Face and politeness: new (insights) for old (concepts)
Category : Paper Articles
Face and politeness: new (insights) for
Many gods have been done away with, but the individual himself stubbornly remains as a deity of
considerable importance. He walks with some dignity and is the recipient of many little offerings.
He is jealous of the worship due him, yet, approached in the right spirit, he is ready to forgive those
who may have offended him. Because of their status relative to his, some persons will find him
contaminating while others will find they contaminate him, in either case finding that they must
treat him with ritual care. Perhaps the individual is so viable as god because he can actually
understand the ceremonial significance of the way he is treated, and quite on his own can respond
dramatically to what is proffered him. In contacts between such deities there is no need for middlemen;
each of these gods is able to serve as his own priest.
E. Goffman 1967 : 95
The article re-examines Erving Goffman’s concepts of face and face-work and their roots in
the ritual and sacred essence of the social order as expounded in the work of the French
sociologist Emile Durkheim. Both Goffman and Durkheim are referred to in Brown and
Levinson’s classic work on politeness but the originality of their ideas has become somewhat
diluted. Using three of Goffman’s early essays, the article argues that his observations on the
interactional order and his sophisticated notions of face and face-work could be the starting
point for a re-appraisal of politeness and its fundamental role in the social order.
#2002 Elsevier B.V. All rights reserved.
Keywords:Face; Politeness; Durkheim, Goffman, Brown and Levinson’s theory
The concluding paragraph of Goffman’s essay ‘On the nature of deference and
demeanor’ is a revealing sketch of a characteristically Western, individualistpersona
which also informs the author’s seminal piece ‘On face-work’, originally published one
year before, in 1955. These, together with ‘Embarrassment and social organization’,
are the three essays republished in 1967 in the volumeInteraction Ritual, which provide
apt material for a critical re-appraisal of Goffman’s early treatment of ‘face’. In
the first two essays, Goffman expounds his early social anthropological model based
on a ritual account of self which is most directly influenced by Emile Durkheim’s ideas
of social solidarity (Manning, 1992; Ditton, 1980). Unlike Durkheim, however, Goffman’s
interactional order ‘‘remainsalmost fundamentally problematic’’ (Ditton, 1980:
36, original emphasis) as the third essay on embarrassment graphically illustrates.
Goffman is acknowledged as a distinct influence on Brown and Levinson’s work. In
the revised edition of their 1978 essay, which they dedicate to Goffman’s memory, the
authors write: ‘‘our notion of face is derived from that of Goffman and from the English
folk term’’ (Brown and Levinson, 1987: 61). Interestingly for the discussion in this
article, Brown and Levinson also make a number of references to Durkheim’s work in
relation to their elaboration of ‘ritual’ and ‘positive and negative rites’ (Brown and
Levinson, 1987: 3, 18, 43, 129, 285, note 8). The re-examination of face and face-work in
the light of Goffman’s original elaboration, heavily indebted to Durkheimian sociology
and somewhat diluted in Brown and Levinson’s account, is the subject of this article.
The concept of ‘face’ is commonly thought to have originated in China, and Goffman
himself acknowledges Chinese sources. A careful reading of his essay on face-work,
reveals some distinctly individualistic traits, which appear to be woven into an (originally?)
socio-psychological construct of ‘face’. This individualistic emphasis has been
picked up and elaborated byBrown and Levinson (1978, 1987) into a cognitive model
of ‘face’ based on Western ethnocentric assumptions such as the existence of a predominantly
rational actor and the strategic, goal-oriented nature of ‘face-work’ and of
social interaction. Hence their model’s obsession with Face Threatening Acts (FTAs).
The sub-title ofGoffman’s Interaction Ritual (1967), ‘The ritual elements of social
interaction’, of Durkheiminan resonance, is suggestive of ‘face-work’1 as the default
in interpersonal behaviour. As well as his treatment of ‘face’, Goffman’s notion of
‘face-work’ needs re-examining with reference to Durkheim’s workThe elementary
forms of religious life,which also inspired Goffman’s ‘deference’ and ‘demeanor.2 In
Goffman’s early discussion of ‘face-work’ may lie the key to a novel understanding of
the widely researched, but never convincingly defined, phenomenon of ‘politeness’.
In this article, I will attempt to:
(1) discuss the social orientation of Goffman’s elaboration of ‘face’, which will be
shown to be closer to the richer Chinese construct of face, which, in turn,
stems from a society traditionally dependent on a highly complex network of
social obligations, where hierarchy, status and prestige require acknowledgement
through normative, as well as strategic, ‘‘face-work’’;
(2) tackle the question of the relationship between ‘‘face-work’’ and ‘politeness’
by arguing that we need to distinguish between the two, and that a deeper
understanding of the roots and workings of ‘politeness’, in its non-linguistic
manifestations, can be gained by placing it within the domain of cultural
theory, in general, and social morality, in particular.
3. Goffman’ version of ‘ace’In his historical study of politeness, Konrad Werkhofer (1992: 178) laments the
selective interpretation of Goffman’s face metaphor in ‘‘unambiguously individualistic
terms, abstracting not only from the dimension of ritual order, but from all kinds
of social order’’, thus privileging cognitive notions implicit in the Gricean understanding
of communication over the social ones favoured by Goffman.Werkhofer
(1992: 176)concedes that Goffman’s notion of face ‘‘has inherent limitations’’, but he
is also critical of Brown and Levinson’s formal adherence to Gricean theory of
communication while, at the same time, postulating a polite speaker with face-threatening,
if not antisocial, intentions. This modern understanding of politeness,Werkhofer
(1992: 180)submits, ‘‘introduces the remarkable premise that there must be, as
a prerequisite for politeness to occur, a fundamental antagonism between the speaker’s
intentions, on the one hand, and social aspects, on the other’’.
A cognitive, individualistic interpretation of ‘face’ meets with further resistance in
research in many non-Anglosaxon cultures. The growing body of criticism on cultural
grounds of Brown and Levinson’s notions of ‘negative face’ and ‘imposition’
also testifies to the limitations of the original notion of ‘face’ and, possibly, of its
subsequent treatment in conjunction with a theory of ‘politeness’.
The cultural relativism of Brown and Levinson’s ‘face’ and their preoccupation
with face-threatening acts (FTAs) has forced researchers in a number of non-Western
languages to re-consider Goffman’s notion of ‘face’ (e.g.Mao, 1994; de Kadt,
1998) so as to engender a version of ‘politeness’ that accommodates both strategic
(volitional) and social indexing behaviours. Whereas the former appear to be preeminent
in so-called Western societies’,3 thus supporting Brown and Levinson’s
interpretation of ‘‘polite behaviour’’ as essentially strategic, the latter are more
typical of hierarchical societies, where relative status determines many of the
‘politeness’ norms in interpersonal contact. I shall return to both the critics of
Brown and Levinson and the cultural alternatives to their model presently.
The quotation at the beginning of this article encapsulates some of the themes
Goffman developed in the essays collected in the volumeInteraction ritual. Essays on
face-to-face behavior(1967), the first of which ‘On face-work’, is often referred to in
the literature of politeness as the source of Brown and Levinson’s notion of ‘face’. In
that volume,Goffman (1967: footnote 1) cites sources for both the Chinese and the
American Indian concept of face, thus indicating some of the influences on his own
thinking. Among them, one stands out: Durkheim’s work on the religious origins
and nature of social activities,The elementary forms of the religious life (1915), to
which Goffman explicitly refers and which influences especially his early essays,
three of which have been chosen as particularly relevant to my discussion. Goffman’s
ritual analysis of social life, to which he is directly indebted to Durkheim, is
predominant in the first four essays collected inInteraction Ritual (Branaman, 1997).
The first two essays deal with the function of ‘face-work’ and demeanour in maintaining
the ritual order of social life, while the third, by contrast, concentrates on
instances where the delicate balance of the ritual is broken. The potential for
embarrassment in every social encounter and the study of what Goffman calls ‘‘dissonance’’
enables the sociologist to study the causes of interactional breakdown and,
by implication, the rules that regulate social encounters (Goffman, 1967: 99). The
choice of Goffman’s essays that build on the ritual metaphor of society will be
shown to be relevant to the discussion of Goffman’s ‘face’ and to Brown and
Levinson’s repeated and, I would maintain, significant references to Durkheim’sThe
Early Forms of Religious Life. In later essays, such as ‘The Presentation of the Self’,
Goffman introduces game theory and the dramaturgical self, which are somewhat
remote from the earlier ritual and sacred characterisation of self that so critically
informs Goffman’s early notions of ‘face’ and ‘‘face-work’’ which are the object of
this article (Ditton, 1980; Drew and Wootton, 1988; Manning, 1992; Lemert, 1997).
After reading Durkheim, it comes as no surprise thatGoffman (1967: 45) should
sum up his essay on ‘face’ and ‘face-work’ with a statement about the function of the
‘‘moral rules’’ that guide choices aimed at maintaining ‘‘ritual equilibrium’’.
Of particular interest to my argument is his subsequent elaboration on the interdependence
of the individual and the social in interaction: ‘‘And if a particular person
or group or society seems to have a unique character all its own, it is because its
standard set of human-nature elements is pitched and combined in a particular way.
Instead of much pride, there may be little. Instead of abiding by the rules, there may
be much effort to break them safely. Butif an encounter or undertaking is to be sustained
as a viable system of interaction organized on ritual principles, then these variations
must be held within certain bounds and nicely counterbalanced by
corresponding modifications in some of the other rules and understandings’’ (ibid.,
According toGoffman (1967: 44, added emphasis) individuals become ‘‘interactants’’
when they agree to be ‘‘mobilised’’ as ‘‘self-regulating participants in social
encounters’’ through, among other things, ritual. Social values such as pride, honour,
dignity, considerateness, tact and poise are learned, as is perceptiveness,and
‘‘feelings attached to self, and a self expressed through face’’. Goffman’s social morality,
like Durkheim’s, is located in social solidarity. Durkheim’s morality is relative
to the group and emanates from the hierarchy within the group. For Durkheim,
morality cannot be imposed but must be ‘‘desired’’ and ‘‘desirable’’ (Steiner, 2000:
82). Of singular importance to Goffman’s ritual metaphor of society, however, is the
Durkheimian fusion of the moral and the religious: ‘‘it is rather difficult to understand
moral life if we do not place it alongside religious life [. . .] Therefore the moral
must be in the religious and the religious in the moral’’ (Durkheim, 1924, quoted in
Steiner, 2000: 82, my translation from the French).
In an attempt to understand the roots and practices of religion, Durkheim turns to
an empirical study of a simple religion: totemism. He concludes that religion should
be stripped of any supernatural and divine element and that beliefs and rites are the
two basic components of any religion. Beliefs rest on a classification of all things into
two categories, which Durkheim labels as ‘‘sacred’’ and ‘‘profane’’, respectively, and
between those he posits an absolute distance. Secondly, these two worlds are mutually
exclusive and in competition. Thirdly, however, Durkheim concedes that there are
relations between these two worlds, the so-called ‘‘rites’’, which enable communication
between the (profane) believers and their (sacred) gods (Steiner, 2000: 84–5).
On the subject of ‘‘rites’’, so very important to the early Goffman, and referred to
by Brown and Levinson in their monograph, Durkheim’s classification is more
complex than the one we have become familiar with through the literature on
politeness. Durkheim underlines that all rites enable social communion while being
of three different types: negative rites, which keep the sacred from contact with the
profane; positive rites, through which the believers communicate with their gods;
and sacrificial rites, performed in the face of disaster or loss. Negative rites are performed
in preparation to positive rites, which, in turn, lead on to asceticism and
ability to cope with suffering (Steiner, 2000: 87). In Durkheim’s social order, the
sacred has a strong emotional association for it represents the ‘‘collective conscience’’
of society (Steiner, 2000: 85). Religion works both as a representation of the
natural and social worlds and as a direction towards an ideal (Steiner, 89). Religion
emanates from the common conscience on which collective beliefs are founded
(Stedman Jones, 2001: 202). In this context, ‘‘a rite guides and anticipates social
experience and governs the terms under which action is undertaken’’ (Stedman
Jones, 2001: 202). A religion that originates in the collective, reinforces group identity
and the direction of action is central to Durkheimian sociology. In her recent
reconsideration of Durkheim’s thought, Stedman Jones observes that modern individualism
is comparable to religion because it is a form of belief, with its symbols
and ideals and its drive to action. When the belief becomes passionate, as in periods
of effervescence, the fixing of collective feelings on an external object come to constitute
a cult, that is a form of religion, such as, indeed, modern individualism
(Stedman Jones, 2001: 213).
In Durkheim’s model of society, solidarity generates rights and duties based on
values that are not inherent in things but are ascribed to things by collective thinking
(Stedman Jones, 2001: 190). Ritual is sustained by the fulfilment of certain duties or
obligations that Durkheim sees as invested in the individual towards society so that
‘‘the duties of the individual towards himself are in reality duties towards society’’
(Durkheim, 1893, quoted in Stedman Jones, 2001: 191).
This dimension of interdependence of social beings is not lost to Goffman, as his
social values of ‘‘deference’’, ‘‘demeanor’’, ‘‘tact’’ etc. suggest. Elsewhere,Goffman
(1967: 19)postulates the sacredness of the interactants’ face, the maintenance of
which requires a ritual order, i.e. ‘‘acts through whose symbolic component the actor
shows how worthy he is of respect or how worthy he feels others are of it’’. The selfaware
interactant has replaced the collective self of Durkheim. Whilst the individualism
of Goffman’s ideal social actor is a sign of the influence of the contemporary
Anglo-Saxon values of independence and privacy, Goffman’s concept of face cannot
be dismissed as simply ego-centric. In fact, an awareness of other interactants’ reactions
and feelings is famously expressed inGoffman’s (1967: 5) definition of face as
‘‘the positive social value a person effectively claims for himself by the line others
assume he has taken during a particular contact’’ where a ‘‘line’’ is the interactants’
own evaluation of the interaction and of all its participants, which includes self-evaluation.
Moreover, an individual’s response to others’ evaluation of his own face is not
purely rational:emotions are involved, so that harm to another’s face causes
‘‘anguish’’, and harm to one’s own face is expressed in ‘‘anger’’ (Goffman 1967: 23).
Social encounters are enacted in such a way that own face and others’ face are
maintained through self-respect and considerateness (Goffman, 1967: 11). However,
according to Goffman, face-maintenance is not usually theobjective of the interaction,
but rather acondition of it; the study of face-saving practices, he continues, is
the study ‘‘of the traffic of rules of social interactions, whilst ‘face-work’ refers to
‘‘the actions taken by a person to make whatever he is doing consistent with face’’
(Goffman, 1967: 12).
Goffman elaborates on ‘‘face-work’’ by saying that such ‘actions’ may be conscious
or unconscious, and often become habitual. Cross-cultural variation in facesaving
practices nevertheless reveals similarities that suggest that there may be a
fixed repertoire of ‘‘possible practices’’ (Goffman 1967: 13–4). Defensive practices
(saving one’s own face) and protective practices (saving others’ face) are seen to be
exercised simultaneously, another indication of the social value attached to ‘face’. It
is true that Goffman describes in some detail what he calls ‘‘avoidance process’’,
‘‘corrective process’’, and ‘‘the aggressive use of face-work’’, all of which have been
given pre-eminence in Brown and Levinson’s model. However, he devotes equal
space to the discussion of tacit co-operation that makes possible the performance of
face-work:4 ‘tact’, ‘reciprocal self-denial’ and ‘negative bargaining’ (making the
terms more favourable to one’s counterpart) are all practices available to interactants
as ‘socialized’ individuals, i.e. abiding by ‘the ground rules of social interaction’
(Goffman, 1967: 30–1).
These are the clear indicators of social awareness and concern for others (albeit
out of possible concern for self-preservation) that Goffman’s ideal interactant is
expected to display, for his nature is that of a ‘ritually delicate object’ (Goffman,
1967: 31). Goffman’s primary interest is in a theory of social interaction rather than
a framework for polite behaviour, and this led him to draw heavily on seminal work
by Durkheim where the symbolic value of social action is seen to have originated in
religious practice. Hence the religious language of some of Goffman’s characterisations,
not least that of the person as a ‘ritual object’, a ‘deity’, and ‘his own priest’,
and that of ‘face’ as ‘sacred’. Moreover,Goffman (1967: 55) distinguishes between
rules of conduct that inform ‘substantive rules and substantive practices’ (law,
morality and ethics) and rules of conduct that pertain to ‘ceremonial rules and ceremonial
expressions’, or etiquette. ‘‘Deference’’ and ‘‘demeanor’’ are the two components
of ceremonial behaviour thatGoffman (1967: 73) ‘translates’ from
Durkheim’s religious notion of ‘‘positive and negative rites’’ (Durkheim, 1915). In
particular, ‘‘deference’’ comprises a positive element (‘‘presentational rituals’’) or
‘‘other appreciation’’, and a negative element (‘‘avoidance rituals’’) through which
actors refrain from doing something so as to avoid invading the others’ personal
space. In the essays that I have chosen, Goffman’s individuals are ‘‘guardians of
face-to-face situations’’, projecting selves with a ‘‘social positive value’’ that must be
protected to preserve the equilibrium of the encounter. When the projected self
cannot be sustained, embarrassment occurs (Manning, 1992: 38–9). In encounters,
embarrassment and flustering are the opposites of comfort and ease and are considered,
in Goffman’s contemporary society, ‘‘evidence of weakness, inferiority, low
status, moral guilt, defeat, and other unenviable attributes’’ (Goffman, 1967: 102)
The moral language that describes the ‘‘flustered individual’’ warns against the
unsustainable position of the interactants who break the ‘‘ritual equilibrium’’ to
which they had committed themselves by moving ‘‘into one another’s immediate
presence’’ (ibid.: 99). Moreover, embarrassment ‘‘seems to be contagious, spreading,
once started, in ever widening circles of discomfiture’’ (ibid.: 106). Goffman chooses
organisational sites to illustrate how embarrassment ensues from a ‘‘conflict of the
selves’’. Engaged in conflicting relationships of equality and distance, individuals
seek to mediate often contrasting organisational principles, to preserve which they
may have to sacrifice their own identities and possibly the encounter, too. In this
perspective, Goffman’s individual is thus sacrificed for the social system, a scenario
that is strongly reminiscent of the Durkheimian solidarity on which the ‘‘collective
conscience’’ is founded. The loss of face through embarrassment is the gain of the
society, or, in Goffman’s laconic ending of his third essay: ‘‘social structure gains
elasticity; the individual merely loses composure’’ (ibid.: 112).
4. Goffman v. Brown and Levinson
Subsequent research on ‘face’ and ‘politeness’ has often been more intent on
quoting Goffman selectively than on examining critically the potential of his analytical
constructs and original observations. The first edition of Brown and Levinson’s
work appeared twenty years after Goffman’s and has been enormously influential,
generating a huge amount of literature. Critics, not surprisingly, are mainly non-
Anglophone researchers who find Brown and Levinson’s particular concept of ‘face’
difficult to apply in their own cultures and, consequently, have argued with the universality
of their definition of ‘polite behaviour’. Rather than retracing the lines of
criticism, a task beyond the scope of this article, I shall attempt to illustrate the ways
in which Brown and Levinson’s understanding of ‘face’ and ‘politeness’ falls short of
Goffman’s original ideas.
At the beginning of their 1987 revised essay,Politeness. Some universals in language
use, Brown and Levinson (1987: 13) point out that their notion of face is ‘‘highly
abstracted’’ and subject to ‘‘cultural elaboration’’. It is their dualistic notion of ‘face’, or
public self-image, with matching positive and negative politeness behaviours, that is at
the heart of their model and that departs most radically from both Goffman’s elaboration
of ‘face’ (and ‘‘face-work’’) and Durkheim’s ‘‘positive and negative rituals’’.
‘‘Avoidance rituals’’ have found extensive application in Brown and Levinson’s
elaboration of ‘‘negative politeness’’, from which a notion of ‘negative face’ emerges
that does not find correspondence in Goffman’s or Durkheim’s work. Brown and
Levinson’s cognitive concept of ‘face’ and the rational actor does not fit into Goffman’s
study of interaction, which he understood to be about ‘‘not the individual and
his psychology, but rather the syntactical relations among the acts ofdifferent persons
mutually present to one another’’ (1967: 2, added emphasis). It was the search
for the ‘‘general properties’’ that individuals share in social interaction, i.e. their
social psychology, that led Goffman to analyse ‘‘not [. . .] men and their moments.
Rather moments and their men’’ (Goffman, 1967: 2–).
In particular, it is ‘‘negative face’’ and, consequently, negative politeness, that
have generated the most criticism, usually on cultural relativistic grounds. Whereas
Goffman views ‘avoidance’ as a process whereby individuals avoid face-threatening
situations, this is not reduced to a clear-cut distinction between ‘‘freedom of action
and freedom from imposition’’ that characterises Brown and Levinson’s negative
face (Brown and Levinson, 1987: 61). The two types of ‘face’ and the notion of
‘imposition’, criticised by many as culture-bound, are not the only difficulties. The
negativity that connotes one side of ‘face’ and a type of ‘politeness’, could not have
been derived from Durkheim, although Brown and Levinson’s debt to the French
sociologist is signalled at least six times in the re-issue of their essay [pp. 1, 3, 18, 39,
61 (note 8 printed on p. 285)]. On p. 43, Brown and Levinson state: ‘‘That there
must be simple and direct links [between interpersonal politeness and formal rites]
we dimly saw when we borrowed the distinction between negative and positive
politeness from Durkheim’s distinction between negative and positive rites’’. In his
The Elementary Forms of the Religious Life, Durkheim points out that ‘‘negative
cult’’5 is ‘‘one means in view of an end: it is a condition of access to the positive
cult’’. (Durkheim, 1915: 309).
A little further on,Durkheim (p. 311) states more unequivocally that ‘‘normally,
the negative cult serves only as an introduction and preparation for the positive one’’.
’Negativity’ is in fact a positive and necessary contribution to an overall worthy
endeavour from which both the individual and society ultimately benefit. This is
hardly in line withBrown and Levinson’s (1987) elaborate description of ‘Face
Threatening Acts’ (FTAs) as a major concern of ‘polite behaviour’. ‘Giving’,
enhancing or maintaining face are less important objectives of interactional practices
than protecting one’s own and the other’s face. It appears that in Brown and
Levinson’s treatment of ‘face’, Goffman’s tendentially individualistic treatment of
the ‘sacred self’ becomes an obsessive attempt by an ideal rational actor to mark and
protect personal territory from potentially harmful interpersonal contact. Emotions
may indeed be present in Brown and Levinson’s model(1987: 61) but they appear to
be mostly concerned with defensiveness and protectiveness.
5. Goffman and the critical literature on ‘ace’Criticism of Brown and Levinson’s model has concentrated on the apparent conceptualization
of negative and positive politeness as mutually exclusive (the unidimensionality
proposition) and on the suggestion that the former is approachbased
and the latter is avoidance-based (the approach-avoidance distinction) (Lim
and Bowers, 1991: 418). Empirical evidence reveals the co-existence of many types of
face-work in situations where many face wants are threatened. Moreover, multiple
face-work is not only a possibility but, sometimes, a requirement and the threat to
face is a much weaker predictor of face-work than the right to perform a certain act
(Lim and Bowers, 1991: 448). From a different angle, criticism has been made of
Brown and Levinson’s assumptions that only one type of face can be threatened at
any given time, and that all FTAs can be analysed by looking at decontextualised
speech acts (Wilson et al., 1991–992: 218).
As early as 1984, findings from experimental research appear to expose ‘a British
cultural bias’ in the typology originally presented byBrown and Levinson in 1978
(Baxter, 1984: 453). Towards the end of the nineties, establishing ‘conceptual
equivalence’, that is uncovering the multiple meanings attached to the construct of
‘face’ which are shared across cultures, remains a fundamental aim of a (universal)
theory of ‘face’ (Ting-Toomey and Kurogi, 1998: 216).
The non-Anglosaxon literature in the eighties and nineties (e.g.Hwang, 1987; Ide,
1989; Lim and Choi, 1996; GarcıLa, 1996, Rathmayr, 1999; Placencia, 1996) seems to
point towards an understanding of face as a socio-psychological/affective construct,
but to date, we are nowhere near a universal construct.6 The Chinese ‘face’ is
essentially a more public and more positive concept, consisting of three positive
face-types (Lim, 1994), and firmly embedded in relations; i.e. it is a situational construct
(Ho, 1994). This clearly echoes Goffman’s understanding that normative and
situational factors determine the degree of sensitivity to face and the concern to be
shown for all faces involved in an interaction (own and others), which share equal
status (Goffman, 1967: 6). Goffman’s expression of ‘‘how much feeling one is to
have for face’’ (ibid.) directly recalls the Chinese notion of quantifiable ‘face’ (Ho,
1994). Apart from having, being or maintaining ‘face’, or being in the wrong face, or
out of face, interpersonal behaviour can be seen to ‘enhance’, that is, to increase
‘face’ (Goffman, 1967: 6–).
The Chinese notion of ‘face’ on which Goffman draws in his essay remains a primary
focus of interest in the situated study of interpersonal behaviour. The lack of original
Chinese discourse studies and the frequent borrowing of Western analytical frameworks
and tools continues to hinder the development of much needed indigenous
theories and empirical work (Chen and Gu, 1997). On the bright side, Chinese scholars
have provided some of the most developed and consistent critiques of Brown and
Levinson’s work to date (Gu, 1990; Chen, 1993; Mao, 1994; Lim, 1994; Gao, 1996; Ji,
2000) as well as research on related aspects of Chinese interpersonal communication
that throw light on the importance and the workings of ‘face’ (Ma, 1996; Chang, 1999).
It is perhaps worth remembering that seminal works on ‘face’ such asHu (1944), and
indeedGoffman (1967) himself, were not intended for application to inter-cultural
communication, but rather to cast some light on important aspects of (intra-cultural)
interpersonal behaviour.Brown and Levinson (1987: 14) claim that their model was
designed to accommodate ‘‘cross-cultural conflicts grounded in different views of what
constitutes ‘good behaviour in interaction’’’. However, their conceptualization of ‘face’
is actually weaker than Goffman’s, precisely with respect to cross-cultural validity.
The discussion on the cultural variability of face is one that leaves Brown and
Levinson’s uneasiness about ethnocentrism, and how to deal with it, an open issue.
Research in various cultures published after 1987 has had the twofold effect of
prompting a redefinition of both the validity and the weighting of ‘face’ as a determinant
of interactional dynamics (e.g.Ide, 1989; GarcıLa, 1996).7 Extant research
appears to indicate a split between those cultures where ‘face’ is an important, if not
central, explanatory key to interpersonal behaviour and those where ‘face’ takes
second place to seemingly more dominant notions such as discernment,respeto
(GarcıLa, 1996) and deference. Socially stratified societies where normative ‘politeness’
is dominant (e.g. Japan, Mexico and the Zulu in South Africa) can be contrasted
to ‘face’ and status-based societies such as China and Korea, where both
normative and strategic ‘politeness’ are present. Finally, in less hierarchical societies
such as the northern European and the North American ones, status is allegedly far
less marked in verbal and non-verbal interaction, and normative politeness is therefore
much less in evidence, but the concern with ‘face’ seems to be as relevant as it
was four decades ago, when Goffman first wrote his essays.
It has been suggested that other factors must be considered in a culture-situated
understanding of ‘face’ and its dynamics: personal values, one’s own self-concept,
self-identity in various groupings, role expectations and normative constraints
(Earley, 1997: 95–6). Among the well-known universal dimensions of cultural
variation (individualism–ollectivism, power distance, masculinity–emininity, relationship
with nature) one other dimension, shame v. guilt, may account for the
dominant role of the controlling and sanctioning groups (e.g. family, fellow workers)
on face-work. In guilt-based societies, for instance, face-work will be affected by
one’s sense of individual responsibility and internal moral standards (Earley, 1997:
139). Early’s conclusions point to the need to understand and compare cultural
conceptualizations of the social self and its relationship to others as an alternative
and possibly more fruitful way of studying the relevance and dynamics of ‘face’ and
‘face-work’ in interpersonal contacts.
6. Goffman’ ‘ace-work’and politeness
In this section, I want to re-visit the notion of ‘face-work’, often (and, I think,
incorrectly) treated as synonymous with (linguistic) politeness in research inspired
by Brown and Levinson (e.g.Lim and Bowers, 1991; Holtgraves, 1992; but see
Watts, 1989). For Goffman, ‘‘face-work’’ has to do with self-presentation in social
encounters, and although individual psychology matters, it is the interactional order
that is the focus of Goffman’s study. Goffman’s ideal social actor is based on a
Western model of interactant, almost obsessively concerned with his own self-image
and self-preservation. The cult language of the opening quotation is an unmistakable
signal of the importance Goffman attaches to the individual personality and its
needs. However, forGoffman (1967: 7) ‘face’ is much more than just verbal behaviour:
‘‘At such times [in interpersonal contact] the person’s face clearly is something
that is not lodged in or on his body, but rather something that is diffusely located in
the flow of events in the encounter [. . .]’’.
Similarly, ‘‘face-work’’ consists of ‘‘the actions taken by a person to make whatever
he is doing consistent with face’’ (Goffman, 1967: 12), as well as verbal behaviour.
‘‘Face-saving practices’’, which for Goffman seem to be equivalent to ‘‘facework’’,
are different for every individual, group or society, though they may be
drawn from a possible common framework. (Goffman, 1967: 13). They are of two
main types: ‘‘defensive’’ (of one’s own ‘face’), and ‘‘protective’’ (of other’s ‘face’),
although ‘‘aggressive face-work’’ is also discussed with reference to ‘‘making points’’
(Goffman, 1967: 24–6). At the end of an essay in which he had set out to identify
the ‘‘natural units of interaction’’ (Goffman, 1967: 1), Goffman finds that interpersonal
behaviour is governed by moral rules imposed on a social actor from outside.
These rules affect self- and other-evaluation, emotional display and ritual
practices (Goffman, 1967: 45). Interpersonal behaviour thus defined appears to
incorporate the notions of ‘‘face’’ (evaluation and emotional response) and ‘‘facework’’
(ritual practices). If ‘‘face-work’’ is an integral aspect of communication,
should it be considered equivalent to ‘politeness’, as much post-Brown and Levinson
The perspective on ‘politeness’ outlined here is seen as inextricably connected with
the social order. As a point of departure, I takeWatts et al.’s (1992) twofold notion of
‘‘first-order politeness’’ and ‘‘second-order politeness’’ as a bridge between research
on verbal and non-verbal politeness (or first order politeness) and what I would define
asthe regulatory dynamics of social order (second order politeness). Watts et al.’s
(1992: 3)own definitions are helpful here: ‘‘We take first-order politeness to correspond
to the various ways in which polite behaviour is perceived and talked about by
members of socio-cultural groups. It encompasses, in other words, commonsense
notions of politeness. Second-order politeness, on the other hand, is a theoretical
construct, a term within a theory of social behaviour and language usage’’.
Despite the variety of studies which focus on linguistic politeness (see, for
instance, the bibliography compiled byDuFon et al., 1994), the field still lacks an
agreed definition of what ‘politeness’ is. Notions of politeness as rational, goaloriented
behaviour (Haverkate, 1988), ‘‘politic behavior’’ (Watts, 1992: 50), or
appropriate behaviour (Meier, 1995) all represent attempts to pin down a complex
phenomenon that intuitively extends well beyond its linguistic manifestations. In the
meantime, politeness typologies (Kasper, 1990) and perspectives on politeness (Fraser,
1990) have been proposed. Politeness has also been elevated to the status of
‘theory’ (Coupland et al., 1988; Holtgraves and Yang, 1990; Arundale, 1999; but see
Kwarciak, 1993for an alternative view). More recently, doubts have been cast on
the need of a ‘model of politeness’ on the grounds that ‘‘[l]inguistic phenomena are
necessarily embedded within a larger framework of social interaction and must also
be explained therein’’ (Meier, 1995: 390), a position forcefully defended byEelen
(2001)in his state-of-the art monograph on politeness theories.
Most research within the last decade or so is primarily concerned with one of the
two types of ‘‘first order politeness’’, namely linguistic politeness. Brown and
Levinson’s own discussion of politeness does not exclude non-verbal behaviour, but
concentrates on linguistic strategies. Similarly, although at times Goffman’s terminology
may be confusing, it seems possible in the end to distinguish between facesaving
practices as ‘the rules of interaction’ and face-work as ‘the actions taken by a
person to make whatever he is doing consistent with face’ (Goffman, 1967: 12).
‘‘face-work’’, according to Goffman, has to do with self-presentation in social
encounters which is dynamically realised in the interactional order.
1464F. Bargiela-Chiappini / Journal of Pragmatics 35 (2003) 1453–469
‘‘Face-work’’, then, deals with norms beyond linguistic and para-linguistic politeness.
Social norms, conventions and expectations undergird its workings in much
the same manner as ‘social behaviour and language usage’ underpin ‘‘second-order
politeness’’. The notion of ‘face’ embedded in ‘‘face-work’’ poses serious problems,
however, if one attempts to introduce this rich construct into a discussion of the
relationship between social order and interaction, as ‘face’ has become a term with a
great deal of theoretical and cultural baggage. For this reason, I propose to use the
label ‘polite behaviour’ instead, and seek to widen and deepen Watts et al.’s definition
of ‘‘second-order politeness’’. I also take on board the scope of Watts et al.’s
enquiry, according to which, ‘[I]n studying politeness, we are automatically studying
social interaction and the appropriacy of certain modes of behaviour in accordance
with socio-cultural conventions’ (Watts et al., 1992: 6). Quite appropriately, I think,
the authors also call for interdisciplinary research as a pre-requisite to capture the
dynamic and context-dependent nature and workings of a concept that remains
elusive (Watts et al., 1992: 10–1).
7. ’olite behaviour’Social-embeddedness and dynamism are two features of ‘polite behaviour’ that the
literature on first-order politeness has treated as given but which need revisiting, as
they illuminate aspects of the debate which have been neglected, and which could
provide the key for a new understanding of the nature of ‘polite behaviour’ and its
place within a theory of social behaviour.
Werkhofer (1992: 191)remarks that politeness exercises both enabling and constraining
functions on behaviour and that a commonly agreed upon and accepted
framework of rights and duties would permit the disabling functions to be neutralised.
The important corollaries of (a) co-existing multiple social identities and (b)
the clash between the rights and duties of these different identities (ibid.) afford a
multi-layered and dynamic conceptualisation of social identities acting within a
moral order that is partly interactionally-constructed and partly regulated by
mutually agreed norms and conventions. Accordingly, ‘polite behaviour’ is a multifacetted
social phenomenon that originates within the moral order.
In turn, a critical appreciation of the construct of ‘culture’ becomes necessary, as
various ‘ideologies’ of politeness are often seen to emanate from specific cultural
settings.8 Eelen (1999: 167) notes that ‘‘[I]t is rather a question of epistemology, of
heuristics, indeed of philosophy, our place as scientists vis-a` -vis ‘everyday life’ [. . .]
our role in defining that reality’’. We cannot escape commitment to an interpretative
framework if we wish to make progress in understanding, which also means that we
need to re-visit concepts such as ‘culture’ and ‘society ‘, ‘group’ and ‘individual’ on
which to build our argument, pace their inevitable ideological nature. In fact, it is
Eelen (1999: 170)himself who provides another useful insight on politeness as an
‘‘inherently ethical’’ phenomenon. In his monograph,Eelen (2001) re-states his
understanding of politeness as a phenomenon that is embedded in social reality and
therefore requires a multi-pronged analytical approach that only multi-disciplinarity
can afford to provide. Significantly, he also brings back individual agency to be the
shaping force behind cultural and societal change.
It is from human values that norms, including ‘politeness norms’, are derived which
constitute the social order expressed in interpersonal and inter-group encounters. It
could be argued that in the post-modern age, the flux of change has swept moral codes
away. And yet, the crisis of ethics has not resulted in the loss of morality. In Zigmunt
Bauman’s (1995: 43, original emphasis)words: ‘‘Postmodernity is an ‘era of morality’
in one sense only:thanks to the ‘disocclusion’—he dispersal of ethical clouds which
tightly wrapped and obscured the reality of moral self and moral responsibility—it is
possible now, nay inevitable, to face the moral issues point-blank. In all their naked
truth, as they emerge from the life experience of men and women, and as they confront
moral selves in all their irreparable and irredeemable ambivalence’’
The nature of ‘polite behaviour’ is then grasped through the system of rules, conventions,
expectations that govern social encounters, or rather, through Goffman’s
‘‘dissonance’’, or impolite behaviour. The moral nature of ‘polite behaviour’, embedded
as it is in the social order (in fact, possibly the canvas of such order), is reminiscent
of the Durkheimian ‘‘moral act’’, which for him is obligatory but also ‘‘desired’’ and
‘‘desirable’’ (Steiner, 2000: 82). These cognitive attributes of the moral act seem to me
tantalising pointers to the possible affective dimension of ‘polite behaviour’.
In the so-called Western societies, strategic politeness reflects the paramount concern
for individual rights, i.e. what is owed to the individual, whereas in many non-
Western societies, normative or indexical politeness signals a concern for duty, what
is owed to the group.Moghaddam et al. (2000: 276) write about ‘‘an age of rights of
both positive and negative valence’’, a definition that reflects Brown and Levinson’s
positive and negative face. Duties have not disappeared, they argue, they have only
been re-interpreted, so that from the earlier orientation of duties towards the group,
we have slowly moved towards duties to self. This rights-based social order is constructed
around and for individuals. It is the manifestation of an individualistic
ontology, as opposed to a ‘‘communitarian ontology’’ that rests on a group-based
society (Moghaddam et al., 2000: 297). A theory of rights and duties illuminates the
individualism-collectivism dimension of cultural variation (Triandis, 1995), often
used as an explanation for ‘politeness’, and provides a normative explanation for it.
The normative nature of politeness rules feeds on the moral order within which
encounters take place and the ontology of which tends to be either individualistic or
communitarian. The concepts of ‘right’ and ‘duty’ are powerful interpretative tools
for an understanding of self-other relationships (Bhatia, 2000: 306) and for an
explanation of behaviours that have been subsumed under the label of ‘politeness’.
The various forms of ‘polite behaviour’ reflect society’s emphasis on right or duty.
Within this broader, ethical framework, ‘‘second-order politeness’’ and its surface
manifestations (‘‘first-order politeness’’) emerges as more than pragma-linguistic
behaviour. Its roots go deep into the history and moral constitution of a society and
as such require more than just attention to verbal and non-verbal manifestations. Its
origins and workings are woven into the social fabric of interpersonal behaviour and
only multidisciplinary research can hope to shed further light on them.
This article has revisited some of the most heavily debated concepts in linguistics
and pragmatics research over the last three decades, namely ‘face’, ‘face-work’ and
‘politeness’. The core of the argument was a critical re-examination of some important
aspects of Erving Goffman’s contribution to politeness theory that had been
either neglected or only partly explored in previous work. The discussion concludes
by underlining the central roˆle played by ‘face’ in the ritual dynamics of a rule-governed
moral order which I have called ‘polite behaviour’.
A future, much needed multi-disciplinary enquiry into the nature of face and impolite
behaviour is sure to lead linguists, pragmaticians, sociologists and social psychologists
to share insights on the very essence of the social order.
Arundale, Robert, 1999. An alternative model and ideology of communication for an alternative politeness
theory. Pragmatics 91, 119–54.
Baxter, Leslie A., 1984. An investigation of compliance-gaining as politeness. Human Communication
Research 103, 427–56.
Bauman, Zigmunt, 1995. Life in Fragments. Blackwell, Oxford.
Bhatia, Sunil, 2000. Can we return to the concept of duty in a culture of rights? Implications for morality
and identity. Culture & Psychology 63, 303–16.
Branaman, Ann, 1997. Goffman’s social theory. In: Lemert, C., Branaman, A. (Eds.), The Goffman
Reader. Blackwell, Malden, MA, pp. xlv–xxxii.
Brown, Penelope, Levinson, Stephen, 1978. Universals in language usage: politeness phenomena. In:
Goody, E. (Ed.), Questions and Politeness: Strategies in Social Interaction. Cambridge University
Press, Cambridge, pp. 56–89.
Brown, Penelope, Levinson, Stephen, 1987. Politeness: Some Universals in Language Usage. Cambridge
University Press, Cambridge.
Chang, Hui-Ching, 1999. The ‘well-defined’ is ‘ambiguous’—ndeterminacy in Chinese conversation.
Journal of Pragmatics 31, 535–56.
Chen, Ping, Gu, Yueguo, 1997. Introduction. Text 174, 431–33.
Chen, Rong, 1993. Responding to compliments. A contrastive study of politeness strategies between
American English and Chinese speakers. Journal of Pragmatics 20, 49–5.
F. Bargiela-Chiappini / Journal of Pragmatics 35 (2003) 1453–4691467
Coupland, Nikolas, Grainger, Karen, Coupland, Justine, 1988. Politeness in context: intergenerational
issues Review article. Language in Society 17, 253–62.
de Kadt, Elizabeth, 1998. The concept of face and its applicability to the Zulu language. Journal of
Pragmatics 292, 173–91.
Ditton, Jason (Ed.), 1980. The View from Goffman. Macmillan, London and Basingstoke.
Drew, Paul, Wootton, Athony, Goffman, Erving, 1988. Exploring the Interaction Order. Polity Press,
DuFon, Margaret A., Kasper, Gabriele, Takahashi, Satomi, Yoshinaga, Naoko, 1994. Bibliography on
linguistic politeness. Journal of Pragmatics 21, 527–78.
Durkheim, Emile, 1803. La Division du Travail. PUF, Paris.
Durkheim, Emile, 1915. The Elementary Forms of the Religious Life. George Allen & Unwin Ltd, London.
Durkheim, Emile, 1924. Sociologie et Philosophie Bougle´ C. PUF, Paris.
Earley, P. Christopher, 1997. Face, Harmony and Social Structure. An Analysis of Organizational Behavior
across Cultures. Oxford University Press, New York.
Eelen, Gino, 1999. Politeness and ideology: a critical review. Pragmatics 91, 163–74.
Eelen, Gino, 2001. A Critique of Politeness Theories. St John’s Publishers, Manchester.
Ehlich, Konrad, 1992. On the historicity of politeness. In: Watts, R.J., Ide, S., Ehlich, K. (Eds.), Politeness
in Language. Studies in Its History, Theory and Practice. Mourton de Gruyter, Berlin and New
York, pp. 71–08.
Fraser, Bruce, 1990. Perspectives on politeness. Journal of Pragmatics 14, 219–36.
Gao, Ge, 1996. Self and OTHER: a Chinese perspective on interpersonal relationships. In: Gudykunst,
W.B., Ting-Toomey, S., Nishida, T. (Eds.), Communication in Personal Relationships Across Cultures.
Sage, Thousand Oaks, pp. 81–01.
GarcıLa, Wintilo, 1996.Respeto: a Mexican base for interpersonal relationships. In: Gudykunst, W.B.,
Ting-Toomey, S., Nishida, T. (Eds.), Communication in Personal Relationships across Cultures. Sage,
Thousand Oaks, pp. 122–36.
Goffman, Erving, 1967. Interaction Ritual. Essays on Face-to-Face Behavior. Allen Lane: The Penguin
Gu, Yueguo, 1990. Politeness phenomena in modern Chinese. Pragmatics 237–57.
Haverkate, Henk, 1988. Toward a typology of politeness strategies in communicative interaction. Multilingua
Herna´ ndez-Flores, Nieves, 1999. Politeness ideology in Spanish colloquial conversation: the case of
advice. Pragmatics 91, 37–9.
Ho, D.Y-F., 1994. Face dynamics: from conceptualization to measurement. In: Ting-Toomey, S. (Ed.),
The Challenge of Facework: Cross-Cultural and Interpersonal Issues. State University of New York
Press, Albany, pp. 3–3.
Holtgraves, Thomas, 1992. The linguistic realization of face management: implications for language production
and comprehension, person perception, and cross-cultural communication. Social Psychology
Quarterly 552, 141–59.
Holtgraves, Thomas, Yang, Joong-Nam, 1990. Politeness as universal: cross-cultural perceptions of
request strategies and inferences based on their use. Journal of Personality and Social Psychology 594,
Hu, Hsien Chin, 1944. The Chinese concepts of ‘‘face’’. American Anthropologist 46, 45–4.
Hwang, Kwang-kuo, 1987. Face and favor. The Chinese power game. American Journal of Sociology
Ide, Sachiko, 1989. Formal forms and discernment: two neglected aspects of universal linguistic politeness.
Multilingua 82/3, 223–48.
Ji, Shaojun, 2000. ‘Face’ and polite verbal behaviors in Chinese culture. Discussion note. Journal of
Pragmatics 32, 1059–062.
Kasper, Gabriele, 1990. Linguistic politeness. Current research issues. Journal of Pragmatics 14, 193–18.
Kwarciak, Barbara J., 1993. The acquisition of linguistic politeness. Multilingua 121, 51–8.
Lemert, Charles, 1997. ‘‘Goffman’’. In: Lemert, C., Branaman, A. (Eds.), The Goffman Reader. Blackwell
Publishers, Malden, MA, pp. ix-xliii.
1468F. Bargiela-Chiappini / Journal of Pragmatics 35 (2003) 1453–469
Lim, Tae-Seop, 1994. Facework and interpersonal relationships. In: Ting-Toomey, S. (Ed.), The Challenge
of Facework: Cross-Cultural and Interpersonal Issues. State University of New York Press,
Albany, pp. 209–29.
Lim, Tae-Seop, Choi, Soo Hyang, 1996. Interpersonal relationships in Korea. In: Gudykunst, W.B., Ting-
Toomey, S., Nishida, T. (Eds.), Communication in Personal Relationships Across Cultures. Sage,
Thousand Oaks, pp. 122–36.
Lim, Tae-Seop, Bowers, Waite, 1991. Facework. Solidarity, approbation and tact. Human Communication
Research 173, 415–50.
Ma, Ringo, 1966. Saying ‘‘yes’’ for ‘‘no’’ and ‘‘no’’ for ‘‘yes’’: a Chinese rule. Journal of Pragmatics 25,
Manning, Philip, 1992. Erving Goffman and Modern Sociology. Polity Press, Cambridge.
Mao, LuMing Robert, 1994. Beyond politeness theory: ‘face’ revisited and renewed. Journal of Pragmatics
Meier, A.J., 1995. Passages of politeness. Journal of Pragmatics 24, 381–92.
Moghaddam, Fathali M., Slocum, Nikki R., Finkel, Norman, Mor, Tzili, Harre´ , Rom, 2000. Toward a
cultural theory of duties. Culture & Psychology 63, 275–02.
Placencia, MarıLa E., 1996. Politeness in Ecuadorian Spanish. Multilingua 151, 13–4.
Rathmayr, Renate, 1999. Metadiscours et realite´ linguistique: l’example de la politesse russe. Pragmatics
Stedman Jones, Susan, 2000. Durkheim Reconsidered. Polity Press, Cambridge.
Steiner, Philippe, 2000. La Sociologie de Durkheim. La De´couverte, Paris.
Terkourafi, Marina, 1999. Frames for politeness: a case study. Pragmatics 91, 97–18.
Ting-Toomey, Stella, Kurogi, Atsuko, 1998. Facework competence in intercultural conflict: an updated
face-negotiation theory. International Journal of Intercultural Relations 222, 187–25.
Triandis, Harry, 1995. Individualism and Collectivism. Westview Press, Boulder.
Watts, Richard J., 1989. Relevance and relational work: linguistic politeness as political behavior. Multilingua
Watts, Richard J., 1992. Linguistic politeness and politic verbal behaviour: reconsidering claims for universality.
In: Watts, R.J., Ehlich, K. (Eds.), Politeness in Language: Studies in Its History, Theory and
Practice. Mouton de Gruyter, Berlin and New York, pp. 43–0.
Watts, Richard J., Ide, Sachiko, Ehlich, Konrad, 1992. Introduction. In: Watts, R.J., Ehlich, K. (Eds.),
Politeness in Language: Studies in Its History, Theory and Practice. Mouton de Gruyter, Berlin and
New York, pp. 1–0.
Werkhofer, Konrad T., 1992. Traditional and modern views: the social constitution and the power of
politeness. In: Watts, R.J., Ehlich, K. (Eds.), Politeness in Language: Studies in Its History, Theory and
Practice. Mouton de Gruyter, Berlin and New York, pp. 155–99.
Wilson, Steven R., Kim, Min-Sun, Meischke, Hendrika, 1991/1992. Evaluating Brown and Levinson’s
politeness theory: a revised analysis of directives and face. Research on Language and Social Interaction
Francesca Bargiela-Chiappiniis senior research fellow in linguistics in the department of English and
Media Studies, Nottingham Trent University. Among her works, Managing Language: The Discourse of
Business Meetings(Benjamins, 1997) and The Languages of Business (EUP, 1997) both with S. Harris; and
Writing Business(Longman, 1999) with C. Nickerson. She is guest editor of three special journal issues: on
Business Discourse for theInternational Review of Applied Linguistics (October 2002), on Intercultural
Business Communication for theJournal of Intercultural Studies (forthcoming, 2003) and on Organizational
Discourse for theInternational Journal of the Sociology of Language (forthcoming, 2004).
F. Bargiela-Chiappini / Journal of Pragmatics 35 (2003) 1453–4691469
Read our customer feedbacks