Tannen, Deborah. 1993.

Gender &1
m.scourse. New York & Oxford: I
.. 5 7 F77· ..

CHAPTER ONE

~~

The Relativity of
Lingu~~tic Strategies:
Rethinking Power and Solidarity
in Gender and Dominance

In this chapter I demonstrate that the theoreticalframework of

power and solidarity is essentialfor understanding genderpat­
terns in language use" and that gender and language is a
fruiiful sitefor investigating the dynamics underlying language
choice" including such dimensions as power and solidarity.
This framework is used to show that gender and language
research cannot he approached as the mechanical search for
specific linguisticphenomena. Using examplesfrom ~onversa­
tion as well as literary creations ofconversations" 1 argue that
each ofthe linguistic strategz·~ that have heen claimed to show';
dorn:inance can also show solidarity. For example" one can talle
while another is tallcing in order to wrest the floor; this can he
seen as a move motivated by power. Yet one can also tallc '
along with another in order'to show support and agreement;
this must he seen as a move motivated by solidarity. The two"
however" are not mutually exclusive. If hoth speakers are en­
gaged in a ritual struggle for thefloor" they might experience
the entire conversation as a pleasurahle· one: an exercise of

)
79
 Gender and Discourse The Relatiyity ofLinguistic Strategies

solidarity on the metalevel. My purpose" then" is not to ques­ means by which men dominate women in interaction. That men
tion that particular linguistic strategies" such as interruption, dO!!'inate 1IJOmen is not in question; what I am problematizing is the
may he used to create dominance" but rather to argue that source and workings of domination and other interpersonal inten­
intention and effict are not always synonymous" andthat there tions and effects. I will show that one cannot locate the source of
is never an endunng one-to-one relationship hetweeen a lin­ domination, or of any interpersonal intention or effect, in linguistic
guistic device and an interactive effiet. In seeking to understand strategies such as interruption, volubility, silence, and topic raising,
indivt.'duals" expen·ences ofconversation" lneluding dominance" as has been claimed. Similarly, one cannot locate the source of
we will have to look more deeply and more suhtly at the women's powerlessness in such linguistic strategies as indirectness,
worlczngs of conversational interaction. taciturnity, silence, and tag questions, as has also been claimed. The
reason one cannot do this is that the same linguistic means can be
used for different, even opposite, purposes and can have different,
INTRODUCTION even opposite, effects in different contexts. Thus, a strategy that
seems, or is, intended to dominate may in another context or in the

I N ANAL YZING DISCOURSE, many researchers operate on

the unstated assumption that all speakers proceed along similar
mouth of another speaker be intended or used to establish connec­
tion. Similarly, a strategy that seems, or is, intended to create con­
lines of interpretation, so a particular example of discourse can be
nection can in another context or in the mouth of another speaker be
taken to represent how discourse works for all speakers. For some
intended or used to establish dominance.
aspects of discourse, this is undoubtedly true. Yet a large body of
Put another way, the "true" intention or motive of any utterance
J
sociolinguistic literature makes clear that, for many aspects of dis­
course, this is so only to the extent that cultural background is
cannot be determined from examination of linguistic form alone.
For one thi~ intentions and effects are not identical. For another,
I
shared. To the extent that cultural backgrounds differ, lines of inter­
pretation and habitual use of many linguistic strategies are likely to
as the s~ciolinguistic literature has dramatized repeatedly (see espe­
cially McDermott and Tylbor 1983; Schegloff 1982, 1988; Erickson
1
jj
diverge. One thinks immediately and minimally of the work of
1986; Durand and Brenneis 1986), human interaction is a "joint
Gumperz (1982), Erickson and Shultz (1982), Scallon and Scollon
production": everything that occurs results from the interaction of
(1981), and Philips (1983). My own research shows that cultural
all participants. The source of the ambiguity and polysemy of lin­
difference is not limited to the gross and apparent levels of country
guistic strategies that I will explore here is the paradoxical relation­
of origin and native language, but also exists at the subcultural
ship between the dynamics of power and solidarity.
levels of ethnic heritage, class, geographic region, age, and gender.
My earlier work (Tannen 1984, 1986) focuses on ethnic and regional
style; my most recent work (Tannen 1990) focuses on gender­ OVERVIEW OF THE CHAPTER
related stylistic variation. I draw on this work here to demonstrate In this chapter I first briefly explain the theoretical paradigm of
that specific linguistic strategies have wide}y divergent potential power and solidarity. Then I show that linguistic strategies are
meanings. 1
potentially ambiguous (they could "mean" either power or soli­
This insight is particularly significant for research on language darity), and polysemous (they could "mean" both). Third, I reex­
and gender, much of which has sought to describe the linguistic amine and expand the power and solidarity framework in light of

20 21
 Gender and Discourse The Relativity ofLinguistic Strategies

cross-cultural research. Finally, I demonstrate the relativity of five involving participants in relation to each other. This creates a close­
linguistic strategies: indirectness, interruption, silence versus vol­ ness that can be contrasted with the, distance of individuals who
ubility, topic raising, and adversativeness (that is, verbal conflict). have no relation to each other at all.
In Brown and Gilman's paradigm, the key to power.is asym­
THEORETICAL BACKGROUND metry, but it is often thought to be formality. This is seen in the
following anecdote. I once entitled a lecture "The Paradox of Power
Power and Solidarity i

and Solidarity." The respondent to my talk app~ared wearing a

Since Brown and Gilman's (1960) introduction of the concept and three-piece suit and a knapsack on his back. The audience was
subsequent elaborations of it, especially those of Friedrich (1972) amused by the as~ociation ofthe suit with power, the knapsack with
and Brown and Levinson ([1978]1987), the dynamics of power and solidarity. There was something immediately recognizable in this
solidarity have heen fundamental to sociolinguistic theory. (Fasold semiotic. Indeed, a professor wearing a -knapsack might well mark
[1990] provides an overview.) Brown and Gilman based their frame­ solidarity with stu~ents at, for example, a protest demonstration.
work on analysis of the use of pronouns in European languages And wearing a three-piece suit to the demonstration might m~rk
which have two forms of the second person pronoun, such as the power by differentiating the w~arer from the demonstrators, per­
French tu and YOUSe In English the closest parallel is to he found in haps even reminding th.em of his dominant position in ~he institu­
forms of address: first name versus title-last name. In Brown and tional fiierarchy. But wearing a three-piece suit to the board meeting
Gilman's system, power is associated with nonreciprocal use of of a corporation would mark sol!darity_-with other board members,
pronouns; in English the parallel would be a situation in which one whereas wearing a knapsack in that'setting would connote not
speaker addresses the other by first ~ame but is addressed by title­ solidarity but disrespect, a move in the power dynamic.
last name (for example, doctor and patient, teacher and student,
boss and secretary, building resident and elevator operator). Soli­
The Ambiguity ofLinguistic Strategies
darity is associated with reciprocal pronoun use or symmetrical
forms of address: both speakers address each other by tu or hy vous As the preceding example shows, the same symbol-a three-piece
(in English, by title-last name or by first name). Power governs suit--can signal either power or solidarity, depending on, at least,
asymmetrical relationships wher:e one is subordinate to another; the setting (for example, a hoard meeting or student demonstra­
solidarity governs symmetrical relationships characterized by social tion), the habitual dress style of the individuf:ll, and the comparison,
equality and similarity. of his clothing with that worn by Qthers in the interaction. (I say
In my previous work exploring the relationship between power "his'~_ intentionally; the range of meanings would be; quite different
and solidarity as it emerges in conversational discourse (Tannen if a man's three-piece suit were worn 'by a woman.) This provides
1984, 1986), I note that power and solidarity are in paradoxical an analogue Jo the ambiguity of linguistic strategies, which are
relation to each other. That is, although power and solidarity, close­ signals in the semiotic system of language. As I have demonstrated
ness and distance, seem at first to he opposites, each also entails the at length in previous books (see eSl?eciaIly Tannen 1984, 1986,
other~ Any show of solidarity necessarily entails power, in that the 1990), all linguistic' strategi~s are potentially ambiguous. The
requirement ofsimilarity and closeness limits freedom and indepen­ power-solidarity dynamicjs one fundamental source of amblguity.
dence. At the same time, any show of power entails solidarity by What appear as attempts to dominate a conversation (an exercise of

22 23
 Gender and Discourse The Relativity ofLinguistic Strategies

power) may actually be intended to establish rapport (an exercise of meanings simultaneously. The question "Where's your coat?"
solidarity). This occurs because (as I have worded it elsewhere) shows frie~dly concern and suggests a parent-ehild constenation~
power and solidarity are bought with the same currency: The same The invitation to contribute a chapter to a book brings editor and
linguistic means can be used to create either or both. contributor closer and suggests a hierarchical relationship.
This ambiguity can be seen in the following fleeting conversa­ One more example will illustrate' the polysemy of strategies
tion. Two women were walking together from one building to signaling power and solidarity. If you have a friend who repeatedly!
another in order to attend a meeting. They were joined by a man picks up the3teck when you dine together, is she being generous
they both knew who had just exited a third building on his way to and sharing her wealth, or is she trying to flaunt her money and
the same meeting. One of the women greeted the man and re­ remind you that she has more of it than you? Although the in~ention
marked, "Where's your coat?" The man responded, "Thanks, may be to make you feel good by her generosity, her repeated
Mom." His response framed the woman's remark as a gambit in a generosity may nonetheless make you feel bad by reminding you
power exchange: a mother tells a child to put on his coat. Yet the that she has more money. Thus, both of you are caught in the web
woman might have intended the remark as showing friendly con­ of the ambiguity of power and solidarity. It is impossible to deter­
cern rather than parental caretaking. Was it power (condescending, mine which was h~r real motive, and whether it justifies your re­
on the model of parent to child) or solidarity (friendly, on the model sponse. On the other hand, even if you believe her motive was
of intimate peers)? Though the man's uptake is clear, the woman's purely generous, you .may nonetheless feel denigrated by her gener­
intention in making the remark is not. osity because the fact that she has this generous impulse is evidence
Another example comes from a letter written to me by a reader that she has more money than you, and her e~pressing the impulse
of You}ust Don"t Understand: Women and Men in Conversation. A reminds you 'ofit. In other words, both interpretations exist at once:
woman was at home when her partner arrived and announced that solidarity (she is paying to be nice) and,power (her being nice in
his archrival had invited him to contribute a chapter to a book. Th~ this way reminds you that she is richer). In this sense, the strategy is
woman remarked cheerfully how nice it was that the rival was not just ambiguous with regard to power and solidarity but poly­
initiating a rapprochement by including her partner in his book. He semous: This polysemy explains another observation that initially
told her she had got it wrong: because the rival would be the editor surprised me: Paules (1991) reports that waitresses in-the restaurant
and he merely a contributor, the rival was actually trying to solidify where she did ethnographic field work were offended not only by
his dominance. She interpreted the invitation in terms of solidarity. tips that were ~oo small, but' also by tips that were too large. The
He interpreted it as an expression of power. Which was right? I customers' inordinate beneficence implies that the amount of money
don't know. The invitation was ambiguous; it could have "meant" left is insignificant to the tipper but significant to the waitress.
either. I suspect it had elements of both. In other words, it was Brown and Gilman are explicit in their assumption that power is
polysemous. associated with asymmetrical relationships in which the power is
held by the person in the one-up position. This is stated in their
definition:
The Polysemy ofPower and Solidarity
If ambiguity denotes meaning one thing or another, polysemy de­ One person may be said to have power over another to the de­
notes meaning one thing and another-that is, having multiple gree that he is able to control the behavior of the other. Power

24 25
 Gender and Discourse
The Relativity of Linguistic Strategies
V
is a relationship between at least two persons, and it is non­
person is obligated to do things for him or her. Similarly, Yamada
reciprocal in the sense that both cannot have power in the same
area of behavior. (255)
(1992) describes the japanese relationship of amae, typified by the
parent-child or employer-employee constellation. It binds two in­
dividuals in a hierarchical interdependence by which both have
I have called attention, however, to the extent to which solidarity in
power in the form ofobligations as well as rights vis-a-vis the other.
itself can be a form of control. For example, a young woman com­
Finally, Wolfowitz (1991) explains that respect/deference is experi­
plained about friends who "don't let you be different." If the friend
enced by Suriname Javanese not as subservience but as an assertion
says she has a particular problem and the woman says, "I don't have of claims.
that problem," her friend is hurt and accuses her of putting her
The Suriname Javanese example is particularly intriguing be­
down, of acting superior. The assumption of similarity requires the
cause it calls into question the association of asymmetry with power
friend to have a matching problem. 2
and distance~ The style W olfowitz calls respect politeness is charac­
Furthermore, although Brown and Gilman acknowledge that
terized by both social closeness and negative politeness. 5 It is hier­
"power superiors may be solidary (parents, elder siblings)" and
archical insofar as it is directional and unequal; however, the crite­
"power inferiors, Similarly, may be as solidary as the old family
rion for directionality is not status but age. The prototypical
retainer" (258), most Americans are inclined to assume that soli­
relationship characterized by respect politeness is grandchild­
darity implies closeness, whereas power implies distance. 3 Thus
grandparent: a relationship that is both highly unequal and very
Americans regard the sibling relationship as the ultimate in soli­
close. M~reover, according to Wolfowitz, the Javanese assume that
darity: "sister" or "brother" are often used metaphorically to indi­
familial relations are inherently hierarchical, including age-graded
cate closeness and equality.4 In contrast, it is often assumed that
siblings. Equality, in contrast, is associated with formal relation­
hierarchy precludes closeness: employers and employees cannot
ships that are also marked by social distance.
"really" be friends. But being linked in a hierarchy necessarily
We can display these dynamics in the following way. The
brings individuals closer. This is an assumption underlying Wata­
model that reflects American assumptions conceptualizes power and
nabe's (1993) observation, in comparing American and Japanese
solidarity as opposite ends of a single continuum simultaneously
group discussions, that whereas the Americans in her study saw
representing symmetry/asymmetry, hierarchy/equality, and dis­
themselves as individuals participating in a joint activity, the Japa­
tance/closeness. (See figure 1.1.) In contrast, the cross-cultural per­
nese saw themselves as members of a group united by hierarchy.
spective suggests a multidimensional grid of at least (and, poten­
When reading Watanabe, I was caught up short by the term
tially and probably, more) intersecting continua. The closeness/
"united." My inclination had been to assume that hierarchy is dis­
tancing, not uniting.
The anthropological literature includes numerous discussions of
cultural contexts in which hierarchical relationships are seen as power
solidarity
close and mutually, not unilaterally, empowering. For example, asymmetry symmetry
hierarchy
Beeman (1986) describes an Iranian interactional pattern he dubs equality
distance closeness
"getting the lower hand." Taking the lower-status position enables
an Iranian to invoke a protector schema by which the higher=status
Figure 1.1. Unidimensional model

26
27

 Gender and Discourse ) The Relativity ofLinguistic Strategies

hierarchy
Now we are conce~ed with a new set of relations which are
symmetrical; for exanlple, attended the same school or have tlte
American: same parents or practice the same profession. If A has the same
Japanese: amae
employer/employee
parents as B, B has the same parents as A. Solidarity is the
Javanese: respect
name we give to the general relationship and solidarity is sym­
metrical. (257; italics in original)

The similarity/difference continuum calls to mind what I have dis­

closeness distance cussed elsewhere (Tannen 1984, 1986) as the double bind of com­
munication. 6 In some ways, we are all the same. But in other ways
we are all different. Communication is a double bind in the sense
that anything we say to honor our similarity violates our difference,
Javanese: and anything we say to honor our difference violates our sameness.
American: siblings formal/polite Thus a complaint can be lodged: "Don't think I'm different." ("If
equality you prick me, do I not bleed?" one might protest, like Shylock.) But
a complaint can also be lodged: "Don't think I'm the same." (Thus,
Figure 1.2. Multidimensional model women who have primary responsibility for the care of small chil­
dren may be effectively excluded from activities and events at which
day care is not provided.) Becker (1982:125) expresses this double
distance dimension can be placed on one axis and the hierarchy! bind as "a matter of continual self-correction between exuberance
equality one on another. (See figure 1.2.) Indeed, the intersection of (that is, friendliness: you are like me) and deficiency (that is, re­
these dimensions-that is, the co-incidence of hierarchy and spect: you are not me)." All these formulations elaborate on the
closeness-may account, at least in part, for what I am calling the tension between similarity and difference, or what Becker and Oka
ambiguity and polysemy of power and solidarity. (1974) call "the cline of person," a semantic dimension they suggest
may be the one most basic to language; that is, one deals with the
world and the objects and people in it in terms of how close (and, I
Similarity/Di.lJerence wo~ld add, similar) they are to oneself.
There is, one mor~ aspect of the dynamics of power and solidarity , As a result of these dynamics, similarity is a threat to hierarchy.
that bears discussion before I demonstrate the relativity of linguistic This': is dramatized in Harold Pinter's play Mountain Language.
strategies. That is the similarity/difference continuum and its rela­ -Composed of four brief scenes, :the play is set in a political prison in
tion to the other dynamics discussed. the capital city of an unnamed" country that is under dictatorial siege.
For Brown and Gilman, solidarity implies sameness, in contrast In the second scene, an old mountain woman is' finally allowed to
to power, about which they observe, "In general terms, the V form visit her son across a table as a guard stands over them. But
is linked with differences between persons" (256). This is explicit in whenever she tries to "speak to her son, the guard silences her,
their definition of "the solidarity semantic": telling the prisoner to tell his mother that it is forbidden to speak

28 29
 Gender and Discourse , The Relativity ofLinguistic Strategies

their mountain language in the capital. Then he continues: (Spaced repeating the guard's words verbatim, the prisoner was then saying,
dots indicate omitted text; unspaced dots are a form of punctuation "I am the same as you~ 'YJ By claiming his humanity and implicitly

included in the original text.) denying'the guard's assertion that he is "a pile of shit," the prisoner
challenged the guard's right to dominate him. 8 Similarity is antithe­
GUARD tical to hierarchy..
. . . And I'll tell you another thing. I've got a wife and three kids. And The ambiguity of closeness, a spatial metaphor representing
you're all a pile of shit. similarity or involvement, emerges in ~ nonverbal aspect of this
Silence. scene. In the performance I saw, the guard- moved steadily closerto
PRISONER
the prisoner as he repeated the question "You've got what?" until
I've got a wife and three kids.
he was bending over him, nose to nose. The guard's moving closer
was. a kinesiclproxemic analogue to the prisoner's statement, but
GUARD

with opposite ~ffeet:,_he was "'closing in." The guard moved closer
You've what?

and brought his face' into contact with the prisoner's not as a sign of
Silence.
affection (which such actions could signify in another context) but
You've got what? as a threat. (:loseness, then, can mean aggression rather than affiI,ia­
Silence. tion in the context ofa hierarchical rather than symmetrical relation­
What did you say to me? You've got what? ship. .
Silence.

You've got what?

THE REI:.ATIVITY O.F LINGUISTIC STRATEGIES
He picks up the telephone and dials one digit.

The potential ambiguity of linguistic strategies to mark both power

Sergeant? I'm in the Blue Room ... yes ... I thought I should report, and solidarity in face-to-face interaction has made .mischief in· lan­
Sergeant ... I think I've got a jo~er in here.
guage':and gender research, wherein it is tempting to assume that
The Sergeant soon enters and asks, "What joker?" The stage whatever women do results from, or creates, their powerlessness
. darkens and the scene ends. The final scene opens on the same and whatever men do -results from, or creates, their dominance. But
setting, with the prisoner bloody and shaking, his mother shocked all the linguistic strategies that have' been taken by analysts as
into speechlessness. evidence ofsubordination can in some circumstances be instru~ents
The prisoner was beaten for saying, "I've got a wife and three of affiliation. For the remainder of this chapter I demonstrate the
kids." This quotidian statement, which would be unremarkable in relativity of linguistic strategies by considering each of the follow­
casual conversation, was insubordinate in the hierarchical context of ing strategies in tum: indi~eetness, interruption, silence versus vol­
brutal oppression because the guard had just made the same state­ ubility, topicraising;-and adversativeness or verbal conflict. All of
ment. When the guard said, "I've got a wife and three kids. And these·' strategies have been "found" by researchers to express or
you're a pile of shit," he was claimi~g, "I am different from you." create domin~nce or subordination. I will demonstrate that they are
One could further interpret his words to imply, "I'm human, and ambiguous or polysemous with regard to dominance/subordination
you're not. Therefore I have a right to dominate and abuse you." By (that is, power) or distance/closeness (that is, solidarity). Once

30 31
 Gender and Discourse The Relativity of Linguistic Strategies

again, I am not arguing that these strategies cannot be used to create course, you should go.'') I argue that this father did not feel power­
dominance or powerlessness, much less that dominance and pow­ less to give his daughter orders. Rather, a communicative system
erlessness do not exist. Rather, my purpose is to demonstrate that was conventionalized by which he and she could both preserve the
the "meaning" of any linguistic strategy can vary, depending at appearance, and possibly the belie£: that she chose not to go rather
least on context, the conversational styles of participants, and the than simply obeying his command.
interaction of participants' styles and strategies. Therefore the oper­ Far from being powerless, this father felt so powerful that he did
ation of specific linguistic strategies must be studied more closely to not nee? to give his daughter orders; he simply needed to let her
understand how dominance and powerlessness are expressed and know his preference, and she would accommodate to it. By this
created in interaction. reasoning, indirectness is a prerogative of the powerful. By the
same reasoning a master who says~ "It's cold in here," may expect a
servant to make a mO,ve to close a window, but a servant who says
Indirectness the same thing is not likely to see his employer rise to correct the
Lakoff (1975) identifies two benefits of indirectness: defensiveness situation and make him more comfortable. Indeed, a Frenchman
and rapport. Defensiveness refers to a speaker's preference not to who was raised in Brittany tells me that his familY'never gave bald
go on record with an idea in order to be able to disclaim, rescind, or commands to their servants but always communicated orders in
modify it if it does not meet with a positive response. The rapport indir~et anq highly polite form. This pattern renders less surprising
benefit of indirectness results from the pleasant experience of get­ the finding of Bellinger and Gleason (1982, reported in Gleason
ting one's way not because one demanded it (power) but because 1987) that fathers' speech to their young children had a higher
the other person wanted the same thing (solidarity). Many re­ incidence than mothers' of both direct imperatives (such as "Turn
searchers have focused on the defensive or power benefit of indi­ the bolt with the wrench'') and implied indirect imperatives (for
rectness and ignored the payoff in rapport or solidarity. example, ."The wheel is going to fall off'').
The claim by Conley, O'Barr, and Lind (1979) that women's The use of indirectness can hardly be understood without the
language is really powerless language has been particularly infl.uen­ cross-cultural perspective. Many Americans find it self-evident that
tial. In this view, women's tendency. to be indirect is taken as evi­ directness is logical and aligned with power whereas indirectness is
dence that women don't feel entitled to make demands. Surely there akin to dishonesty as well as subservience. But for speakers raised in
are cases in which this is true. Yet it can also be demonstrated that most ofthe world's cultures, varieties ofindirectness are the norm in
those who feel entitled to make demands may prefer not to, seeking communication. In Japanese interaction, for example, it is well
the payoff in rapport. Furthermore, the ability to get one's demands known that saying "no" is considered too face-threatening to risk,
met without expressing them directly can be a sign of power rather so negative responses are phrased as positive ones: one never says
than of the lack of it. An example I have used elsewhere (Tannen "no," but listeners understand from the fonn ofthe "yes" whether it
1986, this volume, chapter 5) is the Greek father who answers, "If, is truly a "yes" or a polite ~'no."
you want, you can go," to his daughter's inquiry about going to a The American tendency to associate indirectness with female
party.. Because of the lack of enthusia~m of his response, the Greek style is not culturally universal. The above description of typical
daughter understands that her father would prefer she not go and Japanese style operates for men as well as women. My own research
"chooses" not to go. (A "real" approval would have been "Yes, of (Tannen 1981, 1984, 1986) suggests that Americans of some cultural

32 33
 Gender and Discourse The Relativity ofLinguistic Strategies

and geographic backgrounds, female as well as ,male, are more observation that studies comparing amount of interruption in a11­
likely than others to use relatively direct rather than indirect styles. female versus all-male 'conversations find more interruption, not -i.
In an eariy study (see chapter 5) I compared Greeks and Americans less, in all-female groups. Though initially surprising, this finding ff
with ~egard to their tendency to interpret p question as an indirect reinforces the need to distinguish linguistic strategies by their inter­
means of making a request. I found that whereas American women actional purpose. Does the overlap show support for the speaker, or
were more likely to take an indirect interpretation of a sample does it contradict or chang~ the topicr I explore this phenomenon in
conversation, Greek men were as likely as Greek women, and more detail in chapter 2 of this volume, but I will include a brief summary
likely than American men or women, to take an indirect interpreta­ of the-atgument here.
tion. Greek men, of course, are not less powerful vis-a.-vis women The phenomenon commonly referred to as "interruption," but
than American men. ­
which is more accurately referred to as "overlap," is a paradigm
Perhaps most striking is the finding of Keenan (1974) that in a case of the ambiguity of power and solidarity. This is clearly dem­
Malagasy-speaking village on the island of Madagascar, women are onstrated with reference to a two-and-a-half...hour Thanksgiving
seen as direct and men as indirect. But this in no way implies that dinner conversation that I analyzed at length (Tannen 1984). My
the women are more powerful than men in this society. Quite the analysis makes clear that some speakers consider talking along with
contrary, Malagasy men are socially dominant, and their indirect another to be a show of enthusiastic participation in the conversa­
style is more highly :valued. Keenan found that women were widely tion, of solidarity, c;reating connections; others, however, assume
believed to debase the language with their artless directness, that only one voice should' be heard at a time, so for them any
whereas men's elaborate indirectness was widely admired. overlap is an interruption, an attempt to wrest the floor, a power
Indirectness, then, is not in itself a strategy of subordination. play. The result, in the conversation I analyzed, was that enthusias­
Rather, it can be used either by the powerful or the powerless. The tic listeners who overlapped cooperatively, talking along to estab­
interpretation of a given utterance, and the likely response to it, lish rapport, were perceived by overlap-resistant speakers as inter­
depends on the setting, on individuals' status and their relationship rupting. This doubtless contributed to the impression reported by
to each other, .and also on the linguistic conventions that are ritu­ the overlap-resistant speakers that the cooperative overlappers had
alized in the cultural context. "dominated" the conversation. Indeed, the tape and transcript also
give the impression that the" cooperative overlappers had domi­
Interruption .nated, because the overlap-aversant participants tended to stop
speaking as soon as another voice 'began.
That interruption is a sign of dominance has been as widespread an
It is worth emphasizing the role of symmetry, or balance, in
assumption in research as in conventional wisdom. One rarely en­ determining whether an overlap becomes an interruption in the
counters an article on gender and language that does not make this negative or power-laden sense. If one speaker repeatedly overlaps
claim..Most frequently cited is West and Zimmerman's (1983) find­
_ and another repeatedly gives way, the resulting communication is
ing that men dominate women by interrupting them in conversa­ . unbalanced, or asymmetrical, and the effect (though not necessarily
tion. Tellingly, however, Deborah James and Sandra Clarke (1993), the intent) is domination. But if both speakers avoid overlap, or if
reviewing research on gender and interruption, do not find a clear
both speakers overlap each other and win out equally, there is
pattern of males interrupting females. Especially signifi~ant is their symmetry and no domination, regardless of speakers' intentions. In

34 35
 Gender and Discourse The Relativity ofLinguistic Strategies

an important sense, though-and this will be discussed in the last running through Komarovsky's (1962) classic study of Blue-Collar

section under the rubric of adversativeness-the very engagement Marriage is that many of "the wives interviewed said they talked

in a symmetricai struggie for the floor can be experienced as creat­ more than their husbands: "He's tongue-tied," one woman said

ing rapport, in the spirit of ritual opposition analogous to sports. (13); "My husband has a great habit of not talking," said another

Further, an imbalance can result from differences in the purpose for (162); "He doesn't say much but he means what he says and the

which overlap is used. If one speaker tends to talk along in order to children mind him," said a third (353). Yet there is no question but

show support, and the other chimes in to take the floor, the floor­ o that these husbands are dominant in their marriages, as the last of
taking overlapper will tend to dominate. these quotes indicates.
Thus, to understand whether an overlap is an interruption, one Indeed, taciturnity itself can be an instrument of power. This is

must consider the context (for example, cooperative overlapping is precisely the claim of Sattel (1983), who argues that men use silence

more likely to occur in casual conversation among friends than in a to exercise power over women. Sattel illustrates with a scene from

job interview), speakers' habitual styles (for example, overlaps are Erica Jong's novel Fear of Flying~ only a brief part of which is

more likely not to be interruptions among those with a style I call presented here. The first line of dialogue is spoken by Isadora, the

"high involvement"), and the interaction of their styles (for exam­ second by her husband, Bennett. (Spaced dots indicate omitted text;

ple, an interruption is more likely to occur between speakers whose unspaced dots are a form of punctuation included in the original

styles differ with regard to pausing and overlap). This is not to say text.)

that one cannot use interruption to dominate a conversation or a

person, but only that it is not self-evident from the observation of "Why do you tum on me? What did I do?"

overlap that an interruption has occurred, was intended, or was Silence.

intended to dominate. "What did I do?"

ii",;,'
,~~
He looks at her as if her not knowing were another in­
jury.
Silence Versus Volubility "Look, let's just go to sleep now. Let's just forget it."

The excerpt from Pinter's Mountain Language dramatizes the as­ "Forget what?"

sumption that powerful people do the talking and powerless people He says nothing.

are silenced. This is the trope that underlies the play's title and its
"It was something in the movie, wasn't it?"

~entral theme: By outlawing their language, the oppressors silence

"What, in the movie?"

the mountain people, robbing them of their abiliry to speak and

". . . It was the funeral scene.... The little boy looking at

hence of their humanity. In the same spirit, many scholars (for

his dead mother. Something got you there. That was when you
example, Spender 1980) have claimed that men dominate women by got depressed."

silencing them. There are obviously circumstances in which this is Silence.

accurate. Coates (1986) notes numerous proverbs that instruct "Well, wasn't it?"

women, like children, to be silent. Silence.

Silence alone, however, is not a self-evident sign of powerless­ "Dh come on, Bennett, you're making me furious. ,Please tell

ness, nor volubility a self-evident sign of domination. A theme me. Please."

36
37
 Gender and Discourse The Relativity ofLinguistic Strategies

The painful scene continues in this vein until Bennett tries to leave discussed, there' are cultural and subcultural differences in the length
the room and Isadora tries to detain him. The excerpt certainly of pauses expected between and within.speaking turns. In my study
seems to support Sattei's ciaim that Bennett 7s silence subjugates his of the dinner conversation, those who expected shorter pauses be­
wife, as the scene ends with her literally lowered to the floor, cling­ tween .conversational turns began to feel an uncomfortable silence
ing to his pajama leg. But the reason his silence is an effective ensuing while their longer-pausing friends were simply waiting for
weapon is her insistence that he tell her what's wrong. If she receded what 'they regarded as the "normal" end-of-turn pause. The result
into si.1ence, leaving the room or refusing to talk to him, his silence was that the shorter pausers ended up doing most of the talking,
would be disarmed. The devastation results not from his silence another sign interpreted by their interlocutors as dominating the
alone but from the interaction of his silence and her insistence on conversation. But their intentions had been to fill in what to them
talking, in other words, the interaction of their differing styles.9 were potentially uncomfortable silences, that is, to grease the con­
Researchers have counted numbers of words spoken or timed versational wheels and ensure the success of the conversation. In
length of talk in order to demonstrate that men talk more than ,their view, the, taciturn participants were uncooperative, failing to
women and thereby dominate interactions. (S~e James and Drakich do their part to maintain the conversation.
1993 for a summary of research on amount of talk.) Undoubtedly Thus, silence and volubility cannot always be taken.to "mean"
there is truth to this observation in some settings. But the associa­ power or ppwerlessness, domination or subjugation. Rather, both
tion of volubility with dominance does not hold for all settings and m~y imply either power or solidarity, depending on the dynamics
all cultures. Imagine, for example, an interrogation, in which the discussed. '
interrogator does little of the talking but holds all the power.
The relativity of the "meaning" of taciturnity and volubility is
Topic Raising
highlighted in Margaret Mead's (1977) discussion of "end linkage,"
a concept developed jointly by Mead, Gregory Bateson, and Geof­ Shuy (1982) is typical in assuming that the speaker who raises the
e frey Gorer. The claim is that universal and biologically constructed most topics is dominating a conversation. How~ver, in a study I
relationships, such as parent-ehild, are linked to different behaviors conducted (see this volume, chapter,3) of videotaped conversations
in different cultures. One of their paradigm examples is the appor­ among friends of varying ages· recorded by Dorval (1990), it
tionment of spectatorship and exhibitioni~m. In middle-class Ameri­ emerged that the speaker who raised the most topics was not always
can culture, children, who are obviously the weaker party in the dominant, as judged by other criteria (for example, who took the
constellation, are expected to exhibit while their more powerful lead in, addressing the investigator when he entered the room?). In a
parents are spectators. (Consider, for example, the American child 20-minute conversation' between a pair of sixth-grade girls who
who is prompted to demonstrate how well slhe can recite the alpha­ identified themselves as best friends, Shannon raised the topic of
bet for guests.) In contrast, in middle- and upper-class British cul­ Julia's relationship with Mary by saying, '~Too bad you and Mary
. ture, exhibition is associated with the parental role and spectator­ . are' not good friends··anymore." The conversation proceeded and
, ship with children, who are expected to be seen and not heard. continued to focus almost" exclusively on Julia's troubled relation­
Moreover, volubility and taciturnity, too, can result from style ship with Mary. " f"

differences rather than speakers' intentions. As I (Tannen 1984, Similarly, most of the conversation between two tenth-grade
1985) and others (Scollon and Scollon 1981, Scollo~ 1985) have. girls was about Nancy, but Sally raised the topic of Nancy's prob­

38 39

 Gender and Discourse The Relativity ofLinguistic Strategies .

lerns. In response to Nancy's question "Well, what do you want to how male adversativeness and female cooperation are played out,
talk about?" Sally said, "Your mama. Did you talk to your mama?" complicated, and contradicted in conversational discourse. In -an­
The ensuing conversation focuses on events invoiving Nancy's alyzing Videotapes of friends talking, for example, I found a sixth­
mother and boyfriend. Overall, Sally raised nine topics, Nancy grade boy saying to' his best friend,
seven. However, all but one of the topics Sally raised were 'ques­
tions focused on Nancy. If raising more topics is a sign of domi­ Seems like, if there's a 6ght, me and you are automatically in it.
nance, Sally controlled the conversation when she raised topics, And everyone else wants to go against you and everything. It's
although even this was subject to Nancy's collaboration by picking hard to agree without someone saying something to you.
them up. It ~ay or may not be the case that Sally controlled the
In contrast, girls of the same age (and also of most other ages whose
conversation, but the nature of her dominance is surely other than
talk I examined) spent a great deal of time discussing the dangers of
what is normally assumed by that term if the topics she raised were
all about Nancy. anger aQd c~ntention. In affirming their own friendship, one girl
Finally, the effect of raising topics may also be an effect of told her friend,
differences in pacing and pausing, as discussed above with regard to
Me -and you ~ get in fights hardly,
my study of dinner-table conversation. A speaker who thinks the
other has no more to say on a given topic may try to contribute to' and
the' conversation by raising another topic. But a speaker who was
intending to say more and was simply waiting for the appropriate I mean like if I try to talk to you, you'll say, 'Talk to me!' And'
turn-exchange pause will feel that the floor was taken away and the if you try to talk to me, I'll talk to you~
topic aggressively switched. Yet again, the impression of domi­
nance might result from style differences. These examples of gendered styles of interaction are illuminated by
the insight that power and solidarity are mutually evocative. As,
seen in the statement of the sixth-grade boy, opposing other boys in
Adversativeness: Conflict and Verhal Aggression
teams entails affiliation within the team. The most dramatic instance
Research on gender and language has consistently found male of male affiliation resulting from conflict with others is bonding
speakers to be competitive and more likely to engage in conflict (for among soldiers, a phenomenon explored by Norman (1990).
example, by arguing, issuing commands, and taking opposing . By the same token, girls' efforts to support their friends neces­
stands) and females to be cooperative and more likely to avoid sarily entail exclusion of or opposition to other girls. This emerges
conflict (for example, by agreeing, supporting, and making sugges­ in Hughes' (1988) study of girls playing a street game called four­
tions rather than commands). (Maltz and Borker [1982] summarize square, in which four players occupy one square each and bounce a
some of this research.) Ong (1981:51) argues that "adversativeness" ball into each other's squares. The·objeet of the game is to eliminate
is universal, but "conspicuous or expressed adversativeness is a players by hitting the ball into their square in such a way that they
larger element in the lives of males than of females-." fail to hit it back. But this effort to "get people out" is at odds with
In my analysis of videotapes of male and female friends talking the social injunction under which the girls operate, to be "nice" and
to each other (this volume, chapter 3), I have begun to investig~te not "mean.. " Hughes found that the girls resolved the conflict, and

40 41
',-,
 Gender and Discourse The Relativity of Linguistic Strategies

formed "incipient teams" composed of friends, by claiming that Joseph: 1-1- 1'11- I could poke your eyes out with my gun. I have a
their motivation in eliminating some players was to enable others gun.':
(their friends) to enter the game, since eiiminated players are re­ Den~~: A gun! ~'ll- 1- I-even if­
placed by awaiting players. In the girls' terms, "getting someone Richard: I have a ~n too.
out" was "nice-mean," because it was reframed as "getting some­
Denny: And I have guns too and it's bigger than yours and it poo­
one [a friend] in." This dynamic is also supported by my analysis of
poo down. That's poo-poo.
the sixth-grade girls' conversation: Most oftheir talk was devoted to
(All three boys laugh at Denny~ reference to poo-poo.) fIR!
allying themselves with each other in opposition to another girl
who was not present. So their cooperation (solidarity) also entails Richard:' Now leave.
...
opposition (power). . Joseph: Un-uh. I gonna tell you to put on- on the gun on your _hair
For boys power entails solidarity not only by opposition to and the poo~. ~ilI ,come right out on his face.
another team, but by opposition to each other. In the videotapes of Denny: Well. ,J
friends talking, I found that all the conversations between young Richard: Slinky will snap right on your face too.
boys (and none betwe~n young girls) had numerous examples of Denny: And my gun will snap right-
teasing and mock attack. to In examining preschool conversations
transcribed and analyzed by Corsaro and Rizzo (1990:34), I was Up ~ntil this point I had ~o difficulty interpreting the interaction:
amazed to discover that a fight could initiate rather than preclude The boys were engaged in a fight occasioned by Joseph's Intrusion
friendship. In the following episode, a little boy intrudes on two into Richard ,and penny's play. But what happened next .surprised
others and an angry fight ensues. This is the way Corsaro and Rizzo and, at firs~,p~~le.xed me"! ,~orsaro and ~o describe it this way:
present the dialogue:
At this point a girl (Debbie) enters, says she is Batgirl, and' asks
Two boys (Richard and Denny) have been playing with a slinlcy on the if they have seen Robin. Joseph says he is Robin, but she says
stairway leading to the upstairs playhouse in the school. During their play she is looking for a different RapiD and then runs off.. After
two other boys (Joseph and Martin) enter and stand n.ear the hottom ofthe 'Debbie leaves, Denny and Richard move into the playhouse
stairs. and Joseph follows. From this point to the end of the episode
Denny: Go! the three boys play together.
~

(Martin now runs oJ!; hut]oseph remains and he eventually moves halfway
up the stairs.)
At first.I was inqedulous that,so sqon after their seemingly hostile
~ncounter, the boys played.amit;ably together. I finally came to the
Joseph: These are big shoes.
conclusion that for Joseph picking a fight was a way to enter into
Richard: I'll punch him right in the eye.

interaction. withJhe other~?ys, and engaging him in the fight was

Joseph: I'll punch you right in the nose.

Ri~hard and De~ny'~ way o(accepting him into their inter~etio~

Denny: I'll punch him with my big fist.
at least after he: acquitted himself s.atisfaetorily in the fight. In this
Joseph: 1'11-1-1­ light, I could se~ that the reference to poo-poo, which occasiqned
Richard: And he'll be bumpety, bumpety and punt;hed out all the way generalla~ghter, was the.; begi~ning of a reframing from fighting to
down the stairs. playing. II

42 43
 Gender and Discourse The Relativity of Linguistic Strategies

Folklore provides numerous stories in which fighting precipi­ students belligerent because they provoke arguments about Ameri-­
tates friendship among men. One such is attributed by Bly can foreign policy with A_tnericans they have just met.
(1990:243-44) to Joseph Campbell's account of the Sumerian epic, Greek conversation provides an example of ~ cultural style that
Gi/gamesh. In Bly's rendition, Gilgamesh, ~ young king, wants to places more positive value, for both women and men, on dynamic
befriend a wild man named Enkidu. When Enkidu is told of Gilga­ opposition. Kakava (1989) replicates Schiffrin's findings by showing
mesh, how a Greek family enjoy opposing each other in dinner conversa­
tion. In another study of modern Greek conversation, Tannen and
his heart grew light. He yearned for a friend. "Very well!" he Kakava (1992) find speakers routinely disagryeing when they actu­
said. "And I shall challenge him." ally agree, and using diminutive name forms and other terms of
endearment-markers of closeness-precisely when they are op­
Bly paraphrases the continuation: "Enkidu then travels to the city posing each other. I3 These patterns can be seen in the following
and meets Gilgamesh; the two wrestle, Enkidu wins, and the two excerpt from a conversation that took place in Greece between an
become inseparable friends. "12 older Greek woman. and myself.. The woman, whom I call Ms.
A modem-day academic equivalent to the bonding that results Stella, has just told me that she complained to the police about a
from opposition is to be found in the situation of fruitful collabora­ ~onstruction crew that illegally .continued drilling and pounding
tions that began when an audience member publicly challenged a through the siesta hours, disturbing her nap:
speaker after his talk. Finally, Penelope Eckert (personal communi­
cation) informs me that in her research on high school students Deborah: Echete dildo.
(Eckert 1990) she was told by boys, but never by girls, that their Stella: Ego echo dildo. Kopella mou, den xero an echo dikio i
close friendships began by fighting. . den echo dikio. AUa ego yperaspizomai ta symferonta
These examples call into question the correlation of aggression mou kai ta dikaiomata moue
and power on one hand, and cooperation and solidarity on the Deborah: You're right. .
other. Again the cross-cultural perspective provides an invaluable
Stella: I!!!! right. My dear girl, I don't know if;I'm right or I'm
corrective to the temptation to align aggression with power as dis­ not right. But I am watching out for my interests and my
tinguished from solidarity. Many cultures of the world see arguing ~ rights;
as a pleasurable sign ofintimacy. Schiffrin (1984) shows that among
lower-middle-class men and women of East European Jewish back­ My response to Ms. Stella's complaint is to support her by agreeing.
ground, friendly argument is a means of being sociable. Frank But she disagrees with my agreement by reframing my statement in
(1988) shows a Jewish couple who tend to polarize and take argu­ -her own terms 'lather than simply accepting it by stopping after "I
mentative positions, but they flre not fighting; they are staging a am right." She .also marks her divergence from my- frame with the
kind of public sparring, where both fighters are on the same team. endearment "kopella mou" (literally, "my girl," but idiomatically
Byrnes (1986) claims that Germans find American students unin­ closer to "my dear girl'').
formed and uncommitted because they are reluctant to argue poli­ The following conversation is also taken from Tannen and Ka­
tics with new acquaintances. For their part, Americans find German kava (1992). It is, according to Kakava, typical of her family's socia­

44 45
 Gender and Discourse The Relativity of Linguistic Strategies

ble argument. The younger sister has said that she cannot under­ keeping both images in focus at once, we may at least succeed in

stand why the attractive young woman who is the prime minister sVlitching from one to the other rapidiy and reguiariy enough to

Papandreou's girlfriend would have an affair with such an old man. deepen our understanding of the dynamics underlying interaction

The older sister, Christina, argues that the woman may have felt such as .power and solidarity" as well as gender and language use.

that in having an affCiir with the prime minister she was doing
something notable. Her sister replied,
NOTES
Poly megalo timima re Christinaki na pliroseis pantos. This chapter began as a paper entitled "Rethinking Power and Solidarity in

It's a very high price to pay, Chrissie, anyway. Gender and Dominance," which was published in Proceedings of tke 16th Annual

Meeting oftke Berkeley Linguistics Society, edited by Kira Hall, Jean-Pierre Koenig,

Michael Meacham, Sondra Reinman, and Laurel A. Sutton, 519-29 (Berkeley:

I use the English diminutive form "Chrissie" to reflect the Greek

Linguistics Department, University of Califomia, Berkeley, 1990). A Significantly

diminutive ending -aki~ but the particle re cannot really be trans­ revised and expanded version appears in Gender and Conversational Interaction, a

lated; it is simply a marker of closeness that is typically used when volume I edited, published by Oxford University Press in 1993. That rewriting was

disagreeing, as in the ubiquitously heard expression "Ochi, re" carried out while I was in residence at the Institute for Advanced Study in Prince­

("No, re'j. ton, New Jersey. Further revisions-improvements, I hope-which I made to the

version that appears here (some in response to much-appreciated comments from

Paul Friedrich) were carried out while I was a fellow at the Center for Advanced
j~~

!~
CONCLUSION Study in the Behavioral Sciences in Palo Alto, California. I have lifted the sum­

~
mary of this chapter directly from the overview that appears in the 1993 publica­
The intersection of language and gender provides a rich site
tion.
for analyzing how power and solidarity are created in discourse. 1. I use the term "strategy" in its standard sociolinguistic sense, to refer
im.
But prior research in this area evidences the danger of linking
linguistic forms with interactional intentions such as dominance.
simply to a way ofspeaking. No implication is intended of deliberate planning, as
is the case in the common parlance use ofsuch expressions as "military strategy.»

j~
In trying to understand how speakers use language, we must con­ Neither, however, as Gumperz (1982) observes, are linguistic strategies "uncon­

sider the context (in every sense, including at least textual, re­ scious." Rather, they are best thqught ofas "automatic." That is, people speak in a

particular way without "consciously" thinking it through, but are aware, if qu'es­

lational, and institutional constraints), speakers' conversational

styles, and, most crucially, the interaction of their styles with each
other.
!ioned, of how they spoke and what they were trying to accomplish by talking in
, that way. This is in contrast to the "unconscious" motives of Freudian theory
"about which an individual would be unaware if questioned. (For example, most
'Il
,:j

iWJJ
~
,I

Attempts to understand what goes on between women and men men would vigorously deny that they want to kill their fathers and marry their
in conversation are muddled by the ambiguity and polysemy of mothers, but a strict Freudian might claim that this wish is "unconscious.'') I'
,{i'~ ;"::-;-/2. This example is taken from Tannen (1990). I
power and solidarity. The same linguistic means can accomplish
either, and every utterance combines el~ments of both. Scholars, <;':~-.:" .,-' 3. I myself have made the observation that asymmetry is distancing whereas Ii
:","'symmetry implies closeness, for example, with regard to the ritual of "troubles
however; like individuals in interaction, are likely to see only one
i'.:talk", and the way it often misfues between women and men (Tannen 1990). Many
and not the other, like the picture that cannot be seen for what it #:. ~omen talk about troubles as a way of feeling closer! but many men frequently
is-simultaneously a chalice and two faces-but can only be seen -/ 'interpret the description of troubles as a request for advice, which they kindly
alternately as one or the other. In attempting the impossible task of offer. I have observed that this not only cuts off the troubles talk, which was the

46
47
 Gender and Discourse The Relativity ofLinguistic Strategies

real point of the discourse, but it also introduces asymmetry: If one person says sending a temple prostitute who throws off her clothes atthe appropriate moment.
she has a problem and another says she has the same problem, they are symmetri­ She is simply the vehicle for the two men to get together. Much could be said
cally arrayed and their similarity brings them closer. But if one person has a about this aspect of the epic, but my purpose here is only to draw attention to the
problem and the other: ha~ the solution, the one with the solution is one-up, and way the men use fighting as a means to friendship.,
the asymmetry is distancing-just the opposite of what was sought by initiating 13. Sifianou (1992) independently observes the use ofdiminutives as soli~arity
the ritual. markers in Greek conversation.
4. This assumption is made explicit by Klagsbrun (1992), who, in a book
about sibling relationships, writes, "Unlike the ties between parents and children,
the connection among siblings is a horizontal one. That is, sibs exist on the same REFERENCES
plane, as peers, more or less equals" (12). But Klagsbrun gives a pivotal example
ofhow she was frustrated as a child (and continues to he hampered, as an adult) by Bateson, Gregory. 1972. Steps to an ecology of mind. San Francisco: Chandler.
always being bested by her older brother. It is clear from the example that she and Paperback: New York: Ballantine.
her brother were not equals because of the difference in their ages-and, one Beeker, A. L. 1982. Beyond translation: Esthetics' and language description. Con­
might argue, their genders. temporary perceptions of language: Interdisciplinary dimensions. George­
5. Negative politeness, as discussed by Brown and Levinson ([1978]1987), town University Round Table on Languages and Linguistics 1982, ed.
entails honoring others' needs not to be imposed on. by Heidi Byrnes, 124-38. Washington, DC: Georgetown University
6. Scollon (1982:344-45) explains that all commuriication is a double hind Press.
because one must serve, with every utterance, the ,conflicting needs to be left alone Becker, A. L., and I Gusti Ngurah Oka. 1974. Person in Kawi: Exploration of an
(negative face) and to be accepted as a member ofsociety (positive face). The term elementary semantic dimension..Oceanic Linguistics 13:229-55.
"double bind" traces to Bateson (1972). Beeman, William O. 1986. Language, status, and power in Iran. Bloomington:
7. I have demonstrated at length (Tannen 1987, 1989) that repeating another's Indiana University Press.
words creates rapport on a metalevel: It is a ratification of the other's words, Bellinger, David, and Jean Berko Gleason. 1982. Sex differences in parental direc­
evidence of participation in the same universe ·of discourse. tives to young children. Sex Roles 8:1123-39.
8. Following the oral presentation of this paper at the Berkeley Linguistics Bly, Robert. 1990. Iron John: A book about men. Reading, MA: Addison-Wesley.
Society in 1989, both Gary Holland and Michael Chandler pointed out that the Brown, Roger, and Albert Gilman. 1960. The pronouns of power and solidarity.
prisoner may he heard as implying the second part of the guard's statement: "and Style in language, ed. by Thomas Sebeok, 253-76. Cambridge, MA: MJ.T.
you're a pile of shit." Press.
9. This scene illustrates what Bateson (1972) calls "complementary .schis­ Brown, Penelope, and Stephen Levinson. [1978]1987. Politeness: Some universals
mogenesis": Each person's style drives the other into increasingly exaggerated in I~nguage usage. Cambridge: Cambridge University Press.
forms of the opposing behavior. The more he refuses to tell her what's wrong, the Byrnes, Heidi. 1986. Interactional style in German.and American conversations.
more desperate she becomes to break through his silence. The more she pressures Text 6:2.189-206.
him to tell her, the more adamant he becomes about refusing to do so. Campbell, Joseph. 1964. The masks of god: Occidental mythology. New York:
10. Some examples are given in Tannen (1990). Whereas the boys made such Viking.
gestures as shooting each other with invisible guns, the girls made such gestures Coates, Jennifer. 1986. Women, men and language: London: Longman.
as reaching out and adjusting a friend's headband. Conley, John M., William M. O'Barr, and E. Allen Lind. 1979. The power of
11. Elsewhere (Tannen 1990:163-65) I discuss this example in more detail and language: Presentational stYle in the courtroom. puke Law Journal
note the contrast that the boys fight when they want to play, and the girl avoids 1978:1375-99.
disagreeing even when she in fact disagrees. Corsaro, William, and Thomas Rizzo. 1990. Disputes in the peer cultur~ ofAmeri­
12. Another element of this epic, as Bly recounts it, is that Gilgamesh lures can and Italian nursery school children. Conflict talk, ed. by Allen
Enkidu away from the wild animals with which he had been happily living by Grimshaw, 21-65. Cambridge: Cambridge University P~ess.

48
49
 Gender and Discourse The Relativity ofLinguistic Strategies

Dorval, Bruce (ed.). 1990. Conversational coherence and its development. Nor: KJagsbrun, Francine. 1992. Mixed feelings: Love, hate, rivalry, and reconciliation
wood, Nj: Abiex. among brothers and sisters. New York: Banta~. .
Duranti, Alessandro, and Donald Brenneis (eds.). 1986. The audience as co­ Komarovsky, Mirra. 1962. Blue-collar marriage. New York: Vintage.

author. Special issue of Text 6:3.239-47. Lakof4 Rohin.;-1975. Language and woman's place. New York: Harper and Row.

Erickson, Frederick. 1986. Listening and speaking. Languages and Iinguistics:·The Maltz, Daniel N., and Ruth A. Borker. 1982. A cultural approach tamale-female

interdependence of theory, data, and application. Georgetown University miscommunication~Language and social identity, ed. by John J. Gumpen,
Round Table on Languages and Linguistics 1985, ed. by Deborah Tannen, 196-216. Cambridge: Cambridge University Press.
294-319. Washington, DC: G~orgetown University Press. McDermott, R P., and Henry Tylbor. 1983. On the necessity of collusion in
Erickson, Frederick, and Jeffrey Shultz. 1982. The counselor as gatekeeper: Social conversation. Text 3:3.277-97. .
interaction in interviews. New York: Academic Press. Mead, Margaret. 1977. End linkage: A tool for cross-cultural analysis. About
Fasold, Ralph W. 1990. The sociolinguistics of language. Oxford: Basil Black­ Bateson, ed. by John Brockman, 171-231. New York: Dutton.
well. -­
Norman, Michael. 1990. These good men: Friendships forged from war. New
Frank, Jane. 1988. Communicating("by pairs": Agreeing and disagreeing among York: Crown.
married couples. Unpublished ms., Georgetown University. Dng, Walter J.
1981. Fighting for life: Contest, sexuality, and consciousness.
Friedrich, Paul. 1972. Social context and semantic feature: The Russian pronomi­ Ithaca, NY: Cornell University Press; Amherst: University of Massa­
nal usage. Directions in sociolinguistics, ed. by John J. Gumperz and Dell chusetts Press.
Hymes, 270-300. New York: Holt, Rinehart, and Winston. Reprinted: Paules, Greta Fall: 1991. Dishing it out: Power and resistance among waitresses in
Oxford: Basil Blackwell. a New Jersey restaurant. Philadelphia: Temple University Press.
Gleason, Jean Berko. 1987. Sex differences in parent-child interaction. Language, Philips, Susan Urmston. 1983. The invisible culture: Communication in- class­
gender, and sex in comparative perspective, ed. by Susan U. Philips, Susan room and community on the Warm Springs Indian reservation. New
Steele, and Christine Tanz, 189-99. Cambridge: Cambridge University York and London: Longman. Reprinted: Prospect Heights, IL: Waveland
Press. Press.
Gumperz, John J. 1982. Discourse strategies. Cambridge: Cambridge University Pinter, Harold. 1988. Mountain language. New York: Grove-Press.
Press. Sattel, Jack W. 1983. Men, inexpressiveness, and power. Language, gender and
Hughes, Linda A. 1988. "But that's not really mean": Competing in a cooperative .society, ed. by Barrie Thome, Cheris Kramarae, and Nan,cy Henley, 119­
mode. Sex Roles 19:11/12.669--687. ~24. Rowley, MA: Newbury House. I

James, Deborah, and Sandra Clarke. 1993. Women, men and interruptions: A SChegJoR; Emanuel. 1982. Discourse as an interactional achievement: Some uses
qitica) review. Gender and conversational interaction, ed. by Deborah of 'uhuh' and other things that come between sentences. Analyzing dis:
Tannen; 231-80. New York and Oxford: Oxford University Press. cou~e: Text and talk. Georgetown University Round Table on. Languages
James, Deborah, and Janice Drakich. 1993. Understandi~g gender differences in and Linguistics "1981, ed. by Deborah Tannen, 71-93. Washington, DC:
amount of talk. Gender and conversational interaction, ed.. by Deborah Georgetown University Press.
Tannen, 281-312. New York and Oxford: Oxford University Press. ;l:'SchegJoff;Emanuel. 1988. Discourse as an interactional achievement II: An exer­
Jong, Erica. 1973. Fear of flying. New York: Holt, Rinehart and Winston. L;;<:-L.;;, . • cise in conversation analysis. Linguistics in context: ·Connecting observa­
Kakava, Christina. 1989. Argumentative conversation in a Greek family. Paper tion and understanding, ed. by Deborah Tannen, 135-58. Norwood, NJ:
presented at the Annual Meeting of the Linguistic Society of America, AhJex. .
Washington, DC. £ §chiffrin, Deborah. 1984. Jewish argument as sociability. Language in Society
Keenan, Elinor. 1974. Norm-makers, nOrin-breakers: Uses of speech by men and ""_C'ut;;/ .13:3.311-35.
women in a Malagasy community. Explorations in the ethnography of :ollon, Ron. 1985. The machine stops: Silence in the metaphor of malfunction.
speaking, ed. by Richard Bauman and Joel Sherzer, 125-43. Cambridge: Perspectives on silence, ed. by Deborah Tannen and Muriel Saville­
Cambridge University Press. Troike, 21-30. Norwood, NJ: Ahlex.

50 --
51

 Gender and Discourse

Scallon, Ron, and Suzanne B. K. Scallon. 1981. Narrative, literacy and face in
interethnic communication. Norwood, NJ: Ablex.
Shuy~ Roger W. 1982. Topic as the unit of analysis in a criminal law case. Analyz­
ing discourse: Text and talk. Georgetown University Round Tabie on
Languages and Linguistics 1981, ed. by Deborah Tannen, 113-26. Wash­
ington, DC: Georgetown University Press.
Sifianou, Maria. 1992. The use of diminutives in expressing politeness: Modern
Greek versus English. Journal of Pragmatics 17:2.155-73.
Spender, Dale. 1980. Man made language. London: Routledge and Kegan Paul.
Tannen, Deborah. 1981. Indirectness in discourse: Ethnicity as conversational
style. Discourse Processes 4:3.221-38.
Tannen, Deborah. 1984. Conversational style: Analyzing talk among friends. Nor­
wood, NJ: Ablex.
Tannen, Deborah. 1985. Silence: Anything but. Perspectives on silence, ed. by
Deborah Tannen and Muriel Saville-Troike, 93-111. Norwood, NJ: Ablex.
Tannen, Deborah. 1986. That's not what I meant!: How conversational style
makes or breaks your relations with others. New York: William Morrow.
Paperback: Ballantine.
Tannen, Deborah. 1987. Repetition in conversation: Toward a poetics of talk.
Language 63:3.574-605.
Tannen, Deborah. 1989. Talking voices: Repetition, dialogue and imagery in
conversational discourse. Cambridge: Cambridge University Press.
Tannen, Deborah. 1990. You just don't understand: Women and men in conversa~
tion. New York: William Morrow. Paperback: Ballantine.
Tannen, Deborah, and Christina Kakava. 1992. Power and solidarity in
Greek conversation: Disagreeing to agree. Journal of Modern
Studies 10.12-29.
Watanabe, Suwako. 1993. Cultural differences in framing: American and Japanese,/j
group discussions, 176-208. Framing in discourse, ed. by Deborah
nen. New York and Oxford: Oxford University Press.
Wolfowitz, Clare. 1991. Language style and social space: Stylistic choice in
name Javanese. Urbana and Chicago: University of Illinois Press.
West, Candace, and Don H. Zimmerman. 1983. Small insults: A study
tions in cross-sex conversations between unacquainted persons. Languagei~'~,:'-_;
gender and society, ed. by Barrie Thorne, Cheris Kramarae, and N~n,..ir~·-."
Henley, 103-17. Rowley, MA: Newbury House.
Yamada, Haru.1992. American and Japanese business discourse: A r-ATnT"\-:lrtCAn

interactional styles. Norwood, NJ: Ablex.