한국영어학회편집위원회 위원장황종배 ( 건국대 ) 부위원장정채관 ( 한국교육과정평가원 ) 제 1 분과책임편집위원문안나 ( 인하대 ) 위원 Thomas Cable(U of

Transcription

1

2 ISSN Volume 18, Number 1 Spring 2018

3 한국영어학회편집위원회 위원장황종배 ( 건국대 ) jongbai@konkuk.ac.kr 부위원장정채관 ( 한국교육과정평가원 ) ckjung@kice.re.kr 제 1 분과책임편집위원문안나 ( 인하대 ) anmoon@inha.ac.kr 위원 Thomas Cable(U of Texas, Austin) 고광윤 ( 연세대 ) 신성균 ( 강원대 ) 제 2 분과책임편집위원이진화 ( 중앙대 ) jinhlee@cau.ac.kr 위원박미애 ( 춘천교대 ) 박종원 ( 부경대 ) 전영주 ( 목원대 ) 제3분과 책임편집위원강은경 ( 상명대 ) egkang@smu.ac.kr 위원 Gregory Iverson(U of Wisconsin, Milwaukee) 손형숙 ( 경북대 ) 장세은 ( 해양대 ) 한은주 ( 서울여대 ) 한정임 ( 건국대 ) 제 4 분과책임편집위원김연승 ( 공주대 ) yskim@kongju.ac.kr 위원 Heizo Nakajima(Gakushuin U, Tokyo) 이광호 ( 경상대 ) 홍경선 ( 제주대 ) 제 5 분과책임편집위원위혜경 ( 단국대 ) hkwee@dankook.ac.kr 위원오세랑 ( 경상대 ) 이희철 ( 전북대 ) 홍민표 ( 명지대 )

4 ISSN Volume 18, Number 1 Spring 2018 In Young Yang Pause-tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels 1 Kanghee Lee Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF 30 Tae-Young Kim Ji-Young Kim Remotivation Strategies for Korean Male High School Students' English Learning 50 김연승 트위터누리꾼들의영어차용복합어새말사용및 78 관련태도분석 김현옥 한국영어학습자의유창성과발음에대한비원어 99 민채점자의이해및판단기준분석 안성호 ( 계명 ) 한국인영어학습자의관계대명사사용에관한코퍼스기반연구 119 모든논문은 [ 영어학 ] 의윤리규정을준수하였음을밝힙니다.

5 Korean Journal of English Language and Linguistics Vol. 18, No. 1, Spring 2018, /kjell Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels1 * In Young Yang (Seoul National University) Yang, I. Y Pause-tonic stress interaction in English L2 speech of Korean talkers with different proficiency levels. Korean Journal of English Language and Linguistics 18-1, This study reports on the interaction of pause and tonic stress production in L2 English speech. Pause and tonic stress are important prosodic manifestations of information structures that are crucial in facilitating listeners speech comprehension. Examinations were conducted on the placement of tonic stress and pauses within each tonality of read speech by 35 Korean adult learners. The participants were separated into three groups based on their speech comprehensibility and foreign-accent scores, using a Hierarchical Cluster Analysis. Each cluster revealed distinctive-both quantitative and qualitative-uses of tonic stress and pauses. As pronunciation proficiency increased, learners produced more accurate tonic stress and less frequent pauses within a tonality. In particular, participants with lower proficiency levels frequently placed pauses before less informative words. It is recommended that pronunciation be taught in a way that incorporates information structure, with a focus on pauses and tonic stress. Keywords: English pronunciation, tonic stress, pauses, information structure 1. Introduction With the emergence of the communicative language teaching paradigm, language researchers and practitioners have become more concerned with how learners can improve their communicative ability (Jang 2010, Park and Chang 2017). In * Parts of this research were originally conducted for the author s doctoral dissertation. 1

6 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels line with this expectation, the focus of teaching and researching pronunciation has been on improving learners speaking fluency/intelligibility than accuracy, leading to the neglect of segmental aspects of pronunciation in favor of an increased focus on suprasegmentals (Celce-Murcia, Brinton and Goodwin 2010). This change in interest has been based on assumptions and empirical evidence suggesting that intonation has pragmatic power in communicative circumstances (Celce-Murcia et al. 2010, Hirschberg 2004). In English, the intonation of a specific sentence, a tonality, is composed of tonic stress and its subsidiaries ( non-tonic syllables in Davenport and Hannahs 2010, p. 87). The compositional pattern of these elements patternizes the prosody of the specific stream of speech (Wells 2006). In other words, the pitch variations in a string of words are determined by the location of the tonic syllables. Of more importance is the fact that the choice of tonic syllable mirrors the informativeness of the statement (Halliday 1967); it has a critical effect on the ease or difficulty listeners experience in comprehending the speech in question (Birch and Clifton 1995, Bock and Mazzella 1983; Cutler and Fodor 1979). Another critical feature of speech comprehension is the use of pauses. A substantial volume of documentation has included pauses as variables that affect L2 speech comprehensibility for listeners. Most studies have focused on the quantitative aspects of pauses, such as pause frequency and length (Isaacs and Trofimovich 2012, Kang 2010, Park and Lee 2013, Trofimovich and Baker 2006). Only a handful of studies have addressed the placement of pauses, focusing on the coincidence of pauses and syntactic boundaries (Isaacs and Trofimovich 2012). Some have argued that pause occurrences within a tonality create the type of speech perceived as stuttering, while those that coincide with tonal/clausal boundaries (in more accurate terms, phonemic clauses, (Boomer 1965, 1978 as cited in Laver 1994, p. 536) give an impression of fluent speech (Laver 1994). However, it also has long been assumed that pauses result from the execution of speech planning (Goldman-Eisler 1958), where one run unit 2

7 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels generally makes up a tonal unit that does not always correspond to a syntactic unit (Davenport and Hannahs 2010, Halliday 1967, Roach 2009). In addition, the placement of pauses is known to interact with the placement of tonic stresses within a tonality (Hahn and Dickerson 1999). This background information legitimizes the examination of tonic stress and pause interrelation (as information within a tonality domain) to fully understand the factors relevant to L2 speech. With this purpose in mind, this study addresses the following research questions: 1. How does the way in which Korean learners produce English tonic stress and pauses interrelate with their pronunciation proficiency? 2. What are the characteristics of Korean learners production of pauses in relation to informativeness within a tonality? 2. Background 2.1 Tonic Stress and Tonality English speech is composed of a series of tone units (tonality) (Kreidler, 1997). It is difficult to define tonality in syntactic terms, because it does not always correspond to a specific syntactic unit (Roach 2009). According to Halliday and Greaves (2008), the flow of discourse in English is organized primarily by the systems of TONALITY and TONICITY (p. 97). The following is the description of tonality and tonicity in relation to the information structure. Tonality systems are the choices concerned with how the discourse unfolds as a succession of units of information. A unit of information is a quantum of discourse that is organized into some configuration of given and new material.... The tone unit is a unit of English phonology, and it can be defined as one melodic contour one line of spoken melody.... The information unit, 3

8 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels on the other hand, is a unit of lexicogrammar; it is defined (from below) as that grammatical unit that is realized by one tone unit. Thus the two units, the phonological tone unit and the grammatical information unit, correspond one to one;... Whereas tonality concerns the organization of discourse into quanta of information, tonicity refers to the internal organization of each one of these information quanta. This is called tonicity because it revolves around the location of the tonic, the prominent portion of the tone unit (Halliday and Greaves 2008, pp ). According to Halliday (1967), a string of words builds up units of information. These information units are not always organized within grammatical syntactic boundaries. A single item in an information unit requires a special focus that is prosodically realized as an accent to new information. In sum, tonality prosodically mirrors the information unit of a given speech; the assignment of the tonic syllable is determined within each information unit. Tonic stress is essential in composing a tone unit, which may consist of a single syllable, more than one syllable, or a full sentence (Roach 2009). In ordinary speech, a short sentence is produced as one tone unit, as shown in the material used in this study (presented in Appendix). Tonic stress is usually assigned to the last lexical item in a tone unit, which is said to be unmarked tonic stress (Halliday and Greaves 2008). The location of tonic stress generally falls on an informative content word that provides new information. Sometimes, it falls on a function word that creates new meaning in a given context (Hahn and Dickerson 1999). The assignment of tonic stress interacts with the givenness and contrastiveness of the items in a tone unit and across tone units; the speaker may place tonic stress in the earlier position (i.e., marked tonic stress) in a tone unit (Halliday and Greaves 2008). It has been actively documented that the appropriate placement of tonic stress facilitates speech comprehension (Birch and Clifton 1995, Bock and Mazzella 1983, Cutler 1976, Cutler and Fodor 1979, Terken and Nooteboom 1987). The accentuation of items of new information and the deaccentuation of 4

9 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels given information directly benefits listeners speech comprehension. Inaccurate use of this system confuses listeners, making it difficult for them to evaluate the information provided by the speech. In English, the assignment of tonic stress requires learners to know a language-specific rule: when all of the items in a tone unit are new, the sentence-level accent moves to the last content word (Davenport and Hannahs 2010). Korean does not have such a rule; instead, it frequently employs syntactic or morphological mechanisms (Choi 1997, Jun 2013, Kim 2014, 2016, Sohn 1999). This difference predicts the difficulty that Korean learners tend to experience when acquiring and producing tonic stress and tonality. 2.2 Pause and Information Structure Halliday (1967) was the first to define information structure, a term that had a more technical meaning in Lambrecht (1994), where it was the formal expression of the pragmatic structuring of a proposition in a discourse (p. 5). As mentioned in the previous section, English tonic stress is used to manifest the informativeness of words. Speakers and listeners within given discourse consistently evaluate the informativeness of speech. Pauses are inevitably related to the information structure of speech. The placement of pauses has been investigated from different perspectives, by focusing on syntactic terms, the coincidence of pauses and syntactic boundaries (Hawkins 1971), and semantic terms (O Connell, Kowal and Hörmann 1969). Goldman-Eisler (1958) has investigated within-sentence pauses, focusing on word information. She has hypothesized that a pause within a sentence anticipates the increase of information in subsequent speech and involves acts of choice (p. 96). Her investigations have revealed that within-sentence pauses differentiate between old and new items in a given speech, corresponding to well organized and now organizing speech (p. 105). More specifically, pauses have been observed mostly before lexical words with high semantic content, indicating an interrelation between pauses and content words. The present study will limit its focus to the informativeness of words in a 5

10 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels given context. 2.3 Pause and Tonic Stress in the Development of L2 Information Structure As mentioned in the previous section, the appropriate use of tonic stress critically contributes to speech comprehension in the English language (Cutler 1976). Tonic stress inevitably places prosodic emphasis on new information and unmarked tonic stress marks the end of a prosodic boundary. The attention of listeners is directed more effectively toward important pieces of information when they are aided by tonic stress and tonality (Halliday 1967). Hahn and Dickerson (1999) have claimed that the production of pauses and tonic stress are interrelated. A brief pause is placed before or after a certain word with tonic stress when a speaker wants listeners to pay particular attention to it. Research on the production of English tonic stress has revealed that Korean learners produce too many tonic stresses in a sentence, creating an equal prosodic emphasis on lexical words (Um, Lee and Kim 2001). This may result from an over-use of pauses within a sentence as indicated in Han and Lee (2008). This will impair the comprehension of listeners by hindering their ability to effectively evaluate the information structure of L2 speech. Little is known about the interrelated production of pause and tonic stress in terms of information structure in the development of L2 pronunciation proficiency. As mentioned in the previous sections, pause and tonic stress share the same application domain, information structure. The two necessarily make use of the degree of informativeness of lexical items in a given speech. This study therefore explores the interaction between pauses and tonic stress in this respect to shed new light on the teaching of L2 pronunciation. 6

11 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels 3. Method 3.1 Participants A total of 35 graduate and undergraduate students from a university in Seoul participated in this study on a voluntary basis; they were paid for their participation. The gender of the participants was controlled; to avoid any confounding variables, all participants were female. Their ages ranged from 19 to 39. Among the 35 participants, 18 had lived in an English-speaking country for periods ranging from 6 weeks to 20 years (Length of Residence: LOR). Their age of arrival (AOA) in the country varied from 0 to 35 years. The participant with an AOA of 0 was born in an English-speaking country but was raised by a Korean family; presumably, her first language was Korean, giving rise to her somewhat low foreign-accentedness score. Taking everything into consideration, she was included as a participant. Of the participants, 11 had English-related majors, and the rest had other majors, including social sciences, natural sciences, nursing, medicine, engineering, food & nutrition, arts, humanities, French, and Korean language. For this study, participants were recruited without controlling for their language proficiency, LOR, AOA, major, or age. The acquisition of L2 pronunciation is distinguished from that of other language skills because the developmental patterns of L2 pronunciation should be addressed considering the learners age and the relation between production and perception (Fraser 1999 as cited in Kennedy and Trofimovich 2017). In other words, none of those variables alone can predict the pronunciation proficiency of L2 learners. This study adopted pronunciation-rating procedures to measure the participants pronunciation proficiency levels. They were grouped solely according to their pronunciation abilities. Participants profiles are presented in more detail in Table 1 in section 4.1, along with their overall pronunciation scores. 7

12 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels 3.2 Procedures Recording Recordings were made in a sound-attenuated room using a SONY PCM-M10 recorder with a built-in microphone. A dialogue with nine turns was used in this study, as proposed by Kreidler (2004, pp ); the dialogue is presented in Appendix. This dialogue was chosen for its simple and easy vocabulary, to ensure that participants did not hesitate while reading due to difficult or unfamiliar vocabulary. In addition, an explicit explanation of tonic stress and tonality was included in the reference book, exempting the author from any possible subjectivity in determining tonic stress and tonality. Although the locations of tonic stress and tonality are marked in the dialogue in Appendix, the participants received a version of the dialogue that did not provide information on the placement of tonic stress or tonality. The 35 participants were given a dialogue to record. A male native speaker of American English also participated in the recordings as an interlocutor. The participants were given enough time to familiarize themselves with the text so as to elicit the most natural pronunciation possible. They were first asked to read the dialogue following the lead of the male speaker. The first reading allowed participants to get used to the dialogue and overcome any effect of unfamiliarity. They then read 34 other (very short, thus not affecting the participants fatigue) dialogues for another experiment. After that, the participants were asked to read the target dialogue again, this time taking on the opposite role. Thus, the female participants read the parts that the male speaker had read during the first reading. The second recording was used for the analysis. These procedures were designed to overcome any unfamiliarity effect of the text, as were the direct repetition and imitation of the male native speaker s reading. Ideally, spontaneous speech would be more appropriate for examining pause and tonic stress interaction, especially if the pauses were a by-product of speech planning. However, it is challenging to observe both pauses and tonic 8

13 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels stress because tonic stress is a more subjective phenomenon, making it difficult for the author to determine the appropriate location of tonic stress within a given context. It is the speaker in spontaneous speech who determines the location of tonic stress to convey an appropriate meaning. This study therefore achieved a compromise by using read speech as its experimental material. O Connell et al. (1969) have also used some read speech, showing that pauses also occur in read speech. The pauses were essentially the same as in spontaneous speech. Goldman-Eisler (1972) has compared the pause production of English speakers in read and spontaneous speech, and found that the length of the pauses was different. In read speech, speakers manifest more fluency, with shorter pauses. She has examined any silence longer than 0.25s as pauses. In the production of pauses within clauses, relatively long pauses, such as one lasting 0.75s, disappear in read speech while continuing to occur in spontaneous speech. My study has resolved this issue by defining a pause as any silence longer than 0.1s. A pause is a very short silence that is nevertheless longer than any silence needed for speech production (such as stop closures), following previous research (Kang 2010). Although pauses in read speech may not be the same, or as natural, as those in speech planning, the underlying composition of the tonality of given strings of words is the same in read and spontaneous speech. More proficient speakers are presumably more proficient readers, pausing less frequently. In reading, the informativeness of each word can affect the ease or difficulty of reading a passage aloud. To my knowledge, there has been little empirical research done on the developmental patterns of tonic stress/tonality and pauses with relation to information structure in L2 English. This study therefore provides a significant observational starting point. Future research should adopt a more elaborate experimental design to address this issue in spontaneous speech. 9

14 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels Scoring The recorded dialogues have been scored by native English listeners for their comprehensibility and foreign-accentedness. This procedure has been used to divide the participants into the appropriate levels of overall pronunciation proficiency. Compared with other language skills, such as reading, writing, and vocabulary, which usually correspond to a general proficiency level, the pronunciation of learners often varies within one proficiency level, because many factors interact in pronunciation development. L2 pronunciation develops from interaction with various factors such as age, LOR, AOA, and overall language proficiency. Therefore, none of these factors are direct indications of the participants pronunciation levels (Fraser 1999 as cited in Kennedy and Trofimovich 2017). For example, L2 pronunciation research sometimes shows mixed results from the effect of those variables (e.g., comparisons of experienced and inexperienced learners in Flege, Munro and Skelton 1992, Kang 2007). To avoid this complexity of pronunciation developmental patterns, the present study has divided participants into groups using a Hierarchical Cluster Analysis (HCA) based on their pronunciation comprehensibility and foreign-accentedness scores. The present study assumes that such factors, which determine learners proficiency, are all incorporated into the comprehensibility and foreign-accentedness scores, bypassing the need to rely on a specific set of criteria. Comprehensibility means a listeners perception of the degree of difficulty in understanding an utterance (Munro, Derwing and Holtby 2012, p. 233). Foreign-accentedness refers to the degree of deviation from native pronunciation norms (Munro and Derwing 1994, 1995a, 1995b). Both measures have been used to assess the effects of accuracy and speech comprehension on the participants pronunciation proficiency. A total of 10 native American English speakers, with varying degrees of experience in Korea, were recruited; they were instructed, as listeners, to rate the comprehensibility and foreign-accentedness of each speech sample on a 7-point Likert scale. They 10

15 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels were given five practice scoring examples to familiarize themselves with the process Analysis The recorded materials were analyzed to assess tonic stress accuracy and pause frequency within a tonality. To calculate the accuracy of each participant s tonic stress, the following procedures were used. First, we recruited four American English speakers with linguistic backgrounds (and thus an understanding of sentence level prominence). They listened to the recorded materials, indicating prominent words. This procedure was necessary because tonic stress is a perceptual phenomenon; its placement must be discerned by native listeners. They listened to each sentence of the dialogue separately, in a randomized order, to avoid predicting the location of prominence. The prominence perceived by three or more native listeners was counted as the prominence produced by the participant. Prominent words with the appropriate placement of tonic stress were counted and divided by 13, the total number of expected tonic stress placements in the dialogue. More frequently, the Tone and Break Indices (ToBI) transcription (Silverman, Beckman, Pitrelli, Ostendorf, Wightman, Price, Pierrehumbert and Hirschberg 1992) has been used to indicate the location of tonic stress. This study has adopted the method used by Mo (2010), who has shown that untrained native listeners perception of prominence is no different from the perception of trained native listeners using ToBI. It would be ideal to employ as many native listeners as possible, but this study has used only four native listeners with an understanding of sentence level prominence, due to restrictions in the Korean environment. In the pause analysis, adopting a technique used in previous research (Kang 2010), all silent pauses longer than 0.1s were counted, using Praat speech analysis software, and examined in relation to their distribution. First, the number of pauses within a tonality was counted. The type of words used after the pause (whether function or content words) was then investigated, referring 11

16 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels to the results of Goldman-Eisler (1958). 3.3 Statistical Design This study has adopted three procedures, using SPSS 24 for the statistical analysis. As a first step, to optimally divide participants into groups based on their pronunciation proficiency, a Hierarchical Cluster Analysis (HCA) was carried out, using two pronunciation scores (comprehensibility and foreign-accentedness). The clusters created by HCA were used as groups to show the different proficiency levels (thus, presumably indicating developmental stages) of pronunciation. More often, K-means is used with group participants, but this method requires an exact number of expected divisions, based on a theoretical background. Since there was no way to know how many divisions to expect among the participants in this study, HCA was used instead. Second, the accuracy of tonic stress and the number of pauses were compared among the three clusters, and a MANOVA was conducted to examine whether the differences among the groups reached statistical significance. Finally, a Poisson log-linear distribution was used to test the distributional difference between groups, to see whether the frequency of pause occurrence would have different patterns, reflecting the word types that provided different degrees of information. 4. Results and Discussion 4.1 Hierarchical Cluster Analysis The participants in this study were divided into groups based solely on pronunciation quality, using a Hierarchical Cluster Analysis. Table 1 presents the profiles of the participants. The analysis resulted in three partitions, based on participants comprehensibility and foreign-accentedness levels. The cluster scores are listed from high to low; additional participant information (LOR, 12

17 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels AOA, age, and major) is also summarized. Table 1. Participants Information in the Three Clusters Cluster Case no. Comprehensibility Accent (month) (years) Foreign LOR AOA Age Major A (N=14) English English English English English English English Mean B (N=16) English English English English Mean C

18 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels (N=5) Mean Note. Comprehensibility 7: Very easy to comprehend, 1: Very difficult to comprehend Foreign-Accentedness 7: Native-like accent, 1: Very strong foreign accent LOR: Length of Residence, AOA: Age of Arrival. Cluster A consisted of 14 participants with average scores of 6.4 and 5.9 in their comprehensibility and foreign-accentedness ratings, respectively. 11 participants of them had experience in English-speaking countries, and eight reported a long residence. Half of them had arrived in those countries at early ages. The three participants with short LORs (12-24 months) were all majoring in English-related fields. The main feature of this cluster was the fact that the early learners with long LORs were perceived as native speakers by themselves and others. Cluster B consisted of 16 participants with intermediate-level scores for the two criteria (4.9 and 4.2). Seven of them had spent a short period of time in an English-speaking country, and most were late learners.. Cluster C included the five participants with the lowest scores (3.2 and 2.6). All of the Cluster C participants were non-english majors with no experience in English-speaking countries. The Figure 1 dendrogram shows the grouping results from the HCA. 14

19 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels Figure 1. Dendrogram of the HCA Along the 7-point Likert scale, the participants in this study displayed a high score skew, confirming that they had a relatively high degree of pronunciation proficiency. The MANOVA revealed that differences between the three clusters were significant (F=36.261, p<.001). Tests of between-subject effects showed significant differences for both comprehensibility (F= , p<.001) and foreign-accentedness (F=91.014, p<.001). Pairwise comparisons and post hoc tests (Scheffe) indicated that the significance was relevant to all three clusters. The results of multiple comparisons of any of the two clusters reached a level of significance for both comprehensibility and foreign-accentedness. The rest of this paper will focus on the differences between these groups, exploring this statistically meaningful difference. 4.2 The Production of Tonic Stress and Pause Quantitative differences The main purpose of this work has been to examine whether different learners with varying levels of pronunciation proficiency show different patterns of pause and tonic stress production in relation to tonality reflecting 15

20 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels information unit. The three clusters have been tested on the accuracy of their tonic stresses and the number of pauses within the tonality. This process has attempted to discover whether or not the three clusters manifest quantitatively distinctive productions of pause and tonic stress, thus revealing whether the acquisition of pause and tonic stress are interrelated phenomena in L2 pronunciation development. Table 2 presents the three clusters tonic stress accuracy and the number of pauses, along with means and standard deviations. In general, as the pronunciation quality rating scores dropped, the accuracy of tonic stress production also decreased. At the same time, pause production increased within a tonality, as pronunciation quality decreased. The relatively low accuracy of tonic stress shows that the production of appropriate tonic stress was difficult for the participants in this study. This may have resulted from the fact that the task required the participants to evaluate the overall discourse information to produce both unmarked and marked tonicity appropriately. Table 2. Tonic Stress Accuracy and the Number of Pauses Tonic Stress Accuracy N of Pauses Clusters N Mean S.D. Mean S.D. A B C Total Figure 2 reveals the patterns of pause and tonic stress production in relation to the development of pronunciation proficiency. 16

21 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels Figure 2. The Interaction of the Number of Pauses and Tonic Stress Accuracy In sum, as proficiency increases, the production of pauses of the participants decreases within a tonality, while the accuracy of tonic stress production increases in the opposite situation. A MANOVA was conducted to verify whether the differences among the clusters were meaningful for tonic stress accuracy and the number of pauses. The statistical results confirmed the differences between clusters at the statistical level (F=9.872, p<.001) as shown in Table 3. Table 3. Results of MANOVA: Multivariate Tests Effect Value F Hypothesis df Error df Sig. Noncent. Observed ParameterPower Intercept Wilks' Lambda b Clusters Wilks' Lambda b a. Design: Intercept + Clusters b. Exact statistic c. The statistic is an upper bound on F that yields a lower bound on the significance level. d. Computed using alpha =.05 Table 4 presents the results of the tests of between-subject effects. As the table indicates, there were significant differences among the clusters in both tonic stress accuracy (F=8.314, p=.001) and the number of pauses within a tonality (F=11.669, p<.001). 17

22 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels Source Corrected Model Intercept Clusters Error Total Table 4. Results of MANOVA: Tests of Between-Subjects Effects Type III Sum Mean Noncent. Observed df F Sig. of Squares Square Parameter Power c Tonic Stress.229 a N of Pauses b Tonic Stress N of Pauses Tonic Stress N of Pauses Tonic Stress N of Pauses Tonic Stress N of Pauses Corrected Tonic Stress Total N of Pauses a. R Squared =.342 (Adjusted R Squared =.301) b. R Squared =.422 (Adjusted R Squared =.386) c. Computed using alpha =.05 Post hoc tests (Scheffe) of the number of pauses revealed that the differences between each pair of clusters reached statistical significance. The results of tonic stress accuracy showed that the differences were significant only for Cluster A and C Distributional differences This section presents the distribution of pauses in relation to the informativeness of words. Specifically, comparisons were conducted on the distribution of pauses preceding function and content words within a tonality, based on the findings of Goldman-Eisler (1958). This analysis aims to more directly examine differences in pause-tonic stress production among the clusters. The differences found in the previous section may be by-products of pronunciation development, which do not necessarily suggest that the two are related. To confirm that pauses and tonic stress are interrelated in the acquisition of pronunciation, it is important to prove that the two are related within the acquisition of the L2 information structure. This study has 18

23 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels hypothesized that systematic differences in pause location within a tonality will support, in part, the interconnected pattern of pause and tonic stress in L2 pronunciation development, since both pauses and tonic stress require learners to evaluate the informativeness of items within the tonality. The distribution of pauses in relation to word information has therefore been examined. It was predicted that learners would pause less frequently before function words as their pronunciation (and thus tonic stress accuracy) improved. In counting the number of pauses before function and content words, a problem arose regarding contracted forms, such as don t and doesn t. Such words combine a function word with an adverb (or negative particle). In this study, such words are considered content words. This decision has been based on the following two assumptions. First, negative contractions have both syntactic and semantic salience, which generally indicate a stressed form (Celce-Murcia et al. 2010, p. 210). Second, this study has hypothesized that the pauses before function words will decrease as the learners pronunciation proficiency improves. Learners with lower proficiency levels produce more pauses; contracted forms that include not have therefore been categorized as content words to avoid exaggerating the distributional pattern of pauses. Table 5 compares pause occurrences before function and content words among the three clusters. The use of read speech in this study allowed the researcher to compare the frequency of pauses in each position since the possible distributional locations were the same for all speakers. The number of pauses and their ratios in each location are presented below, grouped by cluster. Table 5. Pause Frequency and Ratio Before Function and Content Words Clusters N Function Words (ratio) Content Words (ratio) Total 1 A 14 3 (0.12) 23 (0.88) 26 B (0.26) 52 (0.74) 70 C 5 11 (0.29) 27 (0.71) 38 Total Total means a total number of pauses that the participants in each group produced 19

24 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels Using the ratio data in Table 5, Figure 3 more clearly shows the extent to which, as pronunciation proficiency decreases, more frequent pauses are found before function words. Pauses occur more often before content words than function words in all three groups. The differences are generally similar to those revealed in the previous quantitative analysis. However, the difference between clusters B and C is not as evident as in the previous analysis. Figure 3. Pause Occurrence Ratio Before Function and Content Words In a more detailed investigation, a statistical analysis has been used to map differences in pause location. The Poisson log-linear distribution has been used to analyze distribution data. Systematic differences among the three groups have statistical significance for both positions: before function words (Wald χ2=11.805, p=.003) and before content words (Wald χ2=12.904, p=.002). Table 6. Results of Poisson Log-linear Distribution Before Function Words: Tests of Model Effects Source Type III Wald Chi-Square df Sig. (Intercept) Clusters Dependent Variable: Before Function Words Model:(Intercept),Clusters as a whole. 20

25 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels Table 7. Results of Poisson Log-linear Distribution Before Content Words: Tests of Model Effects Source Type III Wald Chi-Square df Sig. (Intercept) Clusters Dependent Variable: Before Content Words Model: (Intercept),Clusters Cluster C was set zero in parameter estimates. For function words, there is a significant difference between Clusters A and C (Wald χ2=11.595, p=.001) and the difference between B and C didnot reach significance (Wald χ2=2.785, p=.095). For content words, there is also a significant difference between Clusters A and C (Wald χ2=12.874, p<.001). The difference between Clusters B and C did not reach significance (Wald χ2=3.354, p=.067). Within the general categories of function and content words, each grammatical category was compared. The following three investigations were noteworthy. For function words, pauses before prepositions were found only among Cluster C participants, with the lowest level of proficiency. Pauses before auxiliary verbs were distributed across Clusters B and C. For content words, the relatively low-level learners in Cluster C paused most frequently before nouns, implying that learners retain first and most easily the informativeness of nouns. The analyses in this section are summarized as follows. First, as learners develop their pronunciation, they produce less frequent pauses and more accurate tonic stresses within a tonality. Second, the location of pauses differs with the development of L2 pronunciation: as their pronunciation proficiency improves, learners come to prefer pauses before content words that are more informative or cannot be predicted from the context. In other words, with an increased awareness of tonality, learners come to eliminate pauses before function words that are less informative and usually predictable from the context. 21

26 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels 4.3 General Discussion This study has addressed two questions about Korean learners production of pauses and tonic stress. Question 1 involves the quantitative aspects of pause and tonic stress production in relation to pronunciation development reflected in the different proficiency levels. The developmental progress associated with rising proficiency levels shows that tonic stress accuracy is inversely related to the number of pauses. As learners increase their pronunciation proficiency, they produce fewer pauses within a tonality (thus an information unit) and increase the accuracy of their tonic stress production. The different proficiency levels (clusters) revealed quite systematic change toward more prosodically accurate and fluent (without pauses) speech. It can therefore be deduced that the accuracy of tonic stress is inversely related to the number of pause productions at least in part. It has been assumed that the appropriate placement of tonic stress requires the awareness of a tone unit in the given string of words. Being able to perceive a tone unit means that learners can recognize the string of words as a continuous unit that is produced within a melody. It seems natural that the internalization of a tone unit would accompany the more accurate location of tonic stress, leading to a more fluent production of the unit, without pauses. The second question addresses the distribution of pauses in relation to information. Previous research has shown that native English speakers almost always produce (within-sentence) pauses before lexical words with high semantic content, because they are less predictable, and therefore more informative (Goldman-Eisler 1958). Tonality and tonic stress are interrelated with information structure. Tonality mirrors the information structure of a string of words, and tonic stress moves to the last informative ( new ) word within an information structure (Halliday 1967, Laver 1994). The learners in this study paused less frequently before function words, as their pronunciation proficiency developed. This result of pause distributional change confirms, in part, that a learner s awareness of tonality includes an ability to evaluate the information load of 22

27 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels particular words. In this analysis, the difference between Clusters A and C showed distinctive patterns of tonic stress accuracy and pause frequency data. The distinctive pause distribution of the participants with the different proficiency levels confirms the finding that evaluating the informativeness of words is interrelated with acquiring tonic stress production with an awareness of tonality. As learners pronunciation proficiency increases, they become aware of tone unit and information structure. This, in turn, improves the accuracy of their tonic stress placement, leading to a tendency to pause before highly informative words. However, the differences in pause distribution patterns before function words were not evident between Cluster B and C unlike the quantitative observations in the first analysis. The reason for this discrepancy may be the skewed distribution of participants proficiency in this study. Future research into wider ranges of pronunciation proficiency will reveal more systematic progress in the production of pauses and tonic stress. With the investigation of each grammatical category, lower-level learners paused more frequently before prepositions and auxiliary verbs within the function word category. These pauses were almost unobservable in the speech of high-level learners (Cluster A). This result suggests that learners develop syntactic knowledge along with prosodic awareness. A more in-depth study is needed to support this claim. As for pauses before content words, participants in Cluster C paused more frequently before nouns than any of the other clusters. This suggests that learners become accustomed to the information structure of English through frequent use. Tonic stress is often assigned to the last lexical item in a tone or information unit; these words are usually nouns in the object position of the sentence, since English has SVO word order and tends to move highly informative words to the end of a sentence (Halliday and Greaves 2008). It is therefore easier to understand the informativeness of nouns than other grammatical words. Future research on this topic should uncover fruitful aspects of second language acquisition. The investigation of pause and tonic stress in relation to information is 23

28 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels significant because they both directly affect listeners speech comprehension (Cutler 1976, Hahn and Dickerson 1999). In particular, in L2 English teaching that uses a fluency-based paradigm, comprehensible speech is of particular importance. In this trend in English education, teaching pronunciation should help learners improve their speech comprehensibility. The information structure is primarily realized with prosody in English speaking. This prosodic realization requires the appropriate understanding of new and given information. Thus, L2 learners should learn about pauses and tonic stress production in relation to information structure to make their speech better understood. In addition, this ability is closely related with the effective organization of information in communication, both spoken and written, which has great implications for English language teaching in general. 5. Conclusion The experiments in this study have shown that the development of pronunciation proficiency improves tonic stress assignment and eliminates pauses within a tonality, leading to more fluent and comprehensible speech. In addition, when pauses occur within a tonality, learners grow accustomed to the informativeness of words in the tonality, as they develop better pronunciation. This indicates that learners internalize the information structure of the target language, which is manifested by tonality and tonic stress with more natural pause locations. This study has several limitations. First, it has relied on read speech for its analysis, which may have affected the results, especially in relation to pause occurrences. A more in-depth study, using spontaneous speech samples, is needed to advance future research. This study has also included participants with a relatively high level of English pronunciation proficiency in that the majority of the participants obtained scores over 3.5 along the 7-point Likert scale. In addition, in dividing the participants into groups with different proficiency levels, HCA yielded five participants for Cluster C, which is 24

29 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels relatively small in number for statistical analysis. Future research with wider ranges of pronunciation proficiency will be able to develop more comprehensive patterns of pause and tonic stress interaction in pronunciation development. Even with these methodological limitations, this study has significance; it is the first attempt to investigate pause-related phenomena with a focus on information structure reflected in tonic stress and tonality. The results of this study reveal the potential value of integrating pause lessons into pronunciation and speaking instruction. References Birch, S. and C. Clifton Focus, accent, and argument structure: Effects on language comprehension. Language and Speech 38(4), Bock, J. K. and J. R. Mazzella Intonational marking of given and new information: Some consequences for comprehension. Memory & Cognition 11(1), Boersma, P. and D. Weenink Praat: Doing Phonetics by Computer (version ). [Computer program] Retrieved from Boomer, D. S Hesitation and grammatical encoding. Language and Speech 8, Boomer, D. S The phonemic clause: Speech unit in human communication. In A. W. Siegman and S. Feldstein, eds., Nonverbal Behavior and Communication, NJ: Lawrence Erlbaum Associates. Celce-Murcia, M., D. M. Brinton and J. M. Goodwin Teaching Pronunciation: A Reference for Teachers of English to Speakers of Other Languages. Cambridge: Cambridge University Press. Choi, H-W Topic and focus in Korean: The information partition by phrase structure and morphology. Japanese/Korean Linguistics 6, Cutler, A Beyond parsing and lexical look-up. In R. J. Wales and E. C. T. Walkers, eds., New Approaches to Language Mechanisms: A Collection of Psycholinguistic Studies, Amsterdam: North-Holland Pub. Cutler, A. and J. Fodor. A Semantic focus and sentence comprehension. 25

30 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels Cognition 7(1), Davenport, M. and S. J. Hannahs Introducing Phonetics and Phonology. NY: Routledge. Field, J Psycholinguistics: A Resource Book for Students. Psychology Press. Flege, J. E., M. J. Munro and L. Skelton Production of the word-final English /t/-/d/ contrast by native speakers of English, Mandarin, and Spanish. Journal of the Acoustical Society of America 92, Goldman-Eisler, F Speech production and the predictability of words in context. Quarterly Journal of Experimental Psychology 10(2), Goldman-Eisler, F Pauses, clauses, sentences. Language and Speech 15(2), Hahn, L. D. and W. B. Dickerson Speech Craft: Discourse Pronunciation for Advanced Learners. Ann Arbor: The University of Michigan Press. Halliday, M. A. K Notes on transitivity and theme in English: Part 2. Journal of Linguistics 3(2), Halliday, M. A. K. and W. S. Greaves Intonation in the Grammar of English. CT: Equinox Pub. Han, H. and J-K. Lee Korean speakers phrasing patterns in English clauses. Modern English Education 9(3), Hawkins, P. R The syntactic location of hesitation pauses. Language and Speech 14(3), Hirschberg, J Pragmatics and intonation. In L. R. Horn and G. L. Ward, eds., The Handbook of Pragmatics, Malden, MA: Blackwell Pub. Isaacs, T. and P. Trofimovich Deconstructing comprehensibility. Studies in Second Language Acquisition 34(3), Jang, Y Pedagogical effects of incorporating speech acts found in romantic comedy movies for college English listening class: Sociolinguistic implication for English education. Korean Journal of English Language and Linguistics 10(4), Jun, Y Hangugeo Myeongsagu-ui Uimiron: Hanjeongseong/Teukjingseong, Chongchingseong, Boksuseong (A Semantics for Korean Noun Phrases: Definiteness/Specificity, Genericity, Plurality). Seoul: Seoul National University Press. Kang, O Relative salience of suprasegmental features on judgments of L2 comprehensibility and accentedness. System 38(2), Kang, S A study on Korean students production and perception of English word-final stop voicing. Speech Sciences 14(1),

31 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels Kennedy, S. and P. Trofimovich Pronunciation acquisition. In M. Sato and S. Loewen, eds., The Routledge Handbook of Instructed Second Language Acquisition, NY: Taylor & Francis. Kim, I On the meaning of Korean i/ka. Language and Linguistics, 63, Kim, I Can Korean-(n) un mark (contrastive) focus? Language Sciences 56, Kreidler, C. W Describing Spoken English: An Introduction. NY: Routledge. Kreidler, C. W The Pronunciation of English: A Course Book. Malden, MA: Blackwell Pub. Lambrecht, K Information Structure and Sentence Form: Topic, Focus, and the Mental Representations of Discourse Referents (Vol. 71). Cambridge: Cambridge University Press. Laver, J Principles of Phonetics. Cambridge University Press. Mo, Y Prosody Production and Perception with Conversational Speech (Unpublished doctoral dissertation). University of Illinois at Urbana-Champaign, IL, USA. Munro, M. J. and T. M. Derwing Evaluations of foreign accent in extemporaneous and read material. Language Testing 11(3), Munro, M. J. and T. M. Derwing. 1995a. Foreign accent, comprehensibility, and intelligibility in the speech of second language learners. Language Learning 45(1), Munro, M. J. and T. M. Derwing. 1995b. Processing time, accent, and comprehensibility in the perception of native and foreign-accented speech. Language and Speech 38(3), Munro, M. J., T. M. Derwing and A. K. Holtby Evaluating individual variability in foreign accent comprehension. Proceedings of the 3rd Pronunciation in Second Language Learning and Teaching Conference, O Connell, D. C., S. Kowal and H. Hörmann Semantic determinants of pauses. Psychological Research 33(1), Park, S. and J-H. Lee A study on the comprehensibility of Korean speakers English utterances and affecting factors for native and non-native English listeners. Korean Journal of English Language and Linguistics 13(2), Park, T. and J. Chang A study on high school teachers opinions about issues of communicative English teaching and assessment. Korean Journal of English Language and Linguistics 17(4),

32 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels Roach, P English Phonetics and Phonology. Cambridge: Cambridge University Press. Silverman, K. E. A., M. Beckman, J. F. Pitrelli, M. Ostendorf, C. Wightman, P. Price, J. Pierrehumbert and J. Hirschberg TOBI: A standard for labeling English prosody. Proceedings of the 1992 International Conference on Spoken Language Processing 2, Sohn The Korean Language. Cambridge: Cambridge University Press. Terken, J. and S. G. Nooteboom Opposite effects of accentuation and deaccentuation on verification latencies for given and new information. Language and Cognitive Processes 2(3-4), Trofimovich, P. and W. Baker Learning second language suprasegmentals: Effect of L2 experience on prosody and fluency characteristics of L2 speech. Studies in Second Language Acquisition 28(1), Um, H., H. Lee and K. Kim Korean speakers realization of focus and information structure on English intonation in comparison with English native speakers. Speech Sciences 8(2), Wells, J. C English Intonation: An Introduction. Cambridge: Cambridge University Press. Examples in: English Applicable Languages: English Applicable Level: All In Young Yang Department of English Language Education Seoul National University 1 Gwanak-ro, Gwanak-gu, Seoul, Korea inyoung@snu.ac.kr Received: January, 2018 Revised: February, 2018 Accepted: March,

33 In Young Yang Pause-Tonic Stress Interaction in English L2 Speech of Korean Talkers with Different Proficiency Levels Appendix A Recording Material From Kreidler (2004, pp ) (The locations of tonic stress and tonality are underlined and separated with slashes) Female: / Have you taken your family to the zoo yet, / John? / Male: No, but my kids have been asking me to. I ve heard this city has a pretty big one. Female: / Yes, / it doesn t have a lot of animals, / but it has quite a variety of animals 2. / I think 3 your kids / would enjoy seeing the pandas./ Male: I m sure they would. I d like to see them, too. Female: / Also, / the tigers are worth looking at. / Male: Is it okay to feed them? Female: / No, / they re not used to being fed. / Male: What bus do you take to get there? Female: / Number Twenty-eight. / But don t you have a car? / Male: We used to have one, but we had to sell it. 2 animal is also accentable in this context, if we assume that the speaker is planning the speech, according to Wells (2006). 3 think is also accentable, according to Kreidler (2004). 29

34 Korean Journal of English Language and Linguistics Vol. 18, No. 1, Spring 2018, /kjell Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF Kanghee Lee (Hongik University) Lee, Kanghee Understanding culture and intercultural awareness in intercultural communication through ELF. Korean Journal of English Language and Linguistics 18-1, It is commonly accepted that culture and language are closely intertwined each other. However, the cultural dimension to language has still stayed as a national concept, and the interrelation between the English language and a specific culture based on a particular native speaker variety has been criticized as a problem. This study will revisit the conceptualization of culture and the relationship between culture and language and examine some theoretical concepts such as third place, transcultural flows, critical cultural awareness, and intercultural awareness. This paper argues that the relationship between culture and language in ELF is in a constant tension between individual, local, regional and global contexts, and the concept of culture needs to be approached in a more complex and flexible manner between fluidity and fixity'. Therefore, the nature of culture in ELF, where communication takes place in multilingual and multicultural contexts, is more emergent, situated and dynamic. This would propose that ELF speakers need to develop intercultural awareness and negotiation and mediation skills to effectively manage the diversity and complexity in intercultural communication. Keywords: ELF, intercultural communication, culture, intercultural awareness, complexity 1. Introduction Globalization has led to the change of the role of English, which is more used for international communication for political, economic, business, cultural 30

35 Kanghee Lee Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF and academic purposes. English is used most commonly as a contact language between speakers from different linguistic and cultural backgrounds, and today the linguacultural environment of the world has become more multilingual and multicultural (Kramsch 2009). In addition, the extensive use of English in a various range of settings, domains and purposes calls into question our perception of the ownership and norms of the English language use. Therefore, a nation-state based approach to culture and language, which involves key tenets of monolingual speakers, homogeneous speech community and monolithic views of standard national languages, is no longer appropriate to demonstrate the current sociolinguistic situation of English use. The traditional approach to English language teaching (ELT), whose main goal is to master native-like syntactic, lexical and phonological features of language, is also irrelevant for many L2 speakers who learn and use English for intercultural communication or English as a lingua franca (ELF), because their top priority of learning a language is for effective communication and exchange of messages and information. Therefore, nowadays L2 speakers of English are required to develop the ability to use diverse communicative resources and adapt to a range of communicative situations in the course of interaction through the negotiation of meanings and mediation skills (Kang and Lee 2012, Kim and Chung 2016, D Angelo 2017). It is commonly accepted that culture and language are closely intertwined each other (Kramsch 1993, 1998, Risager 2006). However, the cultural dimension to language has still stayed as a national concept. For example, when we talk about English and culture, a particular linguistic and cultural code such as British English or US culture is often referred as a standard frame of reference. Even though the English language includes to some extent its own culture and conveys particular values and beliefs, today English has been more used as a means of intercultural communication in a global context rather than simply as one of foreign languages used in one specific country. Therefore, an interrelation between the English language and a specific culture based on a particular native speaker variety has been criticized as a problem (Holliday 2010, 2011, Joseph 2004). 31

36 Kanghee Lee Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF This study will revisit the conceptualization of culture and the relationship between culture and language and some theoretical concepts as regards these issues such as third place, transcultural flows, critical cultural awareness, and intercultural awareness. This paper will also include the discussion on how the notion of culture and the relation between language and culture are perceived in ELF communication and intercultural communication and how culture and intercultural awareness can be presented and approached in ELT. Understanding a complex and multifaceted nature of culture in intercultural communication will provide useful pedagogic implications for classroom teaching and language teachers in ELT. 2. The Relationship between Culture & Language Language is viewed as the key semiotic vehicle of characterizing and constructing culture (Greez 1973, Halliday 1979). Language allows people to create a range of types of culture, and socialization and learning in society take place through language. As language use and learning occur in socio-cultural contexts, the relationship between language and context is significantly intertwined. However, language does not govern people s beliefs, values and perspectives in a restricted way. As Baker (2011) puts it, language certainly influences our perception of the world but it does not restrict it (p. 198). In other words, speakers are able to view the world in various ways, and perceptions in culture are less likely to be confined with a specific language or variety of language. Therefore, it is problematic and extremely simplistic to explicate the relationship between language and culture in national terms, because the current sociolinguistic situation of language use has become more dynamic and complicated, and the concept of culture flows across local, national, and global contexts. When it comes to the intercultural communication through ELF in global contexts, where cultural diversity and fluidity are pervasive, this simple and linear view on culture and language is more problematic. ELF communication takes place in multilingual and 32

37 Kanghee Lee Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF multicultural settings, and the diversity and variability are common in terms of speakers lingua-cultural backgrounds, contexts of use, domains and modes of communication. Even different cultures exist according to genders, generations, occupations, races and religions. All these factors have their own cultures, and consequently elements that constitute a culture do not limit to the national based notion but are more likely to vary according to participants, contexts and modes of interaction. However, more importantly, language can never be culturally neutral, even when it is used as a lingua franca, where functional aspects of use are highlighted. Although some researchers view ELF as culturally neutral (House 2014, Kirkpatrick 2007, Meierkord 2002), communication is inevitably embedded in socio-cultural settings, and the language in intercultural communication like any other communications always involves people, places, and purposes, none of which exist in a cultural vacuum (Baker 2012a, p. 64). The concepts of ideology and identity also make language not culturally neutral. As culture is perceived as an ideological process, ideological dimensions to language have been highlighted in intercultural communication with regard to cultural identity and identification (Holliday 2011). Byram (2008) delineates an intercultural citizenship, in which speakers experience intercultural communications across less constrained cultural grouping, as a more idealized identity that speakers need to develop for successful intercultural communication. As intercultural speakers perceive themselves as members of communities of practice in multilingual and multicultural communication, they can make an effective negotiation and mediation of different cultures and languages in intercultural communication. The notion of intercultural citizenship can provide a more relevant and attainable learning model for L2 speakers. In other words, L2 learners and users of English are given an identification which recognizes the importance of their L1 and C1 (first culture) and their resources as bilingual communicators (Baker 2012b, p. 30). By emphasising multilingualism and multiculturalism, intercultural citizenship can help L2 speakers raise awareness on the fluid and emergent nature of correlation between culture, language and identity which are crucial 33

38 Kanghee Lee Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF for intercultural communication in global contexts. In many contexts of use, English has provided L2 speakers with a means of formulating and shaping more fluid and dynamic new identities along with the relatively stable identities of L1/C1. The relationship between culture and language has become more fluid and dynamic in ELF. In other words, ELF takes place in the sociocultural circumstances which are emergent, fluid and situated rather than in fixed, bounded and confined settings, because linguistic and cultural resources in ELF communication continue to move between and across local, national, and global contexts. In this respect, the notion of a third place in L2 communication provides a highly pertinent explanation on the cultural space of language use (Kramsch 1993, p. 233). The concept of third place stresses the L2 speakers capability to mediate and relativize cultures. According to the model, L2 communication operates in a third place, which is another space dominated by neither the L2 speaker s first language (L1) and culture (C1) nor the target language (L2) and culture (C2). In other words, L2 communication is processed along a cultural faultline (p. 25) where communicative practices are governed by the norms of neither L1/C1 nor L2/C2 but new cultural practices and forms are created in intercultural settings. In the contexts of intercultural communication, specific languages and cultures are less likely to provide cultural assumptions and frames of reference, because the language used as a lingua franca is no longer the property of any specific culture or country, but the ownership of the language belongs to all the participants involved in the interaction. When L2 speakers are involved in the intercultural communication in which two or more languages and cultures operate, they continue to make an effort to achieve mediation and negotiation between their own language and culture and those of others. Pennycook (2007) also supports the dynamic fluidity of language and culture, focusing on the case of English. He draws the notion of transcultural flows and argues that English plays a major role in global flows of culture and knowledge in multiple domains of globalization (p. 19). Pennycook views that English has both fluid and fixed nature due to its translocal and transnational 34

39 Kanghee Lee Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF movement, and therefore English involves both localities and correlations in wider social contexts. In other words, as English has spread internationally, linguistic forms and cultural practices of English continue to move and flow across national borders and different communities, and new localized forms of practices are created. The linguistic forms and cultural practices continue to be modified, negotiated and adapted according to the context of use, purposes of communication and interlocutors, and the speakers identities are refashioned in different contexts through the process of borrowing, mixing, observing and revisiting. Pennycook highlights that when the relationship between language and culture is demonstrated with the notion of globalization, the dichotomic view in culture and language is no longer appropriate, that is, the simple distinction between global or local, imperialism or pluralism, and homogeneity or heterogeneity. In other words, the interaction between the global and the local is not a one-way process but both are mutually influenced, and English as a means of intercultural communication has both fluidity and fixity, where language and culture not only move across space, borders, communities, nations but also become localized, indigenized, re-created in the local (Pennycook 2007, p. 7). 3. Cultural Awareness & Intercultural Awareness For successful intercultural communication, speakers need to recognize the significance of the cultural aspect of communicative competence, that is, cultural awareness (CA), which refers to a conscious understanding of the role of culture in language learning and use (Baker 2012a, p. 65). The notion of CA emphasizes the need for language learners and users to have awareness on the culturally based norms, beliefs, and behaviors of their own culture and other cultures (Baker 2012a, p. 65). Speakers need to understand culture as a set of shared behaviors, beliefs, and values as well as the role that culture and communicative context play in interpretation and negotiation of meaning. In 35

40 Kanghee Lee Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF addition, language users need to be aware of the relative nature of cultural norms in interaction and the fact that cultural understanding is temporary and open to revision. As individual speakers are members of many different social groups, multiple voices and perspectives are possible within any cultural groupings and boundaries. Byram (1997) provided the notion of critical CA which is involved in an understanding of the relative nature of cultural norms which leads to an ability to evaluate, critically and on the basis of explicit criteria, perspectives, practices and products in one's own and other cultures and countries (1997, p. 53). The concept of critical CA stresses the understanding of the multi-voiced nature of culture which contains conflicting and contradictory views (Baker 2011, p. 200). From the CA perspectives, the norm of monolingual native speakers is not necessarily the most idealized model in ELT in all contexts of learning, and alternatively the notion of intercultural speaker (Byram 1997, p. 31) is proposed as a more relevant model for L2 speakers. According to critical CA, identity and group membership (affiliation) are important elements in understanding the process of mediation and negotiation of meanings in intercultural communication, where one particular participant cannot provide the norms or target model with which the other interlocutors should comply. More importantly, speakers can expand and develop knowledge, skills, and attitudes of CA as they experience diverse intercultural communication and understand both specific and various cultures and languages. While CA has provided an account of the value and importance of culture in communication, the notion is also constrained to describing cultural groups and practices at the national level from the comparative cross-cultural perspectives. In other words, CA is commonly concerned with comparing one culture and another or influences of a specific culture on another or other cultures. However, this approach is problematic and inappropriate in the current multilingual and multicultural environment, where diversity and heterogeneity are pervasive in language use, because a user or learner of English could not be expected to have a knowledge of all the different cultural contexts of communication they may encounter and even less so the languacultures of the 36

41 Kanghee Lee Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF participants in this communication (Baker 2012a, p. 65). Consequently, whereas a great part of accounts in CA might be helpful in understanding cultural difference and relativization, they need to move beyond the nation-based understanding of culture in intercultural communication, where there is no one specific norm of culture but culture can move, adapt and combine in a more dynamic and fluid way. In other words, cultural influences in intercultural communication and ELF tend to be varied, fragmented and emergent as hybridity is pervasive in interaction in this context and it is constantly involved in dynamic progress with no limited end point. Therefore, the notion of intercultural awareness is suggested as a more relevant supplement for intercultural communication. Intercultural awareness (ICA) is defined as a conscious understanding of the role culturally based forms, practices, and frames of reference can have in intercultural communication, and an ability to put these conceptions into practice in a flexible and context specific manner in real time communication (Baker 2011, p. 202). Under the intercultural awareness, language users are aware that culturally attributed frames of reference, forms and communicative practices are concerned with both specific cultures and hybrid and emergent elements in cultural contact in intercultural communication. Baker (2011, 2012a) presented a comprehensive model of ICA which describes diverse elements of ICA and correlations among them. The model makes a distinction between conceptual ICA and practice-oriented ICA. Conceptual ICA is associated with the types of attitudes to and knowledge of culture that speakers need to participate in intercultural communication and the ability to express these attitudes and knowledge. Practice-oriented ICA is related to how cultural perspective, knowledge and conscious understanding are applied in real-time communication and therefore emphasizes communicative skills of negotiation and mediation of meanings. Whereas practice-oriented ICA is concerned with abilities and skills, they rely on knowledge, ideas, and attitudes developed in conceptual ICA. According to the model, in the initial stage of interaction in intercultural communication, speakers tend to have a general awareness of the role of 37

42 Kanghee Lee Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF cultures on their own and others and can compare cultural differences and articulate their own cultural perspective. As speakers start to have more advanced cultural awareness on the complexity of cultures, they understand the relative nature of cultural norms and perceive individuals as members of cultural groups. They can develop multiple voices or perspectives within cultural grouping and discover common ground between specific cultures and possibilities for misunderstanding and miscommunication between different cultures. On the final level of intercultural awareness, speakers are able to have awareness of culturally based frames of reference, forms and communicative practices as emergent and dynamic in intercultural communication. Speakers may draw cultural stereotypes or use nation-based cultural generalization at the initial level of intercultural awareness but they can develop an ability to go beyond these biased perspectives through the process of negotiation and mediation in interaction and finally view cultures as complex and fluid rather than bounded and fixed entities (Baker 2011, p. 203). Rather than focusing exclusively on one specific culture such as US culture or British culture, cultural contents in ELT should foreground how participants engage in culture-related issues or topics in intercultural communication and how ELF speakers cope with these encounters and achieve successful negotiation of meaning. ELF speakers need to recognize fluid boundaries of language and culture and expand insights into diversity and potential change in communicative practices and cultural references. Intercultural awareness can encourage speakers to understand and negotiate the complexity of culture in intercultural communication, which is less likely to involve a priori identified cultural groups and affiliation but seeks to co-construct shared meaning and mutual understanding. 4. Culture in ELF Although language is closely related to culture, and a specific language can 38

43 Kanghee Lee Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF reflect and represent a distinctive culture, it is overly simplistic to approach and understand the relationships between language and culture in national boundaries, as mentioned earlier. In particular, when a language is used as a lingua franca such as English today, there is no point to name and categorize a native speaker culture. As there is a growing level of multitude and diversity of English in terms of users, domains and contexts of use, there can be no one culture of English (Baker 2009, Jenkins 2007, Seidlhofer 2011). When English is used as a means of intercultural communication, the relations between culture and language have become emergent and fluid, and the boundaries of one language and culture and another are less likely to be strictly distinguished. Such fluid and complex nature of culture in intercultural communication through ELF is supported by a number of studies. For example, Risager (2007) argues that the national paradigm of language and culture needs to move towards a transnational paradigm (2007, p. 222) where language users participate in intercultural communication as a member of a wider global community of practice, and Kramsch (2009) also highlights an approach to language and culture from multilingual and multicultural perspectives where dynamic, flexible and locally contingent communicative competence and practices are important in intercultural communication (p. 200). Rampton s (1995) notion of liminality also delineates that the nature of culture in ELF cannot be represented by one specific cultural practice or forms. Rather, linguistic and cultural practices can have new forms and meanings according to different situations and contexts of use, particularly in intercultural communication such as ELF which is not affiliated in any specific culture or community. Therefore, culture can be perceived as dynamic and fluid resources in intercultural communication that emerge in-situ as more or less relevant to creating understanding (Baker 2011, p. 200), and it might be better to approach and understand culture as a verb rather than a noun which overtly contains static connotations (Roberts et al. 2001). Risager (2006) makes a distinction between the relationship of language and culture in the generic sense and in the differential sense to account for the 39

44 Kanghee Lee Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF complex nature of cultures. In the generic sense, language and culture are interwoven, since languages always express, embody and represent cultural reality. However, in the differential sense, where specific languages and cultures are discussed, language and culture can be separated. All languages can have new cultural meanings, which is referred to as languacultures (Agar 1994), according to speakers, contexts and purposes of communication. Consequently, a language can have as diverse languacultures as the number of speakers, and cultures cannot be clearly identified and distinguished. Consequently, to demonstrate the interrelation between language and culture in ELF, we need to approach the notion of culture from a more complex perspective than culturally neutral or deterministic approach (Baker 2015a). In other words, culture operates under a constant tension between fluidity and fixity (Pennycook 2007) where more traditional normative conceptualization of culture coexists with more emergent and situated cultural practices. Therefore, cultural frames of reference in ELF are more likely to be hybrid, diverse, and multiple, and the relationship between language and culture in ELF cannot be defined and identified in a priori approach without viewing each specific case of interaction. In this respect, Baker (2015a) proposes conceptualizing ELF as transcultural communication rather than intercultural because it is difficult to identify and distinguish what cultures engage in ELF (Baker 2015a, p. 14). The notion of transcultural communication entails less static perspectives of culture where communication takes place through and across different cultures rather than between cultures as in intercultural communication. One of the fundamental principles of understanding culture as a complex system is that social practices are always understood in a partial and situated manner. However, at the same time the social system cannot be understood by dividing it into each individual component, but the interrelations between each component is crucial for understanding the whole system. As the complexity approach emphasizes the interplay between the emergence of structure on one hand and process or change on the other (Larsen-Freeman 2011, p. 52), pre-existing rules and norms, which are currently regular and stable, can be changed and modified as a result of observation and discovery of new rules 40

45 Kanghee Lee Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF and patterns. Sealey and Carter (2004), and Taylor (2001) also support the intersubjectivity of culture and the relationship as an integral part of the system. They argue that cultural elements exist in interrelation to each other and through constant interactions between individual components. From the perspective of complexity theory, Baker (2015a) provides a conceptualization of cultures as follows: Equally importantly given the high degree of individual variation, cultures are constantly in change, with new beliefs, values, attitudes, and practices becoming socially sedimented while other ones fall out of use. At the same time, this social sedimentation provides a degree of stability. However, the constant changes that are part of the system mean a full account is never possible and the system is in a constantly emergent state with no fixed end point. (p. 16) The complexity approach provides not only a theoretical and analytical framework for understanding culture in ELF as a contingent and emergent system but also an opportunity to recognize unpredictability and uncertainty as natural and common phenomena in ELF. From the complexity perspective, we can eschew a dichotomic approach between our culture and other cultures by foregrounding interrelations and dynamic flows of cultural systems. In other words, cultures as a complex system correlate and affect each other and consequently there are no clear boundaries between one system and another. Given that cultures in ELF are enacted in a hybrid, mixed, and liminal manner, drawing on and moving between global, national, local, and individual orientations, cultural forms, practices, and frames of reference in ELF need to be seen as adaptive and emergent resources which are negotiated and context dependent rather than as a priori defined products (Baker 2009, p. 567). 41

46 Kanghee Lee Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF 5. Culture and ELT The lack of integration of culture into ELT pedagogy is criticized by many applied linguists and ELT practitioners (Cortazzi and Jin 1999, Leung 2005, Vettorel 2010, Young and Sachdev 2011). Despite the importance of culture and intercultural awareness in language learning and use, the cultural approach to ELT has been still relegated in teacher training, teaching materials, curriculum development and language assessments. For this reason, culture or intercultural communicative competence has been treated as the additional fifth skill compared to other language skills such as reading, writing, listening and speaking (Tomalin 2008). However, cultural awareness and intercultural competence are an integral part of intercultural communication, and therefore the emphasis has been placed on the significance of cultural dimensions to language teaching (Kramsch 1993, 1998) and a more extensive understanding of the diverse cultural contexts of English use (Porto 2010, Suzuki 2010). The emphasis on the concept of languaculture (Agar 1994), which views language as a cultural practice and therefore the cultural dimensions of language is stressed on language teaching, might be the starting point. One of the crucial tasks of ELT is to help students aware of languacultural diversity and fluidity, including the fact that any speaker develop languacultures throughout the contact and experience of language use and communication. To help students develop languacultural variability in practice, students need to work with a range of activities in classroom learning. For example, teachers can inform students of the current changing sociolinguistic situations in English language use, and can provide texts on relevant topics such as multilingualism, multiculturalism and intercultural communication. The actual intercultural communications can be shown to students for the further group discussion, and the activity can be supplemented with the culture-related vocabulary and sentence-completion tasks to develop semantic and pragmatic languaculture. To make students understand various cultural contexts of English use, teachers can illustrate Kachru s three circle model by demonstrating the notion of the 42

47 Kanghee Lee Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF inner circle, the outer circle, and the expanding circle and show where and how English is used today. The fact that English is no longer a foreign language used in the native speaker context but acts as a global language for intercultural communication can encourage students to expand and develop their intercultural awareness. There have been some attempts to apply the cultural approach to classroom teaching. For example, Galloway (2013) incorporated CA and ICA in her teaching program in a Japanese university, drawing on issues such as global Englishes, ELF, and various socio-cultural contexts of the English use. After the program, students expressed the awareness of multilingual and multicultural dimensions of the English communication and presented positive attitudes to diversity and flexibility in ELF and intercultural communication. The similar result was observed in Baker s (2011) study. Baker incorporated the topics of global Englishes and ICA into his teaching course in a Thai university including the relationship between language and culture, the role of English as a global language for intercultural communication and the diversity and fluidity of culture in communication. Both teachers and students demonstrated the awareness on the fluid and dynamic nature of culture in intercultural communication and the significance of the mediation and negotiation of cultural practices for successful intercultural communication. When language is understood as a cultural practice, awareness on complexity, fluidity, and flexibility is key in intercultural communication. To overcome complexity and achieve flexibility, students need to develop intercultural communicative competence beyond linguistic competence. The key to promoting cultural dimensions of the knowledge, skills and attitudes for successful intercultural communication is more exposure to the intercultural communication situations (Baker 2009, Byram, Nichols and Stevens 2001, Roberts et al. 2001). Intercultural awareness and intercultural communicative competence can be integrated and developed into the classroom teaching practice by providing opportunities for cultural issues and experiences of intercultural contacts. Firstly, learners need to be provided with a range of opportunities to understand the relations between culture, language and communication. For 43

48 Kanghee Lee Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF instance, learners can examine their own culture and the complexity of local cultures as a starting point ahead of understanding other cultures. Learners can discuss the views on their own and other cultures and languages, which may lead to understanding how the knowledge, skills and attitudes can be influenced by cultural contexts, and introducing the notion of cultural diversity and fluidity in intercultural communication can help learners raise awareness on the complex and emergent nature of culture and language. In addition to intercultural awareness, students can develop communicative competence by observing how speakers interact in the actual intercultural communication or ELF and practicing a variety of pragmatic strategies and interactional skills to enhance mutual understanding and effectiveness for intercultural communication. Furthermore, through the analysis of how cultural images are described and presented in a variety of language learning materials and media, learners can develop critical and reflexive perspectives on stereotypical and fragmented cultural images. Learners need to recognize and understand that cultural information and sources can be subjective and partial, and cultural diversities and complexities can be negotiated and mediated between different cultures. By engaging in intercultural communication both face-to-face and online and having more discussions on diverse cultural issues, learners can develop critical awareness on and understanding of cultures in intercultural communication and build adaptive skills and flexible attitudes for successful intercultural communication. Most importantly, L2 learners and teachers of English need to move beyond the understanding of culture as a national concept and recognize wider plurilingual and transcultural ideas of competence, performance, and awareness (Baker 2015a, p. 23). From the ICA perspective, L2 learners need to recognize flexible and context-specific nature of the knowledge, skills and attitudes. 44

49 Kanghee Lee Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF 6. Conclusion Given that English language is no longer viewed as the property of native speakers of the language, the traditional approach to the relationship between language and culture is not appropriate and applicable to the context of intercultural communication through ELF. In other words, the interrelation between language and culture in ELF, which occurs in multilingual and multicultural contexts of communication, is less likely to be static, linear, and homogeneous but more complex and difficult to clearly identify and define. Cultures in ELF need to be perceived as hybrid, emergent, and dynamic entities which are in a constant state of fluidity and flux between local, national and global resources (Baker 2009, p. 568), and new cultural forms, practices, and frames of reference can be created and employed in each situation of intercultural communication. A number of crucial concepts in relation to culture in intercultural communication such as intercultural awareness, intercultural speaker, and intercultural citizenship overtly reject the national paradigm of language and culture and instead accept a transnational paradigm (Risager 2007, p. 222) where speakers participate in intercultural communication as a member of global communities of practice. Accordingly, the concept of culture needs to be approached from multilingual and multicultural perspectives where fluidity, flexibility, and contingency are prevalent, and what is key to understand culture in intercultural communication through ELF is the process of communication rather than the end product (Baker 2011, p. 201). The relationship between language and culture has become emergent and dynamic in the context of intercultural communication through ELF, and boundaries of one language and culture and another have been less strictly demarcated in the current multilingual and multicultural world. Therefore, the view of a specific language and national approach to culture has lost its legitimacy. Instead, language and culture are more likely to modify, adapt and shift according to the need of the speakers and contexts of use. Consequently, 45

50 Kanghee Lee Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF when English is used as a lingua franca in intercultural communication, it is in constant tension between individual, local, regional and global contexts, all of which need to be approached as dynamic and changeable (Baker 2011, p. 199). Therefore, intercultural communication needs to be perceived as a constantly changing and negotiated social process where the cultural dimension is crucial. Accordingly, the role of culture and socio-cultural contexts of use is significantly essential to understand language processes and practices in intercultural communication through English, and cultural and intercultural awareness is as important as acquiring a knowledge of grammar, vocabulary, and phonology in language. In addition, the emergent, dynamic and complex nature of culture in ELF would propose that there is no explicit target culture which speakers aim to acquire and comply with. ELF speakers need develop an ability to apply intercultural awareness to communicative practices and negotiation and mediation skills to effectively manage the diversity and complexity in intercultural communication. Reference Agar, M Language Shock: Understanding the Culture of Conversation. New York: William Morrow. Baker, W The cultures of English as a lingua franca. TESOL Quarterly 43(4), Baker, W Intercultural awareness: Modelling an understanding of cultures in intercultural communication through English as a lingua franca. Language and Intercultural Communication 11(3), Baker, W. 2012a. From cultural awareness to intercultural awareness: Culture in ELT. ELT Journal 66(1), Baker, W. 2012b. Global cultures and identities: Refocusing the aims of ELT in Asia through intercultural awareness. In T. Muller, S. Herder, J. Adamson and P. S. Brown, eds., Innovating EFL Education in Asia, Basingstoke: Palgrave Macmillan. Baker, W. 2015a. Culture and complexity through English as a lingua franca: 46

51 Kanghee Lee Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF Rethinking competences and pedagogy in ELT. Journal of ELF 4(1), Baker, W. 2015b. Research into practice: Cultural and intercultural awareness. Language Teaching 48(1), Block, D Globalisation and language teaching. ELT Journal 58(1) Byram, M Teaching Assessing Intercultural Communicative Competence. Clevedon: Multilingual Matters. Byram, M., From Foreign Language Education to Education for Intercultural Citizenship: Essays and Reflections. Clevedon: Multilingual Matters. Byram, M., A. Nichols and D. Stevens Developing Intercultural Competence in Practice. Clevedon: Multilingual Matters. Cortazzi, M. and L. Jin, Cultural mirrors: Materials ad methods in the EFL classroom. In E. Hinkel, ed., Culture in Second Language Teaching and Learning, Cambridge: Cambridge University Press. D Angelo, J Applying the WE Enterprise to expanding circle ELT. Korean Journal of English Language and Linguistics 17(4), Galloway, N Global Englishes and English language teaching (ELT): Bridging the gap between theory and practice in a Japanese context. System 41(3), Greez, C The Interpretation of Cultures. New York: Basic Books. Halliday, M. A. K Language as Social Semiotic. Victoria: Edward Arnold. Holliday, A Cultural descriptions as political cultural acts: An exploration. Language and Intercultural Communication 10(3), Holliday, A Intercultural Communication and Ideology. London: Sage. House, J English as a global lingua franca: A threat to multilingual communication and translation?. Language Teaching 47(3), Jenkins, J English as a Lingua Franca: Attitude and Identity. Oxford: Oxford University Press. Joseph, J Language and Identity: National, Ethnic, Religious. Basingstoke: Palgrave Macmillan. Kang, S.-S. and S.-K. Lee A survey of Korean English teachers perceptions and attitudes about the common features of English as a lingua franca : Focusing on the lexical and grammatical features. Korean Journal of English Language and Linguistics 12(3), Kim, Y.-K. and H. Chung English as a lingua franca in the Asian context: Indicating and responding to non-understanding in NNS discourse. Korean Journal of English Language and Linguistics 16(2), 47

52 Kanghee Lee Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF Kirkpatrick, A., World Englishes: Implications for International Communication and English Language Teaching. Cambridge: Cambridge University Press. Kramsch, C Context and Culture in Language Teaching. Oxford: Oxford University Press. Kramsch, C Language and Culture. Oxford: Oxford University Press. Kramsch, C The Multilingual Subject. Oxford: Oxford University Press. Larsen-Freeman, D A complexity theory approach to second language development/acquisition. In D. Atkinson, ed., Alternative Approaches to Second Language Acquisition, Abingdon: Routledge. Leung, C Convivial communication: Recontextualizing communicative competence. International Journal of Applied Linguistics 15(2), Meierkord, C Language stripped bare or linguistic masala? Culture in lingua franca conversation. In K. Knapp and C. Meierkord, eds., Lingua Franca Communication, Frankfurt: Peter Lang. Nichols, M. and D. Stevens Developing Intercultural Competence in Practice, Vol. 1. Multilingual Matters. Pennycook, A Global Englishes and Transcultural Flows. London: Routledge. Porto, M Culturally responsive L2 education: An awareness-raising proposal. ELT Journal 64(1) Rampton, B Crossing: Language and Identity among Adolescents. London: Longman. Risager, K Language and Culture: Global Flows and Local Complexity. Clevedon: Multilingual Matters. Risager, K Language and Culture Pedagogy. Clevedon: Multilingual Matters. Roberts, C., M. Byram, M. A. Barro, S. Jordan and B. Street, Language Learners as Ethnographers: Introducing Cultural Processes into Advanced Language Learning. Clevedon: Multilingual Matters. Sealey, A. and B. Carter, Applied Linguistics as Social Science. London: Continuum. Seidlhofer, B Understanding English as a Lingua Franca. Oxford: Oxford University Press. Suzuki, A Introducing diversity of English into ELT: Student teachers responses. ELT Journal 65(2), Taylor, M The Moment of Complexity: Emerging Network Culture. Chicago & London: University of Chicago Press. 48

53 Kanghee Lee Understanding Culture and Intercultural Awareness in Intercultural Communication through ELF Tomalin, B. and S. Templeski Cultural Awareness. Oxford: Oxford University Press. Tomalin, B Culture-the fifth language skill. Teaching English 48(1), Vettorel, P EIL/ELF and representation of culture in textbooks: Only food, fairs, folklore and facts? In C. Gagliardi and A. Maley, eds., EIL, ELF, Global English: Teaching and Learning Issues, Bern: Peter Lang. Yong, T. and I. Sachdev Intercultural communicative competence: Exploring English language teachers beliefs and practices. Language Awareness 20(2), Examples in: English Applicable Languages: English Applicable Level: Secondary and Tertiary Kang-Hee Lee 2639 Sejong-ro, Jochiwon-eup, Sejong 30016, Korea Institute of General Education, Hongik University TEL: 044) Received: October, 2017 Revised: January, 2018 Accepted: March,

54 Korean Journal of English Language and Linguistics Vol. 18, No. 1, Spring 2018, /kjell Remotivation Strategies for Korean Male High School Students English Learning * Tae-Young Kim (Chung-Ang University) ** Ji-Young Kim (Chung-Ang University) Kim, Tae-Young and Ji-Young Kim Remotivation strategies for Korean male high school students English learning. Korean Journal of English Language and Linguistics 18-1, This study examines the EFL learning remotivation components for male Korean high school students who showed improvement from a state of EFL learning demotivation and low in-house English test scores. The participants were four male second-year high school students who had been remotivated in EFL learning and enhanced low English test scores in junior high school. To explore the experiences specific to EFL learning remotivation, this study adopted a qualitative approach. The participants completed open-ended questionnaires and took part in a series of semistructured interviews, one for the preliminary study and two for the main study. The results revealed two remotivation components: instrumental motivation to obtain admission to a university and competitive motivation with peers. The components, motivated behavior and self-efficacy, enhanced English proficiency. Thus, when EFL learning motivated behavior related to remotivation components is strengthened, it increases English proficiency, enabling students to experience a high level of self-efficacy, which in turn helps maintain higher English proficiency. This qualitative research implies that EFL learners perception of remotivation components is a key determinant of L2 proficiency when combined with their motivated behavior and self-efficacy. Keywords: EFL learning remotivation strategies, changes in EFL learning motivation, high school students, qualitative inquiry * This research was supported by the Chung-Ang University Research Scholarship Grants in ** Corresponding Author 50

55 Tae-Young Kim Ji-Young Kim Remotivation Strategies for Korean Male High School Students English Learning 1. Introduction English as a foreign language (EFL) learning in South Korea (henceforth Korea) has played a crucial role in society for ensuring academic development and successful careers. The social contexts in Korea have led people to focus more on how to improve English proficiency (Park 2009). Moreover, in the early stages of EFL learning in junior high and high schools, English proficiency depends on students' intrinsic and extrinsic motivational components (Lee 2002). Students experience motivational changes caused by internal components such as their emotions and personality traits, as well as external components such as teachers, peers, and coursework (Atkinson 2014, Dörnyei 2009a, Kikuchi 2015). L2 learning motivation fluctuates between the state of demotivation and remotivation due to various factors in the L2 learning context (Dörnyei 2001a, Song and Kim 2017). When experiencing L2 learning demotivation, students gradually lose their interest in L2 learning due to either internal and external reasons (Dörnyei and Ushioda 2013, Kikuchi 2015). However, they become interested again in L2 learning in state of remotivation (Falout 2012). In the EFL curricula of public education, motivation changes in students, especially junior high and high school students, should be recognized in order to implement effective strategies for enhancing their English proficiency. As students start junior high school, their motivation in EFL learning declines (Kikuchi 2013); this continues through high school (Kang 2014, Kim 2012a) because the competitive atmosphere makes them feel stressed about preparing for university admission. Nevertheless, students in Korea tend to think that EFL learning is necessary for getting a better grade and entering a high-ranking university (Park 2009). When high school students consider English useful for their future jobs, they could overcome their EFL learning demotivation and achieve a higher level of EFL learning motivation (Song and Kim 2017). Against this backdrop, to raise proficiency and motivation in EFL learning, it 51

56 Tae-Young Kim Ji-Young Kim Remotivation Strategies for Korean Male High School Students English Learning is necessary to observe changes in the dynamic relationship between EFL learning motivation and English proficiency by tracking the EFL learning experience of Korean high school students. Therefore, this study examined four high school students who showed improvements from low EFL learning motivation and low English test scores in junior high school. Interviews with EFL learners provide knowledge of their specific experiences, providing insight into their learning processes (Brinkmann and Kvale 2015). This study focused on male students, taking account of challenges male students face in educational context. Previous studies (e.g., Kim and Kim 2011) have shown that male students in secondary school in Korea exhibit a lower level of EFL-learning motivation compared with their female counterparts. In addition, boys exhibited less adaptive attitudes in learning contexts due to disengagement-oriented strategies (Bonneville-Roussy et al. 2017). Understanding which components encourage EFL male learners to maintain EFL learning motivation and high English test scores is useful to improve low state of motivation and English test scores. The current paper is an exploratory investigation toward the remotivation components in EFL learning by adopting a series of qualitative interview methods. The following research questions were addressed: 1) What are the remotivation components in EFL learning perceived by Korean high school students? 2) What components induce Korean EFL high school students to activate remotivation, resulting in improved regular in-house English test scores? 2. Literature Review 2.1 Process of L2 Learning Motivation Given that motivation is a psychological construct, it is necessary to examine 52

57 Tae-Young Kim Ji-Young Kim Remotivation Strategies for Korean Male High School Students English Learning relevant discussions from a sociopsychological perspective first. Motivation is derived from psychological contexts and social interactions amid individuals, inducing them to achieve their own goals (Nolen-Hoeksema, Fredrickson, Loftus and Wagenaar 2009). In the field of education, motivation as a form of conscious and intentional human learning is a crucial component of academic performance and achievement (Brophy 2010, Ushioda 2013). Second language (L2) learning motivation exhibits fluctuation and complexity depending on each learner's specific learning experiences, and it substantially affects L2 learning behaviors (Atkinson 2014, Dörnyei 2009b, 2014, Kikuchi and Sakai 2009). Dörnyei (2001a, 2009b) indicated that motivational fluctuations occur in an interplay between internal and external components. Likewise, Atkinson (2014) described a sociocognitive approach integrating mind, body, and the world into L2 learning development. This approach emphasizes the dynamic and multifaceted interactions between internal and external components in EFL learning. The internal processing of emotions and cognition is associated with learners, and external components are related to social influences and interactions. When combined, such components interact with the learning environment and affect L2 learning motivation (Ushioda 2009). When learning an L2, learners undergo feelings of frustration, anxiousness, fulfillment, and satisfaction, by either completing tasks or interacting with others (Ehrman and Dörnyei 1998). Learners receive supportive feedback from teachers and peers while using available materials, which positively affect their English learning motivation. Thus, diverse internal and contextual components including educational contexts, learner personality traits, and cognitive ability influence learning effectiveness and change the motivations for L2 learning (Dörnyei 2009a, MacWhinney 2001). 2.2 Components of EFL Learning Remotivation In EFL learning studies, some researchers have focused particularly on demotivation components, which are the negative forces that render learners reluctant to learn EFL (Falout, Elwood and Hood 2009, Kikuchi 2015, Kikuchi 53

58 Tae-Young Kim Ji-Young Kim Remotivation Strategies for Korean Male High School Students English Learning and Sakai 2009, Trang and Baldauf 2007). Demotivation is defined as various negative influences that cancel out existing motivation (Dörnyei 2001b, p. 142). Previous studies have explored the demotivators in EFL learning in order to prevent decreasing motivation. Moreover, researchers have believed that such studies may facilitate finding ways to recover motivation (Falout 2012, Falout et al. 2009, Kikuchi 2015, Kikuchi and Sakai 2009, Trang and Baldauf 2007). Moreover, researchers have gradually focused on not only demotivation but also remotivation. Falout defined remotivation as a process of recovering motivation after losing it (2012, p. 3). From this perspective, researchers have sought remotivational strategies for motivating learners to study an L2 (Carpenter, Falout, Fukuda, Trovela and Murphy 2009, Dörnyei 2001a, Falout 2012, Hamada 2014, Jung 2011). As students start junior high school, their motivation in EFL learning tends to decline (Kikuchi 2013, Kim and Lee 2013, Song and Kim 2017); this trend continues through high school (Kang 2014, Kim 2012a) because the competitive atmosphere makes them feel stressed from preparing for university entrance. Kikuchi and Sakai (2009) investigated high school experiences of 112 university students. They identified five primary external demotivation factors: teachers competence and teaching style, course books, school facilities, test scores, and communicative methods. Hamada (2014) examined 336 Japanese high school students and identified five main factors that remotivate junior high and high school students who are demotivated to learn EFL in a competitive educational atmosphere: teacher s sensitivity, students positive feelings, English usage, traditional teaching style, and goal orientation. Ma and Cho (2014) investigated demotivating and remotivating factors in Korean EFL contexts through a qualitative method. They collected retrospective essays on EFL learning from Korean university students. Their findings suggested that the enhanced value of learning EFL as an internal factor and the need to raise English test scores as an external factor helped students to be remotivated to learn English. Kim and Kim (2017) analyzed retrospective data on high school demotivators and remotivators from 54

59 Tae-Young Kim Ji-Young Kim Remotivation Strategies for Korean Male High School Students English Learning 130 university students using questionnaires. They revealed that while the dominant demotivating factor was difficulty of learning English, which had a negative effect on English test scores, the students were remotivated by the importance of English scores for university admission Most EFL learning demotivation and remotivation studies have used Likert-scale or open-ended questions to analyze the factors that affect demotivation and remotivation. Motivation is a psychological and cognitive component which fluctuates. In order to study motivational change, qualitative research methods such as in-depth interviews addressing participants learning experiences would be appropriate (Dörnyei 2001b, Han and Ha 2006, Kim 2010b, 2012b). Based on previous studies, this study aims to investigate the experience of EFL learning remotivation of male students through interview analysis. 3. Method 3.1. Participants The participants in this study were four male second-year high school students from Gyeonggi-do, Korea (in the vicinity of Seoul). The researchers recruited the participants demonstrating higher English test scores in high school than in junior high school. All four students not only attended the same junior high and high schools but also lived in the same neighborhood. The high school they attended focused on preparing students for college entrance. For the purpose, its students generally aimed to acquire academic skills for university entrance. They came from households where both parents worked, and each had a sibling. They had been attending a private institute, or hagwon, for five years from junior high school to the present. This study adopted homogeneous samples of individuals based on purposeful sampling procedures, in which participants share specific experiences (Creswell 2013). Because it is 55

60 Tae-Young Kim Ji-Young Kim Remotivation Strategies for Korean Male High School Students English Learning necessary to examine data from samples that have a similar background within a narrow band in order to provide a meaningful analysis of a particular case (Duff 2008), only the participants sharing a similar academic background and socioeconomic status were selected. In the context of junior high and high school curricula, the EFL classes that the participants attended were based mostly on textbooks. In both junior high and high schools, they took an in-house test twice every semester. The in-house English test, administered by relative evaluation, was mainly an evaluation of English grammar and reading comprehension. Other than regular tests, they were assigned group projects and individual English tasks for assessing their English proficiency in both junior high and high schools Instruments Before the main study, a preliminary interview was conducted to explore the participants lifelong EFL learning (see Table 1). In the first part of the preliminary interview, the participants drew a graph of their EFL motivational changes and wrote the reasons for high and low motivation in the graph (see Figure 1 in the result section). In the second part of the preliminary interview, they provided their English in-house test scores and explained how they prepared for English exams throughout junior high and high schools. Given the complexities of the subject, this research conducted qualitative research analysis through two preliminary interview sessions. One session involved an interview focusing on EFL learning motivation, and the other session entailed an interview on English exams. After the preliminary interviews, the main study followed, in which two sessions of semi-structured interviews were conducted (see Appendix for the sample interview questions). The main interview questions were developed by referring to Falout (2012), Kim (2006), Lamb (2007), and Ushioda (2001) while taking the preliminary interview into consideration. The first session of the main interview involved investigating the EFL learning atmosphere at school (eight questions) and the participants EFL learning motivation (nine questions). The participants were 56

61 Tae-Young Kim Ji-Young Kim Remotivation Strategies for Korean Male High School Students English Learning asked to answer the questions twice, once respectively for their junior high and high school experiences. The questions regarding EFL learning motivation asked about their experiences influencing their EFL learning motivation. The second session of the main interview focused on their in-house English test, investigating the components that affected English test scores (nine questions) and how the participants prepared for English tests (17 questions). The two sessions, making up the main portion of the study, explored participants specific experiences of EFL learning motivation and their in-house English test scores Data Collection Data collection was comprised of four steps. Table 1 presents each step including the methods, elapsed questionnaire and interview times, study periods, and the contents of collected data. Table 1. Data Collection Process 1. Introduction to the purpose and procedures of the study, with consent forms collected 2. Preliminary study Method Time (minutes) Period Content Open-ended questions - May 2 5 Motivational changes and EFL Follow-up interviews 7 13 May 6 8 exam scores 3. Main interview question development 4. Main study Method Time (minutes) Period Content Semistructured interviews May 9 15 EFL learning motivation Semistructured interviews May EFL exam scores Before collecting data, the first author met with the participants to explain the purpose and procedures of this study, as well as to distribute a consent form regarding the use of their in-house scores and interviews for this study, which they signed. The researchers interviewed the four participants three times to achieve data saturation (Brinkmann and Kvale 2015). This study 57

62 Tae-Young Kim Ji-Young Kim Remotivation Strategies for Korean Male High School Students English Learning started by a preliminary interview for investigating their learning background and their motivational changes represented by graphs. The remaining main study sessions consisted of semistructured interviews for exploring their perspectives on the components that affected their EFL learning motivation and English in-house test scores. As aforementioned, the data collected in the preliminary interview were used to review and modify the main interview questions before the preliminary interviews, thereby enhancing the trustworthiness of participants' EFL learning experience by incorporating their background information. All the interviews were conducted at one of the researcher s office. The preliminary interviews took 7 to 13 minutes, the first main interviews lasted 42 to 47 minutes, and the second main interviews were 49 to 74 minutes long. All the interviews were conducted in Korean, the first language of both the researchers and the participants, and the data were translated into English to be presented in this study Data Analysis After collecting the data, the interview recordings were first transcribed verbatim. The transcribed data totaled 149 pages, comprising 15 pages for the preliminary interview, 57 pages for the first main interviews, and 77 pages for the second main interviews. After transcribing the recorded data, the researchers highlighted the major sentences and expressions to find major themes (coding). Four themes were extracted from the transcribed data (categorization). The themes were interpreted according to their relationship to concepts from previous research. Finally, the verbatim excerpts were organized based on the four themes. Authors translated the excerpts originally written in Korean, their first language into English. Then, a doctoral student majoring in English education was asked to back-translate the English excerpts into Korean ones in order to guarantee the accuracy of data translation. Intersubjective consensus enhances the reliability of a study (Brinkmann and Kvale 2015). To ensure the reliability of this study, the first author used 58

63 Tae-Young Kim Ji-Young Kim Remotivation Strategies for Korean Male High School Students English Learning repeated readings of coding and categorization lists with another coder enrolled in a doctoral program at the authors' institution. The result of intercoder percentage of agreement of coding reached 100 percent. In addition, the second author reviewed coding and categorization and interpreted the data critically. Such analytical cross-validation can ensure the validity of a study (Brinkmann amd Kvale 2015). Lastly, the researchers showed the four organized themes and the related interview data to the participants and checked the accuracy of the contents. Creswell (2009) demonstrated that the credibility of a study is acquired by such member checking. 4. Results 4.1. Preliminary Interview Results: Exploring EFL Learning Demotivation and English In-house Test Scores Before the interview for exploring the components that enhance EFL learning motivation in the main study, the participants were asked to draw their own EFL learning motivational changes and were then interviewed to solicit their responses to the related questions. The data were used to understand how the participants recovered their EFL learning motivation. Figure 1 presents graphs of their motivational changes from elementary to high school, collected using the form originally used by Carpenter et al. (2009). The left column +3 to 3 referred to the EFL learning motivational level of the participants. They were asked to recall their EFL learning experience and draw the level of their motivational change from +3 to 3. As shown in Figure 1 of EFL learning motivational changes, students A and B showed higher motivation than students C and D in elementary school. However, students A and B experienced sharp decreases in their motivation compared to students C and D in junior high school. Students A and B experienced more fluctuations in their motivations than students C and D. 59

64 Tae-Young Kim Ji-Young Kim Remotivation Strategies for Korean Male High School Students English Learning Figure 1. Participants Changes in EFL Learning Motivation In the interview about the reasons for their EFL learning demotivation, the four students found that EFL learning in junior high school is more difficult than that in elementary school. In addition, because they heavily relied on hagwons, they believed that they did not need to study English in class or for themselves. They mainly demonstrated EFL learning demotivation in junior high school because they did not know why they were studying English, and nobody explained it to them. In summary, the participants were passive in EFL learning for three main reasons: EFL learning difficulty, high reliance on a hagwons, and a lack of EFL learning objectives. Table 2. English Exam Scores from Junior High to High School Junior High School High School First year Second year Third year First year Second year 1st sem* 2nd sem 1st sem 2nd sem 1st sem 2nd sem 1st sem 2nd sem 1st sem 2nd sem A** B C D Note. * sem refers to semester; ** A, B, C, and D refer to the participants 60