Cultural Differences Between Japan and the United States in Uses of “Apology” and “Thank You” in Favor Asking Messages

Abstract

Three studies investigated whether apologies and thanks are used differently when asking favors in the United States and Japan and examined whether their use makes a favor asking message less face-threatening. In Study 1, participants (N = 152) composed an email message for a favor asking situation. Next, participants in Study 2 (N = 634) and Study 3 (N = 417) completed one of four versions of a questionnaire regarding a prototype of an email message. Results showed that (a) more Japanese included apologies in their messages while more Americans used thanks and (b) Japanese considered apologies to reduce some face threats while Americans did not consider thanks to reduce face threats. Implications and future research directions are discussed.

Keywords

cross-cultural research apology gratitude face culture

Many communication processes are characterized by culture and language use. A speech act refers to a minimal unit of discourse that is transferable from language to language (Coulmas, 1981). Apologies and gratitude statements are two kinds of extremely frequent and routine speech acts (Ide, 1998). Although these speech acts exist in many different languages, culture can determine what kinds of behaviors and absences of behaviors require these speech acts and how these obligations in communication patterns can be met verbally.

Apologies and/or thanks are a gracious way of favor asking (Coulmas, 1981; Ide, 1998; Searle, 1976). One of the reasons to use them is an attempt to reduce the face threats associated with favor asking. Because favor asking indicates that a speaker (S) fails to avoid impeding the freedom of action of a hearer (H; Brown & Levinson, 1987; Wilson, Aleman, & Leatham, 1998), use of apologies and thanks in favor asking messages may lessen impressions of presumptuousness (Roloff, Janiszewski, McGrath, Burns, & Manrai, 1988). However, it is questionable whether apologies and thanks are used similarly in favor asking across cultures and, furthermore, whether their use serves universal functions in reducing face threats done by favor asking. Even when one knows how to say “I am sorry” and “thank you” in another language, if one does not know the situations in which each should be said, he/she may seem impolite, insincere, or strange. Therefore, understanding the uses and functions of apologies and thanks in favor asking messages across cultures is important for effective intercultural communication.

To articulate cultural differences in uses of “apology” and “thank you,” we focus on a comparison of just two cultures. Regarding apology, Japan has been selected because Japan is known for being “the most apologetic country in the world” (Kristof, 1995, p. 1), and empirical studies have confirmed the importance of apologies in many situations in Japan to include daily conversation and in judicial proceedings (cf. Barnlund & Yoshioka, 1990; Wagatsuma & Rosett, 1986). Also, Japanese are more concerned about the benefits of apology. For example, Japanese frequently say “my apologies” to express indebtedness, whereas, in contrast, Americans are more concerned about the costs of apology (Sugimoto, 1999), and have been characterized by their frequent use of thanks (Kwon, 2000) in various situations. For example, writing “thanks in advance” is common in letters and notes asking favors in North America, whereas writing “thanks in advance” in Japan is viewed as rather demanding and impolite (Ohashi, 2003). Finally, Japan and the United States represent East Asian and Western cultures, respectively (Heine, Lehman, Markus, & Kitayama, 1999; Oyserman, Coon, & Kemmelmeier, 2002). Therefore, we selected Japan and the United States to maximize comparability.

With Japanese and American cultural contexts, the current study has two particular goals. First, it will be confirmed whether Japanese are more likely to use apologies than Americans in favor asking, and whether Americans are more likely to use expressions of gratitude than Japanese. Second, it will be explored how well face concepts relate to preferences for apologies or thanks based on Brown and Levinson’s (1987) theory of politeness, which explains how expressing apologies or thanks along with favor asking is an attempt to restore face by using positive and negative politeness strategies.

Toward these goals, apologies and thanks as speech acts will first be defined and compared with each other. Next, a hypothesis will be presented concerning cultural differences in apologies and thanks. Then, face will be defined and four different face concerns in a favor asking situation will be explained. Finally, a hypothesis and three research questions regarding the cultural differences of face concerns will be proffered.

Apology and Thanks as Speech Acts

An apology as a speech act aims to provide support for H who was actually or potentially offended by a violation (Olshtain, 1989) and to restore balance between H and S (Leech, 1983). One of the functions of apologies is to acknowledge a regrettable offense and accept partial or full responsibility for it (Kramer-Moore & Moore, 2003). Admitting S’s own failure to meet an implicit or explicit obligation to other people is a genuine apology. People also apologize simply to fulfill expectations of others, as in when two people accidentally bump into each other in a crowded place. Even when there is no serious offense, people often say, “I am sorry.” This situation illustrates the routine use of apologies as a ritualistic remedy. An apology used in this way may serve to initiate a conversation or it may represent a gracious way of favor asking.

Thanks aim to provide support for H through the admission of S’s acceptance of a debt and thereby restore equilibrium between H and S. People say “thank you” to express their appreciation of a past act performed by H, which benefits S or S believes to have benefited him/her (Searle, 1976). Additionally, thanks are directed to some action(s) of a benefactor because of a beneficiary’s belief that he/she will receive a benefit (Coulmas, 1981). People sometimes say thank you for the expected benefits they will receive when they make requests or ask favors. S may imply not only his/her appreciation for the future benefit but also exert pressure on H to comply with the request or to grant the favor.

Searle (1976) classified thanks and apologies as expressive illocutionary acts because both speech acts express S’s psychological state toward the actions done by S or H. When they are used to express genuine states, objects of gratitude and regret are two key components of thanks and apologies, respectively (Coulmas, 1981). Because it is sometimes difficult to keep gratitude and regret distinct from each other, a common domain is defined where thanks and apologies are both appropriate (Coulmas, 1981; Kumatoridani, 1999). A favor calling for verbal gratitude could be turned into an offense calling for verbal apology or vice versa by a slight shift in the interpretation of the situation (Coulmas, 1981). Although thanks and apologies are distinct, the objectives of gratitude and the objectives of regret are sometimes similar to each other with regard to indebtedness. Thanks imply the indebtedness of S for his/her own benefit, and apologies indicate S’s actual recognition of his/her indebtedness to H. In other words, apologies are performed based on the event S will bring about that will be inconvenient or offensive to H, whereas thanks are performed based on the event H will bring about to please S. Because favor asking causes perceptions of gratitude and indebtedness (Goei, Roberto, Meyer, & Carlyle, 2007), if S assumes H will bring about the asked for state or event, favor asking is a common situation in which either thanks or apologies can be used.

Cultural Differences in the Usage of Apologies and Thanks

Favor asking is defined as a directive that, like the more prototypical request, indicates S’s wish or need for H to bring about some desired state or event that would not be provided without being asked for (cf. Becker, 1982; Goldschmidt, 1998). Favor asking induces perceptions of gratitude and indebtedness if S assumes H will bring about the asked-for state or event. Thus, favor asking is a common domain where either thanks, apologies, or both can be found. In general, expressions of apology and gratitude exist across cultures, but their uses differ from culture to culture.

When Japanese and American cultures are compared in the use of apology, Japanese tend to apologize more frequently and explicitly than do Americans. To give some examples, Barnlund and Yoshioka (1990) found that Japanese prefer more direct and extreme forms of apology than do Americans. Furthermore, Wagatsuma and Rosett (1986) provide some anecdotal evidence related to social and legal systems to argue that Americans are less likely than Japanese to apologize formally to those that they have injured. Japanese people also use apologies to express their gratitude (Ide, 1998; Kumatoridani, 1999; Sugimoto, 1999). However, an apology in Japan does not necessarily mean that the person is acknowledging a fault and contextual factors play a crucial role in apology situations (Tanaka, Spencer-Oatey, & Cray, 2008).

Different needs for positive self-regard may explain why Japanese use apologies more frequently than North Americans. North Americans demonstrate a high need for self-enhancement orientations. A high need for self-enhancement and a positive self-view are more prevalent among North Americans than among Japanese (Heine & Lehman, 1999; Heine et al., 1999). For North Americans, presenting low self-esteem or admitting negative aspects of self indicates inadequacy. Therefore, publicly admitting individual faults through apologies tends to be avoided.

In contrast, East Asian people are motivated to maintain self-critical and self-improving orientations (cf. Heine et al., 1999; Kim, 2003). Thus, their need for positive self-regard is not as high in comparison. Japanese people generally assume that their selves are inherently incomplete, so they have to improve themselves through constant effort. This orientation might be influenced by Confucianism. Confucianism holds that the self is fundamentally flawed and so emphasizes modesty, which is needed to balance self-improvement, good behavior toward oneself, and good behavior toward others (Kim, 2003). Apologies reflect the critical self-evaluative nature of Japanese people. Sugimoto (1997) also speculated that, for Americans, saying “I am sorry” is an admission of responsibility and subsequently a sign of low self-esteem. However, for Japanese, apologizing may be a sign of care and concern on behalf of H. In addition, apologies indicate S’s ability to see the situation from the other’s point of view. Thus, apologies are a highly regarded virtue in Japan (Sugimoto, 1999).

Although it is expected that Japanese will give apologies more often than Americans, Americans are expected to use gratitude statements more often than Japanese. This may be explained by the cultural difference between promotion goals and prevention goals. Regulatory focus theory (Higgins, 1998) distinguishes promotion goals with approach motivation from prevention goals with avoidance motivation to explain how individuals regulate their behaviors. Promotion goals focus on attaining positive outcomes such as advancement, achievement, and aspirations, but prevention goals emphasize avoiding negative outcomes such as responsibilities, obligations, and security. Americans adopt more promotion goals than East Asians like Koreans (e.g., Elliot, Chirkov, Kim, & Sheldon, 2001) and Chinese (e.g., A. Lee, Aaker, & Gardner, 2000), who instead favor prevention goals. Expressing apologies and thanks mainly admits regret and cost (i.e., prevention goal) and appreciation and benefit (i.e., promotion goal), respectively.

In sum, Japanese use apologies more often than Americans (e.g., Barnlund & Yoshioka, 1990; Sugimoto, 1999), and Americans use thanks very frequently (Kwon, 2000). Finally, Japanese use apologies to express their gratitude (Ide, 1998; Kumatoridani, 1999). These studies focused on one speech act. To understand uses of these speech acts across cultures, it is necessary to examine both apologies and thanks in the same social event of favor seeking. Therefore, the following hypothesis is presented.

Hypothesis 1: Japanese will be more likely than Americans to use apologies in favor asking (Hypothesis 1a), while Americans will be more likely than Japanese to use expressions of gratitude (Hypothesis 1b).

Expressions of Apology and Gratitude as Facework Strategies

Brown and Levinson’s (1987) theory of politeness explains how individuals produce linguistic politeness to minimize any face threats that might be involved in carrying out goal-directed activities. Favor asking is one type of face threatening act (FTA), and expressing apologies or thanks along with favor asking may be explained as an attempt to reduce the face threats associated with favor asking. In particular, positive and negative face concerns are useful in explaining why Japanese prefer to use apologies more than thanks and why Americans prefer to use thanks more than apologies in favor asking.

Building on Goffman’s (1967) classic analysis, Brown and Levinson (1987) suggested face as a key element in the concept of politeness. Face is defined as the public self-image that everyone wants to claim for himself/herself (Brown & Levinson, 1987). Also, face is interactional because identification of behaviors to evaluate or the ability of other(s) to make the evaluations is not possible without interaction (Haugh, 2009). Therefore, face is “the ongoing and ever-changeable level of regard that accrues to persons engaged in interaction” (Lerner, 1996, p. 303). It consists of two types of face concern, negative and positive. Negative face concern is associated with desires for autonomy and freedom from imposition, whereas positive face concern is the desire for appreciation and approval by at least a few others. When two people interact with each other, S’s positive face, S’s negative face, H’s positive face, and H’s negative face are potentially concerned. Because face is the public image that people desire, and because the faces of S and H are constructs of the same order (Goffman, 1967), people want to protect both their own face as well as H’s face in interactions (Brown & Levinson, 1987; Wilson, 1990).

FTAs are defined as acts that violate or fail to satisfy positive and/or negative face concerns. Particularly, favor asking indicates that S fails to avoid impeding the freedom of action of H. It also obligates H to grant the favor, so it threatens H’s negative face (Brown & Levinson, 1987; Wilson et al., 1998). Aside from the threat to H’s negative face, S may threaten his/her own positive face by asking a favor, because doing so may imply that he/she is not capable of handling his/her problem alone (Goldsmith, 1992), or is lazy rather than independent (Craig, Tracy, & Spisak, 1986). S may also hesitate to ask a favor because he/she would be risking becoming indebted to H, which would constrain S’s future autonomy (Roloff et al., 1988).

In an attempt to address these concerns, S engages in various facework strategies to lessen or rectify the damage done by favor asking (Brown & Levinson, 1987; Cupach & Metts, 1994). Favor asking requires softening by redress in the form of politeness strategies. In the context of asking a favor, two such strategies include apologies and thanks. “I am sorry” and “thank you” are the two most popular examples of polite speech acts. Roloff et al. (1988) explained that apologies and gratitude statements are persuasive elements in favor asking messages because they minimize impressions of presumptuousness, create politeness, and finally lead to a greater likelihood of H’s compliance.

Apologies and thanks both represent forms of redress, but they elicit different effects on the face needs of S and H. Apologies can help to decrease the imposition implied by favor asking and lessen the threat to H’s negative face. With regard to S’s positive face, apologies can function in two ways. First, apologies threaten the positive face of S by denigrating oneself through H (Brown & Levinson, 1987; Holtgraves, 2002). Second, apologies can reduce the threat to S’s positive face by displaying S’s integrity to apologize and/or fulfilling H’s expectation to receive the apology and thus restore H’s positive impression of S. It is unclear, however, which of these possibilities are more likely with regard to favor asking between Americans and Japanese.

In contrast to apologies, thanks threaten S’s negative face by indicating S’s acceptance of a debt and restricting his/her future autonomy (Brown & Levinson, 1987). Threat to H’s negative face in a favor asking message might be less salient if S’s negative face is threatened. Thanks may also implicate S’s admission of the debt, so S may be indicating that he/she is a person who appreciates any favor that he/she receives. In this way, thanks may reduce the threat to S’s positive face by admitting S’s debt. Therefore, threats to S’s positive face in favor asking might be reduced by adding thanks. Like apologies, thanks help decrease the damage done by favor asking.

In sum, favor asking is an FTA, so expressing apologies or thanks along with favor asking may be explained as attempts to reduce the face threats associated with favor asking. Specifically, expressions of apologies and thanks may threaten S’s positive face and S’s negative face, respectively, more than favor asking without apologies or thanks. Another possibility is that these speech acts function to reduce the threat to S’s positive face normally sustained through favor asking. As a result, favor asking messages that include expressions of apologies, thanks, or apologies and thanks should threaten H’s negative face less than favor asking messages without such expressions. Therefore, the following hypothesis and research questions are proposed.

Hypothesis 2: Favor asking messages with an apology, a gratitude statement, or an apology and a gratitude statement will threaten H’s negative face less than favor asking messages without any apology or gratitude statement.

Research Question 1: Does a favor asking message with an apology threaten S’s positive face more or less than a favor asking message without an apology?

Research Question 2: Which threatens S’s negative face more (Research Question 2a) and S’s positive face less (Research Question 2b), a favor asking message with a gratitude statement or one without a gratitude statement?

If Japanese and Americans prefer apologies and thanks in favor asking, respectively (Hypothesis 1), omission of the appropriate speech act could be a FTA in each culture. For example, because apologies are expected in favor-asking situations in Japan, failure to fulfill that expectation is a face threat (Spencer-Oatey, 2007). Furthermore, if Japanese and Americans prefer different speech acts in favor asking (Hypothesis 1), they will do so for the same reason or for different reasons. If they favor different speech acts for the same reason (e.g., reduction of threat to S’s positive face), this can be understood from the observation that Americans use thank you to reduce the threat to S’s positive face while Japanese use apologies to mitigate the same face threat. Determining the reasons for the different speech acts has implications for intercultural encounters. For instance, Japanese who want to decrease the threat to their positive face (e.g., by making good impressions) while interacting with Americans may need to use thanks rather than apologies.

If Japanese and Americans favor different speech acts for different reasons (e.g., thank you threatens S’s negative face and apology threatens S’s positive face, and accordingly Americans prefer thanks and Japanese prefer apologies), it indicates that Americans may focus more on S’s negative face, while the Japanese focus more on S’s positive face. Thus, this may indicate that favor asking has different relational consequences for people from different cultures, for example, when the other party either grants or refuses the favor. Therefore, the following research question is suggested.

Research Question 3: What cultural differences exist between Japanese and Americans in the extent to which apology and gratitude statements reduce the amount of face threats perceived in favor asking messages?

Study 1

It was predicted that a greater number of Japanese would use apologies than Americans in favor asking (Hypothesis 1a), while a greater number of Americans would use gratitude statements than Japanese (Hypothesis 2b). Participants composed a favor asking email message for a given situation. Next, the presence of apologies and thanks in each message was coded for further analyses.

Method

Participants

Seventy-eight American participants (age M = 20.01, SD = 1.69, and 39.7% male) were recruited from one large Midwestern university in the United States, and 74 Japanese participants (age M = 20.10, SD = 1.48, and 34.7% male) were recruited from a large university in Japan. American participants included 76.9% Caucasians, 5.1% African Americans, and 2.6% Asian Americans. Japanese participants were all ethnic Japanese.

Questionnaire

A paper-and-pencil survey was conducted for Study 1. The questionnaire consisted of the open-ended direction, “Please imagine that you find yourself in the following situation and write your own message for that situation.” Then, a vignette was given and participants were instructed to compose a message in the given blank space for the situation that followed:

You are scheduled to present your project in class tomorrow as the last presenter. You just found out that your grandmother has been hospitalized, and you would like to visit her. The visiting hours are limited and overlap with tomorrow’s class. When you told the professor about your situation, the professor said you may present earlier, and leave the class earlier, if you can find somebody who is willing to switch presentation times with you. Now, you are emailing all of your classmates on the class list to look for somebody who can switch presentation times with you.

The questionnaire was produced in both English and Japanese. Various methods of back-translation and inspection by speakers fluent in both languages were used to ensure equivalence in meaning. Participants in the study completed the questionnaire in their native languages.

Coding

Two bilingual coders fluent in Japanese and English independently coded the messages. In all cases, the message served as the unit of analysis. First, coders identified whether each message included any form of apology or gratitude phrases. Examples are “I am (really, very, so) sorry,” “I apologize,” “Thanks,” “I would (really) appreciate,” and “Thank you (very much)” in English and Japanese. After identifying them, coders classified each message into one of the four categories, which were “None,” “Apology,” “Thank you,” and “Both.” Intercoder reliabilities (Krippendorff’s alpha) calculated by all data were .97 for the Japanese data and .92 for the U.S. data. Any discrepancies were resolved through discussion by the coders.

Results

It was expected that more Japanese would write apologies in favor asking messages than would Americans, while more Americans would write thanks in favor asking messages than would Japanese. A total of 42 out of 74 Japanese (56.8%) used apologies in their messages, whereas 6 out of 78 Americans (7.7%) used apologies. In contrast, only one Japanese (1.4%) included thanks in his message, while 68 Americans (87.2%) used thanks (see Table 1 for details).

Table 1.

Inclusion of an Apology or Thank You in Favor Asking Messages in Japan and the United States (Study 1)

		Message Types
Culture		None	Apology	Thanks	Both	Total
Japan	N	31	42	1	0	74
	Percentage	41.9	56.8	1.4	0	100
	Adjusted residual	4.0	7.8	−9.8	−2.4
The United States	N	10	0	62	6	78
	Percentage	12.8	0	79.5	7.7	100
	Adjusted residual	−4.0	−7.8	9.8	2.4

Note: Cell_JN = Cell for Japan and None; Cell_JA = Cell for Japan and Apology; Cell_JT = Cell for Japan and Thanks; Cell_JB = Cell for Japan and Both; Cell_UN = Cell for the United States and None; Cell_UA = Cell for the United States and Apology; Cell_UT = Cell for the United States and Thanks; Cell_UB = Cell for the United States and Both.

A chi-square test was conducted to examine the relationship between culture and the use of an apology, thanks, both, or none. This test showed a significant association between culture and the use of apology or thank you, χ²(3) = 117.80, p < .001, Vc = .88. Particularly, the adjusted standardized residuals in Table 1 showed that the obtained frequencies in cell_JN, cell_JA, cell_JT, cell_UN, cell_UA, and cell_UT exceeded the absolute value of 2.74 for significance at the .05 level. The obtained frequencies in cell_JN, cell_JA, and cell_UT were significantly above the expected frequencies under the hypothesis of independence, whereas the obtained frequencies in cell_JT, cell_UN, and cell_UA were significantly below the expected frequencies under the hypothesis of independence. Therefore, the data were consistent with Hypothesis 1.

Additionally, the numbers of apology and gratitude statements were counted in each message. The frequencies were also consistent with the results of the chi-square test. Fourteen Japanese (19%) used an apology more than once in their messages, while none of the Americans did. In contrast, 41 (52.6%) Americans wrote a gratitude statement more than once, while none of the Japanese did. The median use of an apology for the Japanese data and for the U.S. data was 1 and 0, respectively; however, the median use of a gratitude statement for the Japanese data and for the U.S. data was 0 and 2, respectively.

Summary

When asked to compose a favor asking message for the given situation, Japanese included apologies, while Americans used thanks. Furthermore, Japanese tended to repeat their apologies in their messages, and Americans tended to repeat their thanks.

Study 2

Study 2 examined whether or not expressing apologies or thanks along with favor asking is an attempt to reduce the face threats made by favor asking. Accordingly, it was hypothesized that favor asking messages with apology, thanks, or both would threaten H’s negative face less than a favor asking message with neither an apology nor thanks (Hypothesis 2). In addition, it was asked whether a favor asking message with an apology threatened S’s positive face more or less than a favor asking message without an apology (Research Question 1). Next, it was asked whether a favor asking message with thanks threatened S’s negative face more (Research Question 2a) or S’s positive face less than a favor asking message without thanks (Research Question 2b). Finally, it was questioned whether there were any cultural differences between Japanese and Americans in the extent to which apology and thanks reduced the amount of face threat perceived in favor asking messages (Research Question 3).

Method

Participants

Three hundred twenty-two American participants (age M = 20.04, SD = 1.44, and 35.6% men) were recruited from a large Midwestern university in the United States, and 312 Japanese participants (age M = 20.31, SD = 1.59, and 37.5% men) were recruited from a large university in Japan. American participants consisted of 80.7% Caucasians, 7.1% African Americans, and 4.3% Asian Americans. Japanese participants were all ethnically Japanese.

Design

Study 2 was a paper-and-pencil survey that used a 2 (national culture: the United States and Japan) × 4 (speech act: control, apology, gratitude, and apology and gratitude) between subject design. Participants were asked to fill out one of the four versions of the questionnaire assessing their perceptions of face threats as the S. Version A included a prototype of an email message¹ as well as the description of the situation described in Study 1, which was as follows:

Subject: Looking for somebody to switch presentation times with me!

Hello COM *** classmates,

I’m scheduled to present my project last tomorrow, but just found out that my Grandma’s in the hospital and I’d like to visit her. The visiting hours are right at the end of class, so I’d like to know if anyone will switch times with me. The professor said it’d be ok.

Sincerely,

( )

Versions B, C, and D added an apology (i.e., “I am terribly sorry for asking this, but”), thanks (i.e., “I really appreciate this”), and both (i.e., “I am terribly sorry for asking this, but” and “I really appreciate this”), respectively, in the same message as in Version A. The prototype email message as well as apology and thanks used in Study 2 were developed based on the messages participants wrote in the given situation in Study 1.

Measurement

The procedure was conducted in the same manner as in Study 1. All the measures used a 5-point Likert-type scale response format (1 = strongly disagree, 5 = strongly agree). After reading one of the four messages, participants indicated their perceptions of S’s negative (4 items) and positive (5 items) face threats, and H’s negative (5 items) and positive (5 items) face threats.² The items can be viewed in the appendix.

Cronbach’s α ranged from .73 to .96 for each culture. Confirmatory factor analyses (CFAs) showed that items measuring negative face threat and positive face threat of H and S were unidimensional for both cultures. The majority of the major CFA fit indexes yielded .90 for all measures. For the reliabilities, fit indexes, and correlations among the variables, see Table 2.

Table 2.

Reliabilities, CFA Results for Unidimensionality, and Correlations (Study 2)

			Reliabilities and Correlations
Cultures			Hearer’s Negative Face Threat		Hearer’s Positive Face Threat		Speaker’s Negative Face Threat		Speaker’s Positive Face Threat
Japanese
Hearer’s negative face threat			(.80)
Hearer’s positive face threat			.71**		(.87)
Speaker’s negative face threat			.47**		.49**		(.75)
Speaker’s positive face threat			.62**		.67**		.52**		(.93)
Americans
Hearer’s negative face threat			(.83)
Hearer’s positive face threat			.69**		(.83)
Speaker’s negative face threat			.62**		.55**		(.85)
Speaker’s positive face threat			.71**		.70**		.68**		(.90)

CFA Results
	χ²	df	χ²/df	$χ_{difference}^{2}$	df _difference	GFI	CFI	RMSEA	IFI	TLI

Japanese
Baseline (one-factor)	858.98	152	5.65**	—	—	.72	.78	.12	.78	.75
Four-factor	353.14	146	2.42**	505.84**	6	.89	.94	.07	.94	.93
Americans
Baseline (one-factor)	880.94	152	5.80**	—	—	.73	.81	.12	.81	.78
Four-factor	376.41	146	2.58**	504.53**	6	.88	.94	.07	.94	.93

Note: GFI = goodness of fit index; CFI = comparative fit index; RMSEA = root mean square error of approximation; IFI = incremental fit index; TLI = Tucker–Lewis index. Reliabilities are reported in parentheses on the diagonal.

p < .05. **p < .01.

Results

A two-way MANOVA was conducted to determine the effect of culture and speech acts on the four dependent variables because the dependent variables were significantly correlated with each other. Their correlations ranged from .47 to .71 for each culture. MANOVA yielded a significant effect for culture, F(4, 623) = 8.61, p < .001, Wilks’s Λ = .95, and a significant interaction effect for culture × speech acts, F(12, 1649) = 1.96, p < .05, Wilks’s Λ = .96, but no effect for speech acts, F(12, 1649) = 1.47, p = .13, Wilks’s Λ = .97. MANOVA is usually followed by ANOVAs to probe which of the dependent variables were associated with significant effects. Accordingly, separate ANOVAs were conducted as follow-up tests.

H’s negative face threat

A 2 (culture) × 4 (speech act) ANOVA showed a main effect for culture, F(1, 626) = 4.65, p < .05, η² = .01. There were no main effect for speech act, F(3, 626) = 1.59, p = .19, η² = .01, and no interaction between culture and speech act, F(3, 626) = 1.86, p = .14, η² = .01. Thus, the data are not consistent with Hypothesis 2. Messages with speech acts were no more effective in reducing threats to H’s negative face than messages without any speech acts.

H’s positive face threat

A 2 (culture) × 4 (speech act) ANOVA showed no significant main effect for speech act, F(3, 626) = 1.58, p = .19, η² = .01, and no significant main effect for culture, F(1, 626) = 1.11, p = .29, η² = .00. The interaction between culture and speech act was significant, F(3, 626) = 2.79, p < .05, η² = .01 (see Table 3).Post hoc comparisons using Tukey’s HSD test at p < .05 showed that for Japanese, the message with apology (M = 1.78, SD = 0.54) was perceived by Ss to threaten H’s positive face significantly less than the message with neither an apology nor thanks (M = 2.12, SD = 0.75), whereas there was no difference for Americans.

Table 3.

Means and Standard Deviations of Face Threats (Study 2)

Speech Act	None	Apology	Thanks	Both	Culture Mean
Hearer’s negative face threat
Culture
Japan	2.42 (0.73)	2.14 (0.68)	2.18 (0.81)	2.28 (0.60)	2.25_x (0.71)
The United States	2.42 (0.79)	2.41 (0.68)	2.41 (0.69)	2.26 (0.60)	2.37_y (0.69)
Speech act mean	2.42 (0.76)	2.27 (0.69)	2.29 (0.76)	2.27 (0.60)	2.31 (0.71)
Hearer’s positive face threat
Culture
Japan	2.12_a (0.75)	1.78_b(0.54)	1.93 (0.82)	1.95 (0.58)	1.94 (0.65)
The United States	2.02 (0.71)	2.07 (0.63)	1.99 (0.65)	1.91 (0.61)	2.00 (0.69)
Speech act mean	2.07 (0.73)	1.92 (0.61)	1.96 (0.74)	1.93 (0.60)	1.97 (0.67)
Speaker’s negative face threat
Culture
Japan	2.42 (0.77)	2.48 (0.64)	2.41 (0.85)	2.65 (0.70)	2.49 (0.74)
The United States	2.42 (0.88)	2.54 (0.86)	2.37 (0.71)	2.36 (0.81)	2.42 (0.82)
Speech act mean	2.42 (0.83)	2.51 (0.76)	2.39 (0.78)	2.50 (0.77)	2.45 (0.78)
Speaker’s positive face threat
Culture
Japan	2.40 (0.84)	2.16 (0.79)	2.29 (0.97)	2.44_x (0.84)	2.32_x (0.86)
The United States	2.25 (0.78)	2.32 (0.74)	2.11 (0.70)	1.99_y (0.74)	2.17_y (0.75)
Speech act mean	2.32 (0.81)	2.24 (0.77)	2.20 (0.85)	2.21 (0.82)	2.24 (0.81)

Note: Different subscripts (x and y in a column for culture, a and b in a row for speech act in a culture) indicate significant difference based on post hoc comparisons using Tukey’s HSD test at p < .05. Standard deviations are reported in parentheses.

S’s negative face threat

A 2 (culture) × 4 (speech act) ANOVA showed no main effects for speech act, F(3, 626) = 0.97, p = .41, η² = .01, or culture, F(1, 626) = 1.23, p = .27, η² = .00. Moreover, the interaction between culture and speech act was not significant, F(3, 626) = 1.57, p = .20, η² = .01. Hence, regarding Research Question 2a, thanks did not change any threat to S’s negative face.

S’s positive face threat

A 2 (culture) × 4 (speech act) ANOVA showed no main effect for speech act, F(3, 626) = 0.76, p = .52, η² = .00. However, there were a significant main effect for culture, F(1, 626) = 5.81, p < .05, η² = .01, and the significant interaction effect, F(3, 626) =3.96, p < .01, η² = .02 (see Table 3). Post hoc comparisons using Tukey’s HSD test at p < .05 showed that Japanese (M = 2.44, SD = 0.67) perceived that the message with both apology and thanks threatened S’s positive face significantly more than did Americans (M = 1.99, SD = 0.74). Regarding Research Question 1, the message with an apology did not threaten S’s positive face more or less than the message without an apology in either culture. Furthermore, related to Research Question 2b, the message with thanks did not threaten S’s positive threat less than the messages without thanks.

Cultural differences in terms of face threat

Regarding Research Question 3, Japanese from S’s perspective reported that the message with an apology significantly reduced the amount of H’s positive face threat more than the message with neither an apology nor thanks.

Summary

A cultural difference was observed in Study 2. An apology alone reduced H’s positive face threat among Japanese. Other than this, the results showed that neither an apology, nor thanks, nor the use of both resulted in reducing the face threats associated with favor asking.

Study 3

Similar to Study 2, Study 3 was conducted to examine Hypothesis 2 and Research Questions 1, 2a, 2b, and 3, although a few changes were made to the design. Study 3 used the same vignette (asking to switch presentation times) that was used in Studies 1 and 2. However, participants in Study 3 were now instructed to consider themselves the H receiving an email message from their classmate: Ashley in the United States and 伊藤 (Ito) in Japan. In this way, perceptions related to each role (i.e., S and H) could be compared regarding apologies and thanks. Then, the email message in Study 3 was sent to a single person. The email messages in Studies 1 and 2 were sent to all classmates to ask a favor, and Study 2 found that these messages did not cause any serious face threat. If a favor is asked of a single person, that person should perceive more face threat than a person in a group. Therefore, the email message in Study 3 was designed to be sent to each participant as if he/she were the only one to receive it.

Finally, repetitions of apologies and thanks were included in each message in Study 3. In Study 2, the four favor asking messages were only different in whether they included an apology, thanks, or both. An apology and thanks might be a habitual speech act without any genuine regret and gratitude, and/or participants might not be aware of the presence of a single apology and/or thanks in the messages. If apologies and thanks are repeated in a message they are more likely to be noticed, thereby giving them the opportunity to lessen the damage done by favor asking. Indeed Study 1 found that Americans (52.5%) and Japanese (19%) had a tendency to repeat their thanks and apologies, respectively. For these reasons, two apologies and two thanks were used.

Method

Participants

Two hundred American participants (age M = 23.29, SD = 5.46, and 29% men) were recruited from several universities in the United States, and 217 Japanese participants (age M = 21.39, SD = 2.38, and 41.5% men) were recruited from several large universities in Japan. American participants consisted of 80% Caucasians, 5.5% African Americans, and 6% Asian Americans. Japanese participants were all ethnic Japanese.

Design

Study 3 was a survey³ that used a 2 (national culture: the United States and Japan) × 4 (speech act: control, two apologies, two thanks, and both) between subject design. Participants filled out one of the four versions of the questionnaire. Version A included a prototype of an email message¹ as well as slightly modified version of the situation described in Study 2. Versions B, C, and D added two apologies (i.e., “I am terribly sorry for asking this, but” and “Once again, I deeply apologize for this”), two thanks (i.e., “I really appreciate this” and “Thank you so much”), and both two apologies and two thanks (i.e., “I am sorry for asking this, but,” “I really appreciate this,” and “Once again, I deeply apologize for this and thank you in advance”), respectively, in the same message as in Version A.

Measurement

The procedure and measurements were conducted in the same manner as in Study 2. The items can be viewed in the appendix. Cronbach’s α ranged from .70 to .93 for each culture. CFAs showed that items measuring negative face threat and positive face threat of H and S were unidimensional for both cultures. The major CFA fit indexes yielded .90 for all measures. For the reliabilities, fit indexes, and correlations among the variables, see Table 4.

Table 4.

Reliabilities, CFA Results for Unidimensionality, and Correlations (Study 3)

			Reliabilities and Correlations
Cultures			Hearer’s Negative Face Threat		Hearer’s Positive Face Threat		Speaker’s Negative Face Threat		Speaker’s Positive Face Threat
Japanese
Hearer’s negative face threat			(.80)
Hearer’s positive face threat			.57**		(.85)
Speaker’s negative face threat			.33**		.40**		(.70)
Speaker’s positive face threat			.64**		.77**		.27**		(.93)
Americans
Hearer’s negative face threat			(.85)
Hearer’s positive face threat			.64**		(.87)
Speaker’s negative face threat			.39**		.35**		(.83)
Speaker’s positive face threat			.68**		.78**		.33**		(.93)

CFA Results
	χ²	df	χ²/df	$χ_{difference}^{2}$	df _difference	GFI	CFI	RMSEA	IFI	TLI

Japanese
Baseline (one-factor)	663.48	152	4.37**	—	—	.72	.79	.12	.79	.76
Four-factor	340.45	146	2.33**	323.03**	6	.85	.92	.08	.91	.92
Americans
Baseline (one-factor)	961.63	152	6.33**	—	—	.64	.70	.16	.70	.66
Four-factor	429.93	146	2.95**	531.70**	6	.82	.90	.10	.90	.88

p < .05. **p < .01.

Results

A two-way MANOVA was conducted to determine the effect of culture and speech acts on the four dependent variables because the correlations of the dependent variables ranged from .27 to .78. MANOVA yielded a significant effect for culture, F(4, 408) = 4.65, p < .01, Wilks’s Λ = .96, and a significant effect for speech acts, F(12, 1230) = 2.47, p < .01, Wilks’s Λ = .93, but not a significant interaction effect, F(12, 1230) = 0.55, p = .879, Wilks’s Λ = .98. Separate ANOVAs were conducted as follow-up tests.

H’s negative face threat

A 2 (culture) × 4 (speech act) ANOVA showed no main effect for culture, F(1, 409) = 2.42, p = .12, η² = .01, and no interaction effect, F(3, 409) = 0.34, p = .80, η² = .00. However, the main effect for speech act was significant, F(3, 409) = 7.00, p < .001, η² = .05 (see Table 5). Because the main effect for speech act was significant, post hoc tests probed the location of the differences between conditions. Post hoc comparisons using Tukey’s HSD test at p < .05 showed that for Japanese, the message with repeated apologies (M = 2.18, SD = 0.83) threatened H’s negative face significantly less than the messages without any speech acts (M = 2.69, SD = 0.59). In contrast, there was no difference for Americans. Thus, the data are partly consistent with Hypothesis 2.

Table 5.

Means and Standard Deviations of Face Threats (Study 3)

Speech Act	None	Apology	Thanks	Both	Culture Mean
Hearer’s negative face threat
Culture
Japan	2.69_a (0.59)	2.18_b (0.83)	2.24 (0.83)	2.25 (0.81)	2.36 (0.79)
The United States	2.55 (0.79)	2.13 (0.78)	2.18 (0.78)	2.50 (0.88)	2.27 (0.81)
Speech act mean	2.62_a (0.70)	2.27_b (0.81)	2.20_b (0.74)	2.37 (0.85)	2.31 (0.80)
Hearer’s positive face threat
Culture
Japan	2.28_a (0.72)	1.77_b (0.69)	2.01 (0.77)	1.88 (0.74)	1.93 (0.74)
The United States	2.16 (0.79)	1.94 (0.76)	1.89 (0.68)	1.93 (0.75)	1.97 (0.75)
Speech act mean	2.22_a (0.76)	1.83 _b (0.72)	1.93 (0.71)	1.90_b (0.74)	1.95 (0.74)
Speaker’s negative face threat
Culture
Japan	2.59 (0.69)	2.29 (0.78)	2.47 (0.75)	2.49 (0.71)	2.41 (0.75)
The United States	2.60 (0.86)	2.47 (0.88)	2.63 (0.90)	2.55 (0.77)	2.56 (0.85)
Speech act mean	2.59 (0.77)	2.35 (0.82)	2.58 (0.85)	2.52 (0.74)	2.48 (0.80)
Speaker’s positive face threat
Culture
Japan	2.35_a (0.88)	1.83_b (0.82)	2.03 (0.79)	2.12 (1.07)	2.02 (0.90)
The United States	2.41 (0.80)	1.96 (0.79)	1.96 (0.76)	2.13 (0.87)	2.10 (0.82)
Speech Act Mean	2.38_a (0.83)	1.88_b (0.81)	1.98_b (0.77)	2.13 (0.97)	2.06 (0.86)

H’s positive face threat

A 2 (culture) × 4 (speech act) ANOVA showed a significant main effect for speech act, F(3, 409) = 4.97, p < .01, η² = .04 (see Table 5). There were no main effect for culture, F(1, 409) = 0.00, p = .95, η² = .00, and no interaction effect, F(3, 409) = 1.00, p = .39, η² = .01. Post hoc comparisons using Tukey’s HSD test at p < .05 showed that for Japanese, the message with apologies (M = 1.77, SD = 0.69) was perceived by Ss to threaten H’s positive face significantly less than the message with neither an apology nor thanks (M = 2.28, SD = 0.72), whereas there was no difference for Americans.

S’s negative face threat

A 2 (culture) × 4 (speech act) ANOVA showed no main effect for speech act, F(3, 409) = 1.59, p = .19, η² = .01, or culture, F(1, 409) = 1.62, p = .20, η² = .00. Moreover, the interaction effect was not significant, F(3, 409) = 0.27, p = .847, η² = .00. Therefore, regarding Research Question 2, thanks did not change any threat to S’s negative face.

S’s positive face threat

A 2 (culture) × 4 (speech act) ANOVA showed a significant main effect for speech act, F(3, 409) = 6.63, p < .001, η² = .05 (see Table 5). However, there were no main effect for culture, F(1, 409) = 0.12, p = .726, η² = .00, and no interaction effect, F(3, 409) =0.23, p = .872, η² = .00. For Research Question 1, post hoc comparisons using Tukey’s HSD test at p < .05 showed that for Japanese, the message with apologies (M = 1.83, SD = 0.82) was perceived by Hs to threaten S’s positive face significantly less than the message with neither an apology nor thanks (M = 2.35, SD = 0.88), while there was no difference for Americans. Regarding Research Question 2b, accordingly, thanks did not change S’s positive face threat.

Cultural differences in terms of face threat

Regarding Research Question 3, Japanese reported that the favor asking message with repeated apologies significantly reduced the amount of H’s negative and positive face threat and S’s positive face threat more than the favor asking message with neither an apology nor thanks. In contrast, there was no difference for Americans.

Summary

Some cultural differences were found in Study 3. Repeated apologies alone reduced various face threats among Japanese, but not among Americans. Also, repeated thanks or both did not reduce the face threats associated with favor asking for either culture.

General Discussion

The current studies examined the differences between Japanese and Americans with regard to the use of apologies and thanks in favor asking messages, and considered four types of face threats to potentially explain why this would be so. The results of Study 1 showed that Japanese are more likely to include apologies than thanks whereas Americans are more likely to include thanks than apologies in the favor asking message.

Wilson et al. (1998) claimed that potential face threats related to favor asking do not vary across cultures. Indeed, Japanese and Americans perceived threats similarly for all four face threats in the control condition (i.e., favor asking message without any apology or thanks) in both Study 2 and Study 3. Furthermore, expressions of apology in favor asking messages were viewed as attempts to reduce the face threats associated with favor asking in Japan. Japanese as S considered a favor asking message with an apology to threaten H’s positive face less than a message with neither an apology nor thanks in Study 2, and Japanese as H perceived a message with repeated apologies to threaten H’s negative and positive face and S’s positive face less than a message with neither an apology nor thanks. In contrast, for Americans, adding apologies and/or thanks did not change any face threat in favor asking messages. Therefore, for Americans, thanks seem to play a ritualistic role in that Americans do not seem to rely heavily on thanks to accomplish facework when asking favors. Finally, Japanese Ss and Hs tend to focus on the same type of face during favor asking. Their focus is on H’s desire for appreciation, but Japanese Hs’ additional focus is on H’s autonomy and S’s desire for approval. These findings indicate that H and S could perceive different face threats even when the same message is used, and H might perceive a favor asking message with apologies to be less face threatening than S in Japan.

Compared with the Americans, Japanese have been assumed to have a greater concern for other’s face than for self face (Ting-Toomey & Oetzel, 2001). This assumption is evidenced in Study 2 and Study 3. Japanese used apologies in their favor asking messages, which reduced H’s positive face threat in Study 2 and H’s negative and positive face threat in Study 3. The results of the current study are consistent with this assumption, at least with regard to facework associated with apologies and thanks in favor asking situations.

Implications for Cross-Cultural and Intercultural Communication

These findings have several implications. First, the current findings extend Brown and Levinson’s (1987) theory of politeness to a different domain. The theory of politeness is about how individuals produce linguistic politeness in face-to-face interactions (Brown & Levinson, 1987). When sending email to an unknown group of people such as unfamiliar classmates, both Japanese and Americans produce linguistic politeness (i.e., apologies for Japanese and thanks for Americans), so the current study contributes to extend the theory’s applicable domain to include email. Although it has been argued that the absence of face-to-face interaction in computer-mediated communication leads to the use of blatant and hostile language, politeness can be expressed easily in email (Bunz & Campbell, 2004). The results also enhance empirical support for the theory by showing that people across cultures act to mitigate the threats by doing facework if S decides to perform the FTA on record via email.

Brown and Levinson (1987) suggest that such redressive actions are either positive politeness or negative politeness. Positive politeness is oriented toward the positive face of H. Some examples are exaggerating to get sympathy from H, seeking agreements, giving reasons, assuming reciprocity, and so on. Negative politeness is primarily focused on satisfying H’s negative face. Apologizing, being pessimistic, and hedging are some common strategies Brown and Levinson mention. Although apology is introduced as a negative politeness strategy theorized to reduce H’s negative face threat, it was used as a positive politeness strategy as well in Study 2 and Study 3. Brown and Levinson’s model assumes that strategies are mutually exclusive and serve one strategy and only one strategy: either positive or negative politeness. The current findings show that, in Japan, the same speech act can serve positive and negative politeness goals simultaneously (Dillard, Wilson, Tusing, & Kinney, 1997; H. E. Lee & Park, 2011). Furthermore, although gratitude statements are popularly used at least in the United States, politeness theory is silent on expressing gratitude in its positive and negative politeness strategies. H. E. Lee and Park (2011) found that thanks serves as either a positive or negative strategy in a more imposing favor asking situation. However, the current results revealed that thanks did not serve any politeness strategy in a favor asking situation. To identify the boundary conditions of thanks as a politeness strategy, various favor asking situations should be examined.

Second, only statements of apology and gratitude differentiated four conditions in Study 2 and Study 3. The subtlety of the instigating messages is indeed the most minimal manipulation that still produces maximal variance. One way to demonstrate important effects is to show that even the most minimal manipulation of the independent variable accounts for some variance in the dependent variable (Lance & Vandenberg, 2009; Prentice & Miller, 1992). The current study shows how such a minor manipulation of the independent variable still has an effect even under the most inauspicious circumstances.

Third, with regard to favor asking, these findings showed a clear American preference for thanks over apology and a Japanese preference for apology over thanks. When asking a favor of Japanese, Americans may find relationship value in using an apology rather than thank you, because including only thank you in favor asking violates the Japanese norm. Similarly, a Japanese individual who apologizes when asking a favor of an American is perceived to have apologized unnecessarily, which in turn makes H uncomfortable (e.g., Olshtain, 1989) and triggers perceptions that the Japanese individual is somehow unusual or impolite. This finding is of immediate practical use to Japanese teachers of English who teach American politeness norms to their students through English speech acts as well as to Americans teachers of Japanese who teach Japanese politeness norms. Japanese may criticize Americans for expressing gratitude too often, so that expressions of thanks lose their potency. Also, if Americans proffer thanks in favor asking, it may be interpreted by the Japanese hearer as pressure to grant the favor, making Americans appear demanding and impolite.

Finally, the study provided measures of four face concerns with acceptable reliabilities and unidimensional structures. Although several studies have attempted to measure face threats in the past, there have been theoretical and methodological drawbacks with many of these studies. When two people interact with each other, four faces (S’s positive and negative face, H’s positive and negative face) are potentially concerned. Although these four faces are theoretically distinct, some studies have not differentiated between them (e.g., Kellermann, 2004; Oetzel et al., 2001; Ting-Toomey & Oetzel, 2001). Moreover, even when four faces were measured, some studies have reported no factor analyses (e.g., Cai & Wilson, 2000; Wilson et al., 1998), or principle component analyses (PCA) for separate unidimensionality (e.g., Wilson, Kunkel, Robson, Olufowote, & Soliz, 2009) or multidimensionality (e.g., Johnson, Roloff, & Riffee, 2004). PCA is a procedure for data reduction when a researcher does not want to include all the original measures in analyses and no theoretical knowledge can be incorporated (Park, Dailey, & Lemus, 2002). Thus, CFA for multidimensionality should be examined to evaluate how an underlying construct influences the responses on a number of measured variables, and to validate the factors that are responsible for a set of observed responses (Kline, 2005). Park and Guan (2006) measured four faces (as well as mutual face) and conducted CFA for multidimensionality, but self-negative face and other positive face were dropped from their analyses due to low reliability.

The current study theoretically and empirically differentiated four different face concerns and successfully measured them with six to seven items each. Although the current study showed consistent effects of an apology on face threats in Japan only, more valid face research can be conducted with these measures with various cultures in the future. It is possible that an apology and thanks are habitual speech acts without any real meaning. If emotions such as indebtedness and gratitude are measured, this possibility can be tested. A final possibility is that apologies and/or thanks would be effective at receiving a favor regardless of face concerns. Which explanation is correct is the province of future studies.

Limitations and Future Directions

The major limitation of the current research was that the given situation did not cause any serious face threat. The face threats in the control group (i.e., favor asking message without any apology or thanks) ranged from 2.02 to 2.42 on a 5-point scale in Study 2 and from 2.28 to 2.69 in Study 3. Because the favor asking situation was understandable (i.e., not terribly imposing), it did not trigger serious face threats. Thus, adding an apology and/or thanks might not be necessary. Indeed, 31 (41.9%) Japanese participants failed to include an apology or an expression of gratitude as mitigating supporting acts in their favor asking messages in Study 1. Brown and Levinson (1987) claim that the seriousness of an FTA involves the social distance(D) of S and H, the relative power (P) of S and H and the culturally defined ranking (R) of the degree of impositions. Because the current study focused on favor asking between unknown classmates, D, P and R are fixed.

Wilson et al. (1998) found that different Rs such as asking favors, giving advice, and enforcing obligations create distinct combinations of face threats. Also, different P and R would change facework norms. Asking a favor of a close friend versus a stranger and of a supervisor versus a subordinate would trigger different degrees of face threats, thus they might require different politeness strategies and speech acts. Roloff et al. (1988) found that American college students offered fewer apologies when asking a favor of a friend than an acquaintance or stranger, but Park and Guan (2009) found people had stronger intentions to apologize for an offending act to a stranger than to a friend in the United States and China. Considering that the degree of politeness generated in Japanese can be negotiated by D (Haugh, 2007), various Ds have more important implications in Japanese contexts. We expect that more imposing favor asking situations (i.e., higher R) would change a ritualistic function of thanks to a more meaningful facework function. H. E. Lee and Park (2011) found that Americans considered repeated thanks to reduce H’s negative and positive face threat in a more imposing favor request. To strengthen the external validity of the findings, future research should be conducted that experimentally varies D, P, and R.

Finally, this study measured perceived face threats but not face needs. Reduction of face threats is not necessarily the same as face enhancement. Including an apology or thanks did not reduce the face threats associated with favor asking, but it is possible that they might have enhanced face need. Thus, future studies should include face enhancement items, for example, “Asking a favor in the message with an apology could make me look better to my classmates.”

Study 1 attempted to measure actual behaviors, but those behaviors were only for a hypothetical situation. Participants wrote an email message while imagining themselves in the situation of having to ask a favor. Although there is a strong relationship between behavioral intention and actual behavior (cf. Sheeran, 2002; Webb & Sheeran, 2006), caution should be exercised when interpreting hypothetical situation findings. Study 2 used only prototypical statements of apology and gratitude. How apologizing or thanking is done leads to different effects on face. For example, including accounts in an apology such as “I am sorry, but . . .” is not considered an apology to Japanese (e.g., Kotani, 2008; Sugimoto, 1999). Also, “sorry” is less polite than “I’m really terribly sorry” (Turnbull & Saxton, 1997). It is important to delicately balance the cost for S and the benefit to H in making an apology or thanking. Different expressions beyond the prototypical statements used in the current study will increase understanding of the roles of apology and gratitude. Therefore, various situations and designs should be employed in future studies to determine the extent to which the current findings replicate.

In conclusion, this research systemically examined cross-cultural differences in the use of apology and thanks in favor asking messages. Three different studies found that Americans preferred and included thanks in favor asking messages while Japanese preferred and used apologies. Additionally, the research showed that apologies in favor asking messages reduced different face threats in Japan whereas thanks in favor asking messages did not reduce any face threats in Japan or the United States. Future research should empirically determine the extent to which the contexts and forms of communication affect the face threats associated with favor asking situations.

Footnotes

Appendix

Acknowledgements

Special thanks to Howard Giles and anonymous reviewers for their helpful feedback on an earlier version of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Notes

Bios

Hye Eun Lee (PhD, Michigan State University) is an assistant professor in the Department of Communicology at University of Hawaii at Manoa. Her research areas are cross-cultural differences in speech acts and customers’ perceptions, and communication network approaches to organizational attitudes and behaviors at the workplace.

Hee Sun Park (PhD, University of California, Santa Barbara) is an associate professor in the Department of Communication at Michigan State University. Her current research projects examine cross-cultural differences in norms and interaction patterns, multilevel aspects of group and organizational communication, and health-related social influence processes and outcomes.

Tatsuya Imai (MA, Michigan State University) is a doctoral student in the Communication Studies Department at the University of Texas at Austin. His research interest is interpersonal communication in a variety of contexts such as romantic relationship, mental health, and family.

Daniel Dolan (PhD, University of Washington) is a professor in the Graduate School of Accountancy at Waseda University in Japan. His research includes cultural functions of communication in Japan, membering and commitment functions of personal experience storytelling, communication and public policy, and religious communication theory.

References

Barnlund

D. C.

Yoshioka

(1990). Apologies: Japanese and American styles. International Journal of Intercultural Relations, 14, 193-206.

Becker

J. A.

(1982). Children’s strategic use of requests to mark and manipulate social status. In Kuczac

I. S. A.

(Ed.), Language development (pp. 1-35). Hillsdale, NJ: Erlbaum.

Brown

Levinson

S. C.

(1987). Politeness: Some universals in language usages. Cambridge, England: Cambridge University Press.

Bunz

Campbell

(2004). Politeness accommodation in electronic mail. Communication Research Reports, 21, 11-25.

Cai

D. A.

Wilson

S. R.

(2000). Identity implications of influence goals: A cross-cultural comparison of interaction goals and facework. Communication Studies, 51, 307-328.

Coulmas

(1981). Poison to your soul: Thanks and apologies contrastively viewed. In Coulmans

(Ed.), Conversational routine (pp. 69-91). The Hague, Netherlands: Mouton.

Craig

R. T.

Tracy

Spisak

(1986). The discourse of requests: Assessment of a politeness approach. Human Communication Research, 12, 437-468.

Cupach

W. R.

Metts

(1994). Facework. Thousand Oaks, CA: Sage.

Dillard

J. P.

Wilson

S. R.

Tusing

K. J.

Kinney

T. A.

(1997). Politeness judgments in personal relationships. Journal of Language and Social Psychology, 16, 297-325.

10.

Elliot

A. J.

Chirkov

V. I.

Kim

Sheldon

K. M.

(2001). A cross cultural analysis of avoidance (relative to approach) personal goals. Psychological Science, 12, 505-510.

11.

Goei

Roberto

Meyer

Carlyle

(2007). The effects of favor and apology on compliance. Communication Research, 34, 575-595.

12.

Goffman

(1967). Interactional ritual: Essays on face-to-face behavior. New York, NY: Pantheon Books.

13.

Goldschmidt

M. M.

(1998). Do me a favor: A descriptive analysis of favor asking sequences in American English. Journal of Pragmatics, 29, 129-153.

14.

Goldsmith

(1992). Managing conflicting goals in supportive interaction: An integrative theoretical framework. Communication Research, 19, 264-286.

15.

Haugh

(2007). Emic conceptualizations of (im)politeness and face in Japanese: Implications for the discursive negotiation of second language learner identities. Journal of Pragmatics, 39, 657-680.

16.

Haugh

(2009). Face and interaction. In Bargiela-Chiappini

Haugh

(Eds.), Face, communication and social interaction (pp. 1-30). London, England: Equinox.

17.

Heine

Lehman

(1999). Culture, self-discrepancies, and self-satisfaction. Society for Personality and Social Psychology, 25, 915-925.

18.

Heine

S. J.

Lehman

D. R.

Markus

H. R.

Kitayama

(1999). Is there a universal need for positive self-regard? Psychological Review, 106, 766-794.

19.

Higgins

E. T.

(1998). Promotion and prevention: Regulatory focus as a motivational principle. In Zanna

M. P.

(Ed.), Advances in experimental social psychology (Vol. 30, pp. 1-46). New York, NY: Academic Press.

20.

Holtgraves

(2002). Interpersonal underpinnings of talk: Face management and politeness. In Language as social action: Social psychology and language use (pp. 37-63). Mahwah, NJ: Erlbaum.

21.

Ide

(1998). “Sorry for your kindness”: Japanese interactional ritual in public discourse. Journal of Pragmatics, 29, 509-529.

22.

Johnson

D. I.

Roloff

M. E.

Riffee

M. A.

(2004). Politeness theory and refusals of requests: Face threat as a function of expressed obstacles. Communication Studies, 55, 227-238.

23.

Kellermann

(2004). A goal-directed approach to gaining compliance. Communication Research, 31, 397-445.

24.

Kim

(2003). Korean cultural codes and communication. International Area Review, 6, 93-114.

25.

Kline

R. B.

(2005). Principles and practice of structural equation modeling (2nd ed.). New York, NY: Guilford.

26.

Kotani

(2008). Reinforced codes and boundaries: Japanese speakers’ remedial episode avoidance in problematic situations with “Americans.” Research on Language and Social Interaction, 41, 339-363.

27.

Kramer-Moore

Moore

(2003). Pardon me for breathing: Seven types of apology. Research Library Core, 60, 160-169.

28.

Kristof

N. D.

(1995, June 12). A big exception for a nation of apologizers. The New York Times, pp. 1-2.

29.

Kumatoridani

(1999). Alternation and co-occurrence in Japanese thanks. Journal of Pragmatics, 31, 623-642.

30.

Kwon

(2000). Expressions of gratitude in Korean and English: A comparative analysis (Unpublished dissertation). Hankuk University of Foreign Studies, Seoul, South Korea.

31.

Lance

C. E.

Vandenberg

R. J.

(2009). Statistical and methodological myths and urban legends: Doctrine, verity, and fable in organizational and social research. New York, NY: Routledge.

32.

Lee

Aaker

Gardner

(2000). The pleasures and pains of distinct self-construals: The role of interdependence in regulatory focus. Journal of Personality and Social Psychology, 78, 1122-1134.

33.

Lee

H. E.

Park

H. S.

(2011). Why Koreans are more likely to favor “Apology,” while Americans are more likely to favor “Thank You.” Human Communication Research, 37, 125-146.

34.

Leech

(1983). Principles of pragmatics. London, England: Longman.

35.

Lerner

(1996). Finding “face” in the preference structures of talk-in-interaction. Social Psychology Quarterly, 59, 303-321.

36.

Ohashi

(2003). Japanese culture specific face a nd politeness orientation: A pragmatic investigation of yoroshiku onegaishimasu. Multilingua, 22, 257-274.

37.

Oetzel

J. G.

Ting-Toomey

Masumoto

Yokochi

Pan

Takai

Wilcox

(2001). Face and facework in conflict: A cross-cultural comparison of China, Germany, Japan, and the United States. Communication Monographs, 68, 235-258.

38.

Olshtain

(1989). Apologies across languages. In Blum-Kulka

House

Kasper

(Eds.), Cross-cultural pragmatics: Requests and apologies (pp. 155-173). Norwood, NJ: Ablex.

39.

Oyserman

Coon

Kemmelmeier

(2002). Rethinking individualism and collectivism: Evaluation of theoretical assumptions and meta-analyses. Psychological Bulletin, 128, 3-73.

40.

Park

H. S.

Dailey

Lemus

(2002). The use of exploratory factor analysis and principal components analysis in communication research. Human Communication Research, 28, 562-577.

41.

Park

H. S.

Guan

(2006). The effects of national culture and face concerns on intention to apologize: A comparison of the USA and China. Journal of Intercultural Communication Research, 35, 183-204.

42.

Park

H. S.

Guan

(2009). Culture, positive and negative face threats, and apology intentions. Journal of Language and Social Psychology, 28, 244-262.

43.

Prentice

D. A.

Miller

D. T.

(1992). When small effects are impressive. Psychological Bulletin, 112, 160-164.

44.

Roloff

M. E.

Janiszewski

C. A.

McGrath

M. A.

Burns

C. S.

Manrai

L. A.

(1988). Acquiring resources from intimates: When obligation substitutes for persuasion. Human Communication Research, 14, 364-396.

45.

Searle

J. R.

(1976). A classification of illocutionary acts. Language in Society, 5, 1-23.

46.

Sheeran

(2002). Intention-behavior relations: A conceptual and empirical review. In Stroebe

Hewstone

(Eds.), European review of social psychology (Vol. 12, pp. 1-34). Chichester, England: Wiley.

47.

Spencer-Oatey

(2007). Theories of identity and the analysis of face. Journal of Pragmatics, 29, 639-656.

48.

Sugimoto

(1997). A Japan-U.S. comparison of apology styles. Communication Research, 24, 349-369.

49.

Sugimoto

(1999). “Sorry we apologize so much”: Linguistic factors affecting Japanese and U.S. American styles of apology. Intercultural Communication Studies, 13, 71-78.

50.

Tanaka

Spencer-Oatey

Cray

(2008). Apologies in Japanese and English. In Spencer-Oatey

(Ed.), Culturally speaking: Culture, communication and politeness theory (pp. 73-89). London, England: Continuum.

51.

Ting-Toomey

Oetzel

J. G.

(2001). Managing intercultural conflict effectively. Thousand Oaks, CA: Sage.

52.

Turnbull

Saxton

K. L.

(1997). Modal expressions as facework in refusals to comply with requests: I think I should say “no” right now. Journal of Pragmatics, 27, 145-181.

53.

Wagatsuma

Rosett

(1986). The implications of apology: Law and culture in Japan and the United States. Law & Society Review, 20, 461-497.

54.

Webb

T. L.

Sheeran

(2006). Does changing behavioral intentions engender behavior change? A meta-analysis of the experimental evidence. Psychological Bulletin, 132, 249-268.

55.

Wilson

S. R.

(1990). Development and test of a cognitive rules model of interaction goals. Communication Monographs, 57, 81-103.

56.

Wilson

S. R.

Aleman

C. G

Leatham

G. B.

(1998). Identity implications of influence goals: A revised analysis of face-threatening acts and application to seeking compliance with same-sex friends. Human Communication Research, 25, 64-96.

57.

Wilson

S. R.

Kunkel

A. D.

Robson

S. J.

Olufowote

J. O.

Soliz

(2009). Identity implications of relationship (re)definition goals. Journal of Language and Social Psychology, 28, 32-61.