Summary Definitions and examples

Politeness (Brown and Levinson 1987)
Wednesday, April 27, 2011
12:04 PM
Brown, P. and S. Levinson. 1987. Politeness. Cambridge: Cambridge University Press.
With some stuff from:
Watts, R. 2003. Politeness. Cambridge: Cambridge University Press.
Definitions and examples
Brown and Levinson (1987) are showing that
individual's self-esteem (face) motivates strategies
of politeness (solidarity, restraint, avoidance of
unequivocal impositions).
Face: See below (including definitions of
positive/negative face).
Here's the starting point. Among Model Persons, it
is mutual knowledge that for all Model Persons
(verbatim, including bold, from B&L 1987: 59-60):
i) "All MPs have positive face and negative
face, and all MPs are rational agents--i.e.
choose means that will satisfy their ends.
ii) "Given that face consists in a set of wants
satisfiable only by the actions (including
expressions of wants) of others, it will in
general be to the mutual interest of two MPs
to maintain each other's face. So S will want
to maintain H's face, unless he can get H to
maintain S's without recompense, by
coercion, trickery, etc.
iii) "Some acts intrinsically threaten face; these
'face-threatening acts' will be referred to
henceforth as FTAs.
iv) "Unless S's want to do an FTA with maximum
efficiency (defined as bald on record) is
greater than S's want to preserve H's (or S's)
face to any degree, then S will want to
minimize the face threat of the FTA.
v) "Given the following set of strategies, the
more an act threatens S's or H's face, the
more S will want to choose a high-numbered
strategy; this by virtual of the fact that these
strategies afford payoffs or increasingly
minimized risk:
a. [This is an attempt to render a chart on
page 60--it's not verbatim]
b. Do the FTA
i. On the record
1) Without redressive action,
baldly (smallest estimated
risk of face loss)
2) With redressive action
(greater risk than baldly)
Preference organization: What conversation
analysis use to refer to the phenomenon that after
specific kinds of conversational turn, responses are
often strictly non-equivalent--one kind of response
(the preferred), is direct, abbreviated, structurally
simple and immediate. Others are dispreferred and
are indirect, elaborate structurally, and delayed.
• Atkinson and Heritage (1984: Part II)
• Levinson 1983: 332ff)
• Pomerantz (1975, 1978, 1984a)
• For my purposes, what's most interesting is
that a delay gives the first speaker a chance to
adjust--Davidson (1984), Pomerantz (1984b)-or withdraw (Goodwin 1979).
Virtual offence: Goffman's notion "which predicts
that the non-communication of the polite attitude
will be read not merely as the absence of that
attitude, but as the inverse, the holding of an
aggressive attitude" (B&L 1987: 33).
• Schegloff (n.d.) observes that there is a
tendency for "innocent" utterances to be
interpreted as complaints.
• "By orienting to the 'virtual offence', an
offender can display that he has the other's
interests at heart. Equally, a failure to orient
to the virtual offence counts as a diplomatic
breach. Thus is constructed a precise
semiotics of peaceful vs. aggressive intentions
(where the measure of precision is sometimes
in fractions of a second--see e.g. Davidson
1984), which in assigning such momentous
significance to what are often trivial
substantive acts requires a constant vigilance
over the manner in which social interaction is
conducted. This semiotic system is then
responsible for the shaping of much everyday
interaction, and in so shaping it, constitutes a
potent form of social control" (B&L 1987: 1-2).
Reading notes Page 1
(greater risk than baldly)
a) positive politeness
(lesser risk)
b) negative politeness
(greater risk)
ii. Off record (greater risk than on
the record)
c. Don't do the FTA (greatest estimated
risk of face loss)
vi) Since i-v are mutually known to all MPs, our
MP will not choose a strategy less risky than
necessary, as this may be seen as an
indication that the FTA is more threatening
than it actually is."
• In their (1987) forward, B&L say that other
researchers have "persuaded us that we may
have been in error to set up the three superstrategies, positive politeness, negative
politeness, and off record, as ranked
unidimensionally to achieve mutually
exclusivity" (B&L 1987: 18).
"The social valence of linguistic form has two
especially important sources: the intrinsic
potential impact that a specific communicative
intention may have on a social relationship, and
the ways in which by modifying the expression
of that intention participants seek to modify
that impact--such modification measuring for
participants the nature of the social relationship.
On this view a very considerable intentional
and strategic mediation connects linguistic
form with social relationships. In short,
language usages are tied to strategies rather than
directly to relationships, although relationships will
be characterized by the continued use of certain
strategies" (B&L 1987: 281).
Social order
Here's how Gumperz puts it in the forward to
Brown and Levinson (1987):
"A major reason for [interest in politeness], as
the authors define it, is basic to the
production of social order, and a
precondition of human cooperation, so that
any theory which provides an understanding
of this phenomenon at the same time goes to
the foundations of human social life"
(Gumperz 1987: xiii).
"What counts as polite may differ from group
to group, from situation to situation, or from
individual. If we can find some underlying
potent form of social control" (B&L 1987: 1-2).
• Goffman is fond of theater metaphors, but
there's another kind of performer he likes-the diplomat. In his 1971 work he develops the
notion of a virtual offence, which is the
"worst possible reading" of some action by A
that might trespass on B's
interests/equanimity/personal preserve. If
you orient to the virtual offence, you can
avoid a diplomatic breach.
• See Haviland (1977) on how people
Efficiency factor: Lakoff's idea that it is rude to ask
a superior to spend time/energy calculating the
illocutionary potential of an off-record request. B&L
don't see this as universal or intrinsic in negative
politeness as Leech (1983) and Lakoff (1977b) do.
About face
Face: "The public self-image that every member
wants to claim for himself, consisting of two related
• "negative face: the basic claim to territories,
personal preserves, rights to non-distraction-i.e. to freedom of action and freedom from
• "positive face: the positive consistent selfimage or 'personality' (crucially including the
desire that this self-image be appreciated and
approved of) claimed by interactants" (B&L
1987: 61).
• The notion of face is derived from Goffman
(1967) and the English folk term ("losing
face"). Watts (2003), of course, says Goffman's
definition is closer to what's going on than
B&L's interpretation.
• For Goffman, face is "the positive social value
a person effectively claims for himself by the
line others assume he has taken during a
particular image of self delineated
in terms of approved social attributes"
(Goffman 1955/1967, I think, mentioned in
Watts 2003: 124).
• "Face is a socially attributed aspect of self that
is temporarily on loan for the duration of the
interaction in accordance with the line or
lines that the individual has adopted. It is not
our personal construction of the self,
although the different faces we are required
to adopt in different interactions do
contribute towards that construction...If our
constructed role remains relatively stable
across interactions it will result in a form of
Reading notes Page 2
across interactions it will result in a form of
institutionalisation of the self. However, if
face is the 'condition of interaction, not its
objective', it is equally clear that we have an
obligation to maintain the faces of the other
participants in the interaction" (Watts 2003:
individual. If we can find some underlying
grammatical and social regularities which
account both for this type of variation and for
the recurrent patterns, we will have taken a
major step in demonstrating and not just
claiming the basically social nature of human
language" (Gumperz 1987: xiii).
For B&L, conflict and cooperation are fundamental:
"From a gross ethological perspective,
perhaps we can generalize somewhat: the
problem for any social group is to control its
internal aggression while retaining the
potential for aggression both in internal
social control and, especially, in external
competitive relations with other groups"
(Brown and Levinson 1987: 1).
○ See also Maynard-Smith on Origins of
Social Behavior, B&L say.
The hearer
Watts (2003) argues that B&L offer a production
model of politeness that leaves the hearer pretty
much out in the cold (only the hearer's face needs
are considered, says Watts 2003: 51).
Hearers in B&L aren't absent, of course (how could
they be?)
"The uses of each [politeness strategy] are
tied to social determinants, specifically the
relationship between speaker and addressee
and the potential offensiveness of the
message content" (Brown and Levinson 1987:
In their 1987 introduction, B&L say that they would
probably drop some of the speech act stuff (though
they still think it's handy as a shorthand). Part of
the reason has to do with the hearer:
"Speech act theory forces a sentence-based,
speaker-oriented mode of analysis, requiring
attribution of speech act categories where our
own thesis requires that utterances are often
equivocal in force" (Brown and Levinson 1987:
○ See Rosaldo (1982) for a critique of
speech act theory based on the Ilongot
of the Philippines--do they interpret
each other in terms of expectations of
group membership/role
structures/situational constraints and
For Watts, B&L are focused on a model person who
does things to guarantee:
• "The universal need of individual human
beings to be valued, respected, appreciated in
social groups, i.e. that the self-image that an
individual has constructed of her/himself
should be accepted and supported by others,
• "The universal right of individual human
beings to relative freedom of thought and
action, i.e. to perceived 'territory', in both the
literal and metaphorical senses of the term"
(Watts 2003: 101).
• The first is positive face, the second is
negative face.
Watts takes pains to say that B&L get face from
Goffman but don’t use it as he intended. For Watts,
the key to Goffman's notion of face is that it is "not
something that the individual somehow builds for
her/himself, which then needs to be supported and
respected in the course of interaction, but is rather
'public property', something which is only realised
in social interaction and is dependent on others"
(Watts 2003: 107).
• It is a mutual construct (see also de Kadt
1998: 176). "An interactant will not merely
need to avoid certain behaviours, but will be
expected to produce certain other behaviours"
(1998: 177, cited by Watts 2003: 107).
Accepting Goffman's notion of face means "we are
constrained to accepting that we are attributed face
socially in accordance with the line or lines we have
adopted for the purposes of the communicative
interaction. This leads to two logical conclusions,
firstly that we can be assigned different faces on
different occasions of verbal interaction, and
secondly that all social interaction is predicated on
individuals' face needs, i.e. that we can never get
away from negotiating facework" (Watts 2003: 259).
Significance of the work
"We believe that patterns of message construction,
or 'ways of putting things', or simply language
usage, are part of the very stuff that social
Reading notes Page 3
structures/situational constraints and
NOT as expressions of sincere
○ See Clark and Schunk (1980, 1981) for
ranking of indirect speech acts as B&L
suggest (contra Kemper and Thissen
1981). See B&L (1987: 142-144) for the
politeness predictions. See Walters
(1980) and Fraser and Nolan (1981) for
Spanish and Bates (1976) for Italian.
Power (and distance and ranking of
"In broad terms, research seems to support our
claim that three sociological factors are crucial in
determining the level of politeness which a speaker
(S) will use to an addressee (H): these are relative
power (P) of H over S, the social distance (D)
between S and H, and the ranking of the
imposition (R) involved in doing the facethreatening act (FTA)" (Brown and Levinson 1987:
• Grimshaw (1980a, b, c, and 1983) has the same
three factors as do (in different ways) Bates
(1976), Lakoff (1977b), Lakoff and Tannen
(1979) and Leech (1980, 1983).
Notice, in this that they are talking about relative
power of H over S. This is interesting (and where I
first began in my own inquiries), but how do we
define power? Is it always "power-over"? What
about "power-to"?
About power, B&L say to see:
• Falbo and Peplau (1980)
• Baxter (1984)
○ Greater politeness for friends. Slugoski
(1985) says this is because friendships
don't legitimize instrumental goals--so
B&L's "Distance" variable should
distinguish familiarity from affect.
• Holtgraves (1984)
○ High degree of encoded politeness
indicates higher reciprocal liking
between speaker and addressee.
usage, are part of the very stuff that social
relationships are made of (or, as some would
prefer, crucial parts of the expressions of social
relations). Discovering the principles of language
usage may be largely coincident with discovering
the principles out of which social relationships, in
their interactional aspect, are constructed:
dimensions by which individuals manage to relate
to others in particular ways" (B&L 1987: 55).
"In the case of sociolinguistics, the theory argues for
a shift in emphasis from the current preoccupation
with speaker-identity, to a focus on dyadic patterns
of verbal interaction as the expression of social
relationships; and from emphasis on the usage of
linguistic forms, to an emphasis on the relation
between form and complex inference….In the case
of linguistic pragmatics a great deal of the
mismatch between what is 'said' and what is
'implicated' can be attributed to politeness, so that
concerns with the 'presentational functions' of
language should be supplemented with attention to
the 'social functions' of language, which seem to
motivate much linguistic detail" (Brown and
Levinson 1987: 2-3).
"The implication for sociology and anthropology is,
first and most generally, that more attention should
be given to the interactional basis of social life, if
only to aid progress at other analytical levels--this
because the area offers significant links across the
divide between 'macro' and 'micro' levels of
sociological analysis (Gal 1983)" (Brown and
Levinson 1987: 3).
The following quote is very long, but I think it has
some good stuff in it, so I've transcribed all of it:
Rosaldo (1982: 23) argues that power is very
different in egalitarian and hierarchical societies.
B&L are prepared to add other things to P, D, and
R, but they say that the do capture most stuff.
They'd think about adding:
• Liking
• Presence of an audience
Reading notes Page 4
"The key problem in sociolinguistics is always
the origin and nature of the social valence
attached to linguistic form. Some
sociolinguists view this as a relatively
unmediated attribution of value on the basis
of the social value of the group with which
the linguistic forms are associated (Labov
1972c, Trudgill 1974a). Others see the choice
of form determined primarily by the social
characteristics of participants and setting, and
thus the form's valence derives from the way
in which it encapsulates those social
determinants. The form may then be used
outside the social context that usually
determines its use, to invoke 'metaphorical'
allusions to that context (Ervin-Tripp 1972;
allusions to that context (Ervin-Tripp 1972;
Blom and Gumperz 1972). We prefer a
somewhat less mechanical and less arbitrary
source for the social valence of message
forms. For us communicative intentions
have built-in social implications, often of
a threatening sort. What then becomes
interesting is how such communicative
intentions become constrained, for such
constraints, expressed by means of the
pragmatic resources of the language, show in
the construction of messages. Communicative
intentions, like all social goods, do not flow
smoothly in all directions through a social
structure; in deed part of what gives some
particular social structure its form is the
specific nature and distribution of such
constraints, as Lévi-Strauss (1968) has argued.
In language the constraints are more on form
than on content (or at least form provides a
more feasible area of study). The ways in
which messages are hedged, hinted, made
deferential, and embedded in discourse
structures then become crucial areas of study.
But such areas are also the concern of
pragmatics, the study of the systematic
relation of a language to context. The special
interest of sociolinguistics in our view is
in the differential use of such pragmatic
resources by different categories of
speakers in different situations. It is in this
way that we derive our slogan
'Sociolinguistics should be applied
pragmatics'" (B&L 1987: 280-281).
• Presence of an audience
○ Formality
For Watts (2003: 96), power is underspecified in
B&L. He suggests that their equation that "the
weightiness/seriousness of a face-threatening act is
a combination of the social distance between
speaker and hearer, the power differential between
the hearer and speaker, and the ranking of
• Wx=D(S,H)+P(H,S)+Rx, where x is the face
threatening act.
Watts says that people criticize B&L because they
want to use this equation for measuring, but that
B&L weren't really trying to do that--they were
trying to suggest how different politeness strategies
were distinguished. Watts' own criticism is more
that the imposition (Rx) depends on the power and
social distance factors.
• E.g., asking a boss vs. a stranger vs. a friend
for a cigarette. To assess the value for R, you
have to know D and P. Knowing P may or
may not depend on knowing the value of D.
• See also Watts, Ide, and Ehlich (1992a: 9).
• Social distance is probably not as useful
as the affective relationship (Holtgraves
1986; Holtgraves and Yang 1990; Brown and
Gilman 1989).
Building on Gilman (1960) on T/V pronouns:
• Intimacy: A and B both use T
• Social distance, non-intimacy: A and B both
use V
• Dominance: One uses T and receives V
• B&L reiterate that there is "an iconic relation
between asymmetrical social relations and
asymmetrical usage, that alone will not
explain the direction in which the particular
pronouns are used, or why symmetrical T
should have the value it does in contrast to V"
(B&L 1987: 45). B&L are skeptical about
historical conditions.
• This notion of asymmetry is important to me.
• "Symbols of intimacy (commensality,
grooming, approach and propinquity) are
used like the T pronoun, both as symbols of
intimacy and domination. And fitting
neatly into these patterns one can find the
use of positive politeness and bald-on-record
politeness strategies...being used both
symmetrically as symbols of equality and
asymmetrically (downwards, as it were) as
symbols of domination" (B&L 1987: 46).
On "fishing"--Drew (1984) gives the following
• "If A announces the acquisition of some new
furniture, and B then preempts an invitation
to come and see it by requesting permission
to do so, B conveys 'the essence of
sociability'--a pre-emptive display of caring
about what is important to the other. Thus by
reporting events that make such a display
possible and pertinent, A can make relevant
such a preemptive self-invitation without in
any way requiring it--B can quite
appropriately offer congratulations or other
appreciations of a lesser sort. Pre-sequences
and 'fishings' thus allow the off-record
negotiation of business with face implications
well in advance of the possible on-record
transaction" (B&L 1987: 40).
Reading notes Page 5
symbols of domination" (B&L 1987: 46).
• "Intimate stuff used non-intimately takes on a
different, but highly predictable, meaning,
namely the symbolism of dominance (a
prototype for which can be found, perhaps, in
the relation between parent and child)" (B&L
1987: 46).
Watts does an interesting thing combining
emergent networks and theory of practice--the
overlap has to do with power. This seems clear,
though I would probably like to expand the notion
of "power" along the lines of feminist political
philosophers who consider "power to" in addition
to "power over".
"Changing the value and/or structure of
network links in an emergent network is thus
equivalent to the exercise of power by a
member of the network and is what Bourdieu
meant by the term 'symbolic violence'.
Participants in verbal interaction will always
be involved in a struggle over the right to
exercise power over others" (Watts 2003: 155).
Some problems with B&L
"If we take some face-redressive goal like 'be
pessimistic about the success of the FTA', this
suggests that an utterance like 'You don't want to
pass the salt' should be polite; that it is not, of
course, is due to the fact that it attributes impolite
desires to the addressee: in short, our system 'overgenerates' and needs to be complemented with a
set of 'filters' that check that a chosen utterance
form has no impolite implicatures for other in this case it does because of a rather
complex reflexive reasoning that takes account of
the implied presumption about the addressee's
beliefs…We are thus less sanguine now than we
were about the possibility of real precision in this
area because of the enormous complexities of the
reasoning involved" (Brown and Levinson 1987: 11).
"We underplay the influence of other factors,
especially the presence of third parties, which we
now know to have much more profound effects on
verbal interaction than we had thought (see Bell
1984; Goffman 1981, on 'footing'; Levinson, n.d.)"
(Brown and Levinson 1987: 12).
See also Watts (1991, 1992, 1994, 1997a).
Strategies and Gricean assumptions
Grice assumes--as do his followers and most of
pragmatics--that conversation is rational and
efficient. I'm not totally comfortable with this from
a gut-check level, yet I don't really have a counterproposal at the moment. I don't really want to say
that emotion is irrational, which would be one
place I could go. It may be, as Damasio argues, part
of reasoning.
"The [Cooperative Principle] defines an
'unmarked' or socially neutral (indeed
asocial) presumptive framework for
communication; the essential assumption is
'no deviation from rational efficiency without
a reason'" (Brown and Levinson 1987: 5).
Against the assumption of rational/efficient talk,
"polite ways of talking show up as deviations,
requiring rational explanations on the part of the
recipient, who finds in considerations of politeness
reasons for the speaker's apparent irrationality or
inefficiency" (Brown and Levinson 1987: 4).
"In our model, then, it is the mutual
awareness of 'face' sensitivity, and the kinds
of means-ends reasoning that this induces,
Throughout their reassessment section, B&L talk
about overgenerating, for example: "the
asymmetry of strategy choice between
participants in asymmetrical social
relationships of authority/subservience, and
the exact nature of that asymmetry with more faceredressive strategies employed by the lower ranking
participant" (Brown and Levinson 1987: 12).
"Social interaction is remarkable for its emergent
properties which transcend the characteristics of
the individuals that jointly produce it; this
emergent character is not something for which our
current empirical models are well equipped" (B&L
1987: 48).
• See Bateman (1985) and Suchman (in press)
on cognitivism vs. interactionalism. B&L say
they also suffer from too much cognitivism.
Around page 232, B&L acknowledge that they have
treated interactions as if each utterance was its own
thing, not related to each other.
• "Such a view, promoted in no small part by
the philosophy of action and by the theory of
speech acts in particular, has been ably
criticized by Schegloff and Sacks (1973),
Reading notes Page 6
criticized by Schegloff and Sacks (1973),
Turner (1975), and Schegloff (1976). They
argue, essentially, that conversational
location, both in terms of 'local turn-by-turn
organization' (Sacks, Schegloff and Jefferson
1974) and in terms of overall conversational
structure (Schegloff and Sacks 1973), is a
crucial determinant of how an utterance is
understood" (B&L 1987: 232).
• Most important for me, perhaps, "The
conversational organizations that these
workers have discovered are extremely
sensitive to violation; turn-taking violations
(interrupting, ignoring selection of other
speakers, not responding to prior turns) are
all FTAs in themselves, as are violations of
opening and closing procedures" (B&L 1987:
• Here are the basics of how they propose to
remedy the situation:
○ Redefine "face threatening acts" to "face
threatening intentions", which can last
across turns.
○ "Plans--including conversational plans-are hierarchical, and conversational
understanding is achieved by
reconstruction of levels of intent beyond
and above and integrative of those that
lie behind particular utterances or
sentences" (B&L 1987: 233).
○ "If a breach of face respect occurs, this
constitutes a kind of debt that must be
made up by positive reparation if the
original level of face respect is to be
maintained. Reparation should be of an
appropriate kind and paid in a degree
proportionate to the breach" (B&L's
"balance principle, 1987: 236).
○ Perhaps my favorite little example is "if
one spills coffee on someone else's
clothes and they kindly point out how it
could happen to anyone, one feels even
more obligated to stress how sorry one
is" (B&L 1987: 238)--I don't know how
true this is, but I do like it.
of means-ends reasoning that this induces,
that together with the CP allows the inference
of implicatures of politeness" (Brown and
Levinson 1987: 5-6).
"Our framework presupposes the other great
contribution by Grice, namely his account of the
nature of communication as a special kind of
intention designed to be recognized by the
recipient (1971). That account itself presupposes
that what agents do is related systematically to
their intents, and thus that intentions of actors are
reconstructable by observers or recipients of
actions. The systematic relation is presumed to be
given by some rational means-ends-reasoning"
(Brown and Levinson 1987: 7).
• But what about the infinite regress of
recipients figuring out what the sender
figures the recipient figures the sender etc.?
○ See Schiffer (1972) on mutual
knowledge, Sperber and Wilson (1982,
1986) on a weaker idea of mutual
manifestness being sufficient.
• Is this psychologically plausible? See Clark
and Carlson (1982a, b) who have
"energetically defended the model," say B&L
(just use simple heuristics and you don't get
infinite regress).
• But we still run up about the idea that
reconstructing speakers' communicative
intentions by running a logic of practical
reasoning backwards. But no logical system
offers a way from going from conclusions
back to premises (it's not a symmetric kind of
○ "Thus, even if we had a perfect system
of means-ends reasoning, it would
remain a conceptual mystery how we
are able to reconstruct other agents'
intentions from their actions (Levinson
1985). Yet that we do so, or attempt to
do so, is hardly open to question, and is
presupposed by at least some uses of
the term 'strategy', including ours
(whatever its unclarities, see Riley 1981)"
(Brown and Levinson 1987: 8).
• Is our Gricean view of communication just
our folk-theory canonized as philosophy? See
Ochs (1984) and Candlin (1981).
○ Ochs and Duranti and Rosaldo talk
about personhood being different in
non-Western situations, where
intentional agent is dubious in its
applicability. B&L think these points
See Horn (1984) for taking Grice's four maxims and
the nine submaxims and reducing them to:
• Quality
• Quantity
• Relevance (enlarged--Sperber and Wilson
1986 and Wilson and Sperber 1981 try to drop
everything down to relevance: the natural
Reading notes Page 7
applicability. B&L think these points
just need a slight shift in the relative
importance of what is said vs. what is
implicated/attributed. "a shift tied to
the hoary sociological distinctions,
variously conceived, between
communities where positional status is
emphasized and those where persons
are treated as 'individuals'" (Brown and
Levinson 1987: 9-10).
○ B&L say that they will be disproven by
places where "only high-status
individuals have to take account of the
perspective of lower-status individuals"
(Brown and Levinson 1987: 10), but this
is the nature of power relationships that
the powerful have to consider the
people without power (see George
Orwell's Shooting an Elephant, for
example). Are they really saying they'd
need a society where EVERY and ONLY
high-status individuals take account,
Notice that B&L's claims put intrinsic ranking to
politeness strategies and does it in terms of
cost/benefit analysis. (In order of faceredressiveness: Positive < Negative < Off-record.)
everything down to relevance: the natural
human propensity to maximize the
informational value of environmental stimuli)
• Levinson's "Minimization and conversational
inference" takes these perspectives to task, I
As Watts (2003: 85) points out, B&L take up a
Model Person for discussing politeness and this is
in the spirit of Grice. The MP has "the ability to
rationalise from communicative goals to the
optimal means of achieving those goals. In doing
so, the MP has to assess the dangers of threatening
other participants'' (and hence her/his own) face
and to choose the appropriate strategies in order to
minimise any face threats that might be involved in
carrying out the goal-directed is a
production model" (Watts 2003: 85).
Criticisms about the Gricean model of pragmatics
include (Watts 2003: 111, see also his chapter 8):
• It doesn't adequately explain how addressees
derive the implicatures they do
• It doesn't account for other implicatures that
might also be derived
• It ignores the possibility than an addressee
might infer more than one implicature and
thus be faced with a potential dilemma in
deciding which of them is most appropriate
to the context of the utterance
"We wish to demonstrate the role of rationality,
and its mutual assumption by participants, in the
derivation of inferences beyond the initial
significance of words, tone, and gesture. It is our
belief that only a rational or logical use of strategies
provides a unitary explanation of such diverse
kinesic, prosodic, and linguistic usages" (B&L 1987:
• Here they are explicitly trying to counter an
undervaluing of "the complexity of human
planning" in sociological sciences, though
they admit to only scratching surfaces.
Rational decision-making also leads to an
antagonistic relationship between ego and alter,
says Werkhofer:
"This antagonism takes the form of a
dialogue, but of a strange kind of dialogue
that only takes place within the speaker's
mind: s/he generates as a first turn, what s/he
intends to say. This move remains tacit so
that the next move is not the addressee's
answer to the first one, but it is the speaker's
anticipation of what the threat to her or his
Reading notes Page 8
anticipation of what the threat to her or his
face would probably mean to the addressee.
The polite utterance is then the third move or
the speaker's second turn in this fictive
dialogue" (Werkhofer 1992: 166 cited in Watts
2003: 112).
Reading notes Page 9
More on politeness (Brown and Levinson 1987)
Friday, April 29, 2011
2:46 AM
Mixed situations
References to check out
"A possible explanation for the positive-politeness
impact of strategies like irony and understatement,
for example, would lie in two characteristics of
positive politeness: the reliance on mutual knowledge
to decode utterances, such mutual knowledge of
attitudes and values normally obtaining only between
in-group members, and the fact that positive
politeness uniquely allows the introduction of
extraneous material (not relevant to the particular
FTA in hand)" (B&L 1987: 20).
• See also Drew (1984) for notions of "up-take"
and "on-the-record"-ness.
Goffman (1967) on insisting that apologies and
other repairs are interesting.
• Owen (1983) covers apologies and remedial
interchanges more in-depth.
How are troubles broached and received? That is,
what happens when A announces a misfortune? There
are interactional problems. A could be upset, not in
control, not maintaining face. (Laughter is one
technique for dealing with this.) And how do you ever
disengage? (B&L 1987 40-41) cover this issue, but the
in-depth work is:
• Jefferson (1980, 1984a, 1984b)
• Jefferson and Lee (1981)
This is a metaphor worth exploring:
"We have talked about minute adjustments of
social distance, 'breaks and accelerators', as
the sort of fine-grained strategic manipulation
that routinely occurs in interactions" (B&L 1987:
On style:
"A strategic analysis...makes claims about the
non-arbitrary nature of style. On this view
[B&L's, btw] the features that co-occur in a
'style' are determined by strategic choices, and
the coherence of a style lies not necessarily on
the formal level but on the strategic level that
underlies the selection of forms" (B&L 1987:
○ See this built out in P. Brown (1976).
Giddens apparently suggests (1973: 15), that
interactional systematics are a retreat from central
issues of sociology. B&L argue that it's a crucial way
"in which abstract sociological concepts can be
Reading notes Page 10
McLaughlin, Cody, and Rosenstein (1983) and
McLaughlin, Cody and O'Hair (1983) on
"account" sequences after initial hostilities.
Bonikowska (1985a, b) and House and Kasper
(1981) on complaining.
On management of arguments: Boggs (1978),
Lein and Brennis (1978), Goodwin (1980a, b, 1982,
1983), Goodwin and Goodwin (in press), Jackson
and Jacobs (1980, 1981), Jacobs and Jackson
(1982), Maynard (1985).
Preference for agreement in conversation:
Pomerantz (1984a). But I think one should also
see Sacks.
• Intensifiers in AAVE: Labov (1984)
• Discourse particles and evidentials:
Goldberg (1982), James (1983), Gibbons
(1980), Wierzbicka (ed, in prep).
Think about trivializing adjectives in Lakoff. See
empirical results in:
• Dubois and Crouch (1975)
• Crosby and Nyquist (1977)
• Brouwer, Gerritsen and de Haan (1979)
• Edelsky (1979)
• Brouwer (1982)
• Baroni and d'Urso (1984)
Discover social norms by looking at their
systematic violation:
• Garfinkel (1972)
• Heritage (1984b: Ch 4)
• Gumperz
If you assume contra B&L that patterns of
message construction have purely rule-based
origins, then you have "alternations" and "cooccurrence rules" (Ervin-Tripp 1972) or rules for
"in which abstract sociological concepts can be
related in a precise way to social facts. We would like
our endeavor to be seen as an attempt to build one
arch in one bridge linking abstract concepts of social
structure (whether these are analyst's concepts or
member's concepts) to behavioual facts" (B&L 1987:
"The social value of the linguistic form of messages
can only be ascertained by looking at such forms as
tools for doing things, and asking what kinds of
things a given form could be doing" (B&L 1987: 281,
though in an endnote, they point out that Hymes has
repeatedly said this (1974a, for example) and he got it
from Burke, he says--it still gets ignored in research
• This seems in keeping with folks like Paul Drew,
who would like to drop talk of "meaning" and
focus on "action".
How crucial are families as a site for setting up what
happens? I probably ignore this too much. See also
B&L (1987: 44).
One view of pragmatics (Wilson and Sperber, and
probably Levinson?) has pragmatics coming in to help
determine what the proposition being expressed
means and then again to calculate the
indirect/contextual implications of that proposition
(B&L 1987: 49).
• "When we first wrote, the major justification for
the bifurcation of the theory of meaning into
semantics and pragmatics was the basic Gricean
observation that what is 'said' is typically only a
part of what is 'meant', the proposition
expressed by the former providing a basis for
the calculation of the latter" (B&L 1987: 49).
What do we do with these "facts" (B&L 1987: 57, but
maybe they aren't facts at all):
• If I regard something as a small request, I'll
likely stress in-group membership and social
similarity (B&L use the "Let's have another
cookie then" and "Give us a time", though these
rile me, personally).
• Bigger requests gets formal politeness ("the
conventionalized indirect speech acts, hedges,
apologies for intrusion, etc.").
• Requests we maybe shouldn't even be making
we use indirect expressions (implicatures).
• B&L say the same thing happens with criticisms,
offers, complaints, etc. All of these, they say,
have a strategic orientation to participants' face.
Reading notes Page 11
occurrence rules" (Ervin-Tripp 1972) or rules for
speech events (Hypes 1972, articles in Bauman
and Sherzer 1974).
• But social conditions are supposed to
determine the application of rules and
don't do that all the time. Rule violations
should therefore be attended to in
interaction, but many rule violations don't
have any such attending to.
Check out Gordon (1983) in Language in Society:
"Hospital slang for patients".
• I'm not exactly sure why, but I put a big red
flag on this one. So really check it out.
Grice (1973) is a conference paper on
"Probability, desirability, and mood operators"
that looks interesting.
See Ortner (1984) in Comparative Studies in
Society and History: "Theory in anthropology
since the 60s".
Sen (1979) has "Rational fools" in a book by
Hahan and Hollis, but it's also findable in
Philosophy and Public Affairs (1976-1977).
For a plan for more interactional sociolinguistics,
see Gumperz (1982a, b).
"Deference is not encoded in language by the use
of arbitrary forms, but by the use of motivated
forms" (B&L 1987: 23, see Haiman 1985: 154).
Scollon and Scollon (1981) find among
Athabaskans that positive politeness is naturally
escalated and hence unstable, while negative
politeness doesn't escalate and so it's stable.
They conclude that they can't be put on a
unidimensional scale. B&L think they can be
(B&L 1987: 18).
Reading notes Page 12
Critiques of Brown and Levinson by Watts (2003)
Wednesday, April 27, 2011
12:05 PM
Watts, R. 2003. Politeness. Cambridge: Cambridge University Press.
Often in dialogue with:
Brown, P. and S. Levinson. 1987. Politeness. Cambridge: Cambridge University Press.
Definitions and examples
Watts (2003) comes across as fairly cranky--it's a
tone that is fairly off-putting, at least to me.
In definitions of politeness, Greeks stress:
• Expression of intimacy
• Display of warmth and friendliness
But English concepts are broader:
• Consideration towards others
• Formality
• A discrete maintenance of distance
• A wish not to impose upon addressees
• Expression of 'altruism, generosity, morality,
and self-abnegation' (Sifianou 1992a: 88
summarized in Watts 2003: 14-15)
Watts' aim isn't really to be a production model
(what he calls a B&L goal), nor a blueprint for
interpreting some-but-not-other linguistic
expressions as politeness realizations. He doesn't
like politeness being a "social fact"--he's trying to
promote a theory that offers "ways of
recognising when a linguistic utterance might
be open to interpretation by interlocutors as
'(im)polite'...It aims to provide the means of
assessing how lay participants in ongoing verbal
interaction assess social behaviour that they have
classified as (im)polite utterances as positive or
negative" (Watts 2003: 143).
Three kinds of capital:
• Material: money, property, goods, stocks,
profits (wherever goods are exchanged for goods
is a material marketplace)
• Cultural: education, skills, knowledge, etc.
Watts argues that speakers and addressees work
(wherever these things are acquired is a cultural
together to create some form of common
• Social: network of relationships, quality of those
understanding among themselves, even if that
(so social marketplaces are fields where there
means they understand they can't or won't ever
are family, friendship groups, political
organizations, clubs, etc).
"An utterance made by a speaker and directed
• No marketplace is ever uniquely one of these
at an addressee is a social act, and an
things, though.
addressee deriving a set of inferences from
that utterance to enable her/him to respond Politic behavior: "Linguistic behaviour which is
in some way is carrying out another kind of
perceived to be appropriate to the social constraints
social act. But both acts are essential to socio- of the ongoing interaction, i.e., as non-salient" (Watts
communicative verbal interaction, and both 2003: 19 ,see also Watts 1989c, 1992). This seems to be
acts are embedded in the ongoing emergent similar to Fraser and Nolen's Conversational Contract
development of an interpersonal relationship. (though Watts says in Chapter 3 that it's different).
For Watts, polite and impolite salient behavior is
The goal of a theory of linguistic
politeness which takes (im)politeness1 as what goes beyond what is expected (in either a
its starting point should not be to explain positive or a negative direction...I'm not sure how well
this axis is built out).
why speakers say what they say and to
• Here's a claim that's probably a bit too strong:
predict the possible effects of utterances
on addressees. It should aim to explain
• "Most forms of social interaction have become
how all the interactants engaged in an
institutionalised and...the appropriate discursive
ongoing verbal interaction negotiate the
practices are known to us beforehand" (Watts
development of emergent networks and
2003: 20).
evaluate their own position and the
positions of others within those
Conversational Contract: "On entering into a given
Reading notes Page 13
positions of others within those
networks. (Im)politeness then becomes part
of the discursive social practice through
which we create, reproduce and change our
social worlds." (Watts 2003: 255).
One of the things I like best in Watts is the idea of
expectations and the room for differing
interpretations (I'm not as sold on the economic
metaphors that are so central for him, though):
"A speaker might use a linguistic expression
which s/he intends to be heard as more than
is necessary to uphold the levels of linguistic
behaviour appropriate to the discursive
situation, i.e. as polite, but the hearer may
not interpret the utterance in the way it is
intended to be interpreted. Alternatively, s/he
may very well assign the correct
interpretation but derive a set of inferences
from it that display a negative evaluation. The
speaker may be paying with politeness for
devious reasons that give the hearer reason to
doubt the sincerity of the utterance" (Watts
2003: 253).
Conversational Contract: "On entering into a given
conversation, each party brings an understanding of
some initial set of rights and obligations that will
determine, at least for the preliminary stages, the
limits of the interaction" (Fraser and Nolen 1981: 93-94
cited in Watts 2003: 78).
Politeness2: Mutually shared forms of consideration
for others. Watts wants to specify that if I'm the
speaker and you're the "other", we might have
different ideas about consideration involved in the
same act.
Demeanour: The ceremonial way the ego presents
themselves to the alter (dress, bearing, style, etc).
They are a behavioral mask that the ego needs to
adopt to interact successfully with alter, given an alter
with higher social status. They are accoutrements, but
important in conveying social skills.
Symbolic power: "Every power which manages to
impose meanings and to impose them as legitimate by
concealing the power relations which are the basis of
its force" (Bourdieu and Passeron 1990: 4 cited in
Watts 2003: 151).
Watts argues that speakers and addressees work
together to create some form of common
understanding among themselves, even if that
means they understand they can't or won't ever
Habitus: A state of being; a demeanor, manner or
bearing; the style of dress or toilet (in Latin). "The set
of dispositions to act in certain ways, which generates
cognitive and bodily practices in the individual"
(Watts 2003: 149). There are two aspects in Bourdieu's
idea: (i) the habitus shapes how individuals
"An utterance made by a speaker and directed internalize social structures to use them in ongoing
interactions, (ii) in instances of ongoing interaction,
at an addressee is a social act, and an
the habitus generates practices and actions--so it's
addressee deriving a set of inferences from
that utterance to enable her/him to respond responsible for both the reproduction and the change
in some way is carrying out another kind of
of social structures.
social act. But both acts are essential to sociocommunicative verbal interaction, and both M- m- m- money changes everything
acts are embedded in the ongoing emergent At the heart of Watts' work on politeness1 are notions
development of an interpersonal relationship. of metaphorical goods/payment. But he is cautious
The goal of a theory of linguistic politeness
to point out that these metaphors shouldn't imply
which takes (im)politeness1 as its starting
harmony and equilibrium as some have taken them to
point should not be to explain why speakers be (Sifianou, for example).
say what they say and to predict the possible
• See also Werkhofer (1992) and O'Driscoll (1996)
effects of utterances on addressees. It should
for explicit comparisons between the social
aim to explain how all the interactants
power of politeness and the social power of
engaged in an ongoing verbal interaction
money (discussed in Watts 2003: 111).
negotiate the development of emergent
• One way of seeing it is to see politeness as a
networks and evaluate their own position and
factor mediating between individual and the
the positions of others within those networks.
group (like money).
(Im)politeness then becomes part of the
• "As a private good, the key to understanding the
discursive social practice through which we
nature of money is to consider the ways in
create, reproduce and change our social
Reading notes Page 14
create, reproduce and change our social
worlds." (Watts 2003: 255).
Rituals and probabilities
nature of money is to consider the ways in
which the individual maximises its utility as a
symbolic resource in the exchange of goods, but
as a symbolic resource it is 'a social institution
and quite meaningless if restricted to one
individual'" (Watts 2003: 115 quoting Simmel
• Like money, politeness is something socially
constituted that can still "itself motivate and
structure courses of action, feeding into social
processes and, thus, into the very conditions of
its own existence or maintenance" (Werkhofer
1992: 190 cited in Watts 2003: 115).
B&L "play down the importance of politeness
routines by stressing the 'generative' production of
linguistic politeness" (B&L 1987: 43), they
acknowledge that formulae are clearly important in
folk notions (and to Goffman who stresses how
ritualistic politeness is--ritual being repetitive or
pre-patterned behavior). We see this also in the
B&L's discussion of preference organization:
repetition and probabilities are clearly
important. Perhaps more than they realize.
One starting point for a theory of ways in which
• B&L would like to see interpersonal rituals as people maximize politeness devices as a symbolic
the omnipresent model for rituals of all kinds resource is through Werkhofer's analogy of money:
i) Politeness, like money, is "a socially constituted
(contra Goffman who sees interpersonal
ritual as a residue from an earlier ritually
dominated form of public life).
ii) Politeness, like money, is "a symbolic medium in
the sense that its functions originally derive
Watts is specifically pulling ideas of habitus and
from an association to something else, namely
social practice theory from Bourdieu. And this
to values"
iii) Each is "historically constituted and
could connect well to probabilities and
reconstituted; its functions and the values it is
distributions of experiences.
associated with are essentially changeable ones"
"Participants enter verbal interaction in a
iv) "During its history, the functions of politeness
specific social situation with a knowledge
turn into a power of the medium in the sense
gained from previous experiences about
that it may, rather than being only a means to
what forms of social behaviour are
the ends of an individual user, itself motivate
appropriate and inappropriate to that
and structure courses of action"
type of situation. Their knowledge is
v) The chances of the user being able to "master
constructed through their own personal
the medium completely...will be diminished"
history and the way it has been linked in the
(Werkhofer 1992: 190 cited by Watts 2003: 144-past with objectified social structures" (Watts
and also on page 115, wow).
2003: 144-145).
Watts seems to use three sentence moods, which
"Social practice is carried out within social
focus on how sentences are forms of action--or better
fields, and individuals and groups are defined yet interaction. There's something kind of interesting
by their relative positions in them. Fields are in this division, though interrogatives vs. imperatives
probably need to be better defined:
thus arbitrary social organisations of space
and time, and they are the sites of constant
i) "Assertives give a value and can therefore expect
struggles over capital. Capital can thus be
the payment of some other equivalent value.
seen as an incorporation of resources, which
ii) "Interrogatives request a value but cannot
become part of the individual's habitus. They
automatically expect the payment of that value.
can be loosely grouped into material, cultural
If the value is given, however, some form of
or social 'marketplaces', in which three kinds
return payment can be expected by the giver.
of capital are at stake, material capital,
iii) "Imperatives request a value, which may or may
cultural capital and social capital" (Watts
not be in the form of a linguistic utterance, and
2003: 149).
generally do expect the payment of that value."
(Watts 2003: 154)
"Any new occasion of social interaction
enacts and therefore reproduces earlier
This connects Watts to Relevance Theory's idea of a
verbal utterance as an ostensive, informative act. "The
similar forms of interaction, but is at the
utterer is giving something to the addressee on the
same time always open to discursive
Reading notes Page 15
same time always open to discursive
negotiation that might help to reconstruct
the interaction type...the interaction type is
itself always a locus of struggle with respect
to what constitutes that form of social
interaction" (Watts 2003: 20-21).
The equation of social practice Watts uses is:
[(habitus)(capital)] + field = practice
What this means is that the ways people carry out
social acts that make up instances of interaction
depend on the prior histories of those who are
engaging in the interaction. Practice is a product of
the objectified social structures of the field and the
habitus and forms of capital of the participants.
"Among the objectified social structures of
the field are institutionalised forms of
behaviour, rights and obligations of the
individuals interacting within that field and
the power structures that form part of the
field. Practice therefore depends on the
amount of knowledge about those objectified
structures that the individual has internalised
as part of her/his habitus.
"Specific modes of behaviour have become
canonical as part of the objectified structures
of the field and...they represent reproductions
of discursive formats that have become
institutionalised as expectable behaviour for
interaction" (Watts 2003: 256).
(See also Watts' discussion of Bourdieu's equation
on page 150.)
"Falling out of line constitutes a break in the politic
behaviour which is interpretable by the
interactants as an offence and as damage to the
face of one or more of the interactants including
the interactant who has fallen out of line. This kind
of behaviour is often evaluated as rude or impolite."
(Watts 2003: 131 though he points out that facework
isn't 1:1 with (im)politeness--supportive facework
may have nothing to do with (im)politeness, for
"In any society, whenever the physical possibility of
spoken interaction arises, it seems that a system of
practices, conventions, and procedural rules comes
into play which functions as a means of guiding
and organizing the flow of messages. An
understanding will prevail as to when and where it
utterer is giving something to the addressee on the
justified assumption that the addressee will give
something back either in the way of a linguistic
utterance or in some other way, even in a situation of
conflict...thus creating and sharing a common
understanding. Every verbal interaction is therefore
an exchange of utterances, which the interactants can
assume to be in some sense meaningful...Every
utterance either expresses at least one semantic
proposition or directs attention to some semantic
proposition(s) or, alternatively, does both these things
at the same time" (Watts 2003: 153).
More critiques
Watts argues that we have to get rid of Gricean
assumptions about social interactions being geared
towards cooperation (Watts 2003: 20), though I'm not
sure this cooperation can't be saved.
For Watts, only Leech and B&L have positions that are
firm enough to test on real data (2003: 63).
The politeness literature is all over the place with
regards to what counts as a culture (languages,
genders, age groups, Western Europeans, etc.). It
remains undefined, usually (Watts 2003: 101).
More criticism of B&L from Werkhofer (1992,
summarized in Watts 2003: 113):
• What about bystanders? Is it really just the
speaker and the hearer? What about the social
context of the utterance?
○ B&L are ready, in their 1987 foreword, to
add audiences in, they say.
• What's the distinction between ritualized,
formulaic utterances of politeness and
utterances that aren't typically interpreted as
being about politeness but are interpreted as
such in a particular interaction. The distinction
between "a socially and historically prepatterned, highly conventionalised utterance
and an individually or even idiosyncratically
generated one" (Werkhofer 1992: 168).
• What about online processing? Speakers are
monitoring what is said and may go back to
correct assumptions.
• Clark and Schunk (1980, 1981) give evidence that
there is no connection between the weightiness
of the imposition implied by the utterance and
the cultural ranking of the imposition being
placed on the addressee by the speaker. "Costly"
favors don't always give rise to more polite
utterances. It may be more about what's
Reading notes Page 16
utterances. It may be more about what's
considered just and right.
• Power and distance become reified, "taking on
an existence outside the social sphere of the
interactants rather than being themselves
constructed and/or reproduced through and in
the interaction itself. They are not adequately
defined, and Brown and Levinson do not
consider the function that polite behaviour itself
may have in reconstructing them" (Watts 2003:
○ A fun example: a barrister saying to a
witness: "Would you be so kind as to tell
us where you were on the night of the
thirteenth of January last?" isn't really
showing elaborate deference to the
witness. It's irony and
constructs/reproduces relationships of
power and authority. The witness has to
answer the question. The politeness is a
form of social control.
• The scale of politeness strategies is questionable
(but B&L do acknowledge this in their 1987
understanding will prevail as to when and where it
will be permissible to initiate talk, among whom,
and by means of what topics of conversation"
(Goffman 1955/1967: 33-34, cited in Watts 2003:
Imagine someone else is in your theater seats.
There's been some confusion between P51/P52 and
R51/R52 or there's some other code on the tickets,
easily missed. What participants would expect to
happen in the situation is politic behavior (not
Watts objects to politeness being an objective
concept that is universal. He dislikes the taking up
of face-threat mitigation (B&L, here adopted by
Blum-Kulka who discusses how it is realized
"On a theoretical level this means that
systems of politeness manifest a culturally
filtered interpretation of the interaction
between four essential parameters: social
motivations, expressive modes, social
differentials and social meanings...Cultural
notions interfere in determining the
distinctive features of each of the four
parameters and as a result significantly effect
the social understanding of 'politeness' across
societies in the world" (Blum-Kulka 1992: 270
cited in Watts 2003: 71).
For him, B&L don't look at how interlocutors
struggle over politeness (or whatever term they
use). What is polished? What is (in)appropriate?
The universal assumption for Watts is that all
cultures will have forms of behavior that
members classify as mutually shared
consideration for others. And others which
violate these principles of cooperation and
consideration (Watts 2003: 14).
• This is somewhat interesting since B&L have
been criticized as having "an overly
pessimistic, rather paranoid view of human
social interaction" (Schmidt 1980: 104) where
"social interaction becomes an activity of
continuous mutual monitoring of potential
threats to the faces of the interactants, and of
devising strategies for maintaining the
interactants' faces--a view that if always true,
could rob social interaction of all elements of
pleasure" (Nwoye 1992: 311).
"In virtually all the models of linguistic politeness on
the market, (im)politeness2 has become a set of
strategies available to speakers to enable them to
achieve certain communicative goals whilst retaining
interpersonal harmony, enhancing feelings of comity
and goodwill, showing the requisite levels of
cooperation, etc." (Watts 2003: 254-255).
• Watts thinks this is wrongheaded. How do you
process linguistic expressions? What are the
short-term effects and long-term effects? Most
things labeled by researchers as polite aren't
recognized as such by participants. A focus on
classification that doesn't match the
participants' classification is suspect.
"Brown and Levinson work from the concept of wants
based on what they call 'personality', which an
individual has developed prior to the interaction,
whereas Goffman works from a notion of the ongoing
construction of the individual's self-image contingent
on social factors. Brown and Levinson seem to be
thinking of the self as a stable core of values lodged
somewhere in the individual, whereas for Goffman
self is far less 'real' and is constantly renegotiable"
(Watts 2003: 105).
For Goffman, a person makes a claim for a positive
social value "which is constrained by the 'line' others
interpret him to be taking during the course of the
Reading notes Page 17
pleasure" (Nwoye 1992: 311).
• "Being defined as static entities that
determine polite meanings, these variables
[social distance, power, rate of imposition]
represent a narrow approach to social
realities, an approach that neglects the
dynamic aspects of social language use-aspects that may have no systematic status in
the traditional view, but should be at the very
heart of a modern one" (Werkhofer 1992: 176).
• All three of these cited by Watts (2003: 100).
Watts argues also that nothing is inherently
polite--not even compliments are necessarily
going to be evaluated as positive, supportive
behavior (see his chapter 7).
So we can see this all as an argument against the
objectification of a notion of politeness. For Watts,
politeness is a discursively disputed term. "Raising
the term '(im)politeness' to the status of a
theoretical concept in linguistic pragmatics and
sociolinguistics shifts the concern with politeness
phenomena away from individuals and the
everyday social acts they are involved in and places
it above or beyond those individuals...It places
social structures beyond the individuals involved in
social interaction. This, in turn, logically entails
that social structures are in some sense preexistent
factual entities" (Watts 2003: 254).
interpret him to be taking during the course of the
interaction. That social value is dependent on the
other 'members', and it can change from one moment
to the next. It is an image of the self constructed in
accordance with social attributes approved by others,
and it may be unstable and changeable" (Watts 2003:
"The theory of practice as practice insists, contrary to
positivist materialism, that the objects of knowledge
are constructed, not passively recorded, and, contrary
to intellectual idealism, that the principle of this
construction is the system of structured, structuring
dispositions, the habitus, which is constituted in
practice and is always oriented towards practical
functions" (Bourdieu 1990: 52 cited in Watts 2003:
• "Just as the individual only exists by virtue of
her/his own specific history, so too does society
only 'exist' by virtue of the history of previous
social interaction, which lends those forms of
behaviour the impression of objective validity.
Throughout the social history of an individual
s/he constructs the idea of an objectified
'society' with objectified social structures that
sanction the ways in which s/he behaves in
ongoing interaction" (Watts 2003: 143).
• Contrast this with Parsons, who has society
regulating individual instances of social
interaction. Social structure is a set of givens
(you can discover and then manipulate). So
individuals are pawns in a chess game of society,
says Watts. This is a top-down way of looking at
• See Eelen (2001) for more on Parsons.
See Hsien Chin Hu (1944) in American Anthropologist
for the historical development of the notion of 'face'
in Chinese.
Reading notes Page 18