolingual English stories follow in length and evaluation, while bilingual Spanish stories follow in time changes and frequency of connections

Bilingual Adolescent Frog Stories

123

between ideas. All in all, Spanish monolingual stories appear to be considerably richer in evaluation, logical connections and time representation than all other stories. Bilingual stories in both languages display considerably lower means in all four variables than their monolingual counterparts. My monolingual English narrators produced stories that were closer in length to those of the nine-year-old English narrators in other frog story research than to adult narrators’ frog stories (an average length of 50 clauses for my participants, compared to 45=46 clauses for nine-year-olds and 75=79 clauses for adults reported in previous research). Similarly, my monolingual Spanish narrators produced stories that averaged 53.7 clauses, closer in length to Spanish-speaking nine-year-olds’ (50.8 clauses reported by Berman and Slobin) than to adults’ (91 clauses reported by the same researchers). Monolingual 15-year-olds appear, on length alone, to be only part of the way along in the transition from child to adult narrative productions. On the other hand, the mean frequency of evaluation devices per story used in English by the 15-year-olds (46.11) is much nearer the reported adult mean frequency (33.67) than the nine-year-old mean frequency (7.92). Of course the frequencies in my data include more coded evaluative devices that those in Bamberg and Damrad-Frye’s data (1991), but their four types of evaluation devices constitute a large percentage of mine. My data, then, seem to show general very important narrative development in adolescence. In particular, development of the use of evaluative devices in the English frog story narrative is already adult-like, yet is probably still growing in Spanish, as story length still has to develop a lot before adult-like productions are achieved. This is quite interesting, as the expression of evaluation in narrative discourse also marks the use of complex syntactic forms (Labov, 1972). Qualitative discourse analysis of the four kinds of stories produced by the 15-year-olds in my sample helps illustrate the wide range of variability indicated by the holistic ratings and the descriptive statistics in all four groups of stories. It shows stories that maximally display organisation and uses of evaluation, logical connection, tense and aspect, and cohesive devices, as well as stories displaying the minimum of these narrative features. Specifically for poor versions of the frog story, mostly found in the bilingual group, I find two different manifestations of impoverishment: one in which the narrator includes only a bare sequence of events and another one in which the bare sequence is, in addition, composed of only highly generalised, non-specific events. The highest-rated stories Consistent with the common characteristics of proficient frog stories in English and Spanish identified by previous researchers and by the expert raters in the present study, stories produced by 15-year-olds holistically rated as very proficient (4) show great similarities across the four groups. Highly rated stories include most of the events in the pictures. These events are organised in a structure that includes all the narrative elements identified by Labov and Waletzky (1967) and then by Peterson and McCabe (1983) for the classic narrative pattern: an initial orientation to characters and setting, a set of events making up the complicating action and leading to a high point or crisis, and

124

Bilingual Education in South America

a resolution of this crisis. The best stories also refer to all or most of the principal and secondary characters appearing in the pictures, without referential confusion. They show a variety of logical connectors and evaluative remarks, with special emphasis on character mental states and psychological and logical causation. In addition, they make efficient use of verb tense and aspect to represent the varied time perspectives necessary for explaining the complexity of the unfolding events. Natalia’s and Edith’s stories, monolingual Spanish (15;01) and English (15;11) respectively, were scored highest in their groups. Both start their stories with an initial orientation that sets up the context, presenting the main characters and locating the action in a general time and place: the boy’s bedroom at night. (Translations from Spanish are given in bold italics.) Natalia Initial Orientation Tom
as tenı´a un perro que se llamaba Fido y una rana que se llamaba Reina. Tom
as had a dog that was called Fido and a frog that was called Reina. Estaban todos muy felices por la noche y se fueron a acostar. They were all very happy at night, and they went to bed.

Edith Initial Orientation Once upon a time there was a little boy and his dog and his frog. They were all friends. The dog and the frog were the little boy’s friends. And he lived with his parents in a nice house in the country.

Then, a complex set of events constitutes the complicating action in both stories. Both girls narrate in detail all the events depicted in the wordless picture book and include all the characters, successfully shifting the thematic focus among them. Natalia achieves this, in part, by giving names to the main characters and reintroducing them when necessary. She also uses a large number of subordinate clauses to refer to the agency of the secondary characters without diverting attention from the central story. Edith uses pronouns quite clearly and efficiently, and she uses the pronoun he and two different substituting expressions, ‘the boy’ and ‘the little boy’. However, Natalia and Edith do not just narrate what happens and who is involved. The events in the story are accompanied by rich references to the narrator’s perspectives on the characters and the actions, in the form of narrative evaluation. Evaluation is very important in Natalia’s story, especially in the form of references to the mental states of the different characters. They have minds. They get scared and furious, and they want, like and dislike things; they realise, decide, think, and intend to do things. And the narrator constantly reads their thoughts in detail. Edith also represents a heavily evaluated sequence of actions, rich with the thoughts, feelings and intentions of the characters. She finds it especially important to evaluate her story by explaining the reasons why some of the

Bilingual Adolescent Frog Stories

125

events happen. She clarifies why the frog escapes through the window, why the dog got the jar on his head, and why the boy gets upset. Natalia Complicating Action . . .Se asomaron a buscar por la ventana a Reina para ver si se habı´a escapado por allı´, pero a Fido no le salı´a el frasco y le pesaba tanto que se cay
o. Se le rompi
o el frasco. They looked out the window to look for Reina, to see if she had escaped through there, but Fido could not take the jar off and it was so heavy for him that he fell. The jar broke. Tom
as se puso muy furioso porque se asust
o, pero a Fido no le habı´a pasado nada. Tom
as was very furious because he got scared, but nothing had happened to Fido. . .

Edith Complicating Action . . .And then in the morning the boy woke up, and he saw that his frog was gone, and he was very upset. He looked everywhere in his room. . . And the dog had the jar on his head because he was looking inside the jar, but then the dog also fell out of the window. The jar broke. The little boy was somewhat upset because the jar had broken, but it was all right because the dog was very nice. . .

Forms of connection between ideas are varied, frequent and effectively used in both stories. Natalia expresses the basic events in simple sentences and relates them by means of early-acquired connectors like ‘and’. However, since she carefully explains events and character actions, feelings and thoughts, the story is abundant in psychological and logical causation and purpose. As a result, there is rich use of different ways to connect causes, effects, reasons and intentions (but, so, that was why, since. . .). Edith uses because frequently, following her evaluative preference for explaining the causes of events. She also likes adversative relations marked with but, to connect events and to include evaluative remarks in the sequence of events. The result of the use of this varied array of connections between ideas is heavy ‘packaging’ (Berman & Slobin, 1994) of events in both stories. Natalia Complicating Action (Cont.) . . .A Fido se le cay
o el panal y se asust
o mucho al ver que las abejas comenzaban a salir. Fido dropped [per] the hive and got very scared [per] to see that the bees were starting [imp] to get out. Mientras, Tom
as buscaba a su rana en el hueco de un
arbol, pero en el hueco habı´a un bu´ho, el cual sali
o

Edith Complicating Action (Cont.) They looked everywhere, even in a little chipmunk hole, but then the chipmunk came out, and there was obviously no frog there. . . And while the boy was looking in the other places in the forest, like in trees, the bees started chasing the dog. . .. And an owl came out of the hole

126

asustado y hizo caer a Tom
as. Meanwhile, Tom
as looked [imp] for his frog in the hole of a tree, but in the hole there was an owl which emerged scared and made Tom
as fall [per]. Y Fido le pas
o por encima, por lo que estaba siendo perseguido por las abejas. And Fido stepped [per] on him because he was being chased by the bees. . .

Bilingual Education in South America

where the boy had been looking in the tree. . . And he put his hands on what he thought were two sticks just coming out of the ground. . . And then, when the sticks moved, he realised that it was the head of a male deer, so he was on top of the male deer’s head, between the antlers.

Natalia and Edith show remarkable skill in handling time relations and modality. Natalia anchors her narrative in the past and maintains this anchor tense throughout the story, using the perfective and imperfective forms of the past effectively. She is also able to relate complex series of simultaneous events by employing the imperfective to indicate duration in the past and by combining its use with the use of time connectors, past progressive and inceptive forms of verbs (started to get out, meanwhile, was being chased). Furthermore, Natalia skillfully uses the remote past to refer to events that occur before the time of narration (had escaped, had happened, had heard). As in Natalia’s story, the complicating action in Edith’s story includes complex sequences of simultaneous events represented by means of time connectors, simple, progressive, remote past, and inceptive forms of verbs. Edith also uses the remote past effectively and demonstrates skill in the use of complex combinations of these verb forms (. . .while. . .was looking, the bees started chasing. . ., the boy had been looking. . ., they were being chased). The action in both Natalia’s and Edith’s narratives reaches its climax at a heavily evaluated high point when the boy and the dog find the lost frog. Natalia evaluates, building suspense by repeating the key event and carefully describing the thoughts of the characters. Then she releases suspense in the revelation of the reason for the frog’s disappearance. In turn, the high point in Edith’s story is characterised by suspenseful evaluation achieved not only by detailing characters’ thoughts but by making them speak through reported speech and describing their psychological and physical reactions.

Natalia High Point Vieron un tronco. A lo mejor pensaron que ahı´ podı´a estar su rana, pues a ella le gustaba mucho el agua. . . y oyeron ruidos. They saw a log. They probably thought that their frog could be there because she liked

Edith High Point And all of a sudden the boy heard what he thought was croaking. He said, ‘oh that sounds like my frog.’ And the dog heard it too, and their ears perked up. And so the boy told the dog to be very quiet so they wouldn’t scare

Bilingual Adolescent Frog Stories

water very much. . . and they heard noises Decidieron mirar quie´n habı´a ahı´, pero sin asustarlo para que no se fuera. They decided to see who was there but without scaring him, so he wouldn’t escape. . . Por eso era que se habı´a escapado, para poderlos tener y cuidar los hijitos. That was why she had escaped, to be able to have and care for the little children.

127

away what might be the frog, and they peeked over the edge of a log. . .

Finally, Natalia finishes with an elaborate resolution which, again, is rich in her own perspectives as a narrator. She does not leave any event without an explanation. Similarly, Edith’s resolution is well elaborated, as the frog’s and the boy’s decisions are fully explained. Natalia Resolution Como habı´an oido a su mam
a hablar muy bien de Tom
as, saltaron a recibirlo. Since they had heard their mother talk well about Tom
as, they jumped to greet him. Y despue´s de haber conversado alegremente, Fido y Tom
as se fueron con uno de los hijitos, que la rana quiso que lo educara. And after having talked happily, Fido and Tom
as left with one of the little children that the frog wanted Tom
as to educate. Y se fueron. And they left.

Edith Resolution And because the frog was happy with his family and his spouse or her spouse, that was originally the boy’s pet, they gave the boy one of their children because they knew the boy was a good person and should have the frog, ‘cause he liked frogs’ and he was very nice to them. And so the boy went away and was very happy. The End

Santiago, a bilingual 15-year-old, produced stories in both languages which were rated the highest in their groups. The similarities with the best monolingual stories are remarkable, as is the similar skill Santiago displays in both languages. Despite some linguistic inaccuracies in the second language, he achieves in his English story very similar effects to the ones he produces in Spanish.

128

Bilingual Education in South America

In an introductory orientation Santiago presents the main characters. He also decides to give them names, making subsequent textual reference to them in the story easier through the complex unfolding events. Santiago Orientation Juanito tenı´a dos mascotas, Juancho y Juanete. Juanito had [imp] two pets, Juancho, Juanete. Juanete era su sapo, y Juancho era su perro. Juanete was [imp] his frog , and Juancho was [imp] his dog.

There was once a boy named Juanito. His two best friends were his pets Juancho and Juanete. Juanete was a toad and Juancho was his dog.

In the complicating action in both stories, Santiago makes clear reference to all main and secondary characters. He also constantly portrays the characters’ thoughts, feelings and intentions. Instead of the detailed description and evaluation he makes of the broken jar episode in Spanish, Santiago concentrates in English in evaluating by constantly referring to the negative consequences of the characters’ actions and using literary conventions, like reported speech and repetition for emphasis. Santiago Complicating Action Una noche Juanete se escap
o de su jarrita, y por la ma~ nana Juanito y Juancho se preocuparon mucho por d
onde estaba su sapito. . . One night Juanete escaped from his little jar, and in the morning Juanito and Juancho worried a lot about where their little frog was. . . Por error Juancho meti
o la cabeza dentro del jarro de Juanete, y no la pudo sacar. By mistake Juancho put his head inside Juanete’s jar, and he couldn’t take it out. Y cuando se asom
o por la ventana se cay
o y la rompi
o. And when he looked out the window he fell and broke it. Juanito estaba muy bravo con Juancho, pero decidieron que era mucho m
as importante salir a buscar a Juanete.

. . .The next morning, when Juancho and Juanito woke up, they were so very sad about their friend Juanete, so they decided they would go out and look for him. They looked for him in the room, but he was nowhere to be found. . . They started shouting, ‘Juanete, Juanete, where are you?’ But there was no answer. They went out of the house through the window, and they went to some woods that were close to their house. And they continued calling, ‘Juanete, Juanete, where are you?’ Still no answer. . .

Bilingual Adolescent Frog Stories

129

Juanito was very mad at Juancho, but they decided that it was much more important to go out to look for Juanete. . . The complicating action in both of Santiago’s stories also includes complex sets of simultaneous events. Santiago not only achieves a successful description of complex time perspectives by means of connectors of simultaneity and varied use of verb forms, lexical items like ‘still’, inceptive forms of verbs (stop. . .ing, start to. . ., keep. . .ing) and verb repetitions. He also puts events and consequences in parallel adversative (but, instead [of]) and causative (because) structures, so painting a very dynamic picture of all the events happening at concurrent and consecutive times. Santiago Complicating Action (Cont.) . . .Mientras tanto Juancho lo busc
o en un panal, pero las abejas furiosas empezaron a perseguirlo. Meanwhile Juancho looked for it in a hive, but the bees, furious, started to chase him. Juanito entonces intent
o en el hueco de un
arbol, pero en lugar de encontrar a su amigo Juanete encontr
o a un bu´ho furibundo que lo empez
o a perseguir. Juanito then tried in the hole of a tree, but instead of finding his friend Juanete he found a furious owl that started to chase him. Entre tanto Juancho huı´a de las abejas asesinas. Meanwhile Juancho escaped [imp] from the killer bees. . . . . .y Juanito decidi
o que lo prudente era continuar la bu´squeda de Juanete. The owl finally stopped its chase and sat on a tree branch, and Juanito decided that the prudent thing to do was to continue the search for Juanete.

Juanito looked in a hole in the ground, but it turned out to be the home of a mouse. . . Juancho, instead, went barking at a bee hive, but the only thing that came out were angry bees trying to get to him. He ran and ran and ran, but the bees were still after him. Meanwhile Juanito was looking for Juanete in a tree. Instead of the little toad, out came a big owl was very angry [sic] to be disturbed in his nap. And Juancho kept running because the bees were still after him. . .. Both friends kept calling for Juanete, but there was no answer. Instead, a great deer saw them and threw them on a pond.

The high point is punctuated by the characters’ feelings, thoughts and=or words and by an evaluated description of the frog family.

130

Santiago High Point . . .En el arroyo encontraron un tronco enorme. In the pond they found an enormous log. Juanito procedi
o con cuidado: ‘Sshh, c
allate Juancho; puede que allı´ encontremos a nuestro amigo Juanete’. Juanito acted carefully: ‘Shhh, shut up, Juancho; it’s possible that there we will find our friend Juanete. . .’

Bilingual Education in South America

They sat there at the pond, cold and wet. . . They saw a big log in the pond, and they looked behind it. There they finally find Juanete and his girlfriend. They had little toads all over the place.

And finally, Santiago’s resolution, like those in the best monolingual stories, leaves no event without an elaborated, emotional explanation. The explanations are different in each story, but both make the characters’ thoughts and intentions very clear.

Santiago Resolution Juanete habı´a encontrado una familia. Juanete had found a family. Felices de que su amigo Juanete hubiera conseguido la felicidad en el matrimonio, le pidieron a Juanete que les diera uno de sus sapitos para que pudieran tenerlo como mascota. Thrilled that their friend had achieved happiness in matrimony, they asked Juanete to give them one of his little frogs so they could keep it as a pet. Juanete, muy agradecido por la vida que le habı´a dado Juanito, se lo dio con mucho gusto, y quedaron de volverse a ver algun dı´a. Juanete, very grateful for the life Juanito had provided for him, gave it to them very gladly, and they agreed to see each other again some day. Fin The end

Juanito was very happy of finding his friend, and Juanete gave to him one of his children so he could keep as a pet, because he intended to live in the woods forever more. And they were all happy. The end.

Bilingual Adolescent Frog Stories

131

The poorest stories The stories rated at the lowest levels (1 and 2) are also similar across the language groups, revealing two different modalities: the ‘generalised’ story and the ‘bare bones’ story. Both concentrate on the straightforward narration of the series of events, mostly free from other enriching information. In addition, they feature these events in a straight sequence, avoiding the challenge of representing simultaneity and other more complex time and aspect perspectives. Furthermore, the generalised story has an impoverished event structure, jumping between events that encompass whole episodes. Generalising Juan (15; 4), a Spanish monolingual, and Jarrett (15; 6), an English monolingual, provide examples of the generalised story. Although their narratives are structured in a way similar to that in which the proficient stories are, they refer mostly to general events that cover whole sets of actions, thus avoiding detail, complex tense perspectives and difficult textual reference for a variety of characters. Juan Initial Orientation A ver, esta historia trata acerca de un peque~ no ni~ no. . . y le gustan mucho los animales. Let’s see, this story is about a little boy. . .and likes animals a lot. En su casa posee un peque~ no perrito y una rana. El ni~ no es muy feliz con ellos. In his house he possesses a little dog and a frog. The boy is very happy with them.

Jarrett Initial Orientation The little kid is looking at a jar, and his dog seems like it likes the frog since she’s living in.

The complicating action begins in a general way in both stories, but at the same time both Juan and Jarrett try at the start to describe an episode in specific terms. For these boys the beginning of the action is basically composed of two general events, the search in the boy’s room (which Juan practically converts into the general beginning of the search) and the calling out the window. But both get into some specific action related to the dog falling out the window. Juan Complicating Action . . .al levantarse se da cuenta que la rana no est
a. . . .when he gets up he realises that the frog is not there. Entonces comienza la bu´squeda

Jarrett Complicating Action And then the frog sneaks out of his jar that night. And then the little kid wakes up in the morning, and his frog is gone, and he’s sad.

132

de ella. Then he begins the search for her. Sale por la ventana, grita, ojea, pero no encuentra nada. He goes out through the window, shouts, looks, but he doesn’t find anything. Tambie´n su perro. . .tiene un peque~ no accidente y se cae de la ventana. Also his dog . . . has a little accident and falls from the window. Va a recogerlo, se pone sus botas y sale en la bu´squeda. He goes to pick him up, puts his boots on and goes out on the search.

Bilingual Education in South America

. . . so he looks all over his room, and he still can’t find his frog. And he calls his frog out his window, and he still can’t find the frog. . . So they go out on a walk looking for the frog.

The search in the woods is generalised in the two stories. For Juan, it is reduced to two encounters with ‘several animals’ and ‘another series of animals’ and their rejection of the boy. The dog in fact disappears from Juan’s story as the complicating action starts. Jarrett attempts more detail referring to some specific animals, but the basic sequence of the narration is established by general events like going around the woods yelling, finding willow gophers, yelling into holes and trees and finding birds. The specific events (the dog’s knocking off of the hive, the bees’ chasing of the dog and the boy falling from the tree) are isolated from one another and from the general events. They actually get lost in the generality of the sequence. Juan Complicating Action (Cont.) En ella se encuentra con varios animales, con abejas, con peque~ nos castores. . .que. . .furiosos, desesperados o de mal genio, lo rechazan. In it he found several animals, bees, little beavers. . .that. . .furious, desperate or bad-tempered reject him. El ni~ no entonces continu´a con su. . . travesı´a. The boy then continues his journey. Encuentra tambie´n cuervos y otra serie de animales, entre ellos un ciervo que lo empuja por un barranco. He also finds crows and another

Jarrett Complicating Action (Cont.) There’s some bees or something flying around. . . And they find willow gophers but no frogs. The dog actually knocks a bees’ nest from a tree, and it falls on the floor. And they yell into holes and trees trying to find it, and all of a sudden the bees decide to chase the dog. And it looks like the boy falls from the tree, and they find birds, but still no frogs. . .

Bilingual Adolescent Frog Stories

133

series of animals, among them a deer that pushes him over a cliff. . . In the deer=moose episode both boys attempt a more detailed narration again. Jarrett includes more details and some evaluation, but the actions still lack necessary linking events. We never know in either of these stories, for example, how the boy gets attacked by the deer or trapped between its horns. We never know either how he, by himself or with his dog, gets thrown into the lake. Both boys arrive at the high point of their narration abruptly, although Juan at least re-announces the continuation of the search before the frogs are found. But in both stories the final encounter lacks emotion and detail and the final resolution is dry and confusing. We never know if the boy gets his frog back or if it is a different one that he grabs as a substitute. In Juan’s story we never even know if the boy has actually found his own frog. So, again, the end of these stories is marked both by generality and by lack of information. Juan High Point and Resolution . . .Se apoya, se para, y encuentra a unas ranitas que est
an con toda su familia. He leans, he stands, and he finds some little frogs that are with all their family. Y muy feliz coge una de ellas y se despide de las dem
as. And very happy he takes one of them and says good-bye to the others.

Jarrett High Point and Resolution . . .and there’s two frogs, actually a whole family of frogs. And he found his frogs. And then he takes one frog, and he says good bye to the frog family. And I guess he heads home.

Simplification of the narration task is not only present in the event structure and narrator’s perspective but in the reference system, the representation of time and aspect, and the logical connection between ideas. Both narrators avoid complex reference to several characters by using almost exclusively one thematic subject, ‘he’ (the boy) in Juan’s story and ‘they’ (the boy and the dog) in Jarrett’s. The two stories, anchored in the present, limit time to the anchor tense and include very few progressive aspects in the form of lexical items, in Juan’s story (e.g. continues) and continuous present in Jarrett’s (something flying around). Connections between story events are achieved only by simple means. Juan prefers no connection and a few multi-functional ‘and’s. Jarrett repeats ‘and’ continuously, without much apparent meaning. Generalisation is also evident in the bilingual stories. Andre´s (15;6) achieves the most summarised narrative both in Spanish and in English. He does away with all secondary characters and talks only about the general search movements as carried out by the little boy. As a result, the original long, complex story, full of action and characters, becomes a sequence of five basic events: the frog’s escape, the search in the room, the search in the forest, the fall in the lake and the final finding of the frog. Andre´s provides very little extra

134

Bilingual Education in South America

information beyond that. The summary is so succinct that the structure of the story is practically lost. Andre´s Se trata de un ni~ no que tenı´a una rana, que la habı´a atrapado y la tenı´a entre un tarro. It’s about a boy who had a frog, which he had caught and he had inside a jar. Entonces se fue a dormir un dı´a, y a la ma~ nana siguiente encontr
o que la rana se habı´a escapado. Then he went to sleep one day, and the next morning he found that the frog had escaped. Y entonces se preocup
o y empez
oa buscarla por todo el cuarto, y no la encontr
o. And then he worried and began to search for her all around the room, and he didn’t find her. Y entonces sali
o afuera y empez
o a buscarla en el bosque. And then he went outside and began to look for her in the forest. Y en el bosque tuvo un resto de problemas porque habı´a muchos animales peligrosos. And in the forest he had a lot of problems because there were many dangerous animals. Y busc
o por todas partes, en los
arboles, en los huecos, en las piedras. And he looked everywhere, in the trees, in the holes, on the rocks. Y al final se cay
o a un lago, donde encontr
o a la rana, que se habı´a encontrado con otra rana y habı´a tenido hijos. And at the end he fell in a lake, where he found the frog, which had found another frog and had had children.

The story is about a boy that had a frog, and he left it in a tar [sic]. And when he slept, . . .he found that the frog was gone. So he started to search for the frog in his room, but he didn’t find it. And then he decided to go outside and look for it in the woods. . . .but there was a very dangerous place because there were very dangerous animals; and he had many problems looking for it. He looked everywhere, in the trees, in the rocks, and in the caves, and he didn’t found it, until he fell into a lagoon with his dog. And he found it behind a log. And he found out that the frog had found another female frog, and they’ve had children.

Andre´s also uses simplifying reference and event connection strategies. He opts for a thematic single protagonist, making the boy the agent of all events. In this way, he avoids complex shifts of focus among characters’ agency. And

Bilingual Adolescent Frog Stories

135

he uses only multi-meaning, early-acquired connectors (and, so, entonces -then-, y -and-) between ideas. Bare bones In the second version of a poorly told story, the ‘bare bones’ story, the narrator basically recounts a moderately detailed list of events in sequence, without relating them in any other way or richly revealing his personal point of view. This is what Naty (15;16) and Katherine (15;1), monolingual Spanish and English adolescents respectively, do. Even though it is possible to discern the parts of the narrative structure in their stories, the lack of evaluation makes these structural elements unmarked and difficult to identify. Naty Initial Orientation Bueno, era un ni~ no que tenı´a un perrito, y habı´an cogido un sapo, y lo habı´an metido en un frasco. Well, it was a boy that had a little dog, and they had caught a frog and had put him in a jar.

Katherine Initial Orientation In the first picture there’s a boy and a dog at night time, and they are looking at a frog in a jar.

It is in the complicating action that the ‘bare bones’ aspect is most apparent. Katherine attempts a more detailed narrative than Naty does, but both stories feature the same basic events and little more. These events are also barely connected, basically just strung one after the other. This shows in Naty’s story in the lack of logical connectors between ideas and in Katherine’s story in the meaningless repetition of ‘and’ between her sentences. As to reference to the story characters, Naty decides to make both the dog and the boy the agents of all the actions, using a thematic ‘they’, while Katherine refers alternately and monotonously to ‘the boy’ and ‘the dog’. Naty Complicating Action Por la noche, cuando ellos se durmieron, el sapo -que se llamaba Frog- se escap
o. At night, when they fell asleep, the frog – that was called Frog – escaped. Ellos lo buscaron entre las botas, entre el frasco. They looked for him in the boots, in the jar. Salieron, lo llamaron, y el perro, por estarlo buscando, rompi
o el frasco porque se cay
o de la ventana. They went out, called him, and the

Katherine Complicating Action And then the boy and the dog go to sleep, and the frog escapes. And in the morning. . .they see that there’s no frog. And they look all around the room for him. And then they look outside the window. . . And the dog falls out of the window, and the jar breaks; but he’s OK. . . And they look through the forest for the frog. The boy looks in a hole in the ground, and the dog looks at a bee hive. And a little animal comes out of the

136

dog, because of [being] looking for him, broke the glass because he fell from the window. Salieron al bosque, lo llamaron, buscaron entre los huecos del piso, en un panal de abejas, llamaron entre los
arboles. They went out to the forest, called him, looked in the holes in the ground, in a bee hive, called among the trees. . .

Bilingual Education in South America

hole. . .

In the high point and resolution of the story, Naty achieves a slightly greater elaboration of events, in the same way Katherine attempts in the complicating action of her narrative. Naty includes more evaluation here, slowing down the action to create more suspense, describing the frog family in more detail, and explaining that the protagonists find their own frog. Katherine, on the other hand, continues her straightforward narration of events, leaving us asking ourselves if the original frog was found. Naty High Point and Resolution . . .Oyeron un ruido detr
as de un tronco. They heard a noise behind a log. Se asomaron a ver que´ era. They looked to see what it was. Al asomarse vieron dos sapos con hartos hijitos. And when they looked, they saw two frogs with a lot of children. . . Uno de ellos lo reconocieron. . . One of them they recognised. . . Lo cogieron y se despidieron de todos. They took him and said good-bye to all.

Katherine High Point and Resolution And they find a hollow log, and the boy and the dog climb over it. And then they find two frogs and all their frog babies. And then the boy and the dog take a baby frog and go home. The end

Lina (15;10) and Sebasti
an (14;10), bilingual adolescents, do something similar in one of their two stories. Both narratives are limited to the bare events that occur to the main characters and involve each of the secondary characters (the animals in the forest) once. Lina actually eliminates the bees, while Sebasti
an eliminates the mole. Lina Initial Orientation Bueno habı´a una vez un ni~ no que

Sebasti
an Initial Orientation There was a time when there was a

137

Bilingual Adolescent Frog Stories

tenı´a un perro y una ranita como mascotas. Well, there was once a boy that had a dog and a little frog as pets. . .

Complicating Action Esa noche la ranita se escap
o mientras el ni~ no y el perro dormı´an. That night the little frog escaped while the boy and the dog slept [imp]. Y a la ma~ nana siguiente. . .no la encontraron. And the next morning. . .they didn’t find her. Empezaron a llamarla. . .,pero no la encontraron. They began calling her. . .,but they didn’t find her. El ni~ no la busc
o en la madriguera de un topo, pero no la encontraba. The boy looked in the den of a mole, but he didn’t find [imp] her. Despue´s la busc
o en un hueco en un
arbol. . . Then he looked in a hole in a tree. . . Del hueco del
arbol sali
o un bu´ho que lo asust
o, y el ni~ no se fue. . . From the hole in the tree went out an owl that scared him, and the boy went away. . .

boy who had a frog and a dog.

Complicating Action One night, when he was sleeping with his dog, the frog came out of his bottle and escape. . . He came out of his house and began to call for the name of the frog. He began to search for him in every place. The dog find the bee hive and began to bother the bees. And they get out of it so the bee house get down. And the bees began to attack the dog, so they began to run. And the owl also began to bother the boy. . .

Sebasti
an shows a lot of problems with the second language, which make his story sound very confusing. It is possible that these problems show more clearly in this type of story, where the narrator makes fewer attempts to simplify the event structure than in the previous ‘generalised’ stories. Lina is able to convey a clearer story. She also includes more orientation and evaluation. However, she uses repetitive language a lot, instead of adding more detailed events or evaluative comments to the story. Neither of the stories displays complex representation of time or complex reference to characters. To convey duration, Lina uses the imperfective past and the inceptive form ‘begin to. . .’, the same inceptive form that Sebasti
an uses repeatedly. This use of inceptives works in the same way in English and Spanish, so it is possibly the easier way for Sebasti
an to express duration in events. Although his story seems to be anchored in the past, Sebasti
an’s lack

138

Bilingual Education in South America

of control of past forms in the different verbs he uses makes it difficult to decide if he is able to keep this anchor tense throughout. In addition to simplicity and problems with time in the stories, neither uses logical connectors for pairs of ideas, time-related or otherwise. With regard to reference, in Lina’s story the boy is the main character and becomes the single thematic protagonist at the beginning of the search in the woods. In Sebasti
an’s narrative, reference to the characters is limited to the naming of each animal in turn when it appears. In the high point and resolution Sebasti
an attempts a more detailed description of the finding of the frog. His problems with the second language show here in a very long, impossible sentence in English that includes all those details. Lina, in turn, includes more details at the end of her story. But all in all, what these two adolescents produce are two stories that barely tell the basic events that happen to protagonists and secondary characters. Lina High Point and Resolution Y en el laguito, detr
as de un tronco, encontr
o una familia de ranas o de sapos, al pap
a, la mam
a y a los hijitos. And in the little lake, behind a tree, he found a family of frogs or of toads, the dad, the mom and the little children. Y el perro y el ni~ no le pidieron pues un hijito a la familia, y se lo regalaron. And the dog and the boy asked the family for a little child, and they gave it to them. Y dieron las gracias. And they thanked them. Se devolvieron a su casa. They went back to their house. . .

Sebasti
an High Point and Resolution When he get out of the lake, he saw a piece of wood that was in there and get to the other side and saw that there was his frog and that he had found a girlfriend and that also he had many offsprings [sic]. Then he get one of the offsprings [sic] and said good-bye to his frog.

There is a higher incidence of poor stories in the bilingual group. For example, there are four ‘generalised’ stories in each bilingual group vs only one in each monolingual group. And there are only two monolingual English ‘bare bones’ stories and one monolingual Spanish one, while four bilingual teenagers produce ‘bare bones’ stories in both their languages.

Conclusion I have shown how the performance of Spanish and English monolingual 15year-olds in the frog story task varies widely from simple lists of specific or general events to clear and complete accounts of events, highly evaluated

Bilingual Adolescent Frog Stories

139

and connected and containing rich representation of time. Rich and less rich stories do not vary much in their overall organisation, reflecting the fact that all these 15-year-olds were able to produce the basic elements of narrative structure, but they vary in the specific devices used to tell and establish relationships between events. High variability is especially noticeable in evaluation, which is practically absent from some of the poorest stories but is abundant in proficient narratives. Time and connection in the poorest narratives make them poorly marked, simple sequences of events expressed in the same simple tense. This high variability may be a characteristic of adolescent language development and gives an idea of the huge development which is still possible and which may actually occur during this underresearched age span. In comparison to monolinguals, bilinguals show a similar range of variability in length, evaluation, connection and time representation in their frog stories. Yet, they generally received lower holistic ratings, produced shorter stories on average – less rich in all variables indicating narrative proficiency than their monolingual counterparts – and produced more examples of the poorest stories. This gives an initial idea of underdeveloped performance among bilinguals. They seem to avoid complex representation of events, evaluation and time more frquently, and to simplify textual reference by limiting focus to single protagonists in a story originally rich in actions performed by a variety of characters. They also seem to produce considerably less evaluation than do monolinguals, even in the language they chose as their strongest – Spanish. More detailed statistical analysis produces stronger evidence to support this ~ ez, 2000, 2003). However, more initial picture of underdevelopment (Ord
on research is necessary to describe this phenomenon more specifically. Even if we do not and cannot expect bilinguals to be double monolinguals, a more detailed analysis of the possible costs to the first language, of acquiring a foreign one early in school would allow educators to devise ways in which these costs may be minimised. Correspondence ~ ez, Any correspondence should be directed to Dr Claudia Lucı´a Ord
on Universidad de los Andes, Centro de Investigaci
on y Formaci
on en Educaci
on, CIFE, Carrera 1E No. 18 a-70, piso 2, Casita Rosada, Bogot
a Colombia ([email protected]). Notes 1. A clause is a unit containing a single verbal element and its corresponding unified predicate. A single verbal element may include infinitives, participles, semi-auxiliary verbs indicating beginning, end, continuation or repetition of action (e.g. empezar a correr, seguir llamando ¼ begin running, keep calling) and modal verbs (e.g. can. . ., try to. . .). A unified predicate expresses one single situation, activity, event, or state, and includes its location in time and place. 2. I have calculated mean frequencies of the four evaluative types from the total frequencies per type reported by Bamberg and Damrad-Frye (1991).

140

Bilingual Education in South America

References Applebee, A.N. (1978) The Child’s Concept of Story: Ages Two to Seventeen. Chicago: University of Chicago Press. August, D. and Hakuta, K. (eds) (1997) Improving Schooling for Language-minority Children: A Research Agenda. Washington, DC: National Academy Press. Bakeman, R. and Gottman, J. (1986) Observing Interaction. Cambridge: Cambridge University Press. Bamberg, M. and Damrad-Frye, R. (1991) On the ability to provide evaluative comments. Journal of Child Language 18, 689–710. Berman, R.A. (1993) The development of language use: Expressing perspectives on a scene. In E. Dormi (ed.) Language and Cognition: A Developmental Perspective. Norwood, NJ: Ablex. Berman, R. and Slobin, D. (1994) Relating Events in Narrative: A Crosslinguistic Developmental Study. New Jersey: Hillsdale. Carey, S. (1997) Language management, official bilingualism, and multiculturalism in Canada. In W. Grabe (ed.) Annual Review of Applied Linguistics (Vol. 17). New York: Cambridge University Press. Cummins, J. (1976) The influence of bilingualism on cognitive growth: A synthesis of research findings and explanatory hypotheses. Working Papers on Bilingualism 9, 1–43. Cummins, J. (1995) Canadian French immersion programs: A comparison with Swedish immersion programs in Finland. In M. Buss and C. Lauren (eds) Proceedings of the University of Vassa Research Papers – Tutkimuksia No. 92: Language Immersion, Teaching and Second Language Acquisition. Vaasa, Finland: University of Vassa. Crowhurst, M. (1979) On the misinterpretation of syntactic complexity data. English Education 11, 91–97. Halliday, M.A.K. and Hasan, R. (1976) Cohesion in English. New York: Longman. Labov, W. (1972) Language in the Inner City. Philadelphia: University of Pennsylvania Press. Labov, W. and Waletzky, J. (1967) Narrative analysis: Oral versions of personal experience. In J. Helm (ed.) Essays on the Verbal and Visual Arts. Seattle: University of Washington Press. Lambert, W.E. (1974) Culture and language as factors in learning and education. In F.E. Aboud and R.D. Meade (eds) Cultural Factors in Learning and Education. Bellingham: Western Washington State College. Martin-Jones, M. (2000) Bilingual classroom interaction: A review of recent research. Language Teaching (33), 1–9. Met, M. and Lorenz, E.B. (1997) Lessons from US immersion programs: Two decades of experience. In R.K. Johnson and M. Swain (eds) Immersion Education: International Perspectives. Cambridge: Cambridge University Press. Mayer, M. (1969) Frog, Where are You? New York: Dial Books for Young Readers. ~ ez, C.L. (2000) Oral bilingual proficiency of Colombian adolescents. Ord
on Unpublished thesis. Cambridge, MA: Harvard Univeristy Graduate School of Education. ~ ez, C.L. (2003) Habilidades narrativas bilingu¨es en adolescentes colombianos. In Ord
on Memorias del Tercer Simposio Nacional sobre Bilingu¨ismo Y Educaci
on Bilingu¨e. Cali: Universidad del Valle (in press). Peterson, C. and McCabe, A. (1983) Developmental Psycholinguistics: Three Ways of Looking at a Child’s Narrative. New York: Plenum. Scott, C.M. (1988) What happened in that: Structural characteristics of school children’s narratives. Paper presented at the Annual Convention of the American SpeechLanguage-Hearing Association, San Francisco, CA. Sebasti
an, E. and Slobin, D. (1994) Development of linguistic forms: Spanish. In R. Berman and D. Slobin (eds) Relating Events in Narrative: A Crosslinguistic Developmental Study. New Jersey: Hillsdale.

Bilingual Adolescent Frog Stories

141

Snow, C.E. (1990a) Rationales for native language instruction: Evidence from research. In A.M. Padilla, H.H. Fairchild, and C.M. Valadez (eds) Bilingual Education. Newbury Park, California: Sage Publications. Widdowson, H.G. (1978) Teaching Language as Communication. Oxford: Oxford University Press.