Language Learning and Study Abroad: A Critical Reading of Research

Language Learning and Study Abroad A Critical Reading of Research

Celeste Kinginger

Language Learning and Study Abroad

This page intentionally left blank

Language Learning and Study Abroad A Critical Reading of Research Celeste Kinginger The Pennsylvania State University

© Celeste Kinginger 2009 All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No portion of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, Saffron House, 6-10 Kirby Street, London EC1N 8TS. Any person who does any unauthorized act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The author has asserted her right to be identified as the author of this work in accordance with the Copyright, Designs and Patents Act 1988. First published 2009 by PALGRAVE MACMILLAN Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS. Palgrave Macmillan in the US is a division of St Martin’s Press LLC, 175 Fifth Avenue, New York, NY 10010. Palgrave Macmillan is the global academic imprint of the above companies and has companies and representatives throughout the world. Palgrave® and Macmillan® are registered trademarks in the United States, the United Kingdom, Europe and other countries. ISBN-13: 978–0–230–54924–1 hardback This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. Logging, pulping and manufacturing processes are expected to conform to the environmental regulations of the country of origin. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Kinginger, Celeste, 1959– Language learning and study abroad : a critical reading of research / Celeste Kinginger. p. cm. Includes bibliographical references and index. ISBN 978–0–230–54924–1 (alk. paper) 1. Second language acquisition. 2. Language and languages— Study and teaching—Foreign speakers. 3. Foreign study. 4. Intercultural communication. I. Title. P118.2.K53 2009 418.0071—dc22 2009013641 10 9 8 7 6 5 4 3 2 1 18 17 16 15 14 13 12 11 10 09 Printed and bound in Great Britain by CPI Antony Rowe, Chippenham and Eastbourne

To Jon and Sam

This page intentionally left blank

Contents

List of Tables

viii

Acknowledgments

ix

1 Situating Language Learning in Study Abroad

1

2 Measuring Language Acquisition

29

3 Domains of Communicative Competence

69

4 Communicative Settings for Language Learning Abroad

114

5 Language Socialization and Identity

154

6 Interpreting Research on Language Learning in Study Abroad

205

Notes

224

References

225

Name Index

242

Subject Index

246

vii

List of Tables 2.1 2.2 2.3 2.4 3.1 3.2 3.3 3.4 4.1

Assessment of proficiency development Assessment of fluency development Assessment of listening comprehension Assessment of reading and writing Assessment of grammatical competence Research on speech acts Studies of discourse competence Studies of sociolinguistic competence Research on communicative settings for language learning 5.1 Research on language socialization and identity

viii

42 50 59 62 73 84 92 103 116 157

Acknowledgments I would like to thank my colleagues and students in the Department of Applied Linguistics, the Department of French and Francophone Studies, and the Center for Language Acquisition at the Pennsylvania State University for their support of my research and for many inspiring conversations about language learning. I am indebted to David Block and to Aneta Pavlenko for their invaluable comments on the manuscript. My thanks also go to Jill Lake, Priyanka Pathak, and Melanie Blair at Palgrave/Macmillan for taking on this project and seeing it to completion with expertise and humane professionalism. The preparation of this book was partially supported by a grant of sabbatical leave from the College of Liberal Arts at the Pennsylvania State University and by a grant from the US Department of Education (CFDA 84.229, P229A020010) to the Center for Advanced Language Proficiency Education and Research (CALPER) at the Pennsylvania State University. However, the contents do not necessarily represent the policy of the Department of Education and one should not assume endorsement by the Federal Government.

ix

This page intentionally left blank

1 Situating Language Learning in Study Abroad

Preamble In his 1986 autobiographical novel L’étudiant étranger (The Foreign Student), the French journalist and cineaste Philippe Labro recreates and embellishes his college study abroad experience. During the mid-1950s the novel’s 18-year-old protagonist leaves France for a year to become an exchange student in the United States. His destination is a rural college for ‘gentlemen’ located in central Virginia where the White population clings both to antebellum customs of discreet propriety and to its legacy of racial prejudice and segregation. Alternately terrified and thrilled to the core, the young man is plunged into an alternative universe where everything seems new to him, from language, to modes of social organization, to colors of the natural landscape. Eager to blend in, he decides that the best way to learn will be to participate as actively as possible while pretending to have understood not only words, but also gestures, clothing, grooming, flavors, and quotidian pursuits. He shortly discovers that the meanings of linguistic signs are richly intertwined with social convention, sensual experience, and implicit ideologies of gender and race. He gradually discovers, for example, that the word date, as a verb, corresponds to an elaborate, carefully scripted ritual involving drive-in movies and fraternity parties, accessible only to those who own or can borrow a car. In learning to use this word, Labro’s autobiographical ‘I’ profits by a legitimate place in the local world of dating, as well as the desire for this participation. He discovers not only the word’s grammatical functions but also some scripts for the corresponding activity and the implications of variation from those scripts, including the consequences of rule violation, and script variants like blind date and steady date. 1

2

Language Learning and Study Abroad

Understanding the ritual itself is not enough, however: he must also grasp such subtleties as the paradox of 1950s’ moral codes according to which one’s date (the noun) should be a good girl: maximally enticing but virtuously abstemious, tantalizing but inaccessible. Over time, the meaning of such a word might develop multi-sensory levels, linked not only to knowledge of etiquette and proper dress but also to the flavor of certain foods, such as popcorn, hamburgers, or Pabst Blue Ribbon beer, and to a broad array of ‘prior texts’ (Becker 1988), including the portrayal of dating and romance in the pop music of the 1950s (e.g., Marty Robbins’ 1957 ‘A White Sport Coat and a Pink Carnation’ or Danny and the Juniors’ 1955 ‘At the Hop’). In the end, the word might assume a rich emotional resonance associated with specific events in the author’s memory (Kramsch 2006a). Along the way, the foreign student often violates codes whose very existence he does not suspect. At first, these infractions are relatively minor, as when he is brought before a peer disciplinary council (The Assimilation Committee), suspected of failing to smile with adequate sincerity when greeting strangers on campus. Later, his choices lead him far from his initial innocence, and toward shocking realizations of the economic and racial inequities characterizing the American south prior to the civil rights movement. A generation after Labro, as a bookish 19-year-old college student born in central Virginia, I went to France, joining the many American participants in the then traditional Junior Year Abroad. I would leave behind the leafy oasis of Antioch College, having cultivated a sincere smile and an appropriately radical political outlook, and would abandon routine walking meditations on existentialism in the company of my professors. Having dutifully chanted the dialogues in the first-year textbook, put on plays more ridiculous than sublime in high school French class, then crept one-by-one through the pages of Camus’ L’Etranger, leaning heavily on my bilingual dictionary, my plan was to learn French. Apart from knowing that this effort would involve reading more books and talking, I had precious few ideas about how this plan would work. I didn’t know at the time that learning another language would involve learning how to live otherwise: how to dress, how to eat, walk, gesture, and navigate within institutions. Like Labro, I entered the scenes of my first year in France with only vague awareness of my own preconceptions, and I look back on that year as a time not only of considerable malaise but also of continual discovery. Madame Lunes, proprietress of the small family farm where I satisfied Antioch’s cooperative work assignment, taught me to

Situating Language Learning in Study Abroad 3

manufacture goat cheese, to choose the right market piglet, to perfect a vinaigrette, to endure driving uninhibited by the speed limit in a rattling Deux Chevaux, and above all, to talk about family, love, food, religion, and politics. Later on, mainly by way of shared interest in chamber music performance, I acquired friends at school who oversaw my apprenticeship in youth culture and language, patiently shepherding me through the argot-laden pages of their comic books and correcting my pompous, misplaced, or archaic vocabulary. In the classroom, I encountered literary language in a context where the rules for interaction were clearly very different from those in force at Antioch. On the street, in the news, and in diverse conversational settings, I began to recognize that despite lingering echoes of May 1968, history’s yield of bigotry, intolerance, and social inequity was as much in evidence in Montpellier as it had ever been in Virginia. My first study abroad experience, unlike that of Labro’s protagonist, took place under the aegis and direct guidance of my college and its programs: work/study, an agreement with the university in Montpellier, and finally the Paris Seminar run by my beloved late professor, Anna Otten, and intended to acquaint Antioch students personally with Parisian writers and artists such as Roland Barthes, Alain RobbeGrillet, Eugene Ionesco, Nathlie Sarraute, and Fernando Botero. The program appears in retrospect to have been extremely well designed for my purpose. I returned to my studies in the United States with communicative resources in French to match a wide variety of circumstances, because my French-mediated experiences had extended from herding goats and shoveling manure to discussing the viability of poststructuralist approaches to literary analysis. Although attaining this competence had required striking out on my own most of the time, as a member of the cohort from my college, I had also interacted continuously with other Americans and had certainly noticed that my fellow Antioch students, a motley crew of poets, scientists, activists, and magicians, would have been extremely difficult to classify under any rubric, including ‘language learner.’ L’étudiant étranger is a work of fiction based – perhaps loosely – on Labro’s life experiences as a young Frenchman in Virginia, shaped by its narrator’s perspective of time, his mature wisdom, experience, and aesthetic goals. My own reflections on study abroad as a young Virginian in France are similarly crafted by the benefit of hindsight and current intent. Neither of these narratives are works of second-language research, yet at the beginning of a book about language learning in study abroad, it is well to be reminded that language learning cannot

4

Language Learning and Study Abroad

be easily or unproblematically extracted from the whole of lived experience. Autobiographical accounts, literary or otherwise, serve as a vivid reminder that language learners are unique people, ‘not just communicators and problem solvers, but whole persons with hearts, bodies, and minds, with memories, fantasies, loyalties, identities’ (Kramsch 2006b: 251). In reading these accounts and recalling personal experience, we remember that language learning is tied to concrete historical and sociocultural settings and the vast multiplicity of semiotic resources within them (Lantolf and Pavlenko 2001). Language learning is about lexical, grammatical, phonological, or pragmatic forms, to be sure, but it is also about flavors and colors, sights and sounds, in short, embodied experience of the world (van Lier 2008).

Introduction This book outlines the current state-of-the-art in research on language learning abroad, acquainting readers with the variety of approaches in the contemporary literature through scrutiny of the advantages and drawbacks of each. An enhanced awareness of the relationship between study abroad and language learning is much to be desired, for teachers, program designers, researchers, policy makers, parents, and students. Language-teaching professional folklore frames the sojourn abroad as the highlight of academic careers, a time when students emerge from classroom decorum and observe that a foreign language need not always be interpreted as a bloodless academic object, severed from its cultural origin and habitat (Lantolf 2007). Often basing their evaluation on their own transformative experiences, teachers expect students abroad to discover language as a medium of expressive delight and a key to world of difference. However, even if study abroad is a key feature of many language programs, dispassionate evaluation of its effects remains a rare practice. Few programs make the effort to integrate the sojourn abroad into their curriculum (Coleman 1998; Wilkinson 2005). Often, study abroad is taken not only as a learning environment inherently superior to the classroom but also as an excuse not to teach languages (Schneider 2001). From the perspective of pedagogy, then, a clearer understanding of how study abroad functions in the development of students’ language ability might assist in developing reasoned approaches to program design. Among researchers interested in language learning, particularly in academic settings, study abroad represents a compelling but extremely thorny series of questions. A sojourn abroad is normally considered to

Situating Language Learning in Study Abroad 5

be a crucial step in the development of ability to use a language in a range of communicative settings. The design of research on language learning abroad, however, confronts the investigator with a bewildering array of variable features, from the identities, motives, or desires of the learner to the range of chance or deliberate encounters presenting opportunities to learn. Students abroad potentially observe, participate, and communicate in classrooms, homes, personal relationships, service learning, or commercial interactions. Study abroad programs have varying objectives, academic foci, and expectations for student activities. Students abroad may be received with warmth, enthusiasm, and patient assistance, or they may find their presence noted with indifference or even with hostility. They may or may not position themselves wholly or in part as language learners. Nevertheless, researchers in applied linguistics and language education have demonstrated considerable interest in study abroad, especially in the years following the publication of Freed’s landmark edited volume, Second Language Acquisition in a Study Abroad Context (1995a). Numerous projects have explored the language-related outcomes well as the qualities of student experience abroad. The range of approaches characterizing this sub-field of applied linguistics now warrants critical review, to inspire creativity and ensure caution in future research. For policy makers and the public, questions about study abroad often have to do with justifying investment of time and financial resources. Politicians want to know how the common good is served through support of students’ sojourns in other countries, and parents want to know what is gained and what is lost when their children study abroad. In the United States, for example, opinions may be formed not only by the findings of research, but also by a ‘dominant’ policy discourse (Gore 2005) according to which education of true quality can only be obtained at home, or by sensationalist reports of students whose time abroad is spent in rigorous avoidance of study while they bar hop and practice extreme sports with other Americans (Feinberg 2002). It is challenging to sort through the conflicting messages of detractors and supporters of study abroad, and to arrive at a reasonably clear conclusion. In many policy documents or in advertising, meanwhile, study abroad is portrayed as offering numerous benefits, with language learning typically playing a relatively minor part. Students who go abroad, it is claimed, develop greater personal maturity, first-hand knowledge of other lands and peoples, commitment to civic engagement, and intercultural awareness fostering mutual understanding among

6

Language Learning and Study Abroad

nations. Although language is downplayed or made transparent in these documents, as if it were an ancillary skill incidental to other goals, experts on international education such as Falk and Kanach (2002) argue convincingly that foreign language competence is a key prerequisite for intercultural communicative competence and international understanding. It follows from this argument that understanding how languages are learned in study abroad will offer considerable guidance in assessing the general worth of these educational experiences. Much depends, as well, on whether language is understood only as a practical skill or more broadly as social practice and ‘a personal stake which extends one’s identity’ (Murphy-Lejeune 2002: 104). This book explores a variety of research traditions in the research on language learning abroad, including measurement of proficiency, investigation of language acquisition, ethnographic inquiry, and approaches to identity and language socialization. Such a project can aspire to inclusiveness, but not to exhaustiveness. Major exemplars of prominent approaches to research are examined, but I cannot claim to discuss each and every relevant study. Along the way, nonetheless, the reader will discover an emerging research-derived knowledge base useful for the design and assessment of study abroad programs and for projects aiming to promote meaningful student experiences. The reading supplied here is ‘critical’ in the classic scholarly sense. That is, it attempts to move beyond the face value of knowledge presentation and toward recognition of the historical and ideological contexts shaping research agendas and knowledge representation, with the goal of deriving an informed synthesis addressing several key concerns of educators, policy makers, participants, and researchers: What are the contributions of study abroad to language learning? What can educators do to assist sojourners abroad in reaching their language-related goals? What capabilities are necessary, in the contemporary world, to foster successful, satisfying, and educationally relevant student sojourns abroad? Which aspects of language learning abroad have been adequately studied, and which remain insufficiently explored? This chapter includes three sections. The first sets the stage for the project represented in this book, defining study abroad within the larger context of international education and student mobility, and contrasting it with the related phenomena of migration, on the one hand, and tourism, on the other. The second briefly explores policy issues and demographic facts about international education in the United States, Europe, and Japan, whose student citizens participate in study programs abroad in considerable numbers. This discussion shows that even if a

Situating Language Learning in Study Abroad 7

uniform baseline definition of study abroad were to be achieved, the educational meaning of this pursuit, and the research accompanying it, can nevertheless vary considerably from one geopolitical or national context to the next. Finally, the chapter turns to the recent history of second-language research shaping scholars’ agendas in study abroad contexts and to an overview of the book.

Study abroad, academic migration, and tourism If scholars have been mobile throughout the history of scholarship, in the contemporary era of social and economic globalization, study abroad is becoming ever more difficult to apprehend in its entirety. Since the 1950s in particular, cross-border education has expanded in every form. Observers witness annual increases and innovations in program mobility, and record numbers of students enrolled in study abroad programs each year. Over the past three decades, there has a been fourfold increase in the number of students enrolled outside their country of citizenship; from 0.61 million worldwide in 1975 to 2.73 million in 2005 (Organization for Economic Cooperation and Development (OECD) 2007). In its putative benefits for language expertise and intercultural awareness, student mobility has become a priority of policy makers throughout the world. More students are going abroad for a wider range of purposes, and within a broader selection of programs, than ever before. For James Coleman (2006), the first, and somewhat disarming, challenge in defining study abroad is to discern how it is named. Coleman discovered many nomenclatures circulating in the published literature, including: ‘study abroad,’ ‘student mobility,’ ‘residence abroad,’ ‘in-country study,’ ‘overseas language immersion,’ ‘séjour à l’étranger,’ ‘estancia al extranjero,’ ‘auslandsaufenthalt,’ and ‘academic migration.’ Clearly, each of these terms refers in some way to education taking place outside a given student’s home country or region, but each also limits or expands this phenomenon in a specific manner. ‘Residence abroad,’ for example, highlights the value of lived experience in a way that ‘in-country study’ does not, and a ‘séjour à l’étranger’ seems less consequential, both for individuals and for societies, than ‘academic migration.’ Thus when first we encounter it, this plethora of terms suggests real differences not only in ways of organizing study abroad, but also in the concepts underlying this organization. Meanwhile, Block (2007a) noted that the literature on language learning in study abroad over-represents the experiences of students from the

8

Language Learning and Study Abroad

United States and to a lesser extent, Europe and Japan, excluding many conceivable combinations of sending and receiving countries. While we learn a great deal about how US students fare in Europe or Latin America, and we read about the learning trajectories of European students during their year abroad, this literature appears to exclude, for example, African students in Europe, Asian students in Australia, or Latin American students in the United States. An initial step, then, is to define precisely what we mean by study abroad, and to situate this activity and the focus of the book with respect to the related phenomena of migration, on one hand, and tourism, on the other. A look at the larger economic and demographic picture serves at once to clarify how study abroad functions in the broader context of international education worldwide, and to shed some light on the imbalanced representation in research noted by Block. In a publication treating Internationalization and Trade in Higher Education (2004), the Organization for Economic Cooperation and Development1 (OECD) presented statistics on the nature and growth of ‘cross-border education’ as one issue within the wider development of educational internationalization. Cross-border education includes the mobility of people (students and teachers) as well as the mobility of educational programs and institutions. At the level of policy, the report distinguished four national approaches to cross-border education: • A mutual understanding approach affording intellectual and cultural enrichment and stimulus to academic programs and research. Such an approach is exemplified by the European Union’s ERASMUS program funding over a million student exchanges from 1987 to the present. • A revenue-generating approach promoting the services of a country’s higher education system to fee-paying students abroad in an effort to control a large share of the market. This approach has been adopted in Australia, New Zealand, and the United Kingdom, which all have international agencies for the promotion of national interests in the educational marketplace. • A skilled migration approach, such as the report attributes to Germany, aiming to attract highly skilled students who may remain in the host country after their studies, thus countering the economic effects of an ageing society and stimulating academic life and research. • A capacity-building approach within which countries encourage their domestic students to study abroad in order to build or improve the quality of their home educational provisions upon their return.

Situating Language Learning in Study Abroad 9

Countries emphasizing this approach include those where demand for higher education outweighs supply, such as Sri Lanka or Cyprus. The four approaches are not to be understood as mutually exclusive, but rather to exist and to complement one another in various combinations. Identifying these approaches provides a vocabulary with which the report may then characterize both national and regional policies themselves and their occasionally unintended effects. For example, when students from a developing country promoting capacity building take up the opportunities provided in a developed country with a skilled migration approach to recruiting foreign talent, the result, known as brain drain, has more far-reaching consequences than does mere study abroad – it is in fact temporary or permanent migration. Scrutiny of the 2004 OECD report reveals global economic inequities shaping the exigencies of student mobility and the ambitions of student border crossers. For many of the Latin American, African, and Asian students whose experience appears to be excluded from the study abroad literature, cross-border education is driven less by a mutual understanding approach than by economic striving or by a dearth of opportunity for higher education at home. It is not necessarily undertaken in the context of a home institution’s design for academic enrichment but is, potentially at least, a long-term change of locale and life horizons for individual students and for whole populations of skilled professionals. This reality is reflected in the documented purpose and duration of student sojourns abroad by region of origin. Student mobility, rather than institutional mobility, represents the largest share of crossborder education as a worldwide phenomenon. Student mobility further divides into three categories: (1) full study abroad for a foreign degree or qualification; (2) study as part of an academic partnership within a home degree or a joint degree involving institutions at home and abroad; and (3) exchange programs. For students from the Asia-Pacific region, for example, the main modality of cross-border education is the acquisition of a full degree on a fee-paying basis. That is, these students often go abroad for periods of multiple years, earning degrees recognized and valued by their host countries; these degrees may either be imported back to the home country as part of capacity building or they may qualify the student to join the host country’s workforce. In Western Europe, by contrast, only 2 percent of tertiary level students were enrolled in foreign degree programs, and these students came from

10

Language Learning and Study Abroad

countries such as Cyprus or Luxembourg whose educational system did not offer significant opportunity for study at the post-secondary level. The typical form of student mobility in Europe, Canada, and Mexico is a short-term program not exceeding one year in duration. In the United States, 91 percent of study abroad participants were enrolled in programs of a semester’s duration or less at the time the report was issued. In summarizing general trends in student mobility, the report emphasizes overall growth, the attraction of foreign students to OECD countries, and the particular lure of English. By 2004, international student mobility had doubled in the previous 20 years, and the OECD countries hosted the majority of the world’s foreign students (85 percent). Further, of these internationally mobile students in OECD countries, 57 percent were from outside the OECD area. Approximately half of these students were attracted to study in one of the English-speaking countries of the OECD, namely the United Kingdom, Australia, Canada, and the United States. The number of foreign students in Australia had tripled since 1990 and multiplied more than thirteen-fold since 1980. In the United Kingdom, there were three times as many foreign students as there were in 1990, and four times as many as in 1980. In the United States, there had been over half a million new enrollments of foreign students each year since 1999 (Institute for International Education 2007). Thus, while all forms of student mobility are available in principle to all students, in practice the meaning of ‘study abroad’ varies considerably depending on student origin, destination, and ultimate goal. Students from outside the OECD usually enroll in full study abroad for a foreign degree, often in English-speaking countries. For many of these students, study abroad is migration related, particularly in destinations like the United States, where generous scholarship support at the graduate level combines with a capacity-building approach in technical fields. The OECD report of 2004 cited many examples of countries such as Sri Lanka and the Philippines, which were ‘net exporters of skilled labor,’ or of the 50 percent of Chinese and Indian doctoral students in the United States in 2000 who intended to stay upon completion of their studies (OECD 2004: 150). Thus, in the literature on language learning in study abroad, the limited representation of these students’ experience may be due to blurred boundaries between education and other, longer-term activities, and between study abroad and migration. Documentation of these students’ language-learning processes does exist in many forms within the broader literature on second-language acquisition (SLA) and the socialization of international students, but it is not necessarily identified as related primarily to study abroad alone.

Situating Language Learning in Study Abroad 11

The economic perspective represented in the OECD report helps to clarify why the literature on language learning in study abroad primarily represents the experiences of students from Europe, Japan, or Englishspeaking countries such as the United States, Canada, and Australia. These students enjoy relative high levels of economic privilege conveying the right to imagine a geographically stable future. While earning home-based degrees and envisaging mobility as a matter of personal or professional choice, these students may pursue intellectual growth, cultural enrichment, international awareness, and language learning through various forms of study abroad. The possibility of longer-term migration, while not excluded, is rarely the focus of such sojourns. It is, after all, within the context of deliberately temporary sojourns away from home that study abroad becomes a salient phenomenon in and of itself. Typically, then, when these students go abroad in the course of their studies they do so within short-term programs of a year or less, often as part of a local, home-based degree or qualification program. The definition of study abroad that is adopted in this book, therefore, is as follows: a temporary sojourn of pre-defined duration, undertaken for educational purposes. A study abroad experience may fulfill degree requirements or may provide enrichment within a home-based degree program, normally at the post-secondary level. Study abroad, according to this definition, also includes the cases of individuals who go to another country or region temporarily and for educational reasons, often involving language learning. This definition includes the typical experiences of European exchange students, those of Asian students who go abroad in order to learn English, and those of students from the United States, Australia, or Canada who go abroad for a variety of educational purposes, including language learning. This definition helps to situate the field under consideration and to separate it from the broader phenomenon of more explicitly migration-related student mobility. Without excluding students of any national origin, it also demonstrates that language-acquisition research on study abroad tends to focus on the largest populations of students who have recently enjoyed the luxury of sampling foreign realities within a purely educational framework. Study abroad, as defined here, exists on a continuum describing the student’s intended degree of resettling and integration into the host country. On one end of the spectrum lies migration, and on the other, tourism, including opportunities for short-term ‘vacation study abroad’ in which travel is more closely associated with leisure and entertainment than with effort and focused learning. Distinctions are blurred at

12

Language Learning and Study Abroad

both ends, however. In the case of tourism, for example, much depends upon the intentions of individual students. Feinberg (2002) describes cases where short-term study abroad, although situated within an educational framework, was experienced by American students as opportunity for leisure activities such as bungee jumping and recreational drinking, enhanced by exotic backdrops. As English exerts increasing dominance worldwide, Phipps (2006), by contrast, describes the earnest endeavors of Anglophone tourist language learners, arguing for a revised vision of linguistic tourism as a mainstay of intercultural dialogue. In relation to language learning in study abroad, tourism is a state of mind influenced not only by individual proclivities but also by the value societies assign to foreign language competence as a desired outcome of travel. In his critique of the study abroad literature on language learning, Block (2007a) observes a disproportionate emphasis on the experiences of American students. While researchers do examine outcomes and qualities of study abroad by European and Asian students, in the primary English-language research databases the majority of citations referring directly to language learning and study abroad concern students from the United States. A closer look reveals that the foci adopted by researchers from various regions are shaped in part by sociopolitical history and education policy influencing the societal roles ascribed to language and language learning. For American researchers the salience of study abroad is enhanced not only by the life-transforming nature of travel from a geographically isolated country, but also by the part it may play in furthering language acquisition at relatively early stages. European research often investigates issues of intercultural awareness and adaptation for more advanced learners whose border crossing is a routine practice. Researchers examining the experiences of Japanese students abroad are primarily concerned with the acquisition of English for international communication or for academic success at home. The brief historical overviews of related language education policy and practice below illustrate the diversity of purpose represented within research on language learning abroad in the United States, Europe, and Japan.

The United States In the United States, ‘the public and policy makers have traditionally been more inward than outward-looking’ (OECD 2004: 43), and international education is not central to national educational debate. Similarly, language teaching is a marginalized pursuit that has never been in the mainstream of American education (Bernhardt 1998).

Situating Language Learning in Study Abroad 13

Pavlenko (2002a) took a historiographic approach to the emergence of monolingualism in English as a symbol of national unity and identity. In the 18th and early 19th century, the United States was robustly multilingual, with educational policy and practice in support of linguistic diversity, and numerous immigrant groups maintaining their native languages for generations. However, an ideological shift took place during the Great Migration period of 1880–1924, when some 24 million immigrants arrived in the United States. This influx of immigrants from southern and eastern Europe aroused fears of a threat to national unity. In the wake of World War One, these fears combined with anti-German ‘hysteria and xenophobia’ (2002a: 174) to yield a climate conducive to linguistic intolerance. A new rhetoric emerged equating English proficiency with true citizenship, supported by discourses framing English as a language of high ethical and intellectual value. Multilingualism was linked to cognitive deficit, lack of patriotism, and low moral standards. The post-war period witnessed the rise of the Americanization movement, with efforts to assimilate immigrant adults and children through English and civics instruction, policies imposing English as the official language of certain states, and the restriction of foreign language instruction. Efforts were underway to protect the ‘ideological purity’ (2002a: 182) of foreign language classes by assigning them to American-born teachers, withdrawing them from elementary schools, and de-emphasizing interactive skills in favor of reading. In sum, this period gave birth to the pervasive double standard in present-day American ideologies of multilingualism: while middleclass children may be encouraged to acquire some, usually limited, proficiency in French, Spanish, or German, their immigrant peers are discouraged from maintaining their native languages and told to focus on English only. Accordingly, in the overall contemporary picture, internationalization remains marginal on many campuses and is seen as more related to the presence of foreign students and scholars than to internationalizing teaching and learning for American students. At the federal level, efforts at internationalization in the 20th century have typically been dominated by concerns about national security and economic competition. The Cold War was the impetus for the 1958 National Defense Education Act (NDEA), prompted by the Soviet launching of Sputnik and the perceived threat of other nations’ scientific and educational prominence. The NDEA led to considerable investment in developing language and area studies expertise, but did not lead to a sustained push toward internationalization. The Department of Education’s 1965 Title VI program

14

Language Learning and Study Abroad

represented another key initiative, providing grants to institutions for the furthering of language competence and internationalization. The late 1970s and 1980s saw generalized acknowledgment of another threat, this time economic in nature, as President Jimmy Carter’s Commission on Foreign Language issued its report, Strength Through Wisdom (1979). The report linked language ability with global competition, insisting that language instruction be reformed to promote functional abilities. The Bush administration came to power with an agenda for education centered on the reform of elementary and secondary level instruction. In the years following the tragedy of September 11, 2001, however, language education has come into the limelight once again as a response to a threat. In 1991 the Department of Defense had created the National Security Education Program to support the efforts of students learning less-commonly taught languages deemed critical to the armed forces. Current efforts at the federal level include expanding the number of languages offered to include those heretofore less-commonly taught, to devise pathways toward advanced level proficiency, and to increase the number of American students who study abroad. In 2005, the Commission on the Abraham Lincoln Study Abroad Fellowship Program recommended sending at least one million American undergraduate students abroad each year. In this report, widespread participation in study abroad is described as the next major step in the evolution of US higher education, comparable to the establishment of land grant universities in the latter half of the 19th century, or to the Serviceman’s Readjustment Act of 1944 providing financial assistance to veterans pursuing higher education and thereby consolidating the postwar middle class. The report emphasizes the importance of global skill for competitiveness and national security, noting that ‘study abroad is one of the major means of producing foreign language speakers and enhancing foreign language learning’ (p. vi). Enthusiasm for this proposal was such that 2006 was proclaimed the Year of Study Abroad by unanimous Senate Resolution. Even without the Abraham Lincoln Fellowship Program, study abroad participation by American students is on the rise, although its nature is changing. According to the Institute for International Education’s (IIE’s) 2007 Open Doors Report, the most recent available statistics show that the number of US students studying abroad has increased by more than 150 percent over the past decade. In the academic year 2005– 2006, 223,534 American students went abroad on academic programs, an increase of 8.5 percent over the previous year. Meanwhile, however, the Junior Year Abroad, involving foreign language majors, has

Situating Language Learning in Study Abroad 15

passed out of vogue. Most programs involve shorter timeframes and students majoring in social sciences, business, or management. From 1985 to 2003, the percentage of undergraduate students spending an entire academic year abroad dropped from 17.7 to 7.8 percent (IIE 2004). In the 2007 report, the number of students at all levels going abroad for a so-called long-term program of a full academic year account for only 5.5 percent of the total. From 1985 to 2006, the number of business majors going abroad rose from 10.9 to 17.7 percent while the number of foreign language majors dropped from 16.7 to 7.8 percent. Thus, the typical contemporary American study abroad sojourn does not involve a student embarking on a full year of in-country study with language learning expressly prioritized. The IIE’s 2007 report also lists aspects of the American international education experience remaining constant over the history of study abroad. First, a significant percentage of American students abroad choose Anglophone destinations, particularly the United Kingdom (18.2 percent) and Australia (6.1 percent), and the majority opt to study abroad in Europe (58.3 percent in 2005–2006) or Latin America (15.2 percent in 2005–2006). Secondly, study abroad attracts a majority of Caucasian students (83 percent in 2005–2006). The third of four years in college has long been, and remains, the preferred time for a sojourn abroad. Finally, the female to male ratio has remained stable throughout the history of study abroad (Gore 2005), with females in the majority: 65.5 percent female to 34.5 percent male in 2006. In light of these figures, policy documents such as the report of the Lincoln Commission highlight the need to diversify and to democratize the study abroad experience. The typical design and purpose of American study abroad programs merit examination. Since the 1950s, and in fact throughout their history (Hoffa 2006), these programs have involved students of relatively modest foreign language proficiency and limited experience of cross-border travel (Coleman 1997). These programs ‘generally envisage the short-term transfer of cohesive groups of American students to a different geographical base, where they may benefit from formal (classroom) and informal (naturalistic) language learning but without necessarily abandoning an American educational framework and academic/administrative support’ (Coleman 1997: 1). Student and teacher stances toward study abroad are shaped by a view of the college years as a period of transition from childhood to adult phases of life. During this period, the college or university functions in loco parentis to provide oversight of students’ bodies and minds as they discover

16

Language Learning and Study Abroad

their adult predilections and prepare to assume personal and financial responsibilities (Matthews 1997). Whether they are at home or abroad, students receive nurture, guidance, and support for group solidarity and school affiliation. Within the American study abroad experience there is a core dilemma, since students are encouraged to retain a strong identification with their classmates and compatriots while also, independently, seeking out local connections and relationships. Although the number of students participating in study abroad programs has increased dramatically over the past several decades, Gore (2005) notes that study abroad is a feminized phenomenon in the United States and that the proportion of American students abroad remains low, at less than 3 percent of the total number of full-time students enrolled in four-year degree programs. Mystified by the persistent marginality of these programs, despite decades of policy decisions intended to promote them, Gore explores how ‘a constellation of dominant beliefs has coalesced to form an episteme held by the U.S. higher education community’ (2005: 23). Within the dominant discourse, ‘study abroad programs are perceived as attracting wealthy women to academically weak European programs established in a frivolous Grand Tour tradition’ (2005: 24). That is, study abroad is a parenthetical diversion from serious pre-professional endeavors and is appropriate mainly as a decorative finishing touch on the education of elite women. Underlying the perception that study abroad programs are inherently weak is a prejudicial attitude, fueled by poor understanding of overseas educational institutions, to the effect that education of true quality is only available in the United States. Alongside these dominant discourses, Gore also traces the long-term existence of ‘alternative voices’ heard among students willing to face challenge and undergo hardship in exchange for an educational experience unavailable at home. In the alternative view, study abroad is an academically rigorous aspect of training for future careers and for an enlightened perspective by way of the liberal curriculum. Since the tragic events of 9/11, moreover, it is argued that through education abroad, students ‘become effective citizens able to contribute to the nation’s development and the world’s peace efforts, having enhanced their own international knowledge’ (2005: 136). Given the changing demographics of study abroad, the design of mainly short-term programs and their occasionally antagonistic ideological environment as described above, it is clear that language learning is not uniformly prioritized in American study abroad programs. Yet, the English-language research base on language learning abroad includes a disproportionate number of studies involving students from the

Situating Language Learning in Study Abroad 17

United States. To explore this paradox will require a closer examination of this literature to reveal both the salience of study abroad as an environment for language learning and a tendency to focus on the earlier stages of the process. American researchers concentrate on language learning precisely because study abroad is believed to enhance motivation and foster realistic goals in a population with little experience of foreign travel, and because American students often begin their sojourns abroad with relatively low levels of language proficiency.

The European Union By contrast with the American, the Western European2 model of ‘student mobility’ or ‘residence abroad’ has emerged from a vastly different set of cultural, institutional, and political goals. In outlining a European perspective on language learning and study abroad, Coleman (1998) emphasized both the density of ethnic, cultural, and linguistic diversity within Europe and the continent’s history of internal conflict. Following millennia of continuous migration, conquest, and strife, the relatively small geographic landmass of Europe encompasses numerous distinct linguistic and cultural communities, offering readily accessible experience of other lands to young Europeans. Consequently, these young people, or at least those whose socio-economic background is typical of university students, have substantial experience of travel to other countries. Prior to formal periods of residence abroad, 95 percent of students participating in a survey in the mid-1990s had already visited the ‘L2land’ in question, often multiple times. For students of French and German this figure rose above 99 percent (Coleman 1998: 173–174). In their daily lives and family backgrounds, Western European participants in programs of student mobility have often accumulated substantial ‘mobility capital . . . enabling individuals to enhance their skills because of the richness of the international experience gained by living abroad’ (Murphy-Lejeune 2002: 51). These students may come from families of mixed national heritage, or from geographic regions where the crossing of national borders is an everyday affair. Multilingualism is part of their ‘natural habitat’ (2002: 53), and foreign language learning a routine and long-term practice of schooling. Because travel to other lands is easily achieved, and because foreign language study takes place over periods of eight to ten years, these students typically display advanced skills in an additional language or languages enabling them to independently carry out work or study assignments abroad.

18

Language Learning and Study Abroad

Coleman (1998) summarizes the effect of these contrasts in terms of the organization and qualities of European residence abroad versus American study abroad. Whereas American programs supply academic structure and support from the home institution for relatively short sojourns, European models normally call for longer stays of a full academic year. Participants are often placed at universities abroad, either as foreign language assistants or as students, but they may also benefit from work placements. Whereas for American and Continental European students study abroad is usually optional, in the British system highlighted in Coleman’s paper, most students of languages were required to carry out a portion of their studies abroad. Crucially, and in direct contrast to the relatively insular American model, while these students are abroad they are interpreted as self-regulating adults capable of navigating local academic cultures, social situations, and institutional systems. At the level of policy, on a backdrop of its colonial history, of the two World Wars originating in Western Europe, and of more recent conflict in the former Yugoslavia, understanding European international education requires an appreciation of the ‘will to tolerance and integration respecting cultural and linguistic diversity’ (Coleman, 1998: 168). Two institutions in particular, the Council of Europe and the European Union, represent this political will and the practical policies enabling student movement and language learning in Europe. The Council of Europe, formed in aftermath of the World War Two, ‘seeks to develop throughout Europe common and democratic principles based on the European Convention on Human Rights’ (Council of Europe 2008). With 47 member countries, the Council serves as a forum for discussion of numerous social issues, promotes awareness of Europe’s cultural identity and diversity, and plays a significant role in language education policy. The Council’s language policy initiatives are designed to further pluriculturalism and plurilingualism, defined as the lifelong enrichment of communicative repertoires in languages and varieties thereof, to enhance egalitarian approaches to linguistic diversity, and to achieve mutual understanding, democratic citizenship, and social cohesion. The coordinated efforts of the Council’s Language Policy Division and the European Centre for Modern Languages include the development and promotion of universal proficiency standards through the Common European Framework of Reference for Language. Building on earlier and highly influential efforts to articulate functional goals for language learning, the Threshold Level (van Ek 1975), the Framework now shapes language education policy throughout Europe (Byrnes 2007).

Situating Language Learning in Study Abroad 19

The European Union is an economic and political partnership currently involving 27 member states. Inaugurated in 1951 as the European Coal and Steel Company, its original aim was to secure peace between victorious and vanquished nations by turning ‘the raw materials of war’ into ‘instruments of reconciliation and peace’ (Europa 2008). Since the 1950s the original union has expanded geographically and evolved in impact, renamed as the European Economic Community, the European Community, and in 1993, the European Union (EU). While retaining its historic goal of promoting peace, security, and prosperity within Europe, the EU’s explicitly stated contemporary mission includes both continued integration, including the cultivation of citizens’ identification with Europe, and preservation of cultural diversity. A key feature of EU policy, mobility is enhanced by the right of free movement within EU countries as instituted by the Treaty of Maastrict in 1993. Student mobility is interpreted as ‘an instrument of European construction’ (Murphy-Lejeune 2002: 1). Student mobility is supported by a broad array of national and regional initiatives, and by a number of supranational network programs such as TEMPUS (linking Western, Eastern, and South-Eastern countries), the EU–US Cooperation Program, or NORDPLUS furthering cooperation among the five Nordic Countries (Denmark, Finland, Iceland, Norway, and Sweden). However, its most visible political foundation is the EU’s flagship educational initiative, ERASMUS (European Community Action Scheme for the Mobility of University Students). Launched in 1987 with the goal of involving 10 percent of European students in international exchanges, the ERASMUS program provides grants for the organization of student mobility and for the mobility of individual students. It also supports inter-university cooperation and the European Credit Transfer System designed to ensure recognition of studies undertaken abroad. Over the past 20 years, the ERASMUS program has grown considerably: whereas in 1987, 300 universities exchanged approximately 3000 students (Coleman 1998), by 2007 the program involved 90 percent of the universities in the EU and served over 150,000 students and teachers annually (European Commission 2007). The EURODATA project offers statistics on student mobility from the period 1998 to 2003 (Wächter and Wuttig 2006). Participants in ERASMUS programs account for approximately one quarter of all foreign students in Europe, including participants in all other forms of cross-border education. They constitute a majority (87.7 percent) of students on supra-national programs of ‘credit mobility’ (2006: 162) in which participants move from home institutions to host institutions abroad and then return to finish

20

Language Learning and Study Abroad

their studies at home. Students at the undergraduate level account for the majority, with 56.9 percent in 2002–2003 as compared to 41.5 percent at the postgraduate level and 1.6 percent at the level of the PhD. In terms of gender and field of study, the figures appearing in the EURODATA report resemble statistics representing study abroad in the United States, with females in the majority and a preference for specialization in business or technical fields. The distribution of male (40 percent) versus female (60 percent) participants remained steady from 1998 to 2003, with 38.6 percent male and 61.4 percent female students in 2002– 2003. In the same year, the most popular field of study for ERASMUS participants was business and management (21.4 percent), followed by languages and philology (16.3 percent), engineering and technology (10.5 percent), and social sciences (10.3 percent). Debate about student mobility in Europe, not surprisingly, centers around the extent to which it can be said to reach its goals. According to Papatsiba (2006), rationales for EU support in this domain can be summarized under two headings: (1) an economic and professional rationale wherein student mobility would promote the European labor market by predisposing individuals to cross borders more easily in their working lives; and (2) a civic rationale in which student mobility would forge a European consciousness and serve as a means to attain international understanding. Although these goals are to some extent complementary, they also speak to different audiences, with the first rationale appealing to an ‘increasing spirit of utilitarianism’ forged in response to market pressures (Papatsiba 2006: 102) and the second to the humanitarian aspirations of educators. These tensions are highlighted in the context of proposed reforms under the Bologna Declaration of 1999 in which member states agreed to create a barrier-free European Higher Education Area by 2010 through structural convergence of their higher education systems. Examination of the extent to which the ERASMUS program has in fact paved to the way toward such reform reveals disparities in the interpretation of purposes for student mobility along with figures suggesting that the program has much to accomplish before it reaches its goal of 10 percent participation. A study published by a Commission of the European Parliament (cited in Papatsiba 2006: 104) showed that only 1 percent of students took part in ERASMUS mobility schemes during the year in question. Furthermore, under pressure both to foster the excellence of their local institutions and to reinforce national or regional competitiveness, academics do not universally perceive involvement in student mobility projects as a gratifying and professionally relevant activity.

Situating Language Learning in Study Abroad 21

Although there is a growing knowledge base on the linguistic effects of study abroad by European students, particularly at advanced levels, European educational researchers have historically tended to highlight aspects of the civic goal outlined by Papatsiba (2006). Specifically, their concern for the development of intercultural competence in support of harmonious coexistence and mutual understanding overshadows their emphasis on language learning per se. To some extent, this phenomenon may be traceable to simple lack of interest in SLA on the part of European researchers (Coleman 1998). However, given the policy environment of European study abroad, the interpretation of study abroad as a context for autonomous learning, and the strong linguistic preparation of many participants, it may also be the case that the early stages of language acquisition are simply less salient in the Western European context than they are in the United States.

Japan The case of Japan offers another contrasting view of role of study abroad in relation to language learning. In Japan, although in principle various foreign languages are taught, in practice the study of English has become synonymous with foreign language learning and therefore also with the language-related goals of study abroad. According to Kubota (2002), ethnic and linguistic diversity within Japan is growing, but high school level enrollment in foreign languages other than English accounts for less than 1 percent of the total. English is taught at all secondary schools, in courses influenced by the content of university entrance examinations emphasizing reading and grammatical knowledge over speaking and listening skills. University admissions decisions are strongly influenced by scores on these examinations, because ‘it is believed that the English score is highly correlated with students’ analytical and logical thinking skills’ (Butler and Iino 2005: 30). Alongside the formal school system, many privately run institutes teach English conversation skills, thereby addressing an unmet yet widely perceived need. Butler and Iino (2005) outline the history of English-language education in Japan noting a tension, since the latter half of the 19th century, between English for practical purposes and English for university entrance examinations. The US–Japan Treaty of Amity (1858) signaled the end of Japan’s official isolation policy and the beginning of an era celebrating foreign language learning for practical purposes as a hallmark of modernization. When the formal secondary level education system was established in the 1890s, English became the main foreign

22

Language Learning and Study Abroad

language offered at school. During the period of intense nationalism following victory in the Japan-Sino and Japan-Russo wars, communicative English lost much of its luster, yet knowledge of its formal properties retained a role as an ‘academic yardstick’ and ‘screening process for elite education’ (Butler and Iino 2005: 28). Following World War Two, and the reorganization of education by the US military occupation, English regained its role as a tool for communication. However, political and economic recovery from the war was accompanied by yet another return to an emphasis on the role of English in academic gate-keeping. In the latter decades of the 20th century, English-language education became a matter of public and occasionally acrimonious controversy. Among the more notable proposals for reform was Suzuki’s suggestion, in the 1970s, that English-language instruction become independent from the cultures of English speakers and serve primarily as an instrument for the transmission of cultural, economic, and technological information from Japan. Also remarkable was the proposal offered in 2000 by Funubashi and the Prime Minister’s Advisory Cabinet, to make English Japan’s second official language. Certain aspects of Suzuki’s proposal have retained credibility, but the notion that English should gain official status in Japan encountered serious objections, particularly since the emergence of ‘a new type of cultural nationalism’ (Butler and Iino 2005: 32) expressing fears about a negative effect of foreign influence on the purity of Japanese language and culture. Kubota (2002) describes the learning of English as an expression of cultural nationalism in contemporary Japan. At the peak of Japan’s economic development in the 1980s, trade imbalances began to generate international conflict and criticism. These problems were successfully deflected by a bid, on the part of government and industry, for membership in the Western industrialized community. This strategy did not, however, imply a move toward assimilation. Instead, it led to the current situation in which the protection of Japanese identity and cultural practices is juxtaposed with a selective appropriation of Western communicative modes, particularly the varieties of English in use in the United States and in the United Kingdom. Within contemporary kokusaika (internationalization) and gurôbarazêshon (globalization), the purpose of English language learning is to claim power through Westernization and to affirm Japan’s unique identity (Kubota 2002). In policy and practice, a clear aim of foreign language instruction is to express Japanese points of view to the rest of the world, just as Suzuki had suggested. Although foreign language study is promoted as a vehicle for international understanding, the

Situating Language Learning in Study Abroad 23

practice of teaching culture through essentialized images of Japanese versus American or British identity reinforces cultural nationalism and stereotyping. In particular, this practice perpetuates and strengthens ideologies related to nihonjiron, according to which the Japanese constitute a linguistically, culturally, and ethnically homogeneous nation, ‘uniquely unique’ and endowed with a highly distinctive, difficult language (Liddicoat 2007: 34). Overall, the emphasis is ‘based on nationalistic values relating to particular conceptions of Japanese identity rather than engagement with international perspectives and interculturality’ (Liddicoat 2007: 38). Following a period of economic stagnation, current policy once again promotes functional proficiency in English as the language of global economic power. It is commonly believed that the ability to communicate with foreigners, achieved via widespread communicative competence in English, would serve to bolster Japan’s status in the global political and economic arena. In 2003, therefore, the Japanese Ministry of Education, Culture, Science, and Technology (MECST) articulated an Action Plan to Cultivate Japanese with English Abilities (MECST 2003) calling for reform of language curricula, teacher education, and evaluation in the direction of greater focus on communication. Major action items call for assessment of oral communication skills for both teachers and students, increased employment of native-speaking Assistant Language Teachers, creation of opportunities to use English outside of class, and increased participation in study abroad. Among the goals listed under the rubric entitled ‘Improving Motivation for Learning English’ are the following: Promotion of overseas study for high school students With a goal of 10,000 high school students studying abroad annually, support will be provided for overseas programs carried out by high school exchange and overseas study organizations and for the provision of information about locations for overseas study. Promotion of overseas study for university students Short-term overseas study for Japanese students utilizing such means as agreements for international exchanges between universities will be promoted and the provision of information relating to overseas study will be enhanced. The combined effects of longer-term post-war economic growth and government policy in support of international education have in fact

24

Language Learning and Study Abroad

enhanced the accessibility of study abroad for Japanese students (Ayano 2006). The IIE Open Doors report (2007) notes that 1.5 percent of Japanese tertiary students studied abroad in 2004, with the majority in English-speaking countries (40,835 in the United States, 6395 in the United Kingdom, and 3172 in Australia). These figures no doubt include not only those students who participate in sojourns of pre-defined length, but also those enrolled in full degree programs. Despite these developments, however, according to Kobayashi (2007), in Japan most people display only beginner-level competence in English. The above-referenced MECST reforms were sparked in part by comparative data showing that average Japanese scores on the TOEFL (Test of English as a Foreign Language) were second to last among 18 Asian countries (Koike 2003–2004). Although English-language testing plays a major role in educational gate-keeping, ‘what is less publicly known is that English is not the major criterion for career advancement in many Japanese corporate sectors’ (Kobayashi 2007: 63). Moreover, in the corporate workplace, the long-standing tension between the symbolic and the practical value of English emerges once again, with preference assigned to the role of English as a sign of academic excellence. Japanese corporate culture places a premium on youth and academic prestige in hiring practices, with the understanding that employment has traditionally been a lifetime engagement and that workers’ identities and work preferences are partially shaped by their company affiliation. Elite employees, usually male, may benefit from corporate support for the use and learning of English, but the majority of workers and most women find little occasion to develop English-language skills at work. In combination with an ideology in which ageing for women is expected to be accompanied primarily by family accomplishments such as child-rearing and caring for the elderly, this situation has encouraged a population of both career-oriented and non-elite single adult Japanese women to study English abroad for personal reasons, combining language education with the realization of desire for travel and experience of Western culture (Piller and Takahashi 2006; Kobayashi 2007). Overall, conflicting ideologies and values surround language learning in study abroad for Japanese students. Officially, English-language competence is framed as a functional skill required of all citizens in the era of worldwide economic competition, but in practice English serves as a general indicator of academic prowess for access to career opportunities. Historically, moreover, ‘even during times when aspects of foreign culture were being brought to Japan, the influence that was adopted was also adapted’ (Koike 2003/2004: 49). In the case of English, Japanese

Situating Language Learning in Study Abroad 25

students face a fundamental dilemma as they are pushed to develop foreign language skills in order to affirm their own national identity. In parallel to the United States and Europe, the applied linguistics research on Japanese students abroad tends to reflect its own sociocultural context. The main focus is of course on the learning of English, and there is a particular recent emphasis on the pragmatics of that language, that is, the ability to navigate in real-world communicative settings, to manipulate speech acts and to enhance politeness. In adopting this focus, applied linguistics researchers appear to be responding to the general call for functional English skills, scrutinizing the study abroad context for its contribution in this domain.

Overview of the book Having established the distinct nature of study abroad in relation to migration and tourism, this chapter proposed three brief overviews of policy and demographics related to language learning abroad by students from the United States, Europe, and Japan. Exploring the broader contexts of research on language learning abroad, this overview elucidates the over-representation of American students. One obvious explanation for this disparity would of course be that North American researchers have considerable access to publication venues in English, the dominant language of international research. However, even a cursory examination of the distinctions in nature and purpose for study abroad in these three contexts suggests another potential explanation. Although researchers in any context do pursue a wide range of projects, their efforts are also shaped by local priorities and constraints. Paradoxically, it may be the low esteem in which language learning is generally held in the United States, and the unremarkable achievements of American language learners that fuel research interest in this topic. For Americans abroad, particularly in non-Anglophone environments, foreign languages become highly salient because they are relatively unfamiliar. Since opportunities for significant language learning are perceived as rare at home, American foreign language educators and applied linguistics researchers frame the study abroad context as a key venue for the development of language competence, even when they do not necessarily examine its qualities in international comparative perspective or question the motives of participants (Kinginger 2008). Among Western European researchers, by contrast, language is examined less often as a concern distinct from the furtherance of intercultural awareness. On a backdrop of Europe’s turbulent history and

26

Language Learning and Study Abroad

current policy explicitly aimed at achieving supra-national affiliation and cultural convergence, European researchers examine the achievements of students who often present advanced language skills and ‘mobility capital’ (Murphy-Lejeune 2002) developed prior to official sojourns abroad. Japanese students normally go abroad in pursuit of competence in one language, English, whose symbolic value is a complex amalgam of responses to global economic competition, affirmation of cultural nationalism, and academic capital within local educational hierarchies. Under these circumstances, it is hardly surprising to discover that researchers examining Japanese students’ experience abroad focus more on the pragmatic dimension of language skill than, for example, on relations between language and intercultural awareness. Like the phenomenon of study abroad itself, the research on language learning in study abroad is extremely diverse, reflecting a broad array of scholarly agendas, sociopolitical agendas, and educational desiderata. What counts as ‘language,’ ‘learning,’ and ‘study abroad’ (or ‘student mobility,’ ‘residence abroad,’ ‘séjour à l’étranger,’ etc.) may vary considerably from one setting or project to the next. To achieve a coherent account of this research, I will adopt a historical perspective from the point of view of applied linguistics from the latter half of the 20th century to the present, that is, in the period accompanying exponential growth in programs involving students in deliberately temporary sojourns abroad for educational purposes. Although the entire range of approaches illustrated here continue to attract active interest among contemporary researchers, a survey of efforts to elucidate language learning in this context reveals that a historical approach is appropriate for two reasons. First, because research on study abroad has mirrored trends in applied linguistics more generally, it is occasionally useful to place particular studies in the context of their time, explaining how their design responded to contemporary issues. Secondly, using chronology as an organizing principle provides the book with something very like a narrative plot, with sub-plots, involving a progression of attempts to unravel the mystery of language learning in study abroad. The story, in brief, is one of increasing emphasis on particularity (van Lier 2005) in which efforts to arrive at generalizations raise more questions than they answer. Chapter 2 reviews the early history of research on language acquisition in study abroad, a field populated mainly by North American researchers, for the reasons outlined previously. The chapter then focuses on research guided by holistic definitions of proficiency or modalities of language use, including fluency, listening comprehension, and literacy-related skills of reading and

Situating Language Learning in Study Abroad 27

writing. Studies reviewed in this chapter emphasize the ‘products’ of participation in sojourns abroad and normally construe the sojourn itself as analogous to an experimental treatment. Because holistic constructs of language ability seemed ill-suited to capture the subtle gains taking place in study abroad, researchers began to look instead at components of communicative competence, examining linguistic, pragmatic, and sociolinguistic dimensions of language ability in separate studies reviewed in Chapter 3. Although tentative generalizations do emerge from product-oriented research, some results have been puzzling, including findings to the effect that study abroad does not necessarily promote the abilities dear to school teachers, such as mastery of discrete-point grammar, and findings showing an amplification of the individual differences in outcomes. These results inspired some scholars to set aside a purely product-oriented analysis in order to examine the process, that is, the activities of study abroad participants. Chapter 4 scrutinizes two kinds of research as they relate to the nature and impact of participation in study abroad: (1) studies attempting to correlate documented language development with student activity; and (2) ethnographic and other case studies investigating the qualities of student participation in specific communicative settings. Languageacquisition researchers taking an interest in the activities of students abroad have attempted in various ways to correlate use of time with developmental outcomes. The foundation of this research is the concept of ‘time-on-task,’ in which study abroad is believed to dramatically increase involvement with foreign languages, and a greater amount of time devoted to foreign language–mediated activity should predict proficiency gains. However, attempts to link language gains to reported uses of time do not always succeed. The story once again is about increasing specificity both in the instruments used to track student activity, and in general appreciation of study abroad as an environment for learning. Rather than assuming that study abroad constitutes an undifferentiated abundance of learning opportunities, researchers began to explore the particular contexts frequented by study abroad participants, the classroom, the homestay, and unplanned informal or service encounters. Ethnographic and other qualitative studies, often without an explicit focus on language, demonstrated both that the experiences of students are quite various and that it is unrealistic to assume a direct relationship between study abroad and linguistic immersion. In the absence of specific links between the practices of study abroad participants and the nature of their communicative repertoires, researchers had yet to establish precisely how study abroad fosters

28

Language Learning and Study Abroad

language competence in ways distinct from classroom learning, and why some students excel in their language learning whereas others present only modest gains. Consequently, with the rise to prominence of research on language socialization and identity, researchers began to apply these perspectives to inquiry on study abroad. Following an introduction to language socialization research, Chapter 5 is devoted to studies problematizing the identities of study abroad participants, both in actual socializing encounters and in reflection on learning. The first section of the chapter examines studies, based on recordings of exchanges in homestay settings, documenting the precise nature of socializing interactions. The second section examines diverse ways in which two aspects of identity, gender and nationality, affect language socialization abroad. The studies reviewed in this chapter attempt in various ways to link the micro-level processes of language learning with their macro-level social and cultural contexts, thus leading in cyclical fashion back to the policies, values, and worldviews referenced in this chapter. Chapter 6 offers a broad overview of the findings presented in this book, reviewing variations on each of the three key terms in the title: ‘language,’ ‘learning,’ and ‘study abroad.’ The limitations of the contemporary research base are articulated in terms of the scope and design of studies published to date, and implications are outlined for policy, language education, and applied linguistics, with particular emphasis on the need for improvement in the ecological validity of study abroad research and for the interpretation of findings in international perspective.

2 Measuring Language Acquisition

A student cited in Colin Evans’ Language People has the following wisdom to share about the role of residence abroad in language learning: ‘There is only one way to learn a language properly. If you want to get to the heart of a language, you go out there’ (Evans 1988: 43). By observing that a sojourn abroad is the sole means to attain profound knowledge of a language, this student echoes both professional folklore and popular impressions of the study abroad experience. But what does it mean to ‘get to the heart’ of language, and how can this knowledge be demonstrated, documented, assessed – in a word – proven? If popular opinion would have us believe that study abroad leads to impressive changes in students’ language ability, can this opinion be substantiated with facts? What evidence has been gathered to support claims about the crucial role of study abroad as an environment for language learning? This chapter scrutinizes the history of attempts, on the part of applied linguists and language educators, to measure the outcomes of study abroad in terms of language ability. Although research on the language-related impact of study abroad is of course an international effort, it is widely recognized that this subfield of applied linguistics is dominated by US-based studies (e.g., Rees and Klapper 2007) exhibiting a strong commitment to utilitarianism (Kinginger 2002). That is, the design of many studies reflects desire to prove the effectiveness of study abroad, with learners developing language abilities analogous to ‘concrete’ products (Lafford 2004: 202). Thus, researchers define language competence in functional terms, striving to answer the kinds of questions that program administrators and policy makers might ask: How much competence do study abroad participants bring back with them, and is it different in kind from the competence of classroom learners? What is the optimum length of a 29

30

Language Learning and Study Abroad

study abroad program? On balance, does study abroad yield results commensurate with the investment required to support it? The history of US-based study abroad research is also linked to the emergence of SLA as an independent field of scientific inquiry (Ellis 2000). As inhabitants of a relatively new scientific domain, SLA researchers are careful to defend their autonomy, working to ‘disentangle’ SLA from language pedagogy (Gass and Selinker 2001: 2). That is, in contrast to educational research emphasizing answers to practical questions, here SLA is most often interpreted as a universal and generalizable psycholinguistic process that may be influenced by different pedagogical treatments (in positive or negative ways) but that does not change in its fundamental nature according to context. For these researchers the study abroad context is a variable, albeit a complicated one, that may exert an indirect influence on acquisition. Context in SLA research has traditionally been reduced to two major possibilities: instructed versus naturalistic language development. Study abroad learners are a hybrid variety, with access to instruction but also with potential increases in their time-on-task and access to the language input and interaction believed to drive the acquisition process. Indeed, the very complexity of the situation may be one of the reasons why SLA researchers in general, as portrayed by Freed (1995a), have been relatively loathe to tackle study abroad, preferring instead to focus on more easily definable learning circumstances such as classroom interventions. In the end, however, the study of SLA in study abroad is usually understood primarily as the study of SLA and not as an inquiry into differential access to learning opportunities shaped by learners’ subjective desires or access to social networks, however interesting these may be. Desire to prove the effectiveness of study abroad for language learning first led researchers toward definitions of global constructs used in testing, such as ‘proficiency.’ With efforts to enhance the subtlety of findings, however, research constructs have been analyzed as component parts of increasing nuance. For example, general speaking proficiency may be interpreted as consisting in part of ‘fluency,’ and ‘fluency’ itself may be analyzed from a number of cognitive and sociolinguistic angles. Often, the meaning of study abroad as a learning context can best be demonstrated when precise definitions replace general categories and the conceptual tools of research gain in sensitivity to the social and interactive features of context. However, refinement in research constructs also generates new questions.

Measuring Language Acquisition 31

This chapter includes two major sections, focusing initially on the earliest studies of language learning abroad still frequently cited in the contemporary literature, Carroll’s (1967) survey of proficiency in college language specialists, and the Schumanns’ (1977) diary studies. The remainder of the chapter reviews research driven by global constructs of language ability, usually related to testing within particular modes of communication such as speaking or listening. These constructs include ‘proficiency,’ ‘fluency,’ ‘listening comprehension,’ ‘reading,’ and ‘writing.’ Research emphasizing outcomes tends to leave some interesting mysteries unsolved. For example, study abroad appears to be advantageous for the development of competence related to frequent practice in social interactive settings populated by expert speakers. Participants often make greater gains than do their peers in home classroom settings in speech rate, the ability to use formulaic language, and fluency. This finding in itself is not particularly surprising. On the other hand, some studies show that study abroad participants do not make particularly impressive gains in academic writing. In light of the fact that most programs and nearly all American offerings emphasize study abroad, these findings are somewhat puzzling. One may wonder what happens both in the instructional contexts that students encounter abroad and in their daily lives, such that academic abilities seem to take a backseat to social interactive competence in enough cases to produce statistically verifiable effects. Also, many of the outcomes-based studies reveal significant individual differences in language development among study abroad participants, and in some cases the effect of these differences seems to be amplified by the in-country sojourn (Huebner 1995). Variation in outcome is usually interpreted in terms of explanatory schemes where the learner’s identity is reduced to an array of individual difference factors such as attitude, motivation, or personality. Differential success is neatly categorized and the story ends, often with the unsuccessful learner-assigned motivational deficit, poor attitude, or inappropriate learning strategies. These studies are often grounded in the previously described view that context of learning has little appreciable effect on language acquisition except to the extent that it does or does not activate the learner’s internal language-acquisition mechanisms. Thus, the nature of study abroad experiences is infrequently depicted, learners’ own unique dispositions toward language learning are not considered, and we do not learn why some students prevail whereas others display only modest gains in documented language ability.

32

Language Learning and Study Abroad

The early history of research on language learning abroad The roots of contemporary approaches to language learning abroad may be traced to the efforts of researchers working in the 1960s and 1970s to understand factors predicting language ability and to explain the interaction of psychology and social context in language acquisition. These studies (Carroll 1967; Schumann and Schumann 1977) illustrate that in research design, the search for explanatory power leads in one of two general directions. For Carroll (1967), the key to valid interpretation lies in robust findings of statistical significance based on language assessment, with the inclusion of a large pool of research subjects. For the Schumanns (1977), the source of insight is located instead within the subjective appraisal of specific learning experiences by individual learners. In Carroll’s study, 2784 subjects were recruited for short-term involvement in a battery of language tests, whereas the Schumann’s subject pool consisted of two people, namely themselves, providing narrative accounts of their personal experience. This contrast in approaches to evidence is further illustrated throughout the literature on language learning abroad, reflecting a similar sharp distinction between the values and methods of logico-scientific versus interpretive research in the broader field of applied linguistics (Ochsner 1979). It is crucial to recognize, however, that in the case of the studies reviewed here, the researchers were equally committed to discovering the causes underlying language acquisition. The earliest examination of the role of study abroad in developing foreign language competence that is still routinely cited is John B. Carroll’s (1967) ‘Foreign Language Proficiency Levels Attained by Language Majors Near Graduation from College.’ As the title indicates, the project was neither exclusively nor directly concerned with study abroad but was an attempt to characterize the range of proficiency attained by college language majors in the United States along with factors associated with variation in performance. The study was a national assessment project carried out with support from the US Department of Education under the National Defense Education Act (NDEA), legislation signed into law in 1958 by Dwight Eisenhower at the height of the Cold War. In addition to monies for science and mathematics education, the NDEA provided substantial federal funding for study abroad, language and area centers, and teacher training in foreign languages. Carroll’s report appeared in the inaugural issue of Foreign Language Annals, the journal of the newly created American Council on the Teaching of Foreign Languages (ACTFL). At the time, Carroll was affiliated

Measuring Language Acquisition 33

with the Educational Testing Service, the organization that would later pair with ACTFL in the development of the Oral Proficiency Interview assessment instrument. In the study, 2784 college seniors majoring in foreign languages (French, German, Italian, Russian, and Spanish) were tested using four instruments: (1) the MLA Foreign Language Proficiency Tests for Teachers and Advanced Students, consisting of subtests in the ‘four basic foreign language skills, Listening, Speaking, Reading, and Writing’ (Carroll 1967: 133); (2) a Professional Preparation Test in English; (3) the Modern Language Aptitude Test; and (4) a background questionnaire detailing foreign language learning history. A subsidiary study, involving participants in NDEA teacher training institutes, determined equivalences between the MLA Proficiency skill subtests and the ‘S’ (Speaking) and ‘R’ (Reading) proficiency ratings of the Foreign Service Institute (FSI) testing procedure based on individually conducted interviews. Because correlations between the MLA and the FSI tests were determined to be very substantial, the research team interpreted the findings of the study in terms of the FSI equivalences. Thus, though it was impractical to administer the FSI interview to 2784 individuals, results are reported as ‘absolute proficiency ratings’ (Carroll 1967: 133). The major findings of the study were of median attainment of audio-lingual skills considered low (FSI Speaking rating of 2+, between ‘limited working proficiency’ and ‘minimum professional proficiency’), of a weakly significant effect for concurrently tested aptitude, and of a strong association of skill level and study abroad: Time spent abroad is clearly one of the most potent variables we have found, and this is not surprising, for reasons that need not be belabored. Certainly our results provide a strong justification for a ‘year abroad’ as one of the experiences to be recommended for the language majors. Even a tour abroad, or a summer school course abroad, is useful, apparently, in improving the student’s skill. The obverse of this finding may be rather humbling for the foreign language teaching profession: those who do not go abroad do not seem to be able to get very far in their foreign language study, on the average, despite the ministrations of foreign language teachers, language laboratories, audio-lingual methods, and the rest. (Carroll 1967: 137) Thus, the Carroll study offered unqualified support for study abroad as a context for language development, along with an offhanded if rather devastating critique of the home-bound teaching practices of the time.

34

Language Learning and Study Abroad

Reading Carroll’s study 40 years after its publication, the benefit of hindsight highlights several of its features. First is the very fact that Carroll’s team, equipped with NDEA funding, marshaled sufficient resources to test 24 percent of the college language majors in the United States (although the original targeted figure was 50 percent). In applied linguistics, comparable feats have rarely been repeated in the interval since 1967 (cf. Coleman 1996). Secondly, although the Foreign Service Institute’s operational definition of language proficiency would not become a major influence in language education until the early 1980s, it was already guiding the interpretation of research results as early as the 1960s. Finally, study abroad is represented as a quantity (amount of time), and presented as a causal factor: any amount of time abroad is useful in developing gain scores on the MLA tests (or their correlational equivalents). A conclusion had been reached, and Carroll’s authoritative and robustly supported claim about study abroad might well have been the end of the story. Indeed, Carroll’s solid claim about the effectiveness of study abroad remained firmly anchored in the background, supporting researchers’ efforts at bolstering the evidence for language learning abroad, refining research tools, and diversifying the contexts of their efforts. Carroll’s report became a landmark classic in a field that went on to problematize in myriad ways both the operationalization of language proficiency via test scores and the treatment of study abroad as an undifferentiated temporal variable. With further development of second-language research, and in the absence of NDEA funding, scholars began to question the notion that experiences and settings as complex as language learning and study abroad could be validly conceptualized as unitary factors. Large-scale studies such as Carroll’s passed out of vogue as attention was directed away from the product and toward the process of language acquisition in a range of specific contexts. This process-focus is illustrated in the work of SLA scholars who took up the cause of interpretive research based on learner diaries, beginning in the 1970s (Ochsner 1979; Bailey and Ochsner 1983). Studies of individual learners’ own stories of experience violated all expectations of the experimental paradigm dominating second-language research, ignoring the accompanying quest for statistical significance. In discussions surrounding the inclusion of research based on first-person accounts, however, the dichotomous relationship between experiment and interpretation is emphasized; an attitude of humility seeps through the scholars’ petitions for acceptance of non-experimental approaches. Ochsner (1979), for example, set forth a rationale for ‘perceptual

Measuring Language Acquisition 35

bilingualism’ based on the equally prestigious pedigrees of the nomothetic and hermeneutic traditions of intellectual inquiry. According to Ochsner, the nomothetic attitude assumes the existence of one ordered, discoverable reality that causally obeys the Laws of Nature. Hermeneutic science (the art of interpretation) assumes that reality varies and that human events must be interpreted teleologically, that is, according to the value of their results. Bailey and Ochsner (1983) go so far as to suggest that creative SLA research might be modeled after literary criticism, proposing a list of standards for judging diary studies based on such qualities as the studies’ ability to reveal otherwise unobservable, yet believable motives, the extent to which they respond to the needs of their readership, and the provision of thorough descriptions of the data-collection process. In diary studies of their own experiences abroad Schumann and Schumann (1977) and Schumann (1980) attempted to complete a causal model of language acquisition through inclusion of a subjective dimension. The authors scrutinized their own experiences learning Arabic in Tunisia, and Farsi in a classroom setting in the United States and in Iran at a time when J. H. Schumann was attempting to develop a causal model of SLA based on a combination of social and psychological factors. As elaborated in Schumann’s case study (1976) of a notoriously unsuccessful learner (Alberto), this model proposed that success or lack thereof be explained in the first instance by factors contributing to solidarity or social distance between the relevant groups. These factors include, for example, the relative status of the groups in relation to each other, the degree of enclosure of the groups, their size and cohesiveness, attitudes on the part of each group toward the other, and integration patterns (assimilation, accommodation, or preservation of group identity). In the event that social distance does not explain language acquisition, the model posited a second level including a variety of psychological factors such as cognitive style and personality variables (self-esteem, ego permeability, anxiety, motivation, and the like). The diary studies built on the notion that a third, additional level is required to complete the model. Individual learners display idiosyncratic ‘personal variables’ (Schumann and Schumann 1977: 247) based on subjective appraisals of learning situations. Schumann and Schumann therefore pointed to neglect of in-depth longitudinal case studies examining the social-psychological profiles of individuals in addition to their personal variables. Their 1977 publication offers short rewritten versions of the logbook data collected abroad, aiming to clarify, first, what the data revealed about the social psychology of second-language

36

Language Learning and Study Abroad

learning that was not previously known, and secondly, ‘what the data told the subjects as individuals about their language learning’ (1977: 243). Curiously, in contrast to the ease with which Alberto’s failure to develop English-language proficiency had been attributed to a pidginization process arising from social distance (Schumann 1976), in this paper there is no analysis of proficiency development, and the authors glossed over the first research question (on the social psychology of language learning) stating that the diaries did not reveal new information. At the level of personal variables, however, the project ‘revealed idiosyncratic patterns of behavior’ (1977: 243) seriously affecting language learning for both subjects. For Francine Schumann, these factors included a withdrawal from learning elementary Arabic, due to noncompliance, in a classroom dominated by a strict interpretation of the audio-lingual method, and an ‘obsession with nesting’ (1977: 243) interfering with any focus on learning. Her nesting instinct was particularly difficult to satisfy in Tunisia where the lodging provided was infested by ‘a strange type of bug . . . that attacked us all summer causing us to look like smallpox victims throughout our stay’ (1977: 244). For John Schumann, the relevant personal variables turned out to be anxiety surrounding transitions, and preferences for a personal agenda and learning strategies sometimes at odds with the practices of the classroom. In summarizing the main thrust of their argument, the authors proposed the following compelling if now somewhat anachronistic analogy for mapping the complexity of individual learning pathways: (The authors) suggest that language learning be seen as analogous to the operation of a pinball machine. The machine is the total language learning situation. The knobs represent the social, psychological, cognitive, and personality variables mentioned above. In addition, there are knobs that represent age, aptitude, and instructional variables. The balls represent the learners and embody the sorts of personal variables that have been described in this paper. As each ball travels through the machine, it hits certain knobs and not others. In addition, it hits certain knobs more than others. The personal variables in each ball determine the path it takes through the machine and the extent to which it is influenced by each knob. No ball will exactly duplicate the path of any other ball, in the same way that no two learners will acquire a second language in precisely the same way. (Schumann and Schumann 1977: 248)

Measuring Language Acquisition 37

Despite the Schumanns’ initial reluctance to assign particular consequence to their own social situation as language learners abroad, in a reanalysis of the 1977 study, Francine Schumann (1980) reconsidered this stance in light of the particular difficulties she experienced as a female learner of Farsi in Iran. In this more socially situated version of the diary study F. Schumann rather prophetically announced a number of themes of which some would later come to prove constant in the qualitative study of language learning abroad. Specifically, she described constraints emerging from her somewhat ambivalent allegiance to a local expatriate community voicing negative feelings about Iran. She also commented on the disadvantages accruing to language learners who are speakers of English, due to the widespread use of that language, particularly in the service encounters where she might have been most likely to encounter expert users of Farsi. Finally, she noted the significant extent to which her languagelearning opportunities were limited by her gender: ‘I’ve come to believe after keeping a language journal that the task of learning the language of a country like Iran is far greater an endeavor for a woman than for a man’ (Schumann 1980: 55). In order to learn a language, she wrote, it is important to practice that language and to be immersed in it, ideally in the so-called target culture where opportunities for engagement in language-learning activity are believed to abound. In a retrospective examination of her earlier Peace Corps training in Farsi, she stated that ‘no one ever informed the female (italics in the original) language learner that in any given daily contact situation in Iran a good many of these opportunities are “off bounds” ’ (1980: 55). To dispel any potential illusion arising from the earlier publication, she then commented on the extreme rarity of her Farsimediated encounters outside the classroom, her social isolation, and the extent to which her independent movement was limited by fear of harassment. Interpreted in light of subsequent developments in second-language research, the Schumanns’ diary studies represent the epistemological limits of the social-psychological model while also tentatively raising the curtain on a more critical approach. That is, the pursuit of causal variables from the social to the psychological to the individual levels leaves the researchers in a position where the best metaphorical representation of the entire phenomenon is the pinball machine, where a hapless and unconscious object, identical to all its companion objects, runs a chaotic and meaningless course through an immutable social structure. At the same time, and without relinquishing an emphasis

38

Language Learning and Study Abroad

on causal explanations, the analysis suggests intriguing possibilities for links between learner identity and the notion of social distance, and for consideration of the relationship between social practices and learning opportunities. After Carroll and the Schumanns, their conceptual and methodological tools, along with several themes and tensions, were passed on to the profession. Future researchers would base their work on global constructs such as proficiency, fluency, or pragmatic competence, often operationalized as tests. Such research designs aim to define the outcomes and judge the effectiveness of learner sojourns abroad, without necessarily taking into consideration the qualities of the experience. The assumption that study abroad is analogous to an experimental treatment remains in place. Other scholars would provide intriguing descriptive accounts detailing the nature of students’ subjective perceptions. Frequently, an underlying motive of these studies is the search for forces determining language acquisition abroad, but they rarely provide evidence about the specific achievements of study participants. In short, research on study abroad has exhibited tensions between product and process foci, similar to those of SLA more generally (e.g., Ochsner 1979; Block 2003; Firth and Wagner 2007). The expansion of process-oriented and qualitative approaches may be interpreted as a logical response to questions about the results of product-oriented studies, and these will be treated in subsequent chapters. Here, however, we examine the various ways in which the language-related outcomes of study abroad have been defined and assessed in research based on global constructs of language proficiency.

Holistic constructs: proficiency, fluency, listening comprehension, reading, and writing How do we know that students who have participated in programs of study abroad develop language ability? How can we tell that they have become more proficient speakers, better able to comprehend the language they read and hear, more precise in their control of language forms? One of the most obvious ways to find answers to this question is to ask the students, their teachers, or representatives of the speech communities in question. But if we base our claims only on subjective responses, how can we know that these responses are valid and reliable, not biased by personal proclivities, the presentation of the evidence, or the way the question is posed?

Measuring Language Acquisition 39

How do we know that the criteria invoked to judge learner language use are consistent enough to permit comparison? This section examines the ways in which researchers have employed holistic definitions of language ability, including general proficiency, oral proficiency, and literacy-related skills of reading and writing, in attempting to gauge the value of study abroad for language development. Beginning with studies based on subjective ratings of improvement in language skill, the section goes on to explore various global constructs used to assess the performances of study abroad participants. A number of studies have based their findings on systematic gathering of subjective perceptions among the people involved in learning, teaching, and interacting with students abroad. In the Nuffield Inquiry (Meara 1995), for example, 586 respondents from the United Kingdom rated the extent to which their language skills (speaking, listening, reading, and writing) had improved during their year abroad. Results, based on data gathered in the mid-1980s, indicated that students believed that their social interactive skills had improved more than their reading or writing abilities. Seventy-five percent of the students rated their improvement in the spoken language at four or five (on a scale from 1 to 5), and 87 percent gave similar ratings of their listening skill. For reading, however, only 49 percent of the participants rated their ability this highly, and for writing the percentage declined to 33. Meara concluded that ‘the year abroad has its most significant effects on spoken language, with the majority of respondents clearly holding the view that their ability to speak the foreign language and their passive listening skills had improved a lot’ (Meara 1995: C-11). With findings in parallel those of similar studies (e.g., Dyson 1988; Opper, Teichler, and Carlson 1990; Davie 1996), this project helped to lay the groundwork for an emphasis on social interactive competence in study abroad research. This emphasis is reflected in a study by Yager (1998) examining oral proficiency gains in a short-term summer Spanish program in Mexico. This study assessed changes in proficiency by asking a group of nativespeaking judges of varying age, nationality, and educational background to rate speech samples for the extent to which they seemed native-like in general, and in terms of grammar and pronunciation. The results showed that using this criterion, 22 of the 30 participants showed significant gain in at least one category, and 19 improved in all areas. Although these results are encouraging, and although Yager informs the reader that the raters were instructed to privilege nativeness over grammatical correctness, since these two qualities do not always overlap, it is impossible to ascertain based on the report precisely what criteria were

40

Language Learning and Study Abroad

in fact invoked by the judges. If study abroad students tend to sound more like natives at the end of their sojourn abroad, what are the exact qualities of their language use that support this claim? Additional problems associated with subjective judgment criteria are illustrated in Lennon (1995), a study examining improvement in the language skills of four German students who spent two months in England. Subjective evaluations by experienced teachers of English as a foreign language were complemented by quantitative analyses of transcribed recordings. Once again this study showed positive changes in speaking skill, with more talk in the post-test along with greater control of modality. However, different teacher-judges favored different criteria in their ratings, leading to problems in achieving rater agreement. The more objective quantitative measures were also difficult to correlate with the findings of the subjective aspects of the study, leading Lennon to call for better variables to identify development of language skill in short-term programs. Studies based on subjective rating of student progress have asked the students themselves (e.g., Meara 1995), native speakers (e.g., Yager 1998), and experienced teachers (e.g., Lennon 1995) to judge the outcomes of study abroad in terms of language development. Results of the Nuffield Inquiry indicated that students experience greater growth in their social interactive abilities than in skills related to academic literacy, and in fact, the majority of subsequent studies have been devoted to inquiry into speaking abilities. Alongside the development of research based on subjective responses, another line of inquiry emerged in the 1980s, based on the global construct of ‘proficiency’ as defined in holistic testing instruments.

Proficiency The quest to define and document the emergence of proficiency is one of the major themes in applied linguistics research on study abroad, both in the United States and elsewhere. Study abroad research by American scholars of language education has been profoundly influenced by a particular construct of proficiency emerging from the policy-related exigencies of the late 20th century. In the years following Carroll’s (1967) report and the founding of the American Council on the Teaching of Foreign Languages (ACTFL), the procedure used by the Foreign Service Institute (FSI) for the assessment of speaking competence in US government personnel was adapted for the academic community. Impetus for this project was provided by the recommendations of President Carter’s Commission on Foreign Language and International Studies

Measuring Language Acquisition 41

(1979), with a call for the development of practical, functional foreign language skills in US schools. Since the early 1980s, the influence of the ACTFL Oral Proficiency Guidelines has been widespread, both in the practices of foreign language educators and in reflection on the nature of language competence. Proficiency is often defined in opposition to achievement based on criteria established in specific instructional contexts. A proficiency test is assumed to ‘measure an individual’s general competence in a second language, independent of any particular curriculum or course of study’ (Omaggio 1986: 9). A reliable and valid measure of general language competence, not linked to any specific learning context, would offer an immediately accessible instrument for the evaluation or comparison of learning contexts. To derive an acceptable operational definition of general language proficiency, the scale used by the FSI was adapted by ACTFL in cooperation with the Educational Testing Service. Beginning with a scale running from 0 (no functional ability) to 5 (equivalent to an educated native speaker), the levels of oral proficiency were renamed to render them less intimidating, as ‘Novice’ (0), ‘Intermediate’ (1), ‘Advanced’ (2), ‘Superior’ (3 and 4), and ‘Native’ (5). Descriptors of the levels were generated in three categories representing Function (tasks), Content (topics), and Accuracy (acceptability/grammaticality). The ACTFL Oral Proficiency Interview (OPI) was the first standardized procedure for assessing speaking ability made available to the US foreign language teaching profession. Although tests were developed for other modalities, it was the speaking test that proved most influential. The arrival of the OPI and accompanying ACTFL Oral Proficiency Guidelines (1986) launched a veritable industry in materials development, teacher education, and tester training. A phenomenon then dubbed ‘The Proficiency Movement’ was born. In fairly short order, the popularity of the Proficiency Movement began to generate dissent among SLA researchers and progressive scholars in language education. The validity of the proficiency definitions had been defended on the basis of their emergence from observation of language development: they were ‘experientially, rather than hypothetically based; that is, they describe how learners and acquirers of second languages typically function along the whole range of possible levels of competence’ (Omaggio 1986: 35). Scholars took issue with the notion that a valid scale, particularly in the area of Accuracy, could safely ignore ongoing empirical work on the routes to acquisition of grammatical competence (VanPatten 1985a; Kramsch 1987). They also questioned the validity of a scale implying the existence of one idealized, educated, and presumably monolingual Native Speaker model

42

Language Learning and Study Abroad

(Lantolf and Frawley 1988), and that such a model is the proper standard against which to measure the competence of multilinguals or expert speakers of non-standard varieties (Savignon 1985; Barnwell 1989). They critiqued the interactive qualities of the test itself, proposing that interviewer bias could directly influence the performance of test takers and that, far from eliciting samples of authentically conversational language use, the interview was a particularly egregious manipulation of interactional authority on the part of the tester (van Lier 1989). Eventually, however, and despite this controversy, the ACTFL Oral Proficiency Guidelines exerted significant influence on language education in the United States, forming the default understanding of advanced language ability (Leaver and Shekhtman 2002). Researchers were quick to recognize the potential of the OPI as a universally recognized, presumably context-neutral assessment instrument.

Table 2.1 Assessment of proficiency development Study

Origin Destination

Sojourn length

Participants Instrument

Carroll (1967)

US

Varied

Varied

2784

Magnan (1986)

US

Varied

40

Freed (1990) Milleret (1991) Brecht, Davidson, and Ginsburg (1995) Coleman (1996)

US US US

Francophone countries France Brazil Russia

UK

Varied

Allen and Herron (2003)

US

France

Iwasaki (2007) Rees and Klapper (2007) Tschirner (2007)

US UK

US

MLA Proficiency Test OPI

6 weeks 1 summer 1 semester

40 11 658

Varied (normally I year) 6 weeks

18,925

Japan Germany

1 year 6 months to 1 year

5 57

Picture description and OPI role-play OPI C-Test

Germany

4 weeks

15

OPI

25

OPI OPI OPI

C-Test

Measuring Language Acquisition 43

A number of studies reported a potential relationship between time spent abroad and increases in OPI scores (see Table 2.1 and the mixedmethod studies in Chapter 4). In 1986, Magnan published the results of a study aiming to document current proficiency levels among undergraduate students of French at the University of Wisconsin, as a preliminary step toward the establishment of standards for the university’s language curriculum. Forty students were interviewed, including ten each from four groups: first-year, second-year, third-year French minors, and fourth-year French majors. Within these groups, the students who had spent substantial time in a Francophone country were among those with relatively high-proficiency scores, however the author was unable to document all ‘important intervening variables’ (Magnan 1986: 435). For example, at the second-year level of instruction most of the students earned scores in the Intermediate range, but one student who claimed to have spent 1–6 months in a Francophone environment displayed Advanced capabilities even at this early curricular stage. At the level of French majors (fourth year), all but three of the students scored at the Advanced level, and all had spent time abroad, but there is no apparent relationship between the amount of time spent abroad and the proficiency scores: Time in Francophone Country Less than 1 month (N = 2) 1–6 months (N = 3) 6–12 months (N = 4) 18 months (9 years prior, N = 1)

Proficiency Rating Intermediate High (2) Advanced (3) Intermediate Mid, Intermediate High, Advanced, Advanced Plus Advanced (Magnan 1986: 435)

Thus, in addition to pointing out the relevance of individual differences, this study suggested that the relationship between study abroad and proficiency is far from straightforward. Further studies went on to examine the value of this assessment instrument in specific short-term summer programs of study abroad in France (Freed 1990) and Brazil (Milleret 1991). The aim of Freed’s study was to assess development of achievement and proficiency in relation to the type and quality of out-of-class contact. The study enrolled 40 undergraduate participants with varying preparation in language courses who studied for six weeks in Tours, France. Measures of aptitude and motivation were taken along with assessment of out-ofclass contact using the Language Contact Profile, where contact was

44

Language Learning and Study Abroad

defined as interactive (speaking with people) and non-interactive (reading books, watching television). Pre- and post-measures of performance included a discrete point grammar test (the College Entrance Examination Board Language Achievement Test, or CEEB) and the ACTFL OPI. The more advanced students tended to seek out more contact of both kinds (interactive and non-interactive), but there was no relationship between motivation or aptitude and pursuit of language learning in out-of-class contexts. In the case of achievement, interactive contact appeared to assist the lower level students only; for the advanced students, more non-interactive contact led to higher scores on the CEEB whereas interactive contact lowered their scores. These results are interpreted by Freed in relation to specific needs of beginners, who must master the ‘language of daily activities’ (1990: 473), versus advanced students who are working on their second-language literacy. In attempting to correlate out-of-class contact with growth in proficiency, the study encountered a major difficulty: perhaps due to the short duration of the sojourn in France, very little change was documented (only four of the six beginners scored differently in the post-test). Freed therefore suggested that the OPI, with its global, holistic scoring, lacks nuance in detecting short-term development in language ability: The OPI which utilizes one global holistic score for various aspects of language use is not sufficiently refined to capture growth in oral skills, particularly in a six-week period. Except for students at the very beginning level, there was little variation in OPI scores. We therefore found it difficult with this type of analysis to demonstrate any effect of out-of-class contact. In order to demonstrate change, future studies will have to utilize more finely-tuned analyses; those which will reveal, with specificity, development in students’ lexical breadth, syntactic complexity, stylistic sensitivity, sociolinguistic and pragmatic competence, and cohesion and coherence in language use. (Freed 1990: 475) Milleret (1991) developed the Portuguese Speaking Test as a pre-recorded interview format based on the ACTFL OPI. When used to assess gain in speaking skill among 11 participants in a short-term program in Brazil, this test measured significant gains in the cases of lower-scoring students, whose ratings rose by one level on average. However, students who began the study at the Intermediate level did not show measurable gains. Milleret interpreted these results by noting, with Freed, that the OPI is not sensitive to short-term changes, particularly at the upper

Measuring Language Acquisition 45

levels. In fact, the model of proficiency popularized in related publications (e.g., Omaggio 1986) represents this construct as an inverted cone, with the lower levels at the tip, and the volume of required knowledge expanding with each advancement. The higher the level, in other words, the more the student is expected to be able to do. The insensitivity of the scale left Milleret unable to claim that advanced level learners, those who are reputed to profit most from study abroad, had developed their competence in a study abroad program. This was true despite informal observation among students and instructors indicating that advanced students made more progress than beginners. Despite calls for more refined definitions of language competence, and despite the problems encountered in attempts to document proficiency development in study abroad via the ACTFL OPI, use of this instrument, or of variants or concepts based on it, has gone uninterrupted since its introduction. In particular, the development of speaking proficiency is a key aspect of a multi-year and multi-method study examining the development of Russian language ability: Language Learning in Study Abroad: The Case of Russian. Organized under the auspices of the American Council on the Teaching of Russian (ACTR) and the National Foreign Language Center (NFLC), the study focused on predictors of gain in language ability during four-month ACTR programs. The study was originally reported when data for 658 participants had been collected between 1984 and 1990 (Brecht, Davidson, and Ginsburg 1995). In addition to the Oral Proficiency Interview, assessment instruments included Listening and Reading examinations prepared by the Educational Testing Service, qualifying examinations in grammar and reading developed by the ACTR, and the Short Form of the Modern Language Aptitude Test. Demographic and program data along with detailed information about the language-learning histories of the participants were also collected. Findings related to oral proficiency showed that the main predictors of gain scores were experience in learning other languages, strong command of grammar and reading skills, and gender, a highly significant predictive factor, with men more likely than women to reach the ‘2’ (Advanced) level. Of these findings, the first two, according to the authors, speak to widespread views about the value of grammar instruction and the importance of metalinguistic awareness that had hitherto been held only intuitively. The results related to gender were far more controversial. Brecht, Davidson, and Ginsburg (1995) entertained three possibilities for interpretation of their findings related to gender. Beginning with the assumption that there is no real difference between men and women in

46

Language Learning and Study Abroad

language-learning ability, they treated and rejected the hypothesis that there was a gender bias inherent in the testing instruments. They suggested but did not explore the possibility that the sample was skewed, with self-selection of the minority male population representing high motivation or talent for language learning. Finally, they proposed that gender differences may emerge from differential treatment and qualities of interaction while abroad: . . . one might hypothesize that men and women spend their time differently in-country, and that, therefore, the difference in acquisition might be due to a difference in time-on-task. On the other hand, one might imagine a difference between the interactions of American men and women with Russian men and women as a reflection of the two cultural differences involved: American and Russian, as well as male and female. (Brecht, Davidson, and Ginsburg 1995: 57) The authors further suggested that Russia is not the only society where American students study abroad and where women receive differential treatment that may affect language acquisition. In the case of this large-scale study documenting language learning in sojourns of at least one semester, Brecht, Davidson, and Ginsburg (1995) were able to establish a correlation between pre-program predictive variables and proficiency gain scores. In the case of the other studies reviewed here, involving shorter sojourns or smaller cohorts, the levels described in the ACTFL Guidelines proved to be too blunt an assessment instrument, leading to frustration on the part of researchers who observed changes in speaking ability but were unable to document them officially. Beyond noting problems with lack of refinement in the scale, few study abroad researchers seem to question the model itself or otherwise take early criticism of the OPI into account when interpreting their findings. There is but one sour note among all the implicit praise of the OPI procedure: after observing that certain speech acts in Russian, such as proposing a toast at table, tended to be reserved for males, Polanyi (1995) suggested that the speaking tasks imposed upon interviewees in the OPI are likely to present a gendered dimension potentially harmful to the ecological validity of the procedure, that is, the extent to which the observed phenomena are relevant to participants and are representative of their daily lives (Cicourel 2007). Polanyi implied that the test itself placed female students at a disadvantage, asking them

Measuring Language Acquisition 47

to display pragmatic functions that had been off-limits to them during their stay in Russia, in essence, limiting their chances at success because their interactive experiences did not match the demands of the interview. Despite claims to the effect that the OPI measures contextindependent language ability, in other words, each individual test is an assemblage of particular topical or functional tasks that may or may not correspond to students’ history of language use and may or may not take issues of gender and power into consideration. This criticism notwithstanding, the ACTFL Guidelines continue to inform the design of study abroad research, particularly a number of studies reviewed in Chapter 4 attempting to correlate the activity of students abroad with test outcomes. While American researchers tended to favor the ACTFL Guidelines and an emphasis on speaking ability in assessing general language proficiency, the European Language Proficiency Survey (Coleman 1996) and a subsequent study by Rees and Klapper (2007) relied on a holistic instrument known as the C-Test. The C-Test is assumed to measure overall language ability by imitating the conditions under which expert language users reconstitute meaning based on incomplete texts (such as would occur, for example, in the case of illegible handwriting). As a test of reduced redundancy, the C-test is created by deleting parts of words (the second half of every second word) from a series of authentic texts and asking test takers to reconstitute the original message. According to Coleman (1996) students who are the most proficient do so most successfully, ‘by calling on the whole of their foreign language competence. The clues which are unconsciously analysed may be semantic, syntactic, morphological, collocational, etc., and solutions may require inferencing from other parts of the text’ (bold in the original) (1996: 137). Although it has no oral or aural component, the C-Test provided a reliable and efficient snapshot of proficiency for a cross-sectional study involving over 25,000 students of French, Spanish, German, English, and Russian in the United Kingdom and in seven other European counties between 1993 and 1995. Like Carroll’s earlier project, the aim of the European Language Proficiency Survey was not merely to investigate the effects of residence abroad, but to characterize the language abilities of students at all levels along with the nature of their courses and their attitudes and motives for learning. Coleman’s findings for students from the United Kingdom (N = 18,825), in parallel with Carroll’s, suggest an important role for residence abroad, obligatory for the majority of subjects. Across the four languages under study, mean scores for students in the final year of

48

Language Learning and Study Abroad

study, after an in-country sojourn, were substantially higher than those of students who had yet to study abroad. Coleman also reports a significant correlation between time spent abroad (less than a term, a term, a semester, or a year) and scores on the C-Test. In a study involving 57 students of German, Rees and Klapper (2007) sought to replicate Coleman’s findings using a longitudinal rather than a cross-sectional design. This project also investigated the predictive power of several factors, including gender, general educational history, previous language-learning experience, length of stay, and academic ability as measured by IQ scores. Results from two pre- and post-residence abroad proficiency measures, the German version of the C-Test and a test of grammatical knowledge, provided evidence for accelerated proficiency gain during residence abroad, although these gains were less dramatic than those reported in the Coleman (1996) study. In contrast to Brecht, Davidson, and Ginsburg’s (1995) findings, this study indicated no predictive effect for gender. Neither IQ, nor length of residence abroad (six months versus one year), nor previous academic achievement (i.e., grades) predicted gain. However, the cohort included two groups: specialists who had received considerable instruction in discrete-point grammatical features of German, and non-specialists whose instruction had favored naturalistic exposure to grammar. A comparison of these two groups indicates that the latter approach conferred an advantage in residence abroad. Finally, like many other studies based on holistic constructs, this project documented considerable individual differences in rates of progress toward language proficiency. Effort to assess general proficiency in European projects, as illustrated by these two studies, bears some resemblance to its American parallel. In large-scale projects, scores on holistic testing instruments like the C-Test, the ACTFL OPI, or the Modern Language Association Foreign Language Proficiency Test indicate that, in general, study abroad has an important role to play in developing language ability. In studies where measures are multipled and refined, the findings generate numerous questions. What conditions might account for the fact that overall proficiency growth in the more recent study by Rees and Klapper (2007) fails to reach the level previously documented by Coleman (1996)? Why is female gender a predictive factor for American learners of Russian, but not for British learners of German? Why do some students, but not others, seem to reach a plateau in their proficiency development after six months abroad? If there really is no relationship between general academic prowess and language learning abroad, what capabilities are

Measuring Language Acquisition 49

needed to further students’ success? In the meantime, the discovery that tests of general proficiency did not in fact document many of the subtle changes taking place in study abroad participants’ speaking ability led to exploration of another holistic construct, fluency.

Fluency Fluency is defined in the American Heritage Dictionary as ‘effortless,’ ‘flowing,’ or ‘polished’ language use, and is often cited as a major goal and expected outcome of language learning abroad. According to Freed (1995b: 123), in fact, fluency is ‘the term most frequently evoked in discussions of the linguistic benefits of study abroad.’ Although the term is ‘deeply embedded in lay linguistic perceptions’ (McCarthy 2006: 2), in language-learning research the concept has been elusive. Upon launching a multi-year effort to investigate the development of fluency in study abroad, Freed (1995b) noted the absence of a widely shared technical definition suitable to inspire research. For Freed, fluency seemed to consist of an ‘impression’ (1995b: 123) arising from the interaction of hearer perceptions and some combination of speaker attributes, such as speed of delivery, intonation, or accuracy. The study abroad literature includes eight studies focused in various ways upon fluency (see Table 2.2). In the broader applied linguistics literature, definitions of fluency typically focus on the characteristics of individual performance, including rate of speech, pause length, or the naturalness of rhythm and intonation (McCarthy 2006). Accordingly, in the study abroad literature the assessment of fluency is based upon monologic speech samples collected before and after the sojourn abroad via experimental tasks or oral examinations such as the OPI. These samples may then be evaluated by native speakers, analyzed in terms of temporal characteristics of speech delivery, or examined for the presence of routinized formulaic sequences believed to bolster listeners’ impression of fluency. Overall, these studies not only provide evidence in support of claims that study abroad enhances fluency, but also raise questions about the qualities of study abroad experiences and the nature of fluency itself. In the American applied linguistics literature, a series of studies by Freed and colleagues constitute the most prominent investigations of fluency development in study abroad. Freed (1995b) reports on a project involving native-speaker judgments of fluency based on extracts of OPIs administered to 30 students, of whom 15 had studied abroad in France and 15 had remained on campus for one semester. An improbable early

50

Table 2.2 Assessment of fluency development Study

Origin

Destination

Sojourn length

Participants

Data/Analysis

Freed (1995b)

US

France

1 semester

OPI/7 factors of fluency

Towell, Hawkins, and Bazergui (1996) Freed, So, and Lazar (2003)

UK

France

6 months

15 abroad, 15 at home 12

US

France

1 semester

Freed, Segalowitz, and Dewey (2004)

US

France

1 semester

Segalowitz and Freed (2004) O’Brien, Segalowitz, Freed, and Collentine (2007) Juan-Garau and Pérez-Vidal (2007) Wood (2007)

US

Spain

1 semester

US

Spain

1 semester

Spain

Anglophone environment Canada

3 months

15 abroad, 15 at home 8 abroad, 8 at home, 8 in domestic immersion 22 abroad, 18 at home 25 abroad, 18 at home 12

6 months

4

Japan

Film retelling/Quantitative and qualitative Essays/Native speaker judgment OPI, LCP/Overall performance and hesitation

OPI, LCP, Tests of lexical access and attention control OPI, Test of phonological memory Role-play/Fluency, accuracy, formula use Narrative retelling/Temporal variables, Formula-Run ratio

Measuring Language Acquisition 51

finding of the study was that comparison of the two total populations yielded no statistical difference in the native-speaker assessments of gain in fluency. When the research team included only those students whose pre-departure fluency received a low rating, they were able to show a difference in gain scores between the ‘at home’ (AH) and ‘study abroad’ (SA) groups. A subsequent analysis closely scrutinized features of the same utterances heard by the native-speaker judges, for a selected subset of each group. Of the seven ‘factors of fluency’ (1995b: 136) chosen for analysis, only rate of speech proved significant, suggesting that students who had spent a semester abroad ‘spoke both more, and at a significantly faster rate than did those whose learning had been restricted to the language learning classroom at home’ (1995b: 137). Other features examined included the nature of pauses, the length of fluent speech runs, repairs, and ‘clusters of dysfluencies’ defined as the presence of two or more interruptions to the flow of speech. Although between-group comparison of these features per se did not suggest a significant advantage for the study abroad participants, study abroad participants tended to display fewer dysfluent features. That is, their speech runs were longer and characterized by fewer unfilled or unduly long pauses. Moreover, these students demonstrated a relatively high level of linguistic sophistication in their attempts to communicate complex ideas. Having presumably engaged in extensive conversational interaction, these students wanted to do more with their foreign language, and this desire led them into situations where they ‘stumble linguistically, monitor their speech and self-correct along the way’ (1995b: 141). In a companion study, Freed, So, and Lazar (2003) employed a similar methodology in eliciting native-speaker judgment of written fluency based on essays about a memorable trip written at the beginning and end of the semester by the same students whose oral fluency was evaluated in Freed (1995b). Native-speaker judges were again asked to rate the fluency displayed in the essays based on subjective criteria, and the researchers sought correlates of these judgments in the actual texts. Based on ‘grammar,’ ‘vocabulary,’ ‘expression of thought,’ and ‘organization,’ the native speaker judges perceived the written fluency of the AH group to be higher than that of the SA group. In fact, nothing in the judges’ perception ‘seemed to suggest that the SA students’ writing was more fluent as a result of their having spent a semester abroad’ (2003: 37). An independent analysis of textual features showed that at the end of the semester, the SA group essays grew in length and lexical density, whereas those of the AH group did not. (One explanation proposed for

52

Language Learning and Study Abroad

this difference is that the SA participants had more interesting stories to tell.) When considering change from the pre-test to the post-test, there was, however, no difference in the two groups’ performance on measures of grammatical correctness. The authors concluded that while the construct of fluency is supported by identifiable linguistic features, these features do not always contribute in predictable ways to the assignment of fluency ratings. Both of these publications suggest that one valuable outcome of study abroad is growth in students’ communicative ambition, that is, study abroad participants enlarge their repertoire without necessarily refining it in the areas of grammatical or lexical accuracy valued by native judges. This result is matched by the findings of Allen and Herron (2003), for example, whose study of American learners in France also shows significant growth in fluency but only modest improvements in grammatical correctness. In both, however, the authors also point out the striking nature of differences in individual performance that tended to compound the problems associated with achieving statistical significance. Freed (1995b) noted that among the subjects whose data were selected for close analysis, at least one of the AH study participants performed at least as well if not better than the SA subjects, and that two of the SA participants had scores on fluency measures similar to those of AH students. In the study of growth in written fluency (Freed, So, and Lazar 2003), however, these same two students displayed ‘vast jumps in lexical density’ (2003: 38). In these studies, individual differences are absorbed in the effort to document group differences, yet their presence must be noted. Subsequently, Freed and her colleagues have continued to pursue the elusive construct of fluency through further comparative studies and definitions of increasing precision. Freed, Segalowitz, and Dewey (2004) studied the acquisition of oral fluency by students in three different contexts: AH, SA, and enrollment in domestic immersion programs (IM) integrating formal classroom and out-of-class sheltered learning opportunities. The researchers refined the notion of context by focusing on time-on-task factors, including instructional time and time spent speaking with natives and in literacy-related activities. The study involved 28 native speakers of English learning French in one of the three program types for the period of one semester, in the case of the AH and SA programs, and of seven weeks in the summer IM program. The IM program was characterized by abundant French-mediated extracurricular activities, and a pledge to use only French in all contexts, signed by all participants. SA students lived in Paris taking courses either with

Measuring Language Acquisition 53

their peers or at a local institution of higher education. Data collection included pre- and post-test interviews modeled after the OPI administered at the beginning and the end of the session, and a Language Contact Profile in which students reported on the amount and type of their activity in English and French. Assessment of oral fluency was based on data extracted in one-minute segments from the interviews, with two such segments taken from each testing session, yielding four minutes of speech for each participant. Fluency was operationalized: (1) as general measures of oral performance (total words, duration of speaking time, and length of the longest turn); and (2) as hesitation and temporal phenomena (speech rate, hesitation, pauses, mean length of speech run without dysfluencies, repetitions, or repairs). Results showed that the AH group’s performance did not change from pre- to post-test except in the case of three variables where six of the eight students performed worse. By contrast, the IM group registered significant gains for total words, turn length, and rate of speech as well as in the areas of repair and mean length of fluent run of speech. An independent measure of speech fluidity, grouping together the measures of hesitation and absence of interruption, showed the IM group making the greatest gain, the SA group making some gain, and the AH group making no gain at all. Reading these results in light of data gathered in the Language Contact Profile, the authors found that the IM group reported significantly more contact hours in French than the other two groups, and spent more than four times the amount of time in French language–writing activities than did the others. The SA group, by contrast, reported spending more time in English than in French outside of class in all activities except listening. Time-on task, as reported in the Language Contact Profile, therefore appeared to be a potent variable predicting the development of fluency, with more time-on-task involved in domestic immersion than in study abroad. The authors insisted that they did not wish to argue for the inherent superiority of IM programs. They proposed that it is not the context per se that predicts learning, but that it is ‘the nature of the interactions, the quality of the experiences, and the efforts made to use the L2 that render one context superior to another with respect to language gain’ (Freed, Segalowitz, and Dewey 2004: 298). Additional studies by Freed and her colleagues have pursued the cognitive dimension of individual differences in the fluency-related outcomes of study abroad. Segalowitz and Freed (2004), for example, examined the role of cognitive processing abilities (lexical access and

54

Language Learning and Study Abroad

attention control) in fluency development. Students of Spanish enrolled in an SA program in Spain were shown to outperform their AH counterparts on the OPI and in related measures of fluency. Gains on the OPI were significantly related to initial levels of lexical access processing and speed, suggesting that ‘fast, efficient abilities to connect words and meanings facilitated learning to speak more proficiently’ (2004: 194). O’Brien, Segalowitz, Freed, and Collentine (2007) demonstrated a significant relationship between phonological memory (the ability to retain phonological elements and their serial order in the short term) and oral fluency development. These studies suggest that achievement of fluency in study abroad contexts may be partially attributable to the individual learner’s general or language-specific cognitive skill and readiness for learning. While these American researchers interrogated the process of fluency development through increasing focus on individual cognitive differences in students of modest initial proficiency, the UK-based scholars Towell, Hawkins, and Bazergui (1996), working with advanced learners of French, examined both temporal variables and qualitative differences in more- versus less-fluent second-language use. These researchers borrowed Levelt’s (1989) model of language production, positing three sub-processes in language production (the conceptualizer, the formulator, and the articulator) and Anderson’s (1983) view of development as the proceduralization of declarative knowledge. They suggested that close examination of temporal indicators not only measures global fluency but also illustrates the contribution of sub-processes in Levelt’s model, particularly the formulator where procedural knowledge is stored. Twelve learners of French provided speech samples based on a film-retelling task before and after a six-month sojourn in France. The samples were subjected to a quantitative analysis focusing on speaking rate, phonation/time ratio, articulation rate, and mean length of run, defined as the average number of syllables produced between pauses of .28 seconds and above. The evidence indicated that the students became more fluent, as measured by speaking rate, and that increases in the mean length of run accounted for most of this development. Having established the significance of mean length of run, the researchers then performed a qualitative analysis in search of the kinds of linguistic knowledge proceduralized for the development of fluency. In order to produce longer utterances between pauses, the participants were shown to recruit knowledge of syntax, and of fillers, organizers,

Measuring Language Acquisition 55

and formulaic chunks such as lexical phrases and collocations. This knowledge allowed the learners not only to extend the length of their utterances but also to sound ‘more “French” ’ (Towell, Hawkins, and Bazergui 1996: 112). For example, one of the most fluent participants in the study relied heavily on basic Subject-Verb-Object sentence structure prior to study abroad (e.g., Ils ont rencontré, Il était différent, Ils ont commencé) but in the course of her sojourn abroad had learned to employ certain structural variants characteristic of spoken French, such as double marking of topics (l’histoire, ça commence), and presentatives (il arrive un jour qu’il y a un petit rond qui arrive). Hesitation in the pretest was related to issues like uncertainty about the gender of nouns, but in the post-test appeared to signal efforts at greater precision in rendering concepts. The authors concluded that growth in fluency is not due to decreases in pausing or increases in the rapidity of articulation, but is attributable primarily to the length and complexity of utterances. Another approach to the assessment of developing fluency is illustrated in the work of Juan-Garau and Pérez-Vidal (2007), affiliated with the Stay Abroad and Language Acquisition (SALA) Barcelona project analyzing the impact of study abroad on the English-language competence of Spanish-Catalan bilingual students preparing diplomas in Translating and Interpreting. A key innovation of this study is its longitudinal design, involving testing of the same participants at four moments: the beginning of their degree program, the end of a six-month period of formal instruction, their return to the university following a three-month stay in an Anglophone environment, and 15 months later. Juan-Garau and Pérez-Vidal (2007) focus on the oral performance of these participants in role-play and narrative tasks, measuring a variety of factors including fluency (in words per clause or sentence), accuracy (grammatical and lexical errors), complexity (in clauses per sentence and dependent clauses per clause) and rate of formula use. Their findings generally show that following a period abroad students increased their lexical and formulaic repertoires, made fewer errors, and used more and longer clauses. More importantly, with the exception of formulaic language and subordinate clause use, many of these gains were retained after a period without instruction, suggesting that study abroad can have a durable impact on students’ fluency. The specific role of formulaic sequences in fluency development was explored in a study by Wood (2007) in which four Japanese learners of English in a program of study abroad each provided six consecutive samples of speech in narrative retelling tasks. Wood suggested that the

56

Language Learning and Study Abroad

key to fluency is ‘the use of formulaic sequences, strings, and frames of words with specialized functions, mentally stored and retrieved as single words’ (2007: 209). Like other fluency researchers, Wood examined an array of basic temporal variables; however, among these variables he included a Formula/Run calculation indicating the ratio of length of runs and the number of formulaic sequences in the sample. Wood’s results showed strong development on this indicator: the participants did indeed increase their use of formulaic sequences in producing longer runs of speech between pauses. Wood concluded that broad experience with spoken language is an important element in the development of fluency, that study abroad can aid learners in reaching fluency goals, and that further research is needed involving a wider range of discourse genres, including dialogic conversation. Wood’s suggestion is reasonable, given that most efforts to illuminate the development of fluency in study abroad to date involve ‘judging a speaker on monologic performance, or based on an oral examination where assessors hold back from interacting like normal human beings’ (McCarthy 2006: 5). In the case of the Juan-Garau and Pérez-Vidal paper (2007), student utterances are evaluated as sentences, that is, as if they correspond in each case to the norms of writing. Relying on evidence from native-speaker corpora of natural language use, McCarthy takes issue with the notion that fluency is best understood as the display of individual skill judged in terms of speech rate, pause phenomena, or mean length of run. Native speakers providing data for corpora are normally assumed to be fluent, yet their performance often exhibits features such as intersentential pausing, redundancies, or slow pace, that would be considered dysfluent in the models reviewed thus far. McCarthy provides the following example from the Cambridge International Corpus, in which the participants are discussing plans for a trip to Italy: A:

Where would you tell me to go? And then to a two week trip. Where would you tell me to go? B: Okay. Um well let’s see. You’re gonna want to . . . You’re gonna want to see I mean since you are there for two weeks you’re you’re probably gonna you know you’re just gonna have to see the . . . You’re not gonna have time to really wander around and so you’re gonna want to go where the churches are and+ A: Uh huh. B: +the museums are+ A: Uh huh.

Measuring Language Acquisition 57

B: +so I would say go to Rome and go to Florence [. . .] B: ‘Oh I don’t know what this is all about but let’s [laughing] just+ A: Let’s check it out. B: +park and go see’. (McCarthy 2006: 3) In addition to fulfilling the temporal criteria established in the literature on fluency, these speakers use formulaic chunks of high frequency in the corpus (You’re gonna, I would say, I mean). They do not necessarily speak in sentences. Moreover, the conversation itself is fluent, in that the speakers take it to be a joint responsibility and support one another through back channeling (uh huh) and completing each other’s turns. What fluent speakers do in dialogic interaction may therefore be more like establishment of confluence than display of individual skill. To date the literature on development of fluency in study abroad rarely takes the features of spoken language or conversational interaction into account. Instead, researchers have collected and analyzed samples of speaking performance showing that study abroad generally promotes the ability to speak quickly, without hesitation, and using more complex structures (Juan-Garau and Pérez-Vidal 2007). Research involving American students of modest initial proficiency suggests that study abroad enhances communicative ambition, leading them to attempt a greater variety of language functions in ways that do not necessarily please native-speaking judges (Freed 1995b; Freed, So, and Lazar 2003). This research also suggests that time-on-task for fluency development is more typical of programs engineered for language immersion than of study abroad (Freed, Segalowitz, and Dewey 2004), and that individual variation in outcomes may be ascribed in part to cognitive factors (Segalowitz and Freed 2004; O’Brien, Segalowitz, Freed, and Collentine 2007). The projects of Towell, Hawkins, and Bazergui (1996) and of Wood (2007) demonstrates the relevance of formulaic sequences, suggesting that students need exposure to specific patterns and features of spoken language in order to develop fluency. Given that domestic immersion programs are sheltered learning environments, one may wonder if in fact the gains registered for temporal variables in Freed, Segalowitz, and Dewey (2004) would correspond to greater approximation of norms for speaking as documented by Towell, Hawkins, and Bazergui (1996). That is, can domestic immersion programs offer students engagement with a range of authentic varieties of speech, including those practices that may be common but

58

Language Learning and Study Abroad

are not officially recognized in pedagogical materials, or does study abroad offer unique affordances in learning to sound, for example, ‘more “French” ’? Another mystery emerging from these findings concerns the question of time-on-task in domestic immersion versus study abroad programs. Why would study abroad participants report spending less time involved in foreign language–mediated activities, and exhibit less fluency development, than domestic immersion students? Professional folklore surrounding study abroad would equate an in-country sojourn with effective language immersion, but the findings of Freed, Segalowitz, and Dewey (2004) indicate that this view may be somewhat naïve. This section has examined studies organized around the construct of fluency, but fluency is of course much more than speech rate or manipulation of formulaic sequences. Questions about how fluent learners become during study abroad sojourns touch on a wide range of issues, including the extent to which they develop lexical, grammatical, and phonological competence, approximate the manipulation of sociolinguistic variants in the manner of expert speakers, and develop interactive competence permitting them to construct fluent conversations in the manner depicted by McCarthy (2006). Moreover, fluency is matter of impression management in interaction, and is therefore related to the kinds of communication strategies that learners adopt for navigating conversations. Before turning to these issues, however, to complete the review of studies guided by holistic constructs, it is necessary to consider how other global definitions of language proficiency have been investigated in study abroad contexts.

Listening comprehension Study abroad participants usually report more significant development of abilities related to social interaction than in reading and writing (e.g., Meara 1995). In the research on language-related outcomes, there is a corresponding imbalance weighted in favor of investigating speaking ability. Few studies focus explicitly on listening comprehension, and fewer still on the outcomes of study abroad that are related to literacy. The investigation of listening comprehension is limited to five recent studies (see Table 2.3) most of which gathered data via standardized tests. Everyday listening comprehension is likely to present considerable challenges to the typical American classroom language learner abroad. Kaplan (1986), for example, presented the results of a survey asking

Measuring Language Acquisition 59 Table 2.3 Assessment of listening comprehension Study

Origin Destination

Sojourn length

Participants Instrument

Huebner (1995)

US

Japan

9 weeks

12 abroad, 12 in domestic immersion

Allen and Herron (2003)

US

France

6 weeks

25

Tanaka and Ellis (2003)

Japan

US

15 weeks

166

Cubillos, Chieffo, and Fan (2007) Kinginger (2008)

US

Spain and Costa Rica

5 weeks

48 abroad, 92 at home

US

France

1 semester

23

Educational Testing Service Japanese Proficiency Test French Listening Proficiency Test Test of English as a Foreign Language Spanish Advanced Placement Test Test de Français International

students in France to rate the frequency of certain French-mediated activities, such as talk at meals or service encounters, and also to judge these activities in terms of their difficulty and the satisfaction associated with them. One of Kaplan’s claims was that there is a serious mismatch between the functions stressed in language pedagogy and the realities of study abroad. In particular, the comprehension skills traditionally neglected in the classroom were found to be of ‘far greater importance than students might have expected’ (1986: 297). Kaplan’s participants emphasized the value of opportunities to hear French in a variety of settings, in the media, in the classroom, at mealtime and elsewhere, and in fact claimed that comprehension, in both reading and listening, was the domain in which they made the most progress. Most of the studies attempting to document improvement in listening comprehension ability have in fact succeeded. Huebner (1995) administered the Educational Testing Service’s (ETS) Japanese Proficiency Test to a group of beginning language learner in Japan for a nine-week summer immersion and in a domestic immersion program of parallel length

60

Language Learning and Study Abroad

and pedagogical content. On the listening section of the test, the study abroad group performed slightly better than the immersion students, although the sample size was small and these results were not statistically significant. Allen and Herron (2003) designed their own multiple choice listening comprehension test based on video clips from a French television show, and administered this test in a pre-post design to a group of American students in France for six weeks. In this case, results of the post-test were significantly higher than the pre-test scores. Similar results emerged from my own project (Kinginger 2008), involving a group of American students in France for one academic term. Before and after their sojourn, these students took the Test de Français International, a standardized examination administered by the Educational Testing Service and consisting of a reading and a listening section. Posttest scores were significantly higher on both sections, with a more robust finding for the listening section. In a project involving Japanese learners studying in the United States, Tanaka and Ellis (2003) reported minor gains on the Listening sub-section of the TOEFL. Cubillos, Chieffo, and Fan (2007) compared performance on the listening portion of the College Board Spanish Advanced Placement Test by groups of American learners of Spanish on short-term (five-week) programs in Spain and Costa Rica, and by students enrolled in a course of the same length in the United States. No significant differences between the groups were observed. However, in analyzing participants’ self-assessment of comprehension strategies, the authors claimed that the students abroad approached their task in ways different from those of their peers at home. ‘Somewhat surprising’ (2007: 173) is the observation that from the beginning students choosing to pursue foreign language competence abroad employed sophisticated, social, and topdown strategies associated with successful learning. That is, they would attempt to grasp the overall meaning of the texts they heard, or would appeal for assistance from others when required. Their counterparts at home tended to rely instead on bottom-up and occasionally frustrating strategies such as attempting to identify grammatical structures and the meanings of individual words. Cubillos, Chieffo, and Fan therefore cast their lackluster findings in a positive light by claiming that study abroad promotes or reinforces strategies believed to foster learning at the levels beyond the most basic. As in the case of speaking proficiency testing, listening comprehension research presents issues of ecological validity. In every case, the construct of listening comprehension is operationalized via multiplechoice tests, and in all but one of these tests are standard examinations

Measuring Language Acquisition 61

designed to assess the varieties of language and content domains valued in academic or professional settings. That is, students are tested on their ability to carry out academic tasks using standard language normally characterized by formal register. Reading these studies on the backdrop of Kaplan’s observations (1986) about the significance of everyday and informal language in study abroad, it is legitimate to question whether or not these tests truly represent the language that students have learned. Findings of modest or no gains may be attributable in part to a mismatch between the qualities of experience and the subsequent test, both for individual students and for the groups under investigation. Despite this rather serious limitation, most of the studies do provide results confirming students’ own perception that study abroad is a productive context for developing listening comprehension ability. The exception is the case presented by Cubillos, Chieffo, and Fan (2007), where the period of study abroad was limited to five weeks. As the authors argue, it is worthwhile to interrogate the outcomes of such programs, since they represent a growing proportion of American participation in study abroad. However, since the study offered no information about the qualities of students’ engagement in Spanishmediated activities and little insight on their identities or histories, it is difficult to ascertain the extent to which study abroad really functioned as a language-learning environment in this case.

Reading and writing Competence in reading and writing, as previously noted, is remarkably under-represented in the applied linguistics literature related to study abroad. Most contemporary scholars concur that learning to read and write abroad not only tests students’ linguistic and pragmatic skills (e.g., at word decoding or identifying textual genre) but also confronts them with the need to negotiate new academic cultures and to adapt to new forms of literacy (Mauranen 1994; Kline 1998). Academic literacy abroad poses ‘challenging cross-linguistic and cross-cultural questions’ (Taillefer 2005: 521), since development in this domain is as much a question of socialization as it is of language acquisition (Kern 2003). Foreign language literacy socialization is a compelling topic, to which we shall return in Chapter 4. Here we examine the rare studies explicitly focused on the reading and writing skills emerging from student sojourns abroad. These studies include seven projects reporting results on reading ability and two studies examining writing ability (see Table 2.4).

62

Table 2.4 Assessment of reading and writing Study

Origin

Destination

Sojourn length

Participants

Instrument

Brecht, Davidson, and Ginsburg (1995)

US

Russia

1 semester

658

Huebner (1995)

US

Japan

9 weeks

12 abroad, 12 in domestic immersion

Lapkin, Hart, and Swain (1995) Hayden (1998)

Quebec

3 months

89

China

1 semester

21

Dewey (2004)

Anglophone Canada Europe and North America US

Japan

15 abroad, 15 in domestic immersion

Sasaki (2004)

Japan

11 weeks (study abroad), 9 weeks (immersion) 2–8 months

6 abroad, 5 at home

Iwasaki (2007)

US

1 year

4

Sasaki (2007)

Japan

4–9 months

7 abroad, 6 at home

Kinginger (2008)

US

1 semester

23

Educational Testing Service Listening and Reading Exams Educational Testing Service Japanese Proficiency Test Language Test Package for Core French Computer-Adaptive Test for Reading Chinese Free-recall protocols, Vocabulary knowledge, Self-assessment English composition quality/fluency Educational Testing Service Japanese Proficiency Test English composition quality/fluency Test de Français International

Canada or US Japan

Canada or US France

Measuring Language Acquisition 63

Four studies have focused on assessment of reading proficiency, often within broader research agendas, and all of these show that study abroad has a positive impact. Brecht, Davidson, and Ginsburg (1995) reported that students of Russian made significant gains in reading ability while abroad, and also that reading proficiency was an important predictor of gain in other domains (speaking and listening). Similarly, Lapkin, Hart, and Swain (1995) found that Anglophone students participating in a Canadian interprovincial exchange in Quebec improved their scores on a test of French-reading proficiency. As noted previously, American participants in Kinginger’s (2008) study also recorded significantly higher scores on the reading component of the Test de Français International after their sojourn in France. Hayden (1998) conducted a study of reading development in Chinese by European and North American students on a semester-long program in Beijing. Using the ComputerAdaptive Test for Reading Chinese (based on the ACTFL Guidelines), Hayden found that 14 of 21 participants registered higher scores in the post-test. Iwasaki (2007) documented impressive gains on the ETS Japanese Proficiency Test (Character/ Vocabulary, Reading/Grammar) for four American students after a year in Japan. In the case of reading in ‘orthographically complex Asian languages’ (Dewey 2007: 74), the results of research suggest that English-speaking study abroad participants may experience ‘the pang of illiteracy’ more intensely than do students of French or German (Librande 1998: 184) and that sudden loss of the ability to read can intensify desire for upgraded skills. The beginning students participating in Huebner’s (1995) study exhibited greater (though non-significant) gain scores on the reading component of the Japanese Proficiency Test than did their peers in a domestic immersion program. Huebner linked this result to the students’ journals and reactions to instruction. In the domestic program, the students found the introduction of Katakana and Hiragana writing systems in the first week, and Kanji in the second week, to be ‘unreasonable’ (1995: 185), preferring the Romanized instructional system of their textbook, a system used nowhere in Japan. Students abroad greeted this information with enthusiasm; having abruptly become functionally illiterate in their new environment, some among them ‘had a more urgent need to be literate in Japanese’ (1995: 185) and began to deliberately develop their reading ability through exposure to everyday print media within the linguistic landscape. Dewey (2004) compared development of reading proficiency and subjective responses to reading in Japanese among three groups in different learning environments: SA, IM, and AH. Compared with traditional

64

Language Learning and Study Abroad

classroom learners the study abroad group proved better able to comprehend and recall texts, better able to define written words, and more confident as readers. When comparing the domestic immersion students to their counterparts abroad, however, the only advantage accruing to the study abroad students was in their confidence. Dewey notes that the domestic immersion program offered considerable informal and personal access to instruction during which the students could test their reading acumen, but that the study abroad program provided a much broader variety of Japanese-mediated activities than did either of the domestic options. Writing development in study abroad has been examined by Freed, So, and Lazar (2003), and in a more recent pair of longitudinal studies in EFL contexts (Sasaki 2004; 2007). To recall, in the former study, native-speaking judges assigned superiority to the at-home classroom in fostering written fluency, but the study abroad participants were more ambitious in writing than the classroom students. Sasaki (2004) followed a small group of Japanese students majoring in British and American studies throughout their undergraduate careers to document their development as English-language writers. Of the 11 participants, six elected to study abroad in the United States or Canada, thus affording an opportunity to compare the two groups. Combining quantitative and qualitative assessments of writing ability, Sasaki showed that both groups developed the quality of their argumentative compositions and their fluency as writers (measured on site as total number of words and words per minute). However, the planning process evolved more for the study abroad group, and the members of this group were also far more likely than the classroom learners to express their ideas directly in English, without translating from Japanese, or to consider how to enrich the content of their writing while their compositions were underway. In a confirmatory study based on hypotheses emerging from the earlier work, Sasaki (2007) once again compared students abroad with their peers at home. The researcher once again measured general proficiency, essay quality (content, organization, vocabulary, language use, and mechanics), and fluency (words written and words per minute). Additionally, Sasaki collected retrospective accounts of the writing process through stimulated recall based on videotapes of the students as they composed, and interviewed the students about changes in their writing strategies. While both groups improved their overall proficiency, in this case, only the study abroad students showed enhanced writing ability and fluency. Also, the study abroad group became more motivated to

Measuring Language Acquisition 65

write well in English, whereas the students remaining at home did not. This motivation is reflected in the rationales underlying strategy use; for example, although there was no difference in the amount of translation from Japanese, some study abroad participants claimed to use this strategy in a specific effort to refine their composition. Sasaki (2007) concluded by noting the dilemmas faced by Japanese learners of English, as outlined in Chapter 1. If the primary concern is with achieving significant progress in L2 writing, then study abroad should be recommended for this entire population. However, in this case the participants were abroad during their third or fourth years of study, the crucial phase for seeking employment. While the study abroad group was enhancing language skills that may or may not prove relevant in their careers, the group at home may have been more successful in securing jobs, a process normally completed by the middle of the fourth year. A late start at job hunting can be ‘a devastating handicap in Japan’ (2007: 615). If the students at home became less-motivated learners of English, the waning of their enthusiasm can be attributed in part to their socialization into the L1 community. Overall, the scant literature in this domain offers some support for the claim that study abroad enhances writing proficiency and performance on tests of reading comprehension. Moreover, in some cases the perceived relevance of foreign language literacy tends to grow, and this perception alters the approaches that students adopt in their daily reading practices and approaches to academic writing. On the other hand, student responses are not uniform, and some students appear to reap greater rewards than do others. Both Huebner (1995) and Dewey (2004) pointed out that their findings mirror other comparative investigations of language learning in study abroad where individual differences in outcome are highlighted. Referring to student journals, Huebner noted that the participants in his study did not all react to their sudden illiteracy in the same way. Some students seemed to thrive whereas others expressed frustration in being surrounded by ‘indecipherable’ (1995: 186) writing. Among Hayden’s (1998) participants were five who showed no gain and two whose performance declined. Similarly, in my own study (Kinginger 2008) there were several students who, over the course of their time abroad, appeared to have lost some portion of their ability to succeed on a standard test of reading. In Sasaki’s (2007) study, two of the seven study abroad participants did not improve their writing quality scores. As Sasaki’s interpretation illustrates, a broader lens may be required in order to understand these findings, one that takes into consideration the societal role of reading and writing competence, the

66

Language Learning and Study Abroad

literacy activities of students abroad and the significance of literacy in students’ lives.

Summary and conclusion The measurement of language-related outcomes appears, on the surface, to have yielded a fairly consistent picture in which study abroad is a productive environment for language learning. Beginning with Carroll’s (1967) assertion that time abroad is a major predictor of language ability, researchers have documented gain in all modalities of language use. Efforts to show that study abroad enhances general speaking proficiency, although plagued by the imprecision of the OPI scale, succeeded in the case of a relatively large-scale study (Brecht, Davidson, and Ginsburg 1995). Parallel projects in Europe (Coleman 1996; Rees and Klapper 2007) also attribute growth in overall proficiency to study abroad. Attempts to substantiate claims about the development of fluency, although based entirely on monologic, laboratory-like tasks, indicate that study abroad participants generally speak more quickly and with less hesitation than their counterparts at home, and that advanced learners appropriate specific, fluency-enhancing features of spoken language grammar and style. Most test-based investigations of comprehension abilities show gains for students abroad, and Sasaki (2007) illustrated important changes in students’ ability to efficiently compose argumentative essays. Even when improvement is not documented using the available assessment methods, scholars argue that study abroad prompts changes in students’ dispositions toward language learning, boosting their communicative ambition in speaking and writing tasks, attracting their attention to features of the linguistic landscape affording reading practice, reworking their approach to listening in the direction of meaning-oriented strategies, or altering the motives underlying such strategies. Yet, experts on quantitative assessment and experimental research design do not necessarily agree that the linguistic benefits of study abroad have already been proven. A review of methodologies in study abroad research based on holistic constructs prompted Rees and Klapper (2008: 102) to declare that scholarship in this area is ‘embryonic and underdeveloped.’ They noted in particular the general paucity of investigations and the small scale of most study abroad research. The research base is small, and many of the studies reviewed in this chapter involved few participants, typically a number of volunteers from a particular university’s cohort. Rees and Klapper also critiqued several facets of design

Measuring Language Acquisition 67

in study abroad research. Negative appraisal of quantitative research on learning outcomes has focused on the lack of a control group in many quasi-experimental research designs, leading to poorly supported claims about capabilities of study abroad participants versus classroom learners (e.g., Freed 1995a). However, even when control groups are present there is insufficient consideration given to matching the academic and affective profiles of the participants whose performance is compared. This issue leaks into the interpretation of studies like Cubillos, Chieffo, and Fan (2007) or Sasaki (2007), where students who chose to pursue language competence abroad clearly presented or developed different stances toward this pursuit than do those who remained at home. Although it may not be possible, given these conditions, to secure genuine control groups, Rees and Klapper lamented the inadequate basis for comparison in most comparative studies. Finally, these authors pointed out that many quantitative investigations exhibit weaknesses in statistical methods. Some studies fail to address the reliability of testing instruments or to calibrate these tests in order to avoid ceiling effects, and others produce overgenerous significance values based on inappropriate choice of statistical procedure (e.g., use of Analysis of Variance (ANOVA) in the case of repeated measures). While Rees and Klapper (2008: 98) legitimately called for attention to the ‘psychometric pedigree,’ or the proven validity and reliability of tests assessing holistic language skills, a broader question concerns the extent to which current standard definitions of language proficiency, as operationalized in widely distributed tests, reflect academic norms versus the realities of natural language use and learning in all modes of communication, but particularly in speaking, the focus of greatest interest among students and researchers. ‘Proficiency’ may not be best represented as an implicational scale, but may instead be composed of unique, individual repertoires. As the research of Towell, Hawkins, and Bazergui (1996) demonstrated, students abroad may appropriate perfectly legitimate varieties or styles of speech that are not necessarily enshrined as correct in grammars based on written language. Until such time as tests based on empirical study of everyday, informal talk in a variety of languages become available, and until we know more about the language-related activities of study abroad participants, there will always be questions about the ecological validity of research in this domain. The measurement of outcomes for study abroad is ‘caught between a rock and a soft place’ where ‘the rock is rigor and the soft place relevance’ (Bronfenbrenner 1979: 18). Emphasis on rigor may motivate

68

Language Learning and Study Abroad

elegance in design, but the studies often call for short-term observation of unusual behaviors, such as filling in blanks or telling a story to a machine, that are not generalizable to other settings. Similarly, in research attempting a maximum level of generalization to whole populations, a recurrent finding is of striking individual differences whose explanation offers only tantalizing glimpses into the nature of study abroad experiences and the motives of the people involved. It is this dilemma that has inspired the qualitative approaches to language learning abroad examined in Chapters 4 and 5. Meanwhile, neither the subjective judgment of the so-called native speakers nor construct-based language tests, such as the OPI, the C-Test, nor any of the standard measures of comprehension or writing quality necessarily capture the subtle changes occurring as language learners abroad enlarge their communicative repertoires. A purely utilitarian approach to the quantification of products begins to encounter its limits when researchers struggle to portray the development they have observed with instruments that do not measure it. For this reason, study abroad research on language acquisition includes a significant number of studies examining discrete areas of language proficiency, to be reviewed in Chapter 3.

3 Domains of Communicative Competence

In Entre les murs (The Class) (2006), inspiring the film by the same title (Cantet 2008), author François Bégaudeau recounts his everyday experience of teaching in a tough, largely immigrant-populated Parisian neighborhood. Bégaudeau’s students face problems of poverty, violence, and cultural isolation. His Chinese student Wei struggles to learn French while coping with the threat of his mother’s deportation. Souleymane, an immigrant from Mali, plays out his identity confusion through self-destructive rebellion, and is ultimately expelled from school. Since Bégaudeau is a French teacher, the locus of his struggle to assist these students is the language, and particularly the grammar and style of proper, literate French. To the extent that they have in fact learned French, their language use betrays them, marking working-class origins, particularly because they rely on the norms appropriate for informal speech even when they write. In advance of a major gate-keeping test that will determine his students’ access to future schooling, Bégaudeau harangues them on the importance of avoiding spoken forms in writing: Bon, je reviens un peu sur l’oralité, je rappelle que c’est pas parce qu’on vous demande d’écrire un dialogue qu’il faut écrire comme on parle, vous voyez? D’ailleurs, on arrive jamais à écrire comme on parle, c’est impossible, tout ce qu’on peut faire c’est donner une impression d’oralité, c’est tout, alors on évite de commencer les phrases par ‘franchement,’ on évite de dire ‘on’ pour ‘nous,’ on évite d’utiliser ‘sérieux’ comme adverbe, comme vous faites en permanence à l’oral. (Bégaudeau 2006: 257) So, let’s go back a little bit to spoken language, let me remind you that even if they ask you to write a dialogue you shouldn’t write like 69

70

Language Learning and Study Abroad

you talk, do you see? Moreover, we never manage to write like we speak, it’s impossible, all we can do is give an impression of orality, that’s all, so we avoid starting sentences with ‘franchement,’ we avoid using ‘on’ for ‘nous,’ we avoid using ‘sérieux’ as an adverb, they way you constantly do when you are talking. (My translation) The author conveys both the urgency of his message and the contradictions inherent in his mission to expand and reshape his students’ literacy while reaching out to them through the use of the same oral forms he is called upon to condemn. Bégaudeau’s job is to work on his students’ communicative competence, teaching them to use literate and refined styles appropriate for instructional settings. Communicative competence is the parent construct of proficiency. Since the latter third of the 20th century, and particularly in places and at times when the functional value of language ability is highlighted, educators have generally concurred that language ability should be defined in terms of communicative competence, or the ability to express, interpret, and negotiate meaning (Savignon 1983). This concept, traced to the writings of anthropologist Dell Hymes (e.g., 1984), conveys the breadth of abilities, beyond knowledge of language forms, required for successful communication. Definitions of communicative competence normally include reference to appropriate language use, that is, the ability to match language forms to their social contexts in ways that do not unintentionally violate expectations. For Bégaudeau’s students the immediate challenge is to appropriate formal registers of written French. Many study abroad participants face the opposite problem; when their access to the language has been restricted to classrooms and to prepared texts, they need to expand their repertoire of spoken language in order to interact appropriately within informal settings. Pedagogical models of communicative competence normally differentiate between knowledge about language, the focus of traditional classroom learning, and ability to engage in purposeful language use. These models portray communicative competence as a number of interrelated yet separately conceived abilities. In the most recent mainstream version (Celce-Murcia, Dörnyei, and Thurell 1995) discourse competence is at the core, and consists of linguistic resources to create and interpret texts, that is, coherent and cohesive language in use (including all modes of communication such as speaking, writing, or texting). Three additional components influence and are influenced by discourse competence: (1) linguistic competence, or the ability to manipulate the formal features of language, including vocabulary, syntax, morphology,

Domains of Communicative Competence

71

phonology, and orthography; (2) actional competence, or the ability to interpret and produce speech acts (apologizing, requesting, promising, etc.); and (3) sociocultural competence, comprising knowledge of the rules and expectations governing the social contexts of language use, and including knowledge of politeness conventions, speech styles, and register. A fifth component, strategic competence, supports all the others, and consists of knowledge and skill needed to resolve communication difficulties and enhance both learning and communication. As a model for language learning, communicative competence was popularized in the 1970s and 1980s, during the height of political and educational enthusiasm for functionalism in language teaching. Hymes’ reflections on communicative competence had been intended to inspire research in the ethnography of communication, offering holistic, yet detailed analysis of language use in diverse settings. Whether or not the relevant, model speech communities can be easily identified, and before many of these norms were actually described, the construct was abstracted away from its scholarly origins to serve language education (Leung 2005). Because it bears the mark of its utilitarian roots, and because under its banner language learners have been enjoined to emulate the idealized competence of monolingual native speakers, the universal relevance of the model is now widely contested. Kramsch (2006b), for example, argues for recognition of the symbolic power inherent in the negotiation of meaning, and both Cook (1995) and Leung (2005) contend that models for language learners should hold up successful multilingualism as the ideal. Various authors have justly criticized the model’s unexamined, thus politically suspect, definitions of appropriate language behavior (Fairclough 1992) or its overemphasis on individual effectiveness over intercultural awareness and the dialogic qualities of communication (Byram 1997; Zarate 2003). Nonetheless, variations on the theme of communicative competence have guided many projects in applied linguistics over the past several decades, and this model provides a convenient heuristic for organizing a review of studies narrowing their perspective to particular aspects of language. Although it is not always certain, for example, that a given study should be considered most relevant to discourse competence rather than to actional or sociocultural competence, this chapter considers how study abroad researchers treat each component of communicative competence in turn. This research is often framed in relation to the broader field of SLA, a field in search of universal and context-independent psycholinguistic processes (Ellis 2000; Gass and Selinker 2001). Within SLA, study abroad is assumed to resemble an experimental treatment,

72

Language Learning and Study Abroad

providing ‘high quality, contextualized exposure’ to language (Isabelli 2007: 333) that may alter the route or accelerate the rate of development in a range of ways. Only rarely is the nature of this exposure scrutinized. Scholars examining linguistic competence have focused on study abroad participants’ ability to use second-language grammar, lexis, phonology and morphology. More recently, actional competence has become a popular focus of study abroad research, examining students’ ability to do things with words as they perform or comprehend speech acts. Discourse competence is at the core of the model, serving to ground the conceptualization of the other components in relation to meaning and coherence, and therefore cannot be easily abstracted from the whole. However, several studies have examined study abroad participants’ discourse competence in specific areas such as deixis and conversational structure. Research on sociocultural competence offers insight into the ways in which study abroad enhances students’ ability to manipulate sociolinguistic variation. Investigations of strategic competence scrutinize how students learn to compensate for lacunae in the other areas of language ability, how they resolve problems arising in interaction, and how they organize learning activity. Although the great majority of these projects retain a utilitarian flavor in their focus on concrete products, they also raise questions about language-learning processes in study abroad and about the nature of communicative competence itself.

Linguistic competence Linguistic competence involves the basic formal elements of language, including sentence or utterance patterns and types, morphology, vocabulary, and phonology/orthography (Celce-Murcia, Dörnyei, and Thurell 1995). With few exceptions (e.g., Rees and Klapper 2007), contemporary studies of linguistic competence, particularly in the area of morphology and syntax, are less concerned with knowledge about language than with functional abilities to use it. Scholars have looked to study abroad as a context for the development of grammatical and lexical abilities and, to a lesser extent, phonology. In the area of grammatical competence particularly, these studies have produced mixed results, leading some to suggest that ‘there is cause to doubt that SA [study abroad] leads to significant grammatical gains’ (Collentine 2004: 228). Under the influence of SLA, a field primarily devoted to exploring the development of linguistic competence, some scholars interested in grammatical competence (see Table 3.1) view context as incidental to

Table 3.1 Assessment of grammatical competence Study

Origin

Destination

Sojourn length

Participants

Instrument(s)/data

DeKeyser (1991)

US

Spain

6 months (abroad), 1 semester (at home)

7 abroad, 5 at home

Ryan and Lafford (1992) Walsh (1994) Guntermann (1995)

US

Spain

1 semester

16

Interviews, picture descriptions, grammar test OPI

Ireland US

Germany Central America

15 9

Interviews OPI

Collentine (2004)

US

Spain

8–12 months 8–10 weeks training, 1 year Peace Corps service 1 semester

OPI

Howard (2005)

Ireland

France

1 year

26 abroad, 20 at home 18

Isabelli and Nishida (2005) Howard (2006)

US

Spain

1 year

Ireland

France

1 year

29 abroad, 32 at home 12

McMeekin (2006)

US

Japan

8 weeks

5

Isabelli (2007)

US

Hispanophone countries

1 semester–18 months

24 abroad, 19 at home

Sociolinguistic interviews SOPI Sociolinguistic interviews Videotaped interactions SOPI-type questions

73

74

Language Learning and Study Abroad

the unfolding of universal and/or computational psycholinguistic processes. Research involving study abroad has often been driven by major theoretical stances within SLA, including theories about immutable stages in the acquisition of certain formal features of second languages (e.g., VanPatten 1985b), hypotheses regarding the role of Universal Grammar in second-language development (e.g., White 1989) or approaches interpreting the mind as analogous to a computational device, receiving linguistic input tailored for acquisition through interaction (e.g., Long 1981). Thus, investigators have entered the arena of study abroad research without necessarily taking an interest in the nature of study abroad, at least initially, except to the extent that it ostensibly upgrades the quality and amount of exposure to language. In addition to Freed’s studies based on native speaker assessment of fluency (Freed 1995b; Freed, So, and Lazar 2003), reviewed in Chapter 2, several investigations have shown little or no development of grammatical competence in study abroad. One commonly cited study is DeKeyser (1991), a project comparing small groups of US-based students of Spanish at home (N = 5) and abroad (N = 7). DeKeyser explored differences between the groups in monitoring for grammatical accuracy in interviews and picture descriptions. No difference between the groups was recorded, as none of the students possessed the relevant grammatical knowledge for monitoring their output in the areas under consideration (the subjunctive, the conditional and relative clauses) and they displayed their knowledge of the Spanish copula (ser versus estar, a notoriously difficult aspect of Spanish grammar for Anglophones) in similar ways. Similarly, Walsh (1994) reported the findings of a study involving Irish students of German interviewed at three intervals over the course of their year abroad. Analysis of the interview performances prompted the author to claim that students returned ‘having progressed largely in the area of fluency and lexis but only marginally in the area of grammatical accuracy’ (1994: 48). A typical student in this cohort manifested improved speed of delivery and a new repertoire of communication strategies, but these seemed to distract from grammatical and syntactic deficiencies, yielding ‘fluent ungrammaticality’ (1994: 52). Collentine (2004) combined a focus on grammatical and lexical development in US-based students of Spanish at home and abroad. A corpus was constructed from Oral Proficiency Interviews with 20 classroom learners and 26 participants in a program in Spain. These data were coded for accuracy in 17 morphological, syntactic, and

Domains of Communicative Competence

75

morpho-syntactic categories (e.g., copula accuracy, preposition accuracy, object-pronoun accuracy) and the researcher computed the frequency with which learners generated unique words within seven lexical categories (adjectives, adverbs, conjunctions, nouns, prepositions, pronouns, and verbs). Results indicated that the study abroad experience did not produce overall increases in grammatical abilities, and that, in fact, it was the stay-at-home group whose grammatical accuracy increased on the five accuracy types that most distinguished between the groups (copula, present tense verbs, indicative verbs, subordinateconjunction accuracy, and subordinate-clause count). These are areas, namely verbs and subordinate conjunctions, typically emphasized in language pedagogy. In the area of lexical development the study generated similar results, showing that the ‘at home’ group appeared to have acquired more unique lexical items than the study abroad group. Collentine then calculated a narrative score for each of the groups based on the frequently occurring grammatical features of narrative discourse (past-tense verbs, third-person morphology, past and present participles, and public verbs of communication and events). This analysis demonstrated that the study abroad group produced more of the features associated with narrative than did the classroom learners. Having scaled the data for this purpose, the author also applied a measure of informational richness, concluding once again that the discourse produced by the classroom students was more semantically dense than that of the study abroad participants. Thus, students abroad may in fact develop repertoires different from those of classroom learners, based, for example, on experience of interactive story telling, but traditional measures of grammatical accuracy do not necessarily capture the changes taking place when language learners emerge from the classroom. Collentine recommended that future researchers rely more on research in discourse and corpus analysis, where scholars are ‘beginning to uncover metrics for characterizing discursive levels of representation that combine grammatical and lexical features’ (2004: 245). By contrast, studies pinpointing the development of particular grammatical features offer a somewhat rosier outlook. As Howard remarks (2005: 7), ‘it might be hypothesized that grammatical development may be more accelerated on certain components of the learner’s grammatical competence, rather than being equally evident across the learner’s grammatical competence as a general entity.’ Two studies examined the order of acquisition of certain grammatical features of Spanish. Guntermann

76

Language Learning and Study Abroad

(1995) scrutinized Oral Proficiency Interviews with Peace Corps volunteers in Central America for three grammatical features: the copulas ser and estar, the prepositions por and para, and past tense and aspect. After one year, these participants increased the accuracy of their use of the copulas and of the preterit and imperfect. In another such study, Ryan and Lafford (1992) investigated whether the study abroad environment would alter a hypothetical developmental sequence for the acquisition of ser and estar based on observation of classroom learners. The authors concluded that the developmental sequence is the same in both contexts, and that students abroad did in fact gain in their abilities in this domain. These studies point to links between grammatical and lexical development, since the student’s repertoires included both formulaic chunks (e.g., por ejemplo) and lexico-grammatical syntagmas (e.g., ser + adjective, estar + preposition) (Collentine 2004). The research program of Isabelli (Isabelli and Nishida 2005; Isabelli 2007) exemplifies investigation guided by the Universal Grammar framework. Under this framework, inspired by Chomsky’s distinction between underlying competence and surface-level performance, it is assumed that all languages share certain abstract principles of grammar, and that these are innate to humans. Language learning is grammar building (Schwartz 1993), and depends upon access to Universal Grammar, learning procedure, and linguistic input (Isabelli 2007). For SLA, a key question is the extent to which the principles of Universal Grammar remain accessible in learning languages beyond the first. The debate centers around two positions and their consequences for learning procedure. One of these stances is that first- and second-language acquisition are in essence the same process, and that only positive evidence in the input is necessary for grammar building. The second holds that in SLA access to Universal Grammar is indirect, and that learners therefore need negative evidence (i.e., correction) and explicit positive evidence (i.e., instruction) in order to build grammars. One purpose of Isabelli’s forays into the domain of study abroad is to contribute to this debate about the role of Univeral Grammar and its implications for the design and use of learning environments. Isabelli and Nishida (2005) investigated the acquisition of the subjunctive in Spanish for a cohort of US-based students after nine months of experience abroad. The study involved Simulated Oral Proficiency Interviews, or SOPIs (an audiotaped and standardized version of the face-to-face procedure), and revealed that in comparison to their peers at home the study abroad group produced more subjunctive-related structures and more actual forms of the subjunctive. That is, they produced more

Domains of Communicative Competence

77

complex syntactic structures, since the subjunctive normally appears in subordinate clauses. However, the rate at which they produced forms of the subjunctive within relevant structures was only 33 percent. Thus study abroad, with its abundance of positive input, does not lead to ‘a clear notion of what the subjunctive morpohology semantically represents’ (Isabelli 2007: 332) but it provides a degree of readiness and ‘linguistic maturity’ (2007: 336) needed to understand complex syntax. A follow-up study investigated whether or not former study abroad participants would prove more amenable to instruction than classroom learners (2007). In fact, the second study, once again based on relatively spontaneous linguistic production in the SOPI, documented a distinct advantage for those students who had studied abroad, confirming the author’s hypothesis that indirect access to Universal Grammar characterizes second-language learning and the need for a judicious combination of positive evidence and explicit instruction. Another approach to the refinement of research constructs is illustrated in the work of Howard (2001; 2005) on the development of abilities to mark past tense and aspect in advanced learners of French. Speaking accurately about the past presents considerable difficulties for Anglophones, since this feature of French grammar requires skilled use of complex morphology and awareness of concepts underlying perfective versus imperfective aspect. Howard’s (2005) project was based on a corpus of sociolinguistic interviews conducted within a cross-sectional design and involving three groups of Irish participants: second-year students, third-year instructed learners, and students having spent their third year of college in France. This design allowed Howard to describe the commonalities and differences in past time marking between and among the groups. Howard (2005) confirms that the process of mastering appropriate use of these forms is complex and gradual. All three of the groups relied on temporal adverbs and verbs in the present to mark past tense, failed to supply the pluperfect in cases where it would be expected, and used the passé composé (perfective marker) in expression of habituality where the imparfait (imperfective marker) is required. When analyzed from a lexical perspective, Howard’s data showed that the students systematically applied the imparfait to a limited number of verbs, and the passé composé to a larger, different group of verbs. That is, they restricted their usage of particular lexical verbs to one past time marker but not both. In comparing the performance of the study abroad group to the others, however, several differences emerged. The study abroad participants used more past tense forms, matched these forms more successfully to

78

Language Learning and Study Abroad

their functional context, and extended past time marking to a wider range of lexical verbs. They relied less on temporal adverbs, and thus made more autonomous uses of past time marking. Moreover, their use of the imparfait to mark habituality and progressivity, a function posing considerable difficulty to instructed learners, reached a rate of 67 percent (as compared to 25 percent for the second-year students, and 48 percent for the third year classroom group). Thus it would appear that while study abroad participants do not fully acquire the French aspectual system, they increase their range of appropriate use (see also Duperron 2006). A companion project focusing on the expression of person and number in verb morphology, Howard (2006), showed that the marking of plurality in third-person contexts was especially challenging for all the students, but that the study abroad participants best approximated native-speaker norms in this area as well. Conditions for the development of grammatical competence are explored in McMeekin’s (2006) examination of ‘negotiation’ in a Japanese study abroad setting. This study is grounded in the Input– Interaction–Output framework for SLA (Block 2003), based originally on Krashen’s Monitor Model (1982), in which the provision of comprehensible input drives an essentially computational process of language acquisition. Subsequent revisions and refinements of the framework emphasize ‘negotiation of meaning,’ or ‘the modification and restructuring of interaction that occurs when learners and their interlocutors anticipate, perceive, or experience difficulties in message comprehension’ (Pica 1994: 494, cited in McMeekin 2006: 180). Negotiation is assumed to play three roles in furthering language acquisition: (1) making input comprehensible; (2) encouraging learners to modify their output; and (3) focusing learners’ attention on form through positive and negative evidence. McMeekin videotaped five US-based learners in both homestay and classroom settings, then scrutinized the amount and quality of negotiated interactions. The quantitative data show that negotiation was far more frequent in the homestay than in the classroom. Host families clarified their utterances, thus providing ostensibly comprehensible input, more than teachers. However, students did not modify their own utterances as much in the host family setting as they did in the classroom. The host families in McMeekin’s study went to great lengths to explain their utterances to students through repetition, reformulation, explanation and circumlocution. In the following interaction, for example, Mandy receives clarification of the word utsutte iru (to be filming) from her host parents:

Domains of Communicative Competence 1.

F:

2. Ma: 3.

F:

4. Ma: 5.

F:

6. M: 7. Ma: 8.

F:

9. M:

79

mandy utsutte ru kana I wonder if Mandy is being filmed. nani? what? utsutte ru kana? Is she being filmed? utsu-? film-? kore [kore bideo kamera ni utsutte imasu [ka ((points to camera)) this [this are you being filmed by the camera [utsutte ru wa [kochigawa kara yoku utsutte iru [it’s filming [it’s filming well from this side utsutsu wakarimasen donna imi desu ka I don’t understand utsutsu what does it mean? utsuru it’s utsuru bideo ni utsuru utsuru to be filmed by a camera (McMeekin 2006: 189–190)

McMeekin’s data illustrate the process by which everyday interaction in host family settings can be construed as a series of language lessons with ongoing explicit focus on form (see also Chapter 5). Although McMeekin does not document the growth of grammatical competence for her participants, the study offers insight into the potential of home-based interactions to broaden grammatical and lexical repertoires. In addition to studies of grammatical competence, the literature includes two major investigations of vocabulary development in study abroad. The findings of Milton and Meara (1995) and of Ife, Vives Boix, and Meara (2000) indicate that study abroad participants develop more expansive lexical repertoires than do purely instructed learners, and that their lexicons tend to be organized in a more native-like way. Using the Eurocentres Vocabulary Size Test, Milton and Meara produced estimates of vocabulary size in 53 European exchange participants studying in Britain. The group as a whole showed a remarkable improvement in scores, with some students more than doubling their initial vocabularies. The average growth rate during study abroad was about four times as fast as the growth rate at home. However, these results masked significant individual differences; one student’s score remained unchanged, and five students registered lower scores on the final test than on the entry test. Part of this variation can be attributed to the tendency for

80

Language Learning and Study Abroad

students with smaller initial vocabularies to make the greatest gain, and for students with large vocabularies to register less-significant gain scores. The authors also consulted the results of a questionnaire detailing students’ activities while on the exchange. To their surprise, these findings showed that time spent in both formal and informal study (‘locking yourself away in a library,’ 2000: 32) correlated negatively with vocabulary growth. Although friendship patterns did not appear to affect gains in vocabulary, the authors report that the participants adopted an expatriate lifestyle, mixing with other exchange students and only rarely with locals. Subsequently, Ife, Vives Boix, and Meara (2000) acknowledged a possible ceiling effect for the Vocabulary Size Test, limiting the documentation of growth for advanced students. In its place, they proposed the A3VT (three word association test) permitting assessment of lexical organization. This move was warranted by the authors’ view that increased proficiency would promote ‘greater depth and greater interconnection between lexical items so that the learner’s lexicon resembles more closely that of a native speaker’ (2000: 59–60). In the A3VT, students are presented with sets of three words, where two of the words have been identified by natives as strongly associated with each other, while the third word is not associated with the other two. The test taker is instructed to identify the misfit word in each set. Participants in the research were British learners of Spanish of varying initial proficiency levels who were tested before and after a sojourn in Spain using the A3VT and a separate single word translation task. In this case, improvement over time was significant in both tests for all the students, whether they displayed intermediate or advanced initial proficiency. Longer periods abroad yielded higher gains in overall vocabulary and native-like lexical organization, no doubt because added time provides more chances of social integration. A final domain of linguistic competence under consideration in the study abroad literature is phonology, often framed as the extent to which students abroad appropriate native-like pronunciation. In this domain results are mixed, with most studies providing modest support for claims about the benefits of study abroad. Díaz-Campos (2004), for example, set out to investigate consonantal phenomena identified in textbooks as difficult for learners of Spanish (voiceless stops, wordinitial laterals, intervocalic voiced fricatives, and the palatal nasal). He compared the performance of US-based study abroad and classroom students on a reading task targeting these phenomena, and found similar patterns of development in both groups. All of the students made

Domains of Communicative Competence

81

progress in pronouncing voiceless stops and word-initial laterals, no gain was registered for intervocalic voiced fricatives, and the groups had mastered the palatal nasal prior to the pre-test. In a study defining fluency as accurate pronunciation of syllabi nuclei, only two of five participants made appreciable progress while studying in Costa Rica (Simões 1996). In collaboration with the SALA project, Mora (2008) tested the ability of Spanish–Catalan bilinguals based in Barcelona to discriminate between English contrastive sound units and to produce English voiceless oral stops. To recall, the SALA project involved longitudinal data collection over periods of instruction and of study abroad, and over a longer-term of 15 months. Mora’s results show that at home instruction alone significantly improved students’ perceptual phonetic/phonological ability. The ability to produce voiceless stops improved slightly, but not significantly, with study abroad, but these gains were no longer in evidence at the end of the study. On the other hand, Stevens’ (2001) acoustic analysis of pre- and post-test phonetic data from US-based learners of Spanish revealed a significant advantage for study abroad participants over classroom students. Similarly, O’Brien’s (2003) study of second-language German vowels indicated that American students studying in Germany came to perceive and to produce these vowels in native-like ways, as determined via acoustic analysis. Furthermore, native-speaker judges presented with a number of speech samples from the study abroad participants assigned possible nativeness to these speakers, both in the larger study (2003) and in a companion analysis of prosodic features such as rhythm, stress, and intonation (O’Brien 2004). Lord’s (2000) findings suggest that students who receive instruction on issues of pronunciation prior to study abroad make greater gains. In this case, participants who had formally studied Spanish phonology before a sojourn in Mexico attained 28.7 percent accuracy in a problematic area (voiced fricatives) whereas accuracy in the same area among their companions without instruction reached only 5.8 percent. Although it is the most traditional domain of research on developing language ability, scholarship on linguistic competence in study abroad at first glance resembles a hit-or-miss affair or a patchwork of diverse approaches grounded in pedagogical concerns or mainstream approaches to SLA. Concerns with overall accuracy have guided researchers’ unsuccessful attempts to show that study abroad reduces error rates (Walsh 1994) or enhances students’ control of discrete grammatical features (Collentine 2004), and these findings contribute to arguments for a more refined approach to grammar. However, when

82

Language Learning and Study Abroad

pinpointing specific aspects of grammatical competence for scrutiny, researchers may miss the mark, choosing features beyond the current reach of their participants (DeKeyser 1991) or acquired prior to the sojourn abroad (Díaz-Campos 2004). The scholarship in this area is fragmented by researchers’ various allegiances to major theoretical stances within SLA where the main concern is not to examine study abroad as a learning context per se, but to defend and illustrate the role of developmental sequences (Ryan and Lafford 1992; Guntermann 1995), Universal Grammar (Isabelli and Nishida 2005; 2007), or input and interaction (McMeekin 2006). In cases where scholars have chosen to examine relevant components of linguistic competence, such as the subjunctive in Spanish (Isabelli and Nishida 2005; Isabelli 2007), past tense and aspect in French (Howard 2001; 2005), or vocabulary size and organization (Milton and Meara 1995; Ife, Vives Boix, and Meara 2000), findings most often confirm the valuable contribution of study abroad in the context of a longer-term learning process. With one exception (the faux-Native Speakers of German in O’Brien 2003), and particularly in the case of short-term programs, study abroad does not appear to be a cure-all for problems of linguistic competence. However, the research does show accelerated development and heightened linguistic sophistication in many, if not all cases. This research points to a crucial role of language instruction both before students’ departure (Lord 2000) and upon their return, when their linguistic maturity has primed them to receive it (Isabelli 2007). Despite the variety of frameworks guiding this research, and its development to date in parallel with trends in SLA, research on linguistic competence in study abroad is, in effect, quite traditional. Regardless of broader theoretical orientations, researchers choose topics already familiar to language pedagogues and known to chart hazardous territory for students. Overall discrete-point accuracy, ser and estar, the passé composé versus the imparfait, or the voiced fricative are standardissue pedagogical problems, belonging among the concerns of classroom learning, with its emphasis on conventional forms. Especially in the area of interactive abilities such as conversation or texting, it is not clear that these are the aspects of linguistic competence most significant in study abroad settings. In the future, study abroad research will need to come to terms with variation in language in general and particularly with nonpedagogic language use, including the interaction of grammar, lexis, and discourse noted in passing by Collentine (2004) and revealed in corpusbased descriptive approaches (e.g., McCarthy 1998). It will also need to

Domains of Communicative Competence

83

address lingua franca communication and the competence of multilingual speakers (Cook 1995; Leung 2005) who do not necessarily aspire to emulate natives. In the meantime, however, study abroad researchers have moved toward a more contextually meaningful approach through investigation of actional competence.

Speech acts A number of researchers have interrogated the abilities of study abroad participants to manipulate speech acts (see Table 3.2), or ‘verbal schemata that carry illocutionary force’ (Celce-Murcia, Dörnyei, and Thurell 1995: 17). This ability is related to actional competence (or the ability to convey and understand communicative intent) and concerns the ways in which language learners ‘do things with words’ (Austin 1975) such as requesting, apologizing, or responding to compliments. At minimum, competence in the use of speech acts requires that students appropriate a range of formulaic sequences and strategies, matching these linguistic forms to situations of use in recognizable ways. They need to know, for example, that I’m sorry or Please forgive me typically correspond to an apology. Further, they need to grasp the subtleties of indirect speech acts in cases where a speaker’s intent can only be inferred from the interaction of linguistic form and context, as for example when speakers of American English show agreement with ‘Is the Pope Catholic?’ (Bouton 1992) or in the following example from Taguchi (2008): John: How was the wedding? Mary: Well . . . the cake was OK. (Taguchi 2008: 68) In terms of speech acts, at least in principle, study abroad offers an important advantage over formal instruction in that students observe a variety of everyday linguistic actions and begin to do socially consequential things with their own words. The relationship between speech acts and their context is not deterministic, and very often the same speech act can be realized directly or indirectly using a variety of forms. In the above example, Mary might just as easily have said, I thought it was gruesome, or Thank God it’s over or I survived, or in fact many other things. Success in speech act realization is difficult to measure in any straightforward way, and depends both on identity-related issues and on the extent to which students opt

84

Table 3.2 Research on speech acts Study

Origin

Destination

Sojourn length

Participants

Focus/instrument

Kondo (1997)

Japan

US

1 year

Apology/DCT

Hoffman-Hicks (1999)

US

France

16 months

48 NS of Japanese, 40 NS of English, 45 abroad 25 NS of French, 10 NS of English, 14 abroad

Matsumura (2001)

Japan

Canada

1 year

102 at home, 97 abroad

Owen (2001) Rodríguez (2001)

US US

Russia Spain

1 semester 1 semester

Barron (2003)

Ireland

Germany

10 months

60 NS of Russian, 42 abroad 14 NS of Spanish, 11 abroad, 11 at home 33

Shardakova (2005)

US

Russia

Varied

Schauer (2006)

Germany UK

1 year

Cohen and Shively (2007)

US

Magnan and Back (2007a) Matsumura (2007)

1 semester

86, randomly assigned to experimental/control groups

US

Spanish- or French-speaking countries France

Greeting – leave taking, compliments/Dialogue construction task Advice/Multiple choice questionnaire Request/OPI role-play Request/Appropriacy judgment task Request, offer, refusal/Free DCT Apology/Dialogue completion questionnaire Pragmatic infelicity/ Video-and-questionnaire, Post-hoc interviews Request, apology/Speech Act Measure of Language Gain

1 semester

6

Request/OPI role-play

Japan

Canada

8 months

15

Quebec

1 year

US

4 months

20 NS of Quebecois, 17 NS of German, 7 abroad 44

Advice/Multiple choice questionnaire Apology/DCT

Warga and Schölmberger Austria (2007) Taguchi (2008) Japan

41 NS of Russian, 90 learners, 59 abroad 20 NS of English, 17 at home, 16 abroad

Implied meaning/Pragmatic Listening Task

Domains of Communicative Competence

85

to define situations as expert language users do. In learning to perform second-language speech acts, students do not ‘blindly copy NS [native speaking] norms,’ but reserve the right to construct their own identities (Shardakova 2005: 423). In other words, the extent to which students’ speech acts become native-like is question of how much their experience permits them to see things from the point of view of natives, and how much they choose to adopt these perspectives as their own. Given the conceptual complexities associated with research on speech act realization, and the practical challenges of gathering naturally occurring complaints, requests, or apologies, most researchers design projects using elicited data and invoking Native Speaker usage as the norm. A typical project involves collecting data on one or more speech acts using some form of the Discourse Completion Task (DCT), where participants are given a range of situations in which they must play act their own or others’ contribution in conversational slots reserved for specific acts. Preferences for the components and structure of the speech act are abstracted from native-speaker data and applied to the analysis of learner performance. Only rarely do these projects consider how the participants define the situations, moving beyond surface-level comparisons between learner and expert usage to consider the sociocultural or identity-related meaning of these phenomena. Study abroad participants are better able to approximate native norms in their realization of speech acts than classroom learners, but they rarely appropriate an entirely native-like repertoire. Warga and Schölmberger (2007), for example, documented a mixture of native and non-native like apologies in Austrian learners of Quebecois French. In Hoffmann-Hicks’ (1999) research with US-based students of French, the participants displayed their ability to perform requests and compliments, providing dialogues in response to a series of scenarios. These students out-performed their counterparts at home, having decreased their use of some dispreferred compliment structures and begun to appropriate some of the features of native-speaker usage, but this progress was gradual and slight. Moreover, many of the students based their requests on American norms for interaction, having failed to perceive, for example, that French students rarely impose lengthy appeals for help or for meetings outside class upon their professors. In a study of requests, also with American learners of French, Magnan and Back (2007a) found that students whose overall speaking proficiency increased (as measured by the OPI) also developed greater ability to craft an appropriately formal stance and to balance indirect and direct forms in the test’s role-play exercise. While these characteristics brought the

86

Language Learning and Study Abroad

students’ speech closer to native norms, some native baseline request features, such as the negative declaration (La compagnie n’as pas de moyen de me faire partir plus tôt? [‘The company has no way to speed my departure?’]), were not observed in the data at all. Similarly, in Kondo’s (1997) research on the acquisition of apologies by Japanese students of English in the United States, pre- and post-DCT results showed that study abroad participants moved toward the strategy preferred by native speakers (i.e., providing an explanation), but also increased their use of an approach favored in Japanese (i.e., showing concern for the hearer). Kondo ascribed this development to a mismatch between proficiency growth and sociopragmatic ability; in essence, the students became better able to express concern but did not learn to avoid this approach in the way native speakers do. Barron (2003) used a modified DCT to examine the realization of requests, offers, and refusals by German native speakers and a cohort of 33 Irish students in Germany. The participants in this study moved toward the native norm in several ways. For example, they used fewer ritual re-offers, increased their reliance on German pragmatic routines, and discontinued the use of routines transferred from English (e.g., Ich wundere mich [I wonder]). They also increased their use of syntactic and lexical downgraders in some situations. However, overall Barron suggested that even when investing a full year in study abroad, students may not have the level of access to meaningful interaction that would be required to develop truly native-like competence in speech act performance. In studies employing methodologies other than the DCT, results are mixed. Based on a multiple-choice meta-pragmatic judgment test, Matsumura (2001) showed that Japanese university students in Canada became better able than their peers at home to judge the appropriateness of advice given to people of equal or lower status. Schauer (2006) tested the extent to which German students of English were able to detect pragmatic infelicities in a series of videotaped vignettes, and found that a sojourn of nine months abroad honed the students’ judgment to the point where they could no longer be distinguished from native speakers. On the other hand, in Rodríguez’s (2001) longitudinal study both classroom and study abroad participants moved toward the native speaker norm in their judgments of requests in Spanish, and both contexts appeared to be equally valuable for learning in this domain. Owen (2001) examined how requests were accomplished in OPI roleplay data from 75 US-based students in Russia, with findings similar to those of the DCT-based studies. These students learned to perform requests in a more native-like way, but retained a tendency to prefer

Domains of Communicative Competence

87

speaker-oriented head acts (e.g., I want . . .) over the impersonal and hearer-oriented requests of Russians (e.g., Is it possible . . ., Could you . . .). Other scholars have approached the relationship of study abroad and actional competence from a variety of innovative angles. Taguchi (2008), for example, examined how comprehension of indirect speech acts progressed over the course of a sojourn abroad by Japanese learners of English in an intensive US-based program. The study involved a computerized test in which students responded with ‘yes’ or ‘no’ answers about the implied meaning of indirect refusals and opinions, such as Mary’s opinion of the wedding previously cited. Participants manifested increased accuracy in assessing refusals, but only modest gains for less conventional, indirectly expressed opinions. Matsumura (2007) explored the aftereffects of study abroad on pragmatic competence, focusing once again on how Japanese students of English respond to a multiple choice test offering an array of direct and mitigated ways to offer advice. In this case, one year after a sojourn in Canada the respondents’ approach to giving advice diverged from native-speaker norms, especially in scenarios involving higher status interlocutors where they would decline to proffer words of wisdom to their hierarchical superiors. Matsumura’s qualitative findings, based on a focus group interview, suggest that the students were in fact accommodating to a perceived local norm, but that they were also on the verge of joining the workforce and were no longer positioning themselves primarily as students. Insecurity about this impending change heightened their sensitivity to power differentials. Above all, the students had reflected at some length about how to craft an appropriate public persona and had explicitly compared the relatively laid back style of Canadian professors to the typically more formal conventions governing interactions with teachers in Japan. The study abroad experience, bringing exposure to different ways of enacting hierarchy in interaction, had increased the flexibility of their communicative repertoires, that is, their ability to perceive variation in social situations and adapt their language use accordingly. In light of the relatively modest gains documented in the literature, Cohen and Shively (2007) took an interventionist approach to speech act development in study abroad. This project involved 86 US-based students of French and Spanish studying in diverse locales and divided into two groups. The experimental group participated in a pre-departure workshop illustrating the speech act concept, and was assigned regular readings from a manual providing explicit instruction on language and culture learning strategies: Maximizing study abroad: A student’s guide to strategies for language and culture learning and use (Paige, Cohen, Kappler,

88

Language Learning and Study Abroad

Chi, and Lassegard 2002). Both groups completed the Speech Act Measure of Language Gain, a version of the DCT in which they supplied multiple rejoinders within written simulations of requests and apologies, and these responses were rated by native-speaking judges. A comparison of pre- and post-test results showed a significant gain for both the control and the experimental groups, but only minor, non-significant differences between them. In other words, study abroad alone appeared to have enhanced these students’ awareness of speech acts, but the strategy-based intervention did not convey a statistically confirmed advantage. Cohen and Shively argued that perhaps ‘the intervention in this area was not long enough, not intensive enough, not effective enough in terms of its content, or not deemed important enough to pay attention to in the moment of production’ (2007: 202). Another explanation for the low impact of the strategy-based intervention may be that, as previously noted, learning to perform speech acts in a second language is as much a question of identity as is of language acquisition. This perspective is illustrated in the most compelling study in the literature on speech acts in study abroad to date: Shardakova’s (2005) research on Russian-language apologies by US-based students, a work distinguishable from the majority of speech act studies in the depth of interpretation permitted by the research design. The author claimed that speakers of Russian evaluate situations involving intimacy, unfamiliarity, or unequal social status in ways that are distinct from Americans’ perceptions of the same situations. Apologies are crafted in accordance with these views, which means that learning to apologize in Russian is more than merely matching the forms used by natives; it requires the ability to ‘see things from the point of view of a Russian’ (2005: 445). Shardakova’s project involved four groups composed of students at intermediate and advanced levels of proficiency and with or without experience of study abroad in Russia. These groups, along with native informants, first ranked both the severity of offences and the distance and power differentials described in the DCT task used to elicit data. As per her hypothesis, Shardakova found that the American students rated offenses involving strangers as most severe, whereas the Russians assigned greatest severity to transgressions involving power differentials (e.g., speaking with an authority figure). In terms of social distance and power, for the American students, friends and professors formed a single category of middle-distance people, as opposed to strangers, whereas for the Russian participants friends were considered much closer than strangers or professors who formed one category. Shardakova

Domains of Communicative Competence

89

demonstrated that proficiency developed in a domestic immersion program may slightly increase the number of apology tokens that students use, but that ‘increases in proficiency alone seem to have no effect on learners’ socio-pragmatic development’ (2005: 434). That is, only a combination of high proficiency and exposure to Russian culture provided students the option of behaving in a native-like manner. Shardakova’s project thus provides unique insight both on the specific advantages of study abroad and on the conceptual dimensions of second-language speech acts. By comparing students of high proficiency who have different experiences of language learning, namely domestic immersion versus study abroad, this author brings to light the importance of learning to conceptualize communicative settings as expert speakers do. However, this achievement will not necessarily prompt desire to model native speakers’ behavior. As in the case of other variable sociolinguistic features, the performance of speech acts involves an element of choice for all language users. Language learners may opt to assimilate, or they may cling to familiar forms of self-expression conceived as elements of their personal identity. As in the case of Hoffman-Hicks’ (1999) Americans in interaction with French professors, they may not enjoy the depth of language socialization that would permit them to rework their basic understanding of what transpires in particular communicative settings. Alternatively, as illustrated in Matsumura’s (2007) project, they may very well perceive fundamental differences in interactive norms, but elect not to follow the second-language conventions they have observed. Opportunities to observe and participate in speech acts abroad may enhance students’ communicative flexibility, based on awareness of potentially conflicting norms (Matsumura 2007). However, students may or may not fully grasp the organic links between these acts and their local meaning. To the extent that they do achieve depth of understanding, they may or may not choose to apologize as Russians do, offer advice to hierarchical superiors like Canadians, or modify their view of students’ rights in relation to professors. Speech acts are multiple indexicals, pointing at once to both situational and identityrelated dimensions of language use. This multiple indexicality means that students may not believe that speech acts contribute to language proficiency per se rather than to personal self-expression. Moreover, a strict focus on the extent to which students copy native norms can capture neither the creativity, nor the array of options, nor the sociocultural resources involved in second-language speech act performance. Nevertheless, most research on speech acts in relation to study abroad

90

Language Learning and Study Abroad

skirts these issues, preferring to elicit data using DCTs, which have been shown to generate results differing from language use observed in situ (Golato 2003), or the OPI, whose interactive characteristics have been amply critiqued (see Chapter 2). Above all, as elsewhere in the literature, this research normally invokes a deficit model of the non-native speaker (Belz 2002) whose performance is characterized only in terms of failure to match native norms. Clearly, speech acts represent a key area of interest for languagelearning research. The most revealing projects combine assessment with qualitative inquiry, offering both evidence for specific achievements and resources for interpretation of these findings in relation to the meaning of language use for students and expert speakers. It is much to be hoped that future researchers will design projects in recognition of the complex, multiple indexicality of speech acts, emulating Shardakova’s (2005) efforts to apprehend not only performances but also variable interpretation of situations. It is also clear that the performance of speech acts in laboratory-like settings does not necessarily correspond to naturally occurring language use. Hence, process-oriented scrutiny of language students’ routine interactions in study abroad settings (see Chapter 5) will contribute further understanding, both for descriptions of learner language use and for portrayal of developmental processes.

Discourse competence Discourse competence, closely related to actional competence, concerns the ‘selection, sequencing and arrangement of words, structures, sentences and utterances to achieve a unified spoken or written text’ (Celce-Murcia, Dörnyei, and Thurell 1995: 13). The domains of discourse competence highlighted by these authors include cohesion, deixis, coherence, generic structure, and conversational structure. Cohesion is the aspect of discourse most closely associated with linguistic competence, describing the ways in which linguistic elements such as pronouns, articles, or demonstratives signal co-reference within texts. Other cohesive devices include conventions for avoiding undue repetition through substitution and ellipsis, or the use of conjunction, lexical repetitions, and other explicit links within texts. Deixis is the process by which texts index, or point to features of their situational context through devices like personal pronouns (‘I’ or ‘you’) or temporal reference (‘now’ or ‘then’). Coherence, the most elusive of textual qualities, is the extent to which ‘sentences or utterances in a discourse sequence are felt to be related rather than unrelated’ (1995: 13), and depends upon

Domains of Communicative Competence

91

local preferences for text organization, thematic scope and sequencing, and broader situational and sociocultural knowledge. Coherence is achieved, therefore, through observance of generic and conversational structure, the specific cultural norms defining the structure and content of texts, or in the case of conversation, turn-taking. Participation in conversational interaction requires to ability to open and close conversations, introduce and change topics, recognize and contribute to routine adjacency pairs (e.g., Hi, how are you/Fine, thanks) and repair one’s own or others’ utterances. A number of projects in the study abroad literature touch on issues surrounding on the development of discourse competence (see Table 3.3). Although researchers have rarely summoned the courage to tackle a question as thorny as the development of coherence in discourse, results of some studies illustrate the role of study abroad in furthering student’s command of cohesion, deixis, and conversational structure. As noted in Chapter 2, the C-Test is a holistic measure of reduced redundancy in written texts, and therefore calls upon test takers’ ability to derive a general sense of textual meaning from limited resources, and to restore the overall cohesion of the text. Fraser (2002) tested US-based students of German before and after study abroad using a cloze passage, based on the same principles as the C-Test, and an explicit measure of the extent to which students’ reading abilities were enhanced through improved apprehension of textual cohesion. This test required that students identify the anaphoric and cataphoric referents of pronouns in a journal article about German cinema. Fraser’s results showed that students who were actively engaged in local activities and social networks (e.g., a football team, a theater internship) made greater gains in these discourse-level abilities than did their peers who limited their German-mediated activity to formal study. Grieve (2007) studied the acquisition of pragmatic markers for vagueness by a cohort of German students in Australia. Markers of vague language are used primarily in conversation either for transactional functions such as holding the floor while searching for a word or for interpersonal functions such as saving face and maintaining solidarity in an informal, chatty atmosphere. In English, these markers include general extenders (e.g., and all that), approximators (e.g., like), hedges (e.g., sort of ), and shields (e.g., I guess). Following Brinton (1996), Grieve pointed out that the absence of such markers can make an utterance seem unnatural, awkward, unfriendly, or even dogmatic. At the end of the treatment period, experimental and control group participants remaining in Germany used markers for vagueness at similar rates but

92

Table 3.3 Studies of discourse competence Study

Origin

Destination

Sojourn length

Participants

Instrument(s)

Lafford (1995)

US

Spain or Mexico

1 semester

OPI role-play

Maeno (1995) Marriott (1995) DuFon (2000)

US Australia US

Japan Japan Indonesia

Varied 1 year 4 months

13 in Mexico, 16 in Spain, 13 at home 10 8 6

Sax (2001)

US

France

Varied

Fraser (2002)

US

Germany

1 or 2 semesters

Lemée (2002)

Ireland

France

Varied

Barron (2006) Hassall (2006) Grieve (2007)

Ireland Australia Germany

Germany Indonesia Australia

10 months 3 months 5 months

8 advanced classroom learners, 12 abroad 33 1 2 at home, 2 abroad

Kinginger (2008)

US

France

1 semester

23

10 2nd year, 10 4th year, 10 graduate students 30

Elicited narratives Oral interview Naturalistic recorded data Role play

Cloze test, matching anaphora Sociolinguistic interview Free DCT Diaries Semi-structued interview, Narrative retalling Language Awareness Interview, Role play

Domains of Communicative Competence

93

the students abroad developed a wider repertoire in this domain. That is, the control group relied heavily on I think whereas the group abroad, although not matching the usage of native speakers, had begun to naturalize their conversational English with markers such as sort of, and stuff, or whatever, and like. Deixis offers particularly intriguing questions for study abroad researchers, as illustrated in projects examining development of abilities related to pronominal reference. Among these are two studies of address form competence, or the ability to say you. While at first glance this ability would seem a relatively straightforward affair, in languages offering options for you (e.g., the T/V system in European languages) this choice can be quite complex. In his description of ethnographic rich points, or loci of conflict presenting opportunities to learn about language and culture, the American ethnographer Agar (1994) offered his own learning of the difference between du (informal) and Sie (formal) in German as a case in point. Agar’s grammar books had been ‘clear as freshly washed crystal’ (1994: 18) in explaining that du is for children, friends, and relatives, and Sie is for everybody else. However, his attempts to follow these rules led to numerous gaffes and misunderstandings, including at least one case in which Agar failed to notice flirtatious behavior on the part of a colleague who switched from Sie to du in addressing him. Observation of the ‘social acrobatics’ (1994: 19) needed to restore harmony to interactions following these blunders convinced Agar that the address system in German is a rich point. The ability to say you in many European languages in fact requires both grammatical and sociocultural competence, since the choice of form determines verb morphology and the choice itself is linked to personal, situational, and social conventions. In describing the address system in French (tu versus vous), Morford (1997) treats these forms as inherently ambiguous double indexicals. On the one hand, they may point to the formality or informality of social settings, or to respect, deference, or solidarity within relationships. In a service encounter, for example, mutual vous would normally be in order, but mutual tu is the convention for informal encounters among age-peer members of the same group, such as university students. On the other hand, address form choice also indexes personal identity, with the choice of vous displaying political conservatism, nostalgia for the past, high social status or professional competence, and the use of tu indicating an egalitarian or left-leaning ethos, lower class origins, or the desire for a youthful appearance. The complex meaning of address forms is such that Agar’s social acrobats have found their way into the linguistics literature, for

94

Language Learning and Study Abroad

example in Dewaele’s work on the ‘sociolinguistic tightrope’ stretched between tu and vous (2004) to be crossed by learners and expert speakers alike. In the early phases, language learners normally perceive address forms as relatively transparent and easy to manipulate (Dewaele and Planchenault 2006). However, as their access to authentic use of the language increases, they observe the sociolinguistic and interactional complexities of the system and judge address form use increasingly difficult. Two studies have focused on the development of address form competence in learners of German (Barron 2006) and French (Kinginger 2008) abroad. Barron’s study involved Irish learners in Germany and native speakers of German in a Free Discourse Completion Task requesting that they write both sides of dialogues involving offers and refusals, and illustrating intimacy, solidarity, or social distance. Learner uses of du versus Sie and related forms at the beginning and end of the sojourn abroad were tallied and compared with the native-speaker data. Barron then computed the percentage of target-like versus non-target-like address forms in the learner data, showing that target-like forms increased from 62.7 to 80.2 percent while non-target-like forms decreased from 37.3 to 19.7 percent. This change is attributed to greater use of appropriately reciprocal address forms and a decrease in ‘non-functional switches’ (2006: 80) especially within informal interaction. At the end of the study abroad period, however, the participants’ address form use retained a ‘strong learner-like quality’ (2006: 85) with non-functional switching still prevalent in formal situations. Barron proposed that these learnerlike features are traceable to the fact that the students rarely frequented formal social contexts, or to ongoing insecurity due to the complexity of the system. She also suggested that gaining declarative knowledge of the address form system is an important aspect of language learning abroad, and that research should focus not only on performance but also on awareness in this domain. My own studies involving participants from the United States in France (Kinginger and Farrell 2004; Kinginger 2008) examined the development of address form competence in two ways: (1) performance in formal and informal spoken role-play tasks, and (2) metapragmatic knowledge as revealed in a task where the participants chose an address form for each of six situations, and explained their choice. Spoken roleplay exercises are far from ideal elicitation devices, since they require a struggle to define the task while also achieving self-regulation in its performance (Kasper 2001). However, in this instance the role-play data provided a snapshot of the group’s significantly greater ability, at

Domains of Communicative Competence

95

the end of the study abroad period, to match address forms to formal and informal interactive contexts. In particular, the data showed a decrease in apparently unmotivated switching, as had been the case for Barron’s participants, and an increase in appropriate choice of the informal tu form to index age-related group solidarity rather than intimacy. Data from the Language Awareness Interview indicated a decline in certainty as the participants came to perceive the broad array of conventions governing address form choice. In other words, by shedding some of their confidence about how to say you, these students gained in linguistic expertise while abroad. These data also revealed that some participants, particularly those with more advanced general proficiency and significant related experience, had begun to appreciate the concepts underlying address form use and the double indexicality of these forms. For example, Benjamin, a student whose case is discussed in Kinginger and Farrell (2004), was a relatively cosmopolitan and highly proficient user of French. He lived in the well-heeled Parisian suburb of Boulogne, in the home of a baron and baroness who invited him for weekends in their country château and for nightly ‘philosophical conversations’ (2004: 35) over dinner. By the end of his stay, Benjamin understood the identity-related implications of variation from conventional address form use, and claimed the right to display his polite demeanor through the use of vous even in situations normally calling for tu. DuFon (2000) characterizes the choice of address forms in Indonesian as a complex process requiring speakers to match their choice not only to the situation, setting, and age, gender, or status of participants but also to ethnicity and regional origin. Students of Indonesian are confronted with the task of learning a broad array of address terms (kin terms and titles, the intimate and familiar pronouns kamu and engkau, and the more formal and distant pronoun Anda) in a process that may span several years. DuFon observed this process for a group of six beginning and intermediate students on a four-month program in East Java, using learner journals and naturalistic tape-recorded data collected in a variety of speech events such as mealtime conversations, leave-taking interactions, or service encounters. The students were exposed to a reduced number of address terms appropriate for the social settings they frequented, but their interactions with hosts offered numerous opportunities to observe the use of these forms. The participants chose various ways to reduce the complexity of their task. Three of the students made minimal advances in their comprehension and

96

Language Learning and Study Abroad

use of address forms. This group chose Anda and used it extensively as if it were the equivalent of you in English, occasionally mixing the formal pronoun with markers of the familiar kamu in the same utterance, thus sending an ambiguous social message. One student opted to avoid address terms and the social hazards accompanying their use. Two students, however, enlarged their repertoire in this area. The intermediate learner Bruce began to experiment with kamu in conversation with age-peers, approximating native-speaker use. Most remarkable, however, is the progress of a student, Charlene, who began the program with no knowledge of Indonesian and developed a repertoire including most of the basic kin terms and pronouns. DuFon’s scrutiny of Charlene’s journal revealed a particular interest in the expression of respect through language and an earnest effort to observe and imitate the practices of her hosts. Development of ability to manipulate deixis in expert ways is also illustrated in two studies focusing on learning how to say we in French using the norms illustrated in Bégaudeau (2006). Contemporary usage includes two variants of the first person plural: the relatively formal nous and the relatively informal inclusive on, with evolution toward preference for on. Both are grammatically correct, but language learners tend to favor nous, based on their exposure to pedagogical materials (Wolf 2006) whereas corpus-based studies show that expert language users prefer on and almost never use nous in informal speech (e.g., Coveney 2000). Use of nous in informal interactions therefore lends a bookish and slightly anachronistic cast to the speech of learners. The cross-sectional research of Sax (2001) compared the use of on among three groups of US-based learners. The lowest proportion (4 percent) was found among second-year, language students with no experience abroad, an intermediate level (57 percent) obtained among fourth-year students of whom half had studied in France, and graduate students with one to four years of time in Francophone environments used on in 81 percent of relevant contexts. Similar results were obtained in another project involving Irish study abroad participants (Lemée 2002), suggesting that study abroad represents an opportunity to observe and appropriate non-pedagogical usage, ultimately helping learners to sound ‘more “French” ’ (Towell, Hawkins, and Bazergui 1996: 112). Studies examining learners’ awareness of discourse genre, although common in the broader literature on second-language writing (e.g., Swales 2004), are extremely limited in the research on language learning in study abroad. Scholars in language education (e.g., Kern 2003; Taillefer 2005) point out that students abroad encounter new textual

Domains of Communicative Competence

97

genres, and insist that one of the main challenges to learning abroad is the development of cross-cultural academic literacy. Yet, competence in the area of generic structure is most often viewed through the lens of grammar (e.g., Collentine 2004) and is rarely framed as a discourse-level phenomenon. One study (Maeno 1995) interrogated the development of oral narrative skills in Anglophone learners of Japanese, showing that students with advanced proficiency had begun to appropriate some features of narrative style, but the development of academic genre awareness is clearly a neglected area. Questions about conversational structure border on a number of other areas, including communication strategies, in the case of repair, and fluency, in the case of formulaic sequences or conversational gambits. Four studies are directly concerned with opening and closing conversational interactions. Among the earliest of these is Lafford’s (1995) investigation of the communication strategies adopted by learners of Spanish at home (AH) and abroad (SA) in the context of a role-play exercise. The SA participants spent a semester in Mexico or Spain, while the AH group pursued second-year level coursework. The role play, part of the ACTFL Oral Proficiency Interview for the Intermediate level, required the students to imagine asking directions to a museum from a hotel clerk and therefore to initiate a conversation, gather information, and take leave in an appropriate manner. Of particular interest here is Lafford’s analysis of channel openings and closings (Goffman 1976), and of fillers, connectors, and backchannel signals used to maintain the cohesion of conversations. Overall, the SA participants had considerably more to say in this testing situation than did the classroom students, allowing them to display varied repertoires in each of the areas under consideration. Only three of the classroom learners provided suitable conversation openings, whereas all of the SA participants did so, calling attention to their pleas for assistance with expressions such as con permiso or perdón. Similarly, in closing the conversations, the SA group typically combined several expressions (vale, muy bien, gracias, or muchas gracias, muy amable), while the classroom group tended to offer a bare-bones ‘thank you’ (gracias) or no closing at all. Only the most proficient members of the SA group had begun using Spanish fillers, such as pues, or entonces. The most frequent connector in the data was y (‘and’) but the SA group also displayed a number of other discourse adhesives such as después or también. Similarly, to signal their attention to the ongoing conversation, the students with experience abroad had appropriated a variety of backchannel forms like claro and vale for the participants in Spain, where the latter term

98

Language Learning and Study Abroad

is extremely common. As noted in a general critique of studies comparing populations of students at home and abroad (Rees and Klapper 2008), this study is to some extent a comparison of apples and oranges, since the experiences under consideration are clearly very different one from the other, and the language-related motives of SA participants no doubt vary from those of students who stay at home. Nonetheless, Lafford’s project offered empirical support for the widespread belief that study abroad nurtures conversational competence and suggested that students abroad may participate in a range of socializing interactions serving to broaden their repertoire of socially situated language forms. This claim finds further support in Marriott’s (1995) analysis of participation in interviews by Australian secondary exchange students in Japan. Following a year abroad, these students displayed the ability to use polite, routinized expressions for the opening and closing sequences of the interviews, and to do so with apparent ease. As part of the pre- and post-Language Awareness Interview in Kinginger (2008), the participants (US-based students of French) were asked to match a series of leave-taking formulae with an array of briefly described situations (from Ball 1997) and to comment on their choices (e.g., Host Mother speaking to a guest student as s/he retires for the evening, Orator to deceased at a funeral, Leaving a message on a close friend’s answering machine, Shop owner to departing customer). A correspondence analysis revealed that these students, as a group, developed increased sensitivity to formality versus informality in the various situations and ways to say good-bye. At the beginning of their sojourn, many of the participants had yet to appreciate the formality and finality expressed by adieu or the restricted use of informal expressions such as ciao or gros bisous (‘fat kisses’). There were several students whose Host Mothers seemed bizarrely unwilling to give up their guest’s company for the evening (à tout à l’heure, or ‘see you shortly’) and one whose funeral Orator would bid the departed farewell with a rather flippant ciao. By the end, however, most if not all of the students had discovered the meaning of the formulae and learned to assign them to appropriate contexts, especially in the situations they had routinely encountered while abroad. The supposition underlying my research was that students abroad would observe leave-taking activity in a variety of contexts and would participate in encounters where their choice of leave-taking formulae would be of social consequence. This process is the explicit topic of

Domains of Communicative Competence

99

Hassall’s (2006) diary study recounting his efforts to develop competence in leave-taking while studying in Indonesia. Hassall notes that the informal learning environment, even during a short sojourn abroad, is a ‘powerful stimulus for the acquisition of pragmatic knowledge’ (2006: 52). The story unfolding in his diary is one in which the protagonist observes leave-taking formulae in various scenes (on the street, at home, or in television dramas), tests his new knowledge in his own interpersonal encounters, evaluates his level of satisfaction, and repeats the process. Initially he had but one leave-taking formula (permisi) in his repertoire, and required a ‘painful awakening encounter’ (2006: 37) involving failure to acknowledge his interlocutor’s desire to leave, in order to recognize another (the dulu statement). Having learned to take leave via this second formula, he then over-generalized its use, leading to yet another awkward interaction: As I was trudging home on the final stretch after a long hot walk, I was called over to chat by two women in the yard of a house nearby . . . We chatted amiably enough, but I suddenly got the impression that I’d overstayed my welcome – one of them seemed to be casting around rather awkwardly for further questions to ask me. So I rather hastily took my leave, with a dulu statement: ‘Pulang dulu ya’ ‘I’m going home for now, okay’ (literally, ‘Go-home for-now, yes’). I said it a bit tensely and unsmilingly and it felt a bit abrupt as I said it. As I then turned to go, one of them said softly, in English, ‘Excuse me’. I turned around in puzzlement. She then repeated in Indonesian: ‘Permisi’, and laughed. So her ‘excuse me’ had been a gentle correction; a supplying of what I’d omitted to say. (Hassall 2006: 39) Hassall ‘took this correction to heart’ (2006: 39) and continued to cycle through hypotheses about the difference between permisi and the dulu statement, about the role of pre-closing routines in Indonesian, and about the advisability of pre-planning his leave-takings at the expense of attending to subtle interpersonal cues in the conversation itself. Although no empirically based descriptions of leave-taking in Indonesian existed to provide guidance, eventually Hassall understood that permisi conveys a formal and slightly deferential tone, whereas the dulu statement does not. At the end of his sojourn he was aware not only that his conversational repertoire had been refined through continuous social interaction, but also that a great deal remained to be learned.

100

Language Learning and Study Abroad

Situating his work in relation to Bialystok’s theory of pragmatic development among adult L2 learners (1993), Hassall challenged the notion that pragmatic knowledge itself is relatively unproblematic, and that the crucial process is one of gaining control over existing knowledge. Instead, acquiring knowledge itself is a ‘major task’ (2006: 53) for the language learner attempting to develop discourse competence through informal social interaction. The studies devoted to discourse competence in study abroad are few but varied in method and scope. Studies of discourse cohesion include text-based assessment (Fraser 2002) and reviews of data collected via interviews (Lafford 1995; Marriott 1995; Grieve 2007). Deixis has been examined in the area of person – reference in naturalistic data (DuFon 2000) and in tasks designed to elicit both written (Barron 2006) and spoken data (Sax 2001; Lemée 2002; Kinginger 2008). Abilities related to conversational structure, particularly leave-taking, have been examined in the context of a general Language Awareness Interview (Kinginger 2008) and in much greater depth through focused diary study (Hassall 2006). Hassall’s commentary following his quest for information about how to say ‘good-bye’ in Indonesian points to one possible reason for the dearth of studies in this area. No empirical description of leavetaking in Indonesian was available to him, thus in an important sense Hassall could not know at the outset exactly what he was looking for, what forms constituted leave-taking in the first place and what social, situational, or identity-related conventions would prove relevant for choosing among them. In this case, the learner was required to assume the role of ethnographer in order to fill in the crucial information left wanting in communicative pedagogy (Leung 2005). One leaves this text wondering if learners of other, more and less commonly taught languages are similarly bereft of guidance and obliged to cast about in search of generic or deixis-related features, conversational conventions, cohesive devices, and the concepts underlying these aspects of language use. At the same time, all of the studies reviewed in this section portray study abroad as a valuable, if imperfect environment for developing discourse competence. The reader may wonder about the contribution of this development to overall impressions of increased fluency and naturalness in language use following a sojourn abroad. If, despite decades of progressive thinking about the value of discourse-based pedagogies (e.g., Scollon and Scollon 1995; McCarthy 2001), these features remain neglected in textbooks and classrooms, then study abroad in fact represents a uniquely well-suited opportunity to appreciate their use.

Domains of Communicative Competence

101

Sociocultural competence For Celce-Murcia, Dörnyei, and Thurell (1995: 23), sociocultural competence is the ‘speaker’s knowledge of how to express messages appropriately within the overall social and cultural context of communication, in accordance with the pragmatic factors related to variation in language use.’ Included under this category are four sub-components, beginning with social contextual factors, or awareness of the ways in which language use is tailored to social situations and to participant variables such as age, status, and equitable or other relations of power. Stylistic appropriateness, the second category, includes knowledge of norms for politeness and of formal or informal registers. Into the broad category of cultural factors, these authors place awareness of regional or other dialects, intercultural communication, and knowledge of the history and major values, worldviews, and practices of communities where the language in question is used. A final sub-component, non-verbal communicative factors, includes kinesics, gesture, affective markers such as facial expressions and eye contact, proxemics (use of space), haptics (touching), and the meanings associated with silence (e.g., the stony silence or the pregnant pause). While knowledge is the basis for making explicit choices, in this domain as elsewhere language learners may or may not opt to conform to the norms presented to them. In the study abroad literature on language learning, the domain of sociocultural competence is represented in qualitative research on student experience (see Chapter 4) and socialization (see Chapter 5), but there are few studies attempting to evaluate cultural or social contextual awareness per se as it relates to language. Knowledge of language varieties has been identified as a need of study abroad participants (e.g., Schild 1981; Iino 2006), who may be quite unaware that expert users do not limit their repertoires to the decorous forms found in textbooks, or that the language use they observe may be marked for gender, age, region, or social class, but to date there are no investigations directly addressing outcomes in this arena. Gesture, proxemics, haptics, and ‘body language’ in general are often highlighted as key features of communicative ability in the popular media, but only gesture has attracted the attention of researchers working with study abroad participants. Jungheim (2006) compared how Japanese native speakers and study abroad participants in Japan interpret and perform a culturally specific refusal gesture called the Hand Fan. In this gesture, the speaker moves an open palm back and forth in front of the face, as if fanning a flame, and the hand takes over the

102

Language Learning and Study Abroad

role of the lateral head shake. This gesture can serve to disambiguate accompanying spoken language, often the polite kekkou desu (That’s all right) and emphasize that a speech act is in fact a refusal. Jungheim in fact reports that in his own experience as a second-language user of Japanese, his attempts to refuse offers without bluntly saying ‘no’ (iie) or using a refusal gesture had resulted in interlocutors’ trying to give him things he did not want. Jungheim’s findings, based on participants’ interpretations of and responses to soundless video clips, show that while the native speakers clearly recognized the conventional meaning of this gesture, even after six months of study in Japan the students had not acquired it for receptive purposes and could not interpret its contribution to the speech act of refusal. It is unclear, however, as the author himself notes, that the learners had enjoyed a linguistic immersion experience involving significant social interaction with Japanese hosts. By contrast, a study by Kida (2005) examined spontaneous production of gestures by Japanese students residing temporarily in France. Kida documented the appropriation of discourse and contentoriented gestures specific to French culture, particularly for individuals with high proficiency and a strong integrative orientation to life in France. Taken together, these studies suggest that students abroad may begin to develop a repertoire of gestures but that this process requires a significant investment of time in interaction with expert language users. The majority of studies related to sociocultural issues concern the category of stylistic appropriateness, often termed ‘sociolinguistic competence,’ and in particular, students’ awareness and use of variable features marking formal versus informal registers (see Table 3.4). Most SLA research deals with linguistic features believed to be categorical, that is, not subject to variation in native-speaker usage (Howard 2004). Sociolinguistic competence, however, includes the ability to choose from different realizations of similar meaning, as would be the case when choosing du versus Sie to say you in German, or in the case of negation with or without ne in French: Je ne viens pas vs. Je viens pas. I neg come-pres neg vs I come-pres neg. ‘I am not coming’ (Howard 2004: 144) Research on the learning of French dominates this sub-area, no doubt in part because the history of standardization and linguistic purism in

Domains of Communicative Competence

103

Table 3.4 Studies of sociolinguistic competence Study

Origin

Destination Sojourn length

Participants

Focus/data

Regan (1995; 1997; 2004) Dewaele and Regan (2001)

Ireland

France

1 year

5–6

Belgium, Ireland

France

Varied

29 Belgian, 6 Irish

Dewaele and Regan (2002)

Belgium

France

Varied

27

Howard, Lemée, and Regan (2006) Kinginger (2008)

Ireland

France

1 year

15

US

France

1 semester

23

Japan

1 year

5

‘ne’/Sociolinguistic interviews Colloquial words/ Conversations or sociolinguistic interviews ‘ne’/Informal and formal (structured) interviews /l/deletion /Sociolinguistic interviews Colloquial language/Language Awareness Interview Style shifts/OPI

Iwasaki (in press) US

France has yielded a remarkable gap between the conventions of written language taught in school and contemporary spoken usage (Gadet 1997). This research includes work on awareness and use of colloquial forms, the projects on deixis through address forms previously reviewed, consideration of ne-deletion in negation as illustrated above, and one investigations of a phonological variant. As summarized by Howard (2004: 4), this research shows ‘very evident underuse of informal sociolinguistic markers’ by students whose exposure to French is limited to the foreign language classroom, along with highly beneficial effects of the informal interactions characteristic of study abroad. The research also suggests that students can develop awareness of sociolinguistic variation resembling that of native speakers, but that the status of learner may convey with it an attitude of caution and reluctance when it comes to using stigmatized vernacular language. Dewaele and Regan (2001) scrutinized the use of colloquial words in a corpus of learner French collected in interviews provided by 29 students at a monolingual Dutch university and six Irish participants in an ERASMUS-sponsored year in France. The presence of colloquial lexemes increased only modestly in comparison with data from native speakers, and this increase was associated with time spent speaking French. That is, learners who had spent time abroad displayed a broader repertoire

104

Language Learning and Study Abroad

of colloquial words. One of the most striking findings of this study is that the discourse markers hein and ben represented a large proportion of the colloquial words in the corpus (31 percent), suggesting to the authors that students may enter vernacular speech styles through a ‘back door’ (2001: 61) involving words that are not associated with particular second-language concepts. Thus, a period of study abroad can enlarge students’ repertoires in the area of informal language use, but the lexical nuances of these words, and their conceptual representation in implicit memory, will only emerge at very advanced stages of learning. My own project (Kinginger 2008) examined awareness of colloquial language among US-based study abroad participants in France. These students were presented with a list of colloquial words (adapted from Dewaele and Regan 2001) and a series of vernacular phrases characterized by syntactic and lexical features of ‘ordinary French’ (Gadet 1997). They were asked to provide translations of the words and phrases, and to comment on the types of situations where their use might be appropriate. The rationale for this approach stemmed from the general observation that students at lower levels of proficiency tend to avoid vernacular forms, perhaps even in cases where these forms are known. Comprehension of colloquial language, and the ability to identify its association with informal situations or presentations of self, was hypothesized to be a potential outcome of study abroad and an important phase in the development of advanced proficiency. In fact, upon completion of their programs, the students as a group had developed significantly greater awareness of colloquial French, and the task prompted many of them to cite specific instances where informal language use had been observed and the semantic nuances of specific terms highlighted in interactions with French-speaking hosts. Among the more commonly cited studies of sociolinguistic development in study abroad are the investigations by Regan (1995; 1997), into the acquisition of variable ne deletion in French, taken as a barometer of overall acquisition of speech norms. In French, the negative is formed by a particle on either side of the verb, a proclitic ne and a marker of general negation such as pas (plain negation), plus (‘no longer’), or jamais (‘never’). In formal contexts, both particle and marker are retained, but particularly in spoken French ne may be deleted, with its presence or absence serving as a ‘symbolic sociolinguistic indicator’ (Regan 1995: 248) of formality, power, and solidarity. The use of ne can convey a sense of formality, whereas its deletion contributes to casual speaking styles. A command of this feature is necessary in order to manipulate different registers in French.

Domains of Communicative Competence

105

Regan (1995) obtained speech samples through interviews with six Irish students of French before and after their year in France to determine whether or not they would develop awareness of ne-deletion and ability to manipulate it in native-like ways. Regan’s variationist analysis showed a dramatic increase in ne-deletion following the stay abroad, and a tendency for the students to retain the ne in more formal speech, just as native speakers do. However, in comparison with native speakers’ performance, the students tended to over-generalize the deletion of ne in formal contexts. Regan hypothesized that further exposure to native norms would prompt refinement of the students’ performance in this area. In a follow-up study, Regan (1997) demonstrated that the deletion of ne was concentrated in lexicalized phrases such as Je sais pas (‘I don’t know’), c’est pas (‘it’s not’), or il y a pas (‘there are no’) and that, within these phrases, the students deleted ne at a higher rate than native speakers. Regan (1997) argued that the use of such phrases permits students to signal their integration into the French-speaking community. Not only does the study abroad experience allow such display of integration, it also appears to consolidate sociolinguistic skills. Regan (2004) investigated ne-deletion in the speech of the same Irish participants in her earlier studies one year after their return from France, and found their performance essentially unchanged. An additional study by Dewaele and Regan (2002), this time involving Dutch-speaking learners of French, showed that classroom learning had no influence on students’ ability to convey informality through ne-deletion. Only those students who had enjoyed prolonged contact with a Francophone community developed sensitivity to this issue, a finding leading the authors to conclude that residence abroad, or daily contact with speakers of French, are clearly the most effective ways to stimulate acquisition of sociolinguistic competence. A similar conclusion was reached in a study extending this variationist approach to the acquisition of a phonological feature of spoken French, the deletion of /l/ (Howard, Lemée, and Regan 2006). The nonrealization of /l/ is in some ways analogous to the pronunciation of words ending in –ing in informal English (moving vs. movin’), that is, it is a marker of casual speech, attested among speakers of continental French at particularly high levels for the subject pronouns il and elle. Data for the study included sociolinguistic interviews with 19 students of French at an Irish university, of whom 15 had spent a year in France. When tokens of /l/ were extracted from the data and coded, it was found that the study abroad participants deleted /l/ at a much higher

106

Language Learning and Study Abroad

rate (33 percent) than did the students remaining at home (6 percent), although they did not reach the much higher rates of /l/ deletion described in studies of native speakers. An analysis of the contexts in which the students with study abroad experience deleted /l/ (a high rate for impersonal il versus a lower rate for the subject pronoun elle) showed patterns exactly in line with native-speaker usage. The authors conclude that time abroad appears to be crucial for instructed learners’ development of sociolinguistic competence. Iwasaki (in press) explored the ability of American learners of Japanese to manipulate plain and polite styles. In Japanese, there are no neutral forms of politeness, and speakers must choose plain or polite forms constantly, in recognition of social and situational contexts. In simplistic terms, plain forms such as iku (‘I will go’) are used in informal situations or to index closeness, while polite forms such as iku-masu (‘I will go’) index social or psychological distance or formality. Difficulty in learning appropriate style is widely attested in the second-language literature, with many studies documenting overuse of the plain forms. However, many purely quantitative studies fail to explore the meaning of these forms and the fact that their use can be mixed, within the same interaction. For expert users of Japanese, they are a resource for the creation of interactional contexts: for example, a teacher may use style shifts to express friendliness toward a student, and that student may switch to the plain form to express emotion, despite status differences. The five male students in Iwasaki’s study followed different developmental routes toward politeness in Japanese, but they all developed awareness of the social meaning of style and some ability to recruit style-shifting to reach their interactional goals. The work reviewed in this section overlaps with research on discourse competence, in that this research is concerned with the linguistic realization of social and stylistic issues such as formality versus informality in speech, and the ways in which students abroad gradually discover how particular forms map onto canonical social meanings in their host communities. For proponents of study abroad, this research confirms intuitions suggesting that certain aspects of language are relatively inaccessible to classroom learners, and provides support for the claim that a stay abroad offers unique affordances for developing students’ communicative repertoires, particularly for informal interaction. Students abroad are shown to enhance their awareness and use of colloquial lexis, and to approximate native usage of key sociolinguistic features of French. On the other hand, the scant research on gesture indicates that a stay abroad only initiates a much more complex and long-term