Introduction to Discourse Analysis | Week 6 | Semester 4 — Groups 1 & 2 | Faculty of Letters and Human Sciences

Week 6: Language, Gender, and Identity

Source A: Paltridge, B. (2012). Discourse Analysis: An Introduction (2nd ed.). Bloomsbury. Chapter 2, §2.3, pp.
18–22.
Source B: Lakoff, R. (1973). Language and Woman's Place. Language, 49(4), 45–80. Chapters 1–2, pp. 3–12.
Source C: Sadiqi, F. (2003). Women, Gender and Language in Morocco. Brill. Chapter 1, pp. 2–9; Chapter 2, pp.
53–56.

SOURCE A — PALTRIDGE (2012): DISCOURSE AND GENDER

Discourse Analysis: An Introduction, Chapter 2, pp. 18–22
Early work in the analysis of gender and discourse looked at the relationship between the use of language and
the biological category of sex. This has now moved to an examination of the ways language is used in relation
to the social category, or rather the socially constructed category, of gender. Thus, from the moment a female
child is born and someone says 'It's a girl!' that child learns how to do being a girl in the particular society and
culture, from the way she talks through to the way she walks, smiles, dresses and combs her hair (Butler
1993).
Gender, then, is not just a natural and inevitable consequence of one's biological sex. It is, rather, 'part of the
routine, ongoing work of everyday, mundane, social interaction'; that is, 'the product of social practice' (Eckert
and McConnell-Ginet 2003: 5). As Swann (2002: 47) has pointed out, gender 'has come to be seen as highly
fluid, or less well defined than it once appeared. Researchers interested in language and gender have focused
increasingly on plurality and diversity amongst female and male language users, and on gender as
performativity — something that is "done" in context, rather than a fixed attribute.'

Simone de Beauvoir famously said 'one is not born, but rather becomes a woman'. Performativity is based on
the view that in saying something, we do, or "become" it. A person learns, for example, how to do and, in turn
"display", being a woman in a particular social setting. People perform particular identities through their use
of language and other ways of expressing themselves in their interactions with each other. Mostly, this is done
unconsciously as we "repeat acts" such as gestures, movement and ways of using language that signify, or
index a particular identity. These acts are not, however, natural nor are they part of the essential attributes of a
person. They are part of what people acquire in their interactions with each other.
Many of the conversations in the TV show Sex and the City are examples of the way the lead characters,
through their use of language, do gender. In the following extract, Miranda asks Carrie why she accepted her
boyfriend's proposal of marriage. In her response, Carrie both enacts and affirms her gendered identity:

Miranda: I'm going to ask you an unpleasant question now. Why did you ever say yes?
Carrie: Because I love him . . . a man you love kneels in the street, and offers you a ring. You say yes.
That is what you do.
Gender, then, is 'not something a person "has", but something that a person does' (Cameron 2005a: 49).
Gender is not a result of what people already are but a result of, among other things, the way they talk and
what they do. As Eckert and McConnell-Ginet (2003: 4) argue: 'gender doesn't just exist, but is continually
produced, reproduced, and indeed changed through people's performance of gendered acts, as they project
their own claimed gender identities, ratify or challenge other's identities, and in various ways support or
challenge systems of gender relations and privilege.'

Gender identity is a complex construction. All levels of language and discourse are involved in doing gender.
Gender interacts with other factors such as social class and ethnicity. As Holmes observes: 'gender is only one
part of a person's social identity, and it is an aspect which will be more or less salient in different contexts. In
some contexts, it may be more important to emphasise one's professional expertise, one's ethnic identity, or
one's age than one's gender.'

SOURCE B — LAKOFF (1973): LANGUAGE AND WOMAN'S PLACE

p. 1
Introduction to Discourse Analysis | Week 6 | Semester 4 — Groups 1 & 2 | Faculty of Letters and Human Sciences

Chapters 1–2, pp. 3–12

CHAPTER 1 — THE PROBLEM

This book is an attempt to provide diagnostic evidence from language use for one type of inequity that has
been claimed to exist in our society: that between the roles of men and women. We will find, I think, that
women experience linguistic discrimination in two ways: in the way they are taught to use language, and in
the way general language use treats them. Both tend to relegate women to certain subservient functions: that
of sex object, or servant.

Women's speech is peculiarly 'unassertive'. Women are taught, from childhood, to use a form of language
which, in effect, is characteristic of people who are in an inferior position. But women also find themselves in
a bind: if they speak as women should — unassertively, politely, tentatively — they are taken less seriously; if
they speak as men do — forcefully, directly, assertively — they are thought unladylike. Women cannot win.
CHAPTER 2 — TALKING LIKE A LADY
There are several ways in which women's language differs from men's. Consider the use of expletives.
Women's expletives are restricted: 'oh dear', 'oh fudge', 'goodness'. Men have a fuller vocabulary here: many
of the more 'taboo' words are theirs alone. A little girl is corrected and reprimanded if she uses a strong
expletive; a little boy, for the same offence, may be scolded, but the severity of the punishment is noticeably
less. The socialisation of women to use 'proper' language is part of a broader pattern of keeping women in
their place.

Consider also the use of adjectives. Women use certain adjectives in normal conversation which men would
not use. The following list illustrates the difference:

Neutral / both sexes Women only

great adorable
terrific charming

cool sweet
neat lovely
— divine

The words in the right-hand column convey only the speaker's emotional reaction and very little else. 'Oh
divine!' says almost nothing beyond 'I like it'. This is one of the ways in which women's language is
expressive but not forceful.
Women also use hedges far more than men. These are expressions like 'well', 'y'know', 'kinda', 'I think', 'I
guess', 'sort of'. Hedges signal tentativeness: they indicate that the speaker is not fully committed to what she
is saying, or that she wants approval before proceeding. Women use this pattern far more than men — not
because they are genuinely more uncertain, but because they have been conditioned to express uncertainty
even when they are not.

Tag questions also function to express uncertainty or seek approval. 'John is here, isn't he?' is not the same as
'Is John here?' The tag version acknowledges the speaker's view while expressing uncertainty and inviting the
hearer's confirmation. Women use tag questions in contexts where men would simply make a statement —
placing them in the position of always requiring the other's approval.
Rising intonation functions similarly. A woman may answer 'What time is it?' with 'Four o'clock?' as if asking
for confirmation of what she knows perfectly well. This transforms a statement into a request for approval.
These patterns — hedges, tag questions, rising intonation, empty adjectives — constitute a system that keeps
women in a subordinate communicative position: expressive but tentative, present but not authoritative.

p. 2
Introduction to Discourse Analysis | Week 6 | Semester 4 — Groups 1 & 2 | Faculty of Letters and Human Sciences

SOURCE C — SADIQI (2003): WOMEN, GENDER AND LANGUAGE IN

MOROCCO
Chapter 1, pp. 2–9 — The Three Frameworks

ESSENTIALISM AND CONSTRUCTIONISM

There have been two influential views in theorizing about language and gender: the essentialist view and the
constructionist view. Within the essentialist view, gender was based on biological sex and was regarded as
essentially dichotomous. Gender was defined by three major clusters of characteristics: innateness, strict
binarism, and bipolarization. The categories 'male' and 'female' were assumed to be homogeneous, opposite,
fixed, and unchangeable. In most human cultures, the male category was commonly taken to be the unmarked,
universal, primary, superior category, of which the female was the marked, secondary, inferior version.

In the constructionist view, everything about gender has been seen as 'constructed' because gender has been
conceptualized as fluid, not static. This view assumes a wide variation in the spectrum of 'human sex'. Unlike
the essentialist view, which considered gender as a separate category from other social categories such as
ethnicity, age, and class, the constructionist view regards gender as intertwined with these categories. It is the
interaction of gender with other social categories which explains its multiple meanings and continuous
creation.
THE DEFICIT THEORY
As its name indicates, the Deficit theory considered women's language as an essentially 'deficient' version of
men's language. This theory was based on an essentialist view of the relationship between language and
gender. In the field of modern linguistics, the Deficit theory materialized in the work of Robin Lakoff, who
published her landmark work Language and Women's Place in 1973. According to Lakoff, gender inequity in
women's use of language stemmed from their marginalisation in society. This inequity was reflected in the
way women were expected to speak and the way they were spoken of. In other words, women's language was
deficient because their social position was so.

Lakoff's book had three major goals: to answer the question of what language can tell us about gender
inequity; to reflect on whether linguists could correct social inequity by changing language; and to enhance
research on language and gender. Although Lakoff's book was not based on sociolinguistic empirical evidence
— it relied on introspection — it ignited a flurry of empirical research in the domain of language and gender.
Her findings were both questioned and supported by various studies.
THE DOMINANCE THEORY
According to the Dominance theory, linguistic differences between men and women were triggered by
inequitable power relations between the two sexes. The male dominance in language was attributable to the
political and cultural dominance of men over women in society. The way men and women spoke reflected
social reality. Power as dominance and control was established by the socially 'appropriate' use of language by
men and women. The Dominance model differed from Lakoff's approach in the sense that it was based on
empirical evidence — analysis of conversations in natural settings — whereas Lakoff's work relied almost
exclusively on introspection.
Men 'cornered' women in smaller and less significant space on the linguistic floor by: (i) interruptions and
overlaps (Zimmerman and West 1975), (ii) failing to take up women's conversational gambits (Fishman
1983), (iii) the sheer number of words (Spender 1980), and (iv) the semantic derogation of women. However,
there are problems with the Dominance theory. The notion of power is oversimplified. If men talk more, they
are dominant; if women talk more, they are too talkative. Interruptions have many meanings: they can signal
aggression, support, or familiarity. Linguistic variables do not have unified functions in language use.
THE DIFFERENCE THEORY
The Difference theory was largely based on the work of Maltz and Borker (1982); however, Tannen's (1990)
book You Just Don't Understand is its best known proponent. In the Difference theory, women's language was
not a defective copy of men's; it was simply different. The Difference theory was also called the 'two-cultures'
model. Boys and girls, not being socialised in the same way, end up acquiring two different sociolinguistic
subcultures: the male subculture and the female subculture. Difference presupposed neither dominance nor
subordination and de-emphasised power.

However, problems with the Difference theory remain. It does not differ from the Deficit and Dominance
theories in taking males as the norm. It naturalises difference as symmetrical while ignoring that one culture
has more power than the other. Cameron's critique goes further: calling miscommunication 'cross-cultural'

p. 3
Introduction to Discourse Analysis | Week 6 | Semester 4 — Groups 1 & 2 | Faculty of Letters and Human Sciences

hides the fact that one culture has more institutional power, and this framing can be used to explain away
structural inequality as a matter of cultural difference rather than of power.
CHAPTER 2, PP. 53–56 — THE MARRIAGE FORMULA AS SPEECH ACT
Of all the components of Moroccan culture, it is social organization that has the strongest impact on gender
perception and gender construction. Moroccan society is built on clear role assignment for men and women.
The patriarchal system is built on the exclusion of women from spaces of public power. This channelling of
gender roles is largely achieved through the use of language.

The most significant rituals in Moroccan culture are the ones that accompany marriage. Two elements are
indispensable for the validity of a Muslim marriage: the pronouncement of the appropriate linguistic formula
and the mahr (the sum of money that the future groom pays to the family of the future bride). The marriage
formula is made up of the groom's offer and the bride's father's (or male tutor's) acceptance. The words of the
offer must use a term implying marriage, such as:
jewwejt-ak lbent wla sunnati llahi wa rasulih
'I married my daughter to you according to God's and the Prophet's Sunnah'

The linguistic formula has the power of the law: lkelma bhal lweshma — literally 'a word is like a tattoo',
meaning 'a marriage promise is definite'. No marriage contract may be valid without the ritual marriage
formula. Although present and visible, women are often silent and not heard in the marriage ritual. The bride's
father, or in his absence a male close kin, speaks for her. Men speak; women are spoken for.

The role of the marriage formula is reinforced by the fact that marriage can be broken by the pronouncement
of another formula, again pronounced by a man — the husband. In Islamic law, to repudiate a wife, the
husband has only to pronounce three times:
?anti Taliq / nti mTelqa / llefg-am
'You are divorced' — in Standard Arabic, Moroccan Arabic, and Berber respectively

The triple pronouncement of this formula has the legal power of the act of divorce — it is a speech act in the
sense of Austin (1962). The legal power of both the marriage and divorce formulas is effective only when
used by men. Being male is a precondition for the felicity of these speech acts. Language is used here as a
powerful channel through which male power is perpetuated.

Paradoxically, a woman's consent to marriage is necessary but need not be verbal. The bride's oral consent —
her word — is not a requirement in the marriage ritual. A woman's silence or her crying is often considered as
consent. The implication is profound: women's speech is not required to have the force of a speech act in the
contexts that determine their lives most directly. This is what Lakoff, Spender, and Tannen all miss: the
question is not just how women speak — it is whether women's speech is heard as speech at all.

BRINGING IT TOGETHER — DISCOURSE, GENDER, AND IDENTITY

Paltridge (2012), pp. 21–22 · Sadiqi (2003), p. 56
The point here, then, is that no way of speaking has only one potential meaning: the meanings it conveys in
one context are not necessarily the same ones it conveys in another, and it may also acquire new meanings
over time. People do perform gender differently in different contexts, and do sometimes behave in ways we
would normally associate with the 'other' gender. The relationship between language and gender is almost
always indirect, mediated by something else. The ways that people speak are, in the first instance, associated
with particular roles, activities and personality traits — such as being a mother, gossiping, or being modest.
The extent to which these roles, activities and personality traits become associated in a particular culture with
being gendered leads to these ways of speaking pointing to, or indexing, a particular gender identity. Identity
is not something that is pre-assigned in fixed social categories. It is, rather, something that emerges in
practice, through the use of discourse. (Paltridge 2012: 21–22)

p. 4
Introduction to Discourse Analysis | Week 6 | Semester 4 — Groups 1 & 2 | Faculty of Letters and Human Sciences

As Cameron and Kulick (2003: 58) put it: 'The actual balance between them is not determined in advance by
some general principle, but has to be negotiated in specific situations, since meaning is not only in the
language itself, but also in the context where language is used by particular speakers for particular purposes.'
A person, then, will have a multiplicity of identities or personae which may be at play all at the same time, at
different levels of prominence. Not all of these identities will be equally salient at a particular moment.
Different aspects of identity may be foregrounded at different points in time and for different — conscious or
unconscious — reasons.

This is precisely what the Moroccan marriage ritual makes visible. The linguistic formula does not merely
describe a social arrangement — it enacts and produces it. By repeatedly performing a ritual in which women
are spoken for and men speak, the ritual does not just reflect social relations; it produces and reproduces them.
In Austin's terms, the formula is a performative. In Butler's terms, it is an act of gender construction. In
discourse analysis terms, it is evidence that discourse is not a neutral medium for communicating pre-existing
social facts — it is one of the primary mechanisms through which those social facts are created, sustained, and
naturalised. (Sadiqi 2003: 56)
This is why studying the relationship between language and gender matters beyond the linguistics classroom.
Discourse is one of the main mechanisms through which who can speak, who can act, and who counts are
made to seem natural and inevitable rather than constructed and changeable. If gender is constructed through
language, it can also be deconstructed through language — in naming practices, legal language, institutional
discourse, and everyday interaction.

QUESTIONS
Answer in writing. Draw on the three sources where relevant.

1. Sadiqi identifies three frameworks, Deficit, Dominance, and Difference, for explaining gender and
language. In your own words, explain what each framework claims, and identify one specific
limitation of each. Which framework do you find most convincing, and why?
Don't just describe each framework. Engage critically with it.

2. Sadiqi shows that in the Moroccan marriage formula, being male is a felicity condition for the
speech act to succeed. Using Austin's framework (which you studied in Week 4), explain what this
means. Then explain what this reveals about gender and language that Lakoff's Deficit framework
cannot account for.
Be precise: what is a felicity condition, and what happens to a speech act when one is missing?

3. Extended response (exam practice): Using all three sources, write one paragraph of 6–8 sentences
arguing that gender is not a fixed biological property but something constructed and performed
through discourse. Use at least three key terms from the readings and at least one specific example
from each source.

p. 5