Politeness: Is there an East-West divide?

GEOFFREY LEECH

Abstract
Brown and Levinson (1987 [1978]) has remained the most seminal and
influential starting point for studying cross-cultural and interlinguistic
politeness. Yet it has also provoked countervailing arguments (e. g., Ide
1989; 1993; Matsumoto 1989; Gu 1990; Mao 1994), claiming a Western
bias in Brown and Levinson’s model, particularly in their construal of
the concept of ‘face’, in their overemphasis on face-threat and their as-
sumption of individualistic and egalitarian motivations, as opposed to
the more group-centred hierarchy-based ethos of Eastern societies. This
leads to the question: Is there an East-West divide in politeness?
This article presents a pragmatic framework for studying linguistic
politeness phenomena in communication: a common principle of polite-
ness (Leech, 1983; 2005) and a Grand Strategy of Politeness (GSP),
which is exemplified in common linguistic behaviour patterns in the per-
formance of polite speech acts such as requests, offers, compliments,
apologies, thanks, and responses to these. The GSP says simply: In order
to be polite, a speaker communicates meanings which place (a) a high
value on what relates to the other person (typically the addressee), and
(b) a low value on what relates to the speaker. It is clear from many
observations that constraint (a) is more powerful than constraint (b).
The following hypothesis will be put forward, and supported by evi-
dence from four languages: that the GSP provides a very general expla-
nation for communicative politeness phenomena in Eastern languages
such as Chinese, Japanese and Korean, as well as in Western languages
such as English. Since politeness deals with scalar phenomena, this is
not to deny the importance of quantitative and qualitative differences in
the settings of social parameters and linguistic parameters of politeness
in such languages. A framework such as the GSP provides the parame-
ters of variation within which such differences can be studied.

Journal of Politeness Research 3 (2007), 167–206 1612-5681/07/003–0167

DOI 10.1515/PR.2007.009 © Walter de Gruyter

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
168 Geoffrey Leech

Hence this article argues in favour of the conclusion that, despite man-
ifest differences, there is no East-West divide in politeness.

Keywords: politeness, pragmalinguistics, socio-pragmatics, face, Chinese,

Japanese, Korean

1. Introduction
Brown and Levinson’s seminal treatment of politeness, reissued as a
monograph in 1987 (and henceforth abbreviated to B&L) has remained
the most frequently cited publication on language and politeness. In-
deed, since its publication, in spite of heavy criticism, it has held its
ground as the model that other writers turn to as the starting-point
of their own research perspective. My own treatment of politeness in
Principles of Pragmatics (1983) (abbreviated as POP) has also often
been bracketed with B&L as a pioneering, essentially Gricean, treat-
ment of politeness and has been criticized in a similar way.
B&L have been criticized on numerous grounds, and this itself is
something of a tribute to B&L: if it did not have the virtue of providing
an explicit and detailed model of linguistic politeness, it could not have
been attacked so easily. But in keeping with my title, I will here focus
on one major criticism of B&L. It has been argued that B&L’s model
has a Western, or even Anglophone, bias, and therefore cannot claim
to present a universal theory applicable to all languages and cultures.
This Western bias has been argued on a number of levels. On one level
it has been claimed that B&L’s definition of negative politeness in
terms of negative and positive face reflects an Anglo-Western view of
the supremacy of an individual’s desires and right to freedom:

Negative face: the basic claim to territories, personal preserves,

rights to non-distraction – i. e., to freedom of action and freedom
from imposition.
Positive face: the positive consistent self-image or “personality”
(crucially including the desire that this self-image be appreciated
and approved of) claimed by interactants. (B&L 1987: 61)

In particular, it has been strongly maintained that the B&L focus on

the individual, however appropriate to the West, is quite inappropriate
to the group orientation of Eastern cultures, specifically those of China
and Japan (Matsumoto 1989; Gu 1990; Wierzbicka 1991 [2003]; Mao
1994; Ide 1993). On another level, closely related to this, B&L have
been criticized for their above definition of face (subdivided into nega-
tive and positive face), which is at variance both with Goffman’s “face”,

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
 Politeness: Is there an East-West divide? 169

claimed as the source of B&L’s concept, and more importantly with

the Chinese conceptions of “face” (miànzi and liǎn) from which the
Western conception of “face” historically derives. On a third level, B&
L are criticized for their explanation of the whole of politeness as a
palliative to face-threatening acts (FTAs), for the performance of which
B&L famously specify five strategies, ranging from “bald on-record”
performance of the FTA through indirect strategies to its non-perform-
ance. This emphasis on FTAs not only seems to reinforce the Western
orientation of B&L’s model and hence its restriction to a particular
cultural milieu, but also strikes at least one Western commentator as
putting forward a “paranoid” view of Western society (Schmidt 1980:
104, cited in Mao 1994: 456). By rejecting the universalist claim of
B&L, the “Eastern” critique of B&L as biased towards Western values
has appeared to align itself increasingly with a cultural relativism; a
perspective most forcibly championed by Wierzbicka (1991 [2003]). For
example, in her introduction to her 2nd edition, Wierzbicka says:

… since this book was first published the field of cross-cultural

pragmatics has advanced enormously … this progress has not only
not made my 1991 Cross-Cultural Pragmatics dated, … on the con-
trary, its tenets and its overall approach have been essentially vin-
dicated. A decade ago, the “pragmatic” scene was still largely
dominated by the search for the “universals of politeness” and
for the “universal maxims of conversation”. The widely accepted
paradigms were those of Brown and Levinson’s “pan-cultural in-
terpretability of politeness phenomena” (1978: 288), and Grice’s
(1975) theory of conversation … It is heartening to see to what
extent the situation has now changed.

Against the previous Gricean orthodoxy, Wierzbicka espouses “the

idea that interpersonal interaction is governed, to a large extent, by
norms which are culture-specific and which reflect cultural values cher-
ished by a particular society”. Later, she paints me in the same univer-
salist colours as B&L, referring to:

the once popular assumption that the “principles of politeness”

are essentially the same everywhere and can be described in terms
of “universal maxims” such as those listed in Leech (1983: 132) …

Actually, I never used the words in double quotes in this passage: “prin-
ciples of politeness” (in the plural) and “universal maxims”. In fact, I
never made any claim for the universality of my model of politeness,
although I did in the conclusion of the book express the expectation

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
170 Geoffrey Leech

“that the general paradigm presented in these chapters will provide a

framework in which contrastive studies of pragmalinguistic strategies
can be undertaken”.
The problem with writing a “pioneering” work can be that it is cited
by people who have not read it (at least not recently), but have read
citations and discussions of it by other people. The result can be that
the work becomes caricatured as representing a particular oversimpli-
fied position in the debate. In the worst cases you become an “Aunt
Sally” – and Wierzbicka is not the only offender in this respect1. To be
fair to B&L, they have also been caricatured to some extent: although
they did commit themselves to a universalist position, they also empha-
sized the dimensions of cross-cultural/linguistic variation:
The essential idea is this: interactional systematics are based
largely on universal principles. But the application of the princi-
ples differs systematically across cultures, and within cultures
across subcultures, categories and groups. (B&L 1978: 288)
My own position is not very different from this, although I would not
press for “universal principles” (see 9.1).
At this stage, then, let me give a preview of my answer to the ques-
tion posed by my title: no, there is no absolute divide between East and
West in politeness. Consider the concepts of “collective, group culture”
(East) and “individualist, egalitarian culture” (West). These are not
absolutes: they are positions on a scale. All polite communication im-
plies that the speaker is taking account of both individual and group
values. In the East, the group values are more powerful, whereas in
the West, individual values are.
Consider, as another false dichotomy, the distinction between “uni-
versalism” (the position espoused by B&L) and “relativism” (the po-
sition adopted by Wierzbicka). An absolute universalist position is
clearly untenable: it is obvious, from common experience as well as
from studies over the past twenty years, that politeness manifests itself
in different terms in different languages and cultures. On the other
hand, a completely relativist position is equally untenable. If there were
not a common pattern shared by different languages/cultures, it would
be meaningless to apply a word like “politeness” or “face” to different
cultures, and there would be no point in having an international Journal
of Politeness Research.

2. Background: Back to the 1980s

To argue the above position more fully, I will revisit the Gricean basis
for understanding polite behaviour, while presenting (though space is

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
 Politeness: Is there an East-West divide? 171
Table 1. Abbreviations used in this article.
B&L = P. Brown and S. Levinson, Politeness: Some Universals of Language Usage
(see references)
POP = G. Leech, Principles of Pragmatics (see references).
FTA = face threatening act; GSP = grand strategy of politeness
CP = cooperative principle; PP = politeness principle.
S = speaker, self; H = hearer; O = other (i. e., not the speaker)

limited) a revision of the politeness framework of Leech (1983: chs. 5

and 6)2.
I will begin by referring to the two Grice-inspired works on polite-
ness in communication, B&L and POP.

2.1. Criticisms of B&L (1978, 1987) and sometimes also

of Leech’s POP (1983)
The following are among the criticisms that have been levelled against
the above “pioneers”.

a) These treatments of politeness have a Western bias: they emphasize

an individualist ethos versus group orientation (Ide 1989; Matsu-
moto 1989; Wierzbicka 1991 [2003]; Koutlaki 2002). Thus the cor-
nerstone of politeness theorizing, for B&L, has been their individ-
ual-wants concept of face: B&L (1978: 66) restate the components
of face as:
Negative face: “want of every ‘competent adult member’ [of a soci-
ety] that his actions be unimpeded by others.”
Positive face: “want of every member that his wants be desirable to
at least some others.” (B&L 1987: 62)

This fits a Western bias towards individualism: the individual has

rights and wants which need to be respected and indulged; the indi-
vidual is entitled to assert those rights and wants, unless they inter-
fere too much with others’. This concept does not fit the traditional
Eastern (e. g., Chinese, Japanese, Korean) ethos of identifying with
the group, in which each person has a place defined by obligations
and rights in relation to superiors, equals and inferiors.

b) Thus, according to Mao (1994), B&L’s concept of “face” is quite

unlike the Chinese concepts of miànzi and liǎn, and B&L’s concept

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
172 Geoffrey Leech

of politeness does not match the Chinese concept of lı̌mào – ex-

plained by Gu (1990). Similarly, according to Ide (1993), the Japa-
nese concept of wakimae or “discernment”, neglected by B&L, is
needed to explain Japanese socially-constrained politeness or tei-
neisa (cf. the related Korean concept of gongsonham).
c) Because of their focus on face as just defined, B&L give undue
prominence, as just noted, to face threatening acts (FTAs) and in-
deed their whole understanding of politeness concerns the strategies
for dealing with FTAs.
d) B&L distinguish three factors that determine the strength of the
FTA, and hence the politeness strategy needed: P (power), D (social
distance), and R (ranked size of the imposition). There are, how-
ever, different interpretations of the axes of vertical and horizontal
distance (P and D) according to Eastern and Western culture, such
that “power” is a less appropriate term. For example, Spencer-Oatey
(1993) shows that Chinese postgraduate students, compared with
British postgraduates, interpret their relation to their tutors differ-
ently. They see their tutors as closer socially, but more superordinate
in terms of respectfulness on the P axis than do British postgraduate
students. They expect to show deference to their teachers, and yet
to have a close, friendly relation with them – somewhat like a parent
or uncle3.

There have also been criticisms of B&L’s famous formula for calculat-
ing the strength of an FTA: Wx = D(S,H) + P(H,S) + Rx in terms the
three factors D (distance), P (power) and R (rank of imposition). As
B&L themselves discuss (1987: 16), other findings suggest that this for-
mula is too simple. However, I will not discuss this further here.

2.2. Criticisms of Principles of Pragmatics by Brown & Levinson

and others
These include:
a) Leech (1983) takes a maxim-based approach following the maxims
of Grice’s Cooperative Principle (CP). This approach may now be
considered dated: the four Maxims of Quantity, Quality, Relation
and Manner have been criticized for being unclear, overlapping, or
of different statuses (see e. g., Thomas 1995: 168).
b) Leech has too many maxims: the Maxim of Tact, Maxim of Gener-
osity, Maxim of Modesty, etc. – six in all.

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
 Politeness: Is there an East-West divide? 173

B&L complain:
If we are permitted to invent a maxim for every regularity in lan-
guage use, not only will we have an infinite number of maxims, but
pragmatic theory will be too unconstrained to permit the recogni-
tion of any counterexamples (1987: 4)4.
c) Also Leech’s Principles of Pragmatics is like B&L in being Gricean
and biased towards Western values. Most of its examples are from
English, and it gives too much attention to the “tact maxim”, which
chiefly concerns the minimizing of the imposition made on others
in directives – corresponding to B&L’s canonical case of FTAs.

3. Restatement of the treatment of politeness in

Principles of Pragmatics (POP)
As nearly 25 years have passed since POP was published, it is appropri-
ate for me to attempt a restatement, taking account of research on
politeness which has taken place in the mean time.

3.1. The Principle of Politeness

The Principle of Politeness (PP) – analogous to Grice’s CP – is a con-
straint observed in human communicative behaviour, influencing us to
avoid communicative discord or offence, and maintain communicative
concord. What I mean by “communicative discord” is a situation in
which two people, x and y, can be assumed, on the basis of what mean-
ings have been communicated, to entertain mutually incompatible
goals. (Such discord can spill over into more threatening forms of dis-
cord, such as physical conflict.) For example, x has a state of affairs E
as a goal, and y has a state of affairs not-E as a goal. Concord is the
opposite of discord: where both participants explicitly or implicitly pur-
port to pursue the same goals. But both discord and concord are scalar
phenomena, in terms of their degree and significance. Note that polite-
ness itself is scalar. It is also an aspect of goal-oriented behaviour; to
say that S is being “polite” in using a particular utterance is to say that
S’s goal in using that utterance is, in some degree, to convey an impres-
sion of politeness to H. But also politeness is not a matter of the “real”
motives of human agents, which may be other than appears on the
surface. Politeness is concerned with avoiding discord and fostering
concord, only in so far as these are evident in communication, espe-
cially through what meanings are expressed or implicated.
But of course the PP is not always in operation: we can be impolite
as well as polite. In addition, much of our communicative behaviour is

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
174 Geoffrey Leech

neither polite nor impolite: leaving aside honorific systems, the PP

needs to be invoked only for “transactional” discourse (see Section 7),
where as part of the illocutionary force or the perlocutionary effect,
some element of value is transacted between S and O.

3.2. Two kinds of politeness scale

There are two ways of looking at politeness:
(i) Semantic (or absolute) politeness scale: We can order utterances
on a scale of politeness out of context. For example, out of con-
text, on an absolute scale of politeness, we can judge that Can
you help me? is more polite, as a request, than Help me, and is
less polite than Could you possibly help me?. There is a semantic
reason for this: in a “default”5 sense, the more a request offers
choice to H, the more polite it is. Similarly, Thank you very much
is more polite than Thanks, because it intensifies an expression
of gratitude, rather than expressing gratitude in a minimal way6.
This scale is unidirectional, and registers degrees of politeness in
terms of the lexigrammatical form and semantic interpretation of
the utterance.
(ii) Pragmatic (or relative) politeness scale: This is politeness relative
to norms in a given society, group, or situation. Unlike the abso-
lute scale, it is sensitive to context, and is a bi-directional scale.
Hence it is possible that a form considered more polite on the
absolute politeness scale is judged less polite relative to the norms
for the situation. For example, Could I possibly interrupt? could
be understood as “too polite”, say, if spoken to family members
monopolizing the conversation: it would probably be interpreted
as ironic (see 5.3). The pragmatic politeness scale registers “over-
politeness” and “underpoliteness”, as well as “politeness appro-
priate to the situation”.

3.3. Illocutionary goals and social goals

In POP, certain maxims (such as the Tact Maxim and the Modesty
Maxim) represented the goals people pursue in order to maintain com-
municative concord. The POP approach (which I still broadly adhere
to, although I now avoid the term “maxim”)7 is a goal-oriented ap-
proach. It is assumed that we have some illocutionary goals, i. e., the
primary goals we want to achieve in linguistic communication (in ask-
ing permission, giving advice, etc.). We also have social goals, i. e., main-
taining good communicative relations with people. But illocutionary

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
 Politeness: Is there an East-West divide? 175

goals may either support or compete with social goals – especially the
goal of being (to some degree) polite. Thus in paying a compliment,
one’s illocutionary goal is to communicate to H one’s high evaluation
of H or of some attribute of H. Here the illocutionary goal supports a
social goal (saying something polite, in order to maintain good rela-
tions). But in a request, or a criticism of H, the illocutionary goal com-
petes, or is at odds, with that social goal. Both these kinds of utterance
involve politeness, and I will distinguish these by calling them respec-
tively pos-politeness and neg-politeness.
These abbreviations are meant to be a warning that these are not
quite the same as what B&L understand by “positive politeness” and
“negative politeness”. The kind of politeness involved in paying a com-
pliment is pos-politeness (having a positive import of raising the estima-
tion in which the other person is held). But the kind of politeness in-
volved in making a request has a negative import because it is intended
to avoid offence: this is neg-politeness, which means mitigating or less-
ening the degree to which S’s goals are imposed on H.

3.4. Some points about a pragmatic approach to politeness

The starting point of pragmatics is primarily in language: explaining
communicative behaviour. By studying this we keep our feet firmly on
the ground, and avoid getting lost too easily in “the highly abstract
notion of ‘face’” (B&L 1987: 13). The basic question is: What did S
mean [to convey] by saying X?. It is useful to postulate the Politeness
Principle (PP), I claim, not because it explains what we mean by the
word “politeness” (an English word which in any case doesn’t closely
match similar words in other languages), but because it explains certain
pragmatic phenomena, especially the following:

a) Indirectness: Being indirect in saying what you mean apparently vio-

lates maxims of Grice’s CP, because S’s utterance seems to be less
informative, less clear, less truthful, and/or less relevant than it could
be as a way of conveying S’s understood intention. For example, I
wonder if you’d mind carrying this tray? is an indirect request, with
roughly the same illocutionary force as Carry this tray. Compare
indirect requests such as jusyeosseumyeon jokessseumnida (‘it would
be nice if you did …’) in Korean. Indirectness is by no means always
motivated by politeness (i. e., the goal of observing the Politeness
Principle); but very often it is. In this case, S’s indirectness is ex-
plained as purportedly offering H a more favourable deal: i. e., more
freedom of choice regarding the action S is requesting.

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
176 Geoffrey Leech

b) Asymmetries of politeness: Politeness often shows up in opposite

ways of treating S and H in dialogue. Whereas conveying a highly
favourable evaluation of H is “courteous”, conveying the same eval-
uation of S is “discourteous”. Conversely, while conveying an unfa-
vourable evaluation of S is “courteous”, giving the same evaluation
of H is “discourteous”. There are opposing “courtesy” values for S
and H, and moreover to heighten or intensify those values is to
increase the degree of “courtesy” or “discourtesy” in those same
opposing directions. In this section I use the term “courteous” in-
stead of “polite” (which can be easily misapplied – see 3.2) in refer-
ring to the face-value or default interpretation of an utterance,
based on its degree of semantic politeness, i. e., a relatively high
position on the absolute politeness scale out of context. “Discourte-
ous” means the opposite of this. For example:
It is “courteous” to compliment H, but “discourteous” to compli-
ment S (i. e., oneself).
It is “courteous” to criticize S’s own behaviour, appearance, etc.,
but “discourteous” to criticize H’s behaviour.

Almost as a technical term, I use the phrase courteous belief for an

attribution which has some positive value to H and/or of some negative
value to S, whereas a discourteous belief is an attribution of some posi-
tive value to S and/or some negative value H. Compare, for example,
the courtesy of (a) and the discourtesy of (b):

a) You’re coming to have dinner with us next week. I insist!

b) I’m coming to have dinner with you next week. I insist!

There are such asymmetries in Chinese and Japanese honorific usage:

(1) bìxìng wáng, nín guìxìng? 敝姓王，您贵姓？

‘My surname is Wang, your surname?
(2) (Namae wa) Buraun desu. O-namae wa?
‘My name is Brown. And your name?’

A Chinese commentator Fu Pei observes: “A speaker could not refer

to his/her name as guìxìng. (Although in modern China, bì is rarely
used in spoken Chinese, guìxìng is still frequently used when asking
the other person’s name)”. Similarly in Japanese it is not polite to add
the honorific o- to S’s name. Hence it is obvious to S and to H that o-
namae refers to H’s name. But in replying, H could not refer to H’s
own name in the same way, using o-namae. There is something similar

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
 Politeness: Is there an East-West divide? 177

in Korean: the difference between My name is … and Your name …?

can be signalled honorifically: by ireum (‘my name’) and seongham
(‘your name’). S could not refer to S’s own name as seongham.

c) Interpretations of elliptical constructions relying on the Politeness

Principle: Asymmetry between S and H manifests itself also in ellip-
tical utterances: e. g., in English, Good luck! in contrast to Bad luck!
It is polite to wish someone good luck, but not to wish him/her bad
luck. Hence, the interpretation of Good luck! (when nothing else is
added) is “I wish you good luck”. But the interpretation of Bad
luck! is “I regret (or commiserate with) your bad luck”. A visitor
from Mars might ask “Why are the interpretations of these elliptical
utterances so different?” The answer is that, because of the PP,
speakers have a tendency to prefer “courteous” interpretations to
“discourteous” ones. The default meaning is the “courteous” one.
The previous example of o-namae wa in Japanese is of a similar
kind: without any overt indication of whose name is intended, the
honorific form makes it clear. In Chinese, Richard Zhonghua Xiao
pointed out to me that Bié kè-qi, when uttered by the host welcom-
ing a guest, means ‘Don’t stand on ceremony; make yourself at
home’. When uttered by the guest, it means ‘Don’t bother’.
d) Battles for politeness: Politeness makes us behave in ways which
our visitor from Mars would think irrational: e. g., a sequence of
polite utterances such as the following may occur in certain cultures
(traditionally, in Chinese, for example):

INVITATION / REFUSAL / INVITATION / REFUSAL / INVITATION /

8
ACCEPT

In earlier publications I called these “pragmatic paradoxes” or “quasi-

paradoxes”, but I think “battles for politeness” is a more appropriate
term. Here is the example (A had already been at B’s home for several
hours and it was almost the time for dinner):

(3) A: Nı̌ jīnwǎn jìu zài wǒménjiā chīdūn bìanfàn(ba).

You tonight just at our home eat a casual dinner.
‘How about you staying and taking potluck with us tonight?’
B: Bù(le), bù(le), taì máfan(le).
No, no, too much trouble.
‘No, no, please don’t bother.’
A: Máfan shénme? Súibiàn chǎo jı̌ge cài, henkuài(de).
Trouble what? Casually fry a few dishes, very soon.
‘What trouble? I’ll just fry a few dishes and it’ll soon be done.’

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
178 Geoffrey Leech

B: Wǒ háishì húiqù(ba), fǎnzhèng yěhái bù’è. Yı̌hòu zàilái dǎrǎo

nı̌mén.
I still go back, anyway also still not hungry. Later again come
trouble you.
‘I’d better go home today. Anyway I’m still not hungry. I’ll
trouble you next time.’
A: (Aìya), dōu zà zhè’er(le). Fǎngzhèng wǒmen yěyào chī(de), jìu
suíbiàn chīdiǎn(ba).
Already be here. Anyway we also eat, just potluck eat a little
bit.
‘Come on! Since you’ve already been here, just take potluck
with us. We ourselves will have to have something to eat
anyway.’
B: nà … nà …, wǒ jiù búkèqi(la).
Well … well …, I just not polite.
‘Well . . . well …, then I’ll bother you this time’
[Note: The elements in parentheses are softening or strengthening
markers.]

These are battles arising from the asymmetry of politeness: what is a

“courteous” proposition for S to uphold (say, S gives H a meal) is a
“discourteous” proposition for H to uphold at this stage of the proceed-
ings (although eventually H accepts the meal from S). Hence, if both
speakers maintained their determination to be as polite as the other,
they would never reconcile their different “courteous beliefs”. Imagine
two people stuck in an elevator doorway, both of them too polite to
give way to the other, and to walk through the door first: After you. –
No, after you. – No, after you … This is the paradoxical aspect of
polite behaviour!
These battles can be resolved by negotiating with the other person’s
politeness. Thus traditionally, after a third invitation, say, an invitee will
“reluctantly” accept the invitation. Or one person will “reluctantly”
agree to go first through the doorway before the other. Another exam-
ple of a politeness battle is described by Miller (1967: 289–90) as a
dialogue between Japanese women which proceeds as follows:

COMPLIMENT / DENIAL / COMPLIMENT / DENIAL / COMPLIMENT

Present-day Japanese speakers often consider this kind of behaviour

exaggerated. But traditionally, it has been courteous to deny the truth
of a compliment. Another kind of quasi-paradoxical sequence (from
Persian) is found in Koutlaki (2002: 1748–9): in this case, two male
speakers alternate as follows:

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
 Politeness: Is there an East-West divide? 179

INSIST (on repaying a debt) / REFUSE (to accept repayment) /

INSIST / REFUSE / INSIST / REFUSE / INSIST / REFUSE

e) Gradations of politeness: There are different degrees of absolute

(semantic) politeness, which can often be observed in the degree of
indirectness of the utterance. In English, for example:

(4a) Will you stand over there?

(4b) Would you stand over there?
(4c) Would you mind standing over there?
(4d) Would you mind standing over there for a second?
(4e) I wonder if you’d mind just standing over there for a second?

In Chinese, Japanese and Korean, similar series can be constructed,

with greater length and indirectness tending to correlated with greater
(semantic) politeness – although there is no claim that examples in
each series exactly translate one another, or that indirectness and po-
liteness are precisely correlated:

(5a) (Nı̌) zhàndào nàbiān qù.（你）站到那边去。

‘(You) stand over there.’
(5b) Qı̌ng (nı̌) zhàndào nàbiān qù. 请（你）站到那边去。
‘(You) please stand over there.’
(5c) (Nı̌) kěyı̌ zhàndào nàbiān qù ma?
（你）可以站到那边去吗？
(5d) Qı̌ng nı̌zài nà’er zhàn yīhuı̌’er, xíngma?
请你在那儿站一会儿，行吗？
(5e) Nín néng zài nà’er shāozhàn yīxiàma? Xièxie!
您能在那儿稍站一下吗？谢谢！

In Japanese:

(6a) Soko ni tatte kureru? (‘Please stand over there?’)

(6b) Soko ni tatte kurenai?
(6c) Soko ni tatte kudasaimasu ka?
(6d) Soko ni tatte kudasaimasen ka?
(6e) Chotto soko ni tatte kudasaimasen ka?

In Korean9, similarly:

(7a) Jeogi seo julleyo? ?

‘Will you stand over there?’

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
180 Geoffrey Leech

(7b) Jeogi seo jusigesseoyo? ‘Would you stand over there?’

(7c) Jeogi jom seo jusigessseumnikka?
(7d) Joesonghajiman jeogi jom seo jusigessseumnikka?
(7e) Joesonghajiman jeogi jom seo jusimyeon gomapgessseumnida.

Notice that indirectness is associated with neg-politeness (the kind of

politeness that tries to avoid offence). In contrast, an offer is associated
with pos-politeness. Hence an offer in English can be abrupt and in-
sistent:

(8) Help yourself. Do sit down. You must come again.

English is exceptional in the many kinds and degrees of indirect request

it allows. But the fact that an utterance is indirect is less important than
the reasons for its indirectness. Semantically, an indirect request like I
wonder if you’d mind holding this tray for a second? is more polite than
(say) Can you hold this tray? for a number of reasons:

a) ostensibly it doesn’t require a response from H. (I wonder is a report

on the state of S’s mind.)
b) it refers to the H’s desired action in the hypothetical mood (you
would mind holding …) as if it were an unlikely possibility.
c) it refers to H’s desired action negatively and pessimistically, envisag-
ing H’s objection to it (you’d mind), to stress the expectation that
H will NOT oblige.
d) it purports to reduce the degree of imposition on H: for a second
refers to a very short time, and therefore minimizes what H is asked
to do.

Hence, in this way, the amount of indirectness correlates with the de-
gree of (semantic) politeness.

4. Rethinking the maxims of politeness in

Principles of Pragmatics
I will now attempt to reformulate the maxims of politeness in POP.
The six maxims of the PP were discussed there: the Maxims of Tact,
Generosity, Approbation, Modesty, Agreement, Sympathy. But here, I
avoid using the term “maxim” because it is so easily misconstrued (see
Thomas 1995: 168), preferring the term “pragmatic constraint”. How-
ever, I begin with a single super-constraint, which comprehends all the
above maxims, and which I call the Grand Strategy of Politeness (or
GSP):

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
 Politeness: Is there an East-West divide? 181

Grand Strategy of Politeness: In order to be polite, S expresses or

implies meanings which associate a high value with what pertains
to O (O = other person(s), mainly the addressee) or associates a
low value with what pertains to S (S = self, speaker)

Although O typically refers to the addressee, there is also third-person

politeness (i. e., being polite to a person or a group of people distinct
from S or H). So O may be a third person – but in English at least,
politeness seems to be directed at a third person referent O generally
only where O is present, or is in some way associated with H.
By employing the GSP, S attempts to ensure that offence is avoided
because both participants are, as it were, “leaning over backwards” to
avoid the discord that would arise if they each pursued their own
agenda selfishly through language. They are also “leaning forward”, in
an opposite direction, to propitiate O through pos-politeness. Remem-
ber that we are talking about the communication of meanings. So peo-
ple speaking politely may secretly, or even overtly, pursue selfish agen-
das. They may be insincere – e. g., in flattering someone in order to get
a better job. On the other hand, they may be sincere – in compliment-
ing someone they admire. But such psychological motivations are irrel-
evant to the pragmatics of politeness. Pragmatics is interested only in
communicative behaviour, and politeness in a pragmatic sense is a mat-
ter of conveying meanings in accord with the GSP.
The list in 4.1 below reformulates the maxims of politeness presented
in POP. Although the maxims expressed in POP in such words as
“Maximise the cost to S” etc. could easily be misconstrued, we can
still think of “maxims” in a Gricean sense as bringing about default
interpretations – i. e., unless overridden by some other constraint(s) or
contextual factor(s). But there is less risk of misunderstanding if we
talk of pragmatic constraints or, as Spencer-Oatey and Jiang (2003) do,
of “sociopragmatic interaction principles” (SIPs) conceived as more
specific realizations of the GSP. Although I label and define these con-
straints in Table 2 below, using such terms as “Tact” and “Modesty”,
they are just variant manifestations of the same super-constraint, the
GSP.
Below I list the most important four constraint-pairs which display
the asymmetry between S and O noted earlier. Pos-politeness con-
straints precede corresponding neg-politeness constraints. It is useful to
note that the hearer-oriented constraints (A), (C), (E), (G) and (I) are
generally more powerful than the speaker-oriented ones (B), (D), (F),
(H) and (J). I will introduce the list with the following crucial words:

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
182 Geoffrey Leech

4.1. In pursuing the GSP, S will express/imply evaluative

meanings (A)–(J) in Table 2

Table 2. The component constraints of the Grand Strategy of Politeness.

Constraint related pair of label for this typical
constraints constraint speech-act
type(s)
(A) place a high value on Generosity/ Generosity commissives
O’s wants Tact
(B) place a low value on Tact directives
S’s wants
(C) place a high value on Approbation/ Approbation compliments
O’s qualities Modesty
(D) place a low value on Modesty self-
S’s qualities devaluation
(E) place a high value on Obligation Obligation apology,
S’s obligation to O (of S to O) thanks
(F) place a low value on Obligation responses to
O’s obligation to S (of O to S) thanks and
apologies
(G) place a high value on Opinion Agreement agreeing,
O’s opinions disagreeing
(H) place a low value on Opinion- giving
S’s opinions reticence opinions
(I) place a high value on Feeling Sympathy expressing
O’s feelings feelings
(J) place a low value on Feeling- suppressing
S’s feelings reticence feelings

Now we look at some brief illustrations of each of these constraints.

(A) Place a high value on O’s wants (GENEROSITY)

For example, offers, invitations and promises are (in default terms)
“generous” and, in English, can be direct or even “rude”. In (9)–(11),
insistent directive features are underlined:

(9) You must come and stay with us next time. I won’t take “no” for
an answer!

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
 Politeness: Is there an East-West divide? 183

(10) Come on! Sit down and have a nice cup of coffee.
(11) No you don’t! I’ll pay for this. I insist.

Comparable examples in Chinese are:

(12) Nı̌ xiàcì yídìng lái’(a). Bùzhǔn shuō bù’(a)!

  

你下次一定来啊。
(13) Laí, laí, laí, hēbeī chá! 来，来，来， 喝杯茶!
(14) Bié, bié, bié, zhèhuí wǒmaı̌dān! Bié héwǒ qiǎng!
    

别，别，别， 别和我抢！
But refusals of requests are “ungenerous” and often have to be very
indirect or even unspoken.

(B) Place a low value on S’s wants (TACT)

For example, requests are often indirect, tentative, giving an opportu-
nity to refuse, and also minimizing S’s imposition on H. This is such a
familiar aspect of politeness that it needs no further examples – see
(4a–e), (5a–e), (6a–e) and (7a–e) above.

(C) Place a high value on O’s qualities (APPROBATION)

For example, we like to pay (and be paid) compliments, if it seems
appropriate to do so. (Insincere or excessive compliments count as flat-
tery, and receive a more mixed reception.) Run-of-the-mill compli-
ments like Your garden looks so lovely and What a pretty dress! are
familiar occurrences. In some activity types complimentary language
is a virtual necessity – e. g., guest praises host(ess)’s meal; academic
introduces lecture of visiting senior professor:

(15) Tài hǎochī(le)! Nı̌de chùyì hé dàfàndiàn chúshī yǒude bı̌(le)!

   
     

太好吃了！你的
‘They are so delicious! Your cuisine is as good as that of a chef
at any big restaurant!’
(16) Hěn róngxìng yāoqı̌ngdào zài * lı̌ngyù zuòchū jiéchū gòngxiànde
* jiàoshòu láigěi wǒmén zuòbàogào.   
     

很荣幸邀请到在 * 领域做出杰出贡献的 * 教
‘It’s a great honour for us to have invited Professor * who has
made great contribution to * field to give us a lecture.’

On the other hand, criticisms of O are hedged and muted: You could
have done better. It wasn’t terribly good. These criticisms would not take

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
184 Geoffrey Leech

place at all, of course, in the host-guest situations mentioned above.

But they would be more likely where the requirement of politeness is
lessened: either where S had a dominant social role, or where O was a
third party, not the hearer.

(D) Place a low value on S’s qualities (MODESTY)

For example, self-deprecation (if sincere, even if exaggerated) is often
felt to be polite:
(17) A: … I’m so dumb, I don’t even know it. hhh! heh
B: Y-no, y- you’re not du :mb
(Pomerantz, 1975: 93; quoted in Levinson 1983: 338)
In this example, the modesty of S’s utterance elicits a denial from H,
in accordance with Approbation. This kind of gratuitous self-depreca-
tion is sometimes called “fishing for compliments”, and the PP predicts
that it will be followed by an (implied) denial and an (implied) compli-
ment:
(18) A: Wǒ tàibèn(le), zhème jiǎndānde wèntí háixiǎngle bàntiān.
 


我太 , 这么简单的问题还想 !
‘I’m so dumb. I can’t believe it has taken me so long to figure
out such a simple question!’
B: Nı̌ háibèn(na)? Nà zhèshìshàng jiù méiyǒu cōngmíngrén(le).
 

你还笨哪？那这世上就没有聪明 !
‘Come on! If you were dumb, there wouldn’t exist any smart
guy in the world!’
In keeping with Modesty, it is disfavoured to agree with compliments,
but H may pay a compliment in return: Gee, it’s nice of you to say that.
Or H may respond with apparent disbelief: Oh, do you really think so?
Or H may thank the complimenter: Thank you. It’s nice of you to say
so, but … These are favoured as ways of showing you appreciate a
compliment, and at the same time you are suitably modest about it.
Another strategy is to “deflect” a compliment, neither agreeing nor
disagreeing with it, but making a remark which downgrades the attribu-
tion of high value to oneself:
(19) A: I really like your outfit.
B: Oh, it’s just something I picked up in a sale.
In China as in Japan, it is said that traditionally a hearer will disagree
with a compliment. Here is an MA student complimenting another MA
student on her high grades in the examination:

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
 Politeness: Is there an East-West divide? 185

(20) A: Nı̌ kě zhēnbàng! 你可

‘You did really well!’
B: Bù, bù, bù, dōu méishénmeyòng … 不，不，不，都没什么用 …
‘No, no, no, they don’t mean much …’

Here is a Japanese example from Tanaka (2001: 248):

(21) A: totemo oniai desu ne (That suits you very well)

B: sonna koto nia-n-desu kedo (Well, not really …).

Such denials can also occur in English:

(22) Kate: Steven is a terrific chef.

Steven: nah.
(Locher and Watts 2005: 24, omitting prosodic marks).

Another way to mitigate the attribution of value to S is to attribute the

success to luck: Well, yeah, somehow I was lucky enough to win the first
prize. This still seems a little like boasting, and is probably less accepta-
ble in Eastern societies than in Western societies10. Self-praise is im-
modest, and so often has to be reluctant. In this example, the hedging
expressions diminishing self-assertiveness are underlined:

(23) I mean I think I’m a pretty straight sort of a guy.11

This is particularly noticeable, for example, on TV, in interviews with

victorious sportsmen/women, or politicians who have won elections.
Victory speeches often contain we rather than I, to emphasize collect-
ive rather than individual merit. This helps to reduce the impression
that S is being boastful, attributing achievements to himself/herself:

(24) Well, we did a pretty good job, thanks to tremendous efforts by

all the folks here.

The following utterances were from a Chinese table tennis player after
she won a championship, which my informants thought was a very typi-
cal Chinese example:

(25) Gǎnxiè lı̌ngdǎo gělle wǒ hěnduō bı̌sàide jīhuì, gǎnxiè jiàoliàn wèi
wǒ fùchūde hōnshuı̌, gǎnxiè zhōngguó pīngpāngduì zhègè guāng-
róngde jítı̌.          
         

感谢领导给  

感谢中国乒乓 光荣的集体。

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
186 Geoffrey Leech

‘Thanks to the leaders who provided me with many chances of

taking part in the competitions; thanks to the coach who has done
so much for me; thanks to the China Table Tennis Team, this
glorious collective.’

(E) Place a high value on S’s obligation to O

(OBLIGATION of S to O)
Apologies for some offence by S to H are examples of polite speech
acts giving high prominence to S’s fault and obligation to O:

(26) I’m (terribly) sorry. Please excuse me. I’m afraid I’ll have to
leave early.

A similar case is the expression of gratitude for some favour H has

done to S:

(27) Thanks. Thank you very much. Thank you very much indeed.

These can be intensified to express greater obligation. Compare in Chi-

nese:

(28) Xìe-xie. Xìe-xie ni. Hēicháng gǎnxiè.

(F) Place a low value on O’s obligation to S

(OBLIGATION of O to S)
On the other hand, responses to apologies often minimize the fault: It’s
okay. Don’t worry. It was nothing. Similarly, responses to thanks often
minimize the debt: That’s all right. You’re welcome. No problem. Glad
to be of help. It was a pleasure. In Chinese:

(29) A: Duì-bu-qı̌ ‘Sorry’

B: Mei-guàn-xi ‘It’s all right’.

(G) Place a high value on O’s opinions (AGREEMENT)

In responding to others’ opinions or judgements, agreement is the pre-
ferred response and disagreement is dispreferred12:

(30) A: It’s a beautiful view, isn’t it?

B: Yeah, absolutely gorgeous.

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
 Politeness: Is there an East-West divide? 187

Intensification (as in gorgeous above, or more stereotypically in an-

swers like Absolutely!) enhance the polite effect of agreement, whereas
mitigated agreement has the opposite effect (as is Yeah, it’s not bad. I
suppose it’s okay). On the other hand, disagreement, as a dispreferred
response, is frequently preceded by delay, hesitation, or the use of tem-
porizing expressions such as Well … Instead of intensifying disagree-
ment, English speakers tend to opt for mitigation – for partial, hedged
or indirect disagreement: Do you really think so? I would have
thought … Yes, but don’t you think …? I agree, but … In Japanese:

(31) Old lady: Saikin-no-koto-nanka-de-mo, Koizumi-san-wa hontooni

yoku-yatteiru wa yo ne.
‘Mr Koizumi has been doing a really good job dealing with recent
issues, hasn’t he?’
Male student: Soo-da-ne, o-baa-chan. Koizumi-san-wa jieitai-o
haken-shitari, iroiro-yatte-kureteru-to-omou-yo. Taihenna-koto-
da-to omou-ne. ‘Yes, grandma. I think Mr Koizumi has done vari-
ous things like sending the Self-Defence Force and so on. They
seem to be serious issues.’13

Here the student (who, it seems, actually disagreed with the old lady’s
approval of the Japanese prime minister) avoided outright disagree-
ment, but at the same time avoided agreeing with her, which would
have violated Grice’s Quality Maxim.

(H) Place a low value on S’s opinions (OPINION-RETICENCE)

As shown in the last example, people frequently soften the force of
their own opinions, by using propositional hedges such as, I think, I
guess, I don’t suppose, It might be that … In other cases, S consults H’s
opinion, deferring to H’s supposed greater understanding, wisdom, or
experience: Don’t you think …? In contrast, there is a low tolerance of
opinionated behaviour, where people express their opinions forcefully,
as if they matter more than others’. Expressing an opinion in Japanese
society may be seen as potentially offensive especially to superiors, in
that an opinion may imply a criticism. For example, in Western coun-
tries it is felt to be positively helpful to ask questions and express opin-
ions in the discussion period following a lecture: if no such interaction
takes place, the visiting speaker may feel the presentation was a “flop”.
However, in Japan (and I am told to some extent in China) it may be
felt impolite to present a different opinion from that of an “honoured
speaker”.

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
188 Geoffrey Leech

(I) Place a high value on O’s feelings (SYMPATHY)

A constraint of Sympathy or Concern is needed to explain why we put
a high value on other people’s feelings in such speech acts as congratu-
lations and condolences. It is polite to show others that you empathize
with them. Congratulations, good wishes and condolences are all intrin-
sically courteous speech acts, and need no mitigation: Congratulations!
Well done. Have a good time! Enjoy your meal or on a sadder note: I
was so sorry to hear about your father … Similar to condolences are
inquiries about people’s health, showing sympathy and concern: How’s
your mother? I hope she’s feeling better … Since these are all “courte-
ous”, they can be made more so by intensification. That is, it is easy
to make them more extreme, by heightening the degree of gradable
expressions they contain (intensifying expressions are underlined):

(32) Warmest congratulations!

(33) I was so terribly sorry to hear about …
(34) I do hope she’s feeling much better …
(35) Have a wonderful time!”

(J) Place a low value on S’s feelings (FEELING-RETICENCE)

The corresponding negative-politeness constraint places a low value on
one’s own feelings. For example, B&L (1978: 240) say “it appears that
in English one shouldn’t admit that one is feeling too bad”, and quote
the following:

(36) A: Hi, how are you?

B: Oh, fine. Actually though …

The first response to questions like How are you? is likely to suppress
any bad news, even though B may be tempted to share his/her troubles.

4.2. Where further investigation is needed

In concluding this section, it should be noted that the ten constraints
are of different degrees of importance and have variable constraining
power. On the whole, it seems the constraints higher up the list are
more powerful than those lower down. But this needs further investiga-
tion.
Moreover, the list of constraints A–J above may be incomplete.
These are simply the most observable manifestations of the GSP. In

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
 Politeness: Is there an East-West divide? 189

POP, when I called these “maxims”, critics found it easy to criticize me

for presenting a seemingly open-ended set, in contrast to Grice’s four.
But the essential point is that these are not separate, independent con-
straints: they are instances of the operation of the GSP as “super-
maxim”. Other instances, such as manifestations of the GSP in turn-
taking behaviour and other aspects of discourse management, could be
elaborated, but will not be mentioned further here.

5. Important disclaimers and caveats

It is easy to oversimplify or misinterpret the workings of the PP. For
example, some have imagined it as some kind of absolute rule of con-
duct – which, as experience shows, is not in accordance with reality. In
practice, politeness is always a matter of degree, and can be affected
by a number of different factors not yet mentioned. It is therefore
important to bear in mind a number of caveats.

5.1. People are impolite as well as polite

Of course, the PP is not always observed. The degree to which it is
observed is sensitive to the (i) vertical distance (upward), the (ii) hori-
zontal distance between S and O, and (iii) the “weightiness” of the
transaction (B&L’s P, D and R factors, though not restricting R to
impositions), as well as other factors – see 5.5. When horizontal dis-
tance is reduced (e. g., in communication with familiars or intimates)
the need for politeness is also reduced – until we move into the zone
of non-politeness or impoliteness. In addition, like Grice’s CP, the PP
can be violated, flouted or suspended. Although this article has fo-
cussed particularly on politeness, impolite communicative behaviour
has recently been studied in its own right, with particular reference to
activity types in which impoliteness is conventionalized and expected:
see, for example, Culpeper 1996, 2005; Culpeper et al. 2003)14. My posi-
tion, incidentally, is that a well-founded theory of politeness must be
the basis also for a theory of impoliteness, since impoliteness is a non-
observance or violation of the constraints of politeness. See also 5.3
below.

5.2. Positive politeness and pos-politeness

I have already signalled the difference between B&L’s “positive polite-
ness” and mine, terminologically distinguished as “pos-politeness”.

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
190 Geoffrey Leech

B&L (1978: 107–118) define positive politeness so broadly as to include

claims of common ground or solidarity15. I would say that such strate-
gies, such as the use of familiar forms of address, are directed towards
reduction of social distance, so that when P and D are very small the
PP applies minimally if at all. This assertion of familiarity or camarade-
rie is likely to have one effect which resembles pos-politeness: it pro-
motes concord or conviviality. But the reasoning behind it is different:
instead of showing asymmetry (high evaluation or deference to H),
S claims solidarity with H, making deference unnecessary because of
closeness. Hence I am defining “positive politeness” more narrowly
than B&L to exclude the “solidarity strategy” that Robin Lakoff called
camaraderie (Lakoff 1990: 38 – see Section 7 below).
On the other hand, I am giving pos-politeness a bigger role that is
allowed for by B&L. For them, positive politeness is just one means of
redress for an FTA. In this sense, then, positive politeness acts in the
service of a “negative” avoidance principle. For me, however, pos-po-
liteness is chiefly an enhancement of face: by attributing value to H, for
example in offering, complimenting, offering sympathy, S is primarily
performing a face enhancing act (sometimes better described as a face
maintaining act) or FEA (Suzuki 2007), not a face threatening act.

5.3. Irony and banter

In POP it is proposed that the PP can be exploited for special purposes.
I argue that the “Irony principle” and the “Banter principle” are sec-
ond-order principles which are rooted in violations of the CP or the
PP, and which work in contrary directions.
Irony is mock-politeness. For example, one person may say to an-
other, who should be working: That’s right – have a good rest. Banter
is mock-impoliteness. During a card game in the student common
room, one student may say to another:

(37) What a mean, cowardly trick!

Later, passing round snacks over a drink in the bar, one of the students
may remark affably:

(38) Hey, don’t take all of it, you greedy bastard!

B&L (1978: 129) give the example:

(39) How about lending me this old heap of junk?

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
 Politeness: Is there an East-West divide? 191

referring to H’s new Cadillac (regarding the joke, however, as a strategy

of positive politeness rather than of banter.) These are not treated seri-
ously, and addressees probably smile or laugh when they are said.
Irony maintains courtesy on the surface, but at a deeper level is cal-
culated to offend. Banter is offensive on the surface, but at a deeper
level is intended to maintain comity. Chinese examples of irony and
banter are shown respectively in (40) and (41):

(40) (A has got up very late)

B (A’s father): Nı̌ qı̌de tàizǎo(le), tiān dōu hái méiliàng(ne)!


   

你起得太早
‘You got up so early! It’s still dark outside.’
(41) (B shows off her new shoes)
B (A’s close friend) with a smile: Chòuměi! 臭美！
‘Stinky beauty!’, ‘Showy!’

With irony, the “reversal” of interpretation occurs because the courte-

ous remark is made in a context where it is clear that the remark is not
intended to be serious, e. g., That’s right – have a good rest is a polite
thing to say where H has worked hard and is exhausted. But in a situa-
tion where S has been working hard, and H is taking it easy, this could
not possibly be the intended meaning. To avoid any misunderstanding,
of course, irony is often associated with a special unfriendly demeanour
or tone of voice, whereas banter is associated with a friendly de-
meanour, laughter, prosodic markedness, etc.
With banter, again, the reversal of interpretation occurs because the
discourteous remark cannot be treated as serious. Banter is a way of
reinforcing in-group solidarity: it is a way of saying “We do not need
to be polite to one another: I can insult you, and you will think it a
joke. This proves what good friends we are”.
Banter can offend against either speaker-oriented or hearer-oriented
politeness. The following example (occurring in answer to a compli-
ment praising S’s skirt) is a mock-boast, offending against the Modesty
constraint – boastfulness being a form of immodesty:

(42) Nàshì! Yě bùxiǎngxiǎng wǒ shì shénmerén?

那是！也不想想我是什么人？
‘Of course it is! You know what taste I always have!’

On the other hand, (41) above is a mock-criticism offending against

Approbation – since polite criticisms of H are likely be muted or indi-
rect.

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
192 Geoffrey Leech

5.4. The constraints may compete or clash with one another

Stock situations where constraints can compete include:

(i) Arguing over who should pay the bill in a restaurant: Here Gener-
osity competes with Agreement. Unless there is a clear under-
standing that one person is being treated by the other, Generosity
motivates each of the two people having a meal to offer to pay
the bill. Although in general, arguments are not considered “po-
lite” events, in this case the anti-disagreement motive is out-
weighed by the pro-generosity motive. Hence arguing about the
bill may be considered a polite ritual, and in certain situations
may continue for some time.
(ii) Giving advice: Generosity can compete with Agreement and
Modesty: giving advice means offering the benefit of your opin-
ion to O, but it can also imply that you are exalting your own
opinion above that of O. Hence advice is a double-edged speech
act, and is commonly introduced with markers of unreality, condi-
tionality, tentativeness: If I were you, I’d …; Wouldn’t it be better
if …; Could I suggest … Examples in Chinese:

(43) Nı̌ kàn zhèyàng shìbúshì xíngdétōng, …?


    

你 ……？
‘Would you please see whether it would be okay to…?’
(44) Yǒuméiyǒu kěnéng …? 有没有可 ……？ 

‘Would it be feasible if …?’

and in Korean:

(45) Jeoramyeon, … hagesseoyo.

‘If I were you, I would …’
(46) … haneun ge jochi anheulkkayo? … ?
‘Would it be better if …?’

(iii) Offering, inviting: Generosity can compete with Modesty. Thus

after preparing abundant food, a Japanese hostess may say:

(47) Nani mo arimasen ga, dōzo.

‘There’s nothing [special/to eat], but please … ’

This is almost paradoxical. The speaker simultaneously has to

imply Generosity and Modesty. She has to recommend the food,

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
 Politeness: Is there an East-West divide? 193

and not recommend it. There is also a possible conflict between

Generosity and Tact. These two constraints often work hand in
hand: a well-judged offer will be both clearly generous in sacrific-
ing S’s interests and tactful in serving the assumed wants of
H. However, a tentative offer such as the following seems to arise
from a conflict of constraints:

(48) I don’t know if you’d like this dress. I just picked it up in a junk
shop the other day. It’s really nothing.

If this is an offer, then its tentativeness (disparaging the gift)

might make it seem ungenerous. But this impression may occur
because Tact is winning out over Generosity: S may be genuinely
uncertain whether the gift will please, and to “force” an un-
wanted gift on someone would be tactless. In addition, S may be
exaggerating the worthlessness of the offer, partly as a way of
thwarting O’s expected polite reluctance to accept (see 5.6).

(iv) Competition or conflict with the CP: It is assumed (as in POP)

that there can be competition between the maxims/constraints of
the PP and the maxims/constraints of the Cooperative Principle
(CP). For example, an exaggerated compliment, apparently maxi-
mizing Approbation, may be rejected as “flattery” because it con-
flicts with the Maxim of Quality (= truthfulness). Other speech
acts of pos-politeness may also be considered insincere if they
are overdone. Hence, although such speech acts as compliments,
apologies, and offers lend themselves to intensification or exag-
geration, the CP puts a limit on the degree of exaggeration which
is acceptable. A different case of conflict with the CP arises with
speech acts of neg-politeness, for example where Approbation
competes with Quality or Quantity:

(49) A: I’m getting fat hh.

B: [silence]: [implying Agreement] (Levinson 1983: 339).

Potential disagreement may be masked by not speaking the truth,

or by leaving something unsaid.

5.5. We use scales to assess the appropriate degree of politeness

Politeness is itself a matter of degree, and determining the appropriate
degree of (pragmatic) politeness depends on other scales of value. The
most important of these are:

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
194 Geoffrey Leech

(i) Vertical distance between S and O (in terms of status, power,

role, age, etc.) [cf. B&L’s P]
(ii) Horizontal distance between S and O (intimate, familiar, ac-
quaintance, stranger, etc.) [cf. B&L’s D]
(iii) Weight or value: how large is the benefit, the cost, the favour, the
obligation, etc. [cf. B&L’s R], i. e., the real socially-defined value
of what is being transacted.
(iv) Strength of socially-defined rights and obligations (e. g., a
teacher’s obligations to a student; a host’s obligations to a guest,
service providers’ obligations to their clients or customers).
(v) “Self-territory” and “other-territory” (in-group membership vs.
out-group). There are degrees of membership of “self-territory”
and “other-territory”.

This last factor determines who belongs to the domain of S and who
to the domain of O for the purposes of the GSP. An illustration of in-
group membership is the strong group association, in some Eastern
cultures, between members of a family. One result of this is the Mod-
esty requirement to be humble not only about oneself (in addressing
out-groupers), but about other members of one’s family. Hence in Ko-
rean and Japanese, and traditionally in Chinese, different terms are
used for “my wife” and “your wife”, the former being to varying de-
grees uncomplimentary and the latter to varying degrees complimen-
tary (honorific). For example:

(50) anae/jipsaram/ansaram v. buin/samonim in Korean

v.

(51) nyōbō/kanai/tsuma v. okusan/okusama in Japanese.

5.6. Attributing politeness to H

One added complexity to the account of politeness relates to the asym-

metry between S and H, discussed in 3.4 (b). What is polite from S’s
viewpoint is impolite from H’s viewpoint, so that to be successfully
polite, S sometimes needs to override an anticipated polite response
from H. After all, “politeness” is evaluated favourably in people’s be-
haviour, so to attribute politeness to H is a manifestation of Approba-
tion. An example of awareness for one another’s politeness is seen in
the following negotiation over an offer of a lift, quoted by Gu (1990)16:

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
 Politeness: Is there an East-West divide? 195

(52) A: I can drop you in town if you like.

B: It’s very kind of you, but it will cause you some inconvenience,
won’t it?
A: No, not at all. I’m going in that direction.

After A’s offer has been explored, and its degree of Generosity ascer-
tained, B can then go ahead and accept it without offending against
Tact. The interesting point is that A’s reassurance I’m going in that
direction, while it reduces A’s claim to Generosity, is not felt to reduce
A’s politeness in that it shows consideration for B’s politeness and so
prepares the way for B to accept. That is, Generosity, as a politeness
constraint, does not mean overtly claiming generosity for oneself –
which indeed is a form of boasting (immodesty). It means, rather, act-
ing verbally in a way that attends to the assumed wants of H.

6. Interlinguistic and cross-cultural variation in politeness

The use of politeness in communication is dependent on (a) the lan-

guage and (b) the social or cultural milieu. Hence there are (a) linguisti-
cally-oriented and (b) socio-culturally-oriented aspects of politeness. In
POP these are termed (a) pragmalinguistic, and (b) socio-pragmatic
aspects. In general, the linguistic code of politeness and the cultural
values it allows S to express reinforce one another, but there can be
some mismatch.

6.1. The pragmalinguistic plane of politeness

The values of politeness are encoded mainly through the differing mor-
phological, syntactic and lexical resources of languages. The following
list gives a few illustrations:

– Honorific forms, sentence final pragmatic markers, fading voice and

utterance incompleteness in Chinese and Japanese (Ohta 1991).
– Modal verbs (including hypothetical forms would, could etc.), vari-
ous hedges (e. g., a tiny bit) and intensifiers (e. g., really, terribly) in
Chinese and English.
– Varied “Self”-reference forms and “Other”-reference forms in many
languages: e. g., nı̌ 你 and nín 您 in Chinese; tu and vous in French;
du and Sie in German.

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
196 Geoffrey Leech

– Omission of 1st and 2nd person reference in Japanese and Korean;

use, instead (also in Chinese) of respectful nouns for second person 

reference, e. g.: zhǔrèn 主任 (section chief), (wáng) lǎoshī 王老

(teacher Wang), in Chinese; Kacho, Sensei in Japanese; and Sajang-
nim, Seonsaengnim in Korean.

As is well known, the forms encoding politeness have often become

highly conventionalized, and therefore come to have a weakened force.

6.2. The socio-pragmatic plane of politeness

The socio-pragmatic factors which determine the weight of the values
to be communicated are, as noted in 5.5, scalar. It seems likely that the
scales are fairly general to human societies, but the values considered
norms vary from culture to culture. The variation is both quantitative
(i. e., in degree or position on a scale) and qualitative (i. e., in the actual
social content of the scales themselves).

6.2.1. Quantitative differences in the scales influencing the norm

of politeness
Although all statements of quantitative difference in the present state
of knowledge have to be provisional, this politeness framework enables
us to state hypotheses such as that Modesty has a higher rating in
Japanese or Korean than in “Anglo” societies, where Tact has a high
rating. (Notice we are still talking about communicative behaviour: no
claim is made about whether Japanese people, as a character trait, are
generally more modest than others.) Such claims gain support from
certain pragmalinguistic features, such as the use of humiliative forms
– now in decline – in Korean (jolgo ‘my paper’, nuchuhan jip ‘my
home’), the use of humiliative forms (e. g., using the verb mairu instead
of kuru, ‘to come’) in Japanese, and the exotic range of tactful indirect
request forms in English (I was just wondering if you’d mind …). Of
course such stereotypic generalizations at best are over-simplifications,
but the theory does provide a framework for cross-cultural compari-
sons of politeness pragmatics on this level.

6.2.2. Qualitative differences in the scales influencing norms of

politeness
Examples:
Social distance is interpreted differently for Chinese and British culture.
Vertical distance (B&L’s P) involves a number of different factors, such

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
 Politeness: Is there an East-West divide? 197

as power, age, and social status. As reported above in 2.1, Spencer-

Oatey (1993) investigated the vertical distance relation between Chi-
nese graduate students and their tutors, and found it different from
that between British students and their tutors. It is widely reported that
in Chinese, Korean and Japanese culture, age is particularly important
as an index of superiority on the P scale.

Differences between in-group and out-group (e. g., Japanese vs. the Eng-
lish-speaking world). In Japan spouses do not normally praise their
spouses or their children in talking to people of other families. In fact,
there is a tradition of being “modest” in denigrating one’s family mem-
bers. Similar considerations apply traditionally in China. But in Ameri-
can culture many spouses do this without embarrassment. In fact it
might be considered a polite thing to do – polite, that is, to one’s
spouse.

Differences between socially-defined rights and obligations. For exam-

ple, there are rights and obligations between parents and children, be-
tween teachers and students, between hosts and guests, between bus
drivers and bus passengers, etc. In requesting an action which H has a
socially-sanctioned obligation to do, one needs to show less politeness
than in requesting an action for which H has no obligation. In Japanese
society, there is a well-known high degree of deference and obligation-
to-serve shown by service staff (e. g., in hotels and stores) towards cus-
tomers, although, like other politeness traditions in Japan, this is begin-
ning to undergo change under the influence of “youth culture”.

Differences in the evaluation of weight. The constraints of politeness

apply to some transaction of a value which may be weighty to varying
degrees. But the assessment of weightiness may differ from culture to
culture. As an example, some goods may be considered valuable in
one society but not in another. It has been claimed, for instance, that
(relatively speaking) umbrellas in Japan, cigarettes in Russia, cars in
the USA may be easily borrowed or given.

7. A postscript on politeness in relation to honorifics

Honorifics are found widely among human languages, but Japanese and
Korean have particularly rich and complex honorific systems. Honorif-
ics in modern societies concern relations between S and O, and how
these are grammatically encoded taking account of the relative defer-
ence or familiarity appropriate to these relations. Crucial here are
B&L’s two axes of P and D (originally used by Brown and Gilman,

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
198 Geoffrey Leech

1960, for discussing T and V pronouns in European languages) or, as I

prefer to call them, vertical and horizontal distance.
As B&L have noted, three scales (or axes) are involved in measuring
the appropriate degree of politeness required for a given utterance: the
axes of vertical and horizontal distance (P and D), and the axis R
representing the weight of the imposition (or, as I would like to extend
it beyond B&L, of any other value transacted – e. g., the weight of
value attributed to O in a compliment, or the debt attributed to S in
an apology). For example, in deciding whether to say Will you lend
me …? Can I borrow …? Could I just borrow …? I wonder if I could
possibly borrow …, we have to weigh up:

P: how much higher or lower H is than S on a scale of power/status)

D: how distant or intimate is the relation between S and H – this in-
cludes out-group or in-group membership)
R: the cost to the lender, and benefit to the borrower, of what is to
be borrowed17

Hence in politeness communication as I have just considered it here,

there is also a third scale, representing the ranking in terms of value or
weight (R) of what is transacted in any given speech act. Thus, unlike
honorifics, which need to be used appropriately even in transactionally
neutral situations such as weather reporting, this concept of politeness
is relevant only to certain speech acts, those which involve a value
transaction. For example, a request seeks a transaction of value from H
to S. A compliment is a transaction of value from S to H. Another
difference between honorific usage and transactional politeness is that,
arguably, the use of honorifics is not specifically goal-oriented in the
way that transactional politeness is, but is more socially constrained
and dependent on convention.
These factors seem to make transactional politeness different from
honorific usage. This is a major reason for disagreements between B&L
and Ide, Matsumoto and others. It can be noted, however, that honorific
usage is one of the pragmalinguistic resources in Korean, Japanese and
other languages for expressing transactional politeness. So transactional
politeness often involves or requires honorification (Okamoto 1999;
Byon 2006). At the same time, it can be argued that honorification is a
conventionalized implementation of the constraints of Approbation and
Modesty: giving high value to others, and/or low value to oneself.
Whether we extend the use of “politeness” to cover honorification is
partly a matter of terminological decision. But it is also a question of
whether the two “politeness” domains are close enough to be compre-
hended by the same theory. My provisional conclusion is this: the term

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
 Politeness: Is there an East-West divide? 199

“politeness” (and similar terms in other languages) can cover two related
and overlapping kinds of communicative phenomena. To keep these dis-
tinct, we could follow Ide (1993) in calling the honorific and transactional
kinds of politeness first order politeness and second order politeness re-
spectively. But to relate them to the present discussion, I will call them
honorific politeness and transactional politeness. Honorific politeness is
the upper end of a scale, the other end of which is “familiarity” or (using
Lakoff’s term) camaraderie. We may illustrate them from English voca-
tives: sir and madam are honorific, implying that the P and D factors are
non-trivial and large. At the other end of the scale, vocatives of camarade-
rie such as buddy, guys, dude, man and mate imply that the P and D factors
are trivial and small. As an alternative to the GSP, there is a “grand strat-
egy of camaraderie”, a strategy which, according to Lakoff, made its way
into American usage roughly from 1970 onwards, spreading from the
western states of the US. This strategy achieves rapport by taking steps to
minimize social distance, and combining this with attitudinal warmth.
In very broad terms, transactional politeness is three-dimensional,
concerned with the three dimensions corresponding to B&L’s P, D, and
R, although of course each of the dimensions is itself multi-dimen-
sional. Honorific politeness, on the other hand, is two-dimensional –
concerned with only the two B&L dimensions of P and D. However,
the GSP applies to them both, and to this extent they can be compre-
hended within the same theoretical approach.

8. Postscript on face
B&L’s treatment of face has been criticized, and yet the concept of face
is widely assumed to be the basis of politeness. One of the problems
with B&L’s concept of face is the lack of correspondence between their
definitions of negative face and positive face18. I prefer to define face
more straightforwardly and symmetrically as follows:

Face is the self-image or self-esteem that a person maintains as a

reflection of that person’s estimation by others.
Negative face goal: the goal of avoiding loss of face. (Loss of face is
a lowering of that self-esteem, as a result of the lowering of that
person’s estimation in the eyes of others.)
Positive face goal: the goal of enhancing face (i. e., the heightening/
maintaining of a person’s self-esteem, as a result of the heightening/
maintaining of that person’s estimation in the eyes of others.)

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
200 Geoffrey Leech

The constraints of politeness as presented in section 4.1 above can be

re-interpreted as motivated by these negative and positive face goals.
Put simply: neg-politeness serves the negative face goal; pos-politeness
serves a positive face goal. Thus a request is an FTA because it pur-
ports to make demands on H: if H refuses the request, this shows (to
an extent) H’s low evaluation of S’s wants, and hence S will lose face
by being “turned down”. Secondarily, H will probably go down in S’s
estimation, which entails risk of face-loss to H. So the politeness re-
quired here is neg-politeness, intended to avoid loss of face, by reduc-
ing the assessment of imposition. In contrast, an offer is a face-enhanc-
ing act (FEA) because it implies S’s high estimation of H’s needs, and
therefore heightens H’s own self-esteem. This is a case of pos-polite-
ness. Secondarily, as a result of an offer, S is likely to go up in H’s
estimation, so that S’s face is enhanced19. Because politeness is a posi-
tive value attributed to a person who appropriately and successfully
performs a polite speech act, any speech act which attributes a high
value to H (serving H’s face need) has a secondary effect of serving S’s
face need.
In this way, the GSP can be reformulated in cognitive terms as serv-
ing the mutually-dependent face needs of each participant20. It can be
said that there is a social function (serving concord) and a psychological
function (serving face) of politeness. They are closely interconnected,
and there is no need to choose one rather than the other. As always in
linguistics, and particularly in pragmatics, there is a need for both cog-
nitive and societal explanations.

9. Some tentative conclusions

9.1. Universals of politeness?
B&L have been heavily criticized for elevating a culturally-biased the-
ory into a theory claiming universals of politeness. Consequently, much
of the criticism of B&L has had an anti-universalist or relativist bias.
Although I do not want to underestimate the large differences between
polite linguistic behaviour in different societies, I believe that there is
a common pragmatic and behavioural basis for them, so that (for exam-
ple) when Chinese speakers talk of lı̌mào and English speakers talk of
politeness they are not talking about totally unrelated phenomena. I
have tried to present evidence supporting this view in this article – but
the decision for or against must be decided by those who know both
Chinese and English-speaking language and society. It is premature to
talk of universals of politeness, but my position is as follows. Probably
the scales of value in 5.5 above are very widespread in human societies,

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
 Politeness: Is there an East-West divide? 201

but their interpretation differs from society to society, just as their en-
coding differs from language to language. I suggest this is the basis on
which a well-founded cross-cultural pragmatic research could proceed.
The question to ask is: given these scales of value, what socio-cultural
variants of them are found in particular cultures, and what pragmalin-
guistic forms of language are used to encode these variants?

9.2. Eastern group-orientation vs. Western individual-orientation

There is little doubt that the Eastern group-orientation and the West-
ern individual-orientation are felt to be strong influences on polite be-
haviour. But do the East and the West need a different theory of polite-
ness? I would argue that they do not, because the scales of politeness
can be used to express such differences in values, both qualitative and
quantitative. For example, the group orientation of Japanese culture
(as compared with Western cultures) may be expressed in politeness
norms through such factors as these:

(i) Does vertical distance have a higher weighting in assessing polite-

ness?
(ii) Is vertical distance also qualitatively different: more identified
with status, role and seniority, rather than with individual power
alone?
(iii) Do in-group/out-group distinctions have a clearer and more im-
portant role than in the west (Ide 1989)?
(iv) Are socially defined rights and obligations associated more with
group identity than with individual relationships?

I present these as questions to be answered – or rather, as hypotheses

to be investigated – although much evidence seems to support them.
Although there are big differences between Chinese and Japanese
culture, they have been influenced by the same Confucian tradition,
and represent an ethos that “privileges group harmony over individual
freedom” (Mao 1994: 473). Mao goes on: “To be respected, in this
context, is not to be left alone; rather, it is to be included as a reputable
member of the community. An individual’s behaviour becomes mean-
ingful only in the context of the participation of others. When this
happens, one’s ego sings a chorus of union with the rest of the commu-
nity”. This is poetically put, but I would argue that the PP and the GSP
are easy to reconcile with this “chorus of union”, since they both in-
volve explicating the group-oriented goals of the individual.
On the other hand, the individualistic, “egalitarian” orientation of
Western countries is expressed in politeness through such factors as the

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
202 Geoffrey Leech

higher weighting of Tact (avoidance of imposition on O) and a lower

weighting of Modesty (self-assertion being regarded more favourably).
Note, by the way, that the “egalitarian orientation” does not bring an
assumption that Western societies are somehow more egalitarian than
other societies, although they may be. (Would you like to hand in your
papers now, for example, may be totally authoritarian in its intention
when spoken by a British examination invigilator. But it is encoded in
egalitarian language, in that S ostensibly gives O the choice of whether
to hand in their papers or not, depending on their own wishes.) It is
rather an assumption about the meanings which speakers of Western
languages – such as in particular the “Anglophones” – favour for en-
coding in their language. All this argues in favour of the conclusion
that, despite differences, there is no East-West divide in politeness.

Acknowledgments
I am more than usually grateful for the help I have received from in-
formants on the Chinese, Japanese and Korean languages and cultures.
They not only provided rich examples but some detailed discussion on
matters of linguistic and cultural interest. At the same time, I would
like to exonerate them from any errors and misinterpretations I have
made. For Chinese data, I am indebted to Richard Zhonghua Xiao, Fu
Pei and Luo Qing; for Japanese, to Noriko Tanaka and Toshihiko Suz-
uki; and for Korean, to Hyeon Oak and Soo-kyung Kim. This paper
owes a lot to feedback I have received from presentations I have made,
particularly those at Doshisha University, Kyoto, in 2000, at Busan Na-
tional University Institute of Humanities, Korea, in 2001, and at Shang-
hai International Studies University in 2005. A considerable part of
this paper is modelled on earlier articles, Leech (2003, 2005), and I am
grateful to Prof. Akira Hagiwara, general editor of International Jour-
nal of Pragmatics, and Prof. Dingfang Shu, editor of Wai Guo Yu: Jour-
nal of Foreign Languages for agreeing to this reappearance of a revised
version of those articles.

Notes
1. See note 6 below.
2. To many readers this article will seem like a throw-back to an earlier age when
Gricean pragmatics was “cutting edge” and politeness research was in its infancy.
I still believe that the Gricean paradigm has much to offer (in its updated “neo-
Gricean” form), and this article will place strong emphasis on the linguistic, as
well as the social, aspects of politeness. The term “politeness”, of course, has
been hotly contested in the intervening years. For clarity, let me say that I use
the term to refer to a scalar phenomenon of human communicative behaviour,

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
 Politeness: Is there an East-West divide? 203
and I do not wish to draw a line between marked behaviour called “polite” or
“impolite” and routinized, normal behaviour called “politic” (Watts et al. 1992;
Watts 2003: 4). I think it is difficult if not impossible to draw this line. Thus,
for me, politeness can be relatively salient or relatively formulaic and routine.
Regarding another distinction made by Watts (2003) as well as by Eelen (2001),
that between politeness1 and politeness2, I think my use of “politeness” inclines
towards “politeness2” – i. e. towards politeness as a model/theory building con-
cept, rather than as a folk or lay concept, although there is clearly a strong
connection between the two.
3. Richard Zhonghua Xiao has pointed out to me that in old times teachers in
China were called shifu (shi – teacher; fu – father).
4. However, B&L themselves put forward large, apparently open-ended, lists of
regularities: for example, they enumerate 15 positive politeness strategies and
10 negative politeness strategies (1978: 107, 136).
5. Recent neo-Gricean or “post-Gricean” thinking on the semantics-pragmatics di-
vide has stressed the importance of default interpretations, which do not exclude
exceptional or non-default interpretations by means of implicature. See, espe-
cially, Jaszczolt (2005). In this sense, we are justified in recognizing relative de-
grees of politeness in utterances out of context (what in POP I called “absolute
politeness”), while also recognizing the potential for special non-polite interpre-
tations. The view that the term politeness can only be applied to utterances in
context (pragmatic or relative politeness) is, in my view, an unhelpful restriction.
6. I now avoid the term “absolute politeness” as it has been misunderstood – for
example, Spencer-Oatey (2005: 97) claims that “Leech (1983) … takes an ‘abso-
lute’ approach to politeness. He identifies a number of politeness maxims, such
as the Tact Maxim, … and implies that the more a maxim is upheld, the more
polite the person will be. / Numerous authors have challenged this perspective”.
Actually, Leech (1983) never adopted this perspective in the first place, and
carefully distinguished between “absolute politeness” and “relative politeness”
(1983: 83–4), termed “semantic” and “pragmatic” politeness in this article. It is
true, however, that I put more emphasis on absolute politeness as a concern of
“general pragmatics”, and I now recognize that the term “absolute” was unwise.
I am indebted to Toshihiko Suzuki for discussion of this issue.
7. The term “maxim” used in the Gricean sense has, of course, no prescriptive
implications, but there has been such temptation to misunderstanding that it is
best to avoid it.
8. Actually the iterative structure of Chinese invitation/offer – refusal sequences
is more subtle and variable than this suggests. For examples and analyses, see
Gu (1990: 252–3), Mao (1994: 475–9) and Zhu et al. (2000, especially p. 98).
9. However, Byon (2006: 269) argues that in Korean, honorifics are more important
for indicating politeness in requests than indirectness.
10. However, Fu Pei and Luo Qin (personal communication) point out that such
“traditional” forms of polite behaviour may no longer be observed among the
younger generation in China. A similar comment has been made by Noriko
Tanaka about Japan.
11. Spoken by Tony Blair, British Prime Minister, in a BBC interview with John
Humphrys, 15 November 1997.
12. On preference organization in discourse, including the preference for agree-
ment, see Levinson (1983: 332–345). It is strange that Levinson here does not
develop the relation between preference structure and politeness. For me, the
PP provides an obvious explanation for many of the dialogue asymmetries
pointed out in his chapter 6 on conversational structure.

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
204 Geoffrey Leech
13. This example is from Suzuki (2007: 190).
14. A Conference on Linguistic Impoliteness and Rudeness was held at the Univer-
sity of Huddersfield, UK, on 3–4 July, 2006.
15. Perhaps the most extreme case is B&L’s claim that God you’re farty tonight
(1978: 109) is an instance of positive politeness. There has been much discussion
of the flexible meaning of “politeness” and similar terms in other languages, but
this example seems to stretch it beyond breaking point.
16. Gu (1990: 243–4), following a somewhat similar proposal by Liu (1986), argues
from this example that there are really two variants of the Generosity constraint
used in invitations or offers. One is a “content-regulating maximization” and
the other is “speech regulating minimization”. He observes (p. 244) that “If S
offers H something, S will usually minimize by means of speech the cost which
the offer incurs to him, and H will in turn maximize, also by means of speech,
the benefit he receives from S’s offer”.
17. The simple additive relation between P, D and R proposed by B&L has been
more or less discredited (see 2.1. above). However, it remains clear that these
factors play an important role in the functioning of politeness.
18. On this mismatch, see Spencer-Oatey (2000: 12–16), who argues that B&L’s
concept of face should be separated into two distinct concerns: face needs and
sociality rights, both subsumed under the heading of “rapport management”.
B&L’s negative face needs correspond to Spencer-Oatey’s sociality rights.
19. On the other hand, I would argue that B&L (1978: 71–73) go too far in extend-
ing the use of “face-threatening act” not only to orders, requests, and the like,
but also to such face-enhancing acts as thanking, complimenting, inviting and
promising (see Koutlaki 2002: 1737–8).
20. In this respect my conception of ‘face’ finds more resonance in the formulations
of more recent “face-theorists” (such as Bargiela-Chiappini 2003 and Arundale
2006) than in that of B&L.

References
Arundale, Robert B. (2006). Face as relational and interactional: A communication
framework for research on face, facework, and politeness. Journal of Politeness
Research 2 (2): 193–216.
Bargiela-Chiappini, Francesca (2003). Face and politeness: New (insights) for old
(concepts). Journal of Pragmatics 35 (10/11): 1453–1469.
Brown, Penelope and Stephen Levinson (1978). Universals in language usage: Polite-
ness phenomena. In Questions and Politeness, Esther N. Goody (ed.), 56–310.
Cambridge: Cambridge University Press. (Reissued 1987 with corrections, new
introduction and new bibliography, as a book entitled Politeness: Some Univer-
sals in Language Usage.)
Brown, Roger W., and Albert Gilman (1960). Pronouns of power and solidarity. In
Style in Language, Thomas A. Sebeok (ed.), 253–276. Cambridge, MA: MIT
Press. Reprinted in Pier P. Giglioli (ed.) (1972), Language and Social Context,
252–82. Harmondsworth: Penguin Books.
Byon, Andrew S. (2006). The role of linguistic indirectness and honorifics in achiev-
ing linguistic politeness in Korean requests. Journal of Politeness Research 2 (2):
247–276.
Culpeper, Jonathan (1996). Towards an anatomy of impoliteness. Journal of Prag-
matics 25 (3): 349–367.

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
 Politeness: Is there an East-West divide? 205
Culpeper, Jonathan (2005). Impoliteness and entertainment in the television quiz
show: The Weakest Link. Journal of Politeness Research 1 (1): 35–72.
Culpeper, Jonathan, Derek Bousfield, and Anne Wichmann (2003). Impoliteness
revisited: With special reference to dynamic and prosodic aspects. Journal of
Pragmatics 35 (10/11): 1545–1579.
Eelen, Gino (2001). A Critique of Politeness Theory. Manchester, UK: St. Jerome
Publishing.
Grice, H. Paul (1975). Logic and conversation. In Syntax and Semantics III – Speech
Acts, Peter Cole and Jerry L. Morgan (eds.), 41–58. New York: Academic Press.
Gu, Yueguo (1990). Politeness phenomena in modern Chinese. Journal of Pragmatics
14 (2): 237–257.
Ide, Sachiko (1989). Formal forms and discernment: Two neglected aspects of univer-
sals of politeness. Multilingua 8: 223–248.
Ide, Sachiko (1993). The search for integrated universals of linguistic politeness.
Multilingua 12 (1): 7–11.
Koutlaki, Sofia A. (2002). Offers and expressions of thanks as face enhancing acts:
Tæ’arof in Persian. Journal of Pragmatics 34 (12): 1733–1756.
Lakoff, Robin (1990). Talking Power. New York: Basic Books.
Leech, Geoffrey (1983). Principles of Pragmatics. London: Longman.
Leech, Geoffrey (2003). Towards an anatomy of politeness in communication. Inter-
national Journal of Pragmatics 14: 101–123.
Leech, Geoffrey (2005). Politeness: Is there an East-West Divide? Wai Guo Yu: Jour-
nal of Foreign Languages 160 (No.6, November 2005): 3–31.
Levinson, Stephen C. (1983). Pragmatics. Cambridge: Cambridge University Press.
Liu, Runqing (1986). The politeness principle and a dream of red mansions. Unpub-
lished MPhil thesis, Lancaster University.
Locher, Miriam A. and Richard J. Watts (2005). Politeness and relational work. Jour-
nal of Politeness Research 1 (1): 9–33.
Mao, LuMing Robert (1994). Beyond politeness theory: Face revisited and renewed.
Journal of Pragmatics 21 (5): 451–86.
Matsumoto, Yoshiko (1989). Politeness and conversational universals― observations
from Japanese. Multilingua, 8 (2/3): 207–221.
Miller, R. A. (1967). The Japanese Language. Chicago: Chicago University Press.
Ohta, Amy Snyder (1991). Evidentiality and politeness in Japanese. Issues in Applied
Linguistics, 2 (2): 211–238.
Okamoto, Shigeko (1999). Manipulating honorific and non-honorific expressions in
Japanese conversations. Pragmatics 9 (1): 51–74.
Pomerantz, A. (1975). Second assessments: A study of some features of agreements/
disagreements. Unpublished PhD dissertation, University of California, Irvine.
Schmidt, Richard W. (1980). Review of Esther Goody, ed., Questions and Politeness:
Strategies in Social Interaction. RELC Journal (Singapore), 11: 100–114.
Spencer-Oatey, Helen (1993). Conceptions of social relations and pragmatics re-
search. Journal of Pragmatics 20 (1): 27–47.
Spencer-Oatey, Helen (ed.) (2000). Culturally Speaking: Managing Rapport through
Talk across Cultures. London and New York: Continuum.
Spencer-Oatey, Helen (2005). (Im)Politeness, face and perceptions of rapport: Un-
packaging their bases and interrelationships. Journal of Politeness Research 1 (1):
95–119.
Spencer-Oatey, Helen and Wenying Jiang (2003). Explaining cross-cultural prag-
matic findings: Moving from politeness maxims to sociopragmatic interaction
principles (SIPs). Journal of Pragmatics 35 (11): 1633–1650.
Suzuki, Toshihiko (2007). A Pragmatic Approach to the Generation Gap and Gender
Gap in Japanese Politeness Strategies. Tokyo: Hituzi Shobo Publishing.

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM
206 Geoffrey Leech
Tanaka, Noriko (2001). The Pragmatics of Uncertainty: Its Realisation and Interpreta-
tion in English and Japanese. Yokohama: Shumpūsha.
Thomas, Jenny (1995), Meaning in Interaction: An Introduction to Pragmatics. Lon-
don: Longman.
Watts, Richard (2003). Politeness. Cambridge: Cambridge University Press.
Watts, Richard, Sachiko Ide, and Konrad Ehlich (1992) (eds.) Politeness in Language:
Studies in its History, Theory and Practice. Berlin/New York: Mouton de Gruyter.
Wierzbicka, Anna (1991[2003]). Cross-cultural Pragmatics: the Semantics of Human
Interaction. Berlin/New York: Mouton de Gruyter. (Second edition published,
with a new introduction, 2003.)
Zhu Hua, Li Wei and Qian Yuan (2000), The sequential organisation of gift offering
and acceptance in Chinese. Journal of Pragmatics, 32 (1): 81–103.

Brought to you by | Heinrich Heine Universität Düsseldorf

Authenticated | 132.74.1.4
Download Date | 7/19/13 11:55 AM