Second Language Acquisition Studies Observed in “The International Corpus of Japanese as a Second Language” (I-JAS) by Chinese Speakers: From the Perspectives of Pragmatic Transfer

Abstract

This study examines the pragmatic transfer from L1 Chinese to L2 Japanese based on the learners’ corpus. The research addresses the following two research questions: (1) What are the specific tendencies among native Chinese learners of Japanese in “request” expressions, compared with French, Spanish, and English learners? (2) Do Chinese speakers have specific tendencies to be affected by their native language? We analyzed the role play data of learners of Japanese whose native languages are Spanish, French, English, and Chinese. We discovered that “suspended clauses (incomplete sentences)”, such as “I have a favor to ask you, but …” which are frequently used by Japanese native speakers, are rarely used by the learners of Japanese: Spanish, French, English, and Chinese. However, native Chinese learners use the confirmation expressions more often, such as “is it OK?” at the end of the sentence than other language speakers, which native Japanese hardly use. We then examined pairs of Chinese native speakers by having them work on the same tasks in Chinese. We found they use the confirmation expressions often in Chinese to show politeness. Here are the results of this study: (1) the learners rarely used the “suspended clauses”, however, it was not specific to Chinese speakers; and (2) “the confirmation expressions” was observed more frequently among Chinese speakers compared with the other speakers, and it can be considered a negative transfer from learners’ native language, Chinese.

1 Importance of Language Acquisition Studies Using Learner Corpora

Learners make errors in the process of acquiring foreign and second languages. Great numbers of researchers and linguists have investigated the mechanisms behind learners’ errors (Schachter, 1974, Kellerman, 1979, Kellerman and Sharwood Smith, M. 1986, Odlin, 1989, 2003, Ellis, 2008). In the past, researchers used a contrastive analysis called the phenomena “interference” and “language transfer” but the more recent term is “cross linguistic influence” since L1 and L2 are both influenced through the language acquisition process.

Acquisition research on Japanese as L2 focused on language transfer also has been conducted since the 1970s, but no conclusions have been reached yet.

Research by Inaba (1991) targeting English native speakers, using true–false tests, concluded that differences between Japanese and English influenced the acquisition of conditionals. Ikoma and Shimura (1993) also reported pragmatic transfer found in their research on “refusal”, which targeted English native speakers using discourse completion tests and showed the tendency of their “refusals” to differ from those of the Japanese native speakers’ refusals. On the other hand, other research did not specify the influence of native languages, like Sakoda (1998), who tested the use of demonstratives by native speakers of Chinese and Korean, and Sugaya (2004), who tested grammar, targeting native speakers of German, Russian, English, and other languages.

According to Okuno (2000), results may differ if research targets native speakers of one language or multiple languages. She analyzed the influence of native language and indicated that while 83% of research targeting native speakers of one language considered whether native language transfer had occurred, only 46% of the research concluded there was a possibility of language transfer.

The above result suggests that in order to argue for the effect of native language, it is necessary to examine data of learners with a greater variety of native languages. Compared to English learners’ data, there is little data from learners of Japanese, and it is mostly limited to native speakers of English, Chinese, and Korean. Therefore, this study examined whether or not the Chinese language has effects on the acquisition of Japanese, using the “International Corpus of Japanese as a Second Language: I-JAS” which was released in 2020. This research addresses the following two research questions:

• (1) What are the specific tendencies among native Chinese learners of Japanese in “request” expressions in Japanese, compared with native French, Spanish, and English learners?

• (2) Is there a strong possibility that the trends specific to Chinese speakers are affected by their native language?

2 I-JAS: International Corpus of Japanese as a Second Language

2.1 Conventional Japanese Learner Corpora

Corpora are language resources in which a large volume of texts and utterances (by learners) is compiled into a database. Japanese learner corpora started from the collection of error examples and compositions, as Japanese language education began to spread. In the 1990s, corpora of compositions and utterances by Japanese learners began to appear.

The KY corpus includes data from the utterances of 90 learners, collected from the OPI (Oral Proficiency Interview), which is an oral ability assessment. The assessment is divided into 9 levels from novice to superior for 30 English, 30 Chinese, and 30 Korean speakers.

The conversational database (cross-sectional survey) of the National Institute for Japanese Language and Linguistics mainly contains 30 min of conversation by 339 Korean native speakers, as well as native speakers of other languages such as English, Chinese, and Indonesian. C-JAS (Corpus of Japanese as a second language) is a longitudinal corpus of Japanese learners. It is a collection of about one-hour utterance data taken over 3 years from 3 Chinese and 3 Korean speakers. There is another data set, LARP at SCU (Language Acquisition Research Project at Soochow University), which collected mainly composition data of 37 Taiwanese college students over 4 years.

However, there are some issues with conventional Japanese learner corpora. First, the number of learners is low. Second, learners’ native languages are not balanced. (Most corpora contain data from English, Chinese, and Korean native speakers; data from other languages is insufficient.) Third, the proficiency levels of the learners in the data are not clear. Finally, information on the learners’ backgrounds is unavailable.

2.2 Summary and Features of I-JAS

Considering the issues with conventional Japanese learner corpora, we released the “International corpus of Japanese as a second language (I-JAS)” in March 2020. This study analyzes the role play data included in I-JAS. This section gives a summary of the features of I-JAS.

I-JAS has been constructed for the purpose of elucidating the effects on language acquisition processes in different language environments, including differences in learners’ native languages. It consists of data collected through 4 years of research, targeting learners in 20 educational institutions in 17 different countries and areas, including Japan. The following explanation summarizes the five main features of I-JAS:

• 1) I-JAS collects data from a total of 1050 research subjects. It is the largest corpus of its kind, with data collected from 1000 Japanese language learners and 50 native speakers of Japanese. There are 850 JFL learners with 12 different native language backgrounds, 100 Japanese learners studying in educational institutions within Japan, and 50 JSL Japanese learners who do not study at educational institutions, for example, those who are married to Japanese or who work in Japan and with their families. The 12 native languages of the learners in the corpus are English, Chinese, Korean, German, French, Vietnamese, Russian, Spanish, Indonesian, Hungarian, Turkish, and Thai. These languages were chosen to provide balanced data from the viewpoint of linguistic typology.

  The effect on the learning process of different learning environments, whether outside or within Japan, will emerge by comparing the data taken from JFL learners (outside Japan) and JSL learners (those who learn in a classroom environment in Japan). The effect of classroom teaching on the learning process can be clarified by comparing the data taken from 150 JSL learners, of which there are 100 classroom learners and 50 natural environment learners. Furthermore, the data of 50 Japanese native speakers who completed the same tasks is available in I-JAS, and can be compared to the data described above so that the differences between native speakers and learners in each category can be clarified. The role of the learning environment can also be revealed by comparing different learning environments: within or outside Japan, and classroom or natural setting.

• 2) The second feature is that all the targeted learners have taken two kinds of Japanese proficiency tests, and their results are published. The tests are J-CAT (Japanese Computerized Adaptive Test) and SPOT (Simple Performance-Oriented Test). J-CAT is an adaptive test, to which item response theory is applied (the question items are selected and presented based on the test taker’s ability). J-CAT consists of 4 categories, “listening, reading, vocabulary, and grammar”, which makes it similar to the Japanese Language Proficiency Test (JLPT). As J-CAT is available online, test takers are able to take the test anytime and anywhere by applying to register. They are also able to obtain the results when they complete the test (https://j-cat.jalesa.org/). SPOT is one of the tests within the TTBJ (Tsukuba Test Battery) developed by Tsukuba University, which consists of 90 questions. It is a fill-in-the-blanks test, where test takers identify one Hiragana within a sentence played to them (https://ttbj.cegloc.tsukuba.ac.jp/). I-JAS includes information regarding learners’ scores on both tests, which makes it possible for researchers to objectively judge learners’ Japanese language ability and compare data by native language or different learning environment using the data of learners who have equal Japanese ability based on their scores.

• 3) The third feature is that I-JAS contains data from seven different tasks, including utterances and compositions, which can be used according to the objectives and target of the research. The tasks shown in (3) are included in I-JAS.

(3) Tasks included in I-JAS:

a	Story Telling	oral data	Learner narrates a story based on a 4–5 frame comic
b	Interview	oral data	One-on-one 30-min structured conversation between researcher and learner
c	Role Play	oral data	Learner plays the role of a part-time worker, talking to his/her boss. The worker makes a certain request to the boss in one role play and refuses the boss’s request in the other
d	Picture Description	oral data	Learner looks at a drawing and freely describes what appears in it
e	Story Writing	written data	Learner looks at the same 4–5 frame comic and types the story into a computer
f	E-mails	written data	Learner writes 3 different email responses, including “request” and “declining”
g	Essay	written data	Learner writes his/her own opinion on fast food and home cooking in an essay of approximately 600 characters

Data was collected from 1000 learners and 50 Japanese native speakers for tasks a. ~ e. For tasks f. and g., however, as it would take a long time to complete along with the other tasks within the same day, the tasks were given to learners who volunteered. Learners worked on these tasks at home, not with the researcher.

Two topics, “picnic” and “key”, were prepared for tasks a. and e. At first, the pictures were shown to the learners to confirm that they understood the story before they completed the tasks. For story writing (task e.), learners were told they were allowed to write a different story to the one they had narrated first, and that they were allowed to write while they planned the story. These two tasks may be used to study the effect of planning time on language use.

In task c., the role plays were conducted for request and refusal, which are close to realistic communicative contexts.

Since tasks a., ~ c., and e. include many descriptions of movements, a picture description task (task d.) was added, in which “ ~ teiru (~ing)” can be used to describe an ongoing situation. The same task topics were set for tasks a. (speaking) and e. (writing).

Tasks f. and g. are both composition tasks. The learners were given the tasks beforehand so that they could freely choose a place and time to work on them. Task f. consists of 3 email responses and task g. is a task to write approximately 600 characters based on the learner’s opinion. Learners write the task as an entry to an essay contest on the theme, “Our diet: slow food and home cooking”.

• 4) The fourth feature is that I-JAS contains various additional information. For example, for the utterance data, the actual audio data is provided together with a transcript, so that both can be used. The compositions were written only by the learners who volunteered, and the number of compositions data is not the same as that of the utterances. Note that there is only data from those learners who volunteered to complete the tasks. Information regarding the time taken and references used are published along with the compositions themselves.

  The second type of additional information is the learners’ background information. Since learners’ home country, environment, and other factors are different, we conducted a questionnaire on their background beforehand, and published the information related to their Japanese learning, such as their learning environment (classroom, natural setting, etc.), family (family structure, language used, etc.), part-time job (whether they use Japanese at work or not), and their Japanese learning style.

  The third type of additional information is utterance and composition data collected from 50 Japanese native speakers who completed the same tasks. They are a well-balanced group for comparison with a mix of male and female speakers in their twenties to fifties.

• 5) The last feature is that morphological information is added to I-JAS. I-JAS uses the corpus search application, “CHUNAGON”, which enables not only string searches, but also searches using morphological information. This makes it easier to conduct quantitative analysis of Japanese learner corpus data. It also enables analytical research from various fields, such as vocabulary and morphology studies.

This section has given an overview of I-JAS and its 5 main features. The next chapter will focus on analyzing the data from request expressions used in the role play in I-JAS, in order to examine whether or not Chinese learners of Japanese are influenced by their native language.

3 Previous Studies on Japanese Learners’ “Request and Refusal”

Research on Japanese learners’ “request and refusal” increased in the 1990s together with the growth in the number of language acquisition studies from the viewpoints of social linguistics or pragmatics. For instance, Ikoma and Shimura (1993) targeted English native speakers studying Japanese as a foreign language, comparing their refusal expressions in English and Japanese (L2) with that of Japanese native speakers. The influence of English was observed in the learners’ Japanese, which indicated the possibility of pragmatic transfer from their native language. Specifically, it was reported that Japanese native speakers have a tendency to suggest an alternative idea when turning down a request and use incomplete sentences, such as “… desuga [however, but…]”, when turning down their superior’s request. While English native speakers do not suggest alternative ideas and directly turn down their superiors without using incomplete sentences. It was suggested that this tendency indicates a pragmatic transfer from English expressions.

Izaki (2000) researched deviation and unsuitability observed in requests made by French learners of Japanese. The study showed that Japanese native speakers provide a preliminary and introductory step before requesting, such as “In fact, I have something that I’d like to ask you …”, whereas, in the case of the learners, such a preliminary step is almost never provided. Izaki observes that while “asking for changes” is regarded as a “request” in Japan, learners have a tendency to consider it as “negotiation”, thus suggesting that request expressions are influenced by social and cultural differences between Japan and France.

Other studies on request expressions, such as Kashiwazaki (1992), Samejima (1998), and Lee (2008), suggest that Chinese learners of Japanese have a tendency to use “complete sentences” and “direct expressions”.

However, there are some issues with the past studies. First, much of the data was collected from Japanese learners of a single native language. As pointed out in Sect. 1, it is necessary to target Japanese learners of several native languages to discuss the effect of the native language. The results of the previous studies may, therefore, not be accurate. Secondly, many of the studies used written data from conversation completion tests, which may not be the actual language that learners of Japanese use.

4 Data Analysis of I-JAS Based on “Request” Role Play

This chapter will introduce the two research studies (Sakoda et al., 2017) which use.

I-JAS data, taking into consideration the issues indicated in the previous chapter.

4.1 Suspended Clauses by L2 Learners of Japanese

The data for analysis was taken from a total of 60 targeted learners studying Japanese overseas: 15 French native speakers, 15 Spanish native speakers, 15 Chinese native speakers, and 15 English native speakers. As a comparison group, the data of 15 Japanese native speakers was used. The learners’ Japanese levels were measured based on the results of Japanese language ability tests. The 15 learners in each native language group were chosen after their Japanese ability had been shown to be homogeneous, based on statistical tests.

• (4) Extract from an I-JAS role play instruction card

You are working as a part-time staff member at a Japanese restaurant. (…) Now you work three days a week. However, you want to change to working two days a week as you have got busier. Please tell the restaurant manager that you want to change the number of workdays from three days to two days and get his/her permission. (Please indicate to the researcher when you are ready to begin.)

The task is a role play in which a learner and a researcher (a Japanese native speaker) converse one on one. The content of the role play card is shown in (4). The card is written in the learner’s native language. After having the learner read the card silently, we confirmed that the learner understood the content and then began the role play. The Japanese native speaker (researcher) played the role of the restaurant manager, and the learner played the role of the part-time staff member and student. The researchers in their roles as a manager would not immediately accept the student’s request and would come up with several reasons to dissuade the student from decreasing their shifts in order to continue a back and forth discussion. Conversations were recorded and transcribed.

The first half of the role play where the part-time member of staff begins to talk to the manager, brings up the topic, and mentions his/her request, was divided into three sections, “introduction, precondition, and request”, for analysis. The following is the example as shown in (5).

• (5) Introduction (A), e.g. Anoo, go-soodan ga arundesu ga… [Excuse me, well, I have something to discuss with you, and…]

  Precondition (B), e.g. Ima, shuu mikka hataraite irundesu kedo… [Right now, I work three days per week, but …].

  Request (C), e.g. Shuu futsuka ni kaetaindesu kedo… [I would like to change to two days per week, and …].

  The sentences used for each utterance were categorized as in (6).

• (6) Suspended clauses (incomplete sentences),

  e.g. Ohanashi shitai koto ga arundesu ga… [I have something to discuss with you, and …].

  Question sentences: e.g. Ima sukoshi yoroshii deshoo ka. [Do you have a minute now?].

  Declarative sentences: e.g. Tenchoo, hanashiga arimsu. [Mr. –-, I need to talk to you.]

• (7) Breakdown of sentence types in Introduction section (A) by each group of native speakers (Sakoda, 2016: 105) Suspended clause: SC, Question sentence: QS, Declarative sentence: DS

Japanese native speakers			French native speakers			Spanish native speakers			English native speakers			Chinese native speakers
SC	QS	DS	SC	QS	DS	SC	QS	DS	SC	QS	DS	SC	QS	DS
90%	0%	10%	17%	50%	33%	33%	33%	33%	27%	55%	18%	27%	18%	55%

• (8) Breakdown of sentence types in Request section (C) by each group of native speakers (Sakoda, 2016: 106) Suspended clauses: SC, Question sentence: QS, Declarative sentence: DS

Japanese native speakers			French native speakers			Spanish native speakers			English native speakers			Chinese native speakers
SC	QS	DS	SC	QS	DS	SC	QS	DS	SC	QS	DS	SC	QS	DS
73%	20%	7%	27%	53%	20%	13%	47%	40%	7%	73%	20%	0%	80%	20%

The results of (7) and (8) indicated the following:

• ① A high proportion of the sentences used suspended clauses, such as “I am sorry to bother you, but …” by Japanese native speakers in both the introduction section (A) and the request section (C). However, the proportion of such sentences is very low among learners, regardless of their native language.

• ② On the other hand, a high proportion of the sentences that learners use in both the introduction section (A) and the request section (C) are questions or declarative sentences. Examples are shown in (9) and (10).

• (9) Examples of question sentences:

  • a. Tenchoo, ima hima desu ka. [Mr. –-, are you free now?] (Introduction part (A), French native speaker)

  • b. Tenchoo, jikan ga arimasu ka. [Mr. –-, do you have time?] (Introduction part (A), Chinese native speaker)

  • c. Futsuka no hi-dake iideshoo ka. [It is OK to have two days only?] (Request part (C), Spanish native speaker)

• (10) Examples of declarative sentences:

  • a. Onegai ga arimasu. [I have a favor to ask you.] (Introduction part (A), French, English, Spanish, and Chinese native speakers)

  • b. Hanashi ga arimasu. [I have something to talk to you about.] (Introduction part (A), Chinese native speaker)

  • c. Shuu futsuka shitai to omoimasu. [I’d like to work twice a week.] (Request part (C) Chinese native speaker)

  • d. Mikka-kan wo futsuka-kan ni shite kudasai. [Please change three days to two days.] (Request part (C) Spanish native speaker)

This result shows that Japanese learners, when compared with Japanese native speakers, have a tendency to use declarative and question sentences but not suspended clauses (incomplete sentences) in the request role play. This tendency is commonly observed among learners despite different native languages: French, Spanish, English, and Chinese. This is, therefore, not a feature specific to Chinese native speakers, so we cannot assume the possibility of influence from their L1, Chinese.

4.2 Expression of Confirmation by L2 Learners of Japanese

It was shown in the previous section that the low rate of suspended clauses use in the request role play was not a phenomenon unique to native speakers of Chinese. Is it true to say that no specific tendencies can be observed among Chinese learners of Japanese? In the request section (C) of Chinese speakers’ utterances, expressions like (11) were prominent.

• (11) …. since I am busy, working three times a week is not possible for me, and I’d like to work twice. Is it OK?

The learner expressed his/her wishes by saying “I’d like to work twice”, then tried to make sure by saying “Is it OK?” For Japanese native speakers, this expression of “confirmation” gives the impression of forceful reminding, and thus has a high risk of leaving a highly unpleasant impression on the listener if they are older or one’s superior. We analyzed the data to see if such “confirmation” expressions can be observed particularly among Chinese speakers (Sakoda et al., 2017).

The utterances of 90 learners in the request role play were analyzed in total; 20 French speakers, 20 Spanish speakers, 20 English speakers, 30 Chinese speakers, and 15 Japanese native speakers. The Chinese group was divided into three levels according to their J-CAT scores: top, upper, and lower levels. The other groups were divided into upper and lower levels. The average J-CAT scores in each native language group are shown in (12). The Japanese abilities of learners are not homogeneous; the scores of the upper and lower French and Spanish speaker groups are lower than those of the English and Chinese speaker groups. The scores are almost identical, however, between the Spanish and French groups and the English and Chinese groups, respectively.

• (12) Average scores of J-CAT in each native language group (n = 10) (Sakoda et al., 2017: 56)

Level	Lower	Upper	Highest	Level	Lower	Upper	Highest
French	127.4	201.2	-	English	185.1	234.4
Spanish	126.5	198	-	Chinese	184.6	234	296.6

The chart in (13) shows the number and proportion of learners who used “confirmation” expressions in the request role play.

• (13) The number and proportion of “reminder” expressions in the role play utterances

Speakers	Spanish	French	English	Chinese
Lower level (n = 10)	1 (10%)	2 (20%)	1 (10%)	5 (50%)
Upper level (n = 10)	0	1 (10%)	0	1 (10%)
Highest level (n = 10)	-	-	-	1 (10%)

The results above show that the rate is higher among the lower-level Chinese speakers, half of whom used confirmation expressions. Learners in the upper and highest level were still using the confirmation expressions, which indicates that Chinese speakers specifically have a tendency to use this expression, compared with learners of other native languages. Example utterances are given in (14) ~ (16).

• (14) Ettoo, ee, ni, nichi nichi ga hatarai desu, iidesu ka. [Well, let’s see, two, I want to work for two days. Is it OK?] (Lower level).

• (15) Isshuukan, tabun, futsuka, kite, anoo, ikagadesu ka. [Per week, maybe two days, I’ll come. How about this?] (Upper level).

• (16) Isshuukan futsuka ni natte hoshii to omoimasu, iidesuka.[Per week, I’d like it to be two days. Would it be all right?] (Highest level).

The language form changed from a regular form, “how about this?” uttered by a learner of the upper-level group, to a polite form, “Would it be all right?”, spoken by a learner of the highest group. However, in both cases, it sounds as if the speaker is demanding the listener’s response, which gives an unpleasant impression.

Sakoda et al., (2017) conducted an experiment to ascertain whether similar expressions were used in the utterances of native speakers of Chinese. Analysis of conversations by 12 pairs of Chinese native speakers showed that 8 out of 12 speakers (75%) used expressions equivalent to Japanese confirmation expressions. Examples are given in (17) and (18).

• (17) 所以最近有点忙, 啊, 没法再延续之前, 三天, 一周三天的打工了, 想改成两天, 您看可以吗?

  Therefore, I’ve been busy recently, I can’t continue like before working three days, three days a week. I’d like to change it to two days, do you think that’s all right?

• (18) 店长最近由于我学业比较忙, 我想把打工时间由三天改到两天,行吗?

  Boss, I’ve been pretty busy recently with my studies, I’d like to change my work from three days to two days, is that okay?

From this result, we were able to infer that Japanese learners of Chinese speakers state their wishes first, using the Japanese format “I want to ~ ”, then try to show respect to the listener, the restaurant manager, by “asking for permission” or “asking their opinions”.

In Japan, however, speakers tend to avoid directly conveying their wishes to superiors or people older than them. Repeating those wishes using confirmation expressions can frequently leave an even ruder impression. In general, in Japan, frequent use of this kind of “confirmation expressions” can be regarded as pushy. As opposed to “confirmation expressions”, Japanese native speakers frequently use “suspended clauses”, where they omit the last part of the sentence and have the superior or older listener guess their wishes. An example of this kind of suspended clauses is as follows: “If possible, I would like to change to two days a week, but …”. Thus, expressions of “politeness” in Japanese and Chinese clearly differ. Paying attention to grammatical accuracy alone is insufficient. A lack of understanding of differences in communication between different cultures may inadvertently leave a bad impression and lead to trouble in relationships.

5 Discussion on Pragmatic Transfer and Summary

We summarize and discuss findings from the pragmatics data analysis of the two issues shown below:

• (1)’ What are the specific tendencies among native Chinese learners of Japanese in “request” expressions in Japanese, compared with French, Spanish, and English learners?

• (2)’ Is there a strong possibility that specific tendencies are affected by learners’ native language, Chinese?

Using data from the I-JAS request role play, we compared native speakers of French, Spanish, English, and Chinese. We discovered that “suspended clauses”, such as “I have a favor to ask you, but …” and “I’d like to ask you to let me … but…”, which are frequently used by Japanese native speakers, are rarely used by learners. Therefore, it became clear that this was not a tendency specific to Chinese native speakers. Some previous studies point out that the use of suspended clauses is rare, but many of them reached that conclusion by targeting learners of only one native language. Our research, on the other hand, reached the same conclusion after comparing and analyzing data from learners of several native languages, divided by Japanese proficiency levels. Although the previous studies indicated that this tendency (of not using suspended clauses) might be affected by the learner’s native language, our research is significant because it demonstrated that this may not be correct. We found that native speakers of French, Spanish, English, and Chinese all struggled to use “suspended clauses”, and there were no notable differences among the speakers of these languages. However, the effect of native language cannot be discounted unless research clarifies the tendencies of learners whose native language is Korean or Turkish. Predicates in these languages are located at the end of sentences, which is similar to Japanese sentence structure.

As for the confirmation expression such as “Is it OK?”, we found that this expression was observed more frequently among Chinese native speakers, compared with the speakers of other languages. We then examined pairs of Chinese native speakers by having them work on the same tasks in Chinese to see if they showed the same tendency. Chinese native speakers were found to frequently use confirmation expressions such as “Is it OK?” or “Would it be all right?”, after expressing their requests and wishes. Therefore, it is possible that the use of these confirmation expressions is affected by the Chinese language. While this kind of expression can be considered as a form of politeness in China, it gives the impression of disrespect in Japan. The use of confirmation expressions in Japanese by Chinese speakers can be considered a negative effect on learners’ native language, Chinese. This suggests that learning a second or foreign language is not just a matter of learning the language forms, but also requires the teaching of pragmatics, including awareness of differences in politeness between the culture of the target language and that of the native language. We would like to continue our studies in this area.