Positioning gender in discourse: a feminist research methodology
Dr Judith Baxter; University of Reading, UK
Introduction
The aim of this paper is to introduce to you an approach to discourse analysis that I developed while conducting an ethnographic study into classroom talk. It is a brand of data interpretation called feminist post-structuralist discourse analysis, or FPDA for short. I consider this method to be of particular relevance to educational researchers who may be challenged by some of the complexities and subtleties of working with transcripts of spoken discourse in classroom settings.
My intention in this session is to give you a brief introduction to FPDA by describing how it emerged as the outcome of my own research. I will aim to explain (OHT):
- What FPDA is: the key principles and characteristics of this approach
- The provenance of FPDA
- How FPDA works as an approach to discourse analysis
- What its potential value is to educational research methodology.
As an approach that brings together the diverse (and some might say, contradictory) fields of feminism and post-structuralism, FPDA is necessarily a highly theorised methodology. It would need considerable practical demonstration to reveal how the method works in relation to a corpus of data (an hour rather than 20 minutes). Today I can only hope to give you a flavour of FPDA while referring you both to the Appendix* which offers a detailed illustration of the approach, and also to the Bibliography, which includes articles I have written on the subject.
1.What is FPDA? (OHT)
FPDA can be defined as ‘a feminist approach to analysing the ways in which speakersnegotiate their identities, relationships and positions in the world according to the ways they are located by interwoven discourses’.
FPDA has two distinct dimensions: post-structuralist and feminist. The post-structuralist dimension of FPDA is strongly informed by Foucault’s (1972: 49) view of discourses as ‘practices that systematically form the object of which they speak’. FPDA considers that routine spoken interactions in the classroom, whether formally assessed or as part of informal learning, are interwoven with a web of social and institutional discourses. According to this view, discourses act as a means of organising power relations between speakers. Thus, the post-structuralist dimension of FPDA asks analysts to look at the way different discourses work intertextually to position speakers as variously powerful and powerless, often shifting from one position to another in a matter of moments. Furthermore, what we mean by being ‘powerfully positioned’ within a given discursive context should always be open to interpretation. FPDA advocates the importance of multiple-authored accounts within discourse analysis, thus giving space for the competing perspectives of research participants alongside those of analysts.
The feminist dimension of FPDA invites researchers to privilege the social category of gender in terms of the way in which power relations are constructed through spoken interactions. It highlights the continuing ways in which females are constituted as less powerful than males in many educational contexts. It also recognises that there are given contexts when institutional discourses work in reactionary ways to produce girls as more subservient and boys as more dominant. However, FPDA as an integrated approach challenges any simplistic view that girls should always be perceived as helpless victims in the classroom. Rather, it proposes that students of both sexes adopt multiple and often competing subject positions as speakers. FPDA suggests that there are some learning contexts or moments in a lesson when female speakers will be more powerfully positioned than male speakers, and others where the opposite may be true.
So, to sum up, the principal function of FPDA is to highlight key discourses on gender as they are negotiated and performed within specific, localised contexts. It involves making sense of the way competing discourses position speakers as relatively powerful, powerless or a combination of both. It foregrounds moments of strength in female interactions with others, while identifying the continuing reactionary effects of certain institutional discourses on gender relations in the classroom.
2.The provenance of FPDA
Although I am probably the first to coin FPDA as a term, the approach has a clear provenance in feminist post-structuralism. This includes the theoretical work of a number of feminist post-structuralists such as Butler (1991); Bucholtz (1999); and Weedon (1997); the application of feminist post-structuralist ideas to educational contexts by theorists such as Jones (1993; 1997); Davies (1993; 1997); Walkerdine (1990; 1998); and prototypical FPDA research studies by writers such as Bergvall (1998); Davies & Banks (1992); Francis (1998); Simpson (1997); Walkerdine (1998) and Margaret Wetherell 1998.
However, I don’t think I would have decided to use FPDA if it hadn’t emerged ethnographically as the most fitting solution for analysing the complexities of my body of spoken data. So, to give you an insight into the way FPDA helped me to address a particular research methodology problem, I will now give you a brief outline of the study.
The research study
My research aim was to examine the significance of gender for assessing students’ speech in whole class settings for the UK GCSE English examination (NEAB, 1999). I observed a single Year 10 class (15 to 16 year olds) of mixed sex students involved in a range of oral tasks for a period of about four months. My data collection involved observation and video recordings of class activities and examination meetings, as well as interviews with students, teachers and examiners. As many of you will know, students in the UK are assessed on their speaking and listening abilities in a range of different speech contexts. In 1998 however, there was a new requirement to assess students’ oral competence in more public and formal settings such as whole class discussions, giving presentations, and holding debates. My own view at this stage was that these assessment criteria unwittingly supported a normative model of effective male speech. For example, the syllabus stated that to gain a top mark in the ‘discuss, argue, persuade’ category (OHT):
A Grade A* candidate:
‘Will……use language in a dynamic and influential way; make thought-provoking contributions through powerful expressions and command of the situation.’ (My italics)
(NEAB, 1999)
It is a well-known finding in language and gender research, that in whole class discussion, boys are more likely than girls to gain access to the public space of the classroom, and subsequently to ‘hold the floor’. Given the emphasis in the new assessment criteria on dominant speech, I wondered whether there would be implications along gender lines for the assessment of students’ speech in public settings.
I was looking for a method of discourse analysis which took account of the complexities and competing messages arising from analysing classroom talk that was being produced for assessment purposes. Having amassed a considerable bank of spoken data, I initially experimented with a range of more established methods of discourse analysis, such as ethnography of speaking methods and conversation analysis. Yet none seemed to capture or make sense of the richness, complexity and especially, the contradictoriness of the themes and messages emerging from my data. However, as I investigated the data, I began to notice that there were certain repeated sets of expectations, values and assumptions which appeared to shape the social and learning relationships of this class of students, and in turn, their spoken interactions. These sets of expectations seemed to match the Foucauldian description of discourses as ‘practices that systematically form the object of which they speak’ (1972: 49). Through a combination of close observation of verbal and non-verbal language, as well as noting the metalanguage used by participants in interviews, I recorded the ways three particular ‘discourses’ appeared to shape the students’ spoken interactions in whole class settings. These were (OHT):
- Approval:
Peer approval (e.g. being considered ‘popular’ with the peer group)
Teacher approval (e.g. being regularly selected to speak by the teacher)
- Gender differentiation (e.g. teachers and students overtly drawing attention to gender differences in their interviews)
- A model of collaborative talk (e.g. good listening skills; offering support to peers; sharing and exchanging ideas)
What I noticed was that whenever these students spoke in the classroom, whether formally or informally, their speech and behaviour was constantly being negotiated by their competing subject positions within these three interwoven discourses. I began to sense that students’ ability to speak in public settings could not be fully and fairly assessed by teachers and examiners without a concomitant understanding of how their speaking skills were continuously mediated by their subject positions within these three competing discourses. I also became aware that a discourse of gender differentiation was influencing the outcome of the oral assessments of these students, though not always adversely for girls.
3.How FPDA works
Post-structuralist theorists with their more global view, rarely have their noses pressed up against the exigencies of talk-in-interaction. Rarely are they called on to explain how their perspective might apply to what is happening right now, on the ground in this very conversation.
(Wetherell, 1998)
Given this was my sense impression, supported by the data collection process, how can FPDA help to disentangle the complexities of analysing competing discourses within a given stretch of text? In short, how does FPDA work?
One of the more persistent criticisms of post-structuralist theorists, as Wetherell’s comment above illustrates, is their failure to connect abstract principles about the nature of discourses to the practical job of analysing spoken interactions.
My own approach to FPDA tries to overcome this criticism by operating on two distinct yet interrelated levels. These I have named denotative and connotative, from the work of the French semiologist, Roland Barthes (1973). The denotative level has its roots in Conversation Analysis (or CA) with its focus upon making a detailed, close-grained linguistic description of the sequence of talk-in-interaction. (If you scan p.2 in the Appendix you will see that I have carried out a close linguistic study of the extract on p.1, without obviously attempting any interpretative analysis).
The second level is the connotative analysis. This is more similar to Critical Discourse Analysis (or CDA), which has traditionally tended to avoid any detailed study of talk-in-interaction (although this is now changing). CDA is much more concerned with tracing binary power relations within discourse, particularly where these tend to fix certain subjects as dominant and others as subordinate. While FPDA is clearly interested in issues of discourse and power, it places far more emphasis upon the interplay of multiple voices within a discursive context, as well as the complexity and contradictoriness of those voices. Part of its quest is to move away from the single authored account of the researcher and to incorporate the multiple voices of the various participants in the study. So the written up analysis — and I don’t say that I achieve this— would ideally be a collage of voices representing the different perspectives and concerns of, in this case, the students, their teachers, their assessors and of course, the researcher.
On p.3 of the Appendix you will see that the connotative analysis aims to reveal the competing ways in which discourses often position speakers as variously powerful or powerless at different points within a single stretch of text.
I’ll give one illustration of how this works, summarised from the Appendix. Here the class is discussing a controversial subject: how to survive in the desert. According to a discourse of peer approval, I noticed that students who were obviously more ‘popular’ with their peer groups were much more likely to get access to the floor, and then to ‘command’ the floor than students who were less popular. This was true of Anne, Joe, Rebecca and Damion in the transcripts. However, I comment on how this behavioural pattern was simultaneously upheld and undercut by the competing discourse of gender differentiation. In other words, when discussing controversial subjects, the class appeared to divide along gender lines, and this served the interest of boys more than girls. For example, I noticed that popular male speakers like Joe and Damion were forcefully backed up by a support group of male peers who actively blocked contributions from other students (see ll.31—34, first extract). This group was particularly active in heckling and interrupting the more outspoken female speakers like Anne and Rebecca (see ll.37—8). However, when popular female students spoke, they did not appear to have an equivalent female support group to block other students’ contributions, possibly because a discourse of collaborative talk had taught them it was rude to interrupt. So, Anne and Rebecca had an altogether harder job to get access to the floor and then maintain their contribution. What was ultimately impressive however, was that, according to a discourse of teacher approval, the examiners recognised the efforts of Anne in particular to sustain her point of view in the face of persistent male barracking, and she was awarded a high grade for this exercise—indeed, a higher grade than the two popular boys. So, from a feminist perspective, Anne could certainly not be constituted as a victim, but nonetheless she had a much more demanding challenge than equivalent boys to get herself heard. In contrast, Rebecca found it more difficult to resist interruptions from the floor, spoke less, and was awarded a more mediocre grade.
So, by using the FPDA approach to identify and analyse the workings of the three competing discourses in relation to each speaker’s contributions, I was able to show that some students are much better placed than others to take advantage of the examination criteria of being an effective speaker. These were most likely to be male, popular, approved by the teacher and able to code switch between collaborative and commanding styles of speech. But girls like Anne were notable exceptions to this pattern.
Potential value to educational researchers
So, to sum up, I have pointed out the relevance and value of FPDA to my own research. but what might it offer other researchers? I very much see FPDA as supplementary to approaches such as CA and CDA, not as a rival to them. But specifically, FPDA (OHT):
- Combines an in-depth, denotative study of spoken interaction with a more wide-ranging connotative analysis of the patterns of power relations within talk
- Addresses the complexity and fluidity of speaker identity within a single text, by analysing the way this is constructed by interwoven and competing discourses
- Challenges the authority of the single-authored account and suggests ways of incorporating multiple voices in the research account
- Offers an alternative to the emancipatory research paradigm by drawing a range of more open-ended, multi-faceted conclusions
*The Appendix was available as a hand-out at this year’s conference. For copies of the Append ix, or any of the articles listed below, please email me on:
(the quicker option)
or
REFERENCES
Barthes, R. (1973) Mythologies. (London: Granta)
Baxter, J. (1999) Teaching girls to speak out: the female voice in public
contexts, Language and Education, 13, 2, pp.81—98.
Baxter, J. (2000a) Going public: teaching students to speak out in public
contexts, English in Education, 34 (2), pp. 26—34
Baxter. J. (2002) A juggling act: a feminist post-structuralist analysis of girls’ and
boys’ talk in the secondary classroom. Gender & Education 14, (1) pp. 5—19
Baxter, J. (2002) Competing discourses in the classroom: a post-structuralist analysis
of girls’ and boys’ speech in public contexts. Discourse & Society 13 (6) pp.
827—42
Baxter, J. (2003) Positioning Gender in Discourse: a feminist research
methodology. Palgrave: Macmillan
Bergvall, V. (1998) Constructing and enacting gender through discourse: negotiating
multiple roles as female engineering students. In V.L. Bergvall, J.M. Bing & A.F. Freed, Rethinking Language and Gender Research. (Harlow: Longman)
Bucholtz, M.,Liang, A.C., & Sutton, L. (eds), (1999) Reinventing Identities: The
Gendered Self in Discourse. New York: OUP
Davies, B. & Banks, C. (1992) The gender trap: a feminist post-
structuralist analysis of primary school children’s talk about
gender, Curriculum Studies 24, 1, pp. 1-25
Davies, B. (1993) Shards of Glass: Children Reading and Writing beyond Gendered
Identities. (Sydney: Allen & Unwin)
Davies, B. (1997) The subject of post-structuralism: a reply to Alison Jones,
Gender & Education 9, pp. 271—283
Foucault, M. (1972) The Archeology of Knowledge and the Discourse on Language.
(New York: Pantheon)
Foucault, M. (1984) What is enlightenment? In P. Rabinov (Ed) The Foucault
Reader (London: Penguin)
Francis, B. (1998) Power Plays: children’s constructions of gender,
power and adult work. (Stoke-on-Trent, Trentham Books)
Jones, A. (1993) Becoming a ‘girl’: post-structuralist suggestions for
educational research, Gender and Education, 5, 2, pp.157—66
Jones, A. (1997) Teaching post-structuralist theory in education: student
resistances, Gender in Education, 5 (2), pp. 261—269
Northern Examination & Assessment Board (1999) Syllabus for 1999/2000:
English(Manchester: NEAB)
Walkerdine, V. (1990) Schoolgirl Fictions. (London, Verso)
Walkerdine, V.(1998) Counting Girls Out: Girls and Mathematics
(London, Falmer Press).
Weedon, C. (1997) Feminist Practice and Post-structuralist Theory. 2nd edition,
(Oxford, Blackwell)
Wetherell, M. (1998)Positioning and interpretative repertoires: conversation
analysis and post-structuralism in dialogue, Discourse & Society, 9, pp.
387—412
1