Академический Документы
Профессиональный Документы
Культура Документы
Series Editors
Volume 9
By
Angeliki Alvanoudi
LEIDEN | BOSTON
On the cover: “Communication No. 1”, artwork by Michele Meister. Picture courtesy of Michele Meister.
Alvanoudi, Angeliki.
Grammatical gender in interaction : cultural and cognitive aspects / By Angeliki Alvanoudi.
p. cm. — (Brill’s Studies in Language, Cognition and Culture)
Includes bibliographical references and index.
ISBN 978-90-04-28314-5 (hardback : alk. paper) — ISBN 978-90-04-28315-2 (e-book) 1. Greek language.
Modern—Grammar. 2. Greek language, Modern—Noun. 3. Greek language, Modern—Noun phrase.
4. Greek language—Sex differences. 5. Greek language, Modern—Psychological aspects. I. Title.
PA1057.A48 2014
489’.3045—dc23
2014034959
This publication has been typeset in the multilingual ‘Brill’ typeface. With over 5,100 characters covering
Latin, ipa, Greek, and Cyrillic, this typeface is especially suitable for use in the humanities.
For more information, please see brill.com/brill-typeface.
issn 1879-5412
isbn 978-90-04-28314-5 (hardback)
isbn 978-90-04-28315-2 (e-book)
Acknowledgements ix
Phonetic Symbols x
List of Abbreviations xi
Transcription Conventions xii
1 Introduction 1
1.1 Contextualizing the Problem 1
1.2 On Grammatical Gender and Culture 4
1.3 On Grammatical Gender and Cognition 6
1.4 Summary of the Argument to be Presented in the Book 8
2 Grammatical Gender 13
2.1 Introduction 13
2.2 Noun Classes or Genders 13
2.3 Grammatical Gender in Modern Greek 19
2.4 Grammatical Gender and Reference: Preliminary Remarks 26
2.5 Summary 31
References 177
Index of Authors 195
Index of Subjects 198
Acknowledgements
1 first person
2 second person
3 third person
acc accusative
adv adverb
conj conjunction
cop copula
dat dative
fem feminine
fut future
gen genitive
imperat imperative
imperf imperfect tense
indef indefinite
interj interjection
ipfv imperfective aspect
masc masculine
neg negation
neuter neuter
nom nominative
part participle
particle particle
past past tense
pfv perfective aspect
pl plural
poss possessive
prep preposition
pres present tense
pron pronoun
rational rational
rel relative
sg singular
subj subjunctive
voc vocative
Transcription Conventions
3 Other Markings
Introduction
In the beginning of his book We have never been Modern, Bruno Latour (1993)
wonders whether the ozone layer is an object of study for chemistry, meteorol-
ogy, politics or economics, or whether it constitutes a hybrid, that is, a sort of
cultural-natural network that transgresses disciplinary boundaries and com-
bines elements of knowledge from all different disciplines mentioned above.
Speaking about the ozone layer may be a bizarre way to start a book on linguis-
tics but it is not irrelevant. Latour’s rhetorical question is useful for linguists
working on the relation between language, cognition and culture, because it
opens a window to the intersections that characterize the language-cognition-
culture complex and sheds light on its interdisciplinary nature. Based on
Latour’s approach, the interplay between language, cognition and culture can
be conceptualized as a sort of hybrid that transgresses disciplinary boundar-
ies. For example, it can be explored by linguists, anthropologists or cognitive
scientists, and when it comes to linguistics, it can be examined by linguists
specializing in psycholinguistics, sociolinguistics, cognitive linguistics, prag-
matics or anthropological linguistics. Being a mental phenomenon grounded
in sociocultural practices, language creates a wide range of interconnections
with culture and cognition. In this book, I aim to examine how aspects of these
interconnections manifest themselves in interaction, by focusing on a specific
grammatical feature, gender.
Grammatical gender is a noun class system of two or three distinctions,
which always include the feminine and the masculine (Aikhenvald 2000).
It constitutes an inherent property of the noun, which controls grammati-
cal agreement between the noun and other elements in the noun phrase or
the predicate (Aikhenvald 2000; Corbett 1991; Hellinger and Bussmann 2001).
Nouns may be assigned to specific genders according to semantic, morphologi-
cal and phonological rules. In Modern Greek, the grammatical gender system
includes three distinctions, the masculine, the feminine and the neuter, and
it is inflected in a vast number of linguistic items, such as nouns, adjectives,
participles and pronouns (Pavlidou 2003). In Greek, grammatical gender is
considered to be semantically arbitrary, because gender assignment in nouns
denoting inanimate referents does not follow any straightforward semantic
rules. For instance, nouns denoting physical entities can be feminine (βροχή
than a static essence or a given attribute. As Widerberg (1998, 134) notes, the
English word gender was used primarily in grammatical and literary contexts
and was adopted by American feminists in the 1970s to define sex in a social
sense. Contrary to English, the biological vs. social distinction is not codified
in other languages. For example, in Norwegian, Danish and Swedish the words
kjønn, køn, and kön cover the meaning of both ‘sex’ and ‘gender’ (see Braidotti
2002a for a discussion of the translation of the English term ‘gender’ in vari-
ous European languages). A discussion about the equivalent terms in Greek
is found in Pavlidou (2000). The Greek equivalent term of the English term
‘gender’ is γένος [ʝénos]. This word refers to ‘a general concept in whose exten-
sion specific concepts are contained’, ‘a group of people with common descent’
or ‘ethnic group’, and in grammar contexts it refers to the grammatical category
of gender: γραμματικό γένος [ɣramatikó ʝénos] (Pavlidou 2000, 42). However, the
term γένος is not used for reference to social gender. Sex and social gender are
denoted in Greek via the term φύλο [fílo], which is the equivalent of the English
term ‘sex’. For example, the expressions αρσενικό φύλο [arsenikó fílo] ‘male sex’
and θηλυκό φύλο [θilikó fílo] ‘female sex’ refer to the biological classification of
humans as male or female, while the expression κοινωνικό φύλο [cinonikó fílo]
‘social gender’ refers to the social norms and ideologies associated with this
biological classification.
In this book, I examine the relation between grammatical gender and socio-
cultural gender through the lens of the relation between grammatical gender
and cognition. In the following, I present a ‘map’ of the intellectual trajectory
that has been followed in this book as well as the reasons that motivated my
interest in grammatical gender in the first place. To a very large extent, this
trajectory also forms a politics of location, in Braidotti’s (2002b) terms, that is,
a sort of cartography about the spatio-temporal territory that I share together
with other women and feminist scholars, and about my situatedness in
specific disciplinary contexts. My aim in undertaking this research project is
to study the ways in which language contributes to the construction of socio-
cultural gender and the maintenance of gender inequality. My investigation
started with the following questions. How does grammatical gender contribute
to the construction of sociocultural gender and the reproduction of gender
asymmetry? How is this aspect related to the role of language in mediating
speakers’ interpretation of experience more generally? What sorts of answers
can we give if we examine empirically the use of grammatical gender in inter-
action, that is, in the locus where speakers construct various aspects of their
daily social life? In the next two sections, I sketch out some interesting points
raised in the literature with respect to these questions. These points informed
the argument to be presented in this book.
4 chapter 1
The early writings of feminist linguists in the 1970s and the beginning of the
1980s focus on the role of language as a system in reflecting or mirroring socio-
cultural gender and sustaining gender inequality. I have two scholars in mind
here: Robin Lakoff (1975) and Dale Spender (1980). Both scholars addressed
issues of linguistic sexism in their work. Lakoff (1975) pointed to the negative
representation of women in the lexicon and argued that women experience
inequality in the way in which they are treated by language itself. Spender
(1980) uncovered the role of language in constructing and enforcing a man-
made worldview and sustaining men’s domination and women’s subordina-
tion. For these scholars, language contributes to legitimizing power at the
social level. Therefore, language reform is treated as a crucial element in the
feminist struggle for social change (Spender 1980).
One of the key issues that feminist linguists initially explored was the lexical
and grammatical marking of sex in language. Feminist linguists pointed out
that the semantic distinction of female or male sex can be grammaticized or
be part of the noun’s lexical meaning, and they explored the implications that
these lexico-grammatical features have for the way in which the sociocultural
world is represented and reproduced through language (e.g. Eakins and Eakins
1978; Graham 1975; Miller and Swift [1981] 1988). For example, they focused
on the generic use of the masculine gender, that is, the use of the masculine
gender for reference to female plus male referents or referents whose sex is
unknown. They criticized the generic use of the masculine as a prescriptive
practice, which sustains the social hierarchy between men and women, or the
“Patriarchal Universe of Discourse”, in Penelope’s (1990, xxvi) words.
Lakoff’s (1975) and Spender’s (1980) work inaugurated a long tradition of
research on language and gender which, according to Freed (2003, 701), has
been dominated by three major themes: deficit, dominance, and difference
theory. According to deficit theory, women’s language is ineffective in com-
parison to men’s, and reflects their socially inferior position (e.g. Lakoff 1975).
According to dominance theory, the ways in which women and men use lan-
guage reflect power and inequality at the social level (e.g. Thorne and Henley
1975). Finally, according to difference theory, women and men use different
verbal strategies because they have been socialized in same-sex childhood peer
groups (e.g. Maltz and Borker 1982; Tannen 1990). Freed (2003, 702) argues that
in general these theories have approached gender in essentialist ways. They
presuppose a binary way of thinking about gender, treating men and women
as static categories, while little variation is acknowledged within each category.
introduction 5
However, recently there has been a shift in language and gender research
from “essentialist and dichotomous conceptions of gender to a differentiated,
contextualized and performative model which questions generalized claims
about gender” (Holmes and Meyerhoff 2003, 7). This turn has been informed
by Butler’s ([1990] 1999) theory on gender performativity. Instead of treating
gender as a given attribute that exists prior to language and society, linguists
started examining the practices that continuously produce and maintain gen-
der as a fluid and diverse category, and the “linguistic resources” that speakers
“deploy to present themselves as certain kinds of women or men”, in Eckert and
McConnell-Ginet’s words (2003, 5). As Eckert and McConnell-Ginet (1992, 462)
point out, gender cannot be isolated from other aspects of social identity and
relations, such as age, class or ethnicity. The meaning of gender and the lin-
guistic manifestations of that meaning vary across communities, such as fam-
ily, school or work place, in which speakers may participate in different ways.
Linguists working in the context of this new ‘paradigm’ have shifted their focus
from language system to language use (Pavlidou 2011, 412) and from issues of
representation to issues of construction at the micro-level of interaction.
More specifically, interaction and gender is explored by feminist conversa-
tion analytic studies (e.g. Kitzinger 2000; Speer 2005; Speer and Stokoe 2011).
These studies examine gender as an accomplishment, a process that is con-
structed and maintained through the practices that participants employ in
interaction. Speer and Stokoe (2011, 14) summarize some of the questions that
have been explored by this strand of research as follows. What practices do
speakers employ in order to “make a world of two sexes appear natural and
inevitable”, and ‘pass’ as male or female in interaction? How is gender ‘done’ or
accomplished in interaction? What counts as ‘orientation to gender’ in inter-
action? How can analysts show that gender is interactionally relevant? How
does power manifest itself in interaction? If participants do not orient to their
talk or the talk of others as sexist, can analysts claim that sexism has actually
occurred in interaction? These questions will be highly relevant for the topic
examined in this book.
Overall, grammatical gender in interaction emerges in the space created in-
between the two broad tendencies in language and gender research that were
discussed above. Studying grammatical gender in interaction foregrounds both
language system and language use, and invites scholars to ask how the use of
this grammatical feature in interaction structures the practices that speakers
employ, and what are the social and cognitive aspects of this use. Traces of this
line of thinking can be found in the following extract written by McConnell-
Ginet ([1988] 2011, 39):
6 chapter 1
literature as semiotic relativity (Lucy 1996). Is it possible that “the use of the
semiotic form we call language in and of itself fundamentally alters the vision
of the world held by humans in contrast to other species” (Lucy 1996, 38–39)?
In order to reflect on the symbolic component of language, Lucy (1996, 40)
draws on Peirce’s classification of icons, indexes and symbols. According to
Peirce (1940), icons and indexes are signs related to the objects that they denote
because of natural similarity and physical co-presence respectively. Symbols
are signs standing for their objects because of convention, or “by virtue of a
law [. . .] which operates to cause the Symbol to be interpreted as referring to
that Object” (Peirce 1940, 102). Language is a symbolic medium that maintains
aspects of icons and indexes. It socializes or objectifies individual activities,
including thought, and it is a flexible signaling mode that allows for diversity
to occur at all levels across languages (e.g. morphological, semantic but also
functional), among other things (Lucy 1996, 40).
Due to these semiotic properties of language, a certain kind of linguocen-
trism arises in cultural and cognitive phenomena. As Enfield (2000, 126) points
out, it is difficult “to isolate anything cognitive or cultural which is not already
imbued with language at a profound level”. This view is supported by the fact
that culture involves semiotic processes, which help humans share ideas and
beliefs, and that the basic semiotic tool that humans employ for this purpose
is language. In Enfield’s words,
The book is divided into two main parts. Chapters 2, 3, 4 and 5 constitute the
theoretical part of my research, while chapters 6 and 7 constitute the empirical
part (data analysis and conclusions).
More specifically, in chapter 2, I describe the properties of gender, by focus-
ing on the features of the grammatical gender system in Modern Greek and
the ways in which grammatical gender is related to reference in general. The
key point made in this chapter is that grammatical gender in person reference
codifies the semantic distinction of sex.
In chapter 3, I study the relation between grammatical gender and culture
by drawing on sociolinguistic and feminist non-linguistic approaches.
Sociolinguistic studies on language and gender have shown that grammatical
gender is part of a wide range of linguistic resources available to speakers
for denoting and constructing gender in communication (Hellinger and
Bussmann 2001). Linguistic items marked by grammatical gender codify
biological difference between men and women, ascribe sex to referents
(Eckert and McConnell-Ginet 2003; Ochs 1992), and are used by speakers for
the construction of gender identities (Borba and Ostermann 2007; Hall and
O’Donovan 1996; Kulick 1998; Livia 1997). In addition, gender is constructed
introduction 9
on the basis of social hierarchy, as the generic use of the masculine shows
(Hellinger and Busssmann 2001; 2002; 2003). The relation between language
and gender is also explored by feminist non-linguistic approaches (e.g. Butler
1999; Irigaray 1985a; Wittig 1992). These approaches highlight the role of
language, and of grammatical gender in particular, in categorizing subjects as
women or men and constructing the dominant gender order.
In chapter 4, I explore the relation between grammatical gender and cogni-
tion through the lens of cognitive semantics and research on linguistic rela-
tivity. Lakoff (1987) challenges the idea that grammatical gender constitutes
an arbitrary category with no conceptual dimension, by arguing that gender
relates to conceptual categorization. More specifically, he makes the specula-
tion that genders in Dyirbal, an Australian Aboriginal language, are conceptual
categories with prototypical and less prototypical members. Following Lakoff’s
approach, I suggest that grammatical gender in Greek constitutes conceptual
categorization and that the masculine and feminine gender correspond to
conceptual categories in which men and women are prototypical members
respectively.
Moreover, Lakoff argues that conceptual categories are linked with met-
onymic cognitive models. As Köpcke and Zubin (2003) show, grammatical
gender in German may be related with metonymic cognitive models that
incorporate sociocultural stereotypes. Similar to their speculation, I suggest
that the generic use of the masculine gender is linked to a metonymic cog-
nitive model that incorporates the sociocultural stereotype of man with the
category of human/universal.
The grammaticization of the concept of sex in Greek is taken to have sig-
nificant consequences for speakers’ cognition according to linguistic relativity.
More specifically, as Slobin (1996; 2003) argues, language mediates speakers’
thinking before and while speaking. Language sets limits on what speakers
must say, depending on what is grammaticized, and, thus, guides speakers to
attend to specific aspects of experience when they speak. Because gender is a
grammatical category used in an obligatory and systematic way, it is expected
to guide speakers to attend to the sex distinction when they speak, and inter-
pret referents as female or male.
In chapter 5, I formulate the research question and explore the relation
between grammatical gender and person reference in interaction. In addition,
I present the method to be employed for the empirical investigation of my
research question. Based on the different approaches examined in chapters 3
and 4, grammatical gender is shown to have a complex nature. The cultural
and cognitive aspects of grammatical gender in person reference interweave in
interaction. More specifically, the role of grammatical gender in constructing
10 chapter 1
– in individual (εγώ [eɣó] ‘I’), collective (εμείς [emís] ‘we’) and indefinite or
generic second person (εσύ [esí] ‘you’) self-reference (i.e. reference to
speaker), in reference to recipient (εσύ/εσείς [esís] ‘you.sg/you.pl’), and in
third person reference (αυτός/αυτή [aftós/aftí] ‘he/she’, αυτοί/αυτές [aftí/
aftés] ‘they.masc/they.fem’);
– in cases in which participants orient to sociocultural gender, that is, they
recognize gender as a relevant category for interaction;
– in cases of non-match between grammatical gender and referents’ sex, that
is, in cases where the masculine grammatical gender is used for reference to
female persons only;
– and, in cases in which the use of grammatical gender creates problems in
the interpretation of referent’s sex that are resolved in interaction through
practices of repair.
Overall, the empirical analysis yields indications of the cognitive role of gram-
matical gender i) in cases in which the codification of referent’s gender via
grammatical gender becomes important for what participants do in interac-
tion, and ii) in cases in which the information of referent’s gender is simply
made available in interaction. In the first instance, indications are found in
cases of self-repair and exposed and embedded correction, that is, in repairs
initiated by the speaker and a person other than the speaker respectively, in
which the item marked by grammatical gender constitutes the repairable item.
These repairs are classified as direct indications of the cognitive role of gram-
matical gender, because they involve interventions made by speakers in the
interpretation of referent’s sex. In the second instance, indications are found
in next turns, in turns’ recipient-design, in the non-match between grammati-
cal gender and referents’ sex, and in the membership categorization device.
These indications are indirect or covert, because in the latter cases the infor-
mation of referent’s gender codified by grammatical gender passes unnoticed
by participants.
In addition, the analysis shows that specific meanings are produced in
interaction in routine and covert ways because of the compulsory use of gram-
matical gender in the composition of turns. Linguistic items marked by gram-
matical gender are gender membership categories that categorize speakers,
recipients and third persons as women or men, and invoke and sustain the
stereotypical association of man with the norm. When participants orient to
gender, they employ the information of gender that is made available by gram-
matical gender in interaction to construct their own or other people’s gender
identities. When gender is not interactionally relevant, the gendered catego-
rization of referents and the maintenance of gender hierarchy occur together
12 chapter 1
with various social actions, which are not related to participants’ orientation
to gender.
I end the book with a series of conclusions and reflections in chapter 7. The
main points to be discussed in this final chapter are the following: i) interpret-
ing indications of the cognitive role of grammatical gender as indications of
the role of language in mediating speakers’ thinking for speaking, ii) reflecting
on the relation of the generic use of the masculine with the covert reproduc-
tion of sexism in interaction, iii) and reconsidering the concept of gender per-
formativity through the lens of the present study.
CHAPTER 2
Grammatical Gender
2.1 Introduction
Aikhenvald (2000, 1), classifiers refer to “grammatical means for the linguistic
categorization of nouns and nominals” and include a wide range of noun
categorization devices, such as noun classes or genders, noun classifiers,
numeral, possessed, relational, verbal or deictic classifiers.1 This section
examines noun classes or genders. Noun classes or genders are grammaticalized
agreement systems, which correlate with certain semantic characteristics,
such as sex, animacy or humanness, and they are realized outside the noun
itself (Aikhenvald 2000, 19). The number of noun classes across languages
varies. For example, Babungo (Grassfields Bantu, Benue-Congo) has fourteen
noun classes, while Spanish and French have only two. In many Indo-European
languages, including Greek, there are two or three noun classes, which always
include the feminine and the masculine. These small systems are often
referred as ‘genders’ or ‘grammatical genders’ (Hellinger and Bussmann 2001,
7). For instance, in Spanish, nouns are divided into two genders, masculine and
feminine (Nissen 2002), as shown in example 1, while in German, nouns are
divided into three genders, masculine and feminine, plus neuter (Bussmann
and Hellinger 2003), as shown in example 2.
Although the terms ‘noun class’ and ‘gender’ are used interchangeably in
literature,2 Aikhenvald (2000, 19) suggests that the term ‘noun class’ should
be used as the cover term for noun class and gender, and the term ‘gender’
should be reserved for small systems of two to three distinctions, which always
include the masculine and the feminine. This definition will be followed here.
1 For example, South and American Indian languages and inflecting Indic languages have
special morphemes, which correlate with animacy or shape, and are attached to numerals
(Aikhenvald 2000, 105). These are numeral classifiers. Also, Terêna (South Arawak) has special
morphemes attached to verbs, which characterize the S/O argument in terms of its shape,
size, form and animacy (Aikhenvald 2000, 152). These are verbal classifiers. See Aikhenvald
(2000) for a systematic approach to classifiers from a cross-linguistic perspective.
2 For example, Corbett (1991) uses the term gender as a cover term for all noun class systems.
GRAMMATICAL GENDER 15
Noun class assignment has always a semantic basis, which usually relates to
animacy, humanness, and sex, and sometimes also shape and size (Aikhenvald
2006). In certain languages, noun class assignment is based exclusively on
semantic principles. For example, in Dravidian languages such as Kolami,
Parji and Ollari nouns denoting male humans belong to one class and nouns
denoting female humans and other entities belong to a different class (Corbett
1991, 10). The Australian language Diyari distinguishes nouns denoting female
human/animates from nouns denoting the rest, that is, male human/animates
and all inanimates (Austin 1981, 60 cited in Corbett 1991, 11).
In Dyirbal, a language spoken in North Queensland, Australia, nouns are
categorized in four classes: i) bayi for male referents and non-human animates,
ii) balan for female referents, fire, water and battle, iii) balam for non-protein
food, and iv) bala for the rest (Dixon 1972, 306–312). There are deviations from
these noun class assignment rules based on principles of mythological or
conceptual association or on the important-property principle. For example,
most birds are classified as balan rather than bayi, because birds are associated
mythologically with the spirits of dead women, and the moon is classified as
bayi, because in myths it represents the husband of the sun, which is classified
as balan. In addition, the fishing spear belongs to the bayi class rather than
the bala class, because it is linked conceptually with fish, which belongs to
the bayi class. The light and the stars belong to the balan class rather than the
bala class, because they are linked conceptually with fire, which belongs to
the balan class. While most fishes are classified as bayi, harmful fishes are clas-
sified as balan, because they display the important property of harmfulness,
which is linked conceptually with fire and battle (balan).
3 Each noun is assigned to one specific class. However, occasionally, a noun may belong
to more than one class. These nouns are often called nouns of ‘common gender’ (Corbett
1991, 181).
16 CHAPTeR 2
Manambu, a language from the Ndu family, spoken in the Sepik area of
New Guinea, distinguishes nouns into two classes, masculine and feminine
(Aikhenvald 2012). In general, noun class assignment is based on the size and
shape of referents. For example, adult humans are categorized as masculine
or feminine according to their sex, while children and babies are categorized
as masculine, if they are relatively big or old, or as feminine, if they are rela-
tively small or young. In addition, long, large or both inanimate objects are
categorized as masculine, while small and/or round objects are categorized as
feminine.
In many Indo-European languages such as Danish, Icelandic, Spanish,
Polish, German, Italian, Serbian/Croatian/Bosnian, Portuguese, Swedish,
Greek (see section 2.3), but also in Arab and Hebrew, gender assignment is
semantically motivated in person reference, because of the match between
grammatical gender and referent’s sex. As a number of studies included in the
three volumes edited by Hellinger and Bussmann (2001; 2002; 2003) show, in
languages with grammatical gender nouns denoting female humans are usu-
ally grammatically feminine, while nouns denoting male humans are usually
grammatically masculine. In examples 3 and 4, Spanish and German nouns
codify the sex of the person denoted both lexically and grammatically:
mostly feminine (e.g. Orange, Melone, Birne ‘pear’), wines are masculine (e.g.
Riesling, Elbling, Bordeaux) but soft drinks are feminine (e.g. Limonade, Cola,
Fanta), automobiles are masculine (e.g. Mercedes, Honda, Rover) but motor-
cycles are feminine (Harley Davidson, Kawasaki, Suzuki). Semantic motivation
is also found in the case of nouns ending in -mut. Masculine nouns ending in
-mut (e.g. Mut ‘courage’, Hochmut ‘arrogance’, Unmut ‘displeasure, resentment’)
are associated with the semantic characteristic of extroversion, that is, they
denote types of conduct directed towards controlling the outside world. In
contrast, feminine nouns ending in -mut (Wehmut ‘nostalgia’, Schwermut ‘mel-
ancholy’, Demut ‘humility’) are associated with the semantic characteristic of
introversion, that is, they denote types of conduct that place the self under
outside control.
Besides semantic principles, morphological and phonological principles
may also play an important role in gender assignment. In Russian, nouns
denoting humans are divided into feminine (e.g. mat ‘mother’) and masculine
(e.g. otec ‘father’) according to the semantic distinction of sex (Corbett 1991,
34). In all other cases, nouns are divided into masculine, feminine and neuter
according to morphological principles: nouns of declension 1 are masculine
(e.g. zakon ‘law’), nouns of declensions 2 and 3 are feminine (e.g. škola ‘school’,
kost΄ ‘bone’), and nouns of declension 4 are neuter (e.g. vino ‘wine’) (Corbett
1991, 36). In other languages, gender assignment in nouns denoting inanimate
referents may follow phonological principles. For example, in Katcha, a
Kordofanian language, nouns beginning with m- are feminine, unless they
have a male referent (Heine 1982, 200 cited in Aikhenvald 2000, 25).
Mixed principles of gender assignment can also be found in some
languages. For example, in Yimas, a language from the Sepik area in Papua
New Guinea, noun classes 1–4 are semantically motivated, while classes 5–11
are phonologically motivated (Foley 1986, 86ff.; 1991 cited in Aikhenvald 2000,
26). More specifically, nouns denoting male humans belong to class I, nouns
denoting female humans belong to class ii, nouns denoting animals belong to
class iii, and nouns denoting culturally important plants belong to class iv.
The assignment of the remaining classes is based on phonological principles.
Agreement is a definitional property of noun classes, as I mentioned above.
Noun class agreement can occur in noun phrases between modifiers and heads,
and/or in a clause between a predicate and its arguments. In noun phrases the
head determines noun class agreement with adjectives, participles, articles
or demonstratives. As the following example (found in Marcato and Thüne
2002, 194) illustrates, in Italian the adjective shows gender agreement with
the noun:
18 CHAPTeR 2
In clauses there is usually noun class agreement between the verb and the core
arguments, that is, subject and/or direct object. This is the case, for example, in
some prefixing Australian languages (Dixon 2002, 479–485 cited in Aikhenvald
2000, 34). Noun class agreement is also found with adverbs, though rarely (for
example in Lak, a North-east Caucasian language, Aikhenvald 2000, 34).
The assignment of agreement features is based on semantic or mixed seman-
tic and syntactic principles (Aikhenvald 2000, 29 and 37–39). For instance, in
German noun class agreement is syntactic, that is, based on the grammatical
gender of the noun, as shown in example 6 (found in Bussmann and Hellinger
2003, 146).
All nouns, adjectives, articles and passive participles, and certain pronouns
and numerals are inflected for gender. Gender is inflected in singular and plu-
ral number and in all cases, that is, nominative, genitive, accusative and voca-
tive (see Tables 1, 2 and 3).
In Greek, nouns are assigned to specific genders according to semantic and
morphological principles. More specifically, in person reference gender assign-
ment is semantically motivated because of the match between grammatical
gender and referent’s sex. In general, nouns denoting male human beings are
grammatically masculine and nouns denoting female human beings are gram-
matically feminine. This semantic basis is illustrated with examples 8 and 9.
Singular Plural
Singular Plural
Singular Plural
(8) o φοιτητής
o fititís
the.nom.masc student.nom.masc
‘the male student’
(9) η φοιτήτρια
i fitítria
the.nom.fem student.nom.fem
‘the female student’
GRAMMATICAL GENDER 21
In Greek, the match between grammatical gender and referent’s sex in person
reference is almost perfect, as was shown by a study conducted by Pavlidou,
Alvanoudi and Karafoti (2004) on the nouns included in the Dictionary of
Standard Modern Greek (1998). All nouns were codified according to their
grammatical gender and semantic content, i.e. whether they denote person,
animal, plant, abstract inanimate object, etc., and the sex of referent in per-
son reference, among other parameters. This study showed that the total num-
ber of nouns denoting person in Modern Greek included in the Dictionary
of Standard Modern Greek (1998) is 6225. 59.2% of these nouns are mascu-
line, 33.5% are feminine and 6.5% are neuter. 94.2% of the masculine nouns
denotes male humans, and 94.1% of the feminine nouns denotes female
humans (Pavlidou, Alvanoudi, and Karafoti 2004, 546). Thus, the correlation
between grammatical gender and referent’s sex in person reference is almost
perfect.
In addition, this study showed that 8 feminine nouns denote male humans,
half of which refer to male homosexuals with derogatory meaning, e.g.
λουμπίνα [lubína], πούστρα [pústra] and αδελφή [aðelfí], and only 3 masculine
nouns denote female humans, all carrying sexual connotations, e.g. πειρασμός
[pirazmόs] ‘a woman who causes sexual desire’ and κόμματος [kόmatos] ‘very
beautiful woman’. In addition, 49 neuter nouns denote male humans, e.g. το
αγόρι [to aɣόri] ‘boy’, and 30 neuter nouns denote female humans, e.g. το κορίτσι
[to korít͡si] ‘girl’. The assignment of neuter gender in the latter two nouns is due
to morphological principles, as we will see later on in this section.
In Greek, referent’s sex is marked not only grammatically but also lexically.
Certain nouns carry the semantic property of maleness/femaleness, that is, they
have lexical gender. For example, the nouns άντρας [ándras] ‘man’, θείος [θíos]
‘uncle’ and πατέρας [patéras] ‘father’ encode the semantic property of maleness
as part of their lexical meaning, while the nouns γυναίκα [ʝinéka] ‘woman’,
θεία [θía] ‘aunt’ and μητέρα [mitéra] ‘mother’ carry the semantic property of
femaleness. In these nouns there is correspondence between the masculine
and feminine grammatical gender and the lexical marking of the noun as male
or female. The analysis presented in this book focuses on linguistic items that
are grammatically marked as female or male. However, linguistic items that
are lexically marked as female or male will also be examined in a few points
throughout the text, when their use in interaction is considered to be relevant
with the analysis.
Sex-based gender assignment is also found in nouns denoting animals, e.g.
γάτος [ɣátos] ‘cat.masc’, γάτα [ɣáta] ‘cat.fem’. However, for many animals there
is a single form for both sexes, e.g. αλεπού [alepú] ‘fox.fem’, δελφίνι [ðelfíni]
22 CHAPTeR 2
same form for both male and female referents. These are described as nouns of
‘common gender’ (Holton, Mackridge, and Philippaki-Warburton 1997, 61) and
their gender is distinguished by the article or other modifiers, as examples 10
and 11 show.
The degree of ‘overtness’ of gender depends on case and number. For example,
in singular number in nominative case certain masculine, feminine and neuter
nouns may all end in -os, as shown in example 12.
In plural number in nominative case there are masculine and feminine nouns
which both end in -es or in -i (examples 15 and 16).
Also, in plural number in genitive case there are masculine, feminine and neu-
ter nouns which all end in -on (example 17).
The gender of the noun controls agreement with other elements in the noun
phrase or the predicate. In the noun phrases in examples 18 and 19, the gender
of the head noun controls agreement with the demonstrative, the article and
the adjective:
In examples 20 and 21, agreement is realized outside the noun phrase, with the
gender of the subject noun phrase controlling agreement with the adjective
and the participle in the predicate:
GRAMMATICAL GENDER 25
If a noun phrase consists of two nouns, which denote inanimate objects of the
same kind and are assigned to the same grammatical gender, then syntactic or
semantic agreement is possible in the predicate (example 25, found in Chila-
Markopoulou 2003, 153).
Gender distinctions are found in third person pronouns, e.g. the demonstra-
tives αυτός [aftós] ‘this.masc’, αυτή [aftí] ‘this.fem’, αυτό [aftó] ‘this.neuter’
and εκείνος [ecínos] ‘that.masc’, εκείνη [ecíni] ‘that.fem’, εκείνο [ecíno] ‘that.
neuter’.
Finally, it is worth mentioning that in some nouns grammatical gender is
associated with differences at the semantic and pragmatic level (Anastasiadi-
Symeonidi and Chila-Markopoulou 2003, 16–19). For example, some words
found in two genders may differ in meaning, e.g. μάντρα [mándra] ‘fence.fem’/
μαντρί [mandrí] ‘corral.neuter’, show stylistic variation, e.g αέρας [aéras]/
αγέρι [aʝéri] ‘wind.masc/wind.neuter’, or sociolinguistic variation, e.g. φόνος
[fónos]/φονικό [fonikó] ‘murder.masc/murder.neuter’. Semantic differences
can also be found in some nouns which are assigned to different genders in the
singular and the plural, e.g. δεσμός [ðezmós] ‘bond.masc.sg’, δεσμοί [ðezmí]
‘bonding.masc.pl’/δεσμά [ðezmá] ‘constraint.neuter.pl’.
section with respect to the relation between grammatical gender and refer-
ence in general. Grammatical gender is considered to be one of the means
employed by languages for solving reference problems. Such problems may
involve a) achieving generic reference, b) solving coordination problems, and
c) maintaining reference.
Generic or indefinite reference is reference to female plus male referents
or to referents whose sex is unknown. Languages use different means to deal
with this problem. For example, Zande, a language of the Adamawa-Ubangian
branch of languages spoken in Zaire, Sudan and the Central African Republic,
has a special pronoun, ni, which is distinct from other personal pronouns and
is used if speakers refer to a non-specific or unknown person (Claudi 1985,
95–96 cited in Corbett 1991, 223). In Dyirbal there is no specific preference, as
either bayi or balam may be used for reference to female plus male referents
(Dixon 2010a, 238). In numerous Indo-European languages, including Greek,
the masculine grammatical gender is used for generic reference (Hellinger
and Bussmann 2001; 2002; 2003), while in Arawak (Aikhenvald 1999 cited in
Aikhenvald 2000, 54) and in a few Australian languages (Alpher 1987 cited in
Aikhenvald 2000, 54) the feminine gender is the preferred form for generic ref-
erence. The following example illustrates the generic use of the masculine in
Greek. The noun phrase οι καθηγητές that is marked by the masculine gender is
used for reference to all professors, either female or male.
The use of the masculine or the feminine gender for generic reference is asso-
ciated with markedness relations in noun class systems. According to Dixon
(2010a, 237), the notion of functional markedness in grammar “relates to the
situation of use” of a specific term.4 The marked term is used in restricted
situations, while the unmarked term is employed in all other circumstances.5
As Aikhenvald (2000, 51) notes, a noun class is considered to be functionally
4 Markedness can also be formal (Dixon 2010a, 237). A term in a system with zero realization is
taken to be formally unmarked (e.g. singular number in English).
5 For example, singular number in English is functionally (and also formally) unmarked,
because contrary to plural number, which is used only for reference to two or more referents,
28 CHAPTeR 2
Noun class resolution in Latin is semantic in nouns denoting persons, and syn-
tactic in all other nouns (Aikhenvald 2000, 53). Mixed syntactic and seman-
tic resolution is also found in Greek. If a noun phrase consists of two nouns
belonging to different genders and denoting male and female human beings,
the conflict is resolved in favor of the masculine, as the following example
shows.
singular is used for reference to one referent or “in a general sense when no number specifi-
cation is made” (Dixon 2010a, 237).
6 According to Aikhenvald (2000, 51), the other two criteria for defining a noun class as func-
tionally unmarked are the following: the noun class must be used when the noun class
distinction is neutralized or is of no relevance, and it may be used in default or neutral
agreement.
GRAMMATICAL GENDER 29
For nouns that belong to different genders and denote inanimate objects both
syntactic and semantic resolution is possible (example 31).
2.5 Summary
Based on the discussion so far, we can conclude that the semantic distinction
of female/male sex is grammaticized in Greek. More specifically, the Greek
gender system is shown to have a semantic core in nouns denoting persons,
since the masculine and the feminine grammatical gender correlate with
referent’s male and female sex respectively. Grammatical gender constrains
speakers in specific ways. When speakers use their language, that is, when they
speak or write, they must denote systematically their own sex as well as other
referents’ sex. Given that grammatical gender is a feature inherent to the lan-
guage system that marks a vast number of linguistic items in Greek (cf. section
2.3), the codification of referents’ sex is systematic and compulsory. Speakers
simply do not have another choice; they are enforced by language system to
codify the semantic distinction of sex.
However trivial this may sound, it deserves our attention as linguists for
two reasons. First of all, it is not self-evident that all languages grammaticize
referent’s sex. As I mentioned earlier in this chapter, sex is just one of the various
semantic distinctions (e.g. shape, size) according to which nouns are catego-
rized in different languages. For example, in some Dravidian languages nouns
are divided into human and non-human, while in Siouan and Algonquian
languages nouns are divided into animate and inanimate (Aikhenvald 2000,
276). Moreover, there are languages with no gender systems, such as Finnish
or Turkish. Thus, the grammaticization of sex is an important and special fea-
ture of Greek which foregrounds the issue of language diversity and makes
Greek—as well as other Indo-European languages with grammatical gender—
distinctive. Second, and more importantly, gender is a feature of grammar that
displays a semantic basis. Given that grammatical gender is a morphological
category related to meaning and with a compulsory and systematic use in com-
munication, it should interweave with culture and cognition in specific ways.
In the next two chapters, I explore how this is done, by employing various
approaches within linguistics and beyond.
CHAPTER 3
3.1 Introduction
The relation between grammatical gender and culture has been investigated
primarily by sociolinguistic approaches since the mid 1970s. These approaches
explored the different lexico-grammatical tools that languages provide for
the construction of sociocultural gender. Grammatical gender constitutes
one of these tools. By codifying the semantic distinction of female/male sex
(cf. chapter 2), grammatical gender divides persons according to biological/
anatomical differences, and locates them on a bipolar order in which male/
masculine and female/feminine constitute the two poles respectively.
Moreover, this gender bipolarity is structured on the basis of social hierarchy,
which is associated with men’s dominance and women’s subordination. One
of the basic linguistic practices for maintaining gender hierarchy is the generic
use of the masculine. Aspects of the relation between grammatical gender
and culture have also been addressed in the work of feminist theorists work-
ing within the framework of psychoanalysis and philosophy. These theorists
underlined, among other things, that grammatical gender plays a role in con-
structing gendered subjects and sustaining gender inequality.
In this chapter, I examine the cultural aspects of grammatical gender, by
drawing on both linguistic and non-linguistic approaches. The reason for
selecting such an interdisciplinary perspective lies in the complexities that the
relation between grammatical gender and sociocultural gender foregrounds.
This relation has two sides, a ‘linguistic’ one and a ‘social’ one. These sides
are explored by linguistic and non-linguistic approaches in complementary
ways. In general, feminist theory has deeply influenced and shaped the ways in
which linguists theorize gender and its relation to language. For example, early
feminist linguistic research (e.g. Lakoff 1975; Spender 1980) focused on how the
linguistic representation of women and men is associated with issues of power
and equality, by drawing mainly on the agenda set by feminist empiricism
and feminist standpoint theory1 (Harding 1991). These linguistic approaches
1 Feminist empiricism aims at eliminating gender bias and discrimination against women at
research, teaching, implementation, policy-making and dissemination of data and informa-
tion, by adhering to the principle of scientific objectivity (Braidotti 2003, 199–200). Feminist
standpoint theory criticizes scientific objectivity, uncovers the role of power in knowledge
production, and privileges women’s experience in the formation of new paradigms of knowl-
edge (Braidotti 2003, 200–202).
34 chapter 3
3 Abject subjects are described by Butler (1993, 3) as the “ ‘unlivable’ and ‘uninhabitable’ zones
of social life which are nevertheless densely populated by those who do not enjoy the status
of the subject”.
4 Butler criticizes essentialist approaches to gender by drawing on Nietzche’s position that
“there is no ‘being’ behind doing, effecting, becoming; ‘the doer’ is merely a fiction added to
the deed-the deed is everything” (Butler 1999, 33).
36 chapter 3
Speakers have access to a variety of resources, both material and symbolic, for
constructing gender identities. The basic symbolic resource available to speak-
ers is language system. As Ochs (1992) shows, language relates to gender in two
broad ways. On the one hand, there is a straightforward mapping of language
to gender (one-to-one correlation), in the sense that certain linguistic items
index gender referentially, directly and exclusively (more on this topic in chap-
ter 5). These are linguistic items in which gender constitutes “the actual con-
tent of a linguistic sign”, in Eckert and McConnell-Ginet’s words (2003, 60). A
5 Figurations should not be understood as metaphors but rather as lived maps. As Braidotti
(2002b, 3) argues, “being nomadic, homeless, an exile, a refugee, a Bosnian rape-in-war vic-
tim, an itinerant migrant, an illegal immigrant, is no metaphor. [. . .] These are highly specific
geo-political and historical locations—history tattooed on your body. One may be empow-
ered or beautified by it, but most people are not; some just die of it”.
grammatical gender & intersection of sociocultural gender/sex 37
small number of such items exist in every language community. For example,
in English the third person pronouns he and she, and the address terms Mr,
Mrs and Madam, are grammatically and lexically specified for sex respectively
(Ochs 1992, 338–339). In a similar way, linguistic items marked by grammatical
gender in Greek codify morphologically referent’s sex, and, thus, index gender
referentially, directly and exclusively.
On the other hand, language usually relates to gender in a non-straight-
forward manner, that is, non-referentially, indirectly and non-exclusively. For
example, the use of tag questions in English may index speaker’s feminine
gender, because tag questions are taken to express hesitancy or tentativeness,
which are stereotypically associated with feminine behavior (McElhinny 2003,
35, see also Eckert and McConnell-Ginet 2003, 167–173 for a discussion of stud-
ies on tag questions and gender). In this case, indexing gender is non-exclusive,
because women but also men use tag questions. In addition, indexing gender
via tag questions is constitutive and indirect or mediated, because tag ques-
tions correlate with other social information stereotypically associated with
women. This social information is based on speakers’ knowledge about norms
on gender roles and attributes. Although this book focuses on direct referen-
tial aspects of indexing gender, it is important to keep in mind that direct and
indirect indexing of gender coexist in language use. As Eckert and McConnell-
Ginet (2003) show, speakers may employ a large number of linguistic items or
structures from phonology, morphology, lexicon or syntax to construct gender.
Some of these items and structures may index gender directly, such as items
marked by grammatical gender, and some may index gender indirectly.
Given their relation to indexing gender, linguistic items marked by grammat-
ical gender play an important role in the construction of gender. As Hellinger
and Bussmann (2001, 19) point out, the basic function of grammatical gender
in person reference is “the communication of gendered messages of various
types”. As I show in the next sections, these messages involve the construction
of gender identities and the maintenance of gender hierarchy.
the feminine diminutive noun γυναικάκι [ʝinekáci] ‘small woman’ has negative
meaning, i.e. denoting a woman of no importance.7 Moreover, feminine occu-
pational terms with explicit feminine demotic endings, such as -ινα [ina] and
-ισα [isa], e.g. βουλευτίνα [vuleftína] ‘female parliament member’, have under-
gone a process of semantic derogation (Pavlidou 1985). As Pavlidou (1985, see
also Pavlidou 2003 for an overview of the study in English) shows, the use of
these forms is not preferred in formal contexts. In Greek, women in occupations
are denoted by nouns of common gender stemming from katharevousa, e.g.
o/η φιλόλογος [o/i filόloɣοs] ‘philologist’, ο/η δημοσιογράφος [o/i ðimosioɣráfos]
‘journalist’ (cf. section 2.3), and by masculine nouns with the feminine article,
e.g. η ταμίας [i tamías] ‘cashier’, η βουλευτής [i vuleftís] ‘member of the parlia-
ment’. Triantafyllidis ([1953] 1963 cited in Pavlidou 2003, 186) suggested that
feminine occupational terms could be formed via explicit demotic feminine
suffixes. However, this solution was not adopted broadly. The semantic dero-
gation of these occupational terms seems to be related both to their explicit
feminine marking and the tension between the demotic and katharevousa
varieties (Pavlidou 1985).
The semantic derogation of women also involves the representation of
women as sexual objects or prostitutes. For example, Stanley (1977) shows that
in English the stereotypical woman is a prostitute. More specifically, 220 English
words denote women who are sexually available, e.g. slut, honey pot, but only
22 words denote men in a similar way. In addition, many English words origi-
nally denoting female humans, such as Nymph and Nymphet, acquired sexual
connotations and ended up denoting sexually available women or prostitutes
(Schulz 1975, 85). A similar pattern is also found in Finnish (Engelberg 2002).
In Spanish, many feminine words, such as mujer ‘woman’ and chica ‘girl’, carry
negative sexual connotations (Nissen 2002, 263). Yet, few masculine nouns
carry similar connotations. For instance, the masculine noun tío ‘uncle’ is usu-
ally employed by women to refer to a man as a sexual object. With respect
to Greek, Pavlidou, Alvanoudi and Karafoti (2004) found that 34 of the nouns
included in the Dictionary of Standard Modern Greek (1998) denote unethical
women, e.g. πατσαβούρα [pat͡savúra] ‘old bag’, πορνίδιo [porníðio] ‘little tart’, and
7 In addition, as Pavlidou (2006a, 45) points out, in Greek there are many nouns composed by
a lexical morpheme with negative meaning and a lexical morpheme encoding female sex,
such as βρομοθήλυκο [vromoθíliko] and βρομογύναικο [vromoʝíneko], both meaning ‘dirty slut’.
However, this pattern is not found in composite nouns including a lexical morpheme encod-
ing male sex. An exception to this are composite nouns which consist of a lexical morpheme
encoding both referent’s male sex and age, such as βρωμόγερος [vromόʝeros] ‘dirty old man’.
42 chapter 3
only 15 denote unethical men, e.g. σάτυρος [sátiros] ‘satyr, lustful man’, καριόλης
[karʝόlis] ‘bastard’.
Women can also be represented negatively in less direct ways when speakers
refer to the inanimate world. For example, as Mathiot (1979) shows, American
English speakers often refer to typhoons and broken cars via the feminine pro-
noun she, because emotional instability and weakness are stereotypically asso-
ciated with women. However, they usually refer to cars working properly via
the masculine pronoun he, because power is stereotypically associated with
men. Moreover, male speakers often personify inanimate objects as female,
as the following utterance shows: “She gets me where I want to go” (a trucker
refers to his van). According to Mathiot (1979), this use reflects men’s view of
women as objects. A similar pattern exists in the use of the masculine and
the feminine pronoun in Tasmanian English (Pawley 2002, see also Wierzbicka
2002 for a relevant discussion). Speakers tend to refer to the animate and inan-
imate world via the masculine and the feminine pronoun respectively. The
feminine gender is usually employed for reference to vehicles, e.g. “But when
’e first bought the bulldozer ’e told me he took ’er out in the bush”, storms, e.g.
“She made a mess of that crop”, houses, e.g. She’s certainly got a view, this one”
or abstract concepts as profession, e.g. “I’ve given up the morning job I have
’er away”. These patterns reflect the patriarchal worldview, which identifies
the human with the male and objectifies women (Wierzbicka 2002).
The other medium for marking female difference as a negative category
and establishing men as the norm is the generic use of the masculine. No
matter how ‘innocent’ this practice may look on the surface, it is actually a
highly regulatory practice, which reflects gender hierarchy at the social level
and is based on specific assumptions about how the sociocultural world is
organized (Hellinger and Bussmann 2001, 14–15). These assumptions involve
what Hellinger (2001, 108) describes as “the ideology of MAN (male as norm)”.
According to this ideology, the male/masculine is considered to be the higher
and more prestigious category, while the female/feminine is considered to be
secondary and subordinate.
Several researchers exploring the relation between gender and language
point out that the generic use of the masculine is a political choice, which
establishes a man-made perspective of the world. Following up the discussion
in section 2.4, more examples are given with respect to this rather common
pattern across Indo-European languages. In Italian and German, masculine
nouns may refer to both men and women, e.g. cittadini ‘male and female
citizens’ (Marcato and Thüne 2002, 201), and Jeder Wähler sollte von seinem
Stimmrecht Gebrauch machen ‘Every (masculine) voter (masculine) should
grammatical gender & intersection of sociocultural gender/sex 43
exercise his right to vote’ (Bussmann and Hellinger 2003, 158). In addition, in
Icelandic many masculine nouns denoting occupations can refer to either men
or women, e.g. leikari ‘actor’ and prestur ‘priest’ (Grönberg 2002, 173). This is
also found in French, e.g. Elle est l’auteur de deux romans ‘She is the author
(masculine) of two novels’ (Schafroth 2003, 100–101), and in Polish, e.g. Ona
jest dobrym lekarzem ‘This is a good doctor (masculine)’ (Koniuszaniec and
Blaszkowska 2003, 268). The generic use of items grammatically or lexically
specified as male has also been reported for English8 (e.g. Eakins and Eakins
1978; Graham 1975; Μartyna 1983; Miller and Swift 1988; Nilsen 1977). For exam-
ple, Graham (1975) found that masculine pronouns are used more frequently
than feminine pronouns in English textbooks, and that most of these mascu-
line pronouns refer to male humans.
The generic use of the masculine is also found in Greek, as I mentioned in
section 2.4. A few additional points need to be made here. As Pavlidou (2006a)
points out, the feminine gender may sound unusual or incorrect in certain
usages, because of the generic use of the masculine. This is shown in the fol-
lowing example (found in Pavlidou 2006a, 42).
8 In English, third person pronouns he, she and it are grammatically marked as male, female
and inanimate. However, given that no gender agreement is observed between head nouns
and modifiers, English cannot be said to possess a noun class system (Aikhenvald 2000, 21).
See Curzan (2003) for the history of gender in English.
44 chapter 3
Moreover, a link between the use of the masculine noun άνθρωπος [ánθropos]
‘human being’ and false generics is reported in a study conducted by Makri-
Tsilipakou (1989). The study shows that Greek speakers do not use the noun
άνθρωπος to refer to female humans only.
Interestingly in Greek, male dominance is reflected not only in the way in
which language is used but also in the way in which the language system is
organized. The study of the nouns in the Dictionary of Standard Modern Greek
(Pavlidou, Alvanoudi, and Karafoti 2004) showed that overall—in reference to
human/animates and inanimates—the feminine gender is dominant morpho-
logically, since there are twice as many feminine nouns (45%) as masculine
nouns (23.8%). Yet, the masculine nouns denoting humans (3574) are twice
as many as the corresponding feminine (1949), that is, most nouns denoting
humans are grammatically masculine.
Overall, the generic use of the masculine is considered by feminist linguists
as an exclusionary practice, which makes women less visible, and guides speak-
ers to understand human as male or male as human (Engelberg 2002, 114).
Discourse seems to be ‘colonized’ by the masculine gender, forcing women to
undergo a process of negative ‘othering’ and ‘silencing’ (Spender 1980). In mak-
ing such claims, sociolinguistic studies seem to presuppose or implicate that
grammatical gender has some sort of cognitive depth and, therefore, ‘affects’
or guides the way in which speakers understand the world. This assumption
about the cognitive role of language also seems to underlie or motivate femi-
nist linguistic debates on language reform.
The non-sexist use of language has been treated as a practice for eliminat-
ing linguistic sexism, and, therefore, affecting or facilitating social change. In
Cameron’s words (1985, 1), language constitutes “an essential part of the strug-
gle for liberation”. Feminist initiatives on language reform, or what Pauwels
(2003) defined as ‘feminist linguistic activism’, target linguistic sexism and
explore alternative ways in which women will ‘speak’ about their different
locations and experience and female difference will be represented through
language in positive ways. Reform practices involve various sorts of interven-
tions in language use, which concern the semantic derogation of women and
the generic use of the masculine. Guidelines for non-sexist language use can be
found in Nilsen (1977), Miller and Swift (1988), Kramarae and Treichler (1985),
Frank and Treichler (1989), and Tsokalidou (1996) for the Greek language. With
respect to the generic use of the masculine, two reform strategies have been
proposed. These are summarized by Pavlidou (2006a, 51) as following: a) add-
ing the feminine gender whenever the masculine gender is used for generic
reference, and b) avoiding denoting male or female sex. These strategies are
grammatical gender & intersection of sociocultural gender/sex 45
illustrated with the following examples for Greek (found in Pavlidou 2006a, 51).
Instead of employing the masculine gender only (example 35), generic refer-
ence to Greek people can be achieved either by employing the feminine noun
together with the masculine noun (example 36) or by using a noun which does
not codify the male or female sex of referents (example 37).
As Pavlidou (2006a, 53) highlights, avoiding denoting the female/male sex dis-
tinction in person reference may be a successful strategy for German, which
has a plural participle that makes no distinction between the masculine and
the feminine, e.g. die Lehrenden ‘teachers’. Yet, this strategy is ineffective for
Greek in which most linguistic items inflect for gender (cf. chapter 2). For
intance, in Greek the corresponding participle for die Lehrenden is grammati-
cally marked as male or female in a compulsory way, e.g. οι διδάσκοντες και οι
διδάσκουσες [i ðiðáskodes ce i ðiðáskuses] ‘the male teachers and the female
teachers’ (Pavlidou 2006a, 53).
Besides linguistics, the relation between grammatical gender and culture
has been addressed in the context of feminist non-linguistic theories, more
specifically in the work of Irigaray (1985a; 1985b; 1985c; 1993), Wittig (1992)
and Butler (1993; 1997; 1999). These theories offer a systematic philosophical
and psychoanalytic approach on sociocultural gender and associate the use of
grammatical gender with the construction of gendered subjectivity (Alvanoudi
2008). In the following section, I examine how these theories contribute to our
understanding of the relation between grammatical gender and the gender/
sex intersection.
grammatical gender & intersection of sociocultural gender/sex 47
The feminist theories examined in this section are situated within the broader
epistemological framework of feminist postmodernism9 (briefly discussed
in section 3.1), and more specifically within the framework of French post-
structuralism, which considers postmodernism as a critical mode of thinking.
In poststructuralist theory, language is taken to be a medium of constructing
meaning and social reality. Subjectivity is understood as a discursively pro-
duced process, which encompasses conscious and unconscious dimensions,
and is embodied (Braidotti 2003). Poststructuralist thinking criticizes the sub-
ject of the western philosophical tradition as a ‘disembodied’ abstract rational
individual and moves beyond in theorizing the self as a locus of contradictions,
diversity and change. The category WOMAN is challenged as a unified category
and emphasis is given on the differences within the category itself, taking class,
age, race, ethnicity and sexual orientation into account. These differences are
asserted in positive ways and are linked with newly formed empowered sub-
jects (Braidotti 2003).
In this section, I examine three feminist theorists working in the realm of
this epistemological paradigm. More specifically, I focus on the insights they
provide on the cultural aspects of grammatical gender, by taking the general
context of their work into account.
Luce Irigaray is considered to be the basic philosopher of sexual difference
theory. She argues that the patriarchal symbolic economy10 is based on the
production of subordinate others versus the male as the One and the Same
universal subject (Irigaray 1985a). This economy is reproduced via the domi-
nant (Lacanian) psychoanalytic paradigm, which tends to ignore women.
By establishing hierarchical dichotomies such as phallic/non-phallic, penis/
vagina-clitoris, more/less, psychoanalysis theorizes female sexual difference as
a pejorative other. For Irigaray, the issue of representing female difference and
subverting the idea of woman as a negative category is directly linked with
language. Language constructs sexed subjects by categorizing them as female
or male, and locates them in asymmetric positions: man is the universal sub-
ject and woman is the negative subordinate other or “what is left of a mirror
invested by the (masculine) ‘subject’ to reflect himself, to copy himself”, in
Irigaray’s (1985b, 30) words.
For Irigaray (1993) grammatical gender is one of the basic tools that sus-
tain the patriarchal symbolic economy. For example, Irigaray argues that the
generic use of the masculine and the negative evaluation of grammatically
feminine words in French affect the ways in which women experience them-
selves as negative others. For instance, she points to the semantic derogation
of the feminine noun doctoresse and the asymmetries in the semantic pairs un
ordinateur (‘computer’)/la machine à écrire (‘typewriter’) or un avion (‘plane’)/
une voiture (‘car’), in which masculine nouns denote objects of higher value
compared with the objects denoted by feminine nouns. Irigaray treats lan-
guage as a key notion for women’s emancipation. Women need to speak ‘as
women’ and use language in creative ways in order to represent their differ-
ence in positive ways, and, thus, make themselves visible.11
The role of language in the construction of gender is also central in Monique
Wittig’s (1992) work. Wittig criticizes biological determinism for naturalizing
and legitimizing social inequality between men and women, and deconstructs
the category of sex. According to this theorist, sex should not be understood
as a given and natural category, but rather as a political category. In her words,
“there is but sex that is oppressed and sex that oppresses” (Wittig 1992, 2). Sex
divides subjects according to biological difference. At the same time it cate-
gorizes subjects as legitimate or illegitimate, and locates them to hierarchal
11 Irigaray (1985b) suggests that women should employ the strategy of mimesis, that is, re-
appropriate old meanings and start producing new radical ones. She describes mimesis
as a woman’s effort “to recover the place of her exploitation by discourse, without allow-
ing herself to be simply reduced to it” (Irigaray 1985b, 76). Mimesis involves the use of
innovative practices for the expression of fluid pluralistic identities. For instance, Irigaray
(1985b) employs the metaphor the two lips in order to conceptualize a new feminine
imaginary.
grammatical gender & intersection of sociocultural gender/sex 49
positions in the symbolic order. For Wittig, the process of women becoming-
subjects constitutes a cognitive process of rethinking and re-evaluating the
social world. This process is linked with language.
By drawing on a Marxist framework of analysis, Wittig argues that language
belongs to the superstructure, which reproduces the dominant ideology. She
suggests that language shapes speakers’ understanding of the world, by enforc-
ing a specific view on what is ‘real’ and legitimate, as is shown in the following
extract:
Language casts sheaves of reality upon the social body, stamping it and
violently shaping it . . . for there is a plasticity of the real to language: lan-
guage has a plastic action upon the real. (Wittig 1992, 78)
perform what is being said), and he associates their performative force with
certain felicity conditions, which are related to specific contexts and conven-
tions. For example, the utterance I now pronounce you husband and wife is
infelicitous, if it is not uttered by a person designated to do so in an appropri-
ate context, e.g. by a priest at a church. Following Austin, Butler suggests that
words do not simply describe subjects. Words categorize subjects and locate
them to specific social positions. In this way, language interpellates subjects
into social existence in the context of a ritual, material and pre-existing struc-
ture, in Althusser’s (1984) terms. For example, when a baby is born and the
doctor says It’s a girl! a process of ‘girling’ is inaugarated through which
the baby is categorized as female and located in the dominant gender order. The
use of a linguistic item lexically marked for female sex produces the illusion of
an essentialist identity, that is, the effect of female sex. I suggest that this func-
tion can also be extended to all linguistic items marked by grammatical gender.
These items categorize speakers and others according to their sex and sustain
the bipolar gender order. In this sense, the use of grammatical gender consti-
tutes a performative practice, which produces the materiality of sexed bodies.
It is worth mentioning that Butler does not accept a deterministic view on
the role of language in the construction of the social world. By building on
Derrida’s idea that linguistic signs are always subject to re-appropriation, reit-
eration and re-signification, she argues that words can be used in unforeseen
ways and acquire new meanings. This process of re-signification can enable
subjects to form new kinds of political agency.12
3.5 Summary
If we take both linguistic and non-linguistic approaches into account, the cul-
tural aspects of grammatical gender can be defined as following. Grammatical
gender ascribes sex to referents and locates them in asymmetrical positions in
the bipolar gender order. It naturalizes biological difference, which supports
social hierarchy and inequality between women and men. It constructs gender
12 A typical case of how re-signification can become a tool for political action is the way in
which the word queer has been re-appropriated and re-signified by the lgbt movement.
Originally, the word queer denoted the strange, unusual, sexually deviant, that is, it had
negative meanings. However, it acquired positive meaning, when it was re-appropriated
by lgbt activists to denote a critical stance towards the norm of compulsory heterosexu-
ality. Moreover, the concept queer has been used by Butler as a tool for conceptualizating
gender in non-essentialist ways (cf. section 3.2).
grammatical gender & intersection of sociocultural gender/sex 51
identities and reproduces sexism. The latter is mainly achieved through the
generic use of the masculine. In a few words, grammatical gender constructs
sociocultural gender on the basis of social hierarchy.
This social dimension of grammatical gender seems to be interconnected
with a cognitive dimension. The latter dimension is often implied or presup-
posed by sociolinguistic approaches, which associate the generic use of the
masculine with women’s exclusion from speakers’ thinking of the world. The
idea that grammatical gender plays an important role in cognition is also
reflected in numerous initiatives taken by feminist linguists and other theo-
rists with respect to language reform, which aim at increasing women’s vis-
ibility in the social/symbolic sphere. But is there actually a relation between
grammatical gender and speakers’ cognition? This question is examined in the
next chapter.
CHAPTER 4
4.1 Introduction
1 This classification does not cover the rich complexity that is observed within each strand.
2 Contrary to prototype theory, the classical approach (Taylor 2003) views categories as sets of
common features with clear-cut boundaries. For instance, man is defined as ‘male’, ‘human’,
‘adult’, while woman is defined as ‘female’, ‘human’ and ‘adult’.
54 chapter 4
Social stereotypes are typical cases of metonymy in which one member of the
category is used as a vehicle for understanding the category as a whole. For
example, in patriarchal cultures housewife mothers tend to be taken as the
best examples of what mothers should be like. This stereotype defines which
members are considered to be the most representative and less representative
ones. Housewife mothers are prototypical members, while working mothers
are less central members (Lakoff 1987, 79–82).3
Lakoff (1987) investigates the relation between noun classes and conceptual
categorization by using the noun class system in Dyirbal as one of his exam-
ples. As we saw in section 2.2, Dyirbal has four noun classes organized on the
basis of semantic principles: i) bayi for male referents and non-human ani-
mates, ii) balan for female referents, fire, water and battle, iii) balam for non-
protein food (except for meat), and iv) bala for the rest. Lakoff (1987, 91–104)
suggests that these noun classes constitute conceptual categories with more
representative/prototypical and less representative/peripheral members. For
example, male and female humans are prototypical members in the categories
3 For this reason the category mother is said to have a radial structure (Lakoff 1987, 83–84).
grammatical gender and speakers ’ cognition 55
bayi and balan respectively, while the moon and the stars are peripheral mem-
bers in the corresponding categories.
Although Lakoff examines a noun class system in which nouns are assigned
to specific classes according to semantic principles (cf. section 2.2), I suggest
that his approach may apply to grammatical gender systems as well. Although
in these systems gender assignment is mostly based on morphological or
phonological principles, there is a semantic basis when it comes to gen-
der assignment in nouns with human referents. As we saw in section 2.2, in
many languages with grammatical gender the categorization of nouns denot-
ing inanimate objects is semantically arbitrary to a large extent. However, in
person reference it is semantically motivated because of the match between
grammatical gender and referent’s sex. In addition, as section 4.3 shows, the
gender/sex match seems to be so important for speakers’ cognition that its
‘effect’ is not limited to speakers’ thinking of the human/animate world as
female or male but can also be extended to speakers’ thinking of the inanimate
world as ‘female’ or ‘male’. Following Lakoff’s approach, I suggest that in Greek
the concept of sex is grammaticized. Similar to bayi or balan in Dyirbal, the
masculine and feminine grammatical gender categories in Greek constitute
conceptual categories with prototypical and peripheral members. Male and
female humans are prototypical members in the masculine and feminine gen-
der categories respectively. In contrast, sun (masc) or lighter (masc), and rain
(fem) or door (fem), are peripheral members in the corresponding categories.
Additionally, Lakoff’s approach is useful in our understanding of the cogni-
tive role of grammatical gender in the following way. Köpcke and Zubin (2003)
suggest that there is a relation between grammatical gender and metonymy in
language use, by drawing on Lakoff’s idea about cognitive metonymic models.
These scholars argue that grammatically neuter linguistic items referring to
female persons are associated with a metonymic model, which embodies
the sociocultural stereotype of sexual innocence, social naiveté and depen-
dent social status. This mechanism is activated by the mismatch between the
neuter grammatical gender and referent’s female sex. When grammatically
feminine and neuter linguistic items are used in the same context to refer to
a female person, feminine gender is associated with a different metonymic
model that embodies the stereotype of sexual experience, social maturity and
sophistication.
Köpcke and Zubin (2003) examine the use of grammatical gender in refer-
ence to female humans in Karl Weggerl’s story Legende von den drei Pfändern
der Liebe. In this story, a young village pot-maker travels away from home to sell
his wares. His wife who is a young girl gives him three pledges of love, a hair
56 chapter 4
ribbon, a ring and a knife. During his travel a mysterious woman visits him at
night and sleeps with him. Every morning he gives her one of the love pledges.
On his way back home he finds the ribbon, then the ring and finally the girl
with the knife in her breast, and he realizes that his wife was the mysterious
woman that visited him at night. Köpcke and Zubin point to two archetypes
on which this story is based: the archetype of the sexually innocent village
girl and the archetype of the experienced independent and amoral woman of
the city. The former is evoked by the use of grammatically neuter nouns and
pronouns in anaphora, e.g. Mädchen and es. The latter is evoked by the use of
grammatically feminine nouns and pronouns, e.g. Frau and sie. For instance,
in the first part of the story neuter items are used to refer to the girl, and in the
second part of the story feminine items are used to refer to the woman. Yet,
in the third part of the story in which the hero realizes that the girl and the
woman are the same person, reference to the same female person is achieved
via feminine and neuter items. This is shown in the following extract: “when
he entered the main room, there lay his girl (sein Mädchen) on the bier. And
he knew that she (sie) was the one (es) whom (die) he had loved three times,
and three times betrayed and now there was his knife thrust in the middle of
her (ihrer) white breast” (Köpcke and Zubin 2003, 149). Neuter linguistic items
evoke metonymically the archetype of the innocent village girl character, while
feminine linguistic items evoke metonymically the archetype of the mysteri-
ous amoral woman character.4 The referent is the same but each time she is
associated with properties of different stereotypes.
Following Köpcke and Zubin’s argument, I suggest that metonymy may be
linked with other uses of grammatical gender as well. More specifically, meton-
ymy can help us understand how the generic use of the masculine operates. As
I mentioned in section 3.3.2, the generic use of the masculine depends on the
stereotypical assumption that man is the universal category or the norm. This
sociocultural stereotype relates to the cognitive dimension of the masculine
grammatical gender in two ways. First, this stereotype structures the category
human, by defining which members are taken to be prototypical in this cat-
egory and which members are taken to be less representative. Given speakers’
knowledge about gender hierarchy, it is reasonable to assume that male
humans are prototypes or cognitive-reference points for the category human,
while female humans are peripheral members. Second, this stereotype can be
said to form a metonymic cognitive model on which the generic use of the
masculine is based. This model consists of the following elements:
4 The pronoun die refers to the female person non-metonymically (Köpcke and Zubin 2003).
grammatical gender and speakers ’ cognition 57
In this metonymic model, male sex operates as the vehicle through which
the human/universal (target) is accessed and understood. This metonymic
model is associated with the use of the masculine gender in certain contexts,
as shown in example 40. Suppose that an employee at a museum produces the
following utterance:
The speaker refers to all students and teachers of the primary school, that is,
males plus females, who took part in the school trip. In order to do that, she
employs grammatically masculine nouns, which control agreement with the
modifiers, i.e. articles. In this case, generic reference is achieved metonymi-
cally. Male sex, which is morphologically codified, activates metonymically
the category of human/universal and guides speakers to interpret referents as
not exclusively male. Therefore, the category of human/universal is a sort of
58 chapter 4
However, there may be cases in language use in which the masculine gender is
used for indefinite reference but the metonymic model of male sex as human/
universal is not necessarily activated. This is illustrated with example 42.
The noun phrase marked by the masculine gender refers to all taxi drivers. Yet,
these referents are more likely to be interpreted as male only, rather than as
male plus female, given speakers’ knowledge about the job of driving a taxi as
stereotypically masculine.
Overall, based on the approaches mentioned in this section, the following
hypotheses can be made. Sex is a concept that gets grammaticized in Greek
and the use of the masculine grammatical gender is linked with metonymy.
The grammaticization of the concept of sex is expected to have significant con-
sequences for speakers’ cognition according to research on linguistic relativity.
The latter research shows that grammatical categories play a significant role in
mediating speakers’ thinking. This topic is examined in the next section.
5 The roots of linguistic relativity can also be traced in German romanticism and Humboldt’s
philosophy, as well as structuralist paradigms, more specifically Saussure’s theory of language
as a system of signs and Durkheim’s sociology (Gumperz and Levinson 1996b, 4–5).
60 chapter 4
For example, when English speakers produce the utterance Τhe man is sick,
they codify number, tense and definiteness. However, in Siouan speakers must
codify whether the man is standing or moving, in Kwakiutl they must codify
whether the man is visible or not by the speaker and whether he is close to
speaker, hearer or third person. Thus, languages set limits on what speakers
must say and on which particular aspects of experience must be verbalized.
According to Boas, the concepts that get grammaticized in a language tend
to be conceptually salient for its speakers. Moreover, the interpretations of
experience that are associated with specific grammatical categories are uncon-
scious because grammatical categories are automatic, obligatory and system-
atic. In a similar vein, Sapir (1970) argues that language constitutes a ‘guide’ to
social reality which channels the way in which speakers think about the world.
In Sapir’s words:
Building on the ideas of Boas and Sapir, Whorf (1956) describes the role of lan-
guage in guiding speakers’ thinking of the world in the following way:
We dissect nature along lines laid down by our native languages. The cat-
egories and types that we isolate from the world of phenomena we do not
find there because they stare each observer in the face; on the contrary,
the world is presented in a kaleidoscopic flux of impression which has
to be organized by our minds—and this means by the linguistic systems
of our minds. We cut nature up, organize it into concepts, and ascribe
significances as we do, largely because we are parties to an agreement
to organize it in this way—an agreement that holds throughout our
speech community and is codified in the patterns of our language. The
agreement is, of course, an implicit and unstated one, but its terms are
absolutely obligatory; we cannot talk at all except by subscribing to the
organization and classification of data which the agreement decrees.
(Whorf 1956, 213–214)
grammatical gender and speakers ’ cognition 61
formulation of this idea). These patterns of use may involve differences in the
ways in which language is used by subgroups in the same language community
or in contexts of speaking (e.g. Gumperz 1996). Although the perspective taken
by discursive relativity is an interesting one—especially for scholars working
on sociolinguistics—it is not relevant to the question examined in this chap-
ter. The question about the cognitive role of grammatical gender foregrounds
the idea of structural relativity, because it specifically involves the relation of a
feature of grammar to speakers’ cognition.
Issues of structural relativity are examined by a number of “psychologically
informed” (Sidnell and Enfield 2012, 302), or Neo-Whorfian, studies, situated
mainly within the context of psycholinguistics. Some of these studies can be
found in the volumes edited by Gumperz and Levinson (1996a), Niemeier and
Dirven (2000), Pütz and Verspoor (2000), Bowerman and Levinson (2001), and
Gentner and Goldin-Meadow (2003). These studies treat linguistic relativity as
a hypothesis that can be empirically investigated via experimental methods.6
Within the framework of structural relativity Slobin (1996; 2003) formu-
lated the hypothesis that language mediates speakers’ cognition at least at the
time of speaking. Slobin replaced the terms language and thought with the
terms speaking and thinking respectively, and he argued that these phenom-
ena should be conceptualized as mental processes that occur when speakers
formulate utterances, rather than as abstract entities or static phenomena.
When speakers design and produce utterances, they choose to codify particu-
lar aspects of experience depending on what is grammaticized in the language
they speak. Thus, the mental representations constructed at the time of speak-
ing differ across languages. Slobin defines the sort of thinking that emerges
while speaking as ‘thinking for speaking’. In his words:
6 This hypothetico-deductive appproach to linguistic relativity (see Hill and Mannheim 1992
for the term) has been criticized by Hill and Mannheim (1992), Lee (1996) and Enfield (2002)
as deviating from the original Whorfian position. According to this position, linguistic rela-
tivity is a principle or an ‘axiom’, in Foley’s (1997, 192) terms, that is, a sort of background
assumption for addressing questions and conducting research.
grammatical gender and speakers ’ cognition 63
Similar to Boas, Slobin argues that concepts that get grammaticized in lan-
guages are conceptually salient for the speakers of these languages. For exam-
ple, these concepts are used frequently in speech and are acquired by speakers
at an early stage. In addition, speakers are shown to have easy lexical access
to these concepts. According to ‘thinking for speaking’ hypothesis, speakers’
thinking before and at the time of speaking will be attuned to the concepts
that get grammaticized in the language that they speak. The experiments
conducted by Slobin and Berman with English, German, Spanish and Hebrew
speakers (Berman and Slobin 1994 cited in Slobin 1996, 93) provide indica-
tions of the mediation of language structure to speakers’ thinking. Slobin and
Berman examined time and space expressions in speech. Adults and 3–5 years
old children speaking the abovementioned languages were asked to describe
the events presented in the picture book Frog, Where Are You? (Mayer 1969
cited in Slobin 1996, 72). These scholars found that speakers of different lan-
guages describe the same events in different ways depending on how each lan-
guage codifies the concept of time and space. For example, English codifies
the durative/non durative distinction in the verb and, thus, guides speakers to
attend to the duration of an event or to the lack of it in a compulsory way. In
contrast, Hebrew does not grammaticize this distinction and, thus, it does not
guide speakers to attend to this specific feature of experience when they think
for speaking.7
The structural diversity observed across languages provides analysts
with a rich source for the possible different ways in which languages can
mediate speakers’ thinking for speaking. For example, Turkish is a language
that grammaticalizes evidentiality (Aikhenvald 2004). This means that when
Turkish speakers refer to some event that took place in the past, they must
codify whether they witnessed the event they refer to or whether they heard
it from someone else. Therefore, language is expected to guide them to attend
to this particular feature of experience, when they speak. On the contrary,
speaking Greek does not involve the same mental activity, given that this
7 In his later work, Slobin (2003) argues that language mediates speakers’ thinking not only
at the time of speaking but also at the time of hearing, reading and looking, and it involves
mental processes such as imagination and memory.
64 chapter 4
8 According to Levinson (2003), differences among languages may be related with differences
in speakers’ experiencing for speaking. Given that experience will be verbalized at some
point, it is possible that speakers codify experience non-linguistically in ways that are appro-
priate for its verbal expression. Therefore, speakers of different languages are expected to
differ not only in their thinking for speaking, but also in their experiencing for speaking.
9 For example, a weak version of linguistic relativity seems to be supported by Levelt (1989),
whose work informed Slobin’s approach to thinking for speaking. According to Levelt (1989),
speech production is divided into four systems: i) the production of preverbal messages
which consist of the conceptual information that needs to be expressed for realizing the
speakers’ intention (Conceptualizer), ii) the grammatical and phonological codification of
preverbal messages (Formulator), iii) the production of external speech (Articulator), and
iv) the understanding and control of speech (Speech-Comprehension System). Differences
among languages can be found in Formulators, which differ across languages and require
speakers to codify different conceptual distinctions.
grammatical gender and speakers ’ cognition 65
The theoretical hypotheses that I formulated in the previous section are sup-
ported by a number of empirical studies from sociolinguistics and psycholin-
guistics to be discussed in this section. Although these studies have different
aims, are based on different theoretical backgrounds and employ different
methodological tools, they all show a correlation between grammatical gender
and speakers’ cognition.
10 As Bickel (2000, 185) insightfully points out, “sociocultural practices, whether verbal or
nonverbal, [. . .] sustain the cognitive style and bias of awareness that is required by a
particular grammar”.
66 chapter 4
ascribed male sex to nouns. The only exception to this pattern was the noun
κοινωνικός λειτουργός [cinonikόs liturɣόs] ‘social worker’, which was interpreted
by most speakers as female. According to Pavlidou (1985), speakers’ interpre-
tations were associated with the generic use of the masculine, which guides
speakers to the interpretation of referent as male, and with speakers’ knowl-
edge about stereotypically feminine and masculine occupations. The generic
use of the masculine noun άνθρωπος ‘human being’ was examined by Makri-
Tsilipakou (1989) (cf. section 3.3.2). In this study, speakers were given a ques-
tionnaire with sentences consisting of the noun άνθρωπος with no contextual
clues about referent’s sex. They were asked to give a name to the person
denoted by the noun. Makri-Tsilipakou found that speakers tended to interpret
referents as male. A more recent study conducted by Pavlidou and Alvanoudi
(2013) also indicates that the feminine and the masculine grammatical gen-
der in Greek correlate with the interpretation of person as female and male
respectively. In a sex attribution task, Greek speakers were asked to attribute
male or female proper names to human beings depicted in pictures. Speakers
assigned male or female sex to human beings according to the grammatical
gender of the noun denoting these items. Therefore, an almost perfect match
between grammatical gender and referent’s sex was found for Greek. The find-
ings of this study are presented in detail later in this section.
The studies discussed above indicate that there is a correlation between
grammatical gender and the interpretation of referent’s sex. They show that
the generic use of the masculine is linked with speakers’ strong tendency to
interpret referent(s) as male. They also uncover the role that context plays
in these interpretations. For example, if the noun denotes a stereotypically
masculine activity, it is likely that speakers will interpret referent(s) as male.
Additionally, these studies show that the use of feminine gender ensures that
reference to women is achieved.
Interestingly, a male bias in the interpretation of referent’s sex is also reported
for languages with no gender system at all, such as Finnish (Engelberg 2002)
and Turkish (Braun 2001), or in languages with limited gender distinctions,
such as English (MacKay and Fulkerson 1979). For example, a study conducted
by Braun (2001) showed that Turkish speakers interpret the sex of referents
denoted by gender-indefinite nouns, based on their knowledge about stereo-
typical masculine and feminine occupations. For instance, the nouns sekreter
‘secretary’ and kuyumcu ‘goldseller’ denote stereotypically feminine and mas-
culine occupations respectively. Speakers tend to interpret sekreter as referring
to a female person and kuyumcu as referring to a male person. Therefore, even
in a language with no gender system, gender-related associations still exist and
“remain hidden on a deeper semantic level” (Hellinger and Bussmann 2001, 11).
68 chapter 4
11 Another group of studies (Clarke et al. 1981; Ervin 1962; Flaherty 2001; Konishi 1993; Μills
1986) examines whether masculine and feminine nouns carry connotations of masculin-
grammatical gender and speakers ’ cognition 69
ity and femininity respectively, by employing semantic differential tests. In these tests,
speakers are asked to rate the meanings of nouns on bipolar adjective scales that include,
for example, factors of potency (e.g. strong/weak) or evaluation (good/bad), and are asso-
ciated with ‘masculinity’ and ‘femininity’ (see Pavlidou and Alvanoudi 2013, 110 for a sum-
mary of the criticism that has been articulated with respect to these tests).
12 Sex attribution to inanimate objects is one of the different questions investigated by these
studies. For example, Boroditsky, Schmidt and Phillips (2003) examined whether gram-
matical gender ‘affects’ speakers’ memory of word-name pairs, assessment of similarity
between persons and objects, and description of objects. Contrary to the previous stud-
ies that were conducted in speakers’ native language, the tasks employed by Boroditsky,
Schmidt and Phillips were conducted in the same non-native genderless language, i.e.
English, to ensure that speakers understand stimuli and instructions in the same way.
70 chapter 4
4.4 Summary
How does person reference relate to the use of grammatical gender? Do lin-
guistic items marked by grammatical gender merely identify individuals or do
they display more complex functions, because they codify referent’s sex in a
compulsory and systematic way? In this section, I examine these questions
and I show that the interrelation between the social and cognitive dimensions
of grammatical gender manifests itself in interaction through speakers’ covert
assumptions about referent’s sex as an aspect of social context.
grammatical gender, culture and cognition in interaction 75
When speakers refer to person, time, place or object, they use linguistic forms
that will enable hearers to identify the specific person, time, place or object
that speakers intend to be identified by hearers. According to Enfield (2012,
433), “when we say that a speaker makes reference to something in interaction,
we mean that the speaker establishes or maintains a communicative focus on
some entity, usually in order to say something about it”. Speakers have access to
rich lexical resources for making reference. For example, in Greek speakers can
accomplish person reference by employing various lexical means. They can
use proper names, e.g. Αγγελική [anɟelikí] ‘Angeliki.fem’, definite noun phrases,
e.g. ο γιατρός [o ʝatrós] ‘the.masc doctor.masc’, indefinite noun phrases, e.g.
μία γυναίκα [mía ʝinéka] ‘a.fem woman.fem’, demonstratives, e.g. εκείνη [ecíni]
‘that.fem’, vocatives, e.g. κυρία [ciría] ‘lady/Madam.fem’, or first, second and
third person pronouns and verbs, which point to the speaker, to the addressee
or to a person who is neither speaker or addressee in the speech event, e.g.
εγώ [eɣó] ‘I’, εσύ [esí] ‘you’, αυτός/αυτή [aftós/aftí] ‘he/she’, and είμαι [íme] ‘cop.
pres.1sg’, είσαι [íse] ‘cop.pres.2sg’, είναι [íne] ‘cop.pres.3sg’.
The selection of a referential formulation among various alternative ones
in interaction is a contextually bound and cognitively rich phenomenon. As
Hanks (2007) argues, person reference involves a practice of individuating a
referent under a perspective, which presupposes and activates background
knowledge, and should be theorized as a sort of construal. In his words:
The other waiter is able to identify the referent introduced by the definite noun
phrase το έξι, because in restaurants waiters usually refer to customers via the
number of the table where the latter sit.
As Enfield (2012, 434) notes, the selection of a referential formulation is
based on “factors specific to the speech event, including who the speaker is,
who s/he takes the addressee to be, what the relationship is between the two,
and what the speaker’s communicative purpose is—that is, the social action s/
he wants to produce”. For instance, imagine that I ask a woman sitting next to
me on a bus to help me with the time (example 45):
With this utterance I request some bit of information. Because I do not know
the woman, I use the address form κυρία, which encodes social distance
between me—the speaker—and the addressee. Plus, because the request is a
face-threatening act (Brown and Levinson 1987), I employ this specific address
term as a politeness strategy, together with indirecteness and second person
grammatical gender, culture and cognition in interaction 77
plural verbs, which also index social distance between speaker and addressee.
Thus, my selection of the specific referential formulation is shaped by the
social relationship I hold with the third person to be addressed and the com-
municative purpose I have.
This address term does not simply individuate and identify a referent. It also
encodes information about the referent. More specifically, it indexes the social
relation between speaker and referent, and because it is a grammatically femi-
nine noun, it indexes referent’s feminine gender (cf. chapter 3). This aspect of
person reference in interaction will be the focus of this section.
The complex phenomenon of person reference in interaction is examined
by a number of studies on person reference in interaction included in the vol-
ume edited by Enfield and Stivers in 2007. More specifically, the studies con-
ducted by Brown (2007), Enfield (2007), Hanks (2007) and Haviland (2007)
show that reference to third person in interaction is functionally complex,
because it combines, in Hanks’ words (2007, 154), “the referential effect of pick-
ing up the individual” with “the indexical effect” of displaying certain social
information about the individual. In choosing some referential formulation
speakers do more things than just identifying a referent.
For example, Lao speakers employ pragmatically unmarked, or default, and
pragmatically marked formulations for initial third person reference in interac-
tion (Enfield 2007). Default formulations consist of the person’s first name plus
a prefix that appropriately denotes the referent’s social position relative to the
speaker. Marked formulations consist of the name without the prefix or with a
prefix that does not conform to social relations. Default formulations involve
minimum cognitive effort from the part of the speaker and the addressee(s),
and are designed to be interpreted with no special attention given to the man-
ner of formulation selected. In contrast, marked formulations are departures
from the default, which show speaker’s special effort to accomplish some-
thing more than merely identifying the referent. These formulations invite the
addressee to interpret the ‘special’ thing that the speaker tries to accomplish.
However, as Enfield (2007) shows, even default formulations in Lao interaction
always do more than just referring to person. They encode a person’s hierarchi-
cal position relative to others in Lao society, and, thus, “make publicly overt
and thereby instantiate and stabilize cultural values about persons and their
social relations” (Enfield 2007, 119). In a conversation, two male Lao speak-
ers use the proper name tia together with the prefix bak2 to refer to a man
(Enfield 2007, 105). The prefix in this case is a social deictic element indexing
familiarity and lack of respect. The information of familiarity is not relevant
for the interactional business in which participants are engaged. However, this
78 chapter 5
1 Referential and non-referential indexes may also be related to indexical creativity and
presupposition respectively (Silverstein 1976). For example, personal pronouns I and you
(i.e. referential indexes) presuppose a speaker and hearer at context and they also create
these roles as aspects of context, while the mother-in-law vocabulary in Dyirbal (i.e. non-
referential indexes) presupposes the existence of relatives in context.
2 Linguistic items marked by grammatical gender appear to be similar to social deictic items,
due to the fact that in both cases a bit of ‘social’ information about the context is grammati-
cized. However, these items are not identical. According to Levinson (1983, 127–129), social
deixis is related with conventional implicature, i.e. non-truth-conditional inference that is
conventionally associated with certain lexical items or linguistic constructions. In contrast,
referential indexes are truth-conditional. For example, if I replace second person singular
with second person plural in the following sentence Είσαι ψηλός/Είστε ψηλός [íse psilós/íste
psilós] ‘You are.sg tall.masc/You are.pl tall.masc’, truth conditions remain intact, but the
inference about the social relation between speakers and addressee changes. In contrast,
replacing the masculine adjective ψηλός with the feminine adjective ψηλή affects truth condi-
tions, because the referent changes (a person can be either female or male).
grammatical gender, culture and cognition in interaction 81
In example 47, the noun phrase with the masculine head noun and the mas-
culine modifier identifies a third person and at the same time presupposes
masculine gender as part of referent’s identity, that is, it combines referential
effect with indexical effect.
5.2.2 The Generic Use of the Masculine Gender and Default Inferences
Person reference in everyday interaction in English and also in other languages
is organized on the basis of two principles or preferences: recognition and
minimization (Sacks and Schegloff 1979; Stivers, Enfield, and Levinson 2007).
According to the principle of recognition, speakers use a recognitional
reference form that the recipient will know and can use to identify a person.
According to the principle of minimization, speakers use a single form, whether
a name, a description, etc. to refer to person. Levinson (2007) argues that the
principle of minimization is part of the more general principle of economy,
according to which speakers minimize the expressive means in referring to
person. For example, speakers use a single referring expression (minimization)
or a name rather than a description. The generic use of the masculine seems
to align with the principle of economy in person reference in interaction. In
using the masculine forms alone for achieving reference to males plus females,
82 chapter 5
speakers minimize their expressive means and invite their recipients to make
the necessary inference of human/universal. Therefore, the masculine gender
seems to be related with the interpretative strategies that participants employ
in interaction.
As Enfield (2007) shows, there is a relation between interpretative strate-
gies and person reference in Lao interaction. For example, default referential
formulations are designed to pass unnoticed by participants. They also allow
participants to recognize cases in which speakers deviate from the norm and
use pragmatically marked formulations, trying to express additional meanings
besides referring to a specific person. Inferences or conversational implica-
tures play an important role in understanding speakers’ intentions in commu-
nication, according to Grice (1989).
Levinson (2000) explores the role of inferences in communication by focus-
ing on generalized conversational implicatures, that is, implicatures whose
interpretation does not depend on context. More specifically, Levinson argues
that generalized conversational implicatures are associated with default infer-
ences or preferred interpretations in communication. These preferred inter-
pretations include various phenomena in communication, such as preference
structure in interaction, presuppositions, and illocutionary force. According to
Levinson, generalized conversational implicatures are produced on the basis
of three neo-Gricean principles: Quantity (do not say less than is required),
Informativeness (do not say more than is required), and Manner (do not use
a marked expression without reason). The principle of Informativeness is of
special interest here. This principle consists of the maxim of minimization
(speaker’s maxim) and the rule of enrichment (recipient’s corollary) (Levinson
2000, 114). The maxim of minimization enjoins the speaker to say as little as
necessary, that is, produce the minimal linguistic information sufficient to
achieve her/his communicational ends. The rule of enrichment dictates the
recipient to infer that what is generally said is stereotypically and specifically
exemplified, that is, to amplify the informational content of the speaker’s utter-
ance and interpret it according to stereotypical assumptions. For example, the
utterance ‘John said “Hello” to the secretary and then he smiled’ produces
the generalized conversational implicature ‘John said “Hello” to the female sec-
retary and then John smiled’, based on the knowledge that interlocutors share
about the occupation of secretary as stereotypically feminine (Levinson 2000,
117). In other words, the semantically general expression secretary implicates
the semantically specific interpretation ‘female secretary’.
Levinson’s approach sheds light on the way in which the generic use of the
masculine operates in communication. The generic use of the masculine seems
to be related with the systematic pragmatic inferences that interlocutors make
grammatical gender, culture and cognition in interaction 83
3 Commonsense knowledge serves as an ‘umbrella’ category which includes norms for under-
standing behavior, language use and knowledge about the sociocultural world (e.g. social
institutions and relations) that is taken for granted (Garfinkel 1967). This concept resembles
the concepts of common ground proposed by Clark (1996), and social cognition proposed by
van Dijk (2009).
grammatical gender, culture and cognition in interaction 85
4 Sidnell and Enfield (2012) formulate a new version of linguistic relativity, according to which
lexico-grammatical differences among languages correlate with differences in the kinds of
social actions that participants can achieve in social interaction.
5 Schegloff (2006a) criticizes the dominant approach within cognitive science as focusing on
individual mind and separating cognitive processes from interaction. He argues that the
study of language and cognition should be based on empirical observations about interac-
tion, which constitutes the locus of human sociality.
88 chapter 5
gender order and are inference rich. I suggest that these inferences include
speakers’ covert assumptions about referent’s gender, given that items marked
by grammatical gender operate as referential indexes (cf. section 5.2.1).
Inferences also include speakers’ knowledge about the stereotypic association
of man with the norm, and gender hierarchy (cf. section 5.2.2). Defining items
marked by grammatical gender as gender membership categories foregrounds
the contextually grounded nature of these items in interaction. It associates
grammatical gender with practices that participants employ in interaction in
order to accomplish a particular aspect of social life, gender.
As a number of conversation analytic studies (e.g. Kitzinger 2007; Stockill
and Kitzinger 2007)6 show, the deployment of gender membership categories
in English conversation (e.g. woman, man, male, girl) is not necessarily related
to participants’ orientation to gender. Stockill and Kitzinger (2007, 231) argue
that although gender membership categories are gendered linguistically, they
are not always “gendered interactionally”, that is, they are not always used “pri-
marily to foreground gender”. Yet, they make gender available in talk. However,
as Kitzinger (2000; 2005) argues, various aspects of social life are often repro-
duced in interaction in tacit ways, without participants explicitly orienting to
them. For example, the norm of heterosexuality is reproduced as an unques-
tioned assumption in English conversation when participants accomplish
social actions, which are not related to gender (Kitzinger 2005). The gender
membership categories husband/wife denote specific roles in the heterosexual
kin system, locate referents in the heterosexual order and, thus, make the infer-
ence of heterosexuality available in interaction. Moreover, as Kitzinger (2005)
points out, participants’ lack of orientation to the norm of heterosexuality is a
sort of ‘required’ condition ensuring that this norm is reproduced smoothly as
part of the taken-for-granted world. In this way, heterosexuality becomes part
of the tacit presuppositions of everyday life that participants ‘see’ but do not
notice, in Garfinkel’s (1967) terms.
In general, ca favors a mode of analysis that does not impose a-priori
social categories on the data and is based on aspects of interaction to which
participants explicitly orient. According to Schegloff (1997), the analysis of
conversational episodes should examine only those categories that are shown
to be relevant for participants, that is, procedurally consequential for what
participants do in interaction. Yet, this analytic choice, that Enfield (2007, 113)
describes as the ‘Members-Only Filter’, limits the scope of analysis. Consider,
6 Kitzinger (2000) refers to conversation analytic studies on gender via the term ‘feminist con-
versation analysis’. Speer and Stokoe (2011) provide a review of the issues examined by these
studies.
grammatical gender, culture and cognition in interaction 91
for example, person reference in interaction (cf. section 5.2). As Enfield (2007)
shows, default formulations encode information about the social relation
between speaker and third person. Regardless of whether participants
recognize these relations as relevant for what they do in interaction or not,
these relations are made available in interaction and constitute part of the
routine meanings produced in interaction. Following up these arguments,
one should expect that a lot of interesting things may be happening when
“nothing special is happening” in interaction, in Kitzinger’s (2005, 259) words.
Given that power relations are usually reproduced in interaction in tacit and
hidden ways, Kitzinger (2005) suggests that analysts should treat participants’
everyday world as problematic and explore the assumptions on which this
world is based. In her words, we as analysts might ask what is happening:
when the second is in a preferred relation to the first; when the yes—no
question is followed by a yes—no answer; when the recognitional refer-
ent is recognized or the nonrecognitional referent is treated as adequate;
[. . .] when presumed ordinary experiences are treated as ordinary—what
is happening then, how is that done, and what kind of world must we be
living in that these things run off smoothly? This kind of analysis may
enable us better to understand (as activists and as conversation analysts)
the routine achievement of a taken-for-granted world (Schegloff 1986).
(Kitzinger 2005, 259)
Taking the studies discussed in this section into account, in the next chapter,
I examine empirically grammatical gender in interaction both when gender is
relevant interactionally and when it is not.
In this chapter, I discussed the theoretical and methodological tools that
will help me analyze empirically grammatical gender in interaction: indexical-
ity in relation to gender membership category. To put it briefly, linguistic items
marked by grammatical gender are gender membership categories, which are
linked with covert assumptions that participants share about referent’s gender.
In the next chapter, I apply this notion to the analysis of naturally occurring
data in Greek conversation.
CHAPTER 6
6.1 Introduction
As was mentioned in section 5.3, turns consist of at least one tcu, which con-
stitutes at least one recognizable action, such as question, answer, disagree-
ment, offer, request, invitation, announcement, promise or narration. The
action done by some talk is “grounded in its position, not just its composition”
(Schegloff 2007a, 20). A speaker beginning to talk in a turn has the right and
obligation to produce one tcu, and as a speaker approaches the possible com-
pletion of a first tcu in a turn, transition to next speaker can become relevant
(Schegloff 2007a, 4). Current speaker selects next speaker by addressing her/
him “with a turn whose action requires a responsive action next—for example,
with a question that makes an answer relevant next” (Schegloff 2007a, 4). Or,
next speaker self-selects to take the next turn. First pair parts make specific
second pair parts relevant. Second pair parts are divived into preferred and
dispreferred, depending on whether they align with or distance from the first
94 chapter 6
1 Word selection is one of the basic problems of talk in interaction that get to be examined via
ca. According to Schegloff (2006b, 71, 73, 77 and 82), the other problems of talk in interaction
are the following: i) the “turn-taking” problem, which concerns who and when should talk or
act next, and how this affects the construction and understanding of the turns or acts them-
selves, ii) the “sequence-organizational” problem, which concerns how successive turns or
actions are formed up to be ‘coherent’ with the prior one and constitute a “course of action”,
and the nature of that coherence, iii) the “trouble” problem, which concerns how partici-
pants deal with trouble in speaking, hearing, or understanding the talk or other conduct such
that the interaction does not freeze in place, intersubjectivity is maintained or restored, and
that the turn, sequence and activity can progress to possible completion, and iv) the “overall
structural-organizational” problem, which concerns how an occasion of interaction is struc-
tured, and how placement in the overall structure informs the construction and understand-
ing of the talk and other conduct as turns, as sequences of actions, etc.
Empirical Investigation of Grammatical Gender in Interaction 95
The first question to examine empirically is how the use of grammatical gender
contributes to the construction of sociocultural gender in interaction. What
are the implications of the compulsory use of items marked by grammatical
gender in the composition of turns for the sociocultural world that partici-
pants jointly construct in interaction? In the next section, I start exploring this
question by analyzing grammatical gender in self-reference in interaction.
2 Extracts 1, 8, 13, 14 and 27 were analyzed for the first time in a talk that I gave in the Symposium
Gender and the Greek Language at Freie Universität Berlin, February 2, 2012. The written ver-
sion of the analysis can be found in the special issue on gender and the Greek language in the
journal Gender and Language (Alvanoudi forthcoming).
3 For a systematic analysis of pronoun systems across languages and their semantics see
Mühlhäusler and Harré (1990).
96 chapter 6
Katerina rejects Vaso’s invitation at lines 13 and 14. She refers to herself via
the personal pronoun εγώ ([eɣó] ‘I’) and the first person singular verbs νιώθω
([ɲóθo] ‘feel’) and δε μπορώ να πάρω ([ðe boró na páro] ‘cannot move’). The
speaker uses the grammatically feminine participle κουρασμένη ([kurazméni]
‘tired’) to describe herself and give a non-fault ability account (Heritage 1988)
for her rejection. The feminine participle is a gender membership category
that categorizes the speaker as woman. However, the speaker is not using this
category “for its gendered properties or to invoke gendered attributes”, to quote
Kitzinger (2007, 44). Instead, the speaker employs the feminine participle as a
tool for rejecting Vaso’s invitation. In particular, Katerina’s turn is a dispreferred
second pair part that distances from the first pair part that her co-participant
Empirical Investigation of Grammatical Gender in Interaction 97
In a multi-unit turn Dimitris refers to himself via the first person singular verbs
and pronouns έχω αρχίσει να:: χάνω ([éxo arçísi na:: xáno] ‘I have started losing’),
να νυστάζω ([na nistázo] ‘falling asleep’), γράφω ([ɣráfo] ‘I write’), καθαρογράφω
([kaθaroɣráfo] ‘I write up’), δίνω ([ðíno] ‘I give’), εγώ ([eɣó] ‘I’), μου ([mu] ‘my’).
At lines 2 and 15, he uses the masculine adjectives ξύπνιος ([ksípɲos] ‘awake’)
and σίγουρος ([síɣuros] ‘sure’) to describe himself. These adjectives are gen-
der membership categories, which categorize the speaker as man. Similar
to Extract 1, in this conversational episode, sociocultural gender is not fore-
grounded interactionally. The speaker tells a story in which he is the main
protagonist. He took exams and he failed. In delivering the story, he describes
himself. He woke up early in the morning and went to the examination room,
being self-confident. Although he was sure that he had succeeded in passing
the exams, he actually failed. The speaker selects and deploys the masculine
adjectives together with other elements to deliver his story. At the same time,
these adjectives index speaker’s masculine gender and, thus, presuppose it as
an aspect of his identity.
In both Extracts 1 and 2, gender is not relevant interactionally but it is still
made available in interaction because it is morphologically codified. The avail-
ability of gender in interaction is a phenomenon that deserves our attention
as analysts in spite of the fact that participants show no demonstrable orienta-
tion to it (Schegloff 1997; 2007b). Sociocultural gender is constructed in rou-
tine ways via the compulsory use of grammatical gender in the composition
of turns given the role of grammatical gender in indexing sociocultural gender
referentially. Items marked by grammatical gender constitute obligatory gen-
der membership categories that ascribe female or male sex to speakers, cat-
egorize them as women or men, and are associated with participants’ covert
assumptions or presuppositions about the speaker as woman or man. When
participants accomplish social actions such as rejecting an invitation or telling
a story, they assume feminine or masculine gender as part of their identity and
they construct gender as a given aspect of their sociocultural world.
The relation between grammatical gender and collective self-reference is
examined in the next three extracts.4 According to Lerner and Kitzinger (2007,
526–527), these collectivities can be ‘organizational’, e.g. when one speaks on
behalf of an enterprise, ‘relational’, e.g. when one speaks on behalf of a couple,
or ‘circumstantial’, e.g. when one speaks on behalf of persons who happened
to be together somewhere.
In Extract 3, at lines 10 and 13, participants refer to a circumstantial collec-
tivity that includes the speaker, Katerina, and her co-participant, Vaso. In the
4 See Pavlidou (2008; 2012a) for the study of collective self-reference in Greek interaction.
Empirical Investigation of Grammatical Gender in Interaction 99
lines preceding this extract, Vaso, Yana and Katerina refer to the holidays that
Yana had in the summer after her graduation.
At line 1, Polikseni addresses a question to Vasilis and Natasa, and selects them
as next speakers via the address term παιδιά ([peðʝá] ‘hey guys’). The second
pair part made relevant by this first pair part, i.e. the answer, is built in a col-
laborative manner (Lerner 2004). Vasilis delivers an answer at lines 2 and 3,
and employs the first person plural verbs πάμε ([páme] ‘we cope’) and ήμασταν
([ímastan] ‘we were’) to refer to himself collectively. He speaks on behalf of
a relational collectivity that includes himself and his wife, Natasa. At line 4,
Natasa continues the second pair part that Vasilis started and she uses the
masculine participle αποκλεισμέ:νοι ([apoklizmé:ni] ‘blocked’) to describe the
collectivity. The second pair part is brought to completion by Vasilis at lines
5 and 6. Vasilis employs the same masculine participle that Natasa employed
to describe the same relational collectivity. In this conversational episode,
nothing special is happening with respect to gender. However, inferences
are produced in automatic and systematic ways, because participants empoy
the masculine participle for reference to male plus female persons. The male
categorization of referents imposed by the masculine participle is a tool that
participants employ to activate the inference of human/universal and achieve
generic reference. Thus, in this conversational episode, participants tacitly pre-
suppose the stereotypic association of man with the norm as an unquestioned
assumption about the way the social world is organized.
In Extract 5, Dimitris, Tania, Zoi and Melita discuss the relation between
religion and sexism. Tania and Melita disagree on whether Christianity is a
more sexist religion than Islam. They refer to themselves collectively, as mem-
bers of a religion, that is, they refer to an organizational collectivity.
Tania asserts that Christianity is a sexist religion at lines 6–7, and Melita
argues that Islam is even more sexist at lines 10–11. At line 17, Μelita asserts
that Christianity, contrary to Islam, is a religion that allows its believers to
be freer. She refers to herself and other Christians via the first person plural
είμαστε ([ímaste] ‘we are’). Tania re-refers to the same organizational collectiv-
ity at line 21 via the first person plural verb ακολουθούσαμε ([akoluθúsame] ‘we
followed’). At lines 24, 25 and 26, Melita produces a turn composed by two
tcus. In the first tcu >Μ δεν^ γκζέρω.< ([m ðen gzéro] ‘Mm I don’t know.’)
Melita disagrees with Tania. In the second tcu εγώ εδώ στην εστία που: σας λέω
συναναστρέφομαι με άλλες θρησκείες, (.) βλέπω ότι είμαστε πολύ ελεύθεροι. ([eɣó eðó
stin estía pu: sas léo sinanastréfome me áles θriscíes, (.) vlépo óti ímaste polí
eléfθeri] ‘I am telling you, here in the dorm where I mix with other religions, I
see that we are very free.masc.’) she gives an account for her disagreement, by
drawing on her personal experience. Melita keeps reference to the same orga-
nizational collectivity, i.e. all Christians, by employing the first person plural
verbs είμαστε (‘we are’), and describes the collectivity via the masculine adjec-
tive ελεύθεροι (‘free’). She differentiates the collectivity of Christians from the
collectivity of Muslims, and she selects the masculine adjective as a tool for
accomplishing this differentiation: us, Christians, are different (very free) than
the others, Muslims. In addition, the speaker employs the masculine adjec-
tive as a vehicle for achieving generic reference. Therefore, the inference of
human/universal via male sex is part of the routine meanings tacitly produced
in this conversational episode.
104 chapter 6
In her recounting of the trip to Paris, Zoi refers to all non-French tourists
including herself and her partner via the second person singular verbs and
pronouns είσαι ([íse] ‘you are’), βλέπεις ([vlépis] ‘you see’), θα ’λεγες ([θá leʝes]
‘you would say’), βλέπει:ς ([vlépi:s] ‘you see’), α:γχώνεσαι ([a:ŋxónese] ‘you stress
out’), σε ([se] ‘you’), θε:ς ([θe:s] ‘you want’), (ούτε) μπορείς να εμπλακείς ([úte borís
na eblacís] ‘you cannot get involved’), να πεις ([na pís] ‘you say’). At line 15, she
characterizes the abstract group of referents as foreigners and categorizes the
106 chapter 6
group as male by using the masculine noun ξένος ([ksénos] ‘foreigner’) together
with the masculine indefinite article ένα:ς ([éna:s] ‘one’). At line 16, she refers to
herself and her partner via the first person plural δεν^ μπάθαμε ([ðen báθame]
‘we didn’t hurt ourselves’ meaning ‘nothing happened to us’). In this way, she
shows that she understands herself as part of the abstract group of persons
that she referred to in her previous tcu. The masculine items that the speaker
selects to describe the abstract group of persons are part of the tools that the
speaker employs for the delivery of the story at that particular moment. At the
same time, these items constitute gender membership categories that catego-
rize referents as male and, therefore, allow generic reference to be achieved
via the inference of human/universal. Similar to Extracts 4 and 5, in Extract 6,
gender is not foregrounded interactionally. Yet, the stereotypical association of
man with the norm is tacitly presupposed.
At line 1, Dimitris asks Tania how many credits she got, Tania delivers an
answer at line 3, and co-participants produce a set of minimal post-expansions
at lines 4, 5, 6, 7, 8 and 9. Post-expansions are sequences that occur after the
base second pair part (Schegloff 2007a, 115–168). Minimal post-expansions or
sequence-closing thirds are designed to move for sequence closing. In their
minimal post-expansions, participants refer to Tania via the second person
singular verbs and pronouns χτύπησες ([xtípises] ‘you succeeded in entering’),
εσύ είσαι and εσύ ’σαι ([esí íse/esí se] ‘you are’), and assess her via the feminine
adjectives καλή:: ([kalí::] ‘good’) and ικανή ([ikaní] ‘competent’). Participants
employ these adjectives to attribute specific features to the recipient and
evaluate her competence in achieving a high score in the national exams.
Additionally, the feminine adjectives are gender membership categories that
ascribe female sex to recipient and categorize her as woman. The deployment
of these gender membership categories is not related to participants’ orienta-
tion to gender. However, the recipient’s feminine identity is indexed via the
feminine grammatical gender in a compulsory way and presupposed as part of
the commonsense knowledge that participants share.
In Extract 8, Dimitris reconsiders the choices that he would make regarding
his professional career. At lines 5 and 6, Melita and Tania ask Dimitris what he
would do if he was about to finish school now, and initiate a sequence that is
brought to completion by Tania’s minimal post-expansion at line 17.
At line 11, Dimitris replies that he would go to the army and choose a private
school. At line 13, Evagelia asks Dimitris which private school he would choose
and Dimitris delivers an answer to the question at line 15. At line 17, Tania
produces a minimal post-expansion, assessing Dimitris’s choice as nice. She
uses the grammatically masculine adjective ωραίος:: ([oréos::] ‘nice’) to accom-
plish the action of assessment. Yet, besides assessing the recipient in a positive
way, the speaker also categorizes him as man. In Extracts 7 and 8, the gendered
categorization of recipients occurs in the service of the social action of assess-
ment. Speakers assess aspects of recipients’ activities or conduct, which are
not related to their gender identity, for example their exam-score or the profes-
sion they would choose. In both cases, grammatically feminine and masculine
adjectives are employed as vehicles for accomplishing the action of assess-
ment. Yet, in both cases, these items categorize recipients as women and men
respectively. Even though participants do not orient explicitly to gender, they
presuppose it tacitly as part of the taken-for-granted world.
In Extract 9, four female participants, Natasa, Anna, Marina and Elsa,
describe the decorative items that Marina and Chrysanthi have in their
apartment.
Extract 9 [Ι.20.Α.29.3]
1 Νatasa [↑Αυτά από πού τα παίρνετε εδώ] τα ωραία?=
Where do you buy these nice things over here?
2 Anna =>(. .[ . . . . . . . )<]
Empirical Investigation of Grammatical Gender in Interaction 109
At lines 5–6, Μelita asks her co-participants, Tania, Zoi and Dimitris, if they
had a similar experience, and selects them as next speakers via the second
person plural verb and pronoun είστε ([íste] ‘you are’) and σας ([sas] ‘you’).
She employs the masculine adjective σί:γουροι ([sí:ɣuri] ‘sure’) to describe
her recipients. The speaker uses the masculine grammatical gender to refer
to all co-participants, two females and one male, and, thus, accomplish the
social action that the turn is designed to do, i.e. a question addressed to all co-
participants. Note, however, that more things are happening in interaction.
By using the masculine gender for generic reference, the speaker presupposes
Empirical Investigation of Grammatical Gender in Interaction 111
the stereotype of man as the norm and invites her recipients to make the infer-
ence of human/universal. This inference is part of the routine, unremarkable
meanings produced in interaction.
In the conversational episodes examined in this section, sociocultural gen-
der is constructed in tacit and systematic ways due to the compulsory use of
grammatical gender in the composition of turns. In addition, the social hier-
archy between women and men is a feature of the sociocultural world that
participants presuppose when they use the masculine gender for generic
reference.
6 Ζoi =[ Ναι::. ]
Yes.
7 Μelita =[Εσύ τι γρά]μμα είσαι
In which class were you
8 Dimitris =Α:XA,=
A:ha,
9 Tania → =H[: που πήγε κλα]σικό.=
the.nom.fem pron.rel go.3sg.past classic.acc.neuter
The.fem [one] who studied classic literature.
10 Dimitris → [η (προ)- η Άννα?]
((lighter sound))
the.nom.fem the.nom.fem Anna.nom.fem
The.fem (pro)- Anna.fem?
11 Zoi → =Ναι ρε. αυ[τή. ]
yes PARTICLE this.nom.fem
Hey yes. that one.fem.
12 Dimitris → [Η Ά](ννα)? =
the.nom.fem Anna.nom.fem
Anna.fem?
13 Zoi =[>(Όχι όχι.)<]
(No no.)
14 Tania → =[ Η::]:: [αχ::. ]
the.nom.fem interj
The.fem ah::.
15 Ζoi [Που ήταν πολύ] φυτό:.
Who was a real nerd.
16 (.)
17 Τania → Αυτή πρέπει να είχε πάρει και::- να είχε περάσει στους
she.nom must subj have.3sg.past take.pfv and subj
have.3sg.past pass.pfv in.the.acc.masc.pl
She must have taken and- must have been one of the
18 δέκα πρώτους >ξέρω γω< κάτι τέτοιο, όταν: [μπήκαμε.]
first ten [students] when we entered (the school), something
like that so to speak.
19 Μelita [ Υποτρο]φία:?=
[With a]
scholarship?
20 Τania → =Υποτροφία. μπρά°βο. .h (.) πώς τη λένε [ ρε? ]
scholarship.fem interj adv she.acc call.3pl.pres PARTICLE
[With a] scholarship. right. .h (.) hey what’s her name?
Empirical Investigation of Grammatical Gender in Interaction 113
In a question to her co-participants at lines 1 and 2, Zoi refers to a girl that used
to be in the same class as theirs. This question is a possible first pair part in
a pre-sequence, that is, in a sequence occuring before the base first pair part
(Schegloff 2007a, 28–57), which aims to ensure that co-participants know and
are able to identify the third person that Zoi refers to. Zoi employs the femi-
nine indefinite pronoun μί:α ([mí:a] ‘one’) to introduce the female third per-
son. She describes the referent via the prepositional phrases <με γυαλιά, μ’ αγορέ
μαλλιά ([me ʝaʎá, maɣoré maʎá] ‘with glasses, with boyish hair’) and the rela-
tive clause που μιλούθε °λίγο έτσι:¿ ([pu milúθe °líɣo ét͡si:¿] ‘who was speaking
like this’). The gap after the first pair part shows possible trouble encountered
by co-participants in identifying the referent. At lines 4 and 5, Dimitris and
Tania initiate an insert repair sequence to resolve this trouble. Tania employs
the feminine adjectives ψηλή ([psilí] ‘tall’), χοντρή ([xodrí] ‘fat’) to describe the
referent and Zoi confirms her recipients’ understanding at line 6. At lines 9 and
10, Dimitris and Tania deliver second pair parts to Zoi’s first pair part (line 1),
showing that they have been able to identify the referent. Dimitris uses a recog-
nitional reference form, that is, the feminine proper name Άννα ([ána] ‘Anna’),
to identify the referent. At line 11, in a sequence-closing third, Zoi confirms that
this is the correct name of the referent and uses the feminine nominal demon-
strative αυτή ([aftí] ‘this’) to maintain reference to third person. However, the
pre-sequence is further expanded by Dimitris, who challenges this common
understanding and repeats the feminine proper name with rising intonation.
Tania tries to remember the referent’s proper name and uses the feminine
article η:::: ([i::::] ‘the’) at line 14. She gives additional information about the
referent at lines 17 and 18, keeping reference via the feminine personal pro-
noun αυτή (‘she’). At line 20, Tania asks her co-participants about the referent’s
proper name and re-refers to this specific person via the feminine personal
pronoun (clitic form) τη ([ti] ‘she’). Her co-participants do not deliver a second
pair part and, at line 23, Zoi moves to the first pair part that the pre-sequence
projected. She assumes that their classmate must have graduated and main-
tains reference to her via the feminine personal pronoun αυτή (‘she’).
In this conversational episode, participants make an effort to arrive at an
intersubjective understanding of the referent’s identity. The noun phrase μί:α,
114 chapter 6
.hh <με γυαλιά, μ’ αγορέ μαλλιά που μιλούθε °λίγο έτσι:¿> (‘one [girl] .hh with glasses,
with boyish hair that was speaking like this’) is an initial reference form which
consists of the feminine pronoun μί:α, and the feminine pronouns and articles
η, αυτή, τη are subsequent reference forms. Speakers use the feminine gram-
matical gender to keep track of the third person referred to in this sequence.
Linguistic items marked by feminine gender are taken to be co-referential, that
is, feminine subsequent reference forms select as their referent the same per-
son depicted initially by the feminine intitial reference form (cf. section 2.4).
These feminine items in addition to the feminine adjectives ψηλή and χοντρή
that speakers employ to describe the third person are gender membership
categories that ascribe female sex to the referent, categorize her as woman,
and assume feminine gender as part of the referent’s identity. According to
Schegloff (2007c, 125), subsequent reference forms are designed as ‘referring
simpliciter’, that is, doing reference and nothing more. In this episode, par-
ticipants employ feminine linguistic items as subsequent forms to maintain
reference to third person. Yet, these items make reference to third person func-
tionally complex, because they are grammatically feminine and, thus, index
and presuppose referent’s feminine gender. While participants attempt to
build an intersubjective understanding of the referent’s identity, they also con-
struct an intersubjective understanding of the referent as woman.
In Extract 12, Tania describes the first gynecologist that she visited in
Thessaloniki.
5 Melita [ °Απίστευτο. ]
Unbelievable.
6 Tania → μετά από τόσα χρόνια, ((γελάκι)) αυτός ακόμα είναι
. . . . . . . . . . . . .)) ((laughing))
after prep so.many.acc.neuter years.acc.neuter he.nom
still cop.3sg.pres
after so many years, ((giggle)) he is still
7 → γιατρός, ήταν νέος τότε °βέβαια,° .hh >τώρα πια είναι<
((in a laughing tone)) ((laughing . . . .
((lighter sound and noise))
doctor.nom.masc cop.3sg.past young.nom.masc then cer-
tainly now no.longer cop.3sg.pres
a doctor.masc, of course then he was young.masc, .hh now
[he] is
8 → γέρος, .hh ((γελάκι)) .hh ένας τρελό::ς: γυναικολόγος
. . . . . . .)) ((in a laughing tone. . . . . . .
old.nom.masc one.nom.masc crazy.nom.masc gynecolo-
gist.nom.masc
old.masc, .hh ((giggle)) .hh a.masc crazy.masc
gynecologist.masc
9 → ο οποίος έχει: αφίσες του Τσε: όταν είσαι εκεί
. . . . . . . . . . . . . . . . . . . . . . . . . . .)) ((laughing . . . . . .
the.nom.masc who.nom.masc have.3sg.pres posters.acc.
fem the.gen.masc Che when cop.2sg.pres there
who.masc has posters of Che when you are there
10 στη::[ν τέτοια και περι]μένεις. ((γελά[. . . .)) ναι.] ναι:. .hh
. . . . . . . . . . . . . . . . . . . . . . . . . . . .)) ((laughing))
in that one [the waiting room] and wait. ((she laughs)) yes.
yes. .hh
11 Melita [((she laughs . .))]
12 Dimitris [( . . . . . . . )]
In her telling, Tania uses the noun γυναικολόγο ([ʝinekolóɣo] ‘gynecologist’) for
non-recognitional initial reference to third person and keeps track of the refer-
ent via the personal pronouns τον ([ton] ‘him’) (clitic form) and αυτός ([aftós]
‘he’), the noun phrase ένας τρελό::ς γυναικολόγος ([énas treló::s ʝinekolóɣos] ‘a
crazy gynecologist’) and the relative pronoun ο οποίος ([o opíos] ‘who’). She
also uses the noun γιατρός ([ʝatrós] ‘doctor’) and the adjectives νέος ([néos]
‘young’), γέρος ([ʝéros] ‘old’) and τρελό::ς ([treló::s] ‘crazy’) to describe the gyne-
cologist. All linguistic items mentioned above are inflected for the masculine
116 chapter 6
9 (0.7)
((sound during the silence))
10 Tania [ Ντελίβερι:, ]
Delivery.
11 Ζoi [>Εσύ θα πας στο στρατό<] °φαντάζομαι.
I guess that you will go into the army.
12 Dimitris Βασικά. >πρέπει οπωσδήποτε.<
Basically. definitely I have to.
13 (1.6)
14 Ζoi °Είναι κι αυτό που έχουνε (δυ[στυχώ:ς).]
(Unfortunately) they also have to do this.
15 Τania [ Πο:]πο,
Wow,
16 [<αίσχο::ς.> ((γελά . . . . . . . . . [. . . . . . . . . .] . . . . . . . . . . . .]=
((in a laughing tone))
[That is a] disgrace. ((she laughs))
17 Ζoi [>Ε δεν^ μπειράζει. ας έχουμε και κάτι:.]
Uh that’s fine. let us have something as well.
18 Melita → [>↑(Κι εδώ.)< ας είμαστε]=
and here particle cop.1pl.
pres
Here as well. let us be
19 Τania =[. . . . . . . . . . . .[. . . . . . . . .]. . . . . . . . .]. . .)) ε ναι.
((in a laughing tone))
Eh yes.
20 Μelita → =[και κάπου τυχερές. ]
and somewhere lucky.nom.fem.pl
lucky.fem in certain things as well.
21 Ζoi [ Ε ναι. αυτό. ]
Eh yes. exactly.
At lines 1–2, Melita asks her co-participants about their plans after graduation,
and at lines 5, 6, 7 and 10, Dimitris, Evagelia and Tania deliver their answers to
the question. Dimitris replies with a joke, saying that he is thinking of playing
dumb, and Evagelia introduces a collectivity via the first person plural verb
να γίνουμε ([na ʝínume] ‘we become/becoming’). This collectivity includes
the speaker and other female co-participants, as it is shown by the feminine
nouns πωλή:τριες ([polí:tries] ‘saleswomen’) and σερβιτό:ρε::ς ([servitó:re::s]
‘waitresses’) that the speaker employs to describe the collectivity. At line 11,
118 chapter 6
Zoi self-selects as next speaker and introduces military service as what she and
her co-participants know that Dimitris will do after graduation. She addresses
Dimitris by employing the second person singular pronoun εσύ ([esí] ‘you’)
and the second person singular verb θα πας ([θa pas] ‘you will go’), and by refer-
ring to military service, which is a category-bound activity (Sacks 1972b) ste-
reotypically associated with men in Greece. Greek males between the ages of
19 and 45 are required by law to perform military service for at least 9 months.
Thus, by referring to military service Zoi limits the range of next speakers who
can produce a relevant second pair part to Dimitris only. At line 12, Dimitris
confirms that he needs to go to the army. Referring to military service, which
is a category-bound activity, introduces the relevance of the category ‘man’ in
interaction (Schegloff 2007b), as the following turn shows (line 14).
After a gap, Zoi self-selects again as next speaker and produces the first
pair part of a post-expansion (line 14). She asserts that unfortunately men are
obliged to fulfill their military service duties. She refers to this obligation by
employing the deictic pronoun αυτό ([aftó] ‘this’). The anaphoric use of the
pronoun serves as a ‘tying structure’ (Sacks 1995, 540), connecting Zoi’s turn
with the preceding sequence. Zoi refers to men as a group via the third per-
son plural verb έχουνε ([éxune] ‘they have’) and expresses her sympathy or
empathy (Ruusuvuori 2012) for men via the lexical choice δυστυχώς ([ðistixós]
‘unfortunately’). At lines 15–16, Tania supports Zoi’s affective stance by assess-
ing men’s obligation to perform military service in a negative way. However,
at line 17, Zoi takes a different affective stance: she mitigates Tania’s negative
assessment (first tcu) and expresses a wish/desire (second tcu). She intro-
duces the wish via the particle ας and refers to a collectivity, which includes
herself, her female co-participants and all women in general, via the first per-
son plural έχουμε ([éxume] ‘we have’). Similar to English first and third person
plural pronouns we and they which can refer to organizational collectivities to
which the speaker belongs (Sacks 1995, 713–714), the first and third person plu-
ral verbs in these turns refer to two different gendered groups ‘we women’ vs.
‘they men’. Zoi wishes that she and all women had at least some sort of privi-
lege, implying that usually men are in a more privileged position than women.
Melita agrees with Zoi at lines 18 and 20, by employing a similar format. She
introduces a wish via the particle ας [as], refers to self collectively via the first
person plural verb είμαστε ([ímaste] ‘we are’), and describes the collectivity via
the feminine adjective τυχερές ([tiçerés] ‘lucky’). She wishes that women were
lucky in some ways, implying that women are not lucky in general or that they
experience a greater number of difficulties compared to men. Thus, Melita
produces an affiliative response (Lindström and Sorjonen 2012), that is, she
supports Zoi’s lack of empathy for men’s obligation to perform their military
Empirical Investigation of Grammatical Gender in Interaction 119
duty, and endorses Zoi’s desire for more justice or equality for women. The
grammatically feminine item τυχερές, which ascribes female sex to the collec-
tivity and categorizes referents as women, foregrounds gender interactionally.
Tania and Zoi agree with Melita at lines 19 and 21.
In the next extract, Melita, Tania, Zoi, Evagelia and Dimitris discuss the
differrences between girls and boys in the period of adolescence. In the lines
preceding this extract, Tania and Melita tell stories about female friends who
helped boys with their school homework and were sexually harassed by them.
Melita’s story is utilized as a resource for generating further topic talk (Jefferson
1978) on the sexual behavior of boys in childhood and adolescence.
Melita foregrounds the age of the boy as an important factor in the incident
reported (lines 1–2), by using the formulation παιδάκι ([peðáci] ‘little kid’).
Dimitris agrees with Melita at lines 3–4 and prompts her to think about her
own behavior when she was a kid. Melita foregrounds age again as the key
reason for the boy’s behavior at line 8. Τania disagrees with Dimitris and
Melita (lines 10, 12 and 13) and argues that girls behave differently. Initially,
she refers to women by deploying the gender membership category γυναί:κες
([ʝiné:ces] ‘women’). Then, she switches to the gender membership category
κορίτσια ([korít͡sça] ‘girls’), foregrounding both age and gender. She uses the
first person plural verb μπερνάμε ([bernáme] ‘we go through’) to refer to herself
collectively. The collectivity includes the speaker, her female co-participants
and all women in general. In her turn, Tania differentiates girls from boys and
employs linguistic items lexically marked for female sex to foreground femi-
nine gender interactionally. Dimitris disagrees with Tania on whether these
gender differences exist or not at lines 14, 15 and 16. He refers to his female
co-participants and all women via the second person plural περνάτε ([pernáte]
‘you.PL go through’) and εκδηλώνετε ([ekðilónete] ‘you.PL show/manifest’), and
he refers to himself collectively via the first person plural verbs and pronoun
εκδηλώνουμε ([ekðilónume] ‘we show/manifest’), είμαστε ([ímaste] ‘we are’),
and εμείς ([emís] ‘we’). Thus, by employing first person plural, he introduces
a collectivity that is different from the collectivity of women and includes
men only.
Tania disagrees with Dimitris, at lines 17–18, and insists that no woman
experiences her sexuality in such a blunt (or brutal) way. She employs the
grammatically feminine indefinite pronoun καμία: ([kamía:] ‘anyone’) for
non-recognitional reference to women. At the same time, by deploying this
gender membership category, the speaker foregrounds sociocultural gender
interactionally. In the following turns, Melita highlights gender differences in
adolescence and employs gender membership categories to foreground gen-
der interactionally. These categories include linguistic items in which sex is
lexically specified, such as αγό:ρια ([aɣó:rʝa] ‘boys’), αγόρι ([aɣóri] ‘boy’), άντρας
([ándras] ‘man’), κορί:τσι: ([korí:t͡si:] ‘girl’), γυναί:κα ([ʝiné:ka] ‘woman’), κορίτσι
([korít͡si] ‘girl’), αγό:ρι ([aɣó:ri] ‘boy’), γυναίκα ([ʝinéka] ‘woman’), or grammati-
cized, such as τυχερές ([tiçerés] ‘lucky.fem’).
In Extract 15, Panayotis together with his two female friends, Danai and
Adriana, make plans for a party at Panayotis’ house.
At lines 1, 2 and 3, Panayotis refers to his male friends, by using the grammati-
cally masculine reference forms τους σκύλους (tus skílus] ‘the dogs’),5 τους άλλους
5 The noun denotes metaphorically a person who listens to low quality folk music.
Empirical Investigation of Grammatical Gender in Interaction 123
([tus álus] ‘the others’), ο Στάθης ([o stáθis] ‘Stathis’), αυτοί ([aftí] ‘they’). He
associates referents with the stereotypically masculine activity of grilling in
parties. At line 6, Adriana uses the first person plural pronoun εμείς ([emís] ‘we’)
to introduce a new collective referent, which includes the speaker and is dif-
ferent from the referents already introduced by the previous speaker (Pavlidou
2012a). She locates this new collectivity in the kitchen, which is linked with the
stereotypically feminine activity of cooking. The speaker makes the category
‘woman’ interactionally relevant by referring to this category-bound activity. At
lines 9 and 11, Danai keeps reference to the collectivity of women, by repeating
the first person plural pronoun εμείς, and locates referents in the kitchen. She
describes referents by using the verb phrase τάχα μαγειρεύουν ([táxa maʝirévun]
‘pretending to cook’), which refers to the stereotypically feminine activity of
cooking, and the noun phrase οι κατίνε:ς ([i katíne:s] ‘the.fem gossipers.fem’).
This noun phrase consists of a grammatically feminine head noun and modi-
fier, which categorize referents as women. In addition, the noun κατίνες is a
pejorative term denoting uneducated women who like to gossip (Dictionary of
Standard Μodern Greek 1998). Danai’s deployment of these gender member-
ship categories is related to her orientation to her own and her co-participant’s
feminine identity. In this conversational episode, the category ‘woman’ is fore-
grounded interactionally and a collective feminine identity is constructed via
referring to cooking and deploying the first person plural pronoun εμείς and
the grammatically feminine noun κατίνες, which is also semantically marked
for female sex.
In Extracts 13, 14 and 15, linguistic items marked by grammatical gender are
part of the tools that participants employ for foregrounding gender interac-
tionally and constructing gender identities. In all three cases, the information
of feminine gender, which is made available in interaction, becomes relevant
for what participants do in interaction.
Dimitris refers to his co-participants via the second person plural verb θα
συνεχίζατε ([θa sineçízate] ‘you would continue’), and describes them via the
masculine adjective όλοι: ([óli:] ‘all’). That is, he uses the masculine gram-
matical gender for reference to female persons only, based on the inference
of human/universal associated with the use of the masculine gender. The mis-
match between grammatical gender and sex does not create any problems in
interaction and remains unnoticed by participants.
A similar case is found in Extract 17. Natasa delivers a story telling about a
movie screening at a film festival that she attended together with her female
friend Amalia. Natasa uses the masculine gender in reference to female per-
sons only.
At line 1, Natasa uses the first person plural πήγαμε ([píɣame] ‘we went’) to
refer to a collectivity that includes herself and her friend Amalia. At line 21,
she describes the collectivity via the masculine adjective άσχετοι ([ásçeti] ‘out
of place’), which produces the inference of human/universal. The non-match
between the masculine gender and the female sex of referents is not consid-
ered to be a problem by participants. On the contrary, it is used as a tool for
the interactional work that the speaker is doing at that particular moment:
describing referents in a story.
In both Extracts 16 and 17, speakers refer to self or recipient, that is, to indi-
viduals who are physically present in context and can be easily identified given
the knowledge that participants share about the ‘here-and-now’ of interaction.
Second, speakers refer to more than one referent. This seems to motivate the
use of the masculine gender in these extracts. Speakers utilize the inference of
human/universal associated with the masculine gender in generic reference
to achieve reference to more that one female referents. Third, in both extracts,
gender is not relevant interactionally. One could assume that the lack of
126 chapter 6
17 Tania η τρίχα:,=
((in a laughing tone))
[to have] hair,
18 Μelita =Εί[ναι η κοινω]νία τέ:τοια:.
((noise starts))
Society is like that.
19 Τania [((giggle ))]
20 (.)
21 Μelita → Και θα είσαι τυχερός άμα είσαι σε μία κ- ε: οικογένεια::
and fut cop.2sg.pres lucky.nom.masc if cop.2sg.pres in
a.acc.fem family.acc.fem
And you will be lucky.masc if you are in a k- u:h family
22 φυσιολογική. .hh σκέψου να ’ναι ο πατέρας σου κανένας
[that is] normal. .hh imagine if your father was
23 <αυστηρό:[:ς,> °άντε-°] (0.5) °παναγία μου.
strict, get away- (0.5) my goodness.
24 Τania [ °Ναι ναι. ]
Yes yes.
6.2.6 Conclusions
The analysis presented so far shows that the compulsory use of grammatical
gender in the composition of turns has specific implications for the sociocul-
tural world that participants jointly build in interaction. Items marked by gram-
matical gender are obligatory gender membership categories which structure
and constrain participants’ practices. They enforce participants to categorize
referents—self, recipient and other—as women or men on the basis of social
hierarchy in automatic, compulsory and systematic ways, when they accom-
plish various social actions. These actions may be related or not to participants’
orientation to gender. In both cases, the gendered categorization of referents
and the inference of human/universal via the codification of male sex are part
of the routine, unremarkable and subtle meanings produced in interaction in
automatic ways, and are tacitly presupposed by participants.
When participants orient to gender (cf. section 6.2.4), they utilize the infor-
mation of referent’s gender that is made available in interaction via gram-
matical gender in order to construct the bipolar gender order. These social
processes associated with the use of grammatical gender take place even when
participants do not attend to gender as an interactionally relevant category. For
instance, the gendered categorization of referents occurs when participants
do an assessment refer to third person in a locally initial or subsequent posi-
tion or refer to more than one referent (cf. sections 6.2.1, 6.2.2, 6.2.3 and 6.2.5).
In the above cases, grammatically masculine and feminine items are gendered
linguistically but not interactionally. This fact has implications for the routine
meanings produced in interaction, given the indexical effect associated with
grammatically feminine and masculine items. Although participants show no
orientation to sociocultural gender, they assume gender as an aspect of refer-
ent’s identity. Overall, grammatical gender contributes to the process of ‘gen-
dering’ (Pavlidou 2011, 415) or ‘genderizing’ (Eckert and McConnell-Ginet 2003,
254) interaction. As participants take turns in interaction and accomplish vari-
ous aspects of their daily social life, they construct gender bipolarity and gen-
der asymmetry in routine and covert ways because of grammar.
Empirical Investigation of Grammatical Gender in Interaction 129
In the above extract, the speaker initiates repair by cutting off her talk and
effects the repair by replacing a verb uttered in error (ήρ- [ír]) with another
verb (πέρασε [pérase]). However, in the next extract, repair is initiated by the
recipient, who is not able to identify the third person that the speaker refers to.
cited in Sidnell 2010, 115) before (‘pre-framing’) or after the trouble source
(‘post-framing’). Same tcu self-repairs may be accompanied by silences and
delays or the use of apologetic terms (e.g. ‘sorry’), repeats of the trouble-
source, multiple tries or self-talk (Kitzinger 2012, 239–241). As Schegloff (2013)
shows, self-repair in same tcu involves several operations, such as replacing
an item by another, inserting an item into the prior talk, deleting an item from
the prior talk, re-ordering words or phrases, searching names or places, paren-
thesizing, aborting, sequence jumping, recycling or reformatting. Self-repair
can be employed for various interactional uses. Besides fixing a problem in
speaking, hearing or understanding, self-repair can also “be used in the service
of the action the speaker means to be doing with the talk” (Kitzinger 2012, 242),
and, thus, “reveal the work involved in designing talk for the action the speaker
intends it to do” (Kitzinger 2012, 243). In Drew, Walker and Ogden’s words (2013,
93), self-repairs give analysts “access to the work of constructing a turn—they
bring to the interactional surface the work in which speakers engage in order
to construct the action”. Similar to same tcu self-repairs, self-repairs in the
transition space after the possible completion of a turn involve operations
such as replacing and ‘framing’ of repairable items (Kitzinger 2012, 244–246).6
Other-initiated repair is delivered in a sequence (insert or post-expansion),
which “suspends the ongoing action in which the participants are engaged”
(Kitzinger 2012, 249). Repair can be initiated by a party other than the speaker
via the deployment of open class forms, such as what?, huh?, pardon?, sorry?,
interrogatives, such as who?, when?, where?, or repeats of the trouble-source
turn and candidate understandings, such as resayings of the trouble-source
(Kitzinger 2012, 249). In both self-initiated and other-initiated repair, the pro-
gressivity of the talk is interrupted so that intersubjective understanding is
sustained. As I mentioned in section 5.3, progressivity is defined as the moving
from some element to a hearably-next-one with nothing intervening. In repair,
the new element intervening is inspected by participants for its import, that is,
for the ways in which it reaffirms or changes the understanding of the turn so
far or the understanding of the previous turn.
Sometimes speakers “correct” other participants’ talk without employing
the technology of repair. As Kitzinger (2012, 256) notes, in these cases trouble
in speaking, hearing or understanding is resolved without having the trou-
ble “rising to the surface of the talk as overt repair”. Jefferson (1987) defines
this form of ‘covert’ correction as embedded correction and distinguishes it
6 Self-repair can also occur in third turn after intervening talk from another between repair-
able and repair or when speakers display some sort of misunderstanding of prior turn (third
position self-repair) (Kitzinger 2012, 246–249).
132 chapter 6
self-corrections show speakers’ ability to monitor the speech that they pro-
duce, for instance their ability to control whether what they say is appropriate
for their communicative purpose or suitable given the context of the speech
event. Given the role that repair plays in resolving trouble in participants’
understandings, we can theorize repair as a ‘window’ to participants’ cognition
(cf. Schegloff 1991, section 5.3), and, thus, to the cognitive role of grammatical
gender in interaction.
In the data I examined, I found overall 6 cases of repairing grammatical gen-
der: 4 self-initiated self-repairs, 1 exposed correction and 1 embedded correc-
tion. The small number of repairs found in the corpus is not a surpise. Since
grammatical gender is inherent in language structure and, thus, used in an
automatic, systematic and obligatory manner in the composition of turns, it
is expected that speakers will rarely notice the cognitive role that this feature
of grammar plays. Yet, these rare cases of repair are of special interest for the
study of the relation between grammatical gender and cognition, because they
provide indications of the cognitive role of grammatical gender, as I show in
the next section.
In the story telling, Areti refers to the tv show players and describes how they
get out of the studio when they leave the show and how their parents welcome
them. At lines 25–26, she refers to the parents of these players who wait for
their kids outside the studio, and run to them crying. Areti employs a direct
speech report construction (Aikhenvald 2011) to describe the parents’ behavior
(lines 26, 27, 28). According to Aikhenvald (2011, 400), speech reports involve
the “author” of the original speech and the “reporter”. The speech report con-
tent corresponds exactly (more or less) to what the author of the original
speech said. Areti uses the exact words that the original speakers, i.e. the par-
ents, used in the first place to express themselves παιδί μου, είμαι πε↑ρήφανος, .h
(.) περήφανος ή περήφανη για σένα. ([peðí mu, íme pe↑rífanos, .h (.) perífanos í
perífani ʝa séna.] ‘my child I am proud.masc .h proud.masc or proud.fem of
you’). Direct speech report is linked with speakers’ shifts in their role as anima-
tor and author, that is, in their role as the person whose voice is actually being
used to produce a strip of speech, and the person who actually constructed
the phrase said (Clift and Holt 2007; Goffman 1981). In the direct speech report
in this extract, the parents are the authors of the strip of speech reported and
Areti is the animator. The self-repair to be examined occurs within this direct
speech report.
136 chapter 6
ensure that reference to the same third person that was originally introduced
is kept in interaction. Another reason why categorizing referent as female
becomes important at this particular moment in interaction is the fact that the
speaker refers to parents crying. Crying is a category-bound activity, that is, an
activity stereotypically associated with the members of the category ‘mother’.
The rest of the repairs to be examined in this section involve the generic
use of the masculine. In extract 20, Melita, Tania, Zoi and Evagelia discuss the
problems that partners encounter in living together in the same house.
14 στο σπίτι:.
in the house.
15 (0.7)
16 Μelita °Κατάλαβες?° >καταλάβατε [τι εννοώ?<] [((giggle))]
Do you.sg understand? do you.PL see what I mean?
17 Tania [ Μ:. ]
Mm:.
18 Ζoi [Ναι αλλ-] εγώ
Yes but- I
19 νομίζω κατάλαβα. αλλά εκεί είναι και πιο δύσκολο.
think I understood. but in that case it is more difficult.
20 °να κρατηθεί.
to keep it [that way].
21 (0.6)
22 Μelita Μ::.
Mm.
23 (0.9)
24 Ζoi Και όπως επίσης, δε: θα μπορέσεις- (0.6) όχι δε θα μπορέσεις.
And in addition, you will not be able- (0.6) it’s not that you
won’t be able [to do it].
((noise during the silence))
25 θα μπορέσεις. ο άλλος δεν^ γκζέρω πως θα το δεχτεί, (.)
you will be able. I don’t know how the.masc other.masc is
going to deal with it, (.)
26 όπως ας πούμε εσύ θα πεις στην συγκάτοικό σου, ξέρεις τι?
for example you ’ll tell your flatmate.fem, you know what?
27 έλα να: καθαρίσουμε πιχί το σπίτι. έτσι? θα κάνω εγώ αυτό.
come on let’s clean the house for instance. right? I will do this.
28 εσύ [ το ] άλλο. (.) hh όταν θα το πεις στο:ν άλλον που
you will do that. (.) if you tell the other.masc with whom
29 Τania? [°Μ.]
Mm.
30 Ζoi θα μένεις μαζί σου, (.) >μπορεί να-< να γελά:σει, >μπορεί να
you live together, (.) he might laugh, he might
31 πει< (.) [ τι? ]
say (.) what?
32 Εvagelia [°Να] σε βρί:σει.=
He might insult you.
33 Τania? =((she laughs[..))]
Empirical Investigation of Grammatical Gender in Interaction 139
Melita initiates a sequence about the stereotypical roles of men and women
who are couples living under the same roof (line 1). In the following lines, she
argues that there should be certain rules with respect to how partners share
space, so that they do not loose respect for each other (δεν^ μπρέπει να χαθεί
ο σεβασμός στον άλλον, και [το ότι] πρέπει να κάνεις εσύ κάποια πράγματα κι αυτός
κάποια άλλα [ðen brépi na xaθí o sevazmós ston álon, ce to óti prépi na kánis
esí kápça práɣmata ci aftós kápça ála] ‘they should not loose respect for each
other, and [keep in mind] that you have to do certain things and he must do
others’, Δηλαδή να υπάρχει μί:α:: (2.3) ένα τυπικό ρε παιδί μου στο στο σπίτι: [ðilaðí
na ipárçi mía:: éna tipikó re peðí mu sto sto spíti:] ‘In other words there should
be one (2.3) [there should be] rules in the house’). At lines 24, 25, 26, 27, 28,
30, 31, 34 and 35, Ζoi argues that usually men are unwilling to clean the house
and undervalue such tasks (>μπορεί να-< να γελά:σει, >μπορεί να πει< (.) [τι?] [borí
na na ɣelá:si, borí na pi ti?] ‘he might laugh, he might say (.) what?’). Evagelia
agrees with Zoi at line 36. Her affiliative response consists of two tcus and one
increment. In the first tcu Καλά ναι. ([kalá ne.] ‘Yeah right.’), Evagelia agrees
with the previous speaker’s turn. In the second tcu °ποιος το κάνει? ([pçós to
káni?] ‘who.masc does it?’), she asks a rhetorical question, which further shows
her agreement, and then she delivers the increment =ποιος ά[ντρας?] ([pçós
ándras?] ‘which.masc man?’) with latching. In the increment, she does a self-
repair which concerns the preceding tcu. In her rhetorical question, Evagelia
uses the masculine interrogative pronoun ποιος ‘who’ for generic reference. In
the transition space after the grammatical and pragmatic completion of the
140 chapter 6
tcu, the speaker inserts the noun άντρας (‘man’), which is both grammatically
and lexically marked as male, and she pre-frames the repairable item, i.e. by
repeating the item ποιος.
Through her self-repair the speaker aims at making the third person’s male
sex explicit and, thus, solving or anticipating a possible problem with respect
to how repicients understand the turn. The generic use of the masculine in the
second tcu could guide recipients to interpret referent’s sex as indefinite or
not exclusively male. This interpretation would cancel the kind of agreement
that Evagelia’s rhetorical question is designed to do: Evagelia agrees with Zoi
that men do not participate in cleaning the house.
The analysis of this self-repair provides support for the argument presented
in previous chapter about the relation between the generic use of the mas-
culine and the inference of human/universal. More specifically, it shows that
speakers understand the masculine grammatical gender as a tool for achieving
generic reference to humans.
However, things are a bit more complicated when it comes to the generic
use of the masculine in interaction, as the following repairs show. The self-
repair to be examined in Extract 21 occurs at line 257. It comes after a series
of reciprocal or exchange sequences (Schegloff 2007a) in which participants
judge the similarity in appearance between co-participants and public figures,
such as singers, actors, etc. According to Schegloff (2007a, 195), reciprocal or
exchange sequences “refer to episodes in which a sequence which has just
been initiated by A to B (that is, in which A is the first pair part speaker and
B is the second pair part speaker) is then (after it has run its course) recipro-
cated—initiated by B to A”.
8 Ζoi [ Αχ ναι:. ]
Oh yes.
9 Dimitris [↑Η Χούλια.]
Choulia.fem.
10 Ζoi [Μια ]
One.fem
11 Ζoi =Ναι μοιά:[ζει:ς . ]
Yes you look like her.
((98 lines omitted))
110 (1.5)
((noise . . .
111 Τania Έτσι. >για να πάρω μια γεύση.<
Like this. in order to get an idea.
. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .))
112 Μelita → >Η Ευαγγελία με ποια σας μοιάζει?< εμένα μου μοιάζει με
Whom.fem do you think Evagelia looks like? I think that she
looks
113 τη Νάνσυ Αλεξιάδου.
like Nancy.fem Alexiadou.fem.
114 (0.5)
115 Evagelia Ίδιες είμαστε.
((ironically))
We are the same.fem.
((noise))
((41 lines omitted))
157 (1.5)
158 Μelita → Εσάς με ποιους σας έχουν πει?
Whom.masc.pl did they tell you that you look like?
159 (0.6)
160 Τania °Με κανέναν. ((γελά[. . .)) είναι στη μόδα (. . . . . . . . . . . . . . )]
Like nobody.masc. ((she laughs)) it is in fashion ( . . . . .)
161 Melita [↑Τίποτα? κανείς? δεν [ γίνεται. ]]
Nothing? no one? it can’t be.
162 Ζoi [Με την αδερ]φή
Like my sister,
163 [μου, με τον μπαμπά]:: μου[: ]
like my father
164 Τania [((she laughs. . . . .))]
165 Dimitris [Με] [την Ελπίδα,]
Like Elpida.fem,
142 chapter 6
In the lines preceding this extract, participants talk about ΤV series for chil-
dren. At lines 3, 5 and 7, Tania does stepwise topic transition (Jefferson 1984;
Empirical Investigation of Grammatical Gender in Interaction 143
Sidnell 2010, 240–244), by asserting that when she was little other people told
her that she looks like Choulia, an actress in a Brazilian tv series for chil-
dren (μου ’λέγαν [ότι μοιάζω] με μία που παίζει σ’ αυτό [mu léɣan óti mɲázo me
mía pu pézi saftó] ‘they told me that I look like someone.fem who plays in
this series’). Zoi agrees with this assessment (lines 8, 11). Tania’s similarity to
Choulia is topicalized in the following lines that have been omitted. In these
lines, Dimitris agrees with the assessment, and Tania mentions that her mum
and her schoolmates also shared the same view. She also gives a description
of her look when she was a kid to justify this view. At lines 112–113, Melita asks
her co-participants whom they think Evagelia looks like (>Η Ευαγγελία με ποια
σας μοιάζει?< [i evagelía me pça sas mɲázi] ‘Whom.fem do you think Evagelia
looks like?), and she describes Evagelia as similar to the female singer Nancy
Alexiadou (εμένα μου μοιάζει με τη Νάνσυ Αλεξιάδου. [eména mu mɲázi me ti
nánsi aleksiáðu] ‘I think that she looks like Nancy.fem Alexiadou.fem’). Her
first pair part initiates a sequence (the lines are omitted) in which Evagelia
disagrees with Melita’s assessment. Melita initiates a new sequence at line 158,
asking Zoi and Dimitris whom other people told them that they look like (Εσάς
με ποιους σας έχουν πει? [esás me pçus sas exun pi?] ‘Whom.masc.pl did they
tell you that you look like?’). At lines 162–163, Zoi describes herself as similar to
her family members ([Με την αδερ]φή [μου, με τον μπαμπά]:: μου[:] [me tin aðerfí
mu, me ton babá:: mu:] ‘Like my sister, like my father’), and at line 165, Dimitris
describes Zoi as similar to the female person, Elpida ([Με] [την Ελπίδα,] [me
tin elpíða] ‘Like Elpida.fem’). At line 166, Evagelia describes herself as similar
to her mother ([Με τη μα]μά μου, [me ti mamá mu] ‘Like my mum’). In the
lines omitted, participants disagree with Dimitris’ assessment and comment
on actresses playing roles in these tv series. In general so far, participants
describe their female co-participants as similar to female persons, that is, simi-
larity assessment seems to be based on the sex of the person being assessed.
In this sense, assessing similarity in appearance between persons constitutes a
category-bound activity. In particular, it is an activity linked with the category
of female sex that introduces the relevance of female sex in interaction, as
I show in the following.
At line 257, Zoi addresses a question to Melita. Her turn consists of a tcu
Εσένα με ποιον σου είπαν ότι μοιάζεις? ([eséna me pçon su ípan óti mɲázis?]
‘Whom.masc did they tell you that you look like?’), which is grammatically
and pragmatically complete, and an increment με ποια? ([me pça?] ‘whom.
fem?’), which is designed as grammatical continuation of the tcu and deliv-
ers a self-repair. In the tcu, Zoi selects Melita as next speaker by using the sec-
ond person singular verb μοιάζεις (‘you look like’) and the pronouns εσένα, σου
(‘you’), and associates Melita with a third person. Zoi refers to the third person
144 chapter 6
7 The same clitic form is employed for genitive and accusative case in first person plural per-
sonal pronoun.
148 chapter 6
Extract 23 [Ι.14.Α.17.1]
1 Αreti [Ε ναι ρε [συ.] τις έχουν ξεπα↑στρέψει τις <γυναί]κες.>
Uh yes man. women have been kicked off.
2 Yanis [°(. . . . . . . . . . .)]
3 Μarina [Για↑τί μιλάς έτσι για μια <κοπέλα.>]
Why do you speak about a girl like that.
4 Yanis ↑Α[ΦΟΥ Η]ΤΑΝΕ χαζοχαρούμενη η ά[λλη ρε.]
Because she was stupid/lighthearted.
5 Αreti [°(είναι-)]
([She] is-)
6 Μarina [↑ΕΝΤΑ]:ΞΕΙ
Okay.
7 ΟΜΩΣ- h (.) ε?=
but- h (.) eh?
8 Yanis =[(Δηλαδή αυτή δεν] είναι?)=
That is to say that she is not?
9 Αreti =[ Η Γωγώ::? ]
[You mean] Gogo?
10 Μarina = >Όλες οι άλλες όμως< ήτανε: μπουρούχες να πούμε. ήταν
All.fem the others.fem were ugly and stupid so to speak.
11 ήταν η μόνη που: τέλος πάντων έδινε κάποια ζωντάνια. κάτι.=
anyway she was the only one who livened things up. [who was
doing] something.
12 Yanis =°Ναι.
Yes.
13 (.)
14 Yanis Όπως λένε κι οι φα[ντάροι, η Δώρα ήταν η κλαψομούνα.]
As soldiers [in the military service] say, Dora was the whining-
cunt.fem.
150 chapter 6
At line 14, Yanis describes Dora via the feminine pejorative term κλαψομούνα
([klapsomúna] ‘whining-cunt’), that is, as a person who always complains and
does not get pleasure from anything. At line 23, Niki initiates a repair sequence
to deal with her trouble in identifying the person who is described as a whin-
ing-cunt (=> Ποιος είν’ ο κλαψομούνης? [pços in o klapsomúnis?] ‘Who.masc is
the whining-cunt.masc?’). She initiates repair via the masculine interrogative
pronoun ποιος (‘who’) and describes the referent via the masculine adjective
κλαψομούνης (‘whining-cunt’). Niki’s first pair part makes relevant a second pair
part that will deliver the repair. Instead of delivering a second pair part, the
next speaker produces a turn that is designed to correct previous speaker’s
turn (the gap after the first pair part indicates that there might be a prob-
lem with the production of the second pair part). The item of this exposed
correction is the gender membership category κλαψομούνης. Yanis corrects the
Empirical Investigation of Grammatical Gender in Interaction 151
masculine adjective that Niki used in her previous turn by offering the equiva-
lent feminine adjective as an alternative. More specifically, he uses the second
root of the compound word, the feminine item μούνα [múna] that is marked
as female both grammatically and semantically. The semantic marking of sex
is due to the fact that the word μουνί ([muní] ‘cunt.NEUTER’) denotes woman
metonymically. As I mentioned in section 6.3.1, exposed correction consists
of a speaker producing an object X, a subsequent speaker producing an alter-
native object Y, and a prior speaker producing an alternative object Y or the
same object X. The first two parts of this series are found in this conversa-
tional episode. Namely, Niki employs the masculine grammatical gender and
Yanis offers the feminine grammatical gender as an alternative. The correction
sequence is not continued, because Yanis’s correction overlaps with Spyros’s
first pair part, which continues the sequence that was suspended by the
exposed correction.
The exposed correction done by the speaker shows that the speaker inter-
preted the sex of the referent introduced by the previous speaker via the gram-
matically masculine pronoun as male. For this reason, this correction can be
interpreted as an indication of the correlation between grammatical gender
and cognition. Contrary to what we have seen so far, the exposed correction in
this conversational episode does not seem to be contextually motivated. Or, if
context plays some role, this is not shown in the interaction data.
An embedded correction of grammatical gender is examined in the next
extract. The conversational topic here is the curriculum and other issues
related to participants’ undergraduate studies at the School of Philology.
Extract 24 [Ι.20.Α.28.1]
1 Μelita [Καταρχήν] εγώ <πιστεύω,> ότι ε στο πανε- το πεζ- (0.5) ε
First of all I believe, that uh in the uni- the pez- (0.5) uh
2 ↑πτου:. ((γελάκι))=
damn. ((giggle))
3 Τania =((she laughs[. . . . . . . . . . .))]
((5 lines omitted))
9 Μelita [Το <πανεπιστήμιο> δε σου μα]θαίνει παιδιά κάτι
Hey guys the university does not teach you.sg anything
10 για να διδά:ξει::ς. ↑σας έχει μάθει εσάς τ’ αρχαία ξέρω γω: .h=
with respect to teaching. did it teach you.PL [anything] about
ancient Greek for example .h
11 Εvagelia =Ου:. πάρα πολύ.=
interj [irony] a lot.
((in a laughing tone))
152 chapter 6
12 Dimitris =↑Το πιστεύεις ότι κι εγώ αυτό το ι- (.) ίδιο παρατηρώ >ας
Do you believe that I also observe this the i- (.) the same
[thing] so
13 πούμε ότι< δεν έχει[: ένα μά]θημα παιδα- >να μου πεις<
to speak that there is no peda- course anyway
((lighter sound))
14 Melita [( . . . . .)]
15 Dimitris έχει ένα παιδαγωγικό το οποίο είναι εισαγωγή. καπ-
there is a pedagogical [course] that is an introduction. som-
16 ο τρόπος >δηλαδή να διδάξεις το μάθημα.<=
namely the way to teach a course.
17 Melita =Εκτός του- [του τρό:που:. και το αντικείμενο ] ρε παιδί μου.
Besides the- the way. [it is about] the topic as well.
18 Dimitris [Όταν μπεις σε τάξη. έστω μια °πρακτ-]
When you.SG enter a class. some [sort of ] practi- so
be it
19 Melita στ’ αρχαία πιχί κάναμε στο σχολείο, το και το,
for example at school in the course on ancient Greek we did,
this and that,
20 να ξες να το εξηγήσεις, .h ↑τώ:ρα, είναι μόνος σου, ά:μα
((noise. . . . .))
subj know.2sg subj it.acc explain.2sg.pfv now cop.3sg.
pres on.your.own.nom.masc if
you should know how to explain it, .h now, you are on your
own.masc, if
21 [θες >να] κάτσεις να< τα διαβάσει:ς, [.h και να] είσαι και
want.2sg.pres subj sit.2sg.pfv subj them read.2sg.pfv
and subj cop.2sg.pres and
you want to sit down and read them, .h and you will be also
22 Dimitris [ Έτσι. ]
Precisely.
23 Tania [Ναι °ναι.]
Yes yes.
24 Melita τυχερός άμα το περάσεις με ένα πέ:ντε¿ [μια χα]ρά, .h
lucky.nom.masc if it.acc pass.2sg.pfv with a.neuter five
a.fem joy.fem
lucky.masc if you succeed in passing the exams with a five.
that’s fine, .h
25 Tania [°Ναι:.]
Yes.
Empirical Investigation of Grammatical Gender in Interaction 153
Melita claims (lines 9–10) that students do not receive any teaching training
at the university ([Το <πανεπιστήμιο> δε σου μα]θαίνει παιδιά κάτι για να διδά:ξει::ς.
[to <panepistímio> ðe su maθéni peðʝá káti ʝa na ðiðá:ksi::s] ‘Hey guys the uni-
versity does not teach you.SG anything with respect to teaching.’). The speaker
employs the second person singular pronoun σου (‘you’) to refer to an abstract
group of referents, all students of the school of philology, including speaker
and the other co-participants. In the next tcu, she addresses a question to
her co-participants, soliciting confirmation of her claim (↑σας έχει μάθει εσάς
τ’ αρχαία ξέρω γω: .h= [↑sas éçi máθi esás tarçéa kséro ɣo: .h] ‘did it teach you.
PL [anything] about ancient Greek for example .h’). She selects them as pos-
sible next speakers via the second person plural pronouns εσάς, σας (‘you’). In
his reply at lines 12, 13, 15 and 16, Dimitris keeps reference to the same abstract
group of referents via the second person singular verb να διδάξεις ([na ðiðáksis]
‘you teach’). At lines 17, 19, 20, 21, 24 and 26, Melita refers to the same referents
via the first person plural verb κάναμε ([káname] ‘we did’, line 19). Then, she
shifts to second person singular να ξες να το εξηγήσεις ([na kses na to eksiʝísis]
‘you should know how to explain’, line 20), θες >να κάτσεις να< τα διαβάσει:ς
([θes >na kát͡sis na< ta ðʝavási:s] ‘you want to sit down and read them’, line
21), να είσαι ([na íse] ‘you are’, line 21), περάσεις ([perásis] ‘you pass’, line 24),
and categorizes the referent via the masculine second person singular pro-
noun μόνος σου ([mónos su] ‘on your own’, line 20) and the masculine adjective
154 chapter 6
τυχερός ([tiçerós] ‘lucky’, line 24). In their turns, speakers employ the second
person singular to refer to self as part of an abstract group of referents and
make assertions about the collective experience that they share with their co-
participants. In this way, they introduce the relevance of their own and their
co-participants’ experience as students at the school of philology (cf. Schegloff
1996b). Tania keeps generic second person self-reference at lines 27–28 (να
διαβάσεις [na ðʝavásis] ‘you read’). However, she categorizes referents as female
via the feminine pronoun μόνη σου ([móni su] ‘on your own’). In her next turn,
Melita agrees with Tania. She keeps generic second person self-reference and
repeats the gender membership category μόνη σου.
The change of grammatical gender in this conversational episode operates
as embedded correction, in Jefferson’s (1987) terms. The speaker employs a
linguistic item marked by the masculine grammatical gender, that is, she pro-
duces an object X. The subsequent speaker employs a linguistic item marked
by the feminine grammatical gender, that is, she produces an alternative
object Y. Finally, the prior speaker employs the linguistic item marked by the
feminine grammatical gender, that is, she produces the alternative object Y.
The XYY is embedded into the ongoing talk with no attendant activities. The
categorization of referents as exclusively female becomes important for what
participants do in interaction, because of the relevance of speaker’s personal
experience that has been introduced in interaction via the generic second per-
son singular. This experience seems to be hidden by the abstraction and indefi-
niteness that the generic use of the masculine carries. The deployment of the
feminine pronoun limits reference to female persons only and, thus, aims at
foregrounding speaker’s personal experience.
In the extracts examined in this section, the codification of referent’s sex
via grammatical gender becomes important for what participants do in inter-
action. In these cases, the cognitive role of grammatical gender is uncovered
in interaction. In the next section, I show that indications of the cognitive
role of grammatical gender can also be found in cases in which participants
do not ‘notice’ the routine meanings associated with the use of grammatical
gender.
In her reply to Melita’s question, Evagelia refers to self and other female co-par-
ticipants via the first person plural. By employing the grammatically feminine
156 chapter 6
Elsa asks her co-participants when they would like to go to the free pass movies
(lines 1–2). She selects them as next speakers via the second person plural verb
θέλετε ([θélete] ‘you want’), and describes them and herself via the masculine
adjective όλοι ([óli] ‘all’). Thus, she employs the masculine gender to refer to
more than one female referent. The next turns-answers produced by Marina
and Natasa show that female participants interpreted successfully the social
action that Elsa’s turn was designed to accomplish and that they were able to
‘see’ themselves through the inference of human/universal produced by the
masculine grammatical gender. In this way, next turns in this conversational
episode indicate implicitly the correlation betweeen the masculine gram-
matical gender and the interpretation of referents as male-human-gender
indefinite.
The same applies to Extract 10, which was analyzed in section 6.2.2. Melita
addresses a question to her female and male co-participants (lines 5–6),
employing the masculine grammatical gender to refer to them, and Zoi deliv-
ers an answer (lines 7–8). Her next turn shows that she understood herself as
one of the recipients to be addressed and selected as possible next speaker. In
other words, she understood the referent described by the masculine adjective
as not exclusively male.
Indications of the correlation between grammatical gender and the
intepretation of referent’s sex can also be found in the recipient-design fea-
ture of turns. As I mentioned in the beginning of this chapter, turns are always
recipient-designed, that is, speakers design their turns in such a way as to make
them appropriate and relevant for recipients. If we go back to Extracts 7, 8 and
9, we can make the following remark. In these extracts, speakers assess recipi-
ents and use gender membership categories, which match recipients’ sex. In
this way, speakers make their turns appropriate for the specific recipients they
address and show indirectly their interpretation of recipients as women or
men. Extracts 10 and 16 are also interesting from this perspective, because they
implicitly uncover the conceptual processes associated with the generic use of
the masculine. In these extracts, speakers employ linguistic items marked by
the masculine grammatical gender to achieve reference to females plus males,
and, thus, show their understanding of the masculine gender as a tool for acti-
vating the inference of human/universal. The same understanding is uncov-
ered by the use of the masculine gender for reference to female referents only
that was examined in Extracts 16, 17, 18 and 25.
When the generic use of the masculine is found in collective and generic
or indefinite self-reference and reference to recipients, it is easy for analysts
to claim that the inference of human/universal is activated, given that refer-
ents are present at context. Yet, this is not the case in non-recognitional third
158 chapter 6
person reference, as it is not always clear whether referents are exclusively male
or not (cf. section 4.2.1). For instance, if interlocutors refer to a stereotypically
masculine activity, it is very likely that the referents denoted by the masculine
reference form will be interpreted as exclusively male, rather than as male plus
female. In the following two extracts, I show that the membership categoriza-
tion device can provide a partial solution to the ‘ambiguity’ that accompanies
the generic use of the masculine in non-recognitional third person reference.
In Extract 26, Yanis gives advice to his friend Spyros, who is about to join
the army.
16 (.)
17 Niki → Δεν έφευγες με γκανέναν?
neg leave.2sg.imperf with anyone.acc.masc
Didn’t you leave with anyone.masc?
18 Yanis Τς. °μόνος.
Ts [negation]. [I left] alone.masc.
19 (1.3)
20 Yanis → Αλλά γνώρισα πολλούς.
but meet.1sg.past many.acc.masc
But I met many.masc.
Niki shifts topic at line 13, by asking Yanis whether he knew anyone when he
joined the army, and what he was doing with regard to smoking. She addresses
him via the second person singular verbs έκανες ([ékanes] ‘you did’), ήξερες
([íkseres] ‘you knew’) and the pronoun εσύ ([esí] ‘you’). She employs the mas-
culine interrogative pronoun ποιον ([pçon] ‘who’) and the masculine indefi-
nite pronoun κανένα ([kanéna] ‘anyone’) for generic reference. At line 17, she
addresses another question to Yanis. She employs the second person singular
verb έφευγες ([éfevʝes] ‘were leaving’) to refer to Yanis and the masculine indefi-
nite pronoun γκανέναν ([ganénan] ‘anyone’) for generic reference. In his reply
at line 20, Yanis employs the masculine adjective πολλούς ([polús] ‘many’) as a
non-recognitional reference form. In this conversational episode, participants
refer to a category-bound activity, that is, military service, which is stereotypi-
cally associated with men in the Greek context. Referring to the army can intro-
duce the relevance of the category ‘man’ in interaction and, thus, favors the
interpretation of referents as exclusively male. Note that this interpretation is
not shown in the data of interaction. It is an assumption that we analysts can
make, taking membership categorization device into account.
A similar example is found in the following extract. However, in this conver-
sational episode, interaction data provide an indication of the interpretation
of referent’s sex in speaker’s next turn. Melita describes her life in Athens in
the first year of her undergraduate studies and how she was sexually harassed
on the street.
At lines 9–10, Melita describes the fear that she felt when she was in public
spaces at night and, thus, she refers to a stereotypically feminine behavior. At
lines 12–13, she gives an account for her fear: she used to run into people who
Empirical Investigation of Grammatical Gender in Interaction 161
threatened her safety. She refers to these people via the non-recognitional
reference forms κάποιο πρεζόνι ([kápço prezóni] ‘a junkie’) and κάποιο:: α:νώμαλος
([kápço:: anómalos] ‘a sexual predator’). The grammatical marking of male
sex in the second form, plus its lexical meaning, activate a cultural stereotype
according to which men—who have some sort of disorder—usually harasss
women. This stereotype, in combination with the stereotype of the insecure
and vulnerable woman that has been previously introduced in interaction,
guides recipients to interpret referent as exclusively male. This interpretation
is displayed in Evagelia’s next turn at line 15. Evagelia responds with an ironic
comment, characterizing the incident that Melita describes as luck. Thus, she
interprets the behavior of the third person as sexual desire for Melita. Based on
the norm of compulsory heterosexuality, men are the ones expected to have
sexual desire for women.
However, it is not always easy to resolve the ambiguity associated with the
generic use of the masculine in third person reference in interaction. In the
next extract, Dimitris, Melita and Tania describe the problems of the public
sector in Greece. They use masculine non-recognitional third person reference
forms at lines 5, 8 and 9.
8 Dimitris → γιατί αυτός που θα ’ναι εκεί πάνω, που θα περιμένεις >εσύ<
because he.nom pron.rel will cop.3sg there up conj fut
wait.2sg you
because the one.masc who will be up there [in a position of
power], while you are waiting
9 → να κάνεις, >αυτός θα τα παίρνει.<
subj do.2sg he.nom fut them get.3sg.pres
to do something, he will be profiting [from the public sector]
Μelita asserts that the public sector cannot improve if people do not under-
take a collective form of action. At line 1, she refers to self collectively via the
first person plural κάνουμε ([kánume] ‘we do’). This collectivity includes her-
self, co-participants and all members of the community in general. At line 5,
she employs the grammatically masculine referring expression ο καθένας μόνος
του ([o kaθénas mónos tu] ‘each one on his own’) to maintain reference to
the same collectivity, that is, a gender indefinite collectivity. However, things
become fuzzier when it comes to the deployment of the masculine items at
lines 8–9. Dimitris employs the masculine pronoun αυτός ([aftós] ‘he’) twice
to refer to any individual who holds a position of power in the public sector.
Given that superior social position is usually associated with men, referents
may be interpreted as exclusively male, rather than male plus female.
Finally, an indirect indication of the correlation between grammatical gen-
der and the interpretation of referent’s sex is provided by the speaker’s switch
from feminine to masculine gender in Extract 29. This switch looks similar to
self-repair, because it involves the replacement of a feminine item with the
equivalent masculine. However, the mechanism of repair is not employed.
Melita, Evagelia, Tania and Zoi describe the difficulties of living together with
another person. Melita finds it difficult to share space with someone else after
having lived on her own for so long.
5 Melita Μμ.
Mm.
6 (1.5)
7 Tania Όχι. εμείς τέτοια προβλήματα δεν είχαμε ποτέ.
((noise . . .
No. we never had such problems.
8 (.)
. . .
9 Melita Και τώρα που μένω: μό:νη μου:, στην εστία:, (1.2)
. . . . . . . . . . . .))
((noise during the silence))
And now that I live on my own.fem, in the dorm, (1.2)
10 → >τέλος πάντων. μόνη μου. πώς μπορείς να το θεωρείς μόνη σου
anyhow on.my.own.nom.fem how can.2sg.pres subj it.acc
think.2sg.ipfv on.your.own.fem
anyhow. on my own.fem. how can you think of yourself as
being on your own.fem
11 στην εστία.< έχεις τόσα άτομα. ↑αλλά στο χώρο σου
in the dorm. you have so many people [around]. but in your
space
12 → είσαι μόνος σου εν πάση περιπτώσει. (.) φαντάζομαι ότι πολύ
cop.2sg.pres on.your.own.masc anyway reckon.1sg.pres
conj very
you are on your own.masc anyway. (.) I reckon that [it would
be] very
13 δύσκολα θα συγκατοικούσα με κάποιον. γιατί μαθαίνεις
hardly fut live.together.1SG.IMPERF with someone.acc.
masc because learn.2sg.pres
difficult [ for me] to share the house with someone else.
because you learn to
14 να είσαι μό:νος σου. [κάνει]ς ό,τι θες.
((noise. . . . . . . .))
subj cop.2sg on.your.own.masc do.2sg.pres pron.rel
want.2sg.pres
be on your own.masc. you do whatever you want.
15 Zoi [ Μ:. ]
Mm:.
16 Zoi °Παιδιά είναι πολύ καλό αυτό.
Hey guys that’s very good.
164 chapter 6
At line 9, Melita refers to self via the first person singular verb μένω: ([méno] ‘I
stay’), and describes herself via the feminine pronoun μό:νη μου ([móni mu] ‘on
my own’). At line 10, she maintains individual self-reference (μόνη μου ‘on my
own’) but in the following tcus she shifts to generic second person singular
(μπορείς να το θεωρείς μόνη σου [borís na to θeorís móni su] ‘you can think of your-
self as being on your own’) and she refers to herself plus all other students liv-
ing in the dorm. Melita employs the same feminine pronoun and then replaces
it with the masculine pronoun (μόνος σου ‘on your own’). She keeps using the
masculine gender till the end of her turn (γιατί μαθαίνεις να είσαι μό:νος σου [ʝatí
maθénis na íse mó:nos su] ‘because you learn to be on your own’). What is the
problem here? The speaker seems to be restricted by the use of the feminine
grammatical gender at this particular moment in interaction that categorizes
the referent as exclusively female. She uses the masculine gender in order to
activate the inference of human/universal and ensure that generalization or
indefiniteness is expressed. The switch of grammatical gender found in this
conversational episode uncovers indirectly the conceptual processes linked
with the use of the masculine grammatical gender.
6.3.4 Conclusions
The analysis presented in sections 6.3.2 and 6.3.3 shows that grammatical gen-
der guides speakers to interpret referent(s) as female or male. Overall, indi-
cations of the cognitive role of grammatical gender in person reference in
interaction are found when participants recognize or notice the codification
of referent’s sex via grammatical gender, as well as when the codification of sex
via grammatical gender passes unnoticed. Indications of the first kind can be
described as overt and indications of the second kind as covert.
Overt indications are found in self-repairs and exposed and embedded cor-
rections of grammatical gender. Repairs involve switching from one grammati-
cal gender to another and address problems in the interpretation of referent’s
sex encountered by speaker or recipient. The source of trouble is related to
aspects of context, which make the codification of sex via grammatical gen-
der important for what participants do in interaction. Speakers suspend the
progressivity of the turn (in self-repairs) or the sequence (in exposed correc-
tion) in order to modify the interpretation of referent’s sex that is denoted or
described by linguistic items marked by grammatical gender. In addition, self-
repairs of grammatical gender ensure that the action that the speaker’s turn is
designed to accomplish will be interpreted as such by recipients. Self-repairs
occur in an increment after the pragmatic and grammatical completion of a
tcu or within the same tcu, and exposed and embedded corrections occur in
next turn. In other words, repairs of grammatical gender occur after the item
Empirical Investigation of Grammatical Gender in Interaction 165
marked by grammatical gender was produced and the sex of referent was inter-
preted by interlocutors. In this way, repairs can be interpreted by analysts as
indications of the role of grammatical gender in guiding speakers’ thinking of
referent(s) as male or female, and for the conceptual processes that take place
during the production and understanding of turns in interaction. The majority
of the repairs examined in section 6.3.2 involve the generic use of the mascu-
line (see Extracts 20, 21, 22, 23, 24). More specifically, the self-repair examined
in Extract 20 shows that the masculine gender can produce the inference of
human/universal. This inference is based on the male categorization of refer-
ents, as self-repairs and corrections in Extracts 21, 22, 23 and 24 show.
Covert indications of the cognitive role of grammatical gender are found
in i) next turns, ii) turn’s recipient-design, iii) the membership categoriza-
tion device, iv) the mismatch between grammatical gender and referents’ sex,
v) and the switch of grammatical gender. These indications point to the cogni-
tive role of grammatical gender only indirectly. They can be inferred by us ana-
lysts, based on the knowledge that we have about the way in which interaction
is organized and turns are designed.
CHAPTER 7
To sum up, the present study has shown how the interrelation between the
cultural and cognitive aspects of grammatical gender manifests itself in inter-
action. In the last chapter, I present a series of reflections on the complex
interplay between grammatical gender, culture and cognition, based on the
empirical analysis presented in the previous chapter. These reflections con-
cern three issues: i) the relation between language and cognition in interaction
(section 7.1), ii) the covert reproduction of sexism in interaction (section 7.2),
and iii) the approach of gender performativity through the lens of the present
study (section 7.3). Certain reflections on interdisciplinarity in language and
gender research are also presented in the final section.
feminine gender. In these cases, the routine and covert meanings produced
in person reference in interaction come to the surface of the talk and are
noticed by speakers. More specifically, speakers recognize the automatic and
unconscious conceptual processes related with the use of grammatical gender
and modify them. In this sense, repairs show that language mediates speak-
ers’ thinking but does not ‘imprison’ speakers into static conceptualizations
of reality.1
Three points need to be highlighted here with respect to repairs of gram-
matical gender. First, the small number of repairs found in the corpus is justi-
fied by the fact that grammatical gender is inherent in the language system, is
used in a compulsory and systematic way, and guides speakers to unconscious
and automatic interpretations of experience (cf. chapters 2, 4). Based on that,
it should be expected that speakers rarely become aware of the cognitive role
of grammatical gender and notice the indexical effect associated with items
marked by grammatical gender. Second, the analysis of repairs shows that the
hidden and automatic meanings produced in interaction come to the surface
of the talk because aspects of context shed light on them. The information of
referent’s sex becomes relevant for what participants do in interaction and,
thus, it is foregrounded in the conversational episodes examined, and noticed
by participants. This is an important point that should be taken into account
by research on language and cognition. In exploring the ‘effect’ of language
structure on speakers’ cognition, the context in which this structure is put into
use should also be taken into consideration. In other words, we should explore
not only what speakers must say, but also when and why they say it. Third, it is
important to keep in mind that treating repairs of grammatical gender as indi-
cations of the cognitive role of grammatical gender is an interpretation made
by the analyst on the basis of the empirical analysis of interaction data. As long
as this interpretation is subjected to the limits of the theoretical and method-
ological perspectives adopted in this study, repairs cannot be said to ‘prove’ the
cognitive role of grammatical gender in an objective manner.
The findings reported in this study with respect to grammatical gender and
cognition raise an interesting question, given the scope of structural relativity
(cf. chapter 4). If the use of grammatical gender in Greek presupposes a par-
ticular kind of ‘thinking for speaking’, how does this sort of thinking differ from
‘thinking for speaking’ in languages with no gender systems? For example, is
1 This finding is opposed to linguistic determinism, which views language as a medium shap-
ing speakers’ thinking in absolute ways. If language had a deterministic impact on what
speakers actually do or on what they can think, then it would be impossible for them to
intervene in the mental representations constructed in interaction.
168 chapter 7
it possible that sex is a more salient concept for Greek speakers than Finnish
speakers? Or to phrase it differently, is it possible that the Greek language
guides speakers to a bipolar gender categorization of the world more system-
atically than the English language for example? As was shown in the analysis,
grammatical gender structures participants’ practices in interaction by cat-
egorizing referents as women or men and guiding speakers to attend to sex as
an aspect of context. Even if speakers wanted to escape from the static gen-
dered categorization of self and others, the language they speak enforces and
maintains this categorization on the basis of social hierarchy. These habits of
thinking associated with the use of grammatical gender should be absent from
languages in which sex is not grammaticized. If this is true, then genderless
languages should provide speakers with more possibilities for an egalitarian
representation of women and men via the use of gender-indefinite terms, as
Hellinger and Bussmann (2001, 20) note. Things, however, are not that simple.
As Whorf (1956) argues, covert classifications or cryptotypes may be far more
significant in terms of their impact on speakers’ cognition compared with
overt classifications or phenotypes. If Whorf’s idea is correct, then genderless
languages, in which sex is lexically specified or stereotypically assumed, may
guide speakers’ interpretation of the world in more systematic, unconscious
and significant ways than languages in which sex is grammaticized. In addi-
tion, as Hellinger and Bussmann (2001, 20) point out, it may be more difficult
to challenge sexism in genderless languages, because of the “covert male bias”
found in many nouns denoting humans (cf. section 4.3).
In this last section, I turn to the concept of gender performativity that was
introduced in chapter 3, and I present some preliminary remarks on how this
concept can be re-signified through the lens of the present study. I consider this
process of re-signification as an interdisciplinary process between linguistic
and non-linguistic approaches to gender. In my opinion, linguists need to take
this interdisciplinary step in order to gain a fuller understanding of the com-
plexities of the relation between language and gender. As Cameron (1996, 33)
insightfully points out, “language and gender studies, like other subfields
within sociolinguistics, has tended to neglect theoretical questions about its
‘socio’ side (in this instance, gender)”. As a consequence, Cameron (1996, 33)
claims that “linguists are cut off from insights that would be relevant to their
work, while feminists in other disciplines can continue to talk about language
in ways that are not accountable to the specialized knowledge linguistics
makes available”. In the following, I explore how the study presented in this
book contributes to the interdisciplinary dialogue between linguistic and non-
linguistic approaches.
Following Lykke (2004, 97), I understand interdisciplinarity as a process of
transgressing “borders between disciplinary canons and approaches in a theo-
retical and methodological bricolage that allows for new synergies to emerge”.
These synergies can emerge from the exchange or ‘travelling’ (Bal 2002) of con-
cepts between disciplines and their subsequent integration and transforma-
tion in different disciplinary contexts. Interdisciplinarity consists of moving in
and out of disciplines and creating hybrid spaces, in Latour’s (2000) terms, in-
between disciplines. The concept of performativity constitutes a typical case
of travelling from linguistics to feminist philosophy and back to linguistics.
The concept of performativity proposed by Butler has been central to
Austin’s speech act theory, which theorizes language as a tool for ‘doing things’
or performing actions (cf. section 3.4). Based on this idea, Butler argues that
gender is performative, in the sense that it constitutes an accomplishment, a
‘doing’ rather than a ‘being’, an outcome of repeated practices. Butler’s concept
of gender performativity informed research on language and gender. An origi-
nally linguistic concept returned to linguistics in a slightly modified version,
with a renewed feminist philosophical content. The concept was integrated
172 chapter 7
This means that different linguistic practices, oral or written, give rise to dif-
ferent ways of experiencing the world and, therefore, construct different sorts
of subjectivities. The aim of this book is not to provide guidelines on exactly
how this can be done. But it certainly invites feminists to think why language
should be one of the key issues to be raised in our political agenda.
What Can We Conclude ? 175
The study presented in this book shows that the use of grammatical gender is
part of the practices that speakers employ to construct themselves and others
as women or men, and maintain gender bipolarity on the basis of social hierar-
chy. Grammatical gender contributes to the routine achievement of sociocul-
tural gender in interaction. This process is materialized via covert assumptions
that speakers share about referent’s gender. In addition, grammatical gender
is shown to guide speakers to interpret person as female or male, when they
think for speaking. Indications of the cognitive role of grammatical gender in
person reference are found in the interaction data, in participants’ publicly
displayed, intersubjective understandings.
References
Aikhenvald, Alexandra Y. 1999. “The Arawak Language Family.” In The Amazonian Lan-
guages, edited by R.M.W. Dixon and Alexandra Y. Aikhenvald, 65–106. Cambridge:
Cambridge University Press.
———. 2000. Classifiers: A Typology of Noun Categorization Devices. Oxford: Oxford
University Press.
———. 2004. Evidentiality. Oxford: Oxford University Press.
———. 2006. “Classifiers and Noun Classes: Semantics.” In Encyclopedia of Languages
and Linguistics, edited by Keith Brown, 463–471. Oxford: Elsevier.
———. 2011. “Speech Reports: A Cross-linguistic Perspective.” In Language at Large:
Essays on Syntax and Semantics, edited by Alexandra Y. Aikhenvald and R.M.W. Dixon,
290–326. Leiden: Brill.
———. 2012. “Round Women and Long Men: Shape and Size in Gender Choice in
Papua New Guinea and Beyond.” Anthropological Linguistics 54(1): 33–86.
Alpher, Barry. 1987. “Feminine as the Unmarked Grammatical Gender: Buffalo Girls are
no Fools.” Australian Journal of Linguistics 7: 169–187.
Althusser, Louis. 1984. Essays on Ideology. London: Verso.
Alvanoudi, Angeliki. 2008. “Η Σχέση Γλώσσας και Φύλου στις Φεμινιστικές Θεωρίες της
Αποδόμησης: Η Περίπτωση του Γραμματικού Γένους” [The Relationship between Lan-
guage and Gender in Feminist Deconstruction Theories: The Case of Grammatical
Gender]. Studies in Greek Linguistics 28: 89–99.
———. (forthcoming). The routine achievement of gender in Greek conversation
[Special issue]. Gender and Language.
Anastasiadi-Symeonidi, Anna, and Despina Chila-Markopoulou. 2003. “Συγχρονικές και
Διαχρονικές Τάσεις στο Γένος της Ελληνικής: Μια Θεωρητική Πρόταση” [Synchronic
and Diachronic Trends in Gender of Modern Greek: A Theoretical Proposition]. In
Gender, edited by Anna Anastasiadi-Symeonidi, Angeliki Ralli and Despina Chila-
Markopoulou, 13–56. Athens: Patakis.
Andermahr, Sonya, Terry Lovell, and Carol Wolkowitz. 2000. A Glossary of Feminist
Theory. London: Arnold.
Asher, R.E. 1985. Tamil. London: Croom Helm.
Austin, John L. 1962. How to Do Things with Words. Oxford: Oxford University Press.
Austin, Peter. 1981. A Grammar of Diyari, South Australia. Cambridge: Cambridge Uni-
versity Press.
Bal, Mieke. 2002. Travelling Concepts in the Humanities: A Rough Guide. Toronto: Uni-
versity of Toronto Press.
Barcelona, Antonio. 2000. “Introduction: The Cognitive Theory of Metaphor and
Metonymy.” In Metaphor and Metonymy at the Crossroads: A Cognitive Perspective,
edited by Antonio Barcelona, 1–28. Berlin/New York: Mouton de Gruyter.
178 references
Beauvoir, Simone de. [1949] 1972. The Second Sex. Translated and edited by Howard M.
Parshley. Harmondsworth: Penguin.
Bergvall, Victoria L., Janet M. Bing, and Alice F. Freed, eds. 1996. Rethinking Language
and Gender Research: Theory and Practice. London: Longman.
Berman, Ruth A., and Dan Isaac Slobin. 1994. Relating Events in Narrative: A Crossling-
uistic Developmental Study. Hillsdale, nj: Lawrence Erlbaum.
Bickel, Balthasar. 2000. “Grammar and Social Practice: The Role of ‘Culture’ in Linguis-
tic Relativity.” In Evidence for Linguistic Relativity, edited by Susanne Niemeier and
René Dirven, 161–191. Amsterdam/Philadelphia: John Benjamins.
Boas, Franz. [1911] 1991. Introduction to Handbook of American Indian Languages.
Edited by Preston Holder, 1–79. Lincoln: University of Nebraska Press.
———. 1938. “Language.” In General Anthropology, edited by Franz Boas, 124–145.
Boston, New York: D.C. Heath and Company.
Borba, Rodrigo, and Ana C. Ostermann. 2007. “Do Bodies Matter? Travestis’ Embodi-
ment of (Trans)gender Identity through the Manipulation of the Brazilian Portug-
ese Grammatical System.” Gender and Language 1(1): 131–147.
Boroditsky, Lera, Lauren A. Schmidt, and Webb Phillips. 2003. “Sex, Syntax, and Seman-
tics.” In Language in the Mind: Advances in the Study of Language and Thought, edited
by Dedre Gentner and Susan Goldin-Meadow, 61–79. Cambridge, ma: mit Press.
Bowerman, Μelissa, and Stephen C. Levinson. 2001. Language Αcquisition and Concep-
tual Development. Cambridge: Cambridge University Press.
Braidotti, Rosi. 1991. Patterns of Dissonance: A Study of Women in Contemporary Philoso-
phy. Oxford: Polity.
———. 1994. Nomadic Subjects: Embodiment and Sexual Difference in Contemporary
Feminist Theory. New York: Columbia University Press.
———. 2000. “The Relevance of Gender for the Issue of ‘Women and Science’.” In The
Making of European Women’s Studies: A Work in Progress Report on Curriculum
Development and Related Issues, Vol. I, edited by Rosi Braidotti and Esther Vonk,
187–201. ATHENA/Utrecht University: Utrecht.
———. 2002a. “The Uses and Abuses of the Sex/Gender Distinction in European Fem-
inist Practices.” In Thinking Differently: A Reader in European Women’s Studies,
edited by Rosi Braidotti and Gabriele Griffin, 285–307. London/New York: Zed
Books.
———. 2002b. Metamorphoses: Towards a Materialist Theory of Becoming. Cambridge:
Polity Press.
———. 2003. “Feminist Philosophies.” In A Concise Companion to Feminist Theory,
edited by Mary Eagleton, 195–214. Oxford: Blackwell.
Braun, Friederike. 2001. “The Communication of Gender in Turkish.” In Gender across
Languages: The Linguistic Representation of Women and Men, edited by Marlis
Hellinger and Hadumod Bussmann, 283–310. Amsterdam: John Benjamins.
References 179
Braun, Friederike, Anja Gottburgsen, Sabine Sczesny, and Dagmar Stahlberg. 1998.
“Können Geophysiker Frauen sein? Generische Personenbezeichnungen im
Deutschen.” Zeitschrift für Germanistische Linguistik 26: 265–283.
Brown, Penelope. 2007. “Principles of Person Reference in Tzeltal Conversation.” In
Person Reference in Interaction: Linguistic, Cultural and Social Perspectives, edited by
N.J. Enfield and Tanya Stivers, 172–202. Cambridge: Cambridge University Press.
Brown, Penelope, and Stephen C. Levinson. 1987. Politeness: Some Universals in Lan-
guage Usage. Cambridge/New York: Cambridge University Press.
Bucholtz, Mary, and Kira Hall. 2005. “Identity and Interaction: A Sociocultural Linguis-
tic Approach.” Discourse Studies 7(4–5): 584–614.
Bussmann, Hadumod, and Marlis Hellinger. 2003. “Engendering Female Visibility in
German.” In Gender across Languages: The Linguistic Representation of Women and
Men, edited by Marlis Hellinger and Hadumod Bussmann, 141–174. Amsterdam:
John Benjamins.
Butler, Judith. [1990] 1999. Gender Trouble: Feminism and the Subversion of Identity.
London/New York: Routledge.
———. 1993. Bodies that Matter: On the Discursive Limits of Sex. New York: Routledge.
———. 1997. Excitable Speech: A Politics of the Performative. London: Routledge.
Cameron, Deborah. 1985. Feminism and Linguistic Theory. London: Macmillan.
———. [1995] 2005. Verbal Hygiene. London/New York: Routledge.
———. 1996. “The Language-Gender Interface: Challenging Co-optation.” In Rethink-
ing Language and Gender Research: Theory and Practice, edited by Victoria L.
Bergvall, Janet M. Bing and Alice F. Freed, 31–53. London: Longman.
Chila-Markopoulou, Despina. 2003. “Γένος και Συμφωνία στη Νέα Ελληνική” [Gender and
Agreement in Modern Greek]. In Gender, edited by Anna Anastasiadi-Symeonidi,
Angeliki Ralli and Despina Chila-Markopoulou, 132–167. Athens: Patakis.
Chomsky, Noam. 1965. Aspects of the Theory of Syntax. Cambridge, ma: mit Press.
Christidis, Anastasios Ph. 1999. Γλώσσα, Πολιτική, Πολιτισμός [Language, Politics, Cul-
ture]. Athens: Νisos.
Clark, Herbert H. 1996. Using Language. Cambridge: Cambridge University Press.
Clarke, Mark A., Ann Losoff, Margaret Dickenson McCracken, and JoAnn Still. 1981.
“Gender Perception in Arabic and English.” Language Learning 31: 159–169.
Claudi, Ulrike. 1985. Zur Entstehung von Genussystemen: Überlegungen zu einigen Theo-
retischen Aspekten, Verbunden mit einer Fallstudie des Zande: Mit einer Bibliographie
und einer Karte. Hamburg: Buske.
Clift, Rebecca, and Elizabeth Holt. 2007. “Introduction.” In Reporting Talk: Reported
Speech in Interaction, edited by Elizabeth Holt and Rebecca Clift, 1–15. Cambridge:
Cambridge University Press.
Connell, Raewyn. 2002. Gender. Cambridge: Polity Press.
Corbett, Greville, G. 1991. Gender. Cambridge: Cambridge University Press.
180 references
Corpus of Spoken Greek of the Institute of Modern Greek Studies. Accessed June 12, 2014.
http://corpus-ins.lit.auth.gr/corpus/about/symbols.html.
Croft, William, and D. Alan Cruse. 2004. Cognitive Linguistics. Cambridge: Cambridge
University Press.
Curzan, Anne. 2003. Gender Shifts in the History of English. Cambridge: Cambridge Uni-
versity Press.
Derrida, Jacques. [1972] 1988. “Signature, Event, Context.” In Limited, Inc, edited by Ger-
ald Graff. Translated by Samuel Weber and Jeffrey Mehlman. Evanston: Northwest-
ern University Press.
Dictionary of Standard Modern Greek. 1998. Thessaloniki: Institute of Modern Greek
Studies (Manolis Triantafyllidis Foundation).
Dixon, R.M.W. 1972. The Dyirbal Language of North Queensland. London: Cambridge
University Press.
———. 1980. The Languages of Australia. Cambridge: Cambridge University Press.
———. 2002. Australian Languages: Their Nature and Development. Cambridge: Cam-
bridge University Press.
———. 2010a. Basic Linguistic Theory. Vol. 1. Methodology. Oxford: Oxford University
Press.
———. 2010b. Basic Linguistic Theory. Vol. 2. Grammatical Topics. Oxford: Oxford Uni-
versity Press.
Doleschal, Ursula. 1993. “Genus als Grammatische und Textlinguistische Kategorie. Eine
Kognitiv-funktionalistische Untersuchung des Russischen”. Ph.D. diss., University of
Vienna.
———. 1997. “O Vzaimosvjazi Grammatičeskoj Kategorii Roda i Pola” [On the Interde-
pendence of the Grammatical Category Gender and Sex]. In Verbal and Non-Verbal
Interpretation of Gender Characteristics, edited by Aleksandr Xolod, 134–155. Krivoj
Rog: Meždunarodnyj Issledovatel’skij Centr Čelovek-Jazyk-Kul’tura-Poznanie.
Doleschal, Ursula, and Sonja Schmid. 2001. “Doing Gender in Russian: Structure and
Perspective.” In Gender across Languages: The Linguistic Representation of Women
and Men, edited by Marlis Hellinger and Hadumod Bussmann, 253–282. Amster-
dam: John Benjamins.
Drew, Paul, Traci Walker, and Richard Ogden. 2013. “Self-repair and Action Construc-
tion.” In Conversational Repair and Human Understanding, edited by Makoto
Hayashi, Geoffrey Raymond, and Jack Sidnell, 71–94. Cambridge: Cambridge Uni-
versity Press.
Eakins, Barbara W., and Rollin G. Eakins. 1978. Sex Differences in Human Communica-
tion. Boston: Houghton Mifflin.
Eckert, Penelope, and Sally McConnell-Ginet. 1992. “Think Practically and Look
Locally: Language and Gender as Community-Based Practice.” Annual Review of
Anthropology 21: 461–490.
References 181
Freed, Alice F. 2003. “Epilogue: Reflections on Language and Gender Research.” In The
Handbook of Language and Gender, edited by Janet Holmes and Miriam Meyerhoff,
699–721. Oxford: Blackwell.
Garfinkel, Harold. 1967. Studies in Ethnomethodology. Englewood Cliffs, N.J.: Prentice-
Hall.
Gentner, Dedre, and Susan Goldin-Meadow, eds. 2003. Language in the Mind: Advances
in the Study of Language and Thought. Cambridge, ma: mit Press.
Gleitman, Lila, and Anna Papafragou. 2005. “Language and Thought.” In Cambridge
Handbook of Thinking and Reasoning, edited by Keith J. Holyoak and Robert G.
Morrison, 633–661. Cambridge: Cambridge University Press.
Goffman, Erving. 1981. Forms of Talk. Oxford: Blackwell.
———. 1983. “The Interaction Order: American Sociological Association, 1982 Presi-
dential Address.” American Sociological Review 48(1): 1–17.
Graham, Alma. 1975. “The Making of a Nonsexist Dictionary.” In Language and Sex: Dif-
ference and Dominance, edited by Barrie Thorne and Nancy Henley, 57–63. Rowley,
ma: Newbury House.
Grice, H. Paul. 1989. Studies in the Way of Words. London/Cambridge, ma.: Harvard Uni-
versity Press.
Grönberg, Anna G. 2002. “Masculine Generics in Current Icelandic.” In Gender across
Languages: The Linguistic Representation of Women and Men, edited by Marlis
Hellinger and Hadumod Bussmann, 163–185. Amsterdam: John Benjamins.
Gumperz, John J. 1996. “The Linguistic and Cultural Relativity of Conversational Infer-
ence.” In Rethinking Linguistic Relativity, edited by John J. Gumerz and Stephen C.
Levinson, 374–406. Cambridge: Cambridge University Press.
Gumperz, John J., and Stephen C. Levinson, eds. 1996a. Rethinking Linguistic Relativity.
Cambridge: Cambridge University Press.
———. 1996b. “Introduction: Linguistic Relativity Re-examined.” In Rethinking Lin-
guistic Relativity, edited by John J. Gumperz and Stephen C. Levinson, 1–18. Cam-
bridge: Cambridge University Press.
Hall, Kira, and Veronica O’Donovan. 1996. “Shifting Gender Positions among Hindi-
speaking Hijras.” In Rethinking Language and Gender Research: Theory and Practice,
edited by Victoria L. Bergvall, Janet M. Bing and Alice F. Freed, 228–266. London:
Longman.
Hanks, William F. 1990. Referential Practice: Language and Lived Space among the
Maya. Chicago: University of Chicago Press.
———. 1996. “Language Form and Communicative Practices.” In Rethinking Linguistic
Relativity, edited by John J. Gumperz and Stephen C. Levinson, 232–270. Cambridge:
Cambridge University Press.
———. 2007. “Person Reference in Yucatec Maya Conversation.” In Person Reference in
Interaction: Linguistic, Cultural and Social Perspectives, edited by N.J. Enfield and
Tanya Stivers, 149–171. Cambridge: Cambridge University Press.
References 183
Haraway, Donna. 1991. “Situated Knowledges: The Science Question in Feminism and
the Privilege of Partial Perspective.” In Simians, Cyborgs, and Women: The Reinven-
tion of Nature, 183–201. London: Free Association Books.
Harding, Sandra. 1991. “What is Feminist Epistemology?” In Whose Science? Whose
Knowledge, 105–137. Open up.
Haviland, John B. 2007. “Person Reference in Tzotzil Gossip: Referring Dupliciter.” In
Person Reference in Interaction: Linguistic, Cultural and Social Perspectives, edited
by N.J. Enfield and Tanya Stivers, 226–252. Cambridge: Cambridge University
Press.
Hayashi, Makoto, Geoffrey Raymond, and Jack Sidnell, eds. 2013. Conversational Repair
and Human Understanding. Cambridge: Cambridge University Press.
Heine, Bernd. 1982. “African Noun Class Systems.” In Apprehension: Das sprachliche
Erfassen von Gegenständen, i: Bereich und Ordnung der Phänomene, edited by
Hansjakob Seiler and Christian Lehmann, 189–216. Tübingen: Narr.
Hellinger, Marlis. 2001. “English: Gender in a Global Language.” In Gender across Lan-
guages: The Linguistic Representation of Women and Men, edited by Marlis Hellinger
and Hadumod Bussmann, 105–113. Amsterdam: John Benjamins.
Hellinger, Marlis, and Hadumod Bussmann. 2001. “Gender across Languages: The Lin-
guistic Representation of Women and Men.” In Gender across Languages: The Lin-
guistic Representation of Women and Men, edited by Marlis Hellinger and Hadumod
Bussmann, 1–25. Amsterdam: John Benjamins.
———, eds. 2001–2002–2003. Gender across Languages: The Linguistic Representation
of Women and Men. Amsterdam: John Benjamins.
Heritage, John. 1984. Garfinkel and Ethnomethodology. Cambridge: Polity Press.
———. 1988. “Explanations as Accounts: A Conversation Analytic Perspective.” In
Analyzing Everyday Explanation: A Casebook of Methods, edited by Charles Antaki,
127–144. London: Sage.
Heritage, John, and J. Maxwell Atkinson. 1984. “Introduction.” In Structures of Social
Action: Studies in Conversation Analysis, edited by J. Maxwell Atkinson and John
Heritage, 1–15. Cambridge: Cambridge University Press.
Ηeritage, John, and Geoffrey Raymond. 2005. “The Terms of Agreement: Indexing Epis-
temic Authority and Subordination in Talk-in-Interaction.” Social Psychology Quar-
terly 68(1): 15–38.
Hill, Jane H., and Bruce Mannheim. 1992. “Language and World View.” Annual Review of
Anthropology 21: 381–406.
Holmes, Janet, and Miriam Meyerhoff. 2003. “Different Voices, Different Views: An
Introduction to Current Research in Language and Gender.” In The Handbook of
Language and Gender, edited by Janet Holmes and Miriam Meyerhoff, 1–17. Oxford:
Blackwell.
Holton, David, Peter Mackridge, and Irene Philippaki-Warburton. 1997. Greek: A Com-
prehensive Grammar. London: Routledge.
184 references
Horrocks, Geoffrey C. 2010. Greek: A History of the Language and its Speakers. Wiley-
Blackwell.
Hougaard, Anders. 2008. “Compression in Interaction.” In Mental Spaces in Discourse
and Interaction, edited by Todd Oakley and Anders Hougaard, 179–208. Amsterdam/
Philadelphia: John Benjamins.
Huang, Yan. 1994. The Syntax and Pragmatics of Anaphora. Cambridge: Cambridge
University Press.
———. 2007. Pragmatics. Oxford: Oxford University Press.
Hymes, Dell H. 1966. “Two Types of Linguistic Relativity (with Examples from Amerin-
dian Ethnography).” In Sociolinguistics, Proceedings of the ucla Sociolinguistics Con-
ference, edited by William Bright, 114–157. The Hague: Mouton.
Imai, Mutsumi, Lennart Schalk, Henrik Saalbach, and Hiroyuki Okada. 2013. “All Giraffes
have Female-specific Properties: Influence of Grammatical Gender on Deductive
Reasoning about Sex-specific Properties in German Speakers.” Cognitive Science 1–23.
Irigaray, Luce. 1985a. The Speculum of the Other Woman. Ithaca: Cornell University
Press.
———. 1985b. This Sex which is not One. Ithaca: Cornell University Press.
———. 1985c. Speaking is never Neutral. Paris: Editions de Minuit.
———. 1993. Je, Tu, Nous: Towards a Culture of Difference. New York: Routledge.
Ιrmen, Lisa, and Astrid Köhncke. 1996. “Zur Psychologie des ‘generischen’ Maskuli-
nums.” Sprache und Kognition 15: 152–167.
Jefferson, Gail. 1978. “Sequential Aspects of Storytelling in Conversation.” In Studies in
the Organization of Conversational Interaction, edited by Jim Schenkein, 219–248.
New York: Academic Press.
———. 1984. “On Stepwise Transition from Talk about a Trouble to inappropriately
Next-positioned Matters.” In Structures of Social Action: Studies in Conversation
Analysis, edited by J. Maxwell Atkinson and John Heritage, 191–221. Cambridge:
Cambridge University Press.
———. 1987. “On Exposed and Embedded Correction in Conversation.” In Talk and
Social Organization, edited by Graham Button and John R.E. Lee, 86–100. Clevedon,
uk: Multilingual Matters.
———. 1996. “On the Poetics of Ordinary Talk.” Text and Performance Quarterly 16(1):
1–61.
Joseph, Brian, and Irene Philipaki-Warburton. 1987. Modern Greek. London: Croom Helm.
Kati, Dimitra. 2013. Ο Γνωσιακός Ρόλος της Γλώσσας: Η Συμβολή των Επιστημών της Γλώσσας
και του Νου [The Cognitive Role of Language: The Contribution of the Studies on
Language and Mind]. Athens: Nisos.
Kati, Dimitra, Marianna Kondyli, and Kiki Nikiforidou. 1999. “Εισαγωγή” [Introduc-
tion]. In Language and the Mind, edited by Dimitra Kati, Μarianna Κοndyli and Kiki
Νikiforidou, 11–42. Αthens: Αlexandreia.
References 185
Kitagawa, Chisato, and Adrienne Lehrer. 1990. “Impersonal Uses of Personal Pronouns.”
Journal of Pragmatics 14(5): 739–759.
Kitzinger, Celia. 2000. “Doing Feminist Conversation Analysis.” Feminism Psychology
10(2): 163–193.
———. 2005. “Speaking as a Heterosexual: (How) Does Sexuality Matter for Talk-in-
Interaction.” Research on Language and Social Interaction 38(3): 221–265.
———. 2007. “Is ‘Woman’ always relevantly Gendered?” Gender and Language 1(1):
39–49.
———. 2012. “Repair.” In The Handbook of Conversation Analysis, edited by Jack Sidnell
and Tanya Stivers, 229–256. Boston: Wiley-Blackwell.
Klein, Josef. 1988. “Die Benachteiligung der Frau im Generischen Maskulinum—eine
Feministische Schimäre oder Psycholinguistische Realität?” In Vorträge des Ger-
manisten-tages Berlin 1987. Vol. 1: Das Selbstverständnis der Germanistik, edited by
Norbert Oellers, 310–319. Tübingen: Niemeyer.
Konishi, Toshi. 1993. “The Semantics of Grammatical Gender: A Cross-cultural Study”.
Journal of Psycholinguistic Research 22(5): 519–534.
Koniuszaniec, Gabriela, and Hanka Blaszkowska. 2003. “Language and Gender in Pol-
ish.” In Gender across Languages: The Linguistic Representation of Women and Men,
edited by Marlis Hellinger and Hadumod Bussmann, 259–285. Amsterdam: John
Benjamins.
Köpcke, Klaus-Michael, and David A. Zubin. 2003. “Metonymic Pathways to Neuter-
Gender Human Nominals in German.” In Metonymy and Pragmatic Inferencing,
edited by Klaus-Uwe Panther and Linda L. Thornburg, 149–166. Amsterdam/
Philadelphia: John Benjamins.
Köpcke, Klaus-Michael, Klaus-Uwe Panther, and David A. Zubin. 2010. “Motivating
Grammatical and Conceptual Gender Agreement in German.” In Cognitive Founda-
tions of Linguistic Usage Patterns, edited by Hans-Jörg Schmid and Susanne Handl,
171–194. Berlin/New York: Walter de Gruyter.
Kramarae, Cheris, and Paula A. Treichler. 1985. A Feminist Dictionary. London: Pandora
Press.
Kulick, Don. 1998. Travesti: Sex, Gender, and Culture among Brazilian Transgendered
Prostitutes. Chicago/London: University of Chicago Press.
Lakoff, George. 1987. Women, Fire and Dangerous Things: What Categories Reveal about
the Mind. Chicago/London: The University of Chicago Press.
Lakoff, Robin. 1975. Language and Woman’s Place. New York: Harper and Row.
Langacker, Ronald. W. 1993. “Reference-Point Constructions.” Cognitive Linguistics 4(1):
1–38.
Latour, Bruno. 1993. We Have never been Modern. Translated by Catherine Porter. Cam-
bridge, ma.: Harvard University Press.
Lee, Penny. 1996. The Whorf Theory Complex. Amsterdam/Philadelphia: John Benjamins.
186 references
Lerner, Gene H. 1992. “Assisted Story Telling: Deploying Shared Knowledge as a Practi-
cal Matter.” Qualitative Sociology 15(3): 247–271.
———. 2004. “Collaborative Turn Sequences.” In Conversation Analysis: Studies
from the First Generation, edited by Gene H. Lerner, 225–256. Amsterdam: John
Benjamins.
Lerner, Gene H., and Celia Kitzinger. 2007. “Extraction and Aggregation in the Repair
of Individual and Collective Self-reference.” Discourse Studies 9(4): 526–557.
Levelt, Willem. J.M. 1983. “Monitoring and Self-repair in Speech.” Cognition 14: 41–104.
———. 1989. Speaking: From Intention to Articulation. Cambridge, ma: mit Press.
Levinson, Stephen C. 1983. Pragmatics. Cambridge: Cambridge University Press.
———. 2000. Presumptive Meanings: The Theory of Generalized Conversational Impli-
cature. Cambridge, ma: mit Press.
———. 2003. Space in Language and Cognition: Explorations in Cognitive Diversity.
Cambridge: Cambridge University Press.
———. 2004. “Deixis.” In The Handbook of Pragmatics, edited by Laurence Horn and
Gergory Ward, 97–121. Oxford: Blackwell.
———. 2006. “Cognition at the Heart of Human Interaction.” Discourse Studies 8(1):
85–93.
———. 2007. “Optimizing Person Reference: Perspectives from Usage on Rossel
Island.” In Person Reference in Interaction: Linguistic, Cultural and Social Perspec-
tives, edited by N.J. Enfield and Tanya Stivers, 29–72. Cambridge: Cambridge Univer-
sity Press.
Lindström, Anna, and Marja-Leena Sorjonen. 2012. “Affiliation in Conversation.” In The
Handbook of Conversation Analysis, edited by Jack Sidnell and Tanya Stivers, 350–
369. Boston: Wiley-Blackwell.
Livia, Anna. 1997. “Disloyal to Masculinity: Linguistic Gender and Liminal Identity in
French.” In Queerly Phrased: Language, Gender and Sexuality, edited by Anna Livia
and Kira Hall, 349–368. Oxford/New York: Oxford University Press.
———. 2001. Pronoun Envy: Literary Uses of Linguistic Gender. Oxford/New York:
Oxford University Press.
Livia, Anna, and Kira Hall, eds. 1997. Queerly Phrased: Language, Gender and Sexuality.
Oxford/New York: Oxford University Press.
Lucy, John A. 1992. Language Diversity and Thought: A Reformulation of the Linguistic
Relativity Hypothesis. Cambridge: Cambridge University Press.
———. 1996. “The Scope of Linguistic Relativity: An Analysis and Review of Empirical
Research.” In Rethinking Linguistic Relativity, edited by John J. Gumperz and
Stephen C. Levinson, 37–69. Cambridge: Cambridge University Press.
Lykke, Nina. 2004. “Women’s/Gender/Feminist Studies: A Post-Disciplinary Disci-
pline?” In The Making of European Women’s Studies, Vol. V, edited by Rosi Braidotti,
Edyta Just and Marlise Mensink, 91–101. Utrecht: ATHENA.
References 187
———. 2010. Feminist Studies: A Guide to Intersectional Theory, Methodology and Writ-
ing. New York: Routledge.
Lykke, Nina, and Rosi Braidotti, eds. 1996. Between Monsters Goddesses and Cyborgs:
Feminist Confrontations with Science, Medicine and Cyberspace. London: zed.
Lyotard, Jean-François. 1984. The Postmodern Condition: A Report on Knowledge. Trans-
lated by Geoff Bennington and Brian Massumi, foreword by Fredric Jameson. Man-
chester: Manchester University Press.
Mackridge, Peter. 1985. The Modern Greek Language: A Descriptive Analysis of Standard
Modern Greek. Oxford: Oxford University Press.
Makri-Tsilipakou, Μarianthi. 1984. “Απόπειρα Περιγραφής της Νεοελληνικής Προσφώ‑
νησης” [An Αttempt to Describe the Forms of Address in Modern Greek]. Studies in
Greek Linguistics 4: 219–239.
———. 1989. “The Gender of Άνθρωπος: An Exercise in False Generics.” In Proceedings
of the 3rd Symposium on the Description and/or Comparison of English and Greek,
Aristotle University of Thessaloniki, School of English, Department of Theoretical
and Applied Linguistics, 61–83. Thessaloniki: University Studio Press.
Malone, Martin J. 1997. Worlds of Talk: The Presentation of Self in Everyday Conversation.
Cambridge: Polity Press.
Maltz, Daniel N., and Ruth A. Borker. 1982. “A Cultural Approach to Male/Female Mis-
communication.” In Language and Social Identity, edited by John J. Gumperz, 195–
216. Cambridge: Cambridge University Press.
Marcato, Gianna, and Eva-Maria Thüne. 2002. “Gender and Female Visibility in Italian.”
In Gender across Languages: The Linguistic Representation of Women and Men,
edited by Marlis Hellinger and Hadumod Bussmann, 186–217. Amsterdam: John
Benjamins.
Martyna, Wendy. 1983. “Beyond the He/Man Approach: The Case for Nonsexist Lan-
guage.” In Language, Gender and Society, edited by Barrie Thorne, Cheris Kramarae
and Nancy Henley, 25–37. Rowley, ma: Newbury House.
Mathiot, Madeleine. 1979. “Sex Roles as Revealed through Referential Gender in Ame
rican English.” In Ethnolinguistics: Boas, Sapir and Whorf Revisited, edited by
Madeleine Mathiot, 1–47. The Hague/Paris/New York: Mouton Publishers.
Mayer, Mercer. 1969. Frog, Where are You? New York: Dial Press.
McConnell-Ginet, Sally. 2003. “What’s in a Name?: Social Labeling and Gender Prac-
tices.” In The Handbook of Language and Gender, edited by Janet Holmes and
Miriam Meyerhoff, 69–97. Oxford: Blackwell.
———. 2011. “Language and Gender.” In Gender, Sexuality, and Meaning: Linguistic
Practice and Politics, 37–62. New York: Oxford University Press. Originally published
in Frederick J. Newmeyer, ed., Linguistics: The Cambridge Survey. iv: Language: The
Sociocultural Context (Cambridge: Cambridge University Press, 1988).
188 references
———. 2014. “Gender and its Relation to Sex: The Myth of ‘Natural’ Gender”. In The
Expression of Gender, edited by Greville G. Corbett, 3–38. Berlin: De Gruyter
Mouton.
McElhinny, Bonnie. 2003. “Theorizing Gender in Sociolinguistics and Linguistic
Anthropology.” In The Handbook of Language and Gender, edited by Janet Holmes
and Miriam Meyerhoff, 21–42. Oxford: Blackwell.
McKay, Donald G., and David C. Fulkerson. 1979. “On the Comprehension and Produc-
tion of Pronouns.” Journal of Verbal Learning and Verbal Behavior 18(6): 661–673.
Miller, Casey, and Kate Swift. [1981] 1988. The Handbook of Non-Sexist Writing for Writ-
ers, Editors and Speakers. The Women’s Press.
Μills, Anne E. 1986. The Acquisition of Gender: A Study of English and German. Berlin:
Springer-Verlag.
Mühlhäusler, Peter, and Rom Harré. 1990. Pronouns and People: The Linguistic Construc-
tion of Social and Personal Identity. Oxford: Blackwell.
Niemeier, Susanne, and René Dirven, eds. 2000. Evidence for Linguistic Relativity.
Amsterdam/Philadelphia: John Benjamins.
Nilsen, Alleen P. 1977. “Sexism as Shown through the English Vocabulary.” In Sexism
and Language, edited by Alleen P. Nilsen, Haig Bosmajian, H. Lee Gershuny and
Julia P. Stanley, 27–41. Urbana, Ill.: National Council of Teachers of English, U.S.A.
Nissen, Uwe K. 1997. “Do Sex-neutral and Sex-specific Nouns Exist: The Way to Non-
sexist Spanish?” In Kommunikation von Geschlecht [Communication of Gender],
edited by Friederike Braun and Ursula Pasero, 222–241. Pfaffenweiler: Centaurus.
———. 2002. “Gender in Spanish: Tradition and Innovation.” In Gender across Lan-
guages: The Linguistic Representation of Women and Men, edited by Marlis Hellinger
and Hadumod Bussmann, 251–279. Amsterdam: John Benjamins.
Ochs, Elinor. 1992. “Indexing Gender.” In Rethinking Context: Language as an Interactive
Phenomenon, edited by Alessandro Duranti and Charles Goodwin, 335–358. Cam-
bridge: Cambridge University Press.
Pastre, Geneviène. 1997. “Linguistic Gender Play among French Gays and Lesbians.” In
Queerly Phrased: Language, Gender and Sexuality, edited by Anna Livia and Kira
Hall, 369–379. Oxford/New York: Oxford University Press.
Pavlidou, Theodossia-Soula. 1985. “Παρατηρήσεις στα Θηλυκά Επαγγελματικά” [Remarks
on Feminine Occupational Terms]. Studies in Greek Linguistics 5: 201–217.
———. 2000. “Modern Greek Gender.” In The Making of European Women’s Studies: A
Work in Progress Report on Curriculum Development and Related Issues, Vol. I, edited
by Rosi Braidotti and Esther Vonk, 42–43. ATHENA/Utrecht University: Utrecht.
———. 2003. “Women, Gender and Modern Greek.” In Gender across Languages: The
Linguistic Representation of Women and Men, edited by Marlis Hellinger and Hadu-
mod Bussmann, 175–199. Amsterdam: John Benjamins.
References 189
Sacks, Harvey, Emanuel A. Schegloff, and Gail Jefferson. 1974. “A Simplest Systematics
for the Organization of Turn-taking for Conversation.” Language 50(4): 696–735.
Sapir, Edward. [1949] 1970. Selected Writings in Language, Culture and Personality.
Edited by David G. Mandelbaum. Berkley: University of California Press.
Schafroth, Elmar. 2003. “Gender in French.” In Gender across Languages: The Linguistic
Representation of Women and Men, edited by Marlis Hellinger and Hadumod Buss-
mann, 87–117. Amsterdam: John Benjamins.
Schegloff, Emanuel A. 1986. “The Routine as Achievement.” Human Studies 9: 111–151.
———. 1987. “Analyzing Single Episodes of Interaction: An Exercise in Conversation
Analysis.” Social Psychology Quarterly 50(2): 101–114.
———. 1988. “Description in the Social Sciences I: Talk in Interaction.” ipra Papers in
Pragmatics 2: 1–24.
———. 1991. “Conversation Analysis and Socially Shared Cognition.” In Perspectives
on Socially Shared Cognition, edited by Lauren B. Resnick, John M. Levine and
Stephanie D. Teasley, 150–171. Washigton, D.C: American Psychological Association.
———. 1996a. “Issues of Relevance for Discourse Analysis: Contingency in Action,
Interaction and Co-participant Context.” In Computational and Conversational Dis-
course: Burning Issues—An Interdisciplinary Account, edited by Eduard H. Hovy and
Donia R. Scott, 3–38. Heidelberg: Springer Verlag.
———. 1996b. “Some Practices for Referring to Persons in Talk-in-Interaction: A Partial
Sketch of a Systematics.” In Studies in Anaphora, edited by Barbara A. Fox, 437–85.
Amsterdam: John Benjamins.
———. 1997. “Whose Text? Whose Context?” Discourse and Society 8(2): 165–187.
———. 2004. Lectures on Repair and Categorization (244b/266 Conversational Struc-
tures ii), ucla.
———. 2006a. “On Possibles.” Discourse Studies 8(1): 141–157.
———. 2006b. “Interaction: The Infrastructure for Social Institutions, the Natural Eco-
logical Niche for Language, and the Arena in which Culture is Enacted.” In Roots of
Human Sociality: Culture, Cognition and Interaction, edited by N.J. Enfield and Ste-
phen C. Levinson, 70–96. Oxford/New York: Berg.
———. 2007a. Sequence Organization in Interaction. Cambridge: Cambridge Univer-
sity Press.
———. 2007b. “A Tutorial on Membership Categorization.” Journal of Pragmatics 39:
462–482.
———. 2007c. “Conveying Who You Are: The Presentation of Self, strictly Speaking.”
In Person Reference in Interaction: Linguistic, Cultural and Social Perspectives, edited
by N.J. Enfield and Tanya Stivers, 123–148. Cambridge: Cambridge University Press.
———. n.d. “Increments.” Manuscript, ucla.
192 references
Spender, Dale. 1980. Man Made Language. London: Routledge and Kegan Paul.
Stanley, Julia P. 1977. “Paradigmatic Woman: The Prostitute.” In Papers in Language
Variation, edited by David L. Shores and Carole P. Hines, 303–321. Tuscaloosa: The
University of Alabama Press.
Stivers, Tanya, and Jack Sidnell. 2012. “Introduction.” In The Handbook of Conversation
Analysis, edited by Jack Sidnell and Tanya Stivers, 1–8. Boston: Wiley-Blackwell.
Stivers, Τanya, N.J. Enfield, and Stephen C. Levinson. 2007. “Person Reference in Inter-
action.” In Person Reference in Interaction: Linguistic, Cultural and Social Perspec-
tives, edited by N.J. Enfield and Tanya Stivers, 1–20. Cambridge: Cambridge University
Press.
Stockill, Clare, and Celia Κitzinger. 2007. “Gendered ‘People’: How linguistically Non-
Gendered Terms Can Have Gendered Interactional Relevance.” Feminism Psychol-
ogy 17(2): 224–236.
Tannen, Deborah. 1990. You just Don’t Understand: Women and Men in Conversation.
New York: William Morrow.
Taylor, John R. 2003. Linguistic Categorization. New York: Oxford University Press.
Thorne, Barrie, and Nancy Henley, eds. 1975. “Difference and Dominance: An Overview
of Language, Gender, and Society.” In Language and Sex: Difference and Dominance,
5–42. Rowley, ma: Newbury House.
Τopsakal, Maria. 1995. The Influence of the Grammatical Gender on the Perception of
Gender in Modern Greek: Comparison with German and English. ma Thesis, Aristotle
University Thessaloniki.
Triantafyllidis, Manolis. [1941] 2005. Νεοελληνική Γραμματική (της Δημοτικής) [Modern
Greek Grammar (of the Demotic)]. Reprint with corrections of the 1941 edition.
Thessaloniki: Institute of Modern Greek Studies.
———. [1953] 1963. “H ‘Βουλευτίνα’ και ο Σχηματισμός των Θηλυκών Επαγγελματικών
Ουσιαστικών” [The ‘Parliament-member-fem’ and the Formation of Feminine
Occupational Terms]. In Άπαντα Μανόλη Τριανταφυλλίδη [Collected Works of Mano-
lis Triantafyllidis], Vol. 2, 326–334. Thessaloniki: Institute of Modern Greek Studies.
Tsangalidis, Anastasios. 2011. Greek. In The Languages and Linguistics of Europe: A Com-
prehensive Guide, edited by Bernd Kortmann and Johan van der Auwera, 223–239.
Berlin: De Gruyter Mouton.
Tsokalidou, Roula. 1996. Το Φύλο της Γλώσσας: Οδηγός μη Σεξιστικής Γλώσσας για τον
Δημόσιο Ελληνικό Λόγο [The Gender of Language: A Guide to Non-sexist Language for
Public Greek Speech]. Athens: Syndesmos Ellinidon Epistimonon.
van Dijk, Teun A. 2009. “Critical Discourse Studies: A Sociocognitive Approach.” In
Methods of Critical Discourse Studies, edited by Ruth Wodak and Michael Meyer,
62–85. London: Sage.
Weber, David John. 1989. A Grammar of Huallaga (Huánuco) Quechua. Berkley/Los
Angeles: University of California Press.
194 references
Whorf, Benjamin L. 1956. Language, Thought and Reality: Selected Writings of Benjamin
Lee Whorf. Edited by John B. Carroll. Cambridge, ma: mit Press.
Widerberg, Karin. 1998. “Translating Gender.” Nora 2(6): 133–138.
Wierzbicka, Anna. 2002. “Sexism in Grammar: The Semantics of Gender in Australian
English.” Anthropological Linguistics 44(2): 143–177.
Wilkinson, Sue. 2010. “Gender, Routinization and Recipient Design.” In Conversation
and Gender, edited by Susan A. Speer and Elizabeth Stokoe, 112–134. Cambridge:
Cambridge University Press.
Wittig, Monique. 1992. The Straight Mind and Other Essays. Boston: Beacon Press.
Zubin, David A., and Klaus-Michael Köpcke. 1981. “Gender: A Less than Arbitrary
Grammatical Category.” In Papers from the Seventh Regional Meeting, Chicago Lin-
guistic Society. April, 16–18. Chicago: The Society.
———. 1984. “Affect Classification in the German Gender System.” Lingua 63: 41–96.
———. 1986. “Gender and Folk Taxonomy: The Indexical Relation between Grammati-
cal and Lexical Categorization.” In Νoun Classes and Categorization: Symposium on
Categorization and Noun Classification: Papers (Typological Studies in Language
Vol. 7), 139–180. Amsterdam: John Benjamins.
Index of Authors
Agreement 17–18 Orientation 84, 90
Assignment 15–17
Assumption 81. See also Performativity 33, 49–50, 171–174
Presupposition 79–80 performative 35
Presupposition 79–80
Categorization 52–55 Principles, assignment
Category-bound activity 89 morphological 15, 17
Conversation analysis 84–88 phonological 15, 17
Correction 131–132 semantic 15–17
Gender Recipient-design 94
feminine 1–2, 14, 16 Reference
generic use, masculine 4, 27, 42 collective self-reference 95
grammatical 1–2, 13–31 inanimate 15–17, 68–70
hierarchy 40–44 indefinite second person 104
identity 37–40 individual self-reference 95
lexical 2 person 16, 74–79
masculine 1–2, 14, 16 recipient 106
natural 2 third person 111
neuter 1–2, 14, 16 Relativity
Relevance 116 linguistic 59–61
sociocultural 32–36 structural 61–62
Resolution 28–30
Idealized cognitive models 54
Index 36–37, 77–81 Self-repair 129–131
indexicality 75, 79–80 Sex 2, 34–36
Inference 58, 81–83 Sexism 4, 40, 169–171
Interaction 72–74
Thinking for speaking 62–65, 166–167
Membership category 88–90
Metonymy 54–57