Face and Politeness Phenomena in the Changing China

This study aims to investigate the Chinese perceptions about the traditional Confucius concepts mianzi/lian ‘face’ in the context of the world that is changing. In particular, the study focuses on how such perceptions are reflected in their interpersonal communication. Data was collected through direct semi-structured interviews with the respondents, guided by a set of questions regarding the situations they might hypothetically find themselves. Shanghai was chosen as the site for this study because, among other places in China, it has undergone considerable changes. 100 Shanghainese were involved in the study; 50 were city dwellers and 50 villagers, representing their various social backgrounds. The data were analysed by using the Chinese cultural notions underlying the concepts of face: relational, communal, hierarchical, and moral. Analyses show that the respondents often avoid conflicts with their interlocutors, even if they were put in very unfortunate circumstances. These face-favouring acts are indicative of their closed observance of the norms and hence the concepts of face, by which they could gain, maintain, and enhance their own face. In return, they would be regarded as members of the society with polite behaviours and with other good moral characters. The study concludes that although China is changing, the conceptions and practices of the traditional concepts mianzi/lian “face” among Chinese have remained constant. This is because into these concepts are attached the most basic concepts of humanity; the absence of the concepts of face in the mind of human beings can mean the loss of humanity as a whole. 


Introduction
Chinese have often been believed as a community living with a conflict-avoidance culture. Their communication behaviours seem to have been inspired and hence guided by the common belief among participants to save each other's dignity and the sacred social attribute known as mianzi or lian "face". This is a concept derived from the teachings of the most prominent and respected Chinese philosopher Confucius. Cheng (1986: 337) believes that "Confucianism, with its theory and practice, no doubt, is the unequivocal ideological background and foundation of the concept of face and face-work in the Chinese language". This concept of face requires that all parties involved in a communication transaction be obliged to save each other's face as the "positive social value" they will effectively claim for themselves and this is "an image of self delineated in terms of social attributes" (Goffman, 1967: 5). As a social attribute, face is gained on loan from society. This common belief harmonizes their interactions, because facework among Chinese is "conceptualised as a typical Chinese conflict preventive mechanism and a primary means to cultivate harmonious human relations in Chinese social life" (Jia, 1997). Therefore, to understand the philosophical values of face would be of crucial necessity for every member of the society by which they can maintain the harmony of their life.
The values of the concept of face in Chinese culture seem to have been extracted from the most basic value of jen, a concept related to 'humanness' and its associated meanings such as 'humanity', 'benevolence', 'goodness', and 'virtue'. Contained in this jen are all warm human feelings that qualify and constitute an ideal person, who will always be considerate about others. In other words, understanding of and then practicing jen implies an acceptance of the principle of shu 'reciprocity', in which a person will feel and as if experiencing to be the other person by putting him/herself in the other's shoes, so that his/her empathy toward others will spring accordingly (cf. Aziz 2000: 303) According to Confucianism, as noted by Cheng (1986), a perfect ideal human being, reached through selfcultivation, can only be accomplished when the following five relationships are completely understood and then achieved: 1) the relation of closeness (like the father and son); 2) the relation of righteousness (as found in the emperor-subject relation); 3) the relation of distinction (as in husband and wife relation); 4) the relation of hierarchy (as evident in elder-younger relation); and 5) the relation of faithfulness (as in friendfriend relation). Self-cultivation itself is determined by the norms and values of xiao (filial piety), di (brotherliness), li (propriety), and de (moral integrity). It is into de that the concepts of lian and mianzi are attached. Jia (1997) identifies four characteristics attached to the concept of face in Chinese: relational, communal/social, hierarchical, and moral. Relational concept of face is fixedly understood by Chinese as the principal means and mechanism that governs and prescribes all their social conducts so that it guarantees the harmony of their human relations. This relational conception is considerably contradictory with that of Western understanding of face concept which is of highly rational model in nature (Brown&Levinson, 1987), transactional (Scollon&Scollon, 1994) and emphasises individualism on its prime perspective (Chang&Holt, 1994;Gu 1990;Matsumoto, 1988). Ho (1976: 882) further noted that "the Western mentality, deeply ingrained with the values of individuality, is not one which is favourably disposed to the idea of face, for face is never a purely individual thing" (cf. Mao, 1994). Face, according to Jia, is "both the goal and the means for strengthening and expressing harmonisation of human relationships among men in society" (1997) and the central emphasis is on "the human relationship instead of impression management" (Chang&Holt, 1994: 127).
Face is also said to have a communal/social value. This characteristic is rooted on the idea that face is a "public censure" (Hu, 1944:47), the loss of which will severely negatively affect the status of the loser in the community. Therefore, when someone has a feeling of fear of losing face, that strongly indicates that he/she is aware of "the force of social sanction" (1944:50). To keep face means that the normal functioning of the community will prevail, and "the member of who retains it is accepted as a full member of the community" (Jia, 1997).
The Chinese concept of face is grounded upon a hierarchical perspective (Chang&Holt, 1994;Scollon&Scollon, 1994) into which age and blood bonds, relational hierarchy within the family and the hierarchical nature of the family are attributed. This is for example evident in the naming practices that prevail in the Chinese communities (see Lee-Wong, 2000). According to Scollon&Scollon (1994), the concept of hierarchy is obviously noted in the concept of face, and not taking this characteristic into account might lead to a misunderstanding of the concept.
Morality is inherently attached to the concept of face in Chinese culture. In Hu's observation, face finds its place as the sign of "respect of the group [of people in the society] for a man with a good moral reputation" and the loss of it could mean "a condemnation by the group for immoral and socially disagreeable behaviour" (1944: 45-46). Seen from this perspective, the loss of face can be equated with the loss of morality of the loser.
In interpersonal communication, the principles of jen bind all participants to believe that their interest may conflict with others, and it therefore has to be compromised. When taken up appropriately according to the principles, both parties will be felicitous, i.e. 'feeling good and satisfied' because their interest is respected and attended to by the other. In most cases, though not always, this feeling is achieved through politeness or limao. Into this are included a set ofsometime not-negotiable-guidelines, sometimes notnegotiable, for people to follow.
As a concept, limao might be perceived reasonably similar in the minds of the Chinese, although its realisations might vary because these will depend on individuals' understanding, which might have been influenced by some other external factors such as age, education, life experience, intensity of interactions with other members, social status, and so on. The present study was aimed to investigate the influence of those external factors on the realisations of limao in various speech acts. In particular, this study was oriented to portray and explain the on-going changes experienced by the Chinese in the way they view mianzi 'face' as a traditional concept in relation to limao 'politeness' in the new order of the world that is changing. If we believe that the development of information technology has made the world even become smaller and borderless, then we will have to admit that there will no single element of the world remain static. Needless to say, this will also affect the values and norms adopted in our life. This natural process will therefore have to be viewed not as a threat to the traditional values but rather as a complementary that necessitates it to happen.

The Values of Face Concept in Chinese Culture
The fact that Chinese concept of self is different to that of Western has brought about different face orientation in the community. In Chinese society, which is a grouporiented or communal society, interdependence among Chinese is inevitable. Pan (1995), for example believes that as pointed by some anthropologists (e.g., Hsu, 1981;Zhao & Gao, 1990), the Chinese concept of self includes oneself, one's immediate environment such as family, and a larger environment such as those who are related to him or her either by blood, by profession, or by locality. In this sense, a Chinese self is larger than an individual-centered Western self. Therefore, the desire to be independent and be unimpeded in one's actions (negative face) is almost alien to Chinese.
Extending from the fact above, Chinese recognizes two concepts of face which are never independent to society. Lian, which shows the respect of members of society for "a man with a good moral reputation", embodies "the confidence of society in the integrity of ego's moral character," and "it is both a social sanction for enforcing moral standards and an internalized sanction" (Hu, 1944:45). The other kind of face, mianzi, stands for prestige or reputation, which is either achieved through getting on life or ascribed (or even imagined) by other members of one's own community (Hu 1944:45).
Lian and mianzi have become real concern for Chinese since long time ago. Being so, many terms are derived. A person is said to lose lian when he/she performs such immoral or socially disagreeable behaviour as breaking promise, telling lies for one's own profit, exposing crimes, and so on, resulting in criticism from society. The example is, echoing Hu (1944), a girl student who was pregnant hanged herself in her intimate's dormitory after getting desperate because he denied his promise to marry her. The girl's suicide makes him lose lian. Another example is when a new instructor in certain universities was unable to answer students' questions at the first sight, he lost lian. As much of the activity of Chinese life is operated on the basis of trust, losing lian is a real dread for Chinese. The higher the social standing of a person the more dignity he has to maintain, and the more vulnerable this lian becomes (Hu 1944:47). Meanwhile, "to lose mianzi" is to suffer a loss of one's reputation or prestige because of a certain failure or misfortune. "To lose lian" is a far more serious act than "to lose mianzi". It should be noted, however, that once lian is lost, it will be hard to maintain mianzi. In addition, the sudden loss of mianzi might be too much a shock for personality.
The expressions "not to want lian" (being indifferent on what society thinks of his bad character) and "have no lian" are stronger than "to lose lian". The expressions "to have no lian" and "to have no mianzi" are different in meaning. The former means the most severe insult to people's moral character while the latter signifies merely the failure of ego to achieve a reputation through success in life. Person with a thin skin on the lian, i.e. highly sensitive to public opinion, is more favoured by the society than those with thick skin because he/she conforms more readily to society.
It is believed that Chinese concept of face is distinctive from that of Western conception as found in Brown and Levinson's because the former underlying concept of self is different to the latter. In Chinese, moreover, face is not individual but public property; face is not selfimage but public image. It is Goffman (1967) who proposed this construct. He sees face as being "located in the flow of events" and "on loan from society". Also for Goffman, face is a 'public property' that is only assigned to individuals contingent upon their interactional behaviour (Mao 1994:454). This is similar to that of Japanese concept of face in which the underlying interactional focus is centered not upon individualism but upon group identity (Matsumoto 1988: Ide 1989. Here, Mao (1994) gives clear comparison between the two concepts of face.
The importance of face in creating harmony in communication is clearly shown in the research done by Yang (1945), who found that in a Chinese village life, hurting other's "face" turns to be the source of conflict. Generally, this is due to the villagers' high need of face. In other words, conflict is a face war. Some factors involved in losing or gaining face are attributable to the equal or unequal social status between the persons involved. For example, one of two equally popular professors is refused by the other in some request; the former will lose face. Meanwhile, students will not suffer loss of face if they are similarly treated by a professor. The presence of a witness or a third person may increase the risk of losing face before other people. However, the more intimate the person with the third person the lesser the risk of losing face. In relation to the social prestige people acquire, only middle-aged people are serious about losing or gaining face. Lastly, a person's sensibility also contributes to the risk of losing face.
It should be noted that the current conception of politeness in Modern China is derived and largely developed from its concept of face limao. Limao, which morphemically means 'polite appearance', turns out to become the code of conduct for Chinese in their attempt to establish and maintain harmonious and successful communication. To be polite in Chinese discourse is, in many respects, to know how to attend to each other's lian and mianzi and to enact speech acts appropriate to and worthy to such an image (Mao 1994:463). Like Japanese conception of politeness (wakimae), the Chinese concept of limao requires speakers to express deference by humbling themselves and placing themselves in a lower position (Matsumoto 1988). Gu (1990: 239) held that within the Chinese conception of limào comprise four basic notions: 'respectfulness, modesty, attitudinal warmth, and refinement'. Respectfulness refers to the observance of the speaker to positively appreciate or show admiration for the hearer's face, social status, and so on. Modesty is seen as another way of saying 'self-denigration' which may derive from the speaker's timidity. Attitudinal warmth requires the speaker to demonstrate kindness, consideration, and hospitality to others and refinement obligates the speaker to behave to another within certain standards.
In interaction, the four notions above may be implicitly expressed in the forms of maxims as analyzed by Gu, who followed Leech's account of politeness principle. According to Gu (1990), the social norm approach proposed by Leech is more appropriate since its emphasis to the normative aspect of politeness is comparable to Chinese concept of face. Besides, analyzing politeness in terms of maxims is believed to be more appropriate considering the fact that the Chinese conception of politeness is to some extent moralized. Although Gu claims that 'failure to observe politeness will incur social sanctions' (Gu 1990: 242), this claim cannot be applied only to the Chinese culture in particular because this is also true and does exist in other cultures but varied in terms of conditions and sanctions imposed.
The first maxim from Gu, i.e. the self-denigration maxim, contains two sub-maxims, "denigrate self" and "elevate other". If both are breached, people will be perceived respectively as being impolite or rude and being arrogant or boasting. The phenomenon is clearly demonstrated in the introducing-each-other interaction in which Chinese people take the first chance to elevate other, i.e. ask for H's name. The second, the address maxim, deals with addressing the interlocutor with an appropriate address term. A failure to use an appropriate address term is a sign of rudeness. Thus, the choice of it should consider such factors as kin or non-kin, politically superior or inferior, professionally prestigious or non-prestigious, interpersonally familiar or unfamiliar, male or female, old or young, on a formal or informal occasion, family members or relatives, and in public or at home. Unlike English address system, Chinese proper name is arranged in the order of surname + (middle name) + given name. The Chinese surname is non-kin public address term and can be used alone by people outside the family, but the middle + given name and the given name are kin familial address term. Some Chinese kinship terms, for example yeye (grandpa), nainai (grandma), shushu (uncle), and a'yi (aunt), can be used to address people who have no familial relationship whatever with the addresser. Occupational titles, as well, can be used as address terms in Chinese. In unequal encounters, it is usually the inferior who initiates talk exchanges by addressing the superior first. Nevertheless, it should be noted that the implementation of the maxims explained above is, to a certain extent, influenced by some social factors like age differences, setting, level of intimacy and familiarity, and relative power. Scollon&Scollon (1995), as quoted by Pan (1995:466), points out that Chinese politeness behaviour is hierarchical in nature; that is, the strategies used by the interlocutors are not symmetrical. Besides, Chinese people tend to be situation-centered in that emphasis is placed upon an individual's appropriate place and behaviour among his fellowman (Hsu 1981in Pan 1995. Particularly in official settings, age differences are not always significant in performing polite behaviour. It is the institutionalized power of rank that mainly determines polite behaviour. Although subordinates are older in terms of age than deputy secretary (superior), the institutional constraint rank hierarchy gives the superior power to deliver an order directly without using hedges (e.g. is it all right?). Here, speech acts that are performed in workplace are mostly directives. The use of polite hedges in directives is found only in utterances made by speakers when talking to their superiors or equals. As the superior is in a position to make the choice of politeness strategies, he/she can also perform solidarity by attending positive face of his/her subordinates; that is by using deferential address form in a joking manner (Pan 1995:468). That superiors treat the subordinates in such a manner reflects what Goffman (1959) called as demeanour, i.e. the speaker presents him or herself appropriately according to his or her role in the situation. In the other side, subordinates' acts to superiors show a sign of deference, i.e. the appreciation carried out by an act, showing regard for the recipient. To the extent that the interaction happens in family setting, age and gender factors seem to lead the choice of politeness strategies. With regard to the conception of politeness in China, it appears that politeness is culture specific and its realization greatly depends on the situations and participants involved in the interaction. Hence, it is obviously proven that no universality in the matter of politeness.

Selection of Respondents
Sampled purposively, 100 Shanghainese were involved in this study-they consisted of 50 city dwellers and 50 rural residents. The latter group was comprised of 28 respondents residing Songjiang district, a city to the south of Shanghai, and the other 22 people were selected from Jiading district, a city to the north of Shanghai. Of the 100 respondents, 56 were males and 44 females, aged between 17 and 75 years old. Their occupations varied; they include students, marketing officers, security officers, drivers, coaches, consultants, accountants, art designer, etc. There were also respondents who had already retired and those who were unemployed.
Selection of such diverse characteristics of respondents was purposeful, aimed primarily at obtaining various views from the respondents regarding the issue being sought by the study. This judgement sampling, however, should not be seen as being contradictory with Sankoff's view which states that while it is not necessary for a speech community sample to have a large number of members, the sample must be well selected and "representative of all social subsegments about which one wishes to generalize" (1980: 52). This is apparently because, unlike many other kinds of behavioural surveys, linguistic behaviour is comparatively homogenous (cf. Labov 1972). Therefore, the idea of homogeneity will be strengthened, because the study views that homogeneity is not only evident in the language behaviour as realised through acts of speech but it also, and more importantly, relates to the mind of the speakers. In other words, the sample size of 100 respondents was satisfactory, because, as Sankoff pointed out, that " … even for quite complex speech communities, samples of more than about 150 individuals tend to be redundant, bringing increasing data handling problems with diminishing analytical returns" (1980: 51-52). That is because of the homogeneity of peer group speech behaviour.

Data Collection Procedures and the Instruments
Data were collected through the direct semi-structured interviews with the respondents, guided by a set of questions regarding the situations they might hypothetically face themselves. This instrument looked like the discourse completion task (DCT) initially developed by Blum-Kulka (1983) when investigating the realisation of requestive speech acts among Israelis. However, the present instrument differed in its form, because it did not contain the hearer response part as found in Blum-Kulka's. Instead, it provided respondents with some options they could choose from. The options were believed to represent respondents' views about the issues being asked in the situations. Apart from containing statements which represented respondents' opinions, the options also required respondents to exemplify their utterances in the spaces provided. This modified form is closer to and reflects the development of the DCT used in my previous research (Aziz 2000) when searching for the forms and strategies used by speakers of Indonesian in giving refusals. Below is an example of a situation asked to the respondents:

Situation #10
You and some friends are inviting a guest from overseas to have dinner in an expensive restaurant in town. When it turns to pay the bill, you ask everyone to pay, because according to your culture, anyone coming to the dinner will have to pay for themselves, which makes him/her so surprised. On the other hand, in your guest's culture, if somebody is invited, he/she will be treated/paid for the food and drinks. Therefore, your guest did not expect paying anything. What are you going to do/say? a) You pay for him, although you know that you are running out of cash, because …… b) You explain to your guest that everybody will have to contribute, and so will your guest. You say to him/her ……..
This modified version of DCT is believed to be capable of disclosing at least two aspects of interpersonal communication at the same time. Firstly, the format gives respondents enough freedom to express what they believe to be the most appropriate action to be taken when confronted with the situations described in the format. This means that what they choose from the available options clearly reflect and represent their inner conscious belief about human relations. This is so because there is no external force that dictates them about a particular choice. In other words, their decision implies their independence. Secondly, the format requires that after a decision is made, exemplifications are to be provided. Human relations are by nature characterised by the use of spoken language. Requiring the respondents to supply what they are most likely to say when facing their interlocutors enables us to know and then understand how they will verbally regard and then treat their interlocutors. Not only does it concern the how we express our opinion but it is also to do with the what it is inside our utterances. The format we develop here allows us to make inferences whether what a respondent exemplifies actually represents his/her belief.

Types of Data Obtained and Frameworks of Data Analysis
Two types of data were collected through this investigation. The first type is in the forms of expressions of respondents' attitudes towards the issues being asked in the questionnaire, in which reasons were provided by the respondents. This type of data was obtained from the respondents' choice of options provided in the questionnaire. The second type is in the forms of exemplifications of utterances that the respondents would presumably use when confronted with the situations described in the questionnaire. It is believed that the first type of data is capable of disclosing the respondents' concepts of face and the second type is more to do with the realisations of the concepts to which (im)polite acts will be manifested.
Respondents' concepts about face, which are stored in their mind and appear to be so philosophical in nature, have to do with their beliefs and understandings of the concepts by which they will act upon in a situation. These beliefs remained untouched, and we will not be able to disclose them until we leave the respondents to "speak out" their mind. Therefore, analysis of the concepts was based on these confessions. It is believed that to a great extent, politeness realisations are influenced by such beliefs accordingly. Other variables that are potential to affect politeness realisations by speakers in a speech event include their perceptions about their proximity with their interlocutors, contexts of utterances, and level of impositions of a given act. The speakers will take these very seriously to ensure the right conduct and maintain each other's face. All these topics will be used to sketch the phenomena sought in this investigation.

Findings
In this section, the general findings of this investigation will be presented, the basis of which is the variables involved. These include the respondents' age, sex, residence, and occupations. This description is important to see if there are any possible variations of responses given by the respondents.
Age. From the 100 respondents involved in this study, most respondents (47%) were of young age group (below 31 years old), 25% middle-aged (31-45 years old), and the other 28% over 45 years old. Looking at the responses of each age group, we found that there are considerable differences in the way they view a situation of a speech event.
Our data show that in a situation like the one described in Situation #1 where the respondents are faced with a guest who smokes without asking permission, most respondents from all age groups believe that they would leave the smoker finish his smoking. The reasons for such allowance, among others, are mainly because the person is their own guest or friend, or because they themselves smoke. However, when it turns that they had to tell the guest to stop smoking, as indicated in option b), the respondents would express their attitudes only in two forms, i.e. either reminding that smoking is not good for health or saying politely that they could not stand smoke. As for other actions i.e. option c), the respondents said that they would just open the windows in their place or they would cough to indicate that they could not stand smoke.
The phenomena above are comparable to other speech events such as described in situations #4, #5, #10, and #16, where guests are involved, and the settings are in their home. In all these situations, the respondents seem to be more concerned with the maintenance of the faces of their guests even if that is at the cost of their own. This attitude seems to be not without any rationale or logic, because they appear to want to show a positive image about themselves to their guest, let alone they are strangers. This positive image is important for them because that relates to the maintenance of their social status and prestige, into which face is inherent.
The above phenomenon is quite different from what we can find in other situations, such as those described in situations #2 and #4, where more familiar interlocutors such as friends or neighbours are involved although the settings are in their home. Here, all our respondents feel obliged to provide responses which lead them to attend to their interlocutors' faces. This is indicated in the high percentage of responses given by all our age groups. In situation #2, for example, more than two-third respondents chose to lend their interlocutors some money as requested although they knew that the money was to be spent not on something urgent but rather for holiday. There are only few responses that suggest that the respondents are upset by the request from their interlocutor as indicated in option b) that will stimulate them to blame the requestors accordingly. In situation #4, the respondents were receiving a complaint from their children whose food was eaten up by their neighbour's children. Here, they preferred an act of trying to calm down their children. A similar case can also be found in situation #5, which describes the other side of the event, namely it is the respondents' children who had eaten up all the food. What they then did was they, on behalf of their children, requested an apology for the incident while at the same time indicating their readiness to pay off when necessary.
When asked why they did so and so as indicated in their responses to situations #2, #4, and #5, most respondents replied by saying that they took all the actions merely because they would like to maintain their long friendship (situation #2) and good relationship with their neighbours (situations #4 and #5). Lending or borrowing money is not unusual in their society, but that is only done when people have a feeling of very good relationships and trust one another. Similarly, they realize that they live among people in their neighbourhood. Together, they feel obliged to attend to each other's wants and interests so that warmer and stronger bonds of relationships can grow. Therefore, for them, maintaining such good relationships both with their friend and their neighbours is so crucial that they would try hard its everlastingness. It is believed that this is one type of attempts that all members of the society can do and try to gain face from their society.
Responses to situations #6, #8, and #9 are interesting to note. In these situations, a stranger (situation #6), a boss (situation #8), and a teacher (situation #9) are involved. All these parties are deemed to have done something "wrong" in the eyes of the interlocutors. In situation #6, for example, someone, looking at the appearance of the spouse in a doctor's consulting room, had a wrong assumption about the respondent' spouse who was presumed to be his/her "father" or "mother". Interestingly, faced with this situation, many respondents (more than one-third) did not object to such a wrong assumption, although more respondents acted differently, i.e. by telling their interlocutors that the person was actually not their "parent" but instead their spouse. The former group of respondents, instead of blaming their interlocutor for the wrong guess, would continue their conversations with their interlocutors. They did it because they did not want to embarrass their interlocutors for their false assumption. Even for the other groups of respondents who chose to tell their interlocutors about the fact, they began their expressions of objection by a request of apologies "sorry", then told their interlocutors that the person was actually not their "father" nor "mother", but instead their spouse. When an expression of apologies from the side of their interlocutor follows, as indicated in situation #7, almost no respondents objected or at least blamed their interlocutors for their assumptions. Instead, they just said mei guanxi "never mind".
In situations #8 and #9, more powerful people, i.e. a boss and a teacher are involved. When these people did something wrong, only few people were willing and brave enough to argue and correct them openly. Our respondents, if they really had to argue and correct their wrong-doing, would do it tactfully "silently" in a personal sphere so that the people would not lose their face. This is clearly indicated in their strategies of arguing (situation #8) and correcting (situation #9). To the boss, they could only propose to have a company trip in a two-week time, expressed through the use of a suggestion such as How about in two weeks from now, Sir? Is that possible? To some extent, this expression suggests the speaker's "powerlessness". In relation to face-saving acts, this has to be regarded as an attempt made by the speaker not to impose on his/her interlocutor's interests.
When dealing with the teacher, more than two-third respondents preferred not to interrupt the teacher's explanation, but instead they would let him/her finish it. They would only talk to the teacher after a class session in a more private place, not in as a public place as a classroom. Here, they would tell the teacher what was wrong with the explanation just given beforehand. In Chinese community, teachers seem to occupy a special class, and they therefore deserve high respects. When it turns that members of the society, i.e. students in this respect, had to correct the teacher's mistake, they have to find a right and tactful strategy that will not harm either the teacher's face or endanger their future interests as students. Here we found that all groups of respondents developed a strategy that enabled them to conceal their feeling of being correct behind some hedging devices such as a question, uncertainties, etc. Some older respondents, for example, asked the teacher whether the theory he/she had just explained had not changed since it was first spawned. They said a) I don't know if the theory is very old and has not changed since then.
In the case of younger respondents, some of them mentioned to the teacher that they had read other references and found new insights, although they were not certain about their accuracy. They said b) Teacher, in the books I have read, I found something different from what you just explained, but I'm not sure whether this is true.
To a great extent, the above phenomena indicate that attending to the face of a stranger, i.e. someone whom we do not know yet (as described in situation #6) or of those from higher social status (as found in situations #8 and #9) is instrumental, in the sense that they will affect their perceptions about us. Because face is a public image, this cannot be lost publicly. In other words, maintenance of others' faces in a public arena represents the understanding and consideration of people towards them.
Sex. The proportion of male and female respondents involved in the present study is quite balanced: 56% males and 44% females. Yet, this proportion is neither ideal nor representative enough when we look at the percentage of Shanghai population where the number of females outnumbers that of males (but see above on section 3.0.1.).
From our data, we found that the trend of the responses given by both males and females is quite similar. This means that to each situation given in the questionnaire, both males and females indicate similarity in their choice (as indicated in the trend of percentage in each situation). Yet, there are actually considerable differences when further analyses are directed at findings the reasons behind the choice and/or when we look at the exemplifications of their utterances in a given speech event. We found that females gave relatively different reasons and exemplifications from those of males. Take for example the responses and exemplifications given in situations #3, #6, and #10.
In situation #3, both male and female respondents preferred to stop their overseas guest from eating the raw food served during dinner though faced with the risk that the guest would be embarrassed. This is clearly indicated in option c) in the questionnaire. We can find their preference in the data, and this is shown in the high proportion of responses to chose option c). After asking the guest to stop eating, most female respondents would tell that the food is okay to be eaten raw, but the taste would not be as good as when it is cooked. They would say something like a) The food is okay to be eaten raw, but cooking it well is much better and it'll be tastier. b) Are you sure you like raw food? But I prefer it cooked. c) Sorry, it's my fault for not telling you to cook the food before you eat it.
Most male respondents, on the other hand, would inform the guest that the food is actually raw and consequently cannot be eaten yet. Therefore, they asked the guest to stop eating it before it is cooked well. We found expressions like a) I would say that the food is actually raw and you cannot eat it until you cook it. b) Sorry to stop you eating, the food is still raw and needs cooking.
Differences in the exemplifications of responses given by male and female respondents can also be found in situation #6. Having chosen option b), the respondents are required to exemplify their expressions that the person was actually not their "father" or "mother", but rather their spouse. Most female respondents tended to choose an expression that would minimize the cost to their interlocutors. These expressions include the provision of apologies and/or other mitigating devices.

Some of their expressions are like the following a) I'm sorry; he is not my father but my husband. b) I'm sorry to tell you the truth: in fact, he is my husband. c) Does he look old? He is my husband.
Of the 25 sample responses given by female respondents, we found no responses that sounded like blaming or complaining the interlocutors for their wrong assumptions about the spouse. This is considerably different from the fact that we found from the responses given by male respondents. Apart from explaining that the person was actually their spouse, not their mother, male respondents were also mentioning that their interlocutors' assumption was obviously wrong. They, for example, said a) You must be wrong; she is not my mother, but my wife. b) Sorry, you are wrong; she is my wife. c) You make a mistake, you know. She is my wife.
Such differences suggest that male and female respondents differ in the way they regard an incident as described in each situation in the questionnaire. Compared to males, females seem to be more aware of the importance of keeping their interlocutors' faces positive and unintruded. They did not want to appear offensive but instead they wanted to be more tactful. That their interlocutors had a wrong presumption of the fact, that was solely due to their lack of knowledge about the fact; that is not because of their indulgence or their negligence. Consequently, they seem to be able to accept the wrong assumption and act considerately accordingly. Their expression of request for apologies that precedes their responses duibu qi "I'm sorry" indicates their consideration and acceptance. This contrasts with the lines taken by male respondents, where they appear to be more affirmative in showing their objections about the wrong assumption made by their interlocutors. They cannot tolerate this accordingly. Therefore, they told their interlocutors explicitly that they were wrong, and they wanted them to know it. For those male respondents, such a wrong assumption is perceived like a humiliation and they feel like that their face is being attacked. Realizing the importance to maintain their face, they therefore express such objections.
The incident described in situation #10 is quite different from those in #3 and #6. In situation #10, an overseas guest is involved. The respondents, together with their friends, were inviting the guest to have dinner in an expensive restaurant in town. In the guest' culture, if someone is invited to dinner, there is no need that he/she pays the food and drink. By contrast, in the respondents' customs, you will still have to share in paying the bill, although you are formally invited. Confronted with such cultural differences, most respondents, males and females, chose to "adjust" to the guest's customs. They were prepared to pay the bill without asking the guest to share, although they knew that they ran out of cash themselves.
What seems to be interesting and worth noting is the reasons behind their decision to pay the bill. Although both male and female respondents emphasize the need to value a guest, let alone an overseas guest, they differ in interpreting the value. For female respondents, the value of having a guest and inviting him/her to an expensive restaurant relates to their interest to show off their social prestige. Therefore, it is their responsibility to treat their guest well. In other words, by showing to the guest that they fully care about their guest, their prestige can be uplifted and this is one way for them to gain face accordingly. On the other hand, male respondents valued a(n) (overseas) guest not merely for the sake of uplifting their prestige. Trying to adjust to the guest' culture can show their high adaptability to a "new" culture. Having treated the guest well, they expected that their guest would have longer memory of their hospitality and it was intended to seek a possibility of future transactions. In other words, apart from wanting to gain face from their community, they are also aware of their future benefits that they might gain as the effect from their present hospitality shown to their guest.

Residence.
With regards to respondents' residence, two types are recognized by the present study: city and rural. The proportion of the respondents under this category is equal, 50 city dwellers and 50 villagers. The 50 villagers were recruited from two different places, one in the north and the other in the south of Shanghai city. This selection was made to find whether there were differences that the members of these two areas with different characteristics would show on the issues being sought by the study.
With the exception to situations #6 and #11, the responses given by the two groups of respondents into the situations described in the questionnaire are distributed almost similarly. This trend suggests few variations in the preference of strategies used by the respondents in viewing a given situation. To a certain extent, this also suggests considerable similarity in terms of their attitudes in treating an incident indicated in the questionnaire. Nevertheless, we still find some variations in terms of the content of their expressions and the forms of their responses, although the two groups of respondents chose the same option in a given situation. For example, in situation #2, both city and village respondents chose option a) as their answer to the situation. Both respondents said that they would lend the money because they are good friends. In addition to this reason, however, some city respondents added by saying that they understand the situation being experienced by their friend and they regarded having a holiday is necessary for people during a work break. Therefore, being certain that the money will be returned as soon as their friend returns from his/holiday, the city respondents are pleased to lend the money. Village respondents, by contrast, regard having a holiday as not urgent, although, due to their long friendship, they would finally lend the money.
Such a difference suggests a difference in the way the two groups of respondents regard the life, i.e. having holiday in this respect. For city dwellers, to have a holiday during a work break is one way of enjoying their life, and this may uplift their social status. Here, holiday functions instrumentally in their attempt to gain face from other members of their society. Lending money to a friend even to be spent for a holiday is therefore logical, in the sense that that should be understood as a sign of high consideration on the part of the lender; doing it is more than just to maintain friendship but it is also a sign of financial stability, which in turns indicates status. For villagers, on the other hand, spending money for holiday is not so wise, because money can be spent on something more important such as helping families, or just saved. For them, to have holiday, let alone by using money from a loan, is indicative of being wasteful. It is only for the sake of maintaining friendship and (gaining/enhancing) face if they finally lend their money to their interlocutor.
The difference in view of face preservation between city dwellers and villagers is also evident in their responses provided in situation #9. With a considerable difference in the proportion of responses to option a), among the respondents of the two groups agreed that they would be more likely to correct the teacher's mistakes right after the class session rather than directly raising hands during the lecture. The city dweller group would tell the teacher about the mistakes made while referring to the books/journals they had read that contained other views. When asked why they chose the strategies, the two groups had different views. Members of the city dweller group said that they needed to correct the mistake because they did not want other members in their class to be misled by the explanation, although they did the correction outside the class. They would leave to the teacher how the mistake would be corrected and informed to the class. By this way, they would not be regarded as having lost the teacher's face and prestige, but rather having saved it. The villagers, by contrast, felt that they did not have any courage to correct the teacher's mistake by directly pointing to it, because most of them never had experienced it and in their families, they never found such an act of correcting other senior/respected members. Therefore, they would seek other time after the class session to talk to the teacher. As students, they would feel inappropriate to pinpoint the mistake made by the teacher, because the teacher might forget or others; or they might have even misheard what was being explained. Therefore, they had better express it to the teacher as their problem in understanding the issue than telling the teacher that this was a mistake. By this, the teacher may re-check and consider their opinions without a feeling of having been face-lost. In other words, telling the teacher that they had a problem is more appropriate than doing it otherwise i.e. telling that it is the teacher who actually had a problem, let alone telling that the teacher was wrong.
There is a difference in the trend of responses given by the respondents from the urban and rural areas to situation #6. More than a half of city respondents chose option a) as their answer to the incident described in the situation. This means that when their spouses were wrongly assumed and thought as their parents, they seemed not to object to the wrong assumption. When asked about "their parents"' illness, they even continued their conversation with the person asking by telling the illness. This act was taken because they were afraid of embarrassing the person if they said that the person they were accompanying was actually not their parent but rather their spouse. Further interviews with this group of respondents showed that they were actually unhappy about the wrong assumption made by the person about their spouses. Only because they were in a doctor's consulting room, i.e. a public place, did they finally refrain themselves from complaining their interlocutor.
Had it happened in a more private place, they would have scolded their interlocutor. This is clearly indicated in their responses given to situation #7, where some respondents would blame their interlocutors although their interlocutors, having known that the assumed parents were actually their spouses, requested for apologies (indicated in option b) in the questionnaire). This means that in a public arena, the city respondents claimed that they would be more concerned for giving face to their interlocutors than to their own.
Unlike the city respondents, more than three-fourth of village respondents (81.6%) preferred to choose option b) as the answer to situation #6, which provided a chance to the respondents to "correct" the wrong assumption made by their interlocutor. Looking at this figure alone, we would surely think that villagers appear to be more open and prepared to show their objection to their interlocutors. However, if we look further to the responses, we found that almost all their objection was preceded by an expression of apologies or at least regrets. This suggests that they wished to be honest and at the same time tried to avoid conflicts either with their interlocutors because of having embarrassed them of the incident or with themselves because of the wrong assumption about their spouses. Such conflicts were minimized by showing to their interlocutors their politeness, i.e. by the use of an apologizer.
The difference in the trend of responses was also found in situation #11, where the city and village respondents had a slightly different perception. It was described in the situation that when they were in a rush going out from a supermarket for an urgent appointment, they bumped into their boss who was also their neighbour to whom they put respect so much. Faced with such a situation, the city respondents claimed that they (74% of the total responses) would choose to request for an apology and then leave the place immediately. This contrasts the village respondents who, 54% of the total responses, would apologize for the incident and spend sometime to talk to the person (boss-neighbour) before they finally asked to leave.
The incident described in situation #11 above requires the respondents to act highly tactfully because the possibility of face-lost on their part is relatively high; they were dealing with the boss who at the same time was their neighbour. Making a mistake to such a type of person is costly and really face-threatening. However, in the eyes of the city respondents, such an incident was not so critical and should not result in a serious problem. Therefore, asking apologies by saying I'm so sorry, I was in a rush would be sufficient because Shanghai is so densely populated that there is always a possibility that in a busy place like in a supermarket, people accidentally bump into one another. Also, people in a big city like Shanghai tend to be hurried in doing things to catch up with their busy schedules. Therefore, they requested understandings and consideration on the part of the person they bumped into and the person should regard it as not serious. A high mutual understanding in such a situation is therefore needed.
Such a view is remarkably different from that of the villagers where they regard the incident described in situation #11 as more serious. A boss in a workplace has a great power and is so influential in "determining" their (future) life, and they therefore have to put high respect to him/her. Similarly, a neighbour is the party with whom they live, and they have to help and respect each other accordingly. It was such an interconnection that multiplied the need to act tactfully. Expressing a big apology then spending a time to talk to the person of such an important position could become a sign of being polite and that could break the tension that might happen between them after the incident.
Occupation. This variable was included because it was believed that the intense interactions that the respondents had in their daily life would significantly influence their perception about an issue, even if it had to do with an issue as philosophical as face concepts. According to their types of occupation, our respondents were distributed into four categories: professional workers (26%), (lower-class) workers (55%), university students (8%), and retirees (11%).
Based on our data, we noted no further big differences in terms of their preferences, with the exception in situations #1 and #11 (unemployed group) and #8 (student group). This means that they actually had similar views about the acts they had to take when faced with situations described in the questionnaire. However, as has been discussed in earlier sections for other variables, our further analysis on the responses given by each respective group noted that there were some remarkable differences in their responses. This is evident particularly when the settings described in the hypothetical situations were really closed to reality, i.e. professional and (lower-class) workers facing situations of workplaces, students facing students' life situations, and unemployed facing their daily life in their neighbourhood. For example, professional workers often had guests, also from overseas; university students often came across with situations such as described in situation #9, etc. Moreover, most respondents claimed that they often found themselves in the situations described in the questionnaire. Therefore, we can be sure that the data we collected represent the actual beliefs of the respondents with the assumption that the closer the situations to the respondents' real situations, the closer the responses to natural data. Let's take some examples from our data.
In situation #8, for instance, the respondents were required to provide a response to the explanation given by the management who had postponed their group excursion into a month later while they had agreed to have it in a week time. The reason for the postponement, according to the management, was because of a cash-flow problem. On the other hand, they noted that members of management had new cars. Most of our lower-class worker group provided responses that would accept the explanation without giving any comments. They gave reasons that, if they did so, they were afraid of the risk that they might face in the future. They did not have any courage to protest the "policy" because of their weak position. Even if there were respondents from this group who asked the boss to reconsider just a two-week postponement, they mentioned that they were just curious and it was just a proposal. The professional group, on the other hand, appeared to be more notable in raising their objections to the decision made by the management. They argued that one month postponement was too long as they had worked too hard and needed refreshing time. They mentioned that having a company excursion would further lift up their productivity and then add company's benefits.
Student group respondents had a slightly different tendency and view about the cancellation. Half of them responded by saying that they would directly ask the management to convince them for the cancellation and would insist that the trip be made no later than two weeks after the announcement. For them, this was important to show to the management that they also had a good bargaining position that the management could not neglect. Negotiations with the management then had to take place. The position and view taken by this group of respondents can be understood when we connect it with the dynamic and critical position that are always attributed to students. This attitude is also supported by the evidence in their choice of option given to situation #9 where they expressed their objection to the wrong explanation made by their teacher. They conveyed it directly by, for example, saying: a) Teacher, you must be wrong. Other recent references I have read say something different. b) I think your explanation was wrong. The theory you said was no longer valid. c) Excuse me, teacher, I have a different opinion.
In the students' utterances above, directness is evident, which was not found in the responses given by the unemployed group respondents; not only in the contexts of situation described above (situation #9) but in almost all situations too. This group had a strong tendency to express their objections, refusals, and other forms of declinations/disagreements indirectly. We may predict that this inclination closely relates to their maturity, which had given them more experiences and understandings on how to interact with other members of their society.

Discussions
(Im)politeness behaviour shown by a person is believed to have been influenced by his/her perceptions and beliefs about how to behave within his/her society from which he/she would gain prestige, status, and respects or otherwise from other members of the society. These beliefs, in Chinese culture, are associated with the understanding of the concepts of mianzi "face". This mianzi operates and affects all aspects of interactions between members of the society. Therefore, face and politeness limao are two inseparable concepts.
As far as our data have indicated, the interrelationships between the understanding of the concepts of face and politeness realizations have pointed to strengthening the ideas formulated recently by Gu (1990), Mao (1994), andJia (1997). These researchers emphasise the need to relate the discussions of the Chinese concepts of face and politeness with the norms and values that prevail in the society, without which all accounts of the concepts are susceptible to invalidation, as they have seen in the theory developed by Brown&Levinson (1987). These society's norms are rooted in the philosophical values of Confucianism. In the following section, we will examine our data based on the four notions found in Confucianism in relation to the concept of face: relational, communal, hierarchical, and moral. All these concepts will be discussed within the context of politeness realisations.
Relational. This is the principal concept of face in the Chinese culture, and it concerns the affective and emotional human feelings owned by the members of the community in their attempt to promote a harmonious human relationship. The behaviours of the members of the Chinese community are governed by this principle so that they can avoid or at least minimise conflicts that may occur among themselves. Inability to prevent them from happening may result in the lost of face.
Our data show that, as far as they could do it, the respondents seemed to try hard to observe this principle in all their interactions with their interlocutors within the situational contexts described in the questionnaire. This is indicated in their avoidance of having conflicts with their interlocutors although they found themselves in unfortunate circumstances. Take for instance situations #1, #6, and #10. All these situations placed the respondents in the risky circumstances that would possibly endanger their health (situation #1), threat their prestige (situation #6), or even their financial conditions (situation #10). However, due to their closed observance of the relational principle, they put others' interests over their own. Even if they could not bear smoke or might suffer financially, they would leave their guests smoking or paying the bill respectively, only because they wanted to respect their guests. Likewise, they did not get upset when their spouses were assumed to be their parents, which to a greater extent could suggest monstrosity or humiliation. It may be true that their behaviours were not sincere (cf. Gu 1990) or honest, but for the sake of politeness, face-gaining and face-saving, they came up with such actions. In their view, a sincerity principle cannot override or even ignore the need of maintaining social harmony, which is far more important.
Communal. This social principle functions more as a "public censure" that substitutes the law usually used to regulate and punish the wrong-doers. This principle will prevent the members of the society from behaving against the ideal society's values (cf. Mao's (1994) ideal social identity) which have been agreed as collective standard norms. Looking at the examples of the responses to the incidents described in the situational contexts contained in the questionnaire, we found that the respondents appeared to view communal principle as central to their interactional behaviours. This was so because they strongly believed that the force of the sanctions that might be enacted by other members of the society due to their wrong behaviours would be so severe that they might not be able to control it. They therefore preferably chose to comply with the demand from the society's norms that require them to do so rather than thought and behaved more for their own interests. Gu (1990) was right when saying that in Chinese society, politeness, a derivate of the understanding of the concepts of face, functions not only instrumentally but also normatively, i.e. it requires the fulfilment of society's wants.
The incidents described in situations #4, #6, #11, and #14, followed by their respective speech events, allowed the respondents to react to the mistreatments by others to the extent they liked. They could blame, scolded, or even fired their interlocutors, if they wanted to do so, as in the case of situations #12 and #13. However, looking at the reactions indicated in their responses, we noted that they were inclined to accept their interlocutors' "indulgence" and refrained themselves from making a revenge. If they did so, they were likely to lose lian. Further consequences may take place, namely other members of society cannot accept the revenge and would in turn regard them as the people with no prestige. In addition to a possibility of losing their lian, the people of such kind are also vulnerable to exclusion from their society.
Accepting apologies from the people who had accidentally done a wrong doing might lead to the enhancement of face of a person. Other than that, the person will also be regarded as a man of honour, because he/she has observed the interests of both the people and the society. This polite behaviour, Gu called this as "attitudinal warmth" (1990: 239), demonstrates kindness, consideration, and hospitality, which he believes as one of the basic constituents of the notion of limao "politeness" in the Chinese culture.
Hierarchical. Confucianism dictates that the relation of order (hierarchy) be in operative in the Chinese human interactions. This relation, described in elder-younger brothers' relation, suggests the need to respect seniority, which particularly concerns age differences. However, other sources of hierarchy can be identified, and these can include blood bonds and origins, wealth, positions, education, and so on. The most salient evidence and common practices of this societal value can be found in the naming practices among Chinese (cf. Lee-Wong 2000), which are intended as a sign to show different deference and relationship. This relationship in many cases is often built on a hierarchical foundation. In Scollon&Scollon's observation, for example, "Chinese have a rather complex structure of names which depends upon situations and relationships, which includes school names, intimate and family baby names, and even western names, each of which is used just by the people with whom a person has a certain relationship" (1995: 123).
The relative social status attached to a person calls for a different treatment to be given to the person by others. The incidents that caused a person of higher social status to an unfortunate circumstance such as described in our situational contexts in the questionnaire, according to Confucianism, require that the person be more considerate, and vice versa. The person has to take it as less serious and as not harmful to his/her prestige. If he/she tries to do otherwise, he/she may experience losing lian. In Hu's view (1944: 47), "a person of high status … has to be more circumspect in dealing with people of lower status", because even if the person eventually gets the respect from the inferior, "this respect would be impaired if ego lost dignity by behaviour very contrary to the expectation of society" (cf. Gu's (1990) notion on refinement).
Let's have a close look at the choice made by our respondents to situations #12 and #13, where being in a rush, a subordinate bumped into his/her boss. Pretending not to recognise the boss (situation #13), the subordinate just left the place without saying anything.
The questionnaire asked what the respondents would have done if, supposedly, they had been the boss. Surprisingly, majority responded that they would act as if there had been nothing happening and they would greet their subordinate as usual. When asked for the reasons, they generally said that they were afraid of the social sanctions they would have to face if they scolded the person, let alone in a public place. This noble attitude, we can be sure of, does not instantaneously come into existence without the person having achieved a state of being fully aware of the values of maintaining his lian. On the part of the subordinate, on the other hand, the decent behaviour shown by his/her boss, to some extent, may be taken as a serious "attack" to his/her face, which could result in the lost altogether. Therefore, he/she would have to try to regain it from at least his/her boss, and this would not be a simple endeavour.
Moral. Morality is another basic constituent of the Chinese concepts of face, the loss of which is equated with the loss of morality. Seen from this perspective, moral integrity de, I therefore believe, serves as the most basic element that constitutes the Chinese concepts of face. From this morality will all other good human characters spring. On the other hand, morality occurs at the farthest end of a human character continuum and it becomes the utmost goal that all rational human beings will strive to achieve.
With reference to our data, we noted that the respondents seemed to always try avoiding making offence to others, attend more to others' interests than to their own, and highly respect others because of their good reputations. This "self-denigration", in Gu's term (1990) reflects the respondents' adherence to the values that prevail in their society that demand them to do so, which in turn implies their closed observance to the need of achieving high moral integrity. Letting a guest smoke while self does not bear smoke (situation #1), forgiving others for the wrong assumption about self's spouse (situation #6) or for a careless act that results in a bump (situations #11, #12, and #13) while the self has the full power to take revenge represent the examples of acts of a human with high morality. This type of person never thinks of returning others' wrong doing or thinking about him/her, because, if he/she does so, that clearly reflects the low moral standard of him/herself. This will in turn result in the loss of lian of the person.

Conclusions
We witness that the order of the world is changing and the one in China is not an exception. Our primary aim in this study was to look at the effects of the changes that the world is experiencing on the conceptions and the practices of such traditional concepts as mianzi or lian "face" found in the Chinese society. Based on the data we have collected we found the following. Firstly, our interview data suggested that many of our respondents did not really realise the existence of the philosophical concepts of mianzi or lian, although they could find the words in their vocabulary. This was indicated in their inability to explain fully the philosophical concepts. Therefore, their understanding of these two words was only limited to the understanding of common people. Yet, that does not mean that they do not practise the concepts in their daily life.
Secondly, there was a different view of the respondents with regard to their perception on how to interact and treat other people. Among villagers, for example, it was found that their feeling of togetherness was quite high, so that a more harmonious community would be more easily achieved. City dwellers, on the other hand, viewed that harmony could not always be examined from being cooperative with others, but it also had to be seen from the ability to be more independent. In other words, trying to minimise intrusions and impositions to others is one way of creating and maintaining harmony among members of their society. Consequently, there were many instances of their communication transactions which were more driven by instrumental forces than by relational and normative forces, i.e. taking future benefits or the nurturing effects of their communication into their consideration. Therefore, politeness realisations of the groups of these two opposing views were different.
Thirdly, different from Pan's findings (1995), it appeared that age becomes a crucial factor that distinguished the respondents' perceptions about the traditional concepts mianzi and lian, which in turn resulted in the different politeness realisations. Compared to the middle-aged and the older groups, the younger respondents were found to be more straightforward in expressing their illocutionary acts. By contrast, the two former groups appeared to be more careful and hence less direct. Gender was not found as a distinctive variable with regard to the respondents' perceptions about either face concepts or politeness realisations. As far as types of occupations are concerned, this variable contributed to the differences in the respondents' views about the concepts of face as well as their politeness realisations. Professional workers and students seemed to be less hierarchical in their forms of talks, although they showed high deference to their interlocutors. Finally, we found that politeness shown by speakers was mainly intended to maintain harmony among participants and their good image/prestige in the eyes of their interlocutors, particularly when dealing with their closed friends, neighbours, and foreign guests.
Using the parameters of ideal person contained in Confucianism in relation to face, we found that the principles of moral integrity de attached to the concepts of face mianzi and lian appeared to be the most basic aspect that all human beings have to attend to. This is because good morality touches the deepest side of humanity ren. The loss of moral integrity means the loss of humanity, and no more face can be found in such a type of person. Because the Chinese we observed through the present study were found to adhere to this concept, we can conclude that although China is changing, the conceptions and practices of the concepts of face mianzi and lian have remained constant.