Linguistic: Approaches to Gender in Literary Texts
ANNA LIVIA
1 Introduction
The question of gender in literary texts has been approached by linguists in
two different ways. The first involves a comparison of the fiction created by
male and female authors and is typified by the search for "the female sentence"
or a specifically female style of writing. The second involves a study of the
uses to which the linguistic gender system of different languages has been put
in literary works. In the former, gender is seen as a cultural property of the
author, in the latter, a morphological property of the text. A third perspective
on language and gender in literary texts is provided by translators and translation theorists. Translation theorists typically view a text as expressive of a particular time and place as well as being expressed in a particular language
The differences between source and target language may be accompanied
by differences in culture and period, thus translators often work with both
morphological gender and cultural gender. In this chapter, I will discuss men's
and women's style in literature as well as literary uses of linguistic gender.
I will also survey material on translation theory and what it offers to students
of gender.
2 Male and Female Literary Styles
The most prominent modern thinker to discuss the differences between male
and female literary styles is Virginia Woolf, writing at the beginning of the
twentieth century. In a review of Dorothy Richardson's novel Revolving Lights
(1923), she describes the female sentence as "of a more elastic fibre than the
Linguistic Approaches to Gender 143
old, capable of stretching to the extreme, of suspending the frailest particles, of
enveloping the vaguest shapes" (Woolf 1990b: 72). Assuming the traditional
literary sentence to be masculine, she argues that it simply does not fit women,
who need something less pompous and more elastic which they can bend in
different ways to suit their purpose. However, descriptions such as "more
elastic," "too loose, too heavy, too pompous" are annoyingly vague and impossible
to quantify.
Woolf comes closest to giving a more specific evaluation of the female sentence
in a review of Dorothy Richardson's The Tunnel (1919). Here she quotes
a passage of interior monologue as triumphantly escaping "the him and her"
and embedding the reader in the consciousness of the character: "It is like
dropping everything and walking backward to something you know is there.
However far you go out, you come back. I am back now" (Woolf 1990b: 71).
The exact relationship between the pronouns "you" and "I" in this passage is
unclear. They seem to refer to the same person, the self, but also to include the
reader. Because we do not know who "I" is, we have no referent for the
temporal or spatial indicators "now" or "come back" either. This slipperiness
of the referent seems to be what Woolf means by "elasticity."
It is significant that Woolf chose the writings of Dorothy Richardson to
illustrate the female sentence, and specifically, a passage of interior monologue.
Interior monologue has the property of breaking down the boundaries between
character and narrator, so that the angle of focalization (who sees the action)
coincides with the narration of that action (who tells about the action). More
traditional methods of storytelling present a narrator, who recounts, but is
separate from the character whose point of view is related. It was one of the
projects of modernism (and both Richardson and Woolf are considered modernist)
to render the depths of modern experience in an appropriate form,
which meant breaking away from what they considered a smug, self-satisfied
Edwardian frame of social realism and an omniscient narrator. Although we
cannot speak of a "modernist sentence" as such, nevertheless, the other authors
usually included in the modernist canon such as T. S. Eliot, James Joyce, D. H.
Lawrence, Ezra Pound, as well as Woolf and Richardson, have all experimented
with sentence fragments, elimination of predicates, meandering syntax with
many clauses in apposition. These are the very elements which tend also to
typify interior monologue.
We would do best, therefore, to take Woolf's description of the female sentence
as a literary rather than a linguistic commentary. As the stuffy Edwardian
era gave way to greater freedom for women, especially in the inter-war period,
so women novelists felt freer to express themselves in new ways. The literary
movement of modernism coincided with (and was also itself a product of) the
new social developments consequent upon the horror and paradoxical liberty
of the post-First World War period. Woolf's unremitting self-consciousness is
shared by her contemporaries. Indeed her precursor, Henry James, writes of
his own awareness of a fragmented consciousness in a discussion of his novel
Portrait of a Lady (quoted in Millett 1951: v): "'Place the centre of the subject in
144 Anna Livia
the woman's own consciousness,' I said to myself, 'and you get as interesting
and as beautiful a difficulty as you could wish'." The challenge of this "beautiful
difficulty" may be taken up by men or women authors.
Although Woolf's discussion of feminine style is impressionistic and essentialist,
modern theorists have looked at more subtle differences in men's and
women's writing. Sara Mills examines features such as descriptions of characters
and self-descriptions in personal ads. In an analysis of a romance novel by
best-selling author Barbara Taylor Bradford, Mills demonstrates that the
actions performed by the female character are of a different quality from those
performed by the male (1995: 147-9). Parts of the woman's body move without
her volition and she is represented as the passive recipient of the male's
actions. The male acts while the female feels.
That male and female characters in fiction receive very different treatment is
not particularly controversial, but the claim that women's writing differs in
some essential way from that of men is more tendentious. Quoting Woolf's
categorization of the female sentence as loose and accretive. Mills proceeds to
look at some concrete examples to see what proof there may be of these differences.
She concludes that the concept of a female-authored sentence stems
from overgeneralization on the part of the literary critic rather than from any
inherent quality in the writing, but she demonstrates that a female (or male)
affiliation may be a motivating factor in certain texts (1995: 47-8). Comparing
descriptions of a landscape taken from two well-known novels, Anita Brookner's
Hotel du Lac and Malcolm Lowry's Under the Volcano, she shows that the first is
conventionally feminine while the second is conventionally masculine (1995:
58-60). The features which mark the first as feminine include: abundant use of
epistemic modality ("it was supposed," "it could be seen"); grammatically
complex, meandering sentences with many clauses in apposition; and an impressionistic,
subjective vocabulary such as "stiffish," "skimming," and "area
of grey." In contrast, the second landscape is masculine in style, featuring the
absence of an obvious authorial voice; an impersonal, objective tone; the
description of amenities rather than people: "Overlooking one of these valleys,
which is dominated by two volcanoes, lies, six thousand feet above sea-level,
the town of Quauhnahuac" (1995: 60).
Female affiliation, or a distinctly feminist style, is a third possibility, in which
the tone may be ironic or detached; female characters are presented as assertive
and self-confident, and the reader is addressed directly and drawn into the
text to share the narrator's point of view. Mills quotes a passage from Ellen
Galford's Moll Cutpurse to illustrate her point: "She had a voice like a bellowing
ox and a laugh like a love-sick lion" (1995: 60-1). This heroine is clearly very
different from the passive female, mere object of the male's attention. The oxymoronic
(apparently contradictory) quality of the comparison between Moll
and a "love-sick lion" demonstrates the playful, almost parodic nature of the
description. A lion is usually a symbol of masculine strength, but this lion is in
love and therefore emotional. Moll thus combines a traditionally masculine
quality (strength) with a traditionally feminine quality (deep feeling).
Linguistic Approaches to Gender 145
For contemporary critics, it is possible to identify certain features such as
complex sentences with many subordinate clauses and a vocabulary that is
vague and impressionistic as typifying the "female sentence," but there is no
essential link between the fact of being a woman and this type of writing. It is
a style which may be deliberately chosen by either sex. Indeed, if one considers
Marcel Proust's sometimes page-length sentences, and his deliberations about
the exact quality of colors and smells, one is obliged to classify his style as
distinctly feminine:
Jamais je ne m'etais avise qu'elle pouvait avoir une figure rouge, une cravate mauve
comme Mme Sazerat, et I'ovale de ses joues me fit tellement souvenir de personnes que
j'avais vues a la maison que le soupqon m'effleura, pour se dissiper aussitot, que cette
dame, en son principe generateur, en toutes ses molecules n'etait peut-etre pas suhstantiellement
la duchesse de Guermantes, mais que son corps, ignorant du nom qu'on
lui appliquait, appartenait a un certain type feminin qui comprenait aussi des femmes de
medecins et de commerqants.
(I had never imagined that she could have a red face, a mauve scarf like Madame
Sazerat, and her oval cheeks reminded me so much of people I had seen at home
that I had the fleeting suspicion, a suspicion which evaporated immediately
afterwards, that this lady, in her generative principle, in each one of her molecules
was perhaps not in substance the Duchess of Guermantes but that her
body, ignorant of the name she had been given, belonged to a certain feminine
type which also included the wives of doctors and tradespeople.) (Proust 1954:
209-10)
Proust's sentence in the above extract is indisputably long, complex and
meandering, convoluted and concerned with female apparel and appearance -
all traits which have been classified "feminine."
It is equally possible for a woman author to deliberately flout this convention
and write in a recognizably feminist style, or indeed a traditionally masculine
one. The writer James Tiptree Junior was declared by the science fiction author
Robert Silverberg to be a man in the introduction to one of her short story
collections:
For me there is something ineluctably masculine about Tiptree's writing. I don't
think that a woman could have written the short stories of Hemingway, just as I
don't think a man could have written the novels of Jane Austen, and in this way
I think that Tiptree is male. (Silverberg 1975: xii)
Tiptree was invited to participate in a symposium organized by the science
fiction magazine Khatru, the ensuing discussion being published in issues 3
and 4, but "his" style was felt to be so rebarbative that "he" was asked to
withdraw (Lefanu 1988: 105-6). At this point "he" revealed that "he" was
none other than Alice Sheldon, a renowned, and definitely female, author.
The ensuing discussion of each participant's perceptions and misconceptions
turned out to be the most fruitful part of the forum.
146 Anna Livia
Novels may be identified as the work of a woman purely because of their
content. The British feminist publishing company Virago was about to publish
a novel by a young Indian woman, when they learned that the book had in
fact been written by a middle-aged English vicar. Upon hearing this. Virago
stopped publication. As a company that was set up specifically to publish
books by women, they were angry at being hoodwinked into accepting a
manuscript written by a man. Critics of Virago's actions argued that it was the
submissive, downtrodden status of the heroine which had at first convinced
the editors that the novel was written by an Indian woman. This, they said,
was a form of racism as the editors assumed that a victim status was typical of
Asian women. Dinty Moore, a male author, was assumed to be female when
he published a short story in an anthology of reminiscences of a Catholic girls'
school. This also caused hot debate, though the anthology was not withdrawn
(Rubin 1975).
In a study on the micro-level of text-making (looking at the immediate
linguistic environment rather than the whole novel), Susan Ehrlich (1990) has
analyzed the use of reported speech and thought in canonical texts, particularly
the novels of Virginia Woolf. She compares Woolf's style with that of
Henry James and Ernest Hemingway with regard to the types of cohesive
devices each uses (1990: 101-3). James depends heavily on what is known
as grammatical cohesion, or anaphora. This means he introduces a character,
and as soon as the reader has had the chance to form a mental image of this
character, he replaces the character's name with a pronoun (this is, of course,
a very traditional strategy). Hemingway relies instead on lexical cohesion, or
a simple repetition of the character's name. Woolf, in contrast, uses a much
greater variety of cohesive devices including grammatical and lexical cohesion
as well as semantic connectors, temporal linking, and progressive aspect. A
semantic connector tells the reader explicitly to connect two pieces of information
in a particular way: at the same time; in this way; in addition. Temporal
linking gives two clauses the same time reference and is a feature that often
involves hypothetical clauses which have no time reference of their own:
Edith would be sure to know; I would have arrived before the others. Progressive
aspect also links two propositions where one clause provides an anchor for
the other.
The advantage of research like Ehrlich's is that it provides a concrete set of
criteria by which to distinguish different literary styles. We cannot assume
that all women will write like Woolf and all men like James or Hemingway,
but if we know that a researcher has based his or her claims entirely on a
study of canonical texts by male authors, we can predict that certain types of
data will be missing.
Studies of gender in literary texts have not been confined to stylistic analysis
but also include investigations into the representation of men and women and
what these literary models can tell us about conversational expectations in the
real world. In an insightful analysis of the preferred conversational strategies of
a husband and wife at loggerheads with each other, Robin Lakoff and Deborah
Linguistic Approaches to Gender 147
Tannen (1994) propose a new methodology for interpreting communication
between the sexes. They analyze the contrasting conversational strategies of
Johan and Marianne in Ingmar Bergman's film. Scenes from a Marriage.
In this study, they introduce the concepts of pragmatic identity, pragmatic
synonymy, and pragmatic homonymy, which, as they demonstrate, replicate
the semantic relations of synonymy (having the same meaning but a different
form), homonymy (having the same form but a different meaning), and identity
(having the same form and the same meaning) (1994: 148-9). The analysis
shows that the two partners often use similar strategies to very different ends
and, an even more significant finding, that they also achieve the same end
(avoiding conflict) by very different strategies: excessive verbiage on Marianne's
part and pompous pontification on Johan's. Marianne prattles: "Here already!
You weren't coming until tomorrow. What a lovely surprise. Are you hungry?
And me with my hair in curlers" (1994: 152); Johann drones: "I'd been out all
day at the institute with the zombie from the ministry. You wonder sometimes
who those idiots are who sit on the state moneybags" (1994: 154-5). Marianne's
contribution is characterized by short sentences, abrupt changes of topic, and
a homely, domestic tone. Johan's style is more cohesive and elaborate; it concerns
the world of work and is distanced from the current situation. Although
their styles are very different, they share the same goal: each is trying to avoid
a confrontation about their deteriorating marriage.
Justifying their choice of the constructed, non-spontaneous dialogue of a
film script, Lakoff and Tannen explain that "artificial dialog may represent an
internalized model. . . for the production of conversation - a competence model
that speakers have access to" (1994: 137). They later define this type of competence
as "the knowledge a speaker has at his/her disposal to determine what
s/he is reasonably expected to contribute, in terms of the implicitly internalized
assumptions made in her/his speech community" (1994: 139). Although
this type of analysis has not been widely imitated, it demonstrates the utility
of looking at constructed dialogue precisely because such pre-planned scripts
allow us to see what pragmatic roles have been internalized and what expectations
speakers have of patterns of speech appropriate for each sex.
In the French tradition, the e'criture feminine school, made famous by such
writers as Helene Cixous, Chantal Chawaf, and Annie Leclerc in the 1970s,
defines women's writing as corporeal, tied to the workings of the body, and at
the same time multivalent and polysemic, defying syntactic norms. Chawaf
challenges the reader with the rhetorical question "I'aboutissement de Vecriture
n'est-il pas de prononcer le corps?" (1976: 18) ("is not the aim of writing to
articulate the body?"), while Cixous exhorts, "Ecris! L'Ecriture est pour toi, tu es
pour toi, ton corps est toi, prends-le. [ .. . ] Les femmes sont corps. Plus corps done
plus ecriture" (Cixous and Clement 1975: 40, 48) ("Write! Writing is for you, you
are for you, your body is yours, take it. [. . . ] Women are bodies. More body
so more writing"). The assertion that women are bodies is a little puzzling.
Are women, according to Cixous, more corporeal than men? How can writing
be corporeal except in a pen and ink sense?
148 Anna Livia
Ecriture feminine came out of the women's liberation movement as a response
to the complaint that men's writing was increasingly abstract and distanced
from material concerns. Where the prevailing ideology, which dominates most
text forms from highbrow novels to the language of advertising, tended to see
the female body as dirty, messy, shameful, and generally problematic, ecriture
feminine set out to celebrate this body in all its wet, bloody, sticky functions
and by-products from menarche to pregnancy and childbirth to menopause.
Where the subliminal message of mainstream, misogynist discourse was that
women were mired in their own physicality and therefore constitutionally
unable to produce great works of fiction, ecriture feminine saw men as cut off
from their own bodies, decentered and more interested in the play of signifiers
than in their real-world referents.
When we encounter sentences like the following from Cixous's La Jeune ne'e
(The Newly Born Woman), "Alors elle, immobile et apparemment passive, livree aux
regards, qu'elle appelle, qu'elle prend" ("Then she, immobile and apparently passive,
prey to glances, that she calls, that she takes") (Cixous and Clement 1975:
237), which has no main verb and two subordinate clauses, we may feel lost,
confused, or simply impatient. In order to appreciate the innovatory quality of
this style, which provides no object for usually transitive verbs (who does she
call? what does she take?), we need to feel the weight of the well-formed French
sentence and the desire of the feminist writer to wriggle out from under it at
all costs. For the French, their language is "la langue de Moliere" (the language
of Moliere), while English is "la langue de Shakespeare" (the language of
Shakespeare). The apex of literary achievement was apparently achieved many
centuries ago, and perfected by male writers. Ecriture feminine is a reaction to
this assumption of perfection and its attribution to men.
3 Literary Uses of Linguistic Gender
In my own work on the literary uses of linguistic gender, I have examined the
role of gender concord in the creation of particular stylistic effects such as
focalization (or point of view), empathy, and textual cohesion (what makes
everything fit together) (Livia 2000). Insofar as gender concord may be considered
a choice in a given language, and not a morphological or syntactic necessity,
it can be used as a stylistic device to express some aspect of character or
personality. While Judith Butler's research on the performativity of gender
emphasizes the iterative and citational aspects of speech, greatly reducing the
role of speaker agency, my own work on the gender performances of characters
such as drag queens, transsexuals, and hermaphrodites, and those whose gender
is never given, demonstrates that observing (or ignoring) the requirements of
gender concord allows authors to express a wide range of positions.
In her pioneering work Gender Trouble, Judith Butler argues that speakers, or
in her words "culturally intelligible subjects," are the results, rather than the
Linguistic Approaches to Gender 149
creators, "of a rule-bound discourse that inserts itself into the pervasive and
mundane signifying acts of linguistic life" (1990: 145). Although her prose is a
little dense, what this means in simple terms is that she sees individual speakers
as being formed by the discourse they use. This discourse is "performative"
because it is by uttering (or performing) it that speakers, obligatorily, gender
themselves. They are compelled by the syntactic structure and vocabulary
available to position themselves only in certain restricted ways with regard to
gender, that is, the traditional roles of "men" and "women." They are not free
to take up any gender stance they like, for this would not be "culturally intelligible."
Although she does suggest three linguistic strategies by which a speaker
can undermine the system (parody, subversion, and fragmentation), on the
whole Butler sees agency as severely curtailed, limited merely to "variations
on repetition." For her, it is the gender norms themselves which provide the
lynchpins keeping "man" and "woman" in their place. She argues that "the loss
of gender norms would have the effect of proliferating gender configurations,
destabilizing substantive identity, depriving the naturalizing narratives of compulsory
heterosexuality of their cultural protagonists" (1990: 146). Once these
stabilizing norms have been lost, other possibilities become available, moving
beyond the heteronormative lynchpins "man" and "woman."
This view of gender as performative has become a key tenet of queer theory,
which investigates and analyzes "the naturalizing narratives of compulsory
heterosexuality" and the various sexually liminal figures who do not fit into this
traditional framework. Arguing against the linguistic determinism of Butler's
stance, I refute the claim that gender, and particularly linguistic gender, is rigidly
confining and explore the different messages it can convey. My research on a
corpus of literary texts in both English and French, presented in Pronoun Envy
(2000), shows that the realm of what is "culturally intelligible" is much wider
and more diverse than queer theorists have supposed and that the traditional
gender norms are often used as a foil against which more experimental positions
are understood.
Anne Garreta, writing in French, and Maureen Duffy, Sarah Caudwell, and
Jeanette Winterson, writing in English, have each created characters without
gender in at least one of their works. Nowhere in these novels is there any
grammatical clue as to whether the main protagonists are male or female. In
French this is a particularly difficult feat, for gender is usually conveyed not
only by the third-person pronouns il/elle, ils/elles (like the English he/she and
unlike English they) but also in adjectives and past participles. Thus in a sentence
of five words like la vieille femme est assise ("the old woman sat down"),
the gender of the person sitting is conveyed four times: in the definite determiner
la, in the form of the adjective vieille, in the lexical item femme, and in the
form of the adjective assise. In English, the difficulty is decreased by the fact
that morphological (or linguistic) gender is limited to the distinction between
he/she, his/her, his/hers.
Garreta's novel Sphinx features both a genderless narrator and his or her
genderless beloved. The novel is written in the first-person singular je ("I"),
150 Anna Livia
which is gender-neutral. Thus when the narrator describes his or her own
actions, the author can avoid giving gender information by using only genderneutral
adjectives and tenses, like the passe simple rather than the passe compose.
However, gender-neutral adjectives and expressions tend to be less frequently
used than those which agree with the gender of the noun. The use of the passe
simple rather than the more common passe compose also introduces a literary,
almost anachronistic element to the text. Since the novel recounts how a White
Parisian theology student becomes a disc jockey in a seedy bar and falls in
love with a Black American disco dancer, the use of markedly literary tenses
and descriptive expressions seems somewhat out of place. It is as though the
theology student never really left the seminary.
When the narrator describes the actions and attributes of the beloved, the
situation becomes even more complex and the language somewhat convoluted,
for here the use of pronouns must be avoided as well. The beloved can
never simply be referred to as il (he) or elle (she) and various techniques are
introduced to avoid this. Often the proper name. A***, is repeated. This repetition
makes it appear that a new character is being introduced, so that A***
(already confined to an initial and a string of asterisks) never becomes a familiar
figure, but always seems a little strange and distant.
Another technique used by the author to avoid conveying A***'s gender is
to describe A***'s body parts rather than the person himself/herself. Instead of
the more straightforward "Elle az^ait les hanches musculeuses, les cheveux rases
et le visage ainsi rendu a sa pure nudite" ("she had muscular hips, a shaven head
and her face was thus returned to its pure, bare state"), for example, the
author is obliged to avoid mention of gender by describing A***'s body in the
following, far more distanced and depersonalized way: "Le modele musculeux
de ses hanches . . .ses cheveux rases . . . le visage ainsi rendu a sa pure nudite" ("the
muscular moulding of her/his hips . . . her/his shaven hair . . . the face thus
restored to its naked purity") (1986: 27). Because A*** is systematically referred
to by a proper name, or in terms of parts of the body rather than the whole,
this character seems fragmented and static.
Clearly, a text which avoids gender agreement produces a very different
effect from one which follows a more orthodox pattern of reference. But it is
perfectly possible to create a whole novel on this basis, as Garreta's achievement
has shown. One could argue that the style of Sphinx, whether or not it
was initially imposed by the decision to avoid gender, suits the plot of the
novel admirably. Given the different worlds the narrator and the beloved
inhabited prior to their meeting, and the enormous social distance between
them, one a White Parisian intellectual, the other a Black dancer from Harlem,
the presentation of A*** as strange, constantly unfamiliar, and composed of a
series of bodily fragments, creates an exoticism which well suits the story of
infatuation, incomprehension, and loss.
Maureen Duffy's novel Love Child tells the story of the adolescent Kit and
his/her murderous jealousy for Ajax, his/her father's secretary whom he/she
believes to be his/her mother's lover. (In the third person, gender-neutral
Linguistic Approaches to Gender 151
pronominal reference can become extremely clumsy.) While the mother and
father are clearly gendered, Duffy gives no clue as to Kit or Ajax's gender. The
effect of this is rather different for each character since Kit, as first-person
narrator, can use the pronoun "I," while Ajax is never referred to by pronoun.
In this Love Child resembles Sphinx. A character referred to without pronouns
is simultaneously less empathic and less of a coherent whole. Empathy for a
character may be gauged by the types of reference used for that character.
Repetition of the proper name and the use of different lexical items such as
"my father's secretary," "my mother's lover" create the least empathy, while
pronouns and ellipsis create the most. Use of pronouns and ellipsis presuppose
that the reader is already familiar with the referent and can readily access it,
given minimal or zero prompts. In a similar pattern, the linguistic device which
creates the strongest cohesive link is ellipsis followed by pronominalization. If
the proper name is simply repeated, there is no necessary link forged between
each of its appearances. In contrast, in the following sentence: "Ajax spieled,
pattered, manipulated unseen puppets, drew scenes and characters" (1994: 50),
in order to understand that Ajax is the subject not only of "spieled," but also
of "pattered," "manipulated," and "drew," the reader must connect the four
verbs, and this connection creates a strongly cohesive text.
While Kit comes across as a lonely, angry, jealous teenager who causes the
death of his/her mother's lover, Ajax (like A***) seems not quite real, a mere
collection of qualities and attributes, not someone who acts on his/her own
behalf. We never find out if Kit is an adolescent girl witnessing a lesbian affair;
a boy jealous of his mother's male suitor; a boy watching his mother flirt with
another woman; or a girl who is aware of her mother's heterosexual conquests.
Each interpretation gives very different readings to the text. Nevertheless,
Kit is a character for whom the reader can feel some emotional connection
while Ajax is not. It is the presence or absence of pronouns which creates this
contrast, not information about gender, since neither character is gendered.
Jeanette Winterson's Written on the Body and Sarah Caudwell's mysteries
revolve around a genderless narrator, but all third-person characters are
assigned traditional gender markers; these novels do not, therefore, offer
the same degree of complexity as Duffy's or Garreta's.
Science fiction authors, like Ursula Le Guin and Marge Piercy, have used
the possibilities offered by new worlds and new biologies to invent imaginary
communities whose gender positions are very different from those of twentiethcentury
Earth. In The Left Hand of Darkness, Le Guin introduces the ambisexual
Gethenians whose gender status changes at different phases of their life-cycle.
During most of the year their bodies are asexual, but when they enter their
mating phase (called kemmer) they develop either male or female reproductive
organs. They never know in advance which organs will develop and their
gender may change from one period of kemmer to another. For her part, Piercy
has experimented with Utopian worlds in which gender is so insignificant
that it is no longer encoded in the grammar. In the futuristic community of
Mattapoisett, described in Woman on the Edge of Time, people are anatomically
152 Anna Livia
male or female, but this distinction is almost entirely irrelevant in determining
their social roles. To demonstrate the effect this egalitarianism has on the
language they speak, Piercy has invented the pronouns person and per in place
of he/she and his/her/hers. These neologisms are used to describe the futuristic
characters, in contrast with the twentieth-century characters.
Monique Wittig, writing in French, has experimented with a different aspect
of the linguistic gender system in each one of her works. In her first novel,
VOpoponax (1966), she uses on as the voice of the narrator, recounting the daily
lives and relationships among a group of young schoolchildren in a small
village in eastern France. Traditional literary texts in French are narrated
either in the first-person je or in the third-person il or elle. On is grammatically
a third-person singular pronoun which, unlike il/elle, is not marked for gender.
Furthermore, it may be used with the meaning of I, we (inclusive, i.e. I and
you, or exclusive, i.e. I and a third party); "you" (singular or plural); "he" or
"she" or "they" (masculine or feminine). This means that on is both remarkably
flexible to manipulate and remarkably slippery in meaning. Wittig chose
it because it did not encode gender information, but its effect is to neutralize
other oppositions as well.
On refers most often to the narrator, a little girl called Catherine Legrand,
but it is not always clear from the immediate context when it refers exclusively
to Catherine, when it also refers to the other children who are all participating
in the same actions and share the narrator's thoughts and feelings, and when
it includes not only other children but adults as well. In one particularly memorable
scene, a new child arrives at school and is instantly separated from the
other children, sitting on a bench by herself. Subsequently, in a sequence of
increasing violence, she is searched for lice, then beaten on the head by hand
and then with rulers. Who performs each of these acts? It must be the teacher
who seats the girl apart from the others, but does she also participate in, or
even instigate, searching for lice? Wittig states that she uses on to "universalize"
a very specific and somewhat unusual point of view: that of a group of young
children. In fact on does far more than this. Because of its many possible
meanings, it forces the reader to pay close attention not only to assumptions
about gender, but also to assumptions about age appropriateness and common
sense.
In Les Gue'rilleres (1969), Wittig uses the feminine plural elles to tell the story
of a group of women warriors who live a separatist lifestyle away from men.
This feminine plural is less common than the feminine singular elle, the masculine
plural ils, and the masculine singular il, for the following grammatical
reasons. II can refer either to an animate entity such as a person (Eric arrive, il
aime le chocolat, "Eric is coming, he likes chocolate"); to an inanimate object (le
clou m'a gri^, il m'afait de la peine, "the nail scratched me, it hurt me"); or to an
abstract idea (le theoreme est trop abstrait, il est mat explique, "the theorem is too
abstract, it is ill-explained"). II is also used as a "dummy morpheme" or verb
marker in meteorological and modal expressions such as il faut venir ("it is
necessary to come," i.e. you must come); il pleut ("it is raining"). Elle, in contrast.
Linguistic Approaches to Gender 153
refers to a person, inanimate object, or abstract idea, but is never used in
modal or meteorological expressions. The plural ils refers to people, inanimate
objects, abstract ideas, or a combination of these, as does elles. However, ils is
also used for a combination of grammatically masculine and feminine items,
while elles is restricted to feminine items only.
As well as these grammatical reasons for the more limited use of elles, the
French psychoanalyst Luce Irigaray (1987: 81-123) has found that people talk
more rarely about groups of women than about men, mixed groups, or singular
subjects. When asked to finish sample sentences, her respondents were far
more likely to speak of singular, masculine referents than of anyone else.
Although il/elle and ils/elles appear to have contrasting but equal functions in
the pronominal system, their frequency of use is actually steeply graded from
il to ils to elle to elles. A novel in which the least favored pronoun among the
third-person set, elles, is used as the main reference point of narration is a
radical innovation.
For the narrator of Le Corps lesbien ("The Lesbian Body," 1973), Wittig has
invented the pronoun J/e, a divided I who describes and interacts with another
woman. This "barred" spelling is repeated throughout the first-person possessive
paradigm: me is spelled m/e, ma; m/a, mon; m/on, and moi; m/oi. Although,
as we have seen, je is non-gendered, it is clear in The Lesbian Body that the
narrator is a woman since there are frequent, lyrical descriptions of specifically
female body parts such as clitoris, labia, vagina.
As for exactly what this divided J/e represents, Wittig herself has provided
two, rather different explanations. In the "Author's Note" to the English translation
of 1975, Wittig states thatJe, as a feminine subject, is obliged to force her
way into language since what is human is, grammatically, masculine, as elle
and elles are subsumed under il and ils. The female writer must use a language
which is structured to erase her (as elle is erased in il). Wittig explains that the
bar through the J/e is intended as a visual reminder of women's alienation
from (by and within) language. Ten years later, however, Wittig claims: "the
bar in the J/e of the Lesbian Body is a sign of excess. A sign that helps to
imagine an excess of I, an I exalted." This new explanation suggests that, far
from signaling the difficulty for women of taking up the subject position in a
linguistic structure in which the masculine is both the unmarked and the
universal term, the bar through the j/e has the positive value of an exuberance
so powerful it is "like a lava flow that nothing can stop" (ibid.). Within ten
years, J/e has evolved from a mark of alienation to a mark of exuberance.
Members of liminal communities, such as hermaphrodites, transsexuals, drag
queens and drag kings, who do not fit easily into the existing bipartite gender
positions, often use the linguistic gender system to rather different effect from
its traditional function. Drag queens (gay men who wear stereotypically feminine
clothing and use hyper-feminine mannerisms) and drag kings (lesbians who
wear stereotypically masculine clothing and use hyper-masculine mannerisms)
often cross-express, using the pronouns which traditionally refer to the opposite
sex. Thus a drag queen might refer to another drag queen as her and speak
154 Anna Livia
about getting her periods, engaging in a catfight, or putting on her make-up. A
drag king might speak about his butch brothers, getting an erection, or going
home to his wife.
In a study I carried out on the use of linguistic gender by male to female
transsexuals writing in French, I found that although all the authors stated
that they had always felt they were women, in fact they alternated between
masculine and feminine grammatical agreement throughout their autobiographies
(Livia 2000: 168-76). Masculine agreement could indicate variously a
sense of belonging with other males, the gender other people ascribed to them,
or a feeling of power and superiority. Feminine agreement indicated the gender
they felt most comfortable in, isolation and alienation, or a triumphant affirmation.
There was no simple, one-to-one alignment of masculine pronouns
with the rejected gender and feminine pronouns with the desired gender.
When we turn to the descriptions of hermaphrodites in literary texts, we
find that the situation is even more complex. Possessing the sexual organs of
both sexes, hermaphrodites tend to vary in self-presentation far more than the
transsexuals I studied. Feelings of solidarity, isolation, alienation, success, failure,
are all encoded in switches from one gender to another. Indeed, the switch
may be made from one sentence to another with no attempt to naturalize it, or
it may be presented as a positive sign of the fluidity of gender.
4 Gender and Translation
Where the two types of analysis come together (discussion of writing styles, and
discussion of uses of linguistic gender) is in investigations of gender and translation,
a field in which both morphological gender and cultural gender are
highly relevant. Translators work both as interpreters of the original text and,
often, as guides to the culture which produced the text. If the social expectations
of gender in the target culture are very different from those of the source culture,
they need to deal with this anomaly. Similarly, if the languages encode gender
in very different ways, they need to devise a system to encompass the differences.
In their dual role as linguistic interpreters and cultural guides, translators
must decide what to naturalize, what to explain, and what to exoticize.
Studying the role gender plays in translation. Sherry Simon observes that
since as early as the seventeenth century translations themselves have been
seen as belles infideles (beautiful but unfaithful) because, like women, they can
be either beautiful or faithful, but not both (1996: 10-11). Many of the metaphors
for the act or process of translation are highly sexed, and indeed, heterosexed.
One dominant model views translation as a power struggle between author
and translator (both male) over the text (female). In this model, the translator
must wrest the text away from the original author, like a son growing up to
rival his father. George Steiner, himself a prominent translator, describes the
translator as penetrating and capturing the text in a manner very similar to
Linguistic Approaches to Gender 155
erotic possession (1975). Lori Chamberlain, another translation theorist, quotes
Thomas Drant, the sixteenth-century translator of Horace, who claims: "[I have]
done as the people of God were commanded to do with their captive women:
I have shaved off his hair and pared off his nails" (1992: 61-2). For Drant, the
original text must be utterly enslaved and deprived of its foreignness, or, in
his own words, "Englished." In another model, the original author becomes the
translator's mistress whose hidden charms must be revealed and whose blemishes
must be improved. In yet another view, the translator is a submissive,
subjugated, female, alienated, absorbed, ravished, and dispossessed, entirely
taken over by the author (Chamberlain 1992: 57-66). Although the imagined
relationships that prevail among author, text, and translator vary widely, at
the core is the sense that translation is a sexual act.
Given this intense gendering of the process itself, it is hardly surprising that
when it comes to linguistic gender in the original text, the problems posed
are complex and sometimes unanswerable. The novels and poetry of French
Canadian feminist writers such as Nicole Brossard and Louky Bersianik are
characterized by rich alliteration, plays on words, and the creation of portmanteau
words. The title of Brossard's novel L'Amer, for example, is a portmanteau
word containing three others: la mer ("the sea"), la mere ("the mother"),
and amere ("bitter"). Amer is the masculine form of the adjective, while amere
with a grave accent and a terminal -e is the feminine form. In itself amer is a
neologism invented by Brossard. Since the English words sea, mother, and
bitter do not contain the same phonemes as the French words, the neatness of
the alliteration is necessarily lost. The gender play is also lost in English since
the adjective bitter has only one form. Brossard's translator, Barbara Godard,
decided to use a very elaborate graphic representation for the translated title,
composed of three distinct phrases: The Sea Our Mother, Sea (S)mothers, and
(S)our Mothers, all twined around a large S. The English title can therefore
read either These Our Mothers or These Sour Mothers (Simon 1996: 14). This is an
elegant rendition of the original French, but it does not address the practical
problem of how librarians and book catalogues are to refer to the novel.
In my own translation of Lucie Delarue-Mardrus' I'Ange et les Peruers ("The
Angel and the Perverts," Livia 1995), I had to tackle the question of how to
refer to the central character who is a hermaphrodite. Here both linguistic and
cultural gender are at issue. Delarue-Mardrus describes Mario (or Marion, in
her female persona), the main protagonist, as alternately masculine and feminine.
The changes in gender concord in the original French are intended to
produce a sense of shock, requiring the reader to work out how the grammatical
system relates to Mario/n's personality and mental state. The first chapter
introduces us to the young boy and his childhood in a glacial chateau in
Normandy. Here masculine pronouns and concord are used: II avait toujours
e'te seul au monde ("he had always been alone in the world"; Delarue-Mardrus
1930: 19). The second chapter begins in the bedroom of a rich society woman
in an upper-middle-class suburb of Paris. In this section, Marion is described
in the feminine: Elle n'aime rien ni personne ("She loves nothing and no-one";
156 Anna Livia
Delarue-Mardrus 1930: 21). There is no obvious connection between the il of
the first chapter and the elle of the second. Furthermore, both place and social
setting have changed, from Normandy to Paris, and from an old, lonely castle
to a gossipy boudoir. By witholding any explicit link, Delarue-Mardrus forces
readers to make the connection themselves between Mario(n)'s male and female
personae. In this way, they are also implicated in his/her change of gender.
Occasionally, Delarue-Mardrus shocks the reader by referring to Mario/n in
the masculine and then immediately afterwards in the feminine, without providing
any intervening material or a change of context to make this seem more
natural. The River Seine provides a geographical divide between Mario's
bachelor garret and Marion's more luxurious rooms. In one scene we watch
as Mario/n crosses the river and moves from one personality to the other: La
voila chez elle. Le voila chez lui ("She was home. He was home"; Delarue-Mardrus
1930: 38). For a translator the lack of gender concord in English poses a problem.
While the pronouns la and le may easily and effectively be translated as
"she" and "he," their grammatical connection to the expressions chez elle ("at
her house") and chez lui ("at his house") are harder to convey. "There she was
at her house" and "there he was at his house" are more faithful translations
than "she was home," "he was home," and they retain the naturalizing effect
of grammatical necessity. They sound rather stilted in English, however.
In the memoirs of a nineteenth-century hermaphrodite, Herculine Barbin,
recently rediscovered and annotated by Michel Foucault (1980), the narrator's
unusual gender status is conveyed to the reader on the first page. Barbin
begins her self-description in the masculine: soucieux et reveur ("anxious and
dreamy"), but ends in the feminine: fetais froide timide ("I was cold, shy";
Barbin 1978: 9). By this movement from masculine concord in the adjective
soucieux to feminine concord in the adjective froide in the next sentence, Barbin
gets immediately to the crux of the matter. In contrast, in the English translation
it is not until page 58 that reference is made to the grammatical ambiguity
of Herculine's identity: "She took pleasure in using masculine qualifiers for
me, qualifiers which would later suit my official status." The expression
"using masculine qualifiers" is strangely formal, even learned, and stands
out in this plaintive, simply stated autobiography.
5 Implications
We have seen that although many prominent writers have set out to discover the
differences between men's and women's sentences, following in the footsteps
of Virginia Woolf at the beginning of the twentieth century, no convincing
linguistic evidence has yet been provided to indicate the stylistic characteristics
of each. Instead, we have found that there are conventions of masculine and
feminine style which any sophisticated writer, whether male or female, can
follow.
Linguistic Approaches to Gender 157
When we turned to look at linguistic gender, we saw that far from being a
tyrannical system which forces speakers to follow a rigid dualistic structure, it
actually provides means by which speakers may create alternative, oppositional,
or conventional identities. In the realm of science fiction, authors have created
neologistic, non-gendered pronouns to speak of egalitarian Utopias, supplementing
the existing system, which is retained for more traditional worlds.
Authors have experimented with non-gendered protagonists in both the first
and the third person. Although these literary experiments have an effect on
our reading of the novel, it is the lack of pronominal reference, not the lack of
gender markers per se, which causes disturbance.
Finally, in our discussion of the role of the translator and the metaphors used
for the process of translation, we observed that while many different metaphors
exist for the act itself, the dominant metaphors place the translator in a sexual
role in relation to the text and the author. Frequently, when translating from a
language in which there are many linguistic gender markers into a language
which has fewer, either gender information is lost, or it is overstated, overtly
asserted where in the original it is more subtly presupposed.
This research on linguistic approaches to gender in literature demonstrates
the utility for students of gender in society at large to investigate the uses to
which gender may be put in the unspontaneous, carefully planned discourse
of fiction. It reveals not what native speakers naturally do, but what they
are able to understand and the inventions and models that influence their
understanding
