Language Learners in Study Abroad Contexts

Margaret A. DuFon, Eton E. Churchill, Margaret A. DuFon, Eton E. Churchill

• Publisher: Multilingual Matters
• Year: 2006
• ISBN:
  9781847699503
• Topic: Languages & Linguistics
• Subtopic: Literacy
• Index: Languages & Linguistics

Examining the overseas experience of language learners in diverse contexts through a variety of theoretical and methodological approaches, studies in this volume look at the acquisition of language use, socialization processes, learner motivation, identity and learning strategies. In this way, the volume offers a privileged window into learner experiences abroad while addressing current concerns central to second language acquisition.

Chapter 1

Evolving Threads in Study Abroad Research

ETON CHURCHILL and MARGARET A. DUFON

Introduction

For the second language acquisition (SLA) researcher, there are perhaps few contexts as potentially rich and complex as study abroad. On the one hand, concentrated time enjoyed by learners in the host context would appear to facilitate significant linguistic gains. On the other hand, pre-departure individual differences interact in complex ways and are affected by the study abroad context, itself conditioned by cultural norms and factors related to program design. Adding to these dynamics, patterns of acquisition of skills and specific forms are far from linear and have proven difficult to consistently record based on pre–post tests. Given these interactions, it is not surprising that within-group differences are just as frequently reported as between-group differences and that these findings are supported by accounts in qualitative studies (see Coleman, 1995, 1996, 1997, 1998; Freed, 1995a, 1998 for overviews; Huebner 1995b, 1998; Pellegrino, 1998 for discussions of research methods; and Regan, 1998; Pellegrino, 1998 respectively for reviews concentrating on sociolinguistics and conditions of learning experienced by the SA student).

In the present chapter, drawing primarily from studies published subsequent to the reviews cited above, we will focus on what is learned, how individual differences interact with proficiency gains, and how cultural and program related factors shape opportunities for contact with native speakers. We begin with the literature on gains in linguistic skills and then turn our attention to the area of pragmatics. We then address the research on individual differences concentrating on motivation, learning strategies and anxiety. Finally, we focus on the dynamics of language learning highlighting learner involvement in the host context, and on how engagement with native speakers can be enhanced or mitigated by cultural norms and program design. Thus, we hope to delineate what is known regarding the formal aspects of acquisition in the SA context while drawing reader attention to the social conditions in which this learning takes place. Concurrently, we intend to introduce potential directions for future research.

Acquisition of Linguistic Abilities

Reflecting the predominant interest in SLA, the bulk of study abroad research has focused on what is acquired (e.g. forms and skills) by the learner. The majority of these studies have concentrated on gains in specific skills in individual programs. However, returning to an earlier approach taken by DeKeyser (1986) and Lafford (1995), the research agenda has recently expanded to include studies comparing gains in SA contexts with those attained in domestic programs (Bradley, 2003; Collentine, 2004; Dewey, 2004a, 2004b; Díaz-Campos, 2004; Freed et al., 2004; Hoffman-Hicks, 2000; Howard, 2001; Matsumura, 2001; Rodriguez, 2002; Segalowitz & Freed, 2004; Stevens, 2001; Torres, 2003). In the discussion below, we outline the acquisition of linguistic abilities in terms of literacy, listening, speaking (oral proficiency, fluency and pronunciation) and grammar. In each section, we first address the results from single program studies and then review the comparative studies.

Literacy

Relatively few studies of language learners within SA contexts have focused on the acquisition of literacy skills, reflecting a bias in programs themselves and in expectations for gains in the aural/oral skills. To wit, Allen (2002) found that students preferred to obtain information by talking to others (either NSs of their L1 or the L2) rather than by reading or listening to the radio. Although the emphasis has been on oral rather than written language, evidence from both quantitative and qualitative studies suggests that the area of literacy merits more attention.

Researchers have taken different methodological approaches depending on their view of literacy. Some, such as Fraser (2002), who compared the ability of learners to match anaphora and cataphora to their referents, have taken a skills-based approach. Meanwhile, Dewey (2004a, 2004b) used both quantitative (vocabulary tests) and qualitative measures to examine increased word knowledge, improved comprehension, reading processes, habits, strategies and learner perceptions. Elsewhere, Kline (1998) investigated how learners’ literacy-related identities affected literacy behaviors and text-related interaction over time.

Despite the varied methodologies employed, the consistent finding is that SA learners develop their reading skills (Dewey, 2004a, 2004b; Fraser, 2002; Waldbaum, 1997) and literacy (Kline, 1998). For example, Dewey (2004a) reported that his American learners of Japanese became stronger in vocabulary recognition and text comprehension, and developed more confidence in their reading ability. This seemed to be due to the frequency and range of experiences in interacting with text in a naturalistic environment, a claim qualitatively supported by other studies (see Churchill, 2003a: 286–289; Huebner, 1995b: 185; Wilkinson, 1995). Adding further evidence to Dewey’s observations on reader confidence, Kline reported that over the course of a year in France, students gained greater independence in reading texts, learned to look for a ‘deeper meaning’ and began to discuss reading material with their host families. Meanwhile in terms of specific reading skills, learners of German typically advanced in their ability to match anaphora and cataphora to their referents, though the amount of gain depended on pre-program levels and effort expended.

Of these studies, Dewey (2004a, 2004b) is the only one to have compared SA students with those studying at home. SA learners in Japan gained greater confidence than the intensive at-home (AH) group; yet despite increased confidence, there were no significant differences between groups in gains in vocabulary or ability to recall the reading passage. However, there were differences in the way the two groups approached the task of reading. The SA group monitored their understanding of what they were reading more than the AH group, while the AH group more often responded affectively to the text. Dewey concludes that because of the nature of the AH groups’ interaction with their teachers (e.g. more comprehension checks and requests for affective responses), they were able to develop their reading skills as much as their overseas counterparts.

Given the scant attention that literacy has received in SA research and the positive findings summarized above, this would appear to be a rich area for further investigation. In particular, it would be interesting to document the extent to which oral interaction in the host context either supports literacy activities (e.g. reading with younger siblings, sharing of the daily news), or indirectly facilitates reading through increased lexical breadth and grammatical development.

Listening

As with literacy, research on of the acquisition of listening skills in the SA literature has also been scant. However, several studies (Allen, 2002; Kinginger & Whitworth, 2005; Waldbaum, 1997) have found that SA students make improvements in this area, and Tanaka and Ellis (2003) reported minor gains on the listening component of the TOEFL taken by Japanese learners studying in the United States. Investigating the development of listening comprehension during a six-week summer program in France, Allen (2002) reported a significant (p , 0.001) improvement based on a 14-item listening skills test. In a self-assessment questionnaire on listening and speaking tasks, Allen’s learners claimed they made significant gains in listening, but felt they had more success in mastering complicated speaking tasks. Interestingly, at the time of the pre-test, three of the 25 learners reported that they could easily perform certain listening tasks, but on the post-test, they reported that they could not easily perform at least one of those same tasks. Allen attributed this change to a greater awareness of the challenges of interacting with native speakers.

Speaking

Most studies of language acquisition during study abroad have focused on speaking. Below, we begin with a discussion of proficiency and fluency, then turn to pronunciation.

Oral proficiency and fluency

Investigations into gains in oral proficiency have relied heavily on the Oral Proficiency Interview (OPI) or similar interviews. Studies of gains in oral proficiency made by SA learners have revealed that they improve their proficiency over the course of a semester or more abroad (Isabelli-García, 2003; Segalowitz & Freed, 2004). Even learners who go abroad for only a few weeks have been found to improve their oral proficiency, although their gains are not necessarily enough to advance to a higher level on the ACTFL scale (Simôes, 1996).

Fluency has also typically been measured using the OPI or another interview format. The learners’ speech in the interview has then been examined with respect to various temporal and hesitation phenomena. Research using pre-post test designs has revealed that SA learners improve their fluency as a result of an increase in rate and length of fluent runs (Segalowitz & Freed, 2004) and reductions in the number of pauses, fillers, dysfluencies and struggles (Freed et al., 2004; Isabelli-García, 2003; Segalowitz & Freed, 2004; Simôes, 1996; Woodman, 1999). However, it was noted in several of these studies (Segalowitz & Freed, 2004; Simôes, 1996) that not all learners in these SA groups improved in their fluency; rather individual learner differences accounted for considerable variation.

Several studies investigating fluency and proficiency (Isabelli-García, 2003) and grammatical features (Collentine, 2004; Lopez Ortega, 2003; Schell, 2001, Torres, 2003) reported improvements in narrative ability. Collentine (2004), examining language acquisition in Spain, found that learners improved their narrative abilities. By the end of the program, students were able to relate more narrative events and do so with fewer words. In addition, Isabelli-García (2003) reported that two of three L2 learners of Argentine Spanish, unable to produce complex narratives at the beginning of the program, could do so by the end.

In comparative studies on fluency and proficiency, results have been mixed, depending largely on the type of program at home and initial proficiency level. Bradley (2003) compared two groups of Intermediate and Advanced learners of German – an at-home (AH) group in the United States and a SA group – with respect to formulaic speech using a simulated oral proficiency interview. The two groups performed similarly in terms of rate of speech and number of fillers. However, the SA group outperformed the AH group in terms of total number of words and non-filler formulas. Qualitatively, the SA group employed a wider repertoire of fillers for a greater range of functions. Furthermore, in linguistically challenging situations, the SA students were better able to benefit from formulaic language, appearing linguistically more competent than they actually were.

However, Freed et al. (2004) argue that learners might be able to gain just as much – if not more – from an intensive language program at home. Comparing gains in oral fluency and oral proficiency in learners of French in three learning contexts: (1) at home (AH) in formal language classrooms, (2) at home in an intensive all French language summer immersion (IM) program, and (3) abroad in France, Freed and her associates found that gains were related to both learning context and time spent speaking French outside of class. The AH group made no significant progress in any area, while the SA group made significant gains only in speech fluidity. However, the gains of the IM group were greater, attributed in part to the fact that IM students both spoke and wrote significantly more than the other two groups. As a result, Freed and her associates concluded that it is not the learning context per se which determines language gain, but rather the nature and intensity of contact within that context. Segalowitz and Freed (2004) came to similar conclusions based on their investigation of learners of Spanish. Generally, greater gains were made by the SA students both in fluency and in oral proficiency (as measured by the OPI) than the AH group. However, the SA advantage could not be accounted for by out-of-class contact with NSs of the target language alone. At best, this relationship was weak and indirect. Rather, the learners’ initial language abilities played a role in influencing the nature and intensity of out-of-class contact they engaged in, and their ability to quickly and efficiently connect words to meaning affected gains in oral performance.

From these studies on fluency and proficiency, one may posit that the SA experience leads to significant gains in both areas and that these gains tend to be larger than those found in most traditional at-home programs. However, this work also illustrates the importance of the quality and quantity of language use and the fact that initial learner proficiency may be a variable that plays a role in shaping intensity and quality of contact with NSs.

Pronunciation

Recent studies that have examined the effect of SA on the acquisition of pronunciation (Díaz-Campos, 2004; Simôes, 1996; Stevens, 2001) have focused on Spanish pronunciation by native speakers of English. In these studies, the SA learners were able to make changes toward the target norm in terms of vowel quality (Simôes, 1996), tongue placement in word final laterals (Díaz-Campos, 2004), reduced voice onset time for voiceless stops (Díaz-Campos, 2004; Stevens, 2001), and linking between words (Simôes, 1996).

However, there were some inconsistencies in the findings. The subjects in Stevens’ (2001) study improved their pronunciation of voiced intervocalic fricatives while those in the Díaz-Campos’ (2004) study did not. In Stevens’ study, the two groups of learners in Spain improved in their ability to produce the tap and the trill from pre-test to post-test, however these differences were not significant. Simôes (1996) found his learners actually moved away from the target, overgeneralizing the trilled r in contexts where a single tap in Spanish was correct. Problems were also noted with prosodic features. Simôes found that all his students continued to exhibit American English rhythm and intonation patterns in Spanish, though some students indicated awareness of this problem by modulating their intensity (but not their pitch) in order to address this issue.

Of these studies, two (Díaz-Campos, 2004; Stevens, 2001) compared SA learners with those in domestic learning contexts. Stevens (2001) compared the acquisition of Spanish phonology of three groups of learners: one at-home (AH) group in the United States and two SA groups in Madrid, Spain (one summer study group and one semester study group). An acoustic analysis of pre- and post-test phonetic data revealed that the SA groups made significantly greater progress overall than the AH group in acquiring more target-like pronunciation. Both the SA and AH groups improved their pronunciation of sounds that were more similar to English, but only the SA groups demonstrated significant gains in producing phonemes that were less similar to those in English.

Unlike Stevens (2001), Díaz-Campos (2004) found no advantage for the SA group. The discrepancy in these conclusions might be attributable to differences at the outset of the study. In Steven’s study, the AH group had approximately two years of prior Spanish instruction, the summer group three years and the semester group four years. Díaz-Campos does not mention the proficiency level of his students; however, the number of years of prior language instruction was found to affect phono-logical gain in his study. Thus, Stevens’ SA learners may have had an advantage, not only because of the context they were in, but also because of their additional years of formal language instruction.

Factors that might account for differences in pronunciation performance between and within learner groups include not only the learning context, but also the length of time abroad, years of formal language instruction (Díaz-Campos, 2004), proficiency level at the outset of the program (Díaz-Campos, 2004; Simôes, 1996; Stevens, 2001), time spent using the target language prior to and during the SA period or opportunities for input (Díaz-Campos, 2004; Stevens, 2001) and gender (i.e. females using a more careful pronunciation than males) (Díaz-Campos, 2004). Thus, as we concluded from the studies on oral proficiency and fluency, it is not just learning context, but individual learner factors and the nature and intensity of contact with target language speakers which combine to shape linguistic development.

Grammar

Much of the work on grammar acquisition has been done on learners of Spanish or French. Several methods have been used to collect grammatical data to examine the acquisition of features such as subject expression (null subject vs. pronoun vs. full NP), clitics, tense and aspect. Methods used to date include analysis of oral proficiency interview data (Collentine, 2004; Howard, 2001; Isabelli, 2002), written language samples from narrative essays (Schell, 2001), learners’ performance on a worksheet of grammatical exercises (Schell, 2001) and grammaticality judgment tasks (Isabelli, 2002).

In these studies, the consistent finding is that learners move toward target norms, but sometimes their paths of development vary and they do not acquire native proficiency during a semester or even a year abroad. In her examination of the null subject parameter, Isabelli (2002) found that while L2 Spanish learners improved their abilities on all three syntactic properties of the null subject parameter, that-trace violations particularly remained a problem. She concluded that the parameter was not reset. Lopez Ortega (2003) found that whether or not the subject was overtly expressed depended upon a range of discourse conditions. Not only did the traditional aspects of contrast, emphasis, and disambiguation affect subject expression but so did factors such as the speaker’s identity and topic involvement, semantic features of the referents, inter-linguistic narrative structures, type of verb, conjunctions and adverbials. Also illustrating the complex dynamics of grammar acquisition, Torres (2003) investigated the acquisition of Spanish clitics (e.g. direct and indirect object pronouns, reflexives) by 10 novice-high to advanced-low SA learners of Spanish over the course of one semester. The lower proficiency learners rarely used clitics and preferred zero marking. In contrast, the intermediate learners had begun to use clitics, but their contexts of use were quite restricted when compared with NS use due to learner overuse of full NPs as objects and obliques, and limited use and variety of transitive verbs. Finally, in Schell’s (2001) study on the acquisition of Spanish aspect (preterit vs. imperfect) in a nine-month SA program in Spain, all five intermediate to advanced learners demonstrated improved accuracy in morphology and grammatical aspect, revealing that they had basically acquired the preterit–imperfect distinction. While initially the less advanced learners displayed a preference for the preterit over the imperfect, by the end no learners displayed preference for one aspect over another. All subjects were able to identify and correctly mark for aspect habitual events, but variation existed for future/intention events, and events in progress. Moreover, the paths were not quite identical across learners.

Studies to date that compare learning at home and abroad have found little to indicate that the SA context is more advantageous for the acquisition of grammar. Comparing language learning prior to, during, and following study abroad, Longcope (2003) reported that learners of English made gains in fluency but not necessarily in grammatical accuracy or syntactic complexity during their time abroad. Torres (2003) found that the SA context did not seem to be more advantageous than the AH context as far as the acquisition of clitics was concerned; however, the SA learners had an advantage over the AH group in terms of discourse and pragmatics, and were able to use language more effectively. Investigating the performance of two groups on oral proficiency interviews, Collentine (2004) concluded that the AH group in fact demonstrated greater development on discrete grammatical and l...