Refusal Speech Act Response : Differences between South Koreans and North Korean Refugees in Inducing Speech Acts and Directness

The majority of previous studies on North Korean Refugees (NKRs) focused on the linguistic heterogeneity of pronunciation and vocabulary use. Only a handful of studies examined differences between South Koreans’ (SKs) and NKRs’ language use and they concluded that NKRs prefer direct refusal speech acts rather than indirect ones. However, we hypothesized that NKRs’ preference on direct refusal speech acts would vary depending on the types of inducing speech acts; speech acts that induces refusal expressions. 47 SKs and 43 NKRs answered three questions after watching video clips of short conversation in refusal situations. The results were statistically analyzed with independent sample t-tests and multi response analyses, revealing that NKRs preference on direct refusal speech acts varied depending on the types of inducing speech act. In addition, multi response analyses indicated that the NKRs and SKs evaluated the same rejecter’s personality differently and that this difference may result in miscommunication. Lastly, NKRs’ evaluations on the refusal speech acts did not change over time, as NKRs who stayed less than a year in South Korea and those who stayed more than five years did not show significant difference in their evaluation. This result suggests that explicit education on this may be required. Additionally, it is worth to note that current research confirms that SKs and NKRs react differently toward expressions used only in North Korea, and, therefore, not only the refusal strategy, but also the refusal expression itself are both important for successful communication. Additionally, the results showed that refusing with nonavoidable and non-personal reasons were perceived more positively than personal reasons. These findings suggest that for successful communication to happen between two groups with different linguistic backgrounds, pragmatic awareness in language use is helpful.

1) Do SKs' and NKRs' perceive appropriateness and unpleasantness of refusal expressions differently when the types and directness of inducing speech acts vary? 2) Do SKs and NKRs perceive the rejecter's personality differently when the types and directness of inducing speech acts vary? 3) Does the NKRs' length of stay in South Korea have an impact on their perception of appropriateness and unpleasantness of refusal expressions?Would it differ on the basis of the types and directness of inducing speech acts?

BACKGROUND
According to Chung (2001), who made an initial attempt to conduct research on NKRs' linguistic adaptation to South Korea, 90.15% of 90 NKRs struggled due to language barriers.Moon (2006) organized a survey and in-depth interviews with 100 NKRs and concluded that they were psychologically intimidated due to linguistic differences and thus felt obligated to learn the language used in South Korea.Kim (2005) worked with NKRs and South Korean experts2 in South-North Korean relations.Contrary to the expert group's report, NKRs generally recognized the importance of linguistic heterogeneity and reported that they have suffered from linguistic differences.
The results demonstrate general aspects of issues related to NKRs' linguistic adaptation, but they do not identify actual differences in specific language use of SKs and NKRs nor their reasons.Previous investigations on the linguistic heterogeneity of South and North Korea, especially case studies, scrutinized the differences in pronunciation and vocabulary.Only a few investigations have focused on language use, which mostly reported that NKRs prefer direct expressions.For instance, in Jeon (2010)'s survey of 115 NKRs, 34.1% of them identified refusal expressions as most difficult ones to use, followed by compliments (21%), apologizing (20%), and greeting (24%).About 92% of NKRs reported that they and SKs utilized a different refusal speech act strategy and 33% preferred a direct refusal speech act while most SKs favored an indirect one.
In the same research, NKRs also reported that they felt troubled because their language was different from SKs (Jeon, 2010, pp. 12-13).These findings corroborate with Moon's (2006) results.NKRs in this study reported that they felt offended when SKs used euphemisms, such as "나중에 연락할게 najunge yeollakhalke (I will call you later)" or "연락 남길게 yeollak namgilke (I will leave a message)" following their requests.They did not perceive these expressions as refusals and misunderstood the SKs' intentions.It was also mentioned that in North Korea, people tend to use direct expressions rather than euphemisms when they have to use refusal expressions (Moon, 2006, pp. 128-135).Yang and Kwon (2007) also reported cases when NKRs misunderstood SKs' indirect rejections.NKRs expressed a sense of betrayal and were displeased when they faced SKs' indirect refusals.The researchers additionally reported that NKRs evaluated SKs negatively, even as "defrauders" when indirect speech acts were used (pp. 468-469).
While previous studies concentrated on the speaker of the refusal speech act, this study focuses on the "reception of refusal".We hypothesized that an inducing speech act, which leads to refusal expressions plays a crucial role in the difference between SKs' and NKRs' reactions toward that speech act, as well as the expression's directness.As prior research brought to light, NKRs may be uncomfortable with indirect refusals, which are known to be SKs' preferred refusal strategy.On the other hand, SKs can also feel offended when NKRs refuse with direct expressions.These emotional judgments based on the difference of interpreting expressions are likely to cause more severe misunderstandings than using disparate pronunciation and vocabulary3 .Hence, we conduct research on SKs' and NKRs' reactions toward a refusal speech act with the aim to foster communication between these two groups.The present research is distinctive from previously conducted speaker-centric studies as it is listener-oriented, focusing on their interpretations of and reactions to the refusal speech act4 .Additionally, this leads to identification on the cause of NKRs' sociolinguistic miscommunication and misunderstandings in South Korea.

PARTICIPANTS
In this study, 90 participants (43 NKRs and 47 SKs) answered a set of video-based questions.As access to NKRs was limited, all NKRs were recruited through a Hana Center5 .To control unrelated variables, such as age and gender, NKRs were recruited first and South Korean participants matched accordingly.The NKRs' gender ratio corresponded with that of NKRs in South Korea; there are more female NKRs than males.Among 43 NKRs, 15 were male and 28 were female and 19 males and 28 females participated as SKs group.The age was also taken into account to homogenize the two groups6 .The NKRs' length of stay in South Korea varied: Among 43 NKRs, 21 participants had stayed less than one year, 11 one year or more but less than five years, and the remaining 11 five years or more.

INSTRUMENT
A set of video-based questions was designed to determine the SKs' and NKRs' reactions toward a refusal speech act.Lee (2009)'s video-based questionnaire was adapted to develop the instrument and twelve videos were created that presented twelve situations with refusal speech acts.Participants were asked to answer three questions for each video clip.Each situation varied in the inducing speech act type, leading to expressions of refusals with differing directness.Three types of inducing speech acts -compensation, request, and favor -were combined with either direct (conclusive) expressions or indirect (roundabout) expressions.Additionally, the social distance between conversation participants may influence the judgment about the appropriateness of a given speech act.Therefore, we arranged the relationships of conversation between participants -to have one of them with a higher social status or school year than the other.Table 1 shows the survey's outline.Participants were requested to answer three questions8 for each video-recorded situation.The first two were designed to gather data for research question one.
Participants had to answer the questions using 6-point Likert scales9 .The third multiple-choice item pertained to the second research question.To confirm the internal consistency of the first and second question, Cronbach's alpha was calculated.The results indicated that the questionnaire was a reliable measure of research questions: The values were .795for the first question and .743for the second.Figure 1 shows an example of a set of questions.

DATA COLLECTION AND DATA ANALYSIS
The data were collected from the 10 th of March to the 6 th of April 2017 through a "Google online survey form."NKRs who had just moved to South Korea were not familiar with the online survey format, and thus the Hana Center employees were allowed to assist participants with technical issues.However, the staff members did not comment on the contents of the questions which may have influenced the data.
The data were coded and statistically analyzed with the Statistical Package for the Social Sciences (SPSS), version 23 (IBM).As explained in previous sections, three research questions were predetermined to compare South and North Koreans' reactions toward refusal speech acts based on the types of inducing speech acts and their directness.To answer the first research question, the data were statically analyzed using independent sample t-tests comparing SKs' and NKRs' response differences on a refusal speech act.The second research question was statistically analyzed by multiple response analysis where participants could choose multiple answers for the personality judgment of the rejecter.Lastly, to determine whether the NKRs length of stay South Korea influenced their reaction toward a refusal speech act, independent sample t-tests were again calculated.

RESPONSE DIFFERENCES BETWEEN SKS AND NKRS REGARDING REFUSAL EXPRESSIONS
In the present research, we compared SKs' and NKRs' reactions toward six types of refusal situations with varying inducing speech acts and directness.The first research question involved comparing how inducing speech act types and the directness of refusal expressions affected SKs' and NKRs' response differences when they were asked to evaluate the appropriateness and unpleasantness of a given situation.This difference was analyzed with independent sample t-tests and a response close to "one" (1.00) indicated that participants evaluated the situation as "appropriate."Contrary to previous research findings, NKRs did not simply prefer direct refusal expressions.Their preferences depended on different inducing speech acts and the variety of directness.Tables 2 and 3 display the differences in SKs' and NKRs' judgments of appropriateness and unpleasantness about the twelve refusal situations.
Two aspects from the descriptive statistic values, in particular, the mean values shown in these tables are noteworthy.First, except in the situation when the conversation participant (B) was rejecting (A)'s compensation with a direct refusal expression, SKs tended to assess the situation as more positive than NKRs, in which the responses were more "appropriate."This is because SKs perceived "일 없습니다 il eopssumnida (nothing to worry)10 ," which was used to directly reject a compensation, more negatively than NKRs.Secondly, Tables 2 and 3 show that when the rejecter refused (A)'s favor or request with indirect refusal expressions and for inevitable reasons, SKs and NKRs both assessed them as more appropriate and less unpleasant.This finding corresponds with Lee (2009, pp. 218-221).Participants evaluated the situation as more appropriate when the refusal expression was accompanied by inevitable and non-private reasons than when it was used with personal reasons or sententious expressions.When SKs' and NKRs' responses were analyzed with an independent sample ttest as shown in Table 2, SKs did not assess direct rejection as less appropriate nor did NKRs evaluate direct refusal as more positive.This finding does not coincide with previous research, in which NKRs preferred direct refusal expressions over indirect ones.However, prior studies did not consider the influence of inducing speech act types and the speech act's directness.Such results are limited by their research method, as they were based on individual cases of NKRs.As shown in Table 2, SKs' and NKRs' reactions differed only when the rejection was a) indirect and toward a compensation or a request and b) direct toward a favor.Direct refusal toward compensation or a request and indirect refusal toward a favor did not elicit any differences between SKs' and NKRs' reactions.Even in some direct refusal situations, for instance, direct refusal toward a favor (t = 3.108, p = .003),NKRs evaluated the situation as less appropriate than SKs.Therefore, there is only scanty evidence that NKRs prefer direct refusal expressions in contrast to SKs.
Furthermore, the data shown in Table 3 do not correspond with previous research findings, which concluded that NKRs prefer direct refusal speech acts over indirect ones.In our study, when direct refusal was presented toward compensation and requests, although SKs' and NKRs' evaluations of the unpleasantness toward the situation differed, the difference was not statistically significant.Also, even for the direct-rejection situation, NKRs evaluated the situation as more unpleasant when it was toward a favor.

SKS' AND NKRS' JUDGMENT DIFFERENCES ON THE PERSONALITY OF REJECTER
The second research question examined how SKs and NKRs evaluate the personality of a rejecter when different inducing speech act types and refusal expressions of varying directness were displayed.The survey participants were asked to describe the rejecter's personality by choosing from a list (includes determined, kind, rude, cold, polite, arrogant, thoughtful, flattering, tough, sociable) and they were also allowed to depict the rejecter's personality in their own words if the list was not agreeable to them.Multiple responses analysis was used to analyze the data statistically.Interestingly, when SKs and NKRs evaluated the situation in 4.1, they evaluated the rejecters' personality differently.See the appendix for a table with complete data.
As stated above, when the rejecter refused compensations directly, differences in SKs' and NKRs' evaluations about the appropriateness and unpleasantness of the situation were not statistically significant.However, while most SKs evaluated rejecters as determined (42.6%), cold (27.7%), rude (20.2%), and arrogant (18.1%), fewer NKRs described them as determined (20.9%) or cold (19.8%).Instead, 22.1% of the NKRs evaluated the rejecter as a kind person, which is about two times more than SKs (10.6%).This seems to be a result of the of determined expression's influence "il eopssumnida (nothing to worry)" which is only commonly used in North Korea.
On the other hand, when the rejecter refused compensations indirectly, SKs described her as determined (44.7%) and cold (25.5%), but polite (27.7%).However, NKRs described the same person only as determined (27.9%) and rude (22.1%).In addition, although there were only very few responses, in the same situation, NKRs described the rejecter flattering (2.3%) while SKC chose tough (3.2%).Interestingly, none of the NKRs chose "tough" and none of SKs "flattering" to describe the same rejecter.For this conversation, SKs and NKRs exhibited a statistically significant difference when they were asked to evaluate the situation's appropriateness and unpleasantness.Although a minority opinion, this gap in personality evaluation represents a possibility for miscommunication between SKs and NKRs.
For a situation when the rejecter used direct refusal expressions toward a request, SKs and NKRs both indicated that the rejecter was polite.SKs characterized the rejecter as sociable (36.2%) and polite (31.9%) while NKR described her as a kind (30.2%) and polite (25.6%) person.
Also, when the rejecter used indirect refusal expressions toward a request, both SKs and NKRs evaluated her positively with some minor differences.SKs described her slightly more positively than NKRs: They stated polite (46.8%), sociable (38.3%), and kind (33.0%) while NKRs characterized her as kind (38.4%) and polite (34.9%).In addition, only 16.3% of the NKRs perceived the rejecter as a sociable person, which was about half of SKs' perceptions (38.3%).This finding indicates that NKRs perceive indirect rejection toward requests positively, however, they also think that the person who uses this type of rejection lacks social and interpersonal skills.Also, as shown in Tables 2 and 3, SKs' and NKRs' reactions toward this specific situation were different at a statistically significant level.This corresponds with the finding that SKs perceived rejecters more positively than NKRs (refer to the appendix).
Lastly, if asked to evaluate a rejecter's personality when she directly refused a favor, SKs described her as polite (37.2%), but determined (26.6%) and cold (20.2%), while NKRs identified her rather positive: as a polite (23.3%) and kind (23.3) person.It is notable that the percentage of NKRs who described her as arrogant or rude was higher than that of SKs which corresponds with the findings from 4.1, Table 3, where NKRs reported that they felt less pleasant when a favor was directly rejected.We can conclude from these results, that NKRs tend to perceive those who directly reject one's favor somewhat negatively.
Needless to say, perceiving another individual's personality or characteristics is an important feature of interpersonal relationships and may have an impact on pragmatic failures.Therefore, for successful communication, both SKs and NKRs should be aware that some perceptional differences exist among them.An interpretation on personality features like "sociable" may indicate one's perception toward interpersonal relations, which illustrates that these miscommunication issues should not only be handled from a pragmatic-linguistics' perspective, which is based on language habits, but also considered as socio-pragmatics' problems.

CORRELATION BETWEEN NKRS' REACTIONS TOWARD REFUSAL EXPRESSIONS AND THEIR LENGTH OF STAY IN SOUTH KOREA
The third research question investigated whether the NKRs' length of stay in South Korea has an impact on their judgment of the refusal conversation: Do they feel more or less appropriate or unpleasant?To address this question, answers from two groups of NKRs, one with less than a year of residence in South Korea and the other with five years and more, were compared with an independent sample t-test.Table 4 displays whether these two groups of NKR perceived the situations as more or less appropriate, while Table 5 exhibits the differences in the level of unpleasantness they felt.According to the independent sample t-tests, regardless of the inducing speech act types and the refusal expressions' directness, the two North Korean refugee groups did not show any statistically significant difference in their evaluation of the appropriateness and unpleasantness of refusal expressions.Moon (2006, p. 35) reported that 30% of NKRs believe that they overcame the language gap between South and North Korean when they passed their 12t h month of stay in South Korea.However, as shown in Tables 5 and 6, the participants in this study did not show significant changes, even after five years of stay in the RoK.While Moon (2006)'s research was based on NKRs' subjective selfreports, the results did not coincide with the present research, which is based on more objective data analyses.
This result indicates that even if the NKRs' length of stay in South Korea increases, their North Korean language use and attitudes are still retained.Especially their reactions and attitudes towards the refusal speech act, which is different from SKs, may become an obstacle to successful settlement in South Korea.This seems to be a more serious issue than linguistic heterogeneity in vocabulary or pronunciations.NKRs' preservation of North Korean language use and attitudes, even after lengthy residence in South Korea, needs closer attention for successful settlement in South Korea.

DISCUSSION
As more NKRs settle in South Korea, their desire to acquire the country's standard language has increased.However, only a handful of studies with limited variety was conducted in this field.These previous studies focused mostly on examining NKRs' language use, the differences compared to the standard language used in South and North Korea, and NKRs' acquisition of the South Korean standard language.Language gaps between SKs and NKRs may become an obstacle in communication between these two groups, and thus, it is important to identify both the differences in and perceptions of each other's language use.In this regard, the present research yielded four notable findings.
First, our results contradict previous studies on linguistic heterogeneity of SKs and NKRs, which reported that NKRs prefer direct refusal speech acts.That is, it is daring to state that NKRs always prefer direct refusal speech act regardless of the context where the rejection occurs.According to the data gathered in this study, NKRs in general prefer indirect speech acts as do SKs.This becomes clear when we take a look at NKRs' evaluations on appropriateness and unpleasantness of a situation: The mean value did not differ significantly from the SK's value.In addition, NKRs positively evaluated a rejecter's personality not only when they were using direct refusal expressions, but also with indirect expressions.Therefore, we have to depart from the traditional belief that NKRs, at all times, prefer direct speech acts and accept the fact that their preferences vary, based on the types of inducing speech act or the directness of rejection.Secondly, refusal expressions themselves are as important as the refusal strategy.In 4.1, where SKs and NKRs were asked to evaluate the appropriateness of six refusal situation types, depending on the inducing speech act and the rejection's directness, SKs perceived the situation as more appropriate than NKRs, except for the situation of directly rejecting compensation.This was when "il eopssumnida (nothing to worry)," a North Korean refusal expression, was used to directly reject the compensation and it illustrates why we have to take both the refusal strategy and expression into account when studying and teaching NKRs' rejection speech acts.
Furthermore, when the rejecter is using an inevitable reason to refuse compensation and a request, survey participants evaluated the conversation to be more appropriate and less unpleasant.This finding corresponds with the results from previous studies: Non-personal and unavoidable reasons were considered more appropriate for refusals than personal reasons and sententious expressions.
Finally, with the study results, we confirmed that a change in pragmatic perceptions and attitudes requires a long period of time.This is consistent with Bouton's study who asserted that -without instruction -simply spending time in the target language community does not guarantee language acquisition (Bouton, 1994).As mentioned in Rose (2005) and Lee (2015), pedagogical interventions are essential to expedite NKRs' acquiring of pragmatic strategies.An explicit instruction on vocabulary, expression, grammar and pragmatic strategy, as a whole, used in the South Korean pragmatic community will not only facilitate their language acquisition but also aid their adaptation to South Korea.

CONCLUSION
NKRs desire to be fluent in the standard South Korean language; however, their actual language usage is quite different from SKs.Using pragmatic features not preferred in South Korean discourse community may cause NKRs to be involved in pragmatic failures and these failures require special attention as SKs may perceive them as personality issues rather than just linguistic errors.
The current research was based on one single question derived from previous research findings, which concluded that NKRs prefer direct refusal speech acts compared to SKs.However, is this difference, unequivocally statistically significant?Moreover, the hypothesis is that inducing speech act types as well as the expressions' directness may also have an impact on reactions toward rejections.Thus, differences in responses between SKs and NKRs on refusal speech acts, based on the inducing speech acts and directness were examined.
The findings show that NKRs did not simply prefer direct speech act over indirect ones.Statistically significant differences only emerged when the rejecter was refusing compensation and request indirectly, and favor directly.When the two groups were asked to evaluate direct refusal expressions toward compensation and requests as well as indirect rejections toward favors, their evaluation differences were not statistically significant.Also, in some direct refusal situations, such as directly rejecting compensation, NKRs evaluated the situation as less appropriate than SKs.Thus, when researching and teaching NKRs' refusal speech acts, researchers should depart from the dichotomous way of thinking that they will always prefer direct speech acts.Instead, there is a need for an approach with an integrated perspective: Reactions toward refusal speech acts may vary when different inducing speech act types and directness are involved.
It is also confirmed that NKRs, compared to SKs, perceived rejecters who directly refused compensation, as more kind.A few NKRs evaluated a rejecter who indirectly refused compensation as "tough" which was not found at all among the SKs' answers.This result supports the idea that pragmatic differences may perceived as a function of personality (see our second research question).Lastly, NKRs did not show statistically significant differences in pragmatic perceptions even if they had stayed in South Korea for more than five years (see our third research question).This result suggests that to prevent pragmatic failures, explicit instructions are required.
Since judgments about one's personality are an essential part of social relationships, NKRs may face hardships when adapting to South Korea, if they lack understanding about their linguistic differences compared to indigenous South Koreans.Moreover, as their pragmatic perceptions do not change, even after a lengthy stay in the country, our research strongly suggests that education is essential for their adaptation.The findings from the present research will help NKRs to adapt better to South Korean society by fostering their understanding of linguistic perceptions and to prevent communicational barriers between SKs and NKRs.In addition, the current results are expected to inspire future research as the findings from previous investigations could be statistically confirmed, which had been only reported as individual case studies.

FIGURE 1 .
FIGURE 1. Examples of the Survey Questions

TABLE 2 .
SKs' and NKRs' Evaluation of the Conversation: Appropriateness

TABLE 3 .
SKs' and NKRs' Evaluation of the Conversation: Unpleasantness

TABLE 4 .
NKRs' Length of Stay in South Korea and Their Evaluation: Appropriateness

TABLE 5 .
NKRs' Length of Stay in South Korea and Their Evaluation: Unpleasantness