This is an open access article distributed under the terms of the Creative Commons
Attribution License (When an individual receives a benefit from a benefactor, how would this individual respond to the benefactor’s action? It is natural for most people to express gratitude to the benefactor. Past studies have found that such expressed gratitude enhances interpersonal relationships in various ways such as elevated levels of relationship satisfaction in romantic couples (Algoe et al., 2013) and enhanced friendship quality (Lambert et al., 2010). One of the possible reasons for the positive effect of gratitude is that gratitude functions to meet psychological needs of a benefactor; for example, received gratitude enabled a benefactor to feel efficacious, which is one of the basic human needs (Chaudhry & Loewenstein, 2019; Grant & Gino, 2010). However, it is still in question whether gratitude meets psychological needs across cultures. For instance, in Japan where people sometimes use apologies to show their gratitude to a benefactor (Miyake, 1993; Sugimoto, 1999), would expressed apologies meet any of a Japanese benefactor’s psychological needs? This study assesses this question.
It is critical to examine the cultural differences in messages exchanged when help is provided because failing to meet people’s needs during communication could have a significantly negative impact on their relationship (Chaudhry & Loewenstein, 2019; Coulmas, 1981). This study investigates how received gratitude and apologies influence perceptions of a Japanese benefactor regarding their psychological needs. This research specifically focuses on needs of face and competency based on theoretical frameworks such as politeness theory (Brown & Levinson, 1987) and self-determination theory (Deci & Ryan, 2000). The current study also examines the effect of the amount of a benefactor’s effort on the effectiveness of gratitude and apologies as literature suggests that the perceived high imposition on the benefactor influences the beneficiary’s gratitude experiences and expressions (McCullough & Tsang, 2004).
The following section first discusses a general picture of research on the functions of gratitude and apologies in the context of interpersonal relationships. Next, politeness theory and self-determination theory are explained to propose hypotheses regarding how gratitude and apologies could meet psychological needs of Japanese people. Finally, how the amount of effort a benefactor puts into helping a beneficiary could impact the effect of gratitude and apologies is considered.
Gratitude and Apologies
While gratitude is defined as an emotional appreciation of and thankfulness for favors received from others (Bonnie & de Waal, 2004), an apology is defined as an acknowledgement of transgressions together with an expression of remorse (Lazare, 2004). In general, thanking is used by beneficiaries to express gratitude and apologizing is used by offenders to acknowledge their misdeeds. Chaudhry and Loewenstein (2019) developed a responsibility exchange theory arguing that these two distinct types of communications were theoretically two sides of the same coin. Thanking is a communication giving credit to another person for a positive outcome, whereas apologizing is a communication taking blame for a negative outcome on oneself. For example, if a college student is worried about a difficult assignment, he or she may ask a classmate for help, who is also busy doing her or his own assignment. The classmate could spend time and put effort into helping the student. In this case, the classmate provides a positive outcome for the student, so the student may say “thank you.” It is also true that the student bothers the classmate, so the student may say “sorry” for the time and effort the classmate expended. In other words, the fact that an individual receives a benefit from another individual can be considered a situation generating either a positive outcome for a beneficiary or a negative outcome for a benefactor. Depending on how the outcome is evaluated, thanking and apologizing are strategically (but mostly unconsciously) chosen and used (Chaudhry & Loewenstein, 2019).
Coulmas (1981) provided similar theoretical accounts to explain the relationship between gratitude and apologies. Two concepts, which are the object of gratitude and the object of regret, should be considered to analyze thanks and apologies. Thanking is directed to some action of a benefactor and the action is the object of gratitude. Apologizing is directed to some action causing a negative outcome to others and the action is the object of regret. There is a common domain where thanks and apologies are both appropriate and this is where it becomes difficult to keep object of gratitude and that of regret distinct from each other. The link between two types of objects is the concept of indebtedness. A beneficiary could imply her or his indebtedness to a benefactor by thanking. Similarly, a beneficiary could recognize her or his indebtedness to a benefactor by apologizing. Taken together, when a benefit is provided from a benefactor to a beneficiary, depending on how the outcome and the indebtedness are perceived by the beneficiary, either thanks or apologies can be selected and expressed.
Relevantly, another theoretical perspective suggests that beneficiaries use gratitude and apologies to indicate the sign to meet the reciprocity norm that those who are helped should help the benefactor in turn. Research indicates that people experience discomfort when they are helped as they believe it is difficult to repay the benefactor for their help (Walster et al., 1973). In other words, people who are helped would be motivated to behave prosocially back to a benefactor to avoid being in an unequal relationship (McCullough et al., 2001). Whether or not the beneficiaries really help the benefactor later, by expressing gratitude or apologies they could moderate the discomfort caused by the perception that they are in the unequal relationship with the benefactor.
Previous studies suggest that the use of these two communication options varies across cultures. For example, Japanese are more likely to apologize than Westerners (Lee et al., 2012; Miyake, 1993; Sugimoto, 1999). Lee et al. (2012) specifically found that when making a favor-asking message, Japanese were more likely to include apologies than thanks whereas Americans were more likely to include thanks than apologies. Further, Japanese were twice as likely to feel apologetic as were the British in situations where people could feel either grateful or apologetic (Miyake, 1993). One of the reasons for the cultural difference is that Japanese people tend to feel a higher degree of indebtedness in various social interactions (Benedict, 1946; Naito & Sakata, 2010). Naito and Sakata (2010) specifically showed that Japanese people experienced a higher degree of indebtedness along with gratitude when they received a favor from others. As shown above, literature indicates that in some situations thanking and apologizing are interchangeably used based on the perceived outcome and indebtedness and the use varies across cultures. These studies provided insights of how people choose and use these communication styles, but they did not offer knowledge on what effect these different communication styles bring about in interpersonal relationships. The following literature tried to examine such effects.
Effects of Thanking and Apologizing on Receivers
After people help others, what message would they expect and prefer to receive from the beneficiary? Received gratitude almost always has a positive impact on the message receiver. For instance, benefactors who are thanked by a beneficiary are more likely to behave prosocially back to the beneficiary (McCullough et al., 2001). Grant and Gino (2010) also found that benefactors who received gratitude from a beneficiary experienced higher levels of social worth and self-efficacy. Further, Williams and Bartlett (2015) revealed that benefactors who received a note showing gratitude from a beneficiary reported that the impression of the beneficiary was warm. Compared with many studies on effects of received gratitude shown above, there is little research on received apologies after a benefit is provided. A study by Imai (2022) found that Japanese participants who received apologies from a beneficiary reported higher levels of social worth compared with those who received a message without gratitude or apologies. The current study tries to extend the previous research by examining the influence of gratitude and apologies on various aspects of a Japanese benefactor’s perceptions.
Psychological Needs in Interpersonal Relationships
Gratitude and apologies are expected to satisfy two important psychological human needs: face needs and competence needs.
Negative and Positive Face Needs
Goffman (1967) defined face as a positive and socially approved self-image an individual desires to claim through interaction. Extending this definition of face, Brown and Levinson (1987) developed politeness theory describing how people produce messages to maintain their own and others’ face. Based on the theory, there are two types of face: negative face and positive face. Negative face is a desire to maintain autonomy from others and to preserve freedom of action and freedom from imposition, whereas positive face is a desire to gain approval and be positively evaluated by others. People try to meet both types of face needs for themselves and others. For instance, in the previously mentioned case, the classmate helping the other student could meet her or his positive face needs while her or his negative face needs may not be satisfied since helping the student may prevent her or him from preserving freedom of action.
It is critical to meet face needs especially after people receive a favor from others. When a benefit is provided, a beneficiary is expected to show appreciation to a benefactor while recognizing her or his indebtedness to the benefactor. Failing to meet the face needs of a benefactor could lead to serious interpersonal consequences (Chaudhry & Loewenstein, 2019; Imai, 2022). Literature implies that gratitude and apologies help a communicator satisfy these kinds of needs of a benefactor (Brown & Levinson, 1987; Cupach & Metts, 1994; Lee et al., 2012). However, the literature has not fully examined how these communication styles meet face needs when a benefactor provides a benefit to a beneficiary. Lee et al. (2012) focused only on a favor-asking situation to examine the effectiveness of gratitude and apologies to meet face needs. Other research on the effectiveness of received gratitude after a benefit was provided (e.g., Grant & Gino, 2010; Williams & Bartlett, 2015) did not investigate how gratitude meets the face needs of a benefactor.
The current study hypothesizes that for Japanese not only gratitude but also apologies function to meet the negative and positive face needs of a benefactor. As explained in the previous section, Japanese tend to use apologies to show gratitude to a benefactor besides simple thanks (Coulmas, 1981; Lee et al., 2012). Showing apologies could satisfy both negative and positive face of a benefactor because apologizing theoretically functions to recognize a beneficiary’s indebtedness to a benefactor. By showing recognition of the indebtedness, a beneficiary could help a benefactor believe that their freedom (i.e., negative face) is not threatened anymore. Apologies may also function to meet a benefactor’s positive face because the benefactor could believe that a beneficiary recognizes how much effort the benefactor put into helping, and that may help to maintain the benefactor’s positive image. Further, expressed gratitude might also help satisfy both negative and positive face of a benefactor. Received gratitude may imply that a benefactor did enough to meet the desires of a beneficiary, so no more requests imposing on the benefactor are expected. A benefactor who receives gratitude may also believe that their positively evaluated self-image is maintained because they are sure of their contributions to the positive outcome (Grant & Gino, 2010; Imai, 2022).
This study assumes that benefactors’ face needs are met when they believe that their negative and positive faces are not threatened by a beneficiary’s action (Cupach & Metts, 1994). For instance, if a benefactor receives a message containing gratitude from a beneficiary, their negative and positive face needs could be met, which is represented as the benefactor’s perceptions of reduced threats to their face. Further, the current study examines the effect of a message including both gratitude and apologies on a benefactor since it is not uncommon that Japanese use both gratitude and apologies to express gratitude (Kumatoridani, 1991). Based on these theoretical accounts, the following hypotheses were posed:
H1: Benefactors (Japanese participants) who receive a message expressing gratitude, apologies, or both of them perceive that their negative face is less threatened by a beneficiary than those who receive a message without either of them.
H2: Benefactors (Japanese participants) who receive a message expressing gratitude, apologies, or both of them perceive that their positive face is less threatened by a beneficiary than those who receive a message without either of them.
Competence Needs
Other important needs that gratitude and apologies could meet are competence needs. Based on self-determination theory developed by Deci and Ryan (2000), there are three psychological needs people desire to meet: the needs for competence, relatedness, and autonomy, all of which are essential for ongoing psychological growth, integrity, and well-being. The needs for competence are a motivation to have an effect on the environment as well as attain valued outcomes within it. The needs of relatedness are a desire to feel connected to others as well as a desire to love and to be loved. Finally, the needs of autonomy are a desire to self-organize experience and behavior which are consistent with one’s integrated sense of self. This study focuses on the competence needs which are theoretically impacted by receiving gratitude and apologies. When people can meet their competence needs, they experience high levels of self-efficacy, which is defined as people’s beliefs in their capacity and competence to act effectively to produce specific performance attainments (Bandura, 1977).
This study specifically predicts that received gratitude, but not apologies, could meet competence needs of a benefactor, which will be represented as enhanced self-efficacy. A benefactor who receives gratitude signaling that a beneficiary gives credit to the benefactor for a positive outcome (Chaudhry & Loewenstein, 2019) could feel increased competence. However, a benefactor who receives apologies signaling that a beneficiary recognizes her or his indebtedness to the benefactor (Coulmas, 1981) could believe that he or she was not able to have a positive influence on the beneficiary. The belief of failing to provide positive outcomes may not function to meet the competence needs of the benefactor. Grant and Gino (2010) observed that American participants who received gratitude experienced higher levels of self-efficacy compared with those who received a message without gratitude. The current study examines if the same results could be observed among Japanese participants who culturally tend to use apologies to express gratitude more often than Americans (Lee et al., 2012).
H3: Benefactors (Japanese participants) who receive a message expressing gratitude or both gratitude and apologies feel more efficacious than those who receive a message expressing apologies and a message without either of them.
The Influences of Effort
Effects of received gratitude and apologies predicted in this study could be moderated by the degree to which a benefactor puts effort into helping a beneficiary. Literature indicates that people are more likely to feel grateful and express gratitude when a large amount of effort has been expanded on their behalf (McCullough & Tsang, 2004). In a study by Tesser et al. (1968), participants read three benefit-providing scenarios in which three factors were manipulated: the beneficiary's perceptions of the intention of the benefactor, the cost to the benefactor in providing the benefit, and the value of the benefit. The results indicated that the participants felt more grateful toward the benefactor whose effort was more costly than toward the benefactor whose effort was less costly. Further, participants in another study (Okamoto & Robinson, 1997) were most likely to express gratitude when the imposition on the benefactor was highest. Taken together, beneficiaries are more likely to experience and express gratitude especially when the effort by a benefactor was perceived to be substantial.
However, past studies have made little attempt to examine the impact of how costly the effort by a benefactor is on the benefactor’s perception. Based on the previous studies above, it is possible that benefactors who believe that they expended much effort to help a beneficiary perceive received gratitude and apologies to be more meaningful than those who believe that they expended little effort. Specifically, benefactors who put a great deal of effort into helping a beneficiary may be more highly motivated to expect to receive gratitude and apologies than those who put little effort. In such cases, the effect of received gratitude and apologies on the benefactor’s perceptions is predicted to be significant. To examine the possibility, the following hypothesis was developed:
H4: Effects of received gratitude and apologies on benefactors’ negative/positive face and self-efficacy are stronger when their effort to help a beneficiary is more costly than when their effort to help a beneficiary is less costly.
Method
Participants and Procedures
This study was approved by the research ethics committee of the author’s university. The online survey, with all instructions and questions written in Japanese, was completed by 671 Japanese university students. Two recruiting methods were used. First, 587 students participated in this study through a sample recruitment firm. Second, snowball sampling was utilized to recruit 84 participants. After receiving an explanation of the study, the participants gave their informed consent by taking part in it. The mean age of the participants was 21.6 (SD = 1.61) and they were between the ages of 19 and 30. One hundred thirty-seven (20%) of the participants reported they were men while 520 (77%) of the participants reported they were women. Nine (1%) participants answered that their gender was other. Finally, five (1%) did not desire to report their gender.
After answering questions regarding their demographic information above, a following message showed up on the survey screen which was sent from a hypothetical friend attending the same university: “Long time no see! How are you doing? I am writing this message to ask a favor of you. I plan to study abroad next year, so I have prepared some application documents. I have tried my best to prepare the documents because whether or not I could pass the selection process for the study abroad program in my university depends on the quality of the documents. I want to join the program of ABC University for which I have been writing a statement of purpose. It is written in Japanese since it will be read by the selection committee members in my university in Japan. I would be happy if you could check this document to help me find out how I could improve it further. Let me paste and copy here the contents in the statement of purpose. Please let me know if you find the contents I should improve or any mistakes such as typographical errors and omissions! Yoroshiku onegai shimasu (Best regards).”
As shown above, in the message, the hypothetical friend asked the participants to provide feedback on a statement of purpose in which the friend wrote their reasons for applying to a study abroad program at ABC University. After reading the document, the participants typed the feedback for the statement of purpose in a blank box and the reply was analyzed based on the amount of effort that was expended. After finishing writing the feedback, they automatically received a reply from the friend. The content of the reply varied depending on four different conditions to which participants were randomly assigned: control (n = 175), gratitude (n = 160), apologies (n = 169), and both gratitude and apologies (n = 167).
The following reply was read by the participants in the control condition: “I read your advice! ( ) Actually, besides ABC University, I need to submit another statement of purpose for XYZ University. Can you read and give me some advice on the document, too? If you are too busy, that’s ok. Your help would be appreciated!” In the gratitude condition, the following statements were added in the parentheses: “Thank you so much! I appreciate it!” In the apology condition, the following sentence was added in the parentheses: “I’m sorry to bother you when you are busy!” Finally, in the parentheses of the both condition, the following contents were added: “I’m sorry to bother you when you are busy! Thank you so much! I really appreciate it!” After reading the reply, the participants were asked to rate the content of the message as well as the actions and impressions of the hypothetical friend.
The results of the realism check (3 items, α = .88) indicated this hypothetical situation was perceived by the participants to be relatively realistic (M = 3.54 out of 5) without significant differences across different groups (F = 0.94, p = .421, η2 = .004). Further, based on the results of the G*Power analysis, MANOVA, eight groups (4 x 2), three DVs, α = .05, power = .8, f2(V) = .15 according to Cohen (1988), showing that 56 participants were necessary for the analyses, the sample size of this study (n = 671) was considered large enough. (Faul et al., 2009; Faul et al., 2007).
Measures
Participants then answered items of the following measurements: negative face threat, positive face threat, and self-efficacy. Table 1 indicates the reliabilities (Cronbach’s αs), means, and standard deviations for each measure. The back translation method was utilized since original measures were written in English. This process was completed by the principal investigator and a research assistant; both of them were fluent in Japanese and English.
Table 1
Reliabilities, Means, Standard Deviations, and Correlations
| Variable | 1 | 2 | 3 | α | M | SD |
|---|---|---|---|---|---|---|
| 1. Negative face threat | — | .91 | 2.70 | 0.82 | ||
| 2. Positive face threat | .77** | — | .91 | 2.31 | 0.76 | |
| 3. Self-efficacy | -.11** | -.09* | — | .86 | 3.12 | 0.92 |
*p < .05. **p < .01.
Negative and Positive Face Threat
Participants rated a message from a hypothetical friend regarding the degree to which the message threatened their negative and positive face with ten items for each type of face. This scale was adapted from Lee et al. (2012), which assessed face-threatening acts in asking for favors. Two negative face threat examples are “This message will restrict what I do” and “This message is likely to bother me.” Two positive face threat example items are “This message could make me behave inconsiderately” and “This message could make me feel bad about myself.” A 5-point Likert scale was used (1 = strongly disagree, 5 = strongly agree), with higher scores indicating more face threat. The Cronbach’s α values of the negative and positive face threat scales were .91 and .91.
Self-Efficacy
Following the procedure of a study by Grant and Gino (2010), participants’ self-efficacy was assessed by three items adapted from Bandura’s (1990) perceived self-efficacy scale. Example items are “I feel capable to help this friend improve the statement of purpose” and “I feel competent to help this friend improve the statement of purpose.” The higher scores indicated higher levels of self-efficacy and the Cronbach’s α value of this scales was .86.
Effort
The amount of effort that participants put into writing feedback on a statement of purpose for ABC University (the first request) was calculated based on the number of letters they wrote to the friend in a response explaining how to improve the statement of purpose. The mean score of the number of letters was 159. The minimum score was 21 and the maximum score was 727. Two groups were created by splitting all the participants based on the median score, 128. The number of participants who put in less effort was 335 and that of participants who put in more effort was 336.
Results
An experimental manipulation successfully functioned. Participants in the gratitude condition (M = 3.98, SD = 0.81) and those in the both condition (M = 4.00, SD = 0.83) reported that the hypothetical friend expressed more appreciation than those in the apology condition (M = 3.06, SD = 0.93) and those in the control condition (M = 2.19, SD = 0.94), F(3, 667) = 162.93, p < .001, η2 = .423. Further, participants in the apology condition (M = 3.48, SD = 0.96) and those in the both condition (M = 3.28, SD = 1.08) reported that the hypothetical friend expressed more apologies than those in the gratitude condition (M = 2.67, SD = 1.03) and those in the control condition (M = 2.38, SD = 1.03), F(3, 667) = 45.97, p < .001, η2 = .171. Those who participated through snowball sampling (M = 3.45, SD = 0.72) reported higher levels of self-efficacy than those who participated through the other method (M = 3.08, SD = 0.94), so the effect of the different recruiting methods was controlled for in the following analyses.
To test hypotheses 1, 2, 3 and 4, a two-way MANOVA was conducted to investigate the effect of different messages (i.e., control, gratitude, apologies, and both) and the amount of effort (i.e., more/less effort) on negative face threat, positive face threat, and self-efficacy controlling for the effect of different recruiting methods. As shown in Table 2, the results generated a significant effect of different messages, Wilks’ Lambda = .944, F(9, 1603) = 4.30, p < .001, η2 = .019, and the amount of effort, Wilks’ Lambda = .977, F(3, 659) = 5.17, p < .01, η2 = .023. Accordingly, separate ANOVAs were conducted for follow-up tests, all of which also controlled for the effect of different recruiting methods.
Table 2
Significant Multivariate Effects (at p < .05 Level)
| Effect | Wilks’ lambda | F | df | Error df |
|---|---|---|---|---|
| Conditions | .944 | 4.295 | 9 | 1603 |
| Effort | .977 | 5.167 | 3 | 659 |
Negative Face Threat
H1 predicted that benefactors receiving a message showing gratitude, apologies, or both of them would perceive that their negative face was less threatened by a beneficiary than those receiving a message without gratitude or apologies. The results of a 4 (messages) x 2 (effort) ANOVA showed a significant main effect for different messages on negative face threat, F(3, 662) = 10.55, p < .01, η2 = .046. As indicated in Table 3, follow-up tests indicated that participants who received a message expressing gratitude (M = 2.55, SD = 0.80) and those who received a message expressing gratitude and apologies (M = 2.52, SD = 0.82) perceived the message to threaten their negative face significantly less than those who received a message without either of them (M = 2.96, SD = 0.83). These results are partially consistent with H1. The degree to which the participants put effort into helping a hypothetical friend did not have a significant impact on negative face threat.
Table 3
Effects for Conditions Using the Bonferroni Correction
| Dependent Variables | Control | Gratitude | Apologies | Both |
|---|---|---|---|---|
| Negative face threat | 2.96a (0.83) | 2.55b (0.80) | 2.75ab (0.73) | 2.52b (0.82) |
| Positive face threat | 2.56a (0.81) | 2.18bc (0.72) | 2.37ac (0.63) | 2.13b (0.80) |
| Self-efficacy | 3.08 (1.04) | 3.08 (0.85) | 3.10 (0.89) | 3.23 (0.88) |
Note. Standard deviations are reported in parentheses. Different superscripts indicate significant differences based on post hoc analyses using the Bonferroni correction test at p < .05.
Positive Face Threat
H2 predicted that benefactors receiving a message showing gratitude, apologies, or both of them would perceive that their positive face was less threatened by a beneficiary than those receiving a message without gratitude or apologies. The results of a 4 (messages) x 2 (effort) ANOVA showed a significant main effect for different messages on positive face threat, F(3, 662) = 11.23, p < .01, η2 = .048. As indicated in Table 3, follow-up tests indicated that participants who received a message expressing gratitude (M = 2.18, SD = 0.72) and those who received a message expressing gratitude and apologies (M = 2.13, SD = 0.80) perceived the message to threaten their positive face significantly less than those who received a message without either of them (M = 2.55, SD = 0.81). Further, those who received a message expressing gratitude and apologies perceived the message to threaten their positive face significantly less than those who received a message expressing apologies (M = 2.37, SD = 0.63). These results are partially consistent with H2. The degree to which the participants put effort into helping a hypothetical friend did not have a significant impact on positive face threat.
Self-Efficacy
H3 predicted that participants who receive a message showing gratitude or both gratitude and apologies would report higher levels of self-efficacy than those who receive a message showing apologies and a message without either of them. The results of a 4 (messages) x 2 (effort) ANOVA showed a significant main effect for effort, F(1, 661) = 9.13, p < .01, η2 = .017, but not for different messages, F(3, 661) = 1.01, p = .39, η2 = .005. The results were not consistent with H3. As shown in Table 4, follow-up tests indicated that participants who put more effort into helping a hypothetical friend (M = 3.27, SD = 0.90) felt more efficacious than those who put less effort (M = 2.98, SD = 0.92).
Table 4
Effects for Effort Using the Bonferroni Correction
| Dependent Variable | Less | More |
|---|---|---|
| Negative face threat | 2.72 (0.80) | 2.67 (0.83) |
| Positive face threat | 2.37 (0.78) | 2.23 (0.74) |
| Self-efficacy | 2.98a (0.92) | 3.27b (0.90) |
Note. Standard deviations are reported in parentheses. Different superscripts indicate significant differences based on analyses using the Bonferroni correction test at p < .05.
H4 hypothesized that effects of received gratitude and apologies would be stronger when benefactors’ effort to help a beneficiary was more costly than when their effort to help a beneficiary was less costly. The results of a MANOVA did not show the significant interaction effect between messages and effort on negative face threat, F(3, 661) = 0.80, p = .49, η2 = .004, positive face threat, F(3, 661) = 0.76, p = .52, η2 = .003, and self-efficacy, F(3, 661) = 2.62, p = .05, η2 = .012. An F-statistic that a MANOVA provides indicates whether one or more significant differences among group levels exist, but any information about specific group means compared to each other is not obtained by the statistic (Hsu, 1996; Midway et al., 2020). Taking into consideration the fact that self-efficacy was the only variable impacted by the different amounts of effort as indicated in the results of H3 and the interaction effect on self-efficacy was close to the significant level (p = .05), post hoc comparisons using the Bonferroni correction (see Figure 1) were conducted to examine the role of effort moderating the impact of messages on self-efficacy. The results showed that for the participants who received a message with apologies, those who put more effort (M = 3.45, SD = 0.77) felt more efficacious than those who put less effort (M = 2.82, SD = 0.89). Such differences were not observed for those who received a message with gratitude, with both gratitude and apologies, or without either of them. Thus, the results were partially consistent with H4.
Figure 1
Comparison of Mean Scores for Less Effort and More Effort on Different Conditions
*p < .05.
Discussion
Partially consistent with predictions in H1 and H2, the results indicated that participants who received a message showing gratitude or both gratitude and apologies (but not apologies solely) perceived that their negative and positive face were less significantly threatened by a beneficiary than those who received a message without either of them. Based on the conceptualization by Brown and Levinson (1987), this study defines negative face as a desire to maintain autonomy from others and to preserve freedom of action, whereas positive face is defined as a desire to gain approval and be positively evaluated by others. In the current study, after participants (benefactors) provided help to a hypothetical friend (beneficiary), they received a message with or without gratitude/apologies from the friend. If they did not receive gratitude or apologies even after providing help, it was assumed that their negative face would be threatened because they could be afraid of being asked a favor again by the friend who did not recognize their effort. Their positive face would also be threatened because they believed that they failed to have a positive impression on the friend who did not recognize their kindness.
The results showed that gratitude (and both gratitude and apologies) worked to moderate the negative and positive face threats, but apologies did not. These findings may indicate that while it is true that Japanese sometimes use apologies to show gratitude (Coulmas, 1981; Lee et al., 2012), they prefer to receive gratitude when they do something beneficial to others. Therefore, if an individual violates the norm that people should express gratitude to benefactors, people might believe that the individual is not good at face-work to moderate face threats to self and others (Goffman, 1967). The impression is critical since Japan is a tight culture in which ingroup members are expected to behave according to the social norms, and behavioral deviations that would be considered trivial in the West are recognized and criticized (Chan et al., 1996; Triandis, 1989). Taken together, the participants who received a message without gratitude (a message with apologies only and a message without either gratitude or apologies) might perceive that their negative and positive face were threatened by the friend who violated the social norm that people should express gratitude when benefits are provided.
It should be also noted that gratitude and apologies did not uniquely moderate negative face threat and positive face threat. Overall, the results indicated that the participants who received a message expressing gratitude (gratitude only or both gratitude and apologies) perceived that their negative and positive face were less significantly threatened by a beneficiary than those who received a message without either of them. Literature implies that gratitude may function to meet the message receiver’s positive face since gratitude allows a message sender to show the recognition that the message receiver provides positive outcomes, which should help the receiver maintain the socially desired image (Chaudhry & Loewenstein, 2019). Also, apologies could theoretically meet the message receiver’s negative face because apologies allow a message sender to take blame for negative outcomes such as indebtedness, which could help the message receiver believe their freedom will not be threatened anymore (Chaudhry & Loewenstein, 2019). The results in this study did not provide evidence showing that these two types of communication styles specifically impact negative or positive face. A reason for the results could be that received gratitude might simply help the participants believe that a message including gratitude after providing help is socially appropriate. Lee et al. (2012) found that in a favor-asking situation Japanese used apologies as an attempt to reduce face threat, but their results did not reveal the consistent effect of apologies on specific types of face threats either.
The theoretical accounts above may make sense according to the arguments in rapport management proposed by Spencer-Oatey (2008). Based primarily on politeness theory, rapport management focuses on an aspect of language as functions to manage sociality rights and interactional goals. The theory considers perceived sociality rights and obligations important factors influencing interpersonal rapport. Sociality rights and obligations involve the management of social expectancies which are fundamental to the social entitlements that people claim for themselves through interactions with others. Specifically, based on social expectancies, people are concerned about fairness, consideration, and behavioral appropriateness. In the current study, the participants might have expectancies in which beneficiaries should express gratitude after they receive favors. Therefore, a message without gratitude might have the participants report that various types of their face needs were not met through the interaction with the hypothetical friend.
Contrary to expectations in H3, received gratitude or both gratitude and apologies did not help a benefactor feel efficacious in this study. For American participants, received gratitude after a benefit was provided increased a benefactor’s self-efficacy (Grant & Gino, 2010). The discrepancy between these distinct results could be explained by the unique culture in Japan; that is, when Japanese people are positively evaluated by others (such as when they receive compliments), they tend to deny the positive evaluation (Chen, 2010). For example, when Japanese are praised for being competent, one of the typical responses is “That’s not true” (Daikuhara, 1986). Taken together, in this study, gratitude which theoretically functions to help a benefactor be efficacious by recognizing their competence to generate a positive outcome (Chaudhry & Loewenstein, 2019) may not have worked in that manner among Japanese benefactors.
Corroborating the theoretical accounts provided above, the participants’ self-efficacy was higher after they received apologies among those who put more effort into helping a beneficiary than those who expended less effort. Apologies are considered a means for an individual to indicate their recognition of the indebtedness and negative outcomes they bring to others (Coulmas, 1981). Therefore, in this study, received apologies could help the participants believe that the large amount of effort they put was correctly recognized by the beneficiary and the belief may have increased their perceived capacity and competence to produce positive outcomes (i.e., self-efficacy).
On the other hand, received gratitude did not function to enhance the positive impact of the effort on the participants’ self-efficacy. Gratitude technically functions to help a beneficiary signal that a benefactor brings about a positive outcome (Chaudhry & Loewenstein, 2019). However, in this study, gratitude did not work that way. The reason for the results could be that gratitude may not work to help a beneficiary show their recognition of their indebtedness to the benefactor. In order for the participants to enhance their self-efficacy, it seems that their effort should be “confirmed” by the hypothetical friend through expressed apologies. Specifically, when Japanese receive favors, it is highly likely that they feel indebtedness and regret along with gratitude (Benedict, 1946; Naito & Sakata, 2010). If so, it is possible that Japanese benefactors believe that beneficiaries should feel and express indebtedness through apologies. Expressing gratitude may not be effective to indicate the perceived indebtedness and may not help the benefactors confirm that their effort was correctly recognized by the beneficiary. Taken together, various theoretical accounts suggest that apologies rather than gratitude are effective to facilitate the positive influence of the effort on Japanese benefactors’ self-efficacy.
Implications and Limitations
So, when a benefactor provides a benefit for a beneficiary in Japan, should the beneficiary say, ‘Thank you’ or ‘Sorry’ to the benefactor? The answer might be that it is critical to include gratitude in a message, but it would not hurt to include apologies in addition to gratitude. The results related to apologies were mixed: a message with apologies could enhance a benefactor’s self-efficacy (H4), but apologies combined with gratitude in a message did not generate additional positive effects compared with a message with gratitude solely (H1 and H2). If a beneficiary desires to help a benefactor be efficacious, the beneficiary may want to use apologies to show their understanding of their indebtedness to the benefactor. However, the safest way is to put gratitude and apologies together into one message, which is not uncommon in Japanese conversations (Iio, 2017; Kumatoridani, 1991).
One of the limitations in this study is that it did not examine cultural differences directly by testing hypotheses among people from different cultures. For example, the results indicate that apologies were not as effective as gratitude to moderate threats to Japanese participants’ face, but the effects could be considered effective when compared with these among American participants like other studies (e.g., Lee et al., 2012). Another limitation is that this study assumes that the number of letters the participants wrote in a response to a hypothetical friend measured the amount of their effort, but there was no objective basis for the assumption. For instance, it is unclear that participants who wrote a long message to the hypothetical friend really perceived that they put a great deal of effort into helping the friend. Participants who wrote a short massage could believe that they put in considerable effort probably because they carefully chose words with an attempt to help the friend. Therefore, the results obtained in this study related to the effort to help should be interpreted with caution. Future studies could ask the participants how much effort they put into writing a response to the hypothetical friend to measure their perception of effort. It may be also possible to examine the time the participants spend on helping others like other studies (e.g., Bartlett & DeSteno, 2006).
Further, the hypothetical situation in which another request was made after the first request was accommodated might impact the expectancy and appropriateness regarding the use of gratitude and apologies. Specifically, after providing help, the participants received a message from a hypothetical friend that made another request. If this situation was different such that the participants received a message from the friend that only responded to the fact of receiving help without making another request, the participants’ perception toward gratitude or apologies could be different. For example, if the participants receive a message with apologies only and without being made another request, they could perceive the message to effectively moderate their negative and positive face threat unlike the results of this study. The different results could be expected because without a further request, benefactors might not expect their friend to express formal gratitude. In fact, Chinese may experience negative feelings after receiving thanks from a close friend since helping for a close friend is implicitly embedded in their cultural norm and expressing thanks may indicate the relationship is not close enough (Zhang et al., 2018). Therefore, the results of this study should be interpreted cautiously because the appropriateness and expectancy regarding the use of gratitude and apologies could be impacted by the specific situation involving an additional request from a beneficiary.
It should be also noted that theoretical standpoints based on politeness that specifically considers face to be relational, interactional, and culture-specific (Arundale, 2006; Haugh, 2005) suggest the limitation of applying negative face and positive face across cultures. For instance, Haugh (2005) argues that Japanese politeness involves concern more about both the “place one belongs” and the “place one stands” than about positive face and negative face. The term uchi, which means “the place one belongs” and “inside,” indicates being included in a group. The term tachiba, which means “the place one stands,” indicates one's rank and circumstances relative to others. There is some conceptual overlap between positive/negative face and uchi/tachiba such that uchi can be related to positive face because people are concerned about their socially desired image in order for them to be included in a group. However, in general, uchi and tachiba are theoretically distinct from positive face and negative face. The participants in the current study might have been more concerned about the possibility to be included in a relationship with the hypothetical friend or the difference in social statuses between them and the friend than their positive or negative face. In other words, the participants could be concerned about the following questions: “In the first place, how close is this friend to me? (concerns about uchi)” and “Does the friend have a higher status? (concerns about tachiba).” Taken together, the effects of gratitude and apologies were not systematically linked to their positive face and negative face since the two types of communication styles may not be theoretically effective to maintain the needs of uchi and tachiba.
Finally, since the effect sizes observed in this study were small, it is critical to consider other possible variables influencing the effects of different messages on various types of face needs. For example, how close the participants felt toward the hypothetical friend might have played a significant role. Specifically, if the participants believed that they were close to the friend, they would not care or they would care more if they received gratitude or apologies from the close friend since the psychological distance theoretically impacts the necessity of the use of politeness strategies (Brown & Levinson, 1987). Further, the length of the message the participants received from the hypothetical friend could influence the participants’ perceptions of positive and negative face as well as self-efficacy. The mean score of the number of letters the participants wrote to the friend was 159, which was relatively long. However, in response to that, for example, the participants in the control condition received a response composed of only 102 letters, which was mostly for making another request. The relatively short response from the hypothetical friend might have threatened the participants’ face significantly, so the effectiveness of gratitude and apologies could be suppressed. Future research should test other possible factors that could impact benefactors’ perceptions of various types of face needs in the help-providing situation.
Conclusion
The current study attempts to investigate the roles of gratitude and apologies in meeting various aspects of psychological needs of benefactors in Japan. The results indicate unique roles of these communication styles which met the face needs and competence needs of Japanese in some situations. Specifically, received gratitude helped a benefactor meet their negative and positive face needs. Also, received apologies functioned to meet a benefactor’s competence needs when the benefactor put much effort into helping a beneficiary. These results highlight the necessity for further research that recognizes the importance in different roles of gratitude and apologies across cultures.