Making social moves

CHAPTER 4
Making social moves
When people converse with one another, they are making various kinds
of social moves. As we saw in the preceding chapter, this is why con-
versational access is so important and also why it can be problematic.
In this chapter we will look in more detail at different kinds of lin-
guistically mediated social moves, what analysts call speech acts. Speech
acts are firmly embedded in social practice. Each particular utterance
enters into the discourse and into the plans being developed in that
interaction and, in turn, into a larger landscape of social practice, in-
cluding gender practice. The work each utterance does is not a matter
simply of its form, its linguistic properties. Each utterance is part of
the social situation in which it occurs, and its significance unfolds in
the emergent history of the discourse and interaction that it enters.
We have seen that gender structures people’s access to participation in
situations, activities, and events, hence to their opportunity to perform
particular speech acts legitimately. In this chapter, we will see how the
acts themselves accomplish gender.
Talk is often thought of as quite distinct from action. ‘‘He is a man of
action, not words.’’ ‘‘She’s all talk, no action.’’ (The pronouns here reflect
language and gender ideologies familiar to many English speakers.) A
sharp dichotomy between talk and action is, however, problematic. It
is true that simply to say ‘‘Let’s have lunch together sometime soon’’
need not result in any lunchtime meeting. Perhaps the utterance is in
some way a figure of speech, the overt literal proposal to have lunch
not really intended to lead to a lunch but just to indicate that the
relationship between the interlocutors should be seen as continuing
to be cordial. Even in such pro forma cases, however, the words do
something. What precisely those words do on any particular occasion
of their utterance depends on the social relations of the people who
are talking and on what they are doing together, both during this

interaction and more generally. Perhaps one reason that people are
sometimes tempted to identify talk with inaction is that words alone
really do not do anything. Their often considerable force derives from
129
130 Language and Gender
their being embedded in social practice. Not surprisingly, that force is
implicated in gender practice in complex ways.
Aspeech act is a move in a continuing discourse among interactants.
Like other acts, it moves their relationship along one more step, moves
their mutual connection to ideas and ideologies, and it moves their
accomplishment of things in the world as well. Amove can be a com-
pliment, a complaint, an insult, a request, a command, a criticism, a
question, a one-up, an exclamation, a promise -- these are some of the
kinds of speech acts that linguists and philosophers have discussed.
We will sometimes refer to them as social moves in order to emphasize
their place in a larger discourse and as part of socially-oriented plans
and strategies. But we also refer to them as moves because there are
meaningful interactive moves, such as waving, raising one’s eyebrows
or handing someone a pen, that do not use language and are thus not
speech acts as ordinarily understood. As we saw in the last chapter, con-
versational conventions can make silence a meaningful social move. It
might insult or compliment. It might or might not be accompanied by
meaningful facial expressions or other bits of ‘‘body language’’ (which
also, of course, can accompany speech). Sometimes we will talk about
speech acts when we really mean communicatively significant social
moves more generally. That is, our interest is in meaningful interac-
tive moves that often -- perhaps canonically -- involve speech but may
also be made in other ways. The gendered division of labor can mean
that certain kinds of speech acts are seen as more the province of one
sex than the other or that particular ways of performing them enter

into gender practice, or that their effect is different depending on who
performs them.
Repeated moves of a particular type can grow into an activity -- a
series of one-ups can become a competitive conversation, a series of
complaints can become a gripe session, a series of criticisms from one
person to another can become a dressing down, a series of statements
on some topic uttered by the same person can become a lecture.
Speech act theory
Philosopher J. L. Austin (1962) initiated the systematic study of speech
acts in his well-known exploration of ‘‘how to do things with words.’’
To undermine the view that speech and action are opposed to one an-
other, Austin drew attention to what he called performative utterances.
Aperson with the proper institutional authority, he pointed out, can
say ‘‘you’re hired!’’ and thereby give a job to the addressee. The utterance
itself, given the proper institutional setting and a speaker authorized
131 Making social moves
to produce it, brings it about that the addressee has indeed been hired.
Those words start a chain of events that will, if the addressee accepts
the offer, lead to the addressee’s showing up for work and getting a
paycheck some time thereafter. Hiring and firing, naming boats and
babies, pronouncing judgments in a courtroom, marrying two people
or joining them in a domestic partnership: verbal performances are
central to doing such things. And, of course, we have already observed
in the last chapter that gender often affects which people will be in-
stitutionally empowered to bring off particular kinds of verbal perfor-
mances. Although Austin (like most analytic philosophers of his era)
generally spoke of individual speakers as if their social identities and
relations to one another were irrelevant to their status as speakers (or,
more generally, as actors), he spoke of overt performatives like promise
or christen as ‘‘trouser words,’’

1
gendering the notion of performativity
at its birth.
As we noted in chapter one, gendering people can be thought of as
accomplished through a series of acts, many of them linguistically me-
diated. ‘‘It’s a girl,’’ pronounces the medical professional at the moment
of birth, and indeed it is thereby made a girl and kept a girl by sub-
sequent verbal and nonverbal performances of itself and others. In de-
veloping the per
formative theory of gender mentioned in chapter one,
Judith Butler (1990) draws inspiration (and nomenclature) from
Austin’s
theory of perfor
mative utterances. Butler develops Austin’s important
insight that performativity is not just a matter of an individual’s want-
ing to do something by saying something. Verbal as well as other per-
formances come off, acquire their meaning, and do their work, because
they draw on discourse histories of similar performances, reiterating
elements that have worked similarly in the past. In that reiteration,
however, there is the possibility of individuals going beyond the con-
straints of the social or linguistic system they have inherited, perhaps
ultimately thereby contributing to changing it. As Butler (1990, p. 145)
puts it, ‘‘In a sense, all signification takes place within the orbit of the
compulsion to repeat; ‘agency’, then, is to be located within the possi-
bility of a variation of that repetition.” (We will return to Butler’s ideas
about performativity in chapter nine.) Austin focused on the speaker’s
agency but later work has emphasized that what speakers can do with
their words is constrained (though not fully determined) by linguistic
and other social conventions.
1 Trousers were at that time very much masculine apparel and symbolized

authoritative action. Compare ‘‘she wears the pants in that family,’’ a line often used in
the same era to criticize a woman who wielded what the speaker saw as inappropriate
authority in her household, usurping the place of the legitimate pants-wearer, the man
of the house.
132 Language and Gender
Not all utterances affect the world as dramatically as overt perfor-
matives like ‘‘I hereby pronounce you husband and wife’’ (when uttered
by a person licensed by the state to perform marriages). Nonetheless,
Austin observed, all utterances are indeed actions. He distinguished
three different kinds of acts involved whenever someone says some-
thing. There is a locutionary act. The speaker produces an utterance --
a stream of sound or hand gestures or marks on a page or computer
screen -- as a particular linguistic expression with a particular structure
and (literal) meaning. The locutionary act sets the stage for the illocu-
tionary act, what is done in saying whatever has been said. Have you
claimed something or inquired? Have you promised or threatened or
warned? Invited or implored or commanded? Expressed your anger or
your pleasure? Praised or criticized? Apologized or empathized or com-
plained or teased? In saying something and meaning something by it,
a speaker always performs one or more such illocutionary acts. There
will also be perlocutionary acts accomplished by saying something. You
may persuade someone of your views, frighten or annoy them, cheer
them up or comfort them, move them to some kind of action of their
own (e.g. to follow your suggestion or respond to your request), impress
them with your wisdom, fan their love of you.
The literature on speech acts (e.g. Searle 1969) generally focuses on
illocutionary acts (e.g. promising or requesting). If we just pay attention
to illocutionary acts, however, the social character of speech acts may
be underrated. To come off, to work, it looks as if an illocutionary act
needs only to be comprehended (assuming that certain preparatory

conditions are met). So, for example, if the speaker is giving a party
at some future time and says ‘‘Please come to my party,” then the
addressee who understands what is said is thereby invited to the party.
The addressee does not have to welcome or to accept the invitation, but
it has successfully been issued. Comprehension is not trivial, of course.
The speaker cannot always guarantee that the interpreter will figure
out the illocutionary point of what has been said: whether, for example
there is just a report offered by ‘‘I’m thirsty” or a further request for a
drink. But generally, if the illocutionary point is understood, then the
illocutionary act has been performed. (This assumes that the speaker
is indeed empowered to perform the illocutionary act in question, not
always a safe assumption.) Perlocutionary acts, however, are inescapably
social: their coming off as the speaker intends requires very active
participation from the addressee -- for example the addressee’s coming
to the party or getting a glass of water for the speaker. Perlocutionary
acts have to do with effects that go far beyond simple understanding. It
is obvious that gender and other aspects of social standing will affect
133 Making social moves
success in performing intended perlocutionary acts. It is less obvious
but also true that gender and other social attributes of speakers may
enter into success in getting particular locutionary and illocutionary
acts to come off as intended.
2
In our everyday taxonomies, illocutionary and perlocutionary acts
are not so sharply distinguished as this discussion might suggest. For
example, a threat and a promise are two different kinds of illocution-
ary act. But they both commit the speaker to some future course of
action. They are distinguished only by whether or not the addressee
is presumed to be negative or positive about the speaker’s commit-
ment. The person who aims to threaten intends to scare the addressee

in contrast to the promiser, who aims to please. Scaring and pleasing
are distinct desired perlocutionary effects. The same words may be a
threat addressed to one person and a promise addressed to another.
You cannot tell if someone is threatening someone else by simply ob-
serving what words are uttered. And indeed a speaker may be neutral as
to whether the addressee will welcome the commitment made, simply
expressing the commitment with no intention to scare or to please. En-
glish words for speech acts often convey information about both the
kind of illocutionary act and the perlocutionary effects the speaker
hopes to produce.
There is a large literature on apparently gendered speech acts or
speech act types: for example compliments, apologies, insults, one-ups.
As we observed earlier, research in this area has probably raised at least
as many questions as it has answered. To try to sort out some of the
issues involved and think about how research might usefully develop,
we find it useful to see speech acts as kinds of social moves that are part
of larger, socially accomplished plans of action. We will expand this
idea below. First, however, we want to talk about interactional pur-
poses and effects at a very general level.
Functions of talk and motives of talkers:
gender oppositions
In chapter one, we saw that gender is overwhelmingly conceptualized
in terms of oppositions and in the preceding chapter we looked at
2 Inequality of various kinds among speakers can affect interpretation so that even if
comprehension is all that is needed it might not be forthcoming in some situations
(e.g. from someone who thinks that the speaker is not fully competent linguistically or
is ignorant of some fact relevant for interpretation). The importance of interpretation
and its social character are central themes throughout this book.
134 Language and Gender
gender-polarized characterizations of conversational style: cooperative

or other-oriented versus competitive or individualistic. The same or
closely related oppositions are also advanced to describe gender differ-
entiation in linguistic politeness and, more generally, speech-act usage.
Thus women are said to be more polite -- to use more polite language --
than men; and this is said to be because they are more other-oriented,
more collaborative, more affective. Such oppositions are in many ways
an advance over views of women as simply ineffective speakers who
deviate from the (effective) norm set by men’s speech. But these polar-
ized oppositions, however appealing we may find their more flattering
view of women, are ultimately as problematic as the deficit views of
women’s speech that they replaced. And from a linguistic perspective,
notions such as politeness and affectiveness are completely undefined.
How do we identify them in our linguistic data? Is the utterance thank
you always a polite speech act? How about when it is uttered as a re-
sponse to the refusal of a favor?
Politeness
Penelope Brown and Stephen Levinson (1987) have developed a theory
of politeness that builds on Erving Goffman’s (1967) ideas about face-
work (discussed in chapter three), and that has been very influential in
work on gender and politeness. What Brown and Levinson are trying
to do is articulate a theory that will shed light on general principles
of politeness while also showing how it can differ cross-culturally and
offering a framework for doing comparative work on politeness. Each
individual, they argue, has ongoing interests in promoting their pos-
itive face: projecting a self that is affiliated with others, that is liked
and identified with, part of a ‘‘we.” Each individual also cares about
their negative face: projecting a self that is a separate individual, some-
one deserving of respect and freedom from imposition, someone whose
own interests have intrinsic value.
3

An individual’s positive face needs
have to do with need for approval from others, for a sense of being
liked by others, of being connected to them. Negative face needs have
to do with a need to make a place for oneself, a need to pursue one’s
own projects without interference from others, a need to have one’s
own distinctive individuality recognized and respected. Positive and
negative face needs are in tension with one another. The more closely
3 Brown and Levinson’s labels ‘‘positive face’’ and ‘‘negative face’’ are inspired in
part by Durkheim’s (1915) positive and negative rites, along with insights derived from
Goffman.
135 Making social moves
connected we are and the more like one another we see ourselves as,
the harder it may become to protect our own and others’ needs for
separateness and independence of action. The more respect we receive,
the more recognition of our autonomy, the more difficult it may be to
forge intimate bonds linking us to similar others. Brown and Levinson
suggest that people typically have a better chance promoting their own
face interests if they also attend to others’ face interests. Although they
don’t put it this way, it may be most important to seem to care about
helping others preserve and enhance their face needs, whether or not
one in fact does care.
What Brown and Levinson call positive politeness involves addressing
positive face needs: showing that you like or empathize with someone,
that you include them in your ‘‘we,” your ‘‘in-group.” Commiserating
with one another about common problems (interfering parents or a
shared obnoxious boss), admiring the other’s taste in clothes by com-
menting approvingly on their attire, friendly joking and playful banter
marked by profanity and familiar terms of address (sweetie, you old
sonofabitch): such speech moves can exemplify positive politeness.
Much of the behavior that ordinary folk call polite, however, is a mat-

ter of what Brown and Levinson categorize as negative politeness: showing
respect or deference, avoiding imposing or offending, acknowledging
‘‘rights.” Apologies, for example, often try to correct a social wrong
done to another, thanks typically acknowledge that another has been
willing to extend themself for one’s own good, greetings and farewells
offer formulas to ease the strain created for face by the beginning and
ends of interactions. Such speech acts and other linguistic practices
such as the use of relatively formal modes of address and reference
(sir, madam, professor) often convey negative politeness.
Although certain kinds of speech acts do tend to be used to promote
positive face and others to protect negative face, the connections are
not as straightforward as they might at first seem. Brown and Levinson
emphasize that politeness does not lie simply in forms as such but
in what speakers use those forms to do. Of course, forms are not ir-
relevant to politeness. There are, for example, often verbal formulas
that are used to mark speech as conventionally ‘‘polite’’: many a child
acquiring English has learned the magic powers of ‘‘please” as an ac-
companiment to a request. ‘‘Please” conventionally signals recognition
that the request imposes on the addressee, that the speaker cares about
this potential harm to the addressee’s negative face and wants to mit-
igate the imposition. Politeness formulas are often aimed at least as
much at promoting the speaker’s face as protecting the addressee’s.
Following relatively rigid conventions for how one should speak in
136 Language and Gender
particular kinds of situations can be an important part of establishing
one’s own right to respect, showing that one is in the know on social
norms. Similarly, flouting conventions can be a way to show that one
is not socially controlled by those who promote those conventions. For
example where certain politeness routines have been associated with
mothers and women teachers, boys may avoid them as part of present-

ing themselves as independent of that female authority.
On the basis of extensive ethnographic fieldwork in Tenejapa,
Mexico, Brown (1980, 1990) argued that the women of Tenejapa did
more to promote others’ face needs, both positive and negative, than
did the men. Brown had not expected much negative politeness from
women to other women because in Tenejapa women’s subordination to
men in general and to the particular men in their own households was
strongly institutionalized. But she describes women’s relations to one
another as far more complex than she had predicted. She hypothesized
that the women needed to show both negative and positive politeness
to one another because of their extreme vulnerability to one another,
their heavy reliance on the good will of other women in their house-
hold and in the village.
It is not always easy to classify speech acts as promoting positive or
negative politeness or neither. Brown and Levinson’s distinction is not
exactly the same as one between that which aims to make another feel
good and that which aims to lessen the bad feelings someone might
have, to repair actual or potential damage to someone’s face. There
is also a further socially crucial distinction between saying and doing
things to promote one’s own face needs and saying and doing things to
promote someone else’s. Frequently, of course, the same action is in-
tended to play both roles, perhaps even promoting one’s own face needs
by means of promoting the other’s. But considerateness requires atten-
tion to the other’s face needs, whereas politeness as often discussed in
the literature may or may not. What looks like the same kind of act --
for example a compliment -- might be positively polite in one context
but not in another. Sometimes it might be a considerate move to make,
other times not. (Presumably, when a move is not considerate, it is not
really positively polite.)
Drawing on her own and others’ research on gendered distribution of

a number of different kinds of speech acts, Janet Holmes (1995) argues
that women tend to be more (linguistically) polite than men. She found,
for example, women complimenting (and also being complimented)
more than men. She also found women apologizing (and also being
apologized to) more than men. Compliments she treats as positively
polite, apologies as negatively polite. In other words, compliments are
137 Making social moves
seen as aimed at making someone feel liked by others, connected to
them. Apologies are seen as making someone feel that due attention
has been given to their interests and rights, that others respect them.
Holmes and her New Zealand colleagues had observers listen as they
went about their affairs and write down the first twenty instances
of utterances they heard as speech acts of the designated kind. This
method represents a considerable advance on earlier studies in the US
that relied on questionnaires rather than observation of naturally oc-
curring speech acts. Although we can ask on what grounds observers
decided that a compliment or an apology had been proffered (recall
from chapter one how a baby’s cries can be heard differently, depending
on whether one thinks it is a girl or a boy), Holmes’s results and those
of a number of other investigators whose work she discusses strongly
suggest that women predominate as both initiators and recipients of
certain kinds of ‘‘polite’’ speech acts among the populations studied
(mainly New Zealand and US middle-class people of European descent).
Can we conclude that these women are more considerate than the men
with whom they live and work? More interested in strengthening so-
cial ties, in promoting solidarity? More concerned to be seen as ‘‘nice’’?
Less ‘‘sincere’’? Even if we assume that the data represent communica-
tive patterns among these groups fairly accurately, accounting for the
observations is not so straightforward.
What sort of self a person presents in a particular kind of situa-

tion and how they ratify the other’s self-presentation will often be
implicated in constructing gender. Holmes takes the fact that men ap-
parently direct more instances of conventionally ‘‘polite’’ acts towards
women to indicate their recognition that women value these acts more
highly than do men. An alternative explanation might be that (at least
some) men want to project a masculinity that takes a ‘‘protective’’
stance towards women, constructing women as especially vulnerable
creatures in need of special handling. And, of course, both kinds of
motives might be involved, sometimes even for the same man.
Unlike the work by Holmes and her colleagues, much earlier studies
of politeness in service interactions in The Netherlands (Brouwer et al.
1979 and Brouwer 1982) found no difference linked to the speaker’s
sex (as judged by the data collector). Like Holmes’s work, however,
the Dutch studies did find significant differences linked to the sex of
the addressee. But the results go in the opposite direction from those
found in the New Zealand studies. Brouwer and her colleagues looked
at what people said to ticket-sellers in a large train station and in this
public service context found significantly more polite speech to male
ticket-sellers than to female from customers of both sexes. Notice that
138 Language and Gender
differences which depend on the addressee’s sex, however they are to
be explained in particular cases, do point dramatically to the very so-
cial character of gendered facework, which is always framed in relation
to the other participants in an interaction.
To evaluate research on gendered patterns of politeness, it is critical
to see how each researcher has operationalized the notion of polite-
ness. Is it a matter of the incidence of particular forms? Are the forms
in one study comparable to those in another? In what kinds of social
contexts are observations being made?
4

If it is a matter not of forms as
such (e.g. please, thank you) but of speech act types like compliments or
apologies, then it is important to understand how those act types are
identified and in what circumstances they are produced as well as th
e
form they take. Essentiall
y the same kind of act can be performed very
differently in different situations
or by different people. And, like other
features of conversational practice, politeness cannot be understood by
looking just at isolated individual moves or speech acts. Compliments
and apologies, for example, ask for responses from their addressees. Re-
sponses offer important evidence of the kind of facework accomplished
by the speech acts eliciting them, a point that Robert Herbert (1990)
emphasizes in his treatment of complimenting in gender practice.
Each community of practice develops its own expectations about the
facework participants will do on their own behalf and for the other
members of the community, often allocating differential responsibil-
ity for facework to different members of the community. There may
also be expectations about the kinds of means chosen to do that face-
work and how to balance the demands of facework with the other
kinds of things done in talking. What kinds of performance are pos-
sible? The ‘‘separate cultures’’ vie
w of gender discussed in chapter one
proposes that many people spend significant and formative periods en-
gaged in single-sex communities of practice. These separate contexts for
developing expectations about what is expected in the way of facework
are then thought to explain gender-differentiated patterns emerging in
mixed-sex communities of practice. Certainly gender separation at crit-
ical developmental stages is likely to be significant for various kinds

of expectations people have of themselves and of others. In the case of
4 The Dutch study looks only at exchanges between strangers in service transactions,
whereas many other studies have included exchanges between acquaintances and even
intimates. Wolfson (1984) proposes that facework is done most between acquaintances
and is far less consequential between intimates, whose relation is presumably settled,
and between strangers who do not expect to encounter one another again. There is, she
argues, a ‘‘bulge’’ in politeness at the middle distance. Holmes (1995) suggests that
Wolfson’s bulge model fits better with her observations of women than of men.
139 Making social moves
gender separation within a community, however, we have to keep in
mind that those in each group are very much aware of the existence of
the other group. Even more importantly, they are typically exposed to
gender ideologies and gender-differentiated allocations of rights and
responsibilities in mixed-sex communities of practice (e.g. many fami-
lies) during the same period that they are gender-segregated for many
peer activities. Differences that might emerge in ‘‘politeness’’ expec-
tations for women and for men (and in tolerance of failure to meet
expectations) almost certainly have multiple sources and implications
for gender practice far beyond mere marking of ‘‘difference.’’
Affective and instrumental talk
Janet Holmes, who has done a lot of empirical work on gendered
ways of talking among English speakers, associates women’s putatively
greater attention to (politeness-oriented) facework with a greater inter-
est in the affective function of talk. The affective function of talk covers
both the overt expression of emotion (‘‘How sad,’’ ‘‘Damn it,’’ ‘‘What a
sweetie/bastard he is’’) and everything that has to do with the mainte-
nance of social relations. It is generally contrasted with the referential or
instrumental function, conveying information (presumably about things
other than emotional states) or trying to establish ‘‘facts” or get things
accomplished. As Holmes recognizes, virtually all utterances serve both

affective and referential functions. Indeed, these functions intercon-
nect in many intricate ways. Making you feel good by complimenting
your attire may be a move that is part of my strategy to elicit certain
information I need from you in order to clinch a business deal. Or, con-
versely, reporting to you on certain facts may be a way to strengthen
my social bonds to you, to convey that I like you.
The affective/instrumental split has long been associated in the US
and many other English-speaking societies with a female/male divi-
sion of labor not only in talk but also in many other kinds of social
activities. Interestingly, however, people often ignore negative affect
(e.g. anger) in endorsing this gendered view of social life. They often
also ignore certain kinds of instrumental activities, especially what Eva
Feder Kittay (1998) calls ‘‘dependency work”: caring for small children,
the sick and elderly, and others who require near constant assistance.
This work is frequently seen not as work but as just the outpouring of
love; not surprisingly, it is also seen as women’s bailiwick. Caretakers
have to pay great attention to getting things done: cleaning up after the
incontinent elderly, bathing screaming (and slippery) babies, changing
sickbed sheets. Although affection for their charges may keep them