May 2021 – Volume 25, Number 1
Kathleen Bardovi-Harlig
Indiana University, USA
<bardoviindiana.edu>
Yunwen Su
University of Utah, USA
<yunwen.suutah.edu>
Abstract
This exploratory study examines the role of foreign and second language contexts in the acquisition of conventional expressions. A group of 20 ESL learners was compared to 25 EFL learners randomly selected from a larger pool. Both groups completed an aural multiple-choice discourse completion task (MC-DCT), which was developed from a previously validated oral DCT. The aural MC-DCT consisted of 20 items with learner-generated options delivered aurally. A total of 91 native speakers also completed the task as a control group. The results showed an effect of learning environment on learners’ selection of conventional expressions. The ESL group selected the conventional expressions in more items than the EFL group on the aural MC-DCT; the differences in the selections by the two groups were item-specific. The observed effect of learning context is discussed as related to individual items and type and modality of the task. The paper also discusses the special make-up of the ESL group due to the pandemic and expansion of the group for future research.
Keywords: Conventional Expressions, Pragmatic Routines, Learning Environment, Learning Context
Formulaic language has been discussed widely in second language (L2) pragmatics under different labels, such as formulaic sequences, situation-bound utterances, pragmatic routines, and conventional expressions. These terms are sometimes used interchangeably and sometimes with specific definitions (Bardovi-Harlig, 2012a, p. 208). This paper focuses on conventional expressions, which contain at least two morphemes and demonstrate situation-dependent, community-wide use, and can be demonstrated to be linked to specific situations (Bardovi-Harlig, 2009). Examples of conventional expressions include Nice to meet you to greet someone to whom one has just been introduced and No problem to deflect thanks for a small favor.
This paper examines the role of context of second and foreign language contexts in the acquisition of conventional expressions. A relatively small group of studies have investigated the acquisition of formulaic language by ESL versus EFL learners, but they have used a range of tasks that examine different aspects of learners’ pragmatic knowledge (production versus nonproduction) and different modes (oral/aural versus written). Roever (2005) used a written multiple-choice discourse-completion task (MC-DCT) to compare ESL and EFL learners’ recognition of pragmatic routines in different situations. Taguchi (2013) used an oral DCT to test production by learners, with and without study abroad experience, in an English medium university in Japan. Bardovi-Harlig and Bastos (2011) investigated the role of learning context, as measured by the length of residence and intensity of interaction, in ESL learners’ oral production of conventional expressions. Reanalyzing the data collected by Roever (2005), Roever et al. (2014) concluded that the written recognition task (the written MC-DCT) might favor reading proficiency because of its written mode.
This exploratory study compares the performance of a group of ESL speakers with that of a group of traditional EFL students on an empirically-developed aural MC-DCT. The aural MC-DCT reduces the effect of reading, matches the modality of conversation, expands the number of items investigated, and provides information that approximates the information in a production task while reducing the need to transcribe or rate the data. Developed from a previously validated oral DCT, the aural MC-DCT used in this study consists of 20 items with learner-generated options delivered aurally (Bardovi-Harlig & Su, in press), thus enhancing the comparison not only of learners in different learning contexts, but the comparison of tasks as well. Specifically, the present study focuses on individual items and explores what foreign and in-country language learners’ performance on an aural selection task tells us about the development of pragmatics and the acquisition of conventional expressions in different learning contexts.
Literature Review
The boundary between EFL and ESL is often fuzzy. According to Platt and Weber (1979), varieties of nonnative English move along a continuum where the transition from EFL to ESL happens gradually when language is used in everyday communication. The continuum perspective is reflected in the diverse learning contexts in which L2 pragmatic development has been examined empirically (Figure 1), with study abroad programs and domestic formal classroom instruction as dominant sites. This section reviews acquisitional studies of conventional expressions in light of the learning contexts in which they were conducted and the types of tasks that they used.
Figure 1. Continuum of English Learning Contexts.
Contexts for L2 Pragmatic Development
Taguchi and Roever (2017) summarized the potential advantages of learning L2 pragmatics in a host environment as direct observation of local norms, opportunities to apply observed norms in context, real-life consequences, and exposure to various styles and situations. Nevertheless, empirical studies that compare learning in host versus foreign contexts have not uniformly demonstrated the benefits of the host context on pragmatic development (Wyner & Cohen, 2015). The benefits of host environments, especially during study abroad, have been found to be mediated by features specific to input and interactions, to learners, and to target pragmatic features (see Pérez Vidal & Shively, 2019 for a review). Taguchi (2018) classified L2 pragmatic studies that examined the effect of context into three categories according to how context was treated in the research design: 1) context as a black box, coded dichotomously (second versus foreign language contexts, or study abroad versus at-home; e.g., Bardovi-Harlig & Dörnyei, 1998; Roever, 2005) or measured in terms of length of residence (e.g., Félix-Brasdefer, 2004); 2) context as exposure to input (e.g., Bardovi-Harlig & Bastos, 2011; Taguchi et al., 2013); and 3) context as sites for situated practice (typically qualitative studies, e.g., Kinginger, 2008; Shively, 2011).
The lack of agreement about the effect of context on L2 pragmatic development, in general, is reflected in studies on the acquisition of conventional expressions. Taking a “black box” perspective of the target-language context Roever (2005) found an advantage for ESL learners by comparing them with EFL learners on the recognition of pragmatic routines given specific situations or scenarios. Roever (2012) further reported that the advantage of ESL learners was positively related to the length of residence—a two months of residence in the target environment could lead to a substantial increase in the recognition of pragmatic routines with an increase of scores by about 50%. However, the positive effect of length of residence was found to be reduced in a later reanalysis by Roever and his colleagues. When unanswered items were included with other non-targetlike responses, Roever et al. (2014) found length of residence to be a significant predictor for ESL learners’ recognition of pragmatic routines—yet to a much lesser degree than in Roever (2012)—with one month increase in residence leading to only an increase of 0.3% in scores.
Roever et al. (2014) discussed factors that could possibly account for the discrepancy between their findings and Roever’s (2005,2012), one of which is length of residence as “a crude proxy for amount and intensity of interaction” (p. 394). The discussion falls in line with Bardovi-Harlig and Bastos’s (2011) finding that intensity of interaction, but not length of residence, is a significant predictor for ESL learners’ recognition of conventional expressions, a finding which was also supported by Sánchez-Hernández and Alcón-Soler (2019a). In a longitudinal qualitative study using interviews with ESL learners, Dörnyei, Durow, and Zahran (2004) also found that learning of routines was to a large extent related to learners’ involvement in the local community.
Taken together, the literature suggests an advantage of ESL learners over EFL learners in their acquisition of conventional expressions, which is related to the amount of exposure to input (or social contact) rather than the amount of time spent abroad. The effect of context, nevertheless, was found to be mediated by the type and modality of the pragmatic tasks employed and interaction with learner-internal variables like proficiency and attitude.
Tasks Used in the Study of Conventional Expressions and Pragmatic Routines
The type and modality of the tasks used to investigate learner knowledge of conventional expressions in the service of pragmatics vary across studies, and with it, the specific aspects of conventional expressions being investigated and what inferences can be made based on the tasks. Type of task considers whether the task used is a production task (i.e., a simulation of language use) or a nonproduction task (e.g., perception, comprehension, recognition); modality of task considers whether the task is oral/aural or written (i.e., whether input is listened to or read; whether output is spoken or written). This section reviews previous studies that have investigated the influence of learning context on learners’ acquisition of conventional expressions as measured by tasks of different types and modality.
Effects of learning context on the outcome of nonproduction tasks. Production tasks dominate research design in L2 pragmatics, but in the study of pragmatic routines and conventional expressions, Roever’s nonproduction task represents a seminal study. Using a 12-item written MC-DCT, Roever (2005, 2012) found a significant effect of context, but no significant effect of proficiency, on learners’ recognition of pragmatic routines as appropriate in specific contexts.
(1) Jane is at the beach and wants to know what time it is. She sees a man with a watch.
What would Jane probably say?
- “Excuse me, can you say the time?”
- “Excuse me, how late is it?”
- “Excuse me, what’s your watch show?”
- “Excuse me, do you have the time?” (Roever, 2005, Example 304)
In a later study, however, Roever et al. (2014) found a greater influence of proficiency than context. In addition to using length of residence rather than intensity of interaction to represent the variable of context, as is mentioned in the previous section, they also discussed how the written task might have been biased in favor of proficiency over residence. According to Roever et al. (2014), learners at higher general proficiency, which tends to be evaluated more by reading proficiency than residence, were more likely to recognize routines when given options in writing. In fact, using a decontextualized recognition task with aural presentation of conventional expressions and modified non-expressions, Bardovi-Harlig and Bastos (2011) found a significant effect of context, represented by intensity of interaction, but no significant effect of proficiency on their ESL learners’ recognition of conventional expressions in a decontextualized aural recognition task. Learners heard an expression such as “That’d be great” and then reported familiarity by choosing one response of three: “I often/sometimes/never hear this.”
Bardovi-Harlig (2014) examined ESL learners’ knowledge of conventional expressions using a modified Vocabulary Knowledge Scale (Wesche & Paribakht, 1996; henceforth VKS) to test expressions. The 22 items each presented a single aural conventional expression accompanied by a self-reported knowledge scale with opportunities to demonstrate knowledge. Results showed that learners at all levels predominantly reported knowing the expressions with confidence, and most of the time they provided a plausible definition or example. An effect of proficiency could be observed between learners at low and high intermediate levels (Levels 3 and 4) regarding their provision of definitions and examples, but not between the intermediate level (Level 4) and the low-advanced levels (Levels 5 and 6). Using a simplified VKS with written stimuli, Sánchez-Hernández and Alcón-Soler found significant effects of both context (measured by intensity of interaction; 2019a) and proficiency (2019b) on ESL learners’ recognition of pragmatic routines.
Taken together, context or more precisely exposure to input, has been found to play an important role in the development of learners’ ability to recognize conventional expressions. It is natural to assume that learners would have more opportunities to observe the use of the expressions and the sociopragmatic norms for their use in a host environment than in a foreign environment, and that such exposure to input in the target context would benefit their recognition of conventional expressions, which are bound to given situations. Proficiency—the most investigated learner variable in L2 pragmatic development—seems to have a less pronounced role than learning context in recognition tasks, according to Taguchi and Roever (2017), because conventional expressions are mostly short with little linguistic complexity (but see Bardovi-Harlig & Stringer, 2017, for a counterargument regarding syntactic complexity). Nevertheless, proficiency appears to show a more important role in written recognition tasks than aural recognition tasks (e.g., Roever et al., 2012; Sánchez-Hernández & Alcón-Soler, 2019b versus Bardovi-Harlig, 2014; Bardovi-Harlig & Bastos, 2011). Roever et al. (2014) suggest that this is probably due to the prominent role of reading proficiency in helping learners parse and analyze certain expressions in writing. Tasks with aural stimuli, however, demonstrate better ecological validity by matching the mode of real-world tasks because the expressions are used in conversations.
In addition to modality, recognition tasks also differ by whether they present the target expressions in isolation or in context. Bardovi-Harlig (2009, 2014), Bardovi-Harlig and Bastos (2011), and Sánchez-Hernández and Alcón-Soler (2019a, 2019b) presented isolated expressions in order to distinguish pragmalinguistic resources from sociopragmatic knowledge, and Bardovi-Harlig (2009, 2014) investigated use in context separately. The pairing of expressions with given situations in Roever’s (2005, 2012; Roever et al., 2014) written MC-DCT represents an approximation to production tasks in that expressions and alternatives are presented in specific situations presented by scenarios as in DCTs. Given the practice of using oral DCTs in production tasks, Bardovi-Harlig and Su (in press) argued for the integration of the oral mode into controlled pragmatic instruments such as the MC-DCT. They developed an aural MC-DCT as a needs assessment tool for instruction on conventional expressions in L2 English.
Effects of learning context on the outcome of production tasks. Studies of the production of conventional expressions demonstrate a better match of modality with real-world tasks in that most have adopted oral tasks. In general, they showed a combined effect of proficiency and exposure to input. As one of the earliest studies to explore the development of ESL learners’ ability to use conventional expressions, Bardovi-Harlig (2009) found with an empirically developed 32-item oral DCT that some expressions were built through interlanguage grammar (e.g., sorry I’m late and the interlanguage form sorry for lating), and that the use of other conventional expressions depended on the learners’ sociopragmatic knowledge (e.g. You’re welcome to accept thanks versus No problem to deflect thanks), which, presumably, is more related to their exposure to input than their proficiency. The impact of context and proficiency was directly investigated in Bardovi-Harlig and Bastos (2011) interpreting the oral DCT data from the earlier study through the lens of questionnaire data. Learning context was operationalized in two different ways—intensity of interaction and length of residence, and proficiency was decided by ESL learners’ placement in an intensive English program. A logistic regression model showed that both proficiency (p < . 001) and intensity of interaction (p < .05) had significant effects on learner production.
Findings in Bardovi-Harlig (2009) and Bardovi-Harlig and Bastos (2011) were supported by other studies. Using a 4-item oral DCT, Taguchi (2013) examined oral production of pragmatic routines EFL learners in an English-medium university in Japan. The learners were divided into three groups according to their study abroad experience and proficiency: 1) low proficiency with no study abroad experience; 2) high proficiency with no study abroad experience; and 3) high proficiency with more than one year’s study abroad experience. Results showed an overall limited ability of all three groups to produce target routines and a combined effect of proficiency and study abroad experience on production. A post-hoc linguistic analysis showed a stronger impact of proficiency on expressions that required grammatical analysis (e.g. Do you have the time?, which requires the use of article “the”) versus a stronger impact of study abroad experience on fixed expressions (e.g. No problem). In a later study, Osuka (2017) employed a pretest/posttest design and a 24-item oral DCT supported by video (the multi-modal elicitation task designed by Schauer, 2004) to investigate Japanese ESL learners’ development in their production of routines during a one-semester study abroad in the U.S. Results showed significant improvement in learners’ use of only one routine—from Thank you very much before study abroad to Thank you so much after study abroad. A retrospective questionnaire showed that learners with the highest input and output opportunities while abroad were responsible for the change. Nevertheless, overall learners did not show significant improvement in their use of routines, which was attributed to the limited exposure to the target language for only one semester. Studies of target languages other than English also showed item-specific development patterns in terms of learners’ production of formulaic routines during study abroad (e.g. L2 German in Barron, 2019; L2 Chinese in Taguchi et al., 2013).
Establishing Formulaicity
Tasks that have been used to investigate the acquisition of pragmatic routines and conventional expressions also differ in another crucial feature: Whether or not the conventionality or frequency of the expressions they test has been established. In pragmatics, earlier work (e.g., Roever, 2005) made assumptions that certain pragmatic routines were, in fact, conventional routines without providing evidence. The fact that native speakers (NSs) could identify the intended option on the written DCT was intended to indicate that the selection was the pragmatically most appropriate response, but not that the answer contained a formulaic sequence. In contrast, later work established conventionality of formulaic sequences (hence, conventional expressions) through observation, source searches, and extensive piloting with NSs. Following Erman and Warren (2000) conventional expressions are “combinations of at least two words favored by native speakers in preference to an alternative combination which could have been equivalent had there been no conventionalization” (Erman & Warren, 2000, p. 31). Later studies thus operationalized “favored by native speakers in preference to an alternative” as a string that occurs in greater than 50% of the NS responses to specific scenarios in a DCT. This method is described for Spanish (Bardovi-Harlig & Mossman, 2021) and documented in a number of studies for English (Bardovi-Harlig, 2009, 2019; Bardovi-Harlig et al., 2010) and for Chinese (Bardovi-Harlig & Su, 2018; Taguchi et al., 2013; Yang, 2016).
A second means for establishing pragmatic routines (expressions that are linked to speech acts but not specific situations) is to establish the frequency of a candidate string using a corpus that is regionally and pragmatically matched to the area of investigation. In that case, studies generally follow the work of Biber and colleagues who suggested that minimum frequency would be in the range of 10 occurrences per million words (Biber et al., 1999) to 40 occurrences per million words (Biber et al., 2004). Many pragmatic routines are much more frequent. This method was used to determine instructional targets by Bardovi-Harlig et al. (2015) and Bardovi-Harlig et al. (2017). Furniss (2016) also used a corpus, retrieving the most frequent two- and three-word sequences from the Russian National Corpus.
The purpose of establishing the frequency or density of use is to demonstrate that the strings are indeed social formulas (making no claims about the psycholinguistics), and that a learner may encounter them during interaction with competent speakers of the target language. Community-wide use needs to be established before acquisition can be tested. Most speech acts can be realized in a number of felicitous ways. If a particular sequence of words is idiosyncratic to a single speaker or is not favored by a speech community, then there is no reason to expect that learners will pick up that particular wording. Thus, even a low cut-off such as 25% (Osuka, 2017) does not meet a robust level. When there are two targets in the same situation both NS and learner production are inevitably lower (Bardovi-Harlig, 2012b). Moreover, if we do not verify frequency, when learners do not use the target expression one could always question whether the expression is as robust in input as the researcher had thought. Finally, without clearly establishing that conventional expressions or pragmatic formulas are linked to certain situations or speech acts, tasks only test pragmatic appropriateness, but not the acquisition of formulaic language as a pragmalinguistic resource. Conventional expressions and pragmatic routines clearly pose a learnability challenge to learners. Once the conventional expressions are established, along with the contexts in which they occur, the research challenge is to understand why acquisition seems to be delayed even when they are broadly used in the speech community. Rarely used expressions arguably pose a different learning challenge.
On the whole, the effect of learning context on the recognition of conventional expressions seems straightforward in the literature, whereas the role of context is not as clear-cut in production of conventional expressions. However, our understanding of the role of context is based in large measure on the written MC-DCT employed by Roever (2005) and later critiqued by Roever et al. (2014) as drawing on reading competence. Bardovi-Harlig (2013, 2018, inter alia) argues independently for the importance of matching the modality of pragmatics tasks with their real-world correlate, i.e., using oral/aural tasks to simulate conversation, written tasks to simulate writing such as letters, and computers for computer-mediated communication such as email.
To that end, Bardovi-Harlig and Su (in press) developed an aural MC-DCT with learner-generated options in order to have a practical means of assessing instructional needs for foreign-language programs. By comparing EFL learners’ responses to the aural MC-DCT and EFL responses on the oral DCT used by Bardovi-Harlig (2009), Bardovi-Harlig and Su demonstrated that the contextualized aural selection task (cf. the decontextualized presentation of target expressions in the self-reporting recognition task in Bardovi-Harlig, 2009) elicited learner selections that aligned with learner production on the oral DCT in the majority of items. In the present study we use that same task to investigate whether learning context influences EFL and ESL learners’ selections of conventional expressions. The present study hence asks the following research question:
Do EFL and ESL learners differ in their responses to an aural MC-DCT of conventional expressions in L2 English? And if so, how?
Method
Participants
Three groups of participants completed the study: undergraduate NSs (the NS group), EFL learners (the EFL group), and learners with in-country experience (the ESL group). Undergraduate students were recruited from three universities in the United States. Respondents who reported that they spoke English at home (N=74) and those who reported that they spoke English and another language at home and had attended high school in the United States (N=15) comprised the 89-speaker NS group. Sixty-seven students (75%) reported that they had attended high school in the Midwest, thus representing General American speakers (55 from Indiana, 10 from Illinois, and 2 from Michigan). Only two monolingual English speakers had not attended high school in the United States. The remaining 20 students had attended high school in 15 states outside the Midwest.
The EFL learners were 251 Chinese undergraduate EFL students recruited from a Chinese university in the top 10 from a northern province of China. They were all non-English majors who had had no experience in English-speaking countries. They were at the end of their first- (N = 201) or second-year (N = 50) college English classes (4 hours of instruction with nonnative speaking instructors) at the time of the data collection. The university consults China’s Standards of English Language Ability (CSE; Ministry of Education of People’s Republic of China, 2018) in developing its curriculum and assessment for non-English majors (CSE 5 and 6 for first- and second-year courses). CSE 5 and 6 are linked to B1 and B2 (Independent Users) in the Common European Framework of Reference (Papageorgiou, Wu, Hsieh, Tannenbaum, & Cheng, 2019).
Twenty undergraduate learners of English in the U.S. were recruited from three universities. This is a rather heterogeneous group, whose data was gathered during the pandemic. Thirteen were classmates of the NS participants from linguistics and second language studies classes, who reported not attending high school in the U.S. and not speaking English at home. Five were enrolled in an American history and culture course for international students (these students were online from home); only three were not matriculated to the university, and they were enrolled in a low-advanced level of the Intensive English Program at the same university. Like the EFL group, this group had proficiency scores in the B1-B2 range of the Common European Framework of Reference. All were taking courses on-line from American universities and had regular interaction with American peers; 16 were or had been in the US. There was a range of first-languages: Arabic (2), Chinese (7), Korean (1), Serbian (1), Spanish (7), Tagalog (1), and Thai (1).
Because of the difference in the size of the ESL group and the EFL group, we randomly selected 25 learners, or 10%, of the EFL group to represent it for any statistical comparisons. Every 10th learner on the data spreadsheet was selected, thus sampling from all EFL classes that had participated in China.
Instrument
The aural MC-DCT was comprised of 20 scenarios targeting 20 conventional expressions. Each item had four options, all of which were generated by speakers who completed an oral DCT with the same scenarios. The conventional expressions that were the target options were produced by NSs and learners in the earlier production task. The alternatives (forming the other three options) were produced by learners. In some items, the options appear immediately after the scenario (Example 2) and in others there is an intervening turn to which the options reply (Example 3)
(2) Initiating Item: Cellphone
You are at the bus stop. While waiting, you are talking with your friend on your cell phone. The bus arrives and you need to hang up.
(Audio only)
A. Wait for me B. I gotta go C. The bus is coming D. I need to get off
(3) Responding Item: Gave Ride
You give your classmate a ride home. He lives in the building next to yours. He gets out of the car and says: (Learners hear: “Thanks for the ride.”)
(Audio only)
A. Don’t worry B. You’re welcome C. No problem D. My pleasure
All alternatives were grammatical. The options were re-recorded by speakers of American English and were presented in alternating male and female voices. A NS read the scenario while the learners read along, then 4 options were played straight through (a-d), and then they were played a second time. The full script is provided in the Appendix, and the audio instrument is available on IRIS.
Procedure
The aural MC-DCT was delivered following pandemic procedures with remote teacher control of the audio-visual task and digital-delivery of the answer sheet. The learners in China completed the task as an in-class assignment; they accessed an electronic version of the answer sheet, which they printed and completed. They took a picture of the answer sheet and emailed it to their teachers. The NS and ESL groups completed the task as homework or extra credit in linguistics, second language studies, or Intensive English Program courses. The US-based students accessed a web-based answer sheet as they watched an mp4 file hosted on their course support system.
Learners heard the scenario over earphones as they read it on the screen. For responding scenarios, they also heard an audio turn, followed by a screen that said “you say” with four audio icons. As the option was produced, the letter (A-D) appeared on the screen. Learners listened to the options twice and circled the letter of the response they thought was the most appropriate for the context.
Analysis
We calculated the rate of selection for each option on each item. Each option selected received a point, allowing us to calculate both percentage distribution of the response and the mean score. Most results will be reported as percentages.
Because the EFL group of 25 learners was randomly down-sampled from a group of 251 learners, their scores were compared to the larger EFL group. The overall rates of selecting the same conventional expression as NSs did was 47% for the down-sampled group, compared to the larger group’s selection rate of 45%. Next, we checked each item against the larger group. Eleven items were within ±5%, four items were within ±10%, and four items were different by ±11% or more (B8 More food 11%, A3 Movies 15%, A5 Bus 18%, B2 Offer of help, 24%). To put this into perspective, in a sample of 25 learners, the answer of one learner accounts for 4% of the total for any single item. The larger differences will be indicated in the appropriate tables.
NS responses show that they selected the intended conventional expression in 17 of 20 items. Recall that the “intended” conventional expression was the expression produced by over 50% of the NSs in the oral DCT (Bardovi-Harlig, 2009) on which the aural MC-DCT used in this study was based. For A6 25 Minutes Late, NSs selected the distractor I’m really sorry (57%) rather than the NS-produced I’m sorry I’m late (43%); thus I’m really sorry was treated as the target on this task. B7 Dog hit by car (condolences) divided NS responses between I’m sorry and I’m so sorry; these were combined to allow both as the correct answer, in keeping with the NS production on the oral DCT (Bardovi-Harlig, 2009, Table 6). Two items (A1 and A2) failed to meet the cut off for 50% selection, scoring 48% and 47% respectively and were thus eliminated from further analysis.
Results
NSs selected the intended conventional expression in 76% of their responses. The highest rate of selection by NSs was 94% (B6 Party closing, Thanks for inviting me) and the lowest was I’m really sorry selected by 57% of the NS respondents (A6 25 min late).
The overall rate of selection of the target conventional expressions by the ESL group was 72% [(259/(18 x 20)). The highest rate of selection was 100% for You too! as a response to Have a nice day! at a grocery store (B11). The lowest rate of selection was 45% for No problem as a deflection of thanks (B9 Gave Ride) and I’m so sorry as an expression of condolence addressed to someone whose father had passed away (B4 Father).
The overall rate of selection of the target conventional expressions by the EFL participants was 47% (211/(18 x 25)). A Welch t-test was run to determine if there were differences in the selection of conventional expressions by the ESL and EFL groups due to the assumption of homogeneity of variances being violated, as assessed by Levene’s test for equality of variances (p = .017). The ESL learners scored higher on the MC-DCT (M = 12.15 ± 0.35 [mean ± standard error] out of 18 points) than the EFL learners (M = 8.44 ± 0.56), a statistically significant difference of 3.71 (95% CI, 2.37 to 5.05, t (38.814) = 5.60, p < .001) with a large effect size (Cohen’s d = 1.63). The two items with the lowest rate of selection by ESL participants also had the lowest rates of selection by the EFL learners (4% for No problem in B9 Give Ride and 8% for I’m so sorry in B4 Father), but the selection rates of the EFL learners were substantially lower. The EFL participants showed similarly low rates of selection for an additional item, namely the condolence I’m (so) sorry in B7 Dog Hit by Car (8%). This item also received the second lowest rating by the ESL participants (50%).
The ESL participants had a clear preference for target expressions selected by the NSs. In 15 of the items the ESL learners selected the conventional expression at rates greater than 50% (recall that 50% is the cut-off for the identification of conventional expressions). In an additional item, B7 Dog Hit by Car, 50% of the ESL speakers chose I’m (so) sorry. The intensified condolence expression I’m so sorry was chosen by 45% of the ESL speakers in B4 Father Condolences. The remaining item was a deflection of thanking for a ride home (B9 Gave Ride) which also was selected in 45% of the responses. The selection rates for the ESL group fell between 80% and 100% in 8 out of the 18 items, 60% and 79% in 6 items, and 50% and 59% in 2 items (Figure 2). In contrast, the EFL learners showed a clear preference (i.e., over 50%) for NS-selected expressions in only 9 out of the 18 items. Their rate of selection met or exceeded 80% in 2 items (84% for Watch out! in A5 Bus and 80% for Nice to meet you in B12 Introduction), 60% in 4 items, and 50% in 3 items (Figure 2).
The NSs also showed different selection rates according to item and conventional expression (Figure 2). Seven items scored over 80%, an additional 8 between 60-79%, and the final 3 between 50 and 59% (Tables 1-4).
Figure 2. Rate of Selection by NS, ESL, EFL Participants
ESL Target Selection; EFL Alternatives
The ESL participants favored the NS-selected target expressions in all items and scored more than 20% higher than EFL participants in 12 out of the 18 items (Tables 1 and 2). For five of these 12 items, the proportion of the EFL participants who selected a distractor was higher than the proportion of the EFL participants who selected the conventional expression (Table 1). For the two condolence scenarios of B4 Father condolence and B7 Dog Hit by Car, the EFL learners had a clear preference for I’m sorry about that, which may be an apology rather than a condolence, whereas the ESL learners preferred the condolence expressions I’m (so) sorry (45% in B4 Father and 50% in B7 Dog hit by car), though the preference is not as strong as in the NS selection (79% and 61%, respectively). In B9 Gave Ride, in which the NSs and ESL learners deflected the expression of gratitude for a ride home, the EFL learners instead accepted the gratitude and favored You’re welcome (48%) and My pleasure (44%) to accept the thanks. In contrast, NSs favored No problem (90%) to deflect the thanks. No problem was also the favorite selection of the ESL learners (45%), although quite a few of them chose My pleasure (35%). For A3 Movies, the EFL learners were almost equally divided between the alerter Excuse me (48%) and the indirect request Could you be quiet? (40%), whereas the ESL learners, like the NSs, favored Could you be quiet? (ESL: 60%; NS: 67%) to Excuse me (ESL: 40%; NS: 29%). The EFL learners were also divided between Thank you so much (40%) and Thank you for your kindness (44%) in B10 Make-up test, whereas both the ESL learners and NSs showed a clear preference for Thank you so much (ESL: 75%; NS: 82%).
Table 1. Nontargetlike Selection for EFL, Targetlike Selection for ESL.
A3 Movies | Would you please shut up? | Could you be quiet? | Excuse me | Be silent | |
EFL | 12.0 | 40.0 | 48.0 | 0 | |
ESL | 0 | 60.0 | 40.0 | 0 | |
NS | 3.9 | 66.7 | 29.4 | 0 | |
B4 Father | Ok, no problem | I’m sorry about that | I’m sorry | I’m so sorry | |
EFL | 4.0 | 84.0 | 4.0 | 8.0 | |
ESL | 5.0 | 30.0 | 20.0 | 45.0 | |
NS | 1.1 | 12.4 | 7.9 | 78.7 | |
B7 Dog Hit by Car | I’m sorry | How did that happen? | I’m sorry about that | I’m so sorry | |
EFL | 4.0 | 12.0 | 80.0 | 4.0 | |
ESL | 15.0 | 30.0 | 20.0 | 35.0 | |
NS | 20.2 | 18.0 | 21.4 | 40.4 | |
B9 Gave Ride | Don’t worry | You’re welcome | No problem | My pleasure | |
EFL | 4.0 | 48.0 | 4.0 | 44.0 | |
ESL | 0 | 20.0 | 45.0 | 35.0 | |
NS | 1.1 | 3.4 | 89.9 | 5.6 | |
B10 Make-up Test | Thank you for your kindness | Thank you | Thank you so much | I appreciate it | |
EFL | 44.0 | 4.0 | 40.0 | 12.0 | |
ESL | 10.0 | 5.0 | 75.0 | 10.0 | |
NS | 3.4 | 4.5 | 82.0 | 10.1 |
Note. The EFL score of 40% for A3 is 14.6% lower than the larger EFL group.
Targetlike Selection at Different Rates
For the other seven of the 12 items, the EFL learners also showed a clear preference for the target expression, but the ESL learners scored much higher (over 80% for all 7 items: apologies in A4 and B3; refusals in B5 and B8; thanks in B6 [and also B10 in the previous table]; closing the phone conversation in A7; reciprocal greeting in B11). In fact, the ESL learners showed such a high level of familiarity with these expressions that they selected the target expressions at the same rate as (B3, B6, B8) or even to a greater extent (A4, A7, B5, B11) than NS (Table 2).
Table 2. Targetlike Selection for EFL, Higher Targetlike Selection for ESL.
A4 Busy Teacher | I’m sorry to bother you | I’m so sorry | Thank you | Thank you for your time | |
EFL | 16.0 | 0 | 16.0 | 68.0 | |
ESL | 5.0 | 0 | 5.0 | 90.0 | |
NS | 5.6 | 1.1 | 14.6 | 78.7 | |
A7 Cell Phone | Wait for me | I gotta go | The bus is coming | I need to get off | |
EFL | 12.0 | 36.0 | 28.0 | 24.0 | |
ESL | 5.0 | 80.0 | 15.0 | 0 | |
NS | 0 | 76.4 | 20.2 | 3.4 | |
B3 5 Minutes Late | I’m so sorry | Sorry about my delay | I’m sorry I’m late | Thank you | |
EFL | 20.0 | 4.0 | 56.0 | 20.0 | |
ESL | 0 | 20.0 | 80.0 | 0 | |
NS | 6.7 | 10.1 | 80.9 | 2.2 | |
B5 Shopping No Help | No thank you | No thanks, I’m just looking | Well, thanks I’m fine | I don’t need help | |
EFL | 36.0 | 48.0 | 16.0 | 0 | |
ESL | 10.0 | 85.0 | 5.0 | 0 | |
NS | 21.4 | 77.5 | 1.1 | 0 | |
B6 Party Closing | I’m sorry | You’re welcome | Thanks for inviting me | I have to go | |
EFL | 20.0 | 8.0 | 60.0 | 12.0 | |
ESL | 0 | 0 | 95.0 | 5.0 | |
NS | 2.2 | 0 | 94.4 | 3.4 | |
B8 More Food | No thanks | No thanks, I’m full | Really I’m full | That’s enough | |
EFL | 16.0 | 40.0 | 28.0 | 16.0 | |
ESL | 15.0 | 80.0 | 5.0 | 0 | |
NS | 14.6 | 82.0 | 3.3 | 0 | |
B11 Have a nice day! | Thank you | You too | Have a nice day | Good bye | |
EFL | 24.0 | 52.0 | 16.0 | 8.0 | |
ESL | 0 | 100.0 | 0 | 0 | |
NS | 10.1 | 86.5 | 3.4 | 0 |
Note. The EFL score of 40% for B8 is 11% lower than the larger EFL group.
B12 Introduction is the only item that elicited a higher agreement rate among the EFL learners than ESL learners (Table 3). A proportion of the ESL learners (24%) selected Hi Bill, how are you?, which speakers often use to greet someone who is known to them, but which is also an appropriate response to an introduction.
Table 3. Higher Rate of Selection by EFL.
B12 Introduction | Nice to meet you | Hi Bill, how are you doing? | Nice to see you | Hi Bill, how are you? | |
EFL | 80.0 | 4.0 | 8.0 | 8.0 | |
ESL | 65.0 | 10.0 | 0 | 25.0 | |
NS | 78.7 | 7.9 | 1.1 | 12.4 |
Similar Rates of Selection
There were five items where the differences between the EFL and ESL learners’ mean scores were lower than 15% (Table 4). All three groups showed only a moderately high agreement rate (50-70%) in four of the five items. The exception is A5 Bus, where the selection of the target expression Watch out! exceeded 80% in all three groups. In contrast, the agreement rates on Watch out! in A8 Puddle were between 50% and 60%. In A8 Puddle, a restatement of the problem There’s a puddle was the most popular alternative to Watch out in all groups (EFL: 20%; ESL: 25%; NS: 36%).
Table 4. Targetlike Selection for EFL and ESL at Similar Rates.
A5 Bus | The bus is coming | Be careful! | Watch out! | Pay attention! | |
EFL | 4.0 | 0 | 84.0 | 12.0 | |
ESL | 10.0 | 0 | 90.0 | 0 | |
NS | 5.6 | 6.7 | 86.5 | 1.1 | |
A6 25 Minutes Late | I’m sorry about that. | I’m sorry I’m late. | I’m really sorry. | I’m sorry. | |
EFL | 16.0 | 28.0 | 56.0 | 0 | |
ESL | 0 | 25.0 | 75.0 | 5.0 | |
NS | 0 | 42.7 | 60.2 | 0 | |
A8 Puddle | Watch out! | Be careful! | Pay attention! | There’s a puddle | |
EFL | 56.0 | 12.0 | 12.0 | 20.0 | |
ESL | 60.0 | 10.0 | 5.0 | 25.0 | |
NS | 59.6 | 4.5 | 0 | 36.0 | |
B1 Shopping Help | Yes, I need some help | I’m looking for a shirt | I want a shirt | Sure, of course | |
EFL | 32.0 | 44.0 | 8.0 | 16.0 | |
ESL | 35.0 | 55.0 | 0 | 10.0 | |
NS | 32.6 | 65.2 | 0 | 2.2 | |
B2 Offer of Help | Thank you | That’s alright | Ok, I appreciate it | That’d be great | |
EFL | 16.0 | 0 | 20.0 | 64.0 | |
ESL | 10.0 | 0 | 20.0 | 70.0 | |
NS | 9.0 | 6.7 | 24.7 | 59.6 |
Note. The EFL score of 84% for A5 is 17.5% higher and 64% is 24% higher for B2 than the larger EFL group.
On the whole, the ESL group selected the conventional expressions in more items than the EFL group in the aural MC-DCT; the differences in the selections by the two groups were item-specific. The next section discusses the observed effect of learning context as related to the target expressions and the assessment task.
Discussion
The results showed an effect of learning environment on the learners’ selection of conventional expressions in the aural MC-DCT. The performance of 20 ESL learners and 25 EFL learners (down-sampled from a larger group of 251) was compared on the timed task. The ESL learners outperformed the EFL learners, even though not all ESL learners were living in the target language community at the time of the data collection. The ESL learners selected the target conventional expressions in more scenarios than the EFL learners, and they also identified them at higher rates. As in Roever (2005), this study treated the learning context dichotomously, comparing ESL learners to EFL learners. The finding of an advantage of ESL over EFL learners in their selection of conventional expressions is in line with previous studies (Roever, 2005, 2012; Roever et al., 2014).
The study also found evidence that interlanguage grammar is likely to play a role in the selection of some alternatives and that rates of learner selection and native-speaker selection varied by conventional expression and scenario (not all expressions are equally selected in the scenarios given). This section discusses the effect of the learning context and how it interacted with task type and modality. It also further discusses the special make-up of the ESL group as a result of the pandemic, and how learner groups could be expanded for future research.
Pragmalinguistic and Sociopragmatic Differences in ESL and EFL Learner Selections
All the options on the aural MC-DCT were grammatical. Nevertheless, the current state of the EFL learners’ interlanguage grammars seemed to inform their selections. For example, the EFL learners preferred thanking and condolence expressions that had complements. Thank you/thanks for was selected both in contexts in which they were highly preferred (Thank you for your time in A4 Busy Teacher and Thanks for inviting me in B6 Party Closing) and in contexts where no complement was called for. In B10 Make-up Test—for which Thank you so much was the NS selection modeled on NS production on the oral DCT in Bardovi-Harlig (2009)—EFL learners selected Thank you for your kindness. In contrast, the ESL learners preferred the target expression Thank you so much. Similarly, the EFL learners selected I’m sorry about that as the condolence expression in B4 Father and B7 Dog Hit by Car, whereas I’m (so) sorry was selected by the NSs and by over 50% of the ESL learners. The preference for complements was not categorical, however, because the EFL learners selected I’m (so/really) sorry as the apology expression in both Late scenarios (B3 5 Minutes Late and A6 25 Minutes Late), suggesting that they did accept sorry without a complement even when an about-phrase was offered; however, neither selection was the target in those scenarios in which I’m sorry I’m late was preferred.
The ESL learners also aligned more closely with NSs in their sociopragmatics in some situations, which then influenced their selection of the corresponding pragmalinguistic options in the aural MC-DCT. For example, they recognized I’m sorry as both an apology (in B3 and A6) and a condolence (in B4 and B7). They also favored No Problem to deflect gratitude for a small favor in B9 Gave Ride, as NSs would do, over expressions used to accept thanks, although they were not quite there yet, selecting No problem just over half as often as NSs (see Table 1). The findings echoed to a certain degree Taguchi’s (2013) conclusion about the impact of study abroad experience on learners’ production of fixed expressions like No problem, which have little linguistic complexity. While Taguchi (2013) related the impact of different variables only to the linguistic complexity of the expressions, we argue that the effect of learning context on the acquisition of expressions like No problem as a response to thanks should be attributed to its sociopragmatic function as a deflection of thanks, versus acceptance of thanks, rather than its linguistic simplicity alone. Similarly, learners appear to need exposure to acquire I’m sorry as a form of condolence and Thank you so much to express upgraded gratitude. The potential link between exposure and pragmatic development is also supported by Osuka’s (2017) finding that learners shifted from Thank you very much to Thank you so much after a semester’s study abroad. The influence of learning context on the sociopragmatic aspect of L2 pragmatic competence aligns with the review by Wyner and Cohen (2015).
Variability in the Use of Conventional Expressions
Although the learning context may increase (or decrease) learners’ chances of encountering conventional expressions in communication, there is the additional issue of variability in rates of use. In this task variability was realized as rates of selection. A cut-off of 50% agreement in the NS responses was meant to ensure that every selection was the favored option. The target conventional expressions were selected from the single most favored expressions that NSs produced on the corresponding oral DCT, and NSs recognized them as such by selecting the intended target in 18 of 20 scenarios. But they nevertheless showed more robust selection of some conventional expressions than others, parallel to their performance on production tasks. The results also suggest that high NS agreement may predict high rates of selection by ESL learners, but not by EFL learners. A larger ESL group would be necessary to confirm this preliminary observation.
Even high agreement among NSs does not exclude other appropriate responses to the situation. The use of the conventional expressions may be the dominant response, but it is never the only response. Learning conventional expressions has high value for pragmatic comprehension for learners as addressees, but they do not have to use them. However, when they do, they also increase the likelihood that their response will be understood in the speech community in which they are used.
Task Type and Modality
The present study tested learners’ selection of conventional expressions in an aural MC-DCT. The aural mode of the task matches the aural mode of conversations in which learners are likely to encounter the conventional expressions. Thus, it more closely simulates conversation than written MC-DCTs do. However, multiple-choice tasks do not occur naturally. Selecting a conventional expression seems to be easier than producing one, but learners also appear to be attracted by options that reflect their sociopragmatics and interlanguage grammars. The results of the present study show that the effect of learning context on learners’ selection of conventional expressions as observed in the written task in Roever (2005, 2012; Roever et al. 2014) also holds true for responses to the aural task originally developed in Bardovi-Harlig and Su (in press).
Because the aural MC-DCT is a selection task, it does not predict production, even in the same situations. Comparison with previous studies (Bardovi-Harlig, 2009; Bardovi-Harlig & Bastos, 2011; Bardovi-Harlig & Su, in press) shows that both the EFL and ESL learners’ selection rates were higher on the aural MC-DCT than production rates on the oral DCT on the same scenarios. Learners’ relatively low scores on the aural MC-DCT often match low production (e.g., No problem as deflection of thanks, I’m (so) sorry as condolence), which points to areas where instruction could be helpful. However, their high selection scores may predict high production scores (as in the case of You too! and Nice to meet you) or more importantly, they may not (as in the case of the high selection and low production of That’d be great and I gotta go). Using two different tasks Bardovi-Harlig (2009, 2014) and Bardovi-Harlig and Stringer (2017) showed that when responding to an offer of help, ESL learners did not use the expression That’d be great, but rather they attempted the expression by structuring their attempts around the word great. It seems that the learners associated great with the scenario but lacked the syntax and phonotactics to produce (and reduce) the string. Thus, although low scores can be interpreted as areas for instruction, high scores indicate some knowledge upon which teachers and learners can build, but should not be interpreted as needing no attention.
Learner Groups
As noted earlier, the dichotomous physical positioning of learners is a convention that obscures the complexity of place. The divisions between ESL and EFL learners or at-home and study-abroad learners do not always describe the same learners across studies. The EFL learners in this study represented a rather standard EFL situation: They were university students who were not English majors, taught by teachers who were also L2 speakers and shared their L1. Additionally, the EFL students had not been abroad. In contrast, our ESL group is not as typical. Whereas we would usually use the terms “ESL” or “host environment” to describe learners who are taking courses and living in an English-speaking region or country (in this case the U.S.), in this study we have extended the terms to learners who are taking courses that originate at U.S. universities. Due to the corona-virus pandemic, the learners were taking courses online. The learners whom we tested remotely could have been in the U.S. or elsewhere. Even if they had physically returned to campus, interaction outside the classroom is likely to have been diminished, thus reducing the potential benefits of the host environment. We considered using a term other than “ESL” for our pandemic participants, but decided against creating yet another term for a temporary situation. Nevertheless, the pandemic ESL group showed an advantage over the EFL group in selecting conventional expressions in the aural MC-DCT.
Limitations
The use of this same learner group which captures the reality of research during the pandemic and which introduces a new type of learner that might be a permanent part of ESL programs going forward, may, nevertheless, be viewed as a limitation of the present study. We would have sought out “traditional” ESL students, i.e., physically present, host-environment learners, had they been available. We also note that at 20 participants, the ESL group is smaller than the learner groups included in other studies that we have cited here. The diminished size of the ESL learner sample caused us to down-sample our much larger EFL learner group from 251 learners to 25 learners, just 10% of the original group. Although this facilitated comparison of the ESL and EFL learners, it also resulted in the loss of some information, which we attempted to provide in notes following the tables. Certainly, the pandemic was the greatest limitation on recruitment, data collection, and even the ability of international students to enter the United States to attend universities which may not have been holding in-person classes. We hope that this last and greatest limitation resolves soon.
Suggestions for Future Research
In addition to increasing the number of ESL learners and constructing a more homogeneous group of ESL learners in order to address the limitations that we identified in the preceding section, we also suggest testing learners at lower levels of proficiency as well as the relatively advanced learners in the present study. An expanded proficiency range would help illuminate the role that proficiency plays, as well as testing the contribution of the environment to its fullest. Replicating this study with a larger group of ESL learners at different levels of proficiency would allow further investigation of development. Systematically studying a greater variety of ESL learners (such as online learners, resident university students, community language learners, and specific-purpose learners) would facilitate the identification of features of interaction that give ESL learners the edge in building a pragmalinguistic repertoire that includes conventional expressions. In addition, we propose two main avenues for future research.
One avenue would be to expand the study by purposefully including learners with the experience that challenge the binary account of place, discussed earlier in this section. This expansion would be relevant to the acquisition of conventional expressions more generally and would include oral production of conventional expressions and not be restricted to the use of the aural MC-DCT described here. Studies might compare students who have learned or are learning a single L2 to those who are serial language learners. This comparison could be made in the ESL context or the EFL context, as this variable is relevant to both. A different comparison might examine learners who are taking English courses at universities in English-speaking countries whose contact with language classrooms is face-to-face and synchronous online. This might help us understand what aspects of the language classroom are particularly resilient and are conveyed through instruction alone rather than proximity. Heritage learners and second language learners may also share the same environment but have different experiences that lead to different mastery of conventional expressions. Background questionnaires that address such variables would need to be administered after the conventional expressions task or tasks.
A second avenue would be to examine the effect of context by measuring it in terms of intensity of interaction or exposure to input as in Bardovi-Harlig and Bastos (2011). The original questionnaire used to determine the intensity of interaction focused on face-to-face communication and television viewing, two potential sources of oral conventional expressions. Future studies of intensity of interaction should include sources of conversational interactions that have become common since the original study was undertaken, including computer-mediated interactions and exposure to social media, as well as other contexts for interaction that characterize pandemic and post-pandemic learning environments.
Pedagogical Implications
Findings of the study show an effect of the learning environment on learners’ selection of conventional expressions, which in turn adds to the justification for integrating pragmatics instruction into the curriculum of teaching English as a foreign language, as well as other languages taught as foreign languages for that matter, in which contexts learners’ access to authentic input is largely restricted. In the case of conventional expressions, the important pragmalinguistic resources are by definition bound to authentic community-wide use.
There are several resources that foreign language instructors can take advantage of in designing pedagogical materials for teaching conventional expressions. Instructors could start with the target expressions in the present study, the conventionality of which has been empirically validated in previous studies (Bardovi-Harlig, 2009), and extract relevant scenarios from films and TV shows to develop awareness-raising activities. For example, Bardovi-Harlig and Vellenga (2012) used excerpts from Friends to teach many of the conventional expressions used here. Bardovi-Harlig and Nickels (2011) focused exclusively on teaching thanking expressions. In addition, teacher-developed corpus-based materials and teacher-led corpus searches by learners have been found to be effective for the instruction of pragmatic routines in improving the accuracy of learner production and enhancing their engagement (Bardovi-Harlig et al., 2017). Last but not least, the scenarios used in the aural MC-DCT in the present study could serve as the ground for potential production activities in the form of short skits (Bardovi-Harlig & Griffin, 2005, additionally video-recorded the skits, which the class then watched).
Bardovi-Harlig and Mossman (2016) suggest a variety of activities for teaching conventional expressions. There are a number of ways to work with television shows. Online versions of shows are easy to work with, given the ease of locating any scene by its time. Additionally, fan transcriptions of shows posted on the internet are real time-savers for teachers, although they should be checked. Teachers can use TV shows for both input and production. As input, video vignettes featuring the use of target expressions could be used to increase learners awareness of those expressions and the relevant contexts (Derakhshan & Eslami, 2020). For production activities, the scene can be stopped immediately before the conventional expression, and students can supply the next turn (Bardovi-Harlig & Mossman, 2016). The teacher can then play the scene, and learners can compare their production to the original. This has the advantage for EFL teachers, who may be advanced L2 speakers themselves, of having a built-in “answer key.” Teachers may select the variety of English with which their programs or students identify most closely, or they may show TV excerpts from different countries or regions to demonstrate regional variation.
Above all, in foreign language environments, it is essential for classroom instruction to provide learners with exposure to conventional expressions used in real conversations, to raise learners’ awareness of the contexts where certain expressions are called for, and to provide them with opportunities to use the expressions.
Conclusions
This study found that ESL learners scored higher that EFL learners on an aural MC-DCT that tested conventional expressions. The study thus corroborates the results of earlier studies that used a written MC-DCT, while eliminating the confound of reading proficiency. The aural MC-DCT presents spoken options to the participants, thus simulating the way that learners encounter conventional expressions in conversation. Although a multiple-choice task does not predict production, it does indicate situations in which learners need help in identifying the speech act that is preferred in given situations, and it also can identify conventional expressions that are unfamiliar to learners. The aural MC-DCT also compares favorably to the oral DCT with the same scenarios. Comparisons to production tasks suggest that learners may need help producing the conventional expressions that they select.
The effect of the target-language context appears to be advantageous to the acquisition of conventional expressions as measured by a selection task, even when target context is defined generously to include students who are present in the target language community and those who are members of a virtual academic community. Conducting research during the pandemic has made us aware that there are different ways of defining academic communities, which may endure post-pandemic and would be valuable sites for pragmatic research going forward.
About the authors
Kathleen Bardovi-Harlig is Provost Professor of Second Language Studies at Indiana University where she teaches and conducts research on second language acquisition, L2 pragmatics, and tense-aspect systems.Her work on pragmatics has appeared in Language Learning, SSLA, Intercultural Pragmatics, Journal of Pragmatics, The Modern Language Journal, and Language Learning & Technology. She is co-editor of Interlanguage Pragmatics: Exploring Institutional Talk (Erlbaum) and Teaching Pragmatics.
Yunwen Su is Assistant Professor of Chinese Linguistics at University of Utah where she teaches and conducts research on acquisition, pedagogy, and assessment of second language pragmatics. Her work on pragmatics has appeared in Journal of Pragmatics, The Modern Language Journal, Language Learning & Technology, Foreign Language Annals, East Asian Pragmatics, and edited volumes.
To cite this article:
Bardovi-Harlig, K., & Su, Y. (2021). The Effect of Learning Environment on the Selection of Conventional Expressions on an Aural Multiple-Choice DCT. Teaching English as a Second Language Electronic Journal (TESL-EJ), 25(1). https://tesl-ej.org/pdf/ej97/a1.pdf
References
Bardovi-Harlig, K. (2009). Conventional expressions as a pragmalinguistic resource: Recognition and production of conventional expressions in L2 pragmatics. Language Learning, 59(4), 755–795. https://doi.org/10.1111/j.1467-9922.2009.00525.x
Bardovi-Harlig, K. (2012a). Formulas, routines, and conventional expressions in pragmatics research. Annual Review of Applied Linguistics, 32, 206–227. https://doi.org/10.1017/s0267190512000086
Bardovi-Harlig, K. (2012b). Pragmatic variation and conventional expressions. In J. C. Félix–Brasdefer & D. A. Koike (Eds.), Pragmatic variation in first and second language contexts: Methodological issues (pp. 141–173). John Benjamins.
Bardovi-Harlig, K. (2013). Developing L2 pragmatics. Language Learning, 63(Supplement 1), 68–86. https://doi.org/10.1111/j.1467-9922.2012.00738.x
Bardovi-Harlig, K. (2014). Awareness of meaning of conventional expressions in second language pragmatics. Language Awareness, 23(1-2), 41–56. https://doi.org/10.1080/09658416.2013.863894
Bardovi-Harlig, K., & Bastos, M.-T. (2011). Proficiency, length of stay, and intensity of interaction and the acquisition of conventional expressions in L2 pragmatics. Intercultural Pragmatics, 8(3), 347–384. https://doi.org/10.1515/iprg.2011.017
Bardovi‐Harlig, K., & Dörnyei, Z. (1998). Do language learners recognize pragmatic violations? Pragmatic versus grammatical awareness in instructed L2 learning. TESOL Quarterly, 32(2), 233–259. https://doi.org/10.2307/3587583
Bardovi-Harlig, K., & Griffin, R. (2005). L2 pragmatic awareness: Evidence from the ESL classroom. System, 33(3), 401–415. https://doi.org/10.1016/j.system.2005.06.004
Bardovi-Harlig, K., & Mossman, S. (2021). Formulaic language in Spanish pragmatics. In D. A. Koike, & J. C. Félix–Brasdefer (Eds.), The Routledge handbook of Spanish pragmatics (pp. 203–218). Routledge.
Bardovi-Harlig, K., & Mossman, S. (2016). Corpus-based materials development for teaching and learning pragmatic routines. SLA research and materials development for language learning, 250–278.
Bardovi-Harlig, K., Mossman, S., & Su, Y. (2017). The effect of corpus-based instruction on pragmatic routines. Language Learning & Technology, 21(3), 76–103.
Bardovi‐Harlig, K., Mossman, S., & Vellenga, H. E. (2015). Developing corpus‐based materials to teach pragmatic routines. TESOL Journal, 6 (3), 499–526. https://doi.org/10.1002/tesj.177
Bardovi-Harlig, K., & Nickels, E. L. (2011). No thanks, I’m full: Raising awareness of expressions of gratitude and formulaic language. In N. R. Houck & D. H. Tatsuki (Eds.), Pragmatics: Teaching natural conversation (pp. 23–40). TESOL.
Bardovi-Harlig, K., & Stringer, D. (2017). Unconventional expressions: Productive syntax in the L2 acquisition of formulaic language. Second Language Research, 33(1), 61–90. https://doi.org/10.1177/0267658316641725
Bardovi-Harlig, K., & Su, Y. (2018). The acquisition of conventional expressions as a pragmalinguistic resource in Chinese as a foreign language. The Modern Language Journal, 102(4), 732–757. https://doi.org/10.1111/modl.12517
Bardovi-Harlig, K., & Su, Y. (in press). Developing an empirically-driven aural multiple choice DCT for conventional expressions in L2 pragmatics. Applied Pragmatics.
Bardovi-Harlig, K., & Vellenga, H. E. (2012). The effect of instruction on conventional expressions in L2 pragmatics. System, 40(1), 77–89. https://doi.org/10.1016/j.system.2012.01.004
Barron, A. (2019). Using corpus-linguistic methods to track longitudinal development: Routine apologies in the study abroad context. Journal of Pragmatics, 146, 87–105. https://doi.org/10.1016/j.pragma.2018.08.015
Biber, D., Contrad, S., & Cortes, V. (2004). If you look at…: Lexical bundles in university teaching and textbooks. Applied Linguistics, 25 (3), 371–405. https://doi.org/10.1093/applin/25.3.371
Biber, D., Johansson, S., Leech, G., Conrad S., & Finegan, E. (1999). Longman grammar of spoken and written English. Longman Publication Group.
Derakhshan, A., & Eslami, Z. R. (2020). The effects of metapragmatic awareness, interactive translation, and discussion through video-enhanced input on the Iranian upper-intermediate EFL learners’ comprehension of implicatures. Applied Research on English Language, 9(1), 637–664. https://doi.org/10.22108/ARE.2019.118062.1476
Dörnyei, Z., Adolphs, S., & Durow, V. (2004). Individual differences and their effects on formulaic secquence acquisition. In N. Schmitt (Ed.), Formulaic sequences: Acquisition, processing and use (pp. 87–106). John Benjamins.
Erman, B., & Warren, B. (2000). The idiom principle and the open choice principle. Text, 20(1), 29–62. https://doi.org/10.1515/text.1.2000.20.1.29
Félix-Brasdefer, J. C. (2004). Interlanguage refusals: Linguistic politeness and length of residence in the target community. Language Learning, 54(4), 587–653. https://doi.org/10.1111/j.1467-9922.2004.00281.x
Furniss, E. A. (2016). Teaching the pragmatics of Russian conversation using a corpus-referred website. Language Learning & Technology, 20(2), 38–60.
Kinginger, C. (2008). Language learning in study abroad: Case studies of Americans. Modern Language Journal, 92(Supplement 1), 1–124. https://doi.org/10.1111/j.1540-4781.2008.00821.x
Osuka, N. (2017). Development of pragmatic routines by Japanese learners in a study abroad context. In I. Kecskes & S. Assimakopoulos (Eds.), Current issues in intercultural pragmatics (pp. 275–296). Benjamins.
Pérez Vidal, C. P., & Shively, R. L. (2019). L2 pragmatic development in study abroad settings. In N. Taguchi (Ed.), The Routledge handbook of second language acquisition and pragmatics (pp. 355–371). Routledge.
Platt, J., & Weber, H. (1979). The position of two ESL varieties in a model of continuity and tri– dimensionality. In J. C. Richards (Ed.), New varieties of English (pp. 112–130). Regional English Language Center.
Roever, C. (2005). Testing ESL Pragmatics: Development and validation of a web–based assessment battery. Peter Lang.
Roever, C. (2012). What learners get for free: learning of routine formulae in ESL and EFL environments. ELT Journal, 66(1), 10–21. https://doi.org/10.1093/elt/ccq090
Roever, C., Wang, S., & Brophy, S. (2014). Learner background factors and learning of second language pragmatics. International Review of Applied Linguistics, 52(4), 377–401. https://doi.org/10.1515/iral-2014-0016
Sánchez-Hernández, A., & Alcón-Soler, E. (2019a). Pragmatic gains in the study abroad context: Learners’ experiences and recognition of pragmatic routines. Journal of Pragmatics, 146, 54–71. https://doi.org/10.1016/j.pragma.2018.08.006
Sánchez-Hernández, A., & Alcón-Soler, E. (2019b). The role of individual differences on learning pragmatic routines in a study abroad context. In P. Salazar–Campillo & V. Codina–Espurz (Eds.), Investigating the learning of pragmatics across ages and contexts (pp. 141–164). Brill.
Schauer, G. A. (2004). “May you speak louder maybe?” Interlanguage pragmatic development in requests. EUROSLA Yearbook, 4, 253–273.
Shively, R. L. (2011). L2 pragmatic development in study abroad: A longitudinal study of Spanish service encounters. Journal of Pragmatics, 43, 1818–1835. https://doi.org/10.1016/j.pragma.2010.10.030
Taguchi, N. (2013). Production of routines in L2 English: Effect of proficiency and study–abroad experience. System, 41(1), 109–121. https://doi.org/10.1016/j.system.2013.01.003
Taguchi, N. (2018). Contexts and pragmatics learning: Problems and opportunities of the study abroad research. Language Teaching, 51(1), 124–137. https://doi.org/10.1017/s0261444815000440
Taguchi, N., Li, S., & Xiao, F. (2013). Production of formulaic expressions in L2 Chinese: A developmental investigation in a study abroad context. Chinese as a Second Language Research, 2(1), 23–58. https://doi.org/10.1515/caslar-2013-0021
Taguchi, N., & Roever, C. (2017). Second language pragmatics. Oxford University Press.
Wesche, M., & Paribakht, T. (1996). Assessing second language vocabulary knowledge: Depth vs. breath. Canadian Modern Language Review (1), 53, 13–40. https://doi.org/10.3138/cmlr.53.1.13
Wyner, L., & Cohen, A. D. (2015). Second language pragmatic ability: Individual differences according to environment. Studies in Second Language Learning and Teaching, 5(4), 519–556. https://doi.org/10.14746/ssllt.2015.5.4.2
Yang, J. (2016). CFL learners’ recognition and production of pragmatic routine formulae. Chinese as a Second Language, 51(1), 29–61. https://doi.org/10.1075/cld.51.1.02yan
Appendix Full Script for the Aural MC-DCT
Part A
- In this part, you will see a description on the screen. Follow along with the speaker. Imagine that you are speaking to a friend or a teacher. What would you say to him or her in this situation?
- Below each description, you will see the phrase “You Say,” with four possible audio responses. The responses will be played twice. At the first listening, you can take notes if you want. After the second listening, you have 7 seconds to make your choice by circling the correct letter (A, B, C, or D) on the answer sheet.
# | Prompt | A | B | C | D |
Example 1 | Many of your friends are going to the movies, but you don’t have a car. You ask one of your friends for a ride in his car. | Can we go with you? | Would you please give me a ride? | Can I get a ride? | Can you come to pick me up at my apartment? |
Example 2 | You had a birthday party in your home yesterday. The apartment is untidy and you are just cleaning up. Your friend, John, comes by. You invite him in. | Sorry about that. | Sorry about the mess in the party. | Come on in. | Sorry, my room is so dirty. |
A1
Broken Chair |
You see your friend standing on a chair trying to reach a book at the top of the bookshelf. You know that the chair she is standing on has a broken leg. | Be careful! | Can I help you? | The chair is broken. | Watch out! |
A2
Library Closing |
You are in the library and you see an old friend, whom you have not seen for a long time. You talk for a little while, and as you are leaving, | Nice to see you. | Long time no see. | Nice to meet you. | We should get together. |
A3
Movies |
You are in the theater. There is a group of young teenagers sitting behind you. They are talking so loudly that you cannot hear a word. | Would you please shut up? | Could you be quiet? | Excuse me. | Be silent. |
A4
Busy Teacher |
You stop by your teacher’s office to ask a question about the assignment. She takes time to answer your question. You know she is very busy, so before you say goodbye, | I’m sorry to bother you. | I’m so sorry. | Thank you very much. | Thank you for your time. |
A5
Bus |
You and a friend are about to cross the street when you see the campus bus coming. Your friend does not see the bus and is about to step in front of it. | The bus is coming. | Be careful! | Watch out! | Pay attention! |
A6
25 Minutes |
You made an appointment with your teacher. Unfortunately, you arrive 25 minutes after the meeting time, and the teacher is already leaving. | I’m sorry about that. | I’m sorry I’m late. | I’m really sorry. | I’m sorry. |
A7
Cellphone |
You are at the bus stop. While waiting, you are talking with your friend on your cell phone. The bus arrives and you need to hang up. | Wait for me. | I gotta go. | The bus is coming. | I need to get off. |
A8
Puddle |
After class you’re walking to the library with a friend. It’s been raining all morning, and you notice that your friend is about to step into a big puddle. | Watch out! | Be careful! | Pay attention! | There’s a puddle. |
Part B
- In this part, you are talking to your friend or teacher, and they speak first. What would you say to him or her in this scenario?
- Below scenario description, you will see the phrase “You Say”, with four possible responses that are audios. The responses will be played to you twice. At the first listening, you can take notes if you want. After the second listening, you have 7 seconds to make your choice by circling the correct letter (A, B, C, or D) on the answer sheet.
# | Prompt | A | B | C | D |
Example 1 | You borrowed a book from your friend, Kate. You promised to return it today. She needs it for her presentation in class tomorrow. However, you left the book at home. You meet her in class. (Kate: “By the way, did you bring my book?”) | I’m sorry I didn’t. | I’m so sorry. | Oh, I’m sorry I forgot. | Oh I’m sorry I will bring it. |
Example 2 | Your teacher invited the whole class to his house next Saturday. You are very happy that he has invited you, and you would like to go. When you are leaving the class, the teacher says: “How about you? Would you be able to join us this Saturday?” | Of course. I want to go. | Sure, sure, I’d like to. | Sounds good. | Sure, I’d love to. |
B1
Shopping (help) |
You go to a clothing store and you need to find a new shirt. A salesperson approaches you. You want the salesperson’s assistance. (Salesperson: “Can I help you?”) | Yes, I need some help. | I’m looking for a shirt. | I want a shirt. | Sure, of course. |
B2
Offer of help |
You need to pick up a book at the bookstore, but you don’t have any free time today. (Your friend: “I can pick it up for you.”) | Thank you. | That’s alright. | Ok, I appreciate it. | That’d be great. |
B3
5 minutes |
You made an appointment with your teacher. Unfortunately, you arrive 5 minutes late for the meeting. Your teacher says: “Hello. Come on in.” | I’m so sorry. | Sorry about my delay. | I’m sorry I’m late. | Thank you. |
B4
Father (condolences) |
You go to ask your teacher if he will be having office hours tomorrow and he tells you about his father: “I won’t be having office hours tomorrow. My father died and I have to go to the funeral.” | Ok, no problem. | I’m sorry about that. | I’m sorry. | I’m so sorry. |
B5
Shopping (no help) |
You go to a clothing store and you need to find a new shirt. A salesperson approaches you. You don’t want the salesperson’s assistance. (Salesperson: “Can I help you?”) | No thank you. | No thanks, I’m just looking. | Well, thanks. I’m fine. | I don’t need help. |
B6
Party |
There is a reception on campus. The organizer invited you and a few other students as well. It is getting late, and you decide to leave. You go over to the organizer. (The organizer: “Thanks for coming.”) | I’m sorry. | You’re welcome. | Thanks for inviting me. | I have to go. |
B7
Dog hit by car |
You’re talking outside with your long-time neighbor, and she tells you about her dog’s accident: “Last Sunday my dog got hit by a truck.” | I’m sorry. | How did that happen? | I’m sorry about that. | I’m so sorry. |
B8
More food |
You are having dinner at a friend’s house. Your friend offers you more food but you couldn’t possibly eat another bite. (Your friend: “Would you like some more?”) | No thanks. | No thanks, I’m full. | Really I’m full. | That’s enough. |
B9
Ride |
You give your classmate a ride home. He lives in the building next to yours. He gets out of the car and says: “Thanks for the ride.” | Don’t worry. | You’re welcome. | No problem. | My pleasure. |
B10
Make-up Test |
You have been studying very hard for your test, but on the morning of the test your alarm doesn’t go off and you oversleep. You ask your teacher for a make-up test. (Your teacher: “Ok, I’ll give you a make-up this time, but don’t let it happen again.”) | Thank you for your kindness. | Thank you. | Thank you so much. | I appreciate it. |
B11
Have a nice day! |
You are in the supermarket. After you pay, you are ready to pick up your bags. The cashier says: “Have a nice day!” | Thank you. | You too. | Have a nice day. | Good bye. |
B12 Introduction | Your friend introduces you to his new roommate. (Your friend: “This is my new roommate Bill.” | Nice to meet you. | Hi Bill, how are you doing? | Nice to see you. | Hi Bill, how are you? |
[back]
Copyright rests with authors. Please cite TESL-EJ appropriately. Editor’s Note: The HTML version contains no page numbers. Please use the PDF version of this article for citations. |