See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/249870209

New Understandings of Gender and Language Classroom

Research: Texts, Teacher Talk and Student Talk

Article in Language Teaching Research · April 2000

DOI: 10.1177/136216880000400204

CITATIONS READS
143 3,667

1 author:

Jane Sunderland
Lancaster University
71 PUBLICATIONS 1,984 CITATIONS

SEE PROFILE

All content following this page was uploaded by Jane Sunderland on 25 January 2016.

The user has requested enhancement of the downloaded file.

Language Teaching Research 4,2 (2000); pp. 149–173

New understandings of gender and

language classroom research: texts,
teacher talk and student talk
Jane Sunderland Lancaster University

While gender has been an ongoing if sometimes peripheral area of

interest for researchers and practitioners in language education,
conceptualizations of gender itself have developed apace. This means
that, unfortunately, gender is at times viewed in an outdated way in
language education, resulting in oversimplification and unproductive
generalizations. In particular, women and girls are sometimes
simplistically represented as victims of gender bias in language
textbooks, and of male dominance in the classroom. This picture is far
from being the full one, does little, I would argue, to help female
students, and may mislead teachers. In this paper I present a rather
more complex picture. I illustrate some subtleties and complexities of
gender in language education, and suggest some implications of
research for educational practice. I also demonstrate alternatives for
research into gender and language classrooms, showing both how the
more familiar approaches can be fruitfully developed and how
researchers can go beyond them. It is important that both researchers
working in the area of gender and language education, and teachers in
their practice, should be able to engage with considerations of agency,
individuality and diversity, while not losing sight of the still-important
notions of disadvantage and of gender itself.

I Introduction
The 1990s witnessed radical changes in understandings of gender
(in the social, not grammatical sense). Maggie Humm’s definition
of gender in a 1989 publication as ‘a culturally-shaped group of
attributes and behaviours given to the female or the male’ may
have been useful once, but ten years later sounds crude,

Address for correspondence: Jane Sunderland, Department of Linguistics and Modern

English Language, Bowland College, Lancaster University, Lancaster LA1 4YT.

© Arnold 2000 1362–1688(00)LR064.OA

150 New understandings

deterministic and misleading. In particular, Humm’s idea of

attributes and behaviours being ‘given’ raises questions of ‘by what
or whom?’, ‘given once and for all?’ and of how ‘the female’ and
‘the male’ recipients accepted these attributes and behaviours:
passively? even graciously?
Understandings of gender are now more sophisticated, focusing
variously on gender identity, a sense of oneself as ‘masculine’ or
‘feminine’, and as performance (Butler, 1990) – but, crucially, do
not see gender as determined. It is therefore necessary to see
gender in language education in new, non-deterministic ways too.
However, conference papers on gender and education can still be
heard which, though well-intentioned, represent women and girls
as passive victims, and do not allow for the flux, agency, diversity
and individuality which, I suggest, are part of any classroom.
Drawing on both new understandings of gender, and refinements
of familiar ways of looking at gender in classrooms, this paper aims
to raise teachers’ awareness of the complexities and subtleties of
gender in language education, and to suggest implications for
practice and avenues of classroom research. It also aims to
illustrate how research on gender and language teaching can avoid
the pitfalls of representing teachers as predictable, willing,
unquestioning textbook users, and of female learners as passive
victims, and can engage with the notions of flux, agency, diversity
and individuality.
Some years ago I wrote: ‘The effects of gender roles, relations
and identities are everywhere. Ironically, because of this, in much
writing and thinking on English language teaching, gender appears
nowhere’ (Sunderland, 1994: 211). The claim was that,
paradoxically, because gender tends may seem normal and natural,
it often appears not to exist at all. At the start of the new
millennium we are perhaps less blind to the fact that much social
life, including our educational life, is gendered in some way, but
the claim still holds. Phenomena as diverse as literacy practices
(Millard, 1997), language tests (Sunderland, 1995a), performance
on those tests (Hellekant, 1994), self-esteem (Licht and Dweck,
1987), and learning styles and strategies (Oxford, 1994) have all
been shown to be gendered – in the sense that male and female
students tend to be represented, to behave or to feel differently.
(The word tend needs to be stressed.) However, it is now
 Jane Sunderland 151

recognized that these phenomena and the language through which

they are realized may be gendered in the additional sense that they
may play a role in the further gendering of students, that is, in
shaping their masculinities and femininities.
In this paper I look at two particular gendered aspects of foreign
language education: teaching materials, in the form of textbook
texts, and classroom interaction. For both of these our
understandings, shaped by new understandings of gender, have
developed considerably – in some ways, I suggest, in parallel.
Because of the intentional ‘broad brush’ of this paper, I am not
able to go into detail for either of the above aspects, but have
provided a substantial Bibliography.

II Changing understandings of gender in relation to language

textbook texts
I will start by acknowledging past work on gender and language
textbooks. However, I will suggest that the preoccupations inherent
in much of this work should now be left behind, and will illustrate
one way in which this is happening.
Many content analyses of FL textbooks – mainly EFL textbooks
– in the 1970s and 1980s found evidence of gender bias. This was
bias against women in three main senses, and could be categorized
as ‘exclusion’, ‘subordination and distortion’, and ‘degradation’. For
example:
• males were over-represented (e.g., Hellinger, 1980), i.e.,
exclusion;
• men tended to occupy both more powerful and a greater
range of occupational roles than did women (e.g., Porecca, 1984);
both males and females performed gender stereotypical
activities (e.g., Cincotta, 1978), i.e., subordination and distortion;
and
• women tended to be stereotypically emotional and were more
likely than male characters to be the butt of jokes (Hartman
and Judd, 1978) and of ‘implied slurs’ (Talansky, 1986), i.e.,
degradation.
Some teachers argued that this was precisely as it should be,
claiming that that, broadly, language textbooks ‘should’ reflect
152 New understandings

society. Others disagreed – and an interesting question is whether

textbooks should represent a more progressive situation than
actually exists, since while the same books continue to be used,
things move on. (Partly because this is precisely what tends to
happen, there have also been accusations of gender asymmetry in
FL textbooks being greater than the gender asymmetry in the
society they were intended to reflect (Ittzes, 1978).)
Linguistic analyses of language textbooks revealed subtle, even
pernicious effects that content analyses had not. For example:
• verbs associated with female agents reflected ‘some of the
traditional stereotypic female behavioural patterns’ (Hellinger,
1980: 272);
• in dialogues, females spoke less, spoke first less often, and
performed a narrower range of discourse roles (Hartman and
Judd, 1978; Talansky, 1986; Poulou, 1997); and
• the word letter [selected since this suggested the involvement of
a particular letter-writer with a particular recipient] was
associated with male characters, and letters [which suggested
a more general, more superficial level of involvement] with
female characters, as indicated by a computer concordancing
programme (Carroll and Kowitz, 1994).
Content and linguistic analyses of gender in language textbooks
have declined in frequency in the 1990s – partly, I suggest, because
bias is now less evident (Jones, Kitetu and Sunderland, 1997), but
also because theoretical developments suggested that the text itself
may not be the most appropriate focus of study (see below). And
while it would seem to be all to the good for teachers to be aware
of the above forms of bias when selecting textbooks, and for gender
bias to be considered on materials selection sessions in teacher
education, it is also important, I suggest, to be aware that bias may
be found somewhere other than ‘in the text’.

1 Does gender bias ‘in the text’ matter?

One question about gender bias in textbooks is ‘does it matter?’
It may not matter as much as these studies assumed – not because
gender bias is unimportant (it will always matter to those who have
principled objections to sexism in any shape or form), but because
 Jane Sunderland 153

to focus on gender bias in a text may be to miss the point as far

as learners and learning are concerned.
Gender bias in a text may adversely affect language learning,
but I would suggest that this is very hard to prove. Similarly, claims
about the relationship between bias in a textbook and a learner’s
gender identity, or ‘sex role’, such as ‘It can be seen from the few
examples [from a popular French language textbook] how sex-role
formation is influenced by the textbooks with which children come
into contact during the long years of their formal education’
(Cincotta, 1978: 29), are also hard if not impossible to prove.
Effects on learning of any text are impossible to predict because
we cannot predict a given reader’s response to that text, including
what that reader will ‘take’ from it (see Allwright (1984) and
Kebede (1998) for the related notion of ‘uptake’ in language
education). As Peterson and Lach, working in the field of children’s
literature, note:
The theoretical view of what reading is has changed substantially in the last
few years. It used to be thought of as a relatively straightforward process
through which the reader extracted meaning from the text. . . . Today, there is
growing recognition that what readers bring to the text influences how they
take meaning from it. Different readers, with different backgrounds, may take
very different meanings from the same text. In short, reading is a constructive
process.
(Peterson and Lach, 1990: 193)

Thus, though reading may be receptive rather than productive,

it is not a passive activity – rather, in reading constructively, a
reader will be engaging cognitively and affectively with a text. This
includes learners of a FL using a textbook as well as Ph.D. students
researching policy documents. The same can be said of learners as
‘listeners’, for example, when the teacher or another student is
reading from the textbook. Even an agreed case of gender bias in
a text, then, cannot be said in any deterministic way to make people
think in a gender-biased way, or to have such an effect on, say,
female students, that they become completely alienated from the
textbook or from the language learning process itself. While some
students may indeed be alienated from the text, others may accept
the bias and even enjoy it, and still others may recognize it for
what it is and become ‘resistant readers’, rejecting both the gender
representation and any implicitly gendered assumptions. And
154 New understandings

others may have a completely different reading from that which

the author intended, just as three people watching the same film
can come away with three different interpretations of what it was
actually ‘about’. This possibility of different ‘readings’ of the same
text does not mean that a text can be read in just any way.1 What
it does mean is that the individuals in a class are likely to read, or
hear, a given text in a range of different ways.
So, let us consider the possibility that looking at the text alone
may be a fruitless endeavour. This does not mean that we have
reached a dead end in the study of gender and language textbooks.
We can still – and, I suggest, should – look at how a text is used
in class. This may take us one step nearer to students’ learning and
understanding than would looking at the text alone. Use of course
refers to both teacher and students; here, however, I am only
looking at teacher use, or ‘treatment’.

2 Predicting textbook use?

Can teacher treatment of a text be predicted? Let us take as an
example a dialogue, between a male and a female character, which
is being taught in a mixed-sex class. Though I have no empirical
evidence for this, my teaching experience suggests that many
teachers would teach the dialogue by first getting a male and a
female student to demonstrate – male as Joe, female as Sue (or
whoever). Then mixed-sex pairs might practise it together –
probably with males reading Joe and females reading Sue. And
when (or if) this happens, the male students get more practice
in speaking English, albeit in the crude sense of numbers of
words. In other words, the female students’ language learning
opportunities are here arguably inferior to those of the males.
There are, however, at least in principle, other ways of teaching
this dialogue – and the diversity of teachers’ classroom practices
may be as great as the diversity of student uptake. Even in a mixed
sex class, the teacher may not expect mixed-sex pairs to work
together. Alternatively, she might reverse the roles – so that female
students play Joe and male students play Sue. She might give Joe’s
words to Sue and Sue’s words to Joe. She might start a class
discussion about the gender relations implied by this sort of
dialogue. She might even rescript the parts – or she might ask
 Jane Sunderland 155

the students to produce their own ending.

The point is that we cannot predict what a teacher will do from
a text itself. A focus on gendered texts alone, which assumes
teacher treatment of the text is somehow evident from the text,
would thus seem misguided. A text is arguably as good or as bad
as the treatment it receives from the teacher who is using it; in
particular, a text riddled with gender bias can be rescued and that
bias put to good effect, pedagogic and otherwise.

3 Teacher treatment of textbook texts

For this reason, then, empirical studies are needed of teachers’ use
of texts. And, using Literacy Studies’ concept of ‘talk around the
text’, it is possible to observe what a teacher actually does say in
the classroom: Stodolsky ‘found little evidence in . . . our case
studies to support the idea that [maths] teachers teach strictly by
the book’ (1989: 180); Bonkowski claimed that language teachers
often distorted textbook authors’ intentions, interpreting those
intentions through their own models of language, language
learning and language teaching (cited in Alderson, 1997: 12). Such
studies do not tell us about student ‘uptake’ of a given gendered
text, or how learning might be shaped by it. However, use is one
stage nearer to uptake; further, the text alone may point the
researcher in the opposite direction from what is shown empirically
by looking at use (for example, a very progressive text may be dealt
with in a very unprogressive manner).
Teacher treatment of gendered texts can similarly be researched.
I will illustrate this with two brief actual examples. The first is from
a study of teacher use of Upper Intermediate Matters (Shattuck,
1996). Shattuck collected data from an English language class at a
British Council school in Portugal; the unit in question was about
different types of marriage.
Shattuck found that the teacher chose not to stick to the text,
but went beyond it, contributing observations of his own about
weddings in the UK, including that

the bride (.) usually (.) if it’s especially for the church wedding will wear white
(.) and (.) the bridesmaids (.) she will often choose the (.) the outfit for them
(.) usually she chooses something horrible so they (.) don’t look as good as
her . . .
(Shattuck, 1996)
156 New understandings

This is of course a very negative and stereotypical representation

of women: that they are vain, don’t like other women, are jealous
of other women, and that brides are thus prepared to indulge in
such behaviour to make themselves look good. There may have
been a measure of irony here, which may or may not have been
recognized by the students. However, this contribution was very
much the teacher’s not the text’s, and could not have been
predicted.
The second example is that of Leontzakou, who collected data
from an English language school in Greece (1997). She observed
a teacher’s use of a dialogue from Blueprint Intermediate: in it, a
British couple were discussing the evening when their babysitter
had turned up late. The babysitter was a man, Alan. Given that
babysitting is traditionally most often associated with women or
girls, the text was, arguably, a progressive one.
The teacher talked about babysitting in England and in Greece,
but surprisingly ‘seems to ignore the fact that the textbook presents
a man who has been asked to do the babysitting’ (Leontzakou,
1997: 41). Again, this could not have been predicted; on the
contrary, a possible prediction would have been that since the
gender of the babysitter was unusual, the teacher would comment
on it. This is in one sense the reverse of Shattuck’s case, as here a
potentially progressive feature of a text is being ignored by the
teacher.
These two examples illustrate how teacher treatment of
gendered textbook texts is an operationalizable, interesting and
fruitful area of research. So how could this form of teacher talk be
investigated systematically?

4 Beyond the case study

For data collection, a range of lessons could be observed and
audiorecorded, the only criterion being that they would have to
include use of texts in which gender was in some way salient, that
is, they would have to be texts about people. Particularly valuable
would be investigating lessons in which different teachers were
teaching the same unit. The ‘talk around the (gendered) text’ would
then be transcribed. Since professional researchers could not do
this alone, but rather only with the active co-operation of teachers,
 Jane Sunderland 157

this could in fact become a classroom research project for a team

of teachers working in the same institution, with or without a
professional researcher. These teachers could record themselves
(despite knowing the purpose, this may still result in a few
surprises), or others (who, if willing, could be told after the event
about the purpose of the recording).
As regards analysis, the relevant episodes of teacher talk could
then be categorized using a working analytical framework which
takes into account both the text, and the teacher treatment of it.
One possibility is that shown in Figure 1. The top level of this
working framework represents two possible ‘text types’, the boxes
below, the ‘talk around the text’.
Using this framework, Shattuck’s ‘wedding’ episode from Upper
Intermediate Matters would be analysed as:
(1) a text which maintains a traditional representation of gender
roles;
(2) teacher treatment in which the teacher endorses or extends
traditional gender representation.

Text goes beyond Text maintains a

traditional traditional
representation of representation of
gender roles gender roles

Teacher Teacher
Teacher subverts Teacher subverts
ignores non-traditional ignores traditional
gender gender gender gender
issues representation issues representation

Teacher endorses or Teacher endorses or

extends non-traditional extends traditional
gender representation gender representation

Figure 1 An analytical framework for teacher treatment of gendered textbook

texts
158 New understandings

Leontzakou’s ‘babysitter’ episode from Blueprint Intermediate

would be analysed as:
(1) a text which goes beyond traditional representation of gender
roles;
(2) teacher treatment in which the teacher ignores gender issues.
Either text could in principle also have been subverted by the
teacher, the first by, say, starting a discussion of ‘marriage as an
institution which benefits men more than women’, and the second
by, say, suggesting that Alan, as a man, is clearly unreliable as a
babysitter.
If a range of teacher treatments of different texts are observed
and analysed, it may then be possible to establish, for example,
whether texts which maintain a traditional representation of
gender roles are typically subverted or typically endorsed, or
whether the gender issue is simply ignored. Conversely, it would
be possible to establish how more progressive texts, that is, those
which go beyond traditional representations of gender roles, are
actually treated. Different treatments of the same texts in different
educational and cultural contexts could also be investigated.
Such investigations would not tell us about the effect of a
gendered text on a student’s learning, or on her or his gender
identity. It would, however, I suggest, take us further than looking
at the text alone, and/or making completely unempirical
predictions about use, learning and identity. We can then see the
students’ readings or interpretations of a given text, though
potentially many and various, as being ‘filtered’ by the teacher’s
‘talk around the text’. This ‘model’ also allows for teacher diversity
and agency in ways which studies of textbooks alone do not.
This approach to gender in language textbooks has not produced
and will not produce findings which will help teachers evaluate and
select the most appropriate textbooks – for the simple reason that
it is not about the texts themselves. The choice of the most
progressive or gender-neutral language textbook in the world will
not stop gender-biased teacher (and student) treatment of the texts
within it. This is not to say that teachers should be cavalier about
textual bias – rather that a bias-free textbook will not of itself mean
bias-free teaching.
The approach does, however, have implications for writers and
 Jane Sunderland 159

publishers of textbooks. Given the possibility that a progressive

text can always be undermined, clarification of the author’s
intentions in the teacher’s guide may help obviate distortion of
those intentions by teachers in the classroom.
It also has implications for teacher education: trainee teachers
can be shown that a gender-biased text (of which there are still
many) does not have to mean gender-biased teaching. In materials
adaptation sessions on both pre- and in-service courses, teachers
can develop ways of dealing with bias. A text for example can be
critiqued through useful discussion work, and alternatives
suggested; very traditional gender roles in a dialogue can be
reversed. Teachers can also develop ways of doing justice to
progressive texts.

III Changing understandings of gender in relation to

classroom interaction
The second part of this paper concerns gender and classroom
interaction, focusing on teacher-to-student and student-to-teacher
talk (space prevents me from looking also at student-to-student
talk).

1 Differential teacher treatment by gender

Given the question to an imaginary teacher, ‘Do you treat your
male and female students differently?’, a likely answer, I suggest
(if this imaginary teacher were unfamiliar with the literature on
gender and classroom interaction), would be along the lines of
‘No, I’m a professional teacher, I treat my students all the same.’
And he or she is likely to be completely sincere in saying this.
However, in the 1970s and 1980s, many studies of teacher talk
in classrooms – all sorts of classrooms – found that not only did
the teacher talk more than the students, he or she talked far more
to the male than the female students. The researcher who did the
most to popularize such findings was the teacher and writer Dale
Spender, who was aware of the phenomenon and was deeply
disturbed about the possibility that it might happen in her
(secondary school) classroom. She therefore audiorecorded her
own lessons, and noted that
160 New understandings

. . . sometimes I have . . . thought I have gone too far and have spent more
time with the girls than the boys. But the tapes have proved otherwise. Out
of ten taped lessons . . . the maximum time I spent interacting with girls was
42% and on average 38%, and the minimum time with boys 58%. . . . It is
nothing short of a substantial shock to appreciate the discrepancy between
what I thought I was doing and what I actually was doing.
(Spender, 1982: 56)

Though Spender was not a language teacher, and has been

criticized for not being explicit about her methodology, her last
sentence is salutary: your perceptions are not reliable. And, given
that a teacher of any subject is juggling so many balls when
teaching, it is not surprising if giving equal treatment to girls and
boys slips through the net. However, Spender’s findings were in
fact widely echoed elsewhere. Merrett and Wheldall, for example,
found that secondary school boys received significantly more
positive and negative teacher responses than did girls (1992: 73; see
also Croll (1985) and Dart and Clarke (1988)). And writing on her
own meta-analysis of 81 such studies, Kelly noted:
It is now beyond dispute that girls receive less of the teacher’s attention in
class. . . . It applies in all age groups . . . in several countries, in various
socioeconomic groupings, across all subjects in the curriculum, and with both
male and female teachers.
(Kelly, 1988: 20; italics mine)

The findings of such studies were often interpreted as evidence for

and a manifestation of male dominance.
It is important that teachers are aware of the tendency for boys
to receive more attention than girls (and men than women), in
order that they can monitor their own classroom behaviour. They
can also help to obviate it through classroom management
strategies such as alternating between female and male students
when it comes to asking or answering questions. However, this
awareness and these remedial practices need to be kept in
proportion. Giving male students more attention than females is
likely to be unintentional, rather a sort of collaborative process
between teacher and students (Swann and Graddol, 1988), and
such teacher talk is accordingly best referred to as ‘differential
teacher treatment by gender’ rather than ‘discrimination’ or
‘favouritism’. In addition, quantity of attention may not be what
matters for learning, as I will now go on to show.
 Jane Sunderland 161

2 Beyond quantity of attention

A distinction must be made between amount of attention and kind
of attention. Teacher attention in itself may not be useful; some
attention, like being told off, could hinder learning. And Kelly
notes that ‘The discrepancy [between amount of attention boys and
girls receive] is most marked for behavioural criticism’ (Kelly, 1988:
29; my italics). If the teacher does feel that boys are getting more
of his or her attention, he or she should thus not assume that this
means the boys are getting better learning opportunities. (Though
it may be that the girls are being deprived of such opportunities.)
Kind of attention is likely, I would suggest, to be what counts.
While continual behavioural criticism may be a cause of concern
for boys’ education, imbalances in teacher attention which are due
to other forms of utterance (‘boys also get more instructional
contacts, more high level questions, more academic criticism and
slightly more praise than girls’ (Kelly, 1988: 29)) are a cause of
concern for their impact on girls’ learning opportunities.
However, there have been very few studies of this kind in the
language classroom. One underexplored question is whether,
because foreign languages tend to be subjects in which women and
girls do well (Arnot, David and Weiner, 1996), this ‘differential
teacher treatment by gender’ may be less salient, or manifested in
particular ways. This section illustrates how differential kinds of
attention can be explored within the traditional ‘gender
differences’ framework, but in ways that avoid both unwarranted
generalizations and an undue stress on differences.
In my own study of a German as a Foreign Language classroom
(Sunderland, 1996), I looked at a range of specific ways in which
the teacher might treat the girls and boys differently, and at
whether any such differences were statistically significant. The
school was a British comprehensive; the students, 14 boys and 13
girls, then 11 or 12 years old, were in their first year of German;
the data was transcripts of twelve 50-minute lessons.
Using the concept of ‘solicit’ (Bellack et al., 1996; Sinclair and
Coulthard, 1992), I was interested in what the teacher asked the
boys and girls to do and say, and whether this varied with gender.
A solicit might be phrased as an interrogative (‘What is the
German word for train?’), or it might not, (‘the German word for
train – Susan’); it could also be intended to elicit non-academic
162 New understandings

behaviour (‘Can you close the window, please’). In addition to

solicits, I looked at teacher feedback (positive and negative) to
students’ responses to the teacher’s questions, and at the teacher’s
responses to the students’ questions.
Importantly, on most measures there was no or only statistically
non-significant evidence of differential teacher treatment,
suggesting that gender similarities were the norm (as is probably
most often the case). There were, however, two statistically
significant cases of differential teacher-treatment-by-gender, and
two cases approaching significance (see Figure 2).
Since many of the non-academic solicits were in fact disciplinary,
it looked again as if the greater amount of attention the boys were
getting was largely because they were being told off by the teacher.
It is certainly hard to see this as better attention. Indeed, if we look
at the last two cases of differential teacher-treatment, it seems that
the teacher was actually treating – or, arguably, constructing – the
girls as the more academic students: they were being asked
questions which required longer answers, and those answers were
more likely to be expected to be in German. Such findings
illustrate the importance of making a distinction between amount

Taking into consideration the fact that there was one more boy
than there were girls in the class, and accordingly using the
concept of the ‘average boy’ and the ‘average girl’:
(a) The teacher paid more attention to the boys in terms of
* number of solicit words (significant at 5% level)
* proportion of non-academic solicits (approaching
significance at 5% level)
and
(b) in relation to boys, girls were asked
* a greater proportion of academic solicits to which they were
expected to respond in German (approaching significance at
5% level)
* a greater proportion of questions which required an answer
of more than one word (significant at 5% level)
(Sunderland, 1996)

Figure 2 Differences in teacher treatment of girls and of boys

 Jane Sunderland 163

of teacher attention and the nature of this attention, especially as

regards the provision of learning opportunities. They also illustrate
the possibility that while boys may appear to dominate the
classroom in one sense, girls may dominate it in another.

3 Student-to-teacher talk
Studies of the gendered nature of student talk to the teacher have
mostly found that male students tend to talk more to the teacher
than do female students (e.g., Sadker and Sadker, 1985). However,
relatively few studies have been done in the language classroom.
One exception is Batters (1987), who found that male students
studying a range of modern foreign languages were dominant in
‘oral and participatory activities’ – including speaking to the
teacher in the target language. Batters notes ‘[Boys] . . . like [oral
work] as it is an area where they dominate’ (1987: 78) – which
echoes the tendency found in talk in non-language classrooms.
In the (1996) study reported on here, I looked at some of the
different (quantitative and qualitative) ways in which the boys and
girls spoke to the teacher. Though again similarity was more
evident than difference, in two cases the differences were
statistically significant:
(a) the ‘average girl’ produced shorter solicits than the ‘average
boy’

However,
(b) when the teacher asked a question without naming a student
to answer it, the ‘average girl’ volunteered significantly more
answers in German than did the ‘average boy’.

Again, in terms of this second measure of interaction, the girls in

this foreign language classroom look like the more academic
students.
It is therefore important for both teachers and researchers when
looking at classroom talk not only to be wary of a ‘more is better’
interpretation, but also to look at what is done and accomplished
in that talk. In this data, just as the boys did not seem to get most
of what seemed to be the really useful academic teacher attention,
164 New understandings

and though the boys may have talked more on some measures,
they also talked less than the girls when it came to volunteering
answers in German. Together, the findings suggest that the type of
femininity being ‘performed’, or constructed, in this class, by the
teacher and the girls, was a distinctly academic one relative to the
boys’ masculinity.
In looking at kinds of talk, and using tests of statistical
significance to identify those gender differences which have not
occurred by chance, I hope to have shown how a quantitative
approach, traditionally ‘variationist’ at least in student-to-teacher
talk, still has value, if the analysis of classroom language is at high
levels of specificity. This enables the avoidance of crude
generalizations about male dominance and female disadvantage in
the classroom. However, the danger even with such relatively
specific studies and talk of ‘tendencies’ is that girls and/or women
are still represented as a single homogeneous group, as are boys and
men. If the aim is to minimize gender tendencies, suggesting even
implicitly that there is a clear ‘gender binary’ may not only be
inaccurate, but also counter-productive.

4 Diversity: context and individuals

It is thus important to look for diversity. Findings will vary with
context, and neither differential teacher treatment by gender nor
male students’ verbosity should therefore be seen as automatic or
universal classroom phenomena. Some studies (e.g., Yepez, 1994)
indeed did not find differential-teacher-treatment at all. (The
students in Yepez’ study were adults, and age may be an important
variable here.) Clearly, what happens in a secondary school
classroom may not happen in higher or primary education;
classrooms in different cultural contexts with different discourses
surrounding gender are themselves likely to be gendered
differently from each other. However, even within any one
classroom, even when students are the same age and of comparable
socio-economic and educational backgrounds, gender will not be a
straightforward masculine–feminine binary. There will always be
diversity within each gender group and in all probability overlap
between them.2 Acknowledgement of this is crucial in any
discussion or study of gender and classroom talk.
 Jane Sunderland 165

The early findings from my own (1996) study were based on data
for the ‘average girl’ and the ‘average boy’. However, looking
additionally at the number of utterances in the teacher talk
addressed to individual students, and talk from them to her,
diversity was evident (see Table 1).
The first four columns show the number of times the teacher
addressed each girl and each boy over the twelve lessons; the
second four columns, the number of times each boy and each girl
addressed the teacher.3 There are large differences within girls and
within boys. Two boys, Gus and Len, received a huge amount of
the teacher’s attention – Gus more than ten times as much as Max
(as well as almost twice as much as Kay, the girl who received the
most).
If Gus and Len are excluded, the boys as a group and the girls
as a group get broadly the same range of attention, that is, each
student ‘received’ between 10 and 65 teacher utterances. This
echoes the findings of other studies, for example, French and
French (1984), that if ‘boys’ get more attention than ‘girls’, this is
usually in fact because of a small subset of boys. A generalization
in this classroom about ‘boys taking all the teacher’s attention’

Table 1 Number of teacher-directed utterances produced by each student, and

student-directed utterances produced by the teacher

Utterances Utterances Girls’ utterances Boys’ utterances

directed to girls directed to boys to teacher to teacher

Kay 63 Gus 119 Kay 264 Gus 245

Lia 42 Len 100 Lyn (1) 117.5 Len 78
May (1) 39.5 Ken 65 Lyn (2) 117.5 Dan 74
May (2) 39.5 Dan 35 Lia 109 Ken 51
Ann 36 Don 34 May (1) 93 Don 50
Lyn (1) 22 Bob (1) 25 May (2) 93 Ray (1) 46.5
Lyn (2) 22 Bob (2) 25 Sue 92 Ray (2) 46.5
Ros 21 Pip 23 Ros 53 Sam 45
Bea 21 Sam 17 Bea 52 Bob (1) 37.5
Una 18 Ray (1) 16.5 Ann 49 Bob (2) 37.5
Sue 15 Ray (2) 16.5 Eve 32 Jim 36
Liz 11 Jim 15 Liz 30 Max 31
Eve 10 Guy 11 Una 15 Guy 30
Max 10 Pip 15

Totals 360 512 1117 823

Means 27.69 36.57 85.92 58.79
166 New understandings

would be simply inaccurate.

Returning to the last four columns of Table 1, of the eight
students who address most utterances to the teacher, seven are
girls. (Again, Gus, just as he is frequently addressed by the teacher,
talks a lot himself.) The findings about the talk produced by the
girls, therefore, are broadly due to the contributions of half the
girls – more than just a ‘small subset’. Gender is thus arguably, and
interestingly, a more important factor in student-to-teacher talk in
this classroom than in teacher-to-student talk.
Without losing sight of gender, then, the diversity in classroom
talk produced by diversity between individuals – between girls,
between boys, and between students as a whole – needs to be
recognized. Even this simple quantitative study of individuals
shows that gender in this classroom, as elsewhere, is thus not a
simple binary.
This is only a ‘snapshot’, of course, and one which does not see
gender as a continuous process, that is, as gendering. Against this
characterization, I would like to revisit the earlier definition of
gender as ‘a culturally-shaped group of attributes and behaviours
given to the female or the male’ (Humm, 1989). I suggested this
definition was unsatisfactory largely because of the word given:
what about the agency of the ‘recipient’? Diversity and gender
tendencies suggest that ‘given’ is, at best, only part of the story, and
that individuals have their own part to play which is never more
than partially or temporarily pre-scripted. The final section
illustrates learners’ (and a teacher’s) own agency and, I hope,
gendering as a continuous process.

5 Agency, identities and classroom discourse

The idea of individuals having agency suggests that learners’ and
teachers’ (and others’) gender identities are actually constructed
by their own (and others’) social practices (and not just reflected
in these practices). This essentially post-structuralist view of gender
thus sees language use, one form of social practice, as not just a
characteristic of gender (the traditional sociolinguistic paradigm),
but as one of the influences that may itself shape gender. We can
thus see gender identities themselves as processes, shaped by
gendered talk – by gendered discourse and discourses, including
 Jane Sunderland 167

those of the classroom.4 This echoes the earlier point concerning

language textbook texts: that a learner’s gender identity may be
shaped but cannot be determined by gendered texts (or indeed by
any other sort of text or social practice), since an individual is an
agent who always has some capacity for contestation. In particular,
a teacher can treat a gendered text in a range of possible,
unpredicted ways.
I would like to conclude this section with an illustration of how
classroom discourse (one social practice) has its own role to play
in the shaping of learners’ gender – as girls, as boys, and as
gendered learners of a foreign language. The example also serves
to illustrate how gender can be investigated in the classroom
without using a ‘differences’ framework, and how ‘disadvantage’
may not always be the most relevant concept to work with.
In the extract that follows (Sunderland, 1995b, 1996), the
students from the German class are reading out dialogues they
have been writing. They are working in single-sex pairs, by choice.
The teacher – who is aware of the danger of talking to boys more
than girls – is alternating between pairs of boys and pairs of girls.
It is now the turn of two boys:
T: we’re going to have two more boys I think . two more
boys . what about Ray and Max
Ray/Max: no
T: no . why not
Lia/May: we’re boys
Kay/Bea: we’re boys
Kay: we’re boys miss
T: all right we’ll have two more girls and then we’ll see if
the boys have got any courage
So here three girls put up their hand to say ‘We’re boys’ – no
one laughs, and the teacher lets them have the next turn. On the
simplest level of analysis, girl, in English, is a term of abuse for a
boy. So a boy would never say ‘I’m a girl’ without risking extreme
ridicule. The reverse however is not true – boy in English is not a
term of abuse for a girl. However, the situation is I think more
complex than this.
I interviewed two boys about this event, and in particular about
whether they thought the reverse situation could ever obtain. Their
168 New understandings

feelings were extremely clear: ‘no way’ would they have said ‘we’re
girls, miss’, not because girl is a term of abuse for a boy, but because
‘boys have a limit what they can do’, and if they had said it
‘everybody laughed their heads off’. The boys here are
representing gender as a clear male/female binary. They also
illustrate how gender identities are in many ways asymmetrical –
girls can ‘cross’ gender boundaries with impunity, whereas boys
cannot. Boys’ boundaries of masculinity thus appear more rigid –
it is most definitely not OK for them to ‘become’ girls, even
temporarily, strategically and jokily. This problematic ‘Boy-as-
OK/Girl-as-not-OK’ discourse (problematic because although girls
seem to be the ‘winners’ in this episode, there also appears to be
something ‘wrong’ with being a girl), may do more than reflect the
students’ beliefs, but may also shape them – and, accordingly, their
identities. Gender identities thus are not fixed, but rather are in a
constant state of flux, shifting over time (Weedon, 1987). Cameron
(1997: 49) puts it like this:
Whereas sociolinguistics traditionally assumes that people talk the way they
do because of who they (already) are, the postmodernist approach suggests
that people are who they are because of (among other things) the way they
talk. This shifts the focus away from a simple cataloguing of differences
between men and women to a subtler and more complex inquiry into how
people use linguistic resources to produce gender differentiation.

Both the reflection and the shaping that go on in the classroom

may be as much to do with the world which operates outside the
classroom as with the classroom itself. There are few clear
implications here for teacher intervention, or even for whether
intervention is appropriate. This unpredictable and unusual event
is, however, a reminder of the complexity of gender, and is
accordingly a warning for both teachers and researchers of the
unlikelihood of any straightforward deterministic relationship
between classroom talk and learner gender.

IV Conclusion
While the study of gender is advancing apace, often in conceptually
complex ways, new and even experienced language teachers are
still continuing to discover its importance for the first time. In this
paper I have therefore tried to illustrate for practitioners as well
 Jane Sunderland 169

as researchers some of the complexities of gender, and thus how

some unhelpful generalizations about and representations of
female and male learners can be avoided.
For practitioners whose interest in gender is related to a concern
with learning and with possible disadvantage, it may be hard to
integrate these new understandings of gender into their own
understanding and practice of language education. Yet, despite the
complexities, practitioners can of course take active professional
steps to ameliorate disadvantage (though these will always be
dependent on the actual classroom, and cannot be simply
prescribed through a list of ‘Do’s and Don’ts’). For example,
consciously alternating between female and male students when
asking questions may be a useful classroom management strategy,
and teacher educators can usefully take on board questions of
treatment of progressive and gender-biased texts in sessions on
materials selection, evaluation, design and use.
If schools or individual teachers have a ‘research agenda’,
teachers can also carry out their own classroom (or action)
research. I have already suggested observing and audiorecording
colleagues’ treatment of gendered texts, and transcribing just the
‘talk around the text’ part of the lesson (since transcribing and
analysing whole lessons is very time-consuming). Teachers can also
observe each other and list, say, the questions asked to girls and
those asked to boys, or the questions each student asks or answers.
The number and nature of those questions can then be compared.
It is even possible for students themselves to do such a study, the
findings of which could then be presented and discussed as a valid
language learning activity.
I have implicitly represented current research into gender and
language education as falling into three areas, which, though
related, are probably best kept conceptually distinct. These are as
follows:
(1) gender tendencies and similarities;
(2) gender identities; and
(3) educational disadvantage.
Specific manifestations of gender tendencies/similarities and
identities may be related to educational disadvantage. For example,
if in a given classroom boys tend to ask more questions than girls,
170 New understandings

this may mean girls have fewer opportunities to test their

hypotheses about the target language. However, it would seem
foolish to always see gender differentiation as disadvantage, with
girls as victims. In this case, rather than failing to ask, or to get
their questions answered, those girls who are silent may rather be
consciously attending to and learning from the teacher’s answers
to the boys’ questions Dick Allwright, personal communication.
And the girls who said ‘we’re boys, miss!’ were not straight-
forwardly expressing a regret that they were born female, but were
rather using this phrase as a strategy to get what they wanted –
and succeeded in doing this.
At the same time, it would seem important to be equally open
to the possibility of a relationship between gender tendencies and
identities, and educational disadvantage, as to the possibility that
a given set of gender tendencies/similarities and identities may not
straight-forwardly equate with disadvantage, either of girls or of
boys.

V Endnotes
1
As Sara Mills writes in her ‘Introduction’ to Gendering the Reader:

. . . texts are not simply ‘open’ in Umberto Eco’s sense. . . . Rather, it

would seem that texts must structure the reader’s response to some extent
through certain clues and frames which signal to the reader the range of
readings which are possible. . . . Most of the essays in this volume propose
that the struggle for meaning is a complex negotiation between text and
reader.
(Mills, 1994: 8–9)

2
Even the stereotype ‘Girls have neater writing than boys’ draws on nothing
more than a tendency. It does not mean that all girls write neatly or equally
neatly and all boys untidily or equally untidily, and certainly not that all
girls write more neatly than all boys. It simply means that there is a
tendency for girls to write more neatly than boys – and accordingly that
there will be a lot of overlap, some boys being neater than some girls.
3
The names are fictional. Some appear twice because there were several
‘pairs’ of students in the class with the same names.
4
The term discourse is used in different ways. Here, I am using it to mean
‘ways of seeing and behaving in the world’, or, as Fairclough (1992: 3) puts
it, ‘different ways of structuring areas of knowledge and social practice’.
 Jane Sunderland 171

VI References
Abraham, J. 1989: Teacher ideology and sex roles in curriculum texts.
British Journal of Sociology of Education 10(1): 33–51.
Alderson, C. 1997: Models of language: whose? what for? what use?
In Ryan, A. and A. Wray, editors, Evolving models of language.
British Studies in Applied Linguistics 12. Clevedon, OH:
BAAL/Multilingual Matters.
Allwright, D. 1984: Why don’t learners learn what teachers teach? – the
interaction hypothesis. Language Learning in Formal and Informal
Contexts. IRAAL.
Arnot, M., David, M. and Weiner, G. 1996: Educational reforms and
gender equality in schools. Equal Opportunities Commission
Research Discussion Series No. 17. Manchester: EOC.
Batters, J. 1987: Pupil and teacher perceptions of foreign language
learning. Ph.D. thesis, University of Bath.
Bellack, A., Kliebard, H., Hyman, R. and Smith, F. 1966: The language of
the classroom. New York: Teachers’ College Press.
Bonkowski, F. 1995: Teacher use and interpretation of textbook materials
in the secondary ESL classroom in Quebec. Ph.D. dissertation,
Lancaster University.
Butler, J. 1990: Gender trouble: feminism and the subversion of identity.
New York: Routledge.
Cameron, D. 1997: Performing gender identity: young men’s talk and the
construction of heterosexual masculinity. In Johnson, S. and
Meinhof, U., editors, Language and masculinity. Oxford: Blackwell.
Carroll, D. and Kowitz, J. 1994: Using concordancing techniques to study
gender stereotyping in ELT textbooks. In Sunderland, J., editor,
Exploring gender: questions and implications for English language
education. Hemel Hempstead: Prentice Hall.
Cincotta, M.S. 1978: Textbooks and their influence on sex-role stereotype
formation. Babel: Journal of the Australian Federation of MLTS
Association 14(3).
Croll, P. 1985: Teacher interactions with individual male and female pupils
in junior age classrooms. Educational Research 27(3): 220–23.
Dart, B. and Clarke, J. 1988. Sexism in schools: a new look. Educational
Review 40(1): 41–49.
Fairclough, N. 1992: Discourse and social change. London: Polity.
French, J. and French, P. 1984: Gender imbalances in the primary
classroom: an interactional account. Educational Research 26(2):
127–36.
Gaff, R. 1982: Sex-stereotyping in modern language teaching – an aspect
of the hidden curriculum. British Journal of Language Teaching
20(3): 71–78.
172 New understandings

Hartman, P. and Judd, E. 1978: Sexism and TESOL materials. TESOL

Quarterly 12(4): 383–92.
Hellekant, J. 1994: Are multiple choice tests unfair to girls? System 22(3):
349–52.
Hellinger, M. 1980: ‘For men must work and women must weep’: sexism
in English language textbooks used in German schools. Women’s
Studies International Quarterly 3: 267–75.
Humm, M. 1989: The dictionary of feminist theory. Hemel Hempstead:
Harvester Wheatsheaf.
Ittzes, K. 1978: Hungarian women in reality. In Freudenstein, R., editor,
The role of women in foreign language textbooks: a collection of
essays. Collection d’Etudes Linguistiques, No. 24. Ghent: Fédération
Internationale des Professeurs de Langues Vivante.
Jones, M., Kitetu, C. and Sunderland, J. 1997: Discourse roles, gender and
language textbook dialogues: who learns what from John and Sally?
Gender and Education 9(4): 469–90.
Kebede, S. 1998: The relationship between uptake and questioning.
Centre for Research in Language Education Working Paper 32.
Dept. of Linguistics and Modern English Language, Lancaster
University.
Kelly, A. 1988: Gender differences in teacher–pupil interactions: a meta-
analytic review. Research in Education 39: 1–23.
Leontzakou, C. 1997: How teachers deal with gendered EFL textbook
material. M.A. dissertation, Lancaster University.
Licht, B. and Dweck, C. 1987: Sex differences in achievement orientations.
In Arnot, M. and Weiner, G., editors, Gender and the politics of
schooling. London: Unwin Hyman.
Merrett, F. and Wheldall, K. 1992: Teachers’ use of praise and reprimands
to boys and girls. Educational Review 44(1): 73–79.
Millard, E. 1997: Differently literate: boys, girls and the schooling of
literacy. London: Falmer Press.
Mills, S., editor, 1994: Gendering the reader. Hemel Hempstead: Harvester
Wheatsheaf.
Oxford, R. 1994: La differénce continue . . .: gender differences in
second/foreign language learning styles and strategies. In
Sunderland, J., editor, Exploring gender: questions and implications
for English language education. Hemel Hempstead: Prentice Hall.
Peterson, S.B. and Lach, M.A. 1990: Gender stereotypes in children’s
books: their prevalence and influence on cognitive and affective
development. Gender and Education 212: 185–97.
Porecca, K. 1984: Sexism in current ESL textbooks. TESOL Quarterly
18(4): 705–24.
Poulou, S. 1997: Sexism in the discourse roles of textbook dialogues.
Language Learning Journal 15 (March).
 Jane Sunderland 173

Sadker, M. and Sadker, D. 1985: Sexism in the schoolroom of the ’80s.

Psychology Today, March: 54–7.
Shattuck, J. 1996: The interplay between EFL textbooks, teacher
behaviour and gender. M.A. dissertation, Lancaster University.
Sinclair, J.M. and Coulthard, R.M. 1992: Towards an analysis of discourse.
In Coulthard, M., editor, Advances in spoken discourse analysis.
London: Routledge.
Spender, D. 1982: Invisible women: the schooling scandal. London: The
Women’s Press.
Stodolsky, S. 1989: Is teaching really by the book? In Jackson, P. and
Harountian-Gordon, S., editors, From Socrates to software: the
teacher as text and the text as teacher. The National Society for the
Study of Education (USA): 159–84.
Sunderland, J., editor, 1994: Exploring gender: questions and implications
for English language education. Hemel Hempstead: Prentice Hall.
–––– 1995a: Gender and language testing. Language Testing Update 17:
24–35.
–––– 1995b: ‘We’re boys, miss!’: finding gendered identities and looking
for gendering of identities in the foreign language classroom. In
Mills, S., editor, Language and gender: interdisciplinary perspectives.
Harlow: Longman.
–––– 1996: Gendered discourse in the foreign language classroom:
Teacher–student and student–teacher talk, and the social construction
of children’s femininities and masculinities. Ph.D. dissertation,
Lancaster University.
–––– 1998: Girls being quiet: a problem for the foreign language
classroom? Language Teaching Research 2(1): 48–82.
Swann, J. and Graddol, D. 1988: Gender inequalities in classroom talk.
English in Education 22(1): 48–65.
Talansky, S. 1986: Sex role stereotyping in TEFL teaching materials.
Perspectives XI(3): 32–41.
Weedon, C. 1987: Feminist practice and poststructuralist theory. Oxford:
Blackledge.
Yepez, M. 1994: An observation of gender-specific teacher behaviour in
the ESL classroom. Sex Roles 30(1/2): 121–33.

View publication stats