Proceedings of the 4th Annual JALT Pan-SIG Conference. May 14-15, 2005. Tokyo, Japan: Tokyo Keizai University.

Appreciation strategies of German and Japanese native speakers and German learners of Japanese
(Japanese Title)

by Kayoko Nakamura (Graduate School of the University of Tokyo)



Using a discourse completion test and a multiple choice test, this study detected cultural differences in the appreciation strategies of native German and Japanese speakers; the German speakers predominantly used thanking expressions in almost all situations, whereas the Japanese speakers used both thanking and apologetic expressions in the same situations. For the German learners of Japanese, while their discourse completion test data identified some pragmatic transfer, they used apologetic expressions more frequently than did the native Japanese on the whole as a result of hypercorrection. Furthermore, the effects of learning contexts, length of residence, and situational variables were also investigated. The learners of Japanese as a foreign language presented clearer evidence of pragmatic transfer than did those learning Japanese as a second language. The study also revealed the Japanese native speakers' great sensitivity to the situational variables, whereas these variables barely affected the strategy selections of the German native speakers and the learners of Japanese as a foreign language.

Keywords: appreciation strategies, pragmatic transfer, hypercorrection, length of residence, situational variables

It has been suggested that the goal of learning a second language involves the acquisition of communicative competence, which allows us an appropriate use of the target language in a given communication encounter (Tanaka & Kawade, 1982). But this is not an easy task even for advanced level learners since there are cross cultural differences in language values and customs, and employment of L1 communication strategies in L2, namely pragmatic transfer, can lead to various misunderstandings (Coulmas, 1981).

[ p. 44 ]

It has also been pointed out that Japanese frequently use apologetic expressions in thanking situations, which often perplexes learners of Japanese from Western cultures since giving thanks and apologizing seem to be two totally different speech acts to them. Coulmas (1981) maintains, however, that these speech acts are linked by the concept of "indebtedness," and that Japanese often apologize in thanking situations since they tend to focus on the trouble they have caused the hearer rather than the benefits they have received. Kumatoridani (1988) also explains this shift of viewpoint as a polite action since it involves 'switching into the addressee's point of view'. Other researchers such as Nakata (1989), Ogawa (1995), and Sakuma (1983) view that thanks and apology in Japanese form a continuum. Sumimasenis one such expression very frequently used in both situations. Today sumimasenis used in almost every encounter on the street (McClure, 2000), which gives the impression to non-Japanese that Japanese are constantly apologizing. On the other hand, Coulmas points out that Japanese learners of European languages often make the mistake of apologizing where not necessary. This point inspired me to try to find out whether the reverse is the case with German learners of Japanese.
". . . native Japanese speakers very often use apologetic expressions in non-apologetic situations . . . "

Miyake (1994a) compared the appreciation strategies of native Japanese and English speakers and concluded that native Japanese speakers very often use apologetic expressions in non-apologetic situations and that Japanese tend to feel apologetic more often than English speakers. Nakata (1989) also compared thanking and apologizing scenes from movies and TV dramas in Japanese and English and claims that Japanese apology rituals often refer to the hearer's voluntary favor, whereas English speakers usually show only gratitude for such favors. Therefore, the first aim of this study was to investigate whether any cultural differences could be seen in the appreciation strategies of native German and Japanese speakers.
The second focus of this study involved the interlanguage pragmatics of the German learners of Japanese. More precisely, it investigated whether they would show pragmatic transfer from their L1 in their selections of appreciation strategies. Moreover, in a similar comparative study, Takahashi & Beebe (1987) tried to investigate effects of learning contexts and identified more pragmatic transfer in the EFL (English as a foreign language) groups than in the ESL (English as a second language) groups. Furthermore, effects of length of residence also have been investigated by Olshtain & Blum-Kulka (1985), who report that learners' response patterns change over time as the function of the speakers' length of stay in the target language community. Therefore, these aspects were also explored in the current study using two learner groups, one with and one without living experience in Japan.
This study also attempts to examine the influences of situational variables such as social distance, power, and size of imposition suggested by Brown & Levinson (1987). Takahashi & Beebe (1987), Beebe, Takahashi, & Uliss-Weltz (1990) as well as Miyake (1994b) identified their Japanese subjects' greater style shifting according to the interlocutor's status compared to their American or English subjects' responses. Miyake also contends that the use of apologetic expressions by Japanese is more strongly affected by the relationship to the hearer rather than the size of the imposition. Thus, the current study investigates if this gap in the perceptions of social distance and power also applies between native Japanese and German speakers.

Research questions and hypotheses

Based on the above-mentioned past studies, the following research questions were addressed.
  1. What kind of differences can be found in the L1 thanking strategies of native German (GNS) and Japanese (JNS) subjects in their use of thanking and apologetic expressions?
  2. What are the appreciation strategies of German learners of Japanese? How much pragmatic transfer from L1 can be observed in their use of appreciation strategies in L2?
  3. What differences do the JSL (Japanese as a second language) group and the JFL (Japanese as a foreign language) group show in terms of the degree of assimilation to native Japanese norms?
  4. How much do situational factors (social distance, power difference, and size of imposition) affect the strategies of the four groups (GNS, JNS, JFL, and JSL)?

[ p. 45 ]

Accordingly, the following hypotheses were tested.


Four groups of subjects participated in this study and these subjects were confined to undergraduate and graduate students (Table 1). The JNS group had 62 native speakers of Japanese at the University of Tokyo and the GNS group also consisted of 62 native speakers of German at the Universities of Hamburg, Frankfurt, and Berlin. The JFL group was made up of 34 German learners of Japanese language at Universities of Hamburg, Frankfurt, Heidelberg, and Leipzig and all members had less than a month's experience living in Japan. Finally, the JSL group consisted of 44 German learners of Japanese language with more than a year's residency in Japan.

[ p. 46 ]

Table 1. Number of respondents in each subject group for this study.
Male        32 31 11 25 99
Female      30 31 23 19  103
Total       62 62 34 44 202

Materials and procedures

Three instruments were used for data collection: (1) a discourse completion test (DCT), (2) a multiple-choice test (MCT), and (3) a retrospective questionnaire. Each of those will be briefly explained.
The DCT listed in Appendix A contained twelve thanking situations in which the subjects were offered a favor. The subjects were encouraged to write down their spontaneous reactions and were also suggested to feel free to opt out when they wished to. Three versions of the questionnaires were used: a Japanese version for JNS, a German version for GNS, and a bilingual version for JSL and JFL in which the instructions and the situations were explained in German and the interlocutors' utterances were written in Japanese.

Table 2. Twelve situations and the situational variables represented in the Discourse Completion Test.

Familiarity/status Big favor Small favor
F a m i l i a r
Lower (1) A younger student offers you a ride home (2) A school boy you tutor brings you coffee
Equal (3) Your roommate pays for and receives your parcel (4) A friend lets you borrow his/her notebook before a test
Higher (5) Your professor writes a recommendation for you (6) Your section manager pays for your coffee
U n f a m i l i a r
Lower (7) A high school boy helps you with your luggage at a station (8) A junior high school student offers you a seat in a train
Equal (9) An unknown student fixes a copy machine for you (10) An unknown student offers you a change of seats
Higher (11) A stranger helps you put some tire chains on (12) A middle-aged person offers you a seat in a train

These twelve situations were designed to reflect combinations of different degrees of social distance, power, and size of imposition. As power difference could imply differences in both age and status, the familiar addressees in the situations were designed to be lower, equal, or higher in both age and status compared to undergraduate and graduate students. As to the unfamiliar addressees, who were not supposed to have any status relationship to the subjects, lower/higher power was indicated only in terms of age difference. A big favor indicates hard work requiring considerable time and/or financial/physical burden for the hearer, whereas a small favor involves only momentary actions or an insignificant expense.
The MCT listed in Appendix B was given to the JNS, JSL, and JFL groups after the DCT. For each of the same twelve situations, four alternative answers (thanks, thanks + apology, apology, and opting out) were given, and the subjects were asked to choose the one that sounded most appropriate to them. In order to reduce the influences of the MCT choices on the DCT, the subjects were requested not to look at the MCT until they had finished the DCT.

[ p. 47 ]

Lastly, after the DCT and the MCT, the retrospective questionnaires listed in Appendix C were given to all four groups in order to obtain information about the subjects' background data which might have influenced their answers; for example, the questionnaires for the learner groups inquired their age, sex, length of learning Japanese, length of stay in Japan, Japanese courses they were enrolled in, purpose of learning Japanese, whether they currently had chances to talk with native Japanese speakers or had a Japanese partner, whether they had learned in class the use of apologetic expressions when thanking, which situations had been difficult to answer, and the reason for changing their answers in the MCT if they had done so. In order to examine the proficiency levels of the learners, a small cloze-test of Japanese (Appendix D) was given at the end of the questionnaire.


For quantitative analyses, the DCT answers from the JNS, JSL, and JFL groups were first categorized into the three groups in Table 3 following Miyake's (1993b) categorization.

Table 3. Categorization of the DCT answers.

Semantic formulas Examples
1. A routinized thanking and
apologetic expression only
Doumo arigatou gozaimasu. ("Thank you very much.")
Doumo sumimasen. ("I'm very sorry.")
2. A routinized expression and
a supporting expression
Arigatou! Tasukatta yo. ("Thank you! You helped me a lot.")
Sonna ni te wo yogoshite shimatte...Gomennasai. ("Your hands got so dirty. I'm sorry.")
3. A supporting expression only Onegai suru yo. ("I'll accept your offer.")
Ah, honto... Ikura datta? ("Oh, really... How much was it?")

When the answers included a routinized thanking/apologetic expression – in other words, when the answers were categorized into either 1. or 2. of Table 3 – they were further sorted into the following three groups: "thanks", "thanks + apology", or "apology." In order to focus solely on the selections of these three appreciation strategies, the numbers of the subjects who used only supporting expressions (category 3. of Table 3) were excluded in the analyses. As the next step, ratios of the strategy selections were calculated and statistical analyses (Chi-square tests using SPSS) were also conducted on the numbers of the subjects who chose each of the appreciation strategies. In addition, with the MCT data, the ratios of strategy selections were calculated and the same statistical analyses were carried out.

[ p. 48 ]

Since the lengths of residence of all the learners were significantly correlated with their proficiency levels based on the cloze-test scores (r = .54), it became difficult to attribute the differences in the strategy selections solely to the different lengths of residence in Japan. The subjects were, therefore, regrouped into high proficiency (JHP) and low proficiency (JLP) groups by their cloze-test scores in order to investigate the effects of proficiency. The results of this examination will be discussed at the end of the next section.

Results and discussion

"The [German Native Speaker Group] almost exclusively used thanking expressions and barely used apologetic expressions . . . "
Appreciation strategies of native Japanese and German speakers

The two native speaker groups showed distinct differences in their strategy selections. The GNS almost exclusively used thanking expressions and barely used apologetic expressions (See Appendix E-1: DCT); almost 100% of them used thanks in 10 out of the 12 situations and the "apology" strategy was used in only 1 situation, which still accounted for merely 5% (3 out of 56 answers).
Compared to the GNS, the JNS showed a more varied selection of answers; the JNS group used thanking and/or apologetic expressions according to the situations. The results of the chi-square tests conducted to compare the use of the three appreciation strategies by the two groups revealed a statistically significant difference (p <.01) in all but one of the situations (Situation 2). Therefore, it can be claimed that the German subjects do not use as many apologetic expressions as Japanese.

Pragmatic transfer of the German learners of Japanese

In the DCT data, clear pragmatic transfer from German was identified in the learners' responses. The ratios of their strategy selections mostly fell between those of the GNS and the JNS; both learner groups used more thanks and fewer thanks + apology than the JNS group in most of the 12 situations. Moreover, the JFL group used fewer apologies than the JNS group in 9 situations. These results are evidence of pragmatic transfer. It should be noted, however, that the JSL group surpassed the JNS group in their use of apology strategy in 8 situations. It seems that these subjects tried to be polite and native-like but as a result surpassed the actual norms of the native Japanese.
On the other hand, the MCT results failed to prove the existence of pragmatic transfer. Instead they revealed a more striking use of hypercorrection of both learner groups. The JSL group, for example, used thanks less frequently than did the JNS group in 11 situations and they used apologies more frequently in all 12 situations.

The effects of learning contexts (JFL vs. JSL)

An examination of the ratios of strategy selections by all four groups revealed that, in comparison to the JSL group, the strategy selections of the JFL group were relatively more similar to those of the GNS in the DCT data. The results of the chi-square tests to compare the strategy selections of the two native speaker groups and the two learner groups also indicate a stronger resemblance of the JSL to the JNS as well as slightly more similarity of the JFL to the GNS. As the DCT table in Appendix E-1 suggests, the selections of thanks, thanks + apology, and apology by JSL & JNS all share more situations which are statistically non-significant than do the JFL & JNS. In the same way, JFL & GNS presented more similarity than the JSL & GNS.

[ p. 49 ]

These results agree with Takahashi & Beebe's (1987) findings that their EFL group revealed more pragmatic transfer than the ESL group did.
In the MCT results, however, the JSL revealed a stronger overuse of apologies than the JFL, which makes the overall strategy selection patterns of the JSL group less similar than the JFL to the JNS group, as evidenced by the results of the chi-square tests comparing the strategy selections of the JNS and the two learner groups (Appendix E-1: MCT); in the selections of thanks and apology, the JFL & JNS share more situations without any statistically significant difference than do the JSL and JNS.

The effects of length of residence

In order to find out the relationships between length of residence and pragmatic transfer, the ratios of the strategy selections were compared among the JNS; the JFL; 17 subjects from the JSL group with experience living in Japan for one year or less (shorter residence group, SRG); and 16 subjects with experience living in Japan for two or more years (longer residence group, LRG; average length of stay 3.7 years). This analysis of the DCT results revealed a clear tendency of pragmatic transfer of the JFL in most of the 12 situations as well as quite a distinct tendency of hypercorrection of the LRG. As a result, the strategy selection patterns of the SRG turned out to be the most similar to those of the JNS on the whole. The results of the chi-square tests comparing the strategy selections of the JNS and each of the above-mentioned three learner groups also suggested the greatest similarity of the SRG to the JNS followed by the LRG and the JFL (Appendix E-2: DCT).
While the MCT data indicated hypercorrection by all three learner groups, the SRG showed the most similar strategy selection patterns to the JNS in terms of the ratios in 7 situations, whereas the LRG showed the strongest tendency towards hypercorrection in 9 situations. The results of the chi-square tests (Appendix E-2: MCT) again suggest the greatest similarity of the SRG to the JNS followed by the JFL and the LRG.
From these results and the fact that many of the JSL subjects reported that they had never learned the use of apologetic expressions in thanking situations in class, it can be claimed that shorter residence contributed to the positive change for native-like strategy selections. Through contacts with native speakers in Japan, these subjects seem to have learned the usage of apologetic expressions in thanking situations. Hence, Hypothesis 3 (Assimilation will be correlated to the length of residence in Japan) was weakly supported. It should not be forgotten, however, that an even longer stay in Japan led to the hypercorrection and many of the subjects in the LRG reported that they were still not sure of the usage of each apologetic expression. The data is, therefore, too weak to support Olshtain & Blum-Kulka's (1985) findings on assimilation to the target norms and this issue merits further investigation.

The effects of situational variables

As explained earlier, each of the 12 situations represents a different combination of the three situational variables: social distance, power difference, and size of imposition. In order to focus on the influence of one of these factors by controlling the other two, the situations were paired up into 21 different combinations and chi-square tests were performed on the selection patterns of the three appreciation strategies by each group of subjects. Though the alpha level may have to be adjusted for more precise statistical analyses, this paper examines only the tendencies of the JFL and JSL subjects' reactions to the situational variables. The results are summarized in Tables 4 and 5; when the difference between two situations is statistically significant, the numbers of the subjects who chose each strategy in the two situations are given.

[ p. 50 ]

Table 4. Situational differences of the subjects' responses (DCT).
Table 4 DATA

[ p. 51 ]

As indicated in the table, the GNS and JFL groups barely changed their strategy selections according to these variables since they used thanks almost exclusively. The subjects of these groups used considerable ratios of thanks + apology or apology only in Situation 9, in which an unfamiliar student fixes a copy machine with great difficulty. In contrast, the response patterns of the JSL were much more similar to those of the JNS, who most clearly diversified their answers depending on the situational differences.
Just as in the DCT, the MCT results also revealed the JNS subjects' great sensitivity to the situational factors (Table 5). Although the JSL surpassed the JNS in the numbers of the statistically significant situational differences, they seem to have diversified their answers in the wrong places, which makes the JFL more similar to the native Japanese norms solely in figures.

[ p. 52 ]

Table 5. Situational differences of the subjects' responses (MCT).
Table 5 DATA

[ p. 53 ]

Moreover, as can be observed in Tables 4 and 5, it was found that the JNS reacted more sensitively to familiarity and power differences than to size of imposition. While they also changed their responses according to the size of imposition, a closer look reveals that they used more thanks for a big favor (Situation 7) and more apologies for a small favor (Situation 8). It should also be noted that the responses of the Japanese subjects showed no significant differences between hard labor (fixing a copy machine and putting snow chains on tires) and an offer of a seat on a train. This suggests that size of imposition does not play a major role in the thanking strategy selections of Japanese, which agrees with Miyake's (1993a) claim discussed earlier.
"The [Japanese Native Speaker Group] reacted more sensitively to familiarity and power differences than to size of imposition."

Qualitative analyses of the DCT data

The JNS, JSL, and JFL used very limited explicit thanking/apologetic expressions except that these expressions were sometimes accompanied by polite forms of copulas (e.g. -gozaimashita) and intensifiers (e.g. doumo, hontou ni) according to the situations. This also applied to the answers of the GNS; variations of Danke and Vielen Dank für... (= "thank you very much for...") were exclusively used.
Though apologetic expressions were barely used in German, it should not be overlooked that the GNS tried to express their eagerness to compensate for the favors they received. For example, in Situation 9, where an unfamiliar student fixes a copy machine, 23 GNS subjects out of 62 offered the addressee a handkerchief, coffee, cola, a chocolate bar, or soap and another 5 subjects asked how they could make up for the favor, whereas only 1 JNS subject out of 62 offered a handkerchief in the same situation. Transfer of this tendency to offer compensation was observed in 8 out of 44 JSL subjects as well as 7 out of 34 JFL subjects. This shows that the use of Japanese apologetic expressions in thanking situations functions as a remedial strategy that is considered to be as effective as a token of appreciation.

Responses to the retrospective questionnaires

The learners' responses to the retrospective questionnaires revealed uncertainty about the usage and distinctions of apologetic expressions used in thanking situations. 16 out of the 44 JSL subjects (36%) had not learned them as appreciation strategies in class, and another 4 subjects (9%) reported that they had learned only sumimasenused in thanking situations. More than half of the JFL subjects also reported that they had never known or learned such expressions before, while they had learned Japanese for three years on the average. However, with the given MCT choices introducing them to the possibility of using apologetic expressions in thanking situations, they showed much less pragmatic transfer in the MCT than in the DCT. Their improvement in the MCT data could also be explained by the nature of the test; since it requires the subjects to only assess the given choices rather than actually produce a response, it elicited more native-like responses which were above their proficiency levels.

The effects of language proficiency

As stated at the end of the previous section (in Analyses), the learners were regrouped into high proficiency (JHP) and low proficiency (JLP) groups by their cloze-test scores to examine the effects of proficiency. As a result, the JLP consisted of almost all the JFL subjects and 12 JSL subjects with lower scores, and the JHP was made up of the rest of the JSL subjects with higher scores. Surprisingly, it turned out that the JHP subjects (mean score 21.9/30 and mean length of stay 2.4 years) were responsible for the strong hypercorrection of the JSL, whereas the 12 JSL subjects with lower proficiency (mean score 9.8 and mean length of stay 1.1 years) brought positive changes to the JLP. Since the mean proficiency level of these 12 learners was almost the same as the JFL (mean score 6.6 and mean length of stay 10 days), it can be claimed that their length of residence rather than their proficiency contributed to the positive change towards native-like responses.

[ p. 54 ]


The results of this study suggest that the great variety of appreciation strategies in Japanese seem to present difficulties for most of the learners. As discussed above, the learners expressed unfamiliarity with the usage of apologizing expressions in thanking situations in the retrospective questionnaires. The DCT and MCT results also revealed the learners' overuse of apologetic expressions, which probably reflect their uncertainty regarding the speech acts. Comparative analyses of the three learner groups categorized in terms of length of residence displayed evidence of positive effects of a short stay in Japan, but those with a longer residency still do not seem to have a good command of the usage of apologetic expressions in thanking situations. All these results may suggest that such usage can be learned only after learners come in contact with native speakers in Japan and it is too complicated to be mastered in a few years. Therefore, it may be suggested that a systematic introduction of situationally appropriate response patterns in the target language should be adopted into second and foreign language classrooms.
"[Many Japanese] learners expressed unfamiliarity with the usage of apologizing expressions in thanking situations . . ."

This study was designed to include all the combinations of the three variables (social distance, power differences, and size of imposition) in all different settings. This design, however, made the examination of the effects of each variable difficult. Another weakness of the DCT and MCT tests was the restrictions on the number of the situations for the feasibility of the test; it would have been desirable to add more situations to obtain a better control of the variables.
The relationship between the development of pragmatic ability and the length of residence in a target language community also remains an area for further investigation. Since there are no written rules or correct answers of performing certain speech acts, their acquisition is expected to require considerable time. Since the average length of residence of the JSL subjects in this study was only two years and it was not possible to investigate how an even longer stay in Japan would affect learners' performances, replication of the study is called for to illuminate learners' further pragmatic development.


Beebe, L.M., Takahashi, T., & Uliss-Weltz, R. (1990). Pragmatic transfer in ESL Refusals. In R. Scarcella, E. Andersen & S. Krashen (Eds.). Developing communicative competence in a second language (pp. 55-73). Boston, MA: Heinle & Heinle.

Brown, P., & Levinson, S. (1987). Politeness: some universals in language usage. Cambridge: CUP.

Coulmas, F. (1981). 'Poison to your soul': Thanks and apologies contrastively viewed. In F. Coulmas (Ed.). Conversational routine (pp. 69-91). The Hague: Mouton.

Kumatoridani, T. (1988). Hatsuwakoui riron to danwa koudou kara mita nihongo no wabi to kansha [Apology and thanks in Japanese from the perspective of speech act theory and discourse behavior]. Education Dept., Hiroshima University, 37, (2), 223-234.

McClure, W. (2000). Using Japanese - A guide to contemporary usage. Cambridge: CUP.

Miyake, K. (1994a). Kansha no taisho kenkyu, Nichi-ei taisho kenkyu: Bunka shakai wo hanei suru gengo koudou. [A contrastive study of thanks between Japanese and English: Linguistic behavior reflecting cultures and societies].Nihongogaku, 13, (7), 10-18.

Miyake, K. (1994b). 'Wabi' igai de tsukawareru wabi hyougen: Sono tayouka no jittai to uchi, soto, yoso no kankei. [Apologetic expressions used for other functions: The actual conditions of its diversification and relationship to uchi, soto, and yoso]. Nihongo Kyouiku, 82, 134-146.

Nakata, T. (1989). Hatsuwa koui to shiteno chinsha to kansha: Nichi-ei hikaku [Apology and thanks as speech acts: Comparison between Japanese and English]. Nihongo Kyouiku, 68, 191-203.

Ogawa, H. (1995). Kansha to wabi no teishiki hyougen: Bogo washa no shiyou jittai no chousa karano bunseki. [Formulaic expressions of thanks and apologies: Analyses of the actual usages by native speakers]. Nihongo Kyouiku, 85, 38-52.

Olshtain, E., & Blum-Kulka, S. (1985). Degree of approximation; Nonnative reactions to native speech act behavior. In S. Gass & C. Madden (Eds.). Input in second language acquisition (pp. 303-325). Rowley, MA: Newbury House.

Sakuma, K. (1983). Kansha to wabi [Thanks and apology]. In O. Mizutani (Ed.). Hanashi kotoba no hyougen. (Kouza nihongo no hyougen 3) [Expressions in a spoken language (Seminar of Japanese expressions 3)] (pp. 54-66). Tokyo: Chikumashobou.

Takahashi, T., & Beebe, L.M. (1987). The development of pragmatic competence by Japanese learners of English. JALT Journal, 8, 131-155.

Tanaka, S., & Kawade, S. (1982). Politeness strategies and second language acquisition. Studies in Second Language Acquisition, 5, 18-33.

Main Article Appendix A Appendix B Appendix C Appendix D Appendix E

2005 Pan SIG-Proceedings: Topic Index Author Index Page Index Title Index Main Index
Complete Pan SIG-Proceedings: Topic Index Author Index Page Index Title Index Main Index

Last [ p. 55 ] Next