A native-like adult L2 Korean learner’s agentive acquisition of Korean relative clauses

RC tasks

Pierre performed RC comprehension and production tasks which were based on Nishikawa (2014). While Jeon and Kim (2007) and Lee-Ellis (2011) addressed L2 Korean RCs, they were focused predominantly (Jeon and Kim) or entirely (Lee-Ellis) on heritage speakers. Further, the four experts on Korean education (mentioned earlier in the subsection “Participant”) agreed—after closely examining the RC types and the RC sentences in Nishikawa as well as those in Jeon and Kim, and Lee-Ellis—that the seven Japanese RC types and the sentences in Nishikawa could also be used (adaptively and profitably) for this study.

The RCs in the comprehension and production tasks were adapted from Nishikawa (2014). ⁶ This researcher had brief sessions of task practice with Pierre, who had been found to be aware of all the vocabulary. Additionally, his knowledge of Korean nominative and accusative cases, and adnominal verbal suffixes was verified by a 15-item truth-value judgment test designed by the four experts on Korean education.

As in Nishikawa, the comprehension and production tasks each had 35 items—five tokens for each of the seven types of RC. The 35 items were distributed at random throughout the tasks. Pierre took approximately 25 minutes to complete the tasks. Concerning each item in the comprehension task, Pierre circled what he perceived to be the correct picture among a set of three pictures ⁷ presented to him, in accordance with recorded instructions. As for the production task, Pierre orally indicated the circled human/animal/inanimate object on a sheet of paper on which a pair of almost identical pictures ⁸ had been drawn. For example, he described the circled baby as in (6) after being shown two pictures where a mother is washing a (circled) baby and where a father is washing a baby.

(6) [eomma-ga ssiseoju-neun] aegi

 [mom-nom wash-rel.pres] baby

 “the baby whom a mother is washing”

Pierre’s responses in the comprehension task and those in the production task (audio-recorded) were reviewed for their accuracy. First, this researcher gave a score of 1 if Pierre circled the correct picture and a score of 0 if he did not in the comprehension task. Second, the audio-recorded responses in the production task—transcribed by this researcher—were scored by one of the above-mentioned experts on Korean education and the researcher. The inter-rater reliability estimate (kappa coefficient) was .99. Because scoring criteria were not clearly given in any of Nishikawa, Jeon and Kim (2007), and Lee-Ellis (2011), this researcher—after consulting with the four experts on Korean education—used the error types in L2 Korean RC production reported in Jeon and Kim. Specifically, a score of 0.2 was assigned to each of the four error types: adnominal inflection error, argument omission, tense error, and case marker error/omission. It was decided that Pierre’s response to an item in the production task would be scored 1 when it was native-like (error-free), scored 0.8 when it contained an error (or errors) of one of the types, scored 0.6 when it contained an error (or errors) of two of the types, and so forth; further, a score of 0 would be given when he failed to produce a response. The maximum score for the production and comprehension tasks was 35 each. Pierre’s task scores were not further analyzed because they were judged to be adequate for the purpose of the current study.

Phase one

This phase ran from the moment of Pierre’s arrival in Korea to that of his enrollment in a Korean learning program in a college. During this phase, he made every effort to learn Korean to attain his goals notwithstanding perceived prejudice and economic exploitation. Pierre’s investment resulted from his “learner agency in the face of larger social structures” (Menard-Warwick, 2005, p. 168). According to Norton Peirce (1995), language learners undertake an investment in the target language in the knowledge that they will obtain a wider variety of symbolic and material resources through such an investment. These learner backgrounds of Pierre, which are different from those of child immigrants in Nishikawa (2014) and of heritage speakers in Lee-Ellis (2011), seem to have led to his idiosyncratic learner agency. Pierre’s disempowering position as an outsider did not weaken his agency, consistent with Miller’s claim (2012). The following is what he commented:

Excerpt 1: Pierre’s report

Initially, I viewed myself as a powerless, poor foreign factory worker. This was what prompted me to be an active learner of Korean. I wanted to make money by keeping my factory job in this country of opportunities. To keep my job, I needed to become a fluent Korean speaker. I eagerly sought maximum opportunities to learn Korean. I even talked to myself in Korean in make-believe dialogues. I also found myself trying to learn Korean in dreams. While learning Korean through conversations with my Korean co-workers and other Koreans, I gradually shifted myself to being, in their/my eyes, a learner of Korean who could confidently and successfully participate in the conversations. I had a good co-worker and a good supervisor, who not only taught me well but supported my Korean learning well. Incidentally, I had to focus on listening and speaking abilities because I needed them most of all.

From Excerpt 1, it appears that Pierre’s agentive endeavors, including engagement in dialogues with himself (see Jenks, 2015), to learn Korean resulted in improved self- and interactive positionings and in reconstructed identity. Further, Pierre secured the help of Korean language informants among his colleagues/supervisors, though these were few in number. Miller (2012) contended that past/present affordances and the constraints that adult L2 learners have experienced contribute to their achievement of desired goals. Pierre’s learner agency also informed his language anxiety. He observed in the interview (not excerpted here) that his eager efforts to learn Korean kept him from having any anxiety about learning the language.

Interestingly, Pierre’s Korean language informants provided him with affordances restricted to L2 listening and speaking. He prioritized these two language skills because, although they are required most frequently in daily life (e.g. Flowerdew & Miller, 2005), he lacked them. Erickson (1986) contended that learners’ choices—predicated on their perception of what is valuable—heavily influence what they learn. In addition, his past experience of learning English in his home country was found to contribute to his focus on learning the two skills. As discussed in the literature review section earlier, agentive adult L2 learners’ history in person involves their L2 learning experiences.

Concerning history in person, it seems that the concept of trajectory can be beneficially considered. Trajectory is defined as “a path of development, often through a variety of social contexts, in which each step (or learning event) builds on the previous ones, though sometimes in unpredictable ways” (Menard-Warwick, 2005, p. 169). The gradual development of Pierre’s L2 Korean listening and speaking abilities led to more and more of his focus being placed on his reading ability, claimed to be considerably more difficult to develop than listening and speaking abilities (e.g. Moxley, 1982). According to Pierre, it seems that his first noticing and awareness of, and attention to (see Mackey, 2007), Korean RCs occurred through reading books for Korean children. Below is what Pierre stated:

Excerpt 2: Pierre’s report

While reading Korean children’s books, I repeatedly found certain new patterns in the sentences hard to understand. The patterns prevented me from fully understanding the contents of the books. I had not been exposed to such patterns during my Korean learning through conversations. The patterns included that of an object noun followed by what seemed like a verb and then by a noun. I knew, from my experience of learning Korean through conversations, that eul or reul attaches itself to a noun and makes the noun an object; but I did not know what the verb lookalike was and what it did in the pattern. Further, I realized that the pattern was not found in French/English. These caught my attention and interest. I repeatedly asked my Korean language supporters about the pattern and the verb lookalikes using different examples. While their explanations were not organized, I could slowly and gradually understand how the pattern worked.

The first, fourth, and fifth sentences in Excerpt 2 can be taken as indicating Pierre’s noticing of Korean RCs. Mackey (2006, p. 417) operationalized noticing as follows: “incidences of noticing of form were identified when learners’ reports indicated that they were aware of the fact that their production or comprehension of form was problematic or that the form was new to them.” Additionally, the sixth and seventh sentences seem to represent Pierre’s awareness of and attention to Korean RCs respectively. According to Swain and Suzuki (2010, p. 558), awareness is defined as “the subjective experience of noticing” and attention as “the process of selecting information for further processing.” Many L2 researchers have emphasized the importance of noticing, awareness, and attention (e.g. Gass & Mackey, 2006).

Pierre’s agency was found to have influenced and been influenced by these three processes related to learning of L2 Korean RCs. Specifically, the processes were facilitated by his knowledge of patterns in Korean which, similar to the RCs, consist of an object noun, a suffixed verb, and a noun. The attainment of this knowledge had been facilitated by his agency for developing L2 Korean reading ability. According to Pierre, he made agentive efforts to gain the knowledge of Korean RCs through negotiation of meaning with his Korean language informants and to organize the knowledge in his memory. That is, Pierre’s agency was activated by his noticing and awareness of, and attention to, Korean RCs. The informants’ explanations in Korean, however, were inadequate for him to grasp Korean RCs: RCs have been claimed to be one of the most difficult L2 structures (e.g. Wiechmann, 2014). Pierre observed that his love for reading books for Korean children, originally intended to develop his L2 reading ability and L2 cultural competence, had deepened to enable him to identify examples of RCs in use and to try to understand them using the inadequate explanations from the Korean language informants. His reading activity represented reading for pleasure or extensive reading (Duncan, 2010) which relates to agency (Cremin, 2014). As suggested in Williams (2005), Pierre’s reading for pleasure seems to have improved his knowledge of RCs to a degree.

Pierre commented that he desired to position himself as someone who could change Koreans’ stereotypes of low-paid foreign workers, including him: Koreans, in common with citizens of other industrialized countries, have stereotypically categorized these foreigners (e.g. Kim, 2009).

Excerpt 3: Pierre’s report

I want to be the one who can shift the way Koreans invariably view us foreign workers: they disdain us foreigners working in low-skilled jobs for low wages also because our Korean abilities are poor.

Shuck (2006) found that, in the US, L2 English speakers’ lack of proficiency in English was a co-determinant of L1 English speakers’ biased views of them. Pierre had believed that his fluent and accurate comprehension of complex Korean sentences involving patterns, which were mentioned above (see Kenstowicz & Sohn, 2008, for the patterns) and were different from English or French, would begin to help achieve his goal. Paiva (2011) asserted that a language learner agent affects, and is affected by, her or his social practices. It appears that Pierre’s agency in relation to his noticing and awareness of, and attention to, Korean RCs was exercised to influence the social practice of stereotyping (Leslie & Storey, 2003), albeit on a partial, contextual, and micro level.

Phase two

Phase two of Pierre’s acquisition of Korean RCs ran from the moment of his enrollment in a university Korean education program to that immediately preceding his participation in the current study. Concerning the formal education, Pierre agentively searched the internet for the relevant information. He then contacted a university which ran a free Korean education program for foreign workers on weekends. While the program cost him precious rests from his hard weekday work, Pierre gladly tolerated four-hour round trips to the university. He stressed that his intention to learn Korean RCs in a systematic way had been one of the prime causes of his enrollment in the program. Below is what he commented in the interview:

Excerpt 4: Pierre’s report

Before attending the language program, my understanding of the pattern was sketchy. That is, I understood it as something in which an object noun and a verb lookalike describe a noun that follows them. The teachers at the university taught me, by my badgering and imploring, the characteristics, roles, and examples of this and similar patterns. They taught me that the noun and the verb lookalike make up something called a relative clause. The teachers also confirmed my suspicion that relative clauses are very useful and important.

During his six-month participation in the intermediate-level Korean course, which along with other courses in the program focused not on grammar but on verbal communication, Pierre was enabled to strengthen/increase his knowledge of Korean RCs and produce them. Formal instruction has been found to have a positive effect on RC acquisition. Pavesi (1986) claimed that formal (instructed) learners acquired marked RCs better than informal learners. Further, Ammar and Lightbown (2004) found that L2 learners who had been taught relativization performed better on RCs than those who had not been. Pierre’s trajectory of learning was facilitated by his agency (as indicated in Excerpt 4). In particular, his enlistment of the help of the teachers of Korean in his learning of RCs led to his attention to his Korean writing ability, the reason being that the greatest hindrance to reading books (for Korean children and young adolescents)—that is, his inadequate knowledge of RCs—was overcome. Among the various factors involved in the complex mechanism of L2 writing (see Norris & Manchón, 2012), an ability to produce RCs appeared to have increased the most his enthusiasm for writing in Korean. Pierre also attributed this increase in enthusiasm to his perception of the high incidence of RCs in language use, which his Korean teachers had affirmed. It has been asserted that higher-proficiency L2 writers produce a greater number of, and more accurate, RCs (e.g. Ferris, 1994). Further, the quality of L2 writing has been measured by an increase in the use of formal features notably involving RCs (see Hyland, 2003).

To improve his production of RCs in writing (and in conversations), Pierre agentively made extensive use of a notebook. The preparation and use of a grammar notebook has been contended to be a helpful strategy for grammar learning (e.g. Oxford, 2011). In the notebook, he entered examples of RCs and his thoughts about them in Korean. Pierre observed that his repeated and cumulative reviews of the contents in the notebook contributed to what could be viewed as his internalization, ¹⁰ which refers to changing symbolic artifacts such as language into mental processes (Lantolf, 2006), of Korean RCs. The following is what he stated:

Excerpt 5: Pierre’s report

I often thought hard and deeply about the relative clauses while and after studying what I had written in my notebook. You see, I also used notebooks for my English learning while in my country. Additionally, I tried to use many relative clauses in my Korean writing as well as in conversations with Koreans. Concerning my Korean writing, no one asked or advised me to do it. I just did it independently to improve my Korean writing skill. I showed my writing to my Korean teachers. They corrected the incorrect relative clauses. I did my best to recognize the differences between the relative clauses of mine and those of my Korean teachers and conversation partners. I strived to work on the differences with the help of the Koreans. By the time the university Korean course was completed, I realized that Koreans had changed the way they treated me and my fellow foreign workers in a much more positive way. I believe that my efforts to learn Korean relative clauses have helped a lot.

In Excerpt 5, it seems that Pierre’s agentive production of many RCs could be considered from the perspective of the comprehensible output hypothesis (see Swain & Lapkin, 2001). According to this hypothesis, successful SLA requires comprehensible output as well as comprehensible input. Specifically, SLA can occur when learners are pushed to produce output designed to communicate their messages, because this can place them in a position to better notice the gap between their productions and proficient speakers’ productions (see Richards & Schmidt, 2010).

Pierre reported that he had perceived a change for the better in Koreans’ view of him and his fellow foreign workers because of his accurate and appropriate use of Korean RCs in writing and speaking. ¹¹ This change seems to indicate how the differences (i.e. among Pierre, and Nishikawa’s (2014) child immigrants and Lee-Ellis’ (2011) heritage speakers) in learner backgrounds and in learner agency could mediate the results/influence of L2 RC acquisition. Pierre’s Korean language informants, who had been engaged in explaining Korean RCs, recognized a substantial improvement in his knowledge of RCs and in the quality of his writing; the writing included notes for his colleagues/supervisors. The informants conveyed news of this improvement to other Koreans, who consequently expressed support/encouragement for Pierre and then for other foreign workers. Casanave (2012) indicated that the idiosyncrasies of L2 learning are formed by events in real life that involve workplace politics. Further, Norton and Toohey (2011) claimed that failure to invest in the language practices of workplaces causes discriminatory practices. Pierre believed that working conditions unfavorable to foreign workers, including him, could be improved, albeit partially, by the workers’ successful communication with their Korean colleagues/supervisors.

Pierre was found to have enjoyed his efforts on the trajectory of his acquisition of Korean RCs. Casanave (2012) reported that she had enjoyed her Japanese learning in Japan because she had been the agent of her own learning activities. It seems that Pierre’s agentive efforts, similarly, gave him the leverage to endure considerable difficulty and sacrifice inflicted on/required of him.