Language, power and gendered identities: the reflexive social worker.

Bagshaw, Dale

Abstract: This paper explores the implications of feminist poststructuralist ideas for the theory and practice of social work, with a particular emphasis on the links between power, knowledge and language as discourse, and the constitution of gendered identities. The author explores the power of language to define our gendered identities, with illustrations drawn from research she has conducted in South Australia in the areas of domestic violence, verbal abuse and the construction of gendered adolescent identities in secondary schools. She argues that social workers should emphasise the importance of reflexivity in their practice by developing awareness of the effect of dominant cultural and professional discourses on their lives, in order to more fully understand the standpoint of the 'other', and by empowering their clients to identify, resist and challenge discourses and their effects. By conceiving of masculinities and femininities as discursive phenomena, offering a range of possible subject positions, there is greater potential for provoking change than if masculinity or femininity is conceived as an essence.

Introduction

Postmodernism is not a singular, unambiguous enterprise but is embedded in a variety of theoretical and political orientations. Penna and O'Brien believe that the term postmodernism describes the chaotic and confusing cultural, social and political changes which are occurring in Western society. It can be seen as a reaction to modernism, which until recently has dominated much of social work theory and practice. A modernist culture, built over the last few centuries around forms of rationality, self-discipline and bourgeois values is succumbing to the effects of rapid technological and economic change (Penna & O'Brien, 1996).

The weakness of modernist thinking is the search for unitary definitions and the reduction under one label of complex clusters of thought (Game, 1991). However there is now a realisation that the grand or meta-theories of modernity (such as Marxism) do not adequately explain the human condition. New cultural forms are emerging, challenging traditional values and ways of thinking.

Jessup and Rogerson locate poststructuralism as the theoretical stream within postmodernism. It has grown out of the ideas and practices of postmodernism but deals with the implications of language as the producer of the social and cultural world (Jessup & Rogerson, 1999). Most of the postmodernist literature is heavily influenced by poststructuralism, which brings into focus the cultural dimensions of politics and the political dimensions of culture. Poststructuralists are concerned about relations of domination and submission located within cultural processes, the different origins and sites of power, whose interests are served by the exercise of power, and how concepts of 'normality' are fashioned and subjectivities are positioned in a range of different historical locations (Penna & O'Brien, 1996). Positivist research is seen by poststructuralists to have reproduced a colonising discourse of the 'other' without acknowledging that representations of 'self' and 'other' are always politically situated. Fine, for example, points out that 'self' and 'other' are "knottily entangled" and positivist paradigms obscure the nature of the dominant-subordinate, imperialistic relationships between social science researchers and the researched under a veil of 'neutrality' or 'objectivity' (Fine, 1994).

Modernists tend to value scientific explanations, objectivity, rationality and neutrality and the search for universal 'truths'. Positivist theorists and researchers adhering to modernist ideas generally only accept knowledge that can be seen through evidence--our own experience or observation. Essentialism (categorising) and dualistic thinking underpin these theories. These ways of thinking tend to promote simplistic ways of viewing the world and ignore the complexity of concepts such as 'power', 'neutrality' and 'identity'. They also promote adversarial ways of thinking about problems, issues or conflicts, encourage structured, linear, solution-oriented approaches such as problem-solving, and ignore transformative processes and the power of language to define reality and meaning (Bagshaw, 2000).

Postmodern feminist theorists offer a more complex understanding of post-industrial society and value conflict, complexity, diversity and the co-existence of multiple 'truths' and identities (Haraway, 1988; Hartsock, 1996; Hoff, 1996; Weedon, 1987; Worell, 1996). In particular, they acknowledge the links made by Foucault between knowledge, language and power and emphasise how power relations are expressed in discourse. They highlight the power of language (dominant discourses or stories) to define our subjectivity or identity, reality and meaning (McHoul & Grace, 1993). For example before domestic violence was formally recognised by professionals in Australia in the 1980's, and discussed publicly, victims did not have a language to give meaning to, or describe, their experience and their plight was largely ignored. Postmodernist approaches, such as those based on transformative and narrative ideas, require social workers to think and work in very different ways to those based on modernist ideas (Payne, 2005).

Language, power and social work

The French philosopher, Michel Foucault, highlighted the central function that professions such as social work play as points of organisation and distribution of power and control (Foucault, 1977). He also highlighted the power of language (or discourse) to reflect and shape the world. Foucault's concept of power is closely aligned to knowledge and refers to the relationship between categories of thought and categories of experience. Systems of definition and codification of knowledge, he demonstrated, are often developed within institutions, such as universities, schools, prisons, courts and government departments, whose agents have vested interests in defining what counts as knowledge and in codifying knowledge in particular ways. He argued that systems of codification are political and lead to norms, deviations, patterns, understanding and action and define what is appropriate learning and knowledge (Foucault, 1970, 1977a, 1980a; McHoul & Grace, 1993).

Social constructionism (Burr, 2003) forms part of the postmodernist and poststructuralist complex of ideas. Social constructionists suggest that 'truths' or ideas about reality are social constructions--that is, products of social discourses that both emerge out of and shape social processes. Payne points out three areas of social construction important to social work; the political-social-ideological arena which forms the policies that guide agencies, the agency-professional arena and the client-worker-agency arena, with each influencing the other (Payne, 2005, p. 17). For example, he sees 'client hood' as a social construction, a matter of perception held by a variety of people, including policy makers, social workers and the clients themselves. "In this sense, social work is a reflexive process in which the clients change workers and the nature of social work and therefore also change the 'theory' of social work" (Payne 2005, p.19).

Poststructuralists assert that people's lives and identities are shaped by the meaning they give to their experiences, which is in turn shaped by and reflected in dominant language and cultural practices ('normalising' discourses) in society and their historical position in the social structures (Weedon, 1987). Foucault identified the crucial role of discourse in producing and sustaining hegemonic power, such as patriarchy (Foucault, 1980). Particular power relations are seen to produce ways of thinking and knowing that inevitably reflect the power relations. He emphasised the challenges contained within marginalised or unrecognised discourses and highlighted that not all discourses are equal. Stories and 'truths' are inevitably framed by dominant cultural discourses that specify what is 'normal' or 'healthy' and what can or cannot be talked about, by whom, when and in what contexts. The voices of disempowered people tend to be marginalised, subjugated or silenced by the voices of people in power, whose knowledge tends to be privileged and accepted as 'the truth' (McHoul & Grace, 1993).

The dominant discourses in our society have tended to be ageist, racist, sexist and homophobic (Hartman, 1991) and throughout recorded history such discourses have been used by legal and social science professionals to justify categorising people as '(un)deserving', '(ab)normal', '(dys)functional', '(in)competent', '(mal)adjusted', 'subversive, 'delinquent' or 'deviant'. Poststructuralists seek to understand how power and ideology work through systems of discourse and how words and their meaning play a part in shaping attitudes and behaviours, or 'performances', for example of gender. From this perspective words carry intentionality and can serve to harass, disparage and marginalise people from relatively powerless groups such as people from culturally and linguistically diverse backgrounds, indigenous people, lesbians and gay men. Hartman stresses that practitioners can never discount the power of language (as discourse) to inflict irreparable harm. To quote:

 indeed, we, along with the people who are oppressed must continue
 to challenge the dominant discourses that attempt to marginalise
 groups on the basis of such categories as colour, sex, age or
 sexual orientation (Hartman, 1991, p.276).

Discourse and discourse analysis

In the early 1970's, Michel Foucault reconceived language as discourse--the producer of meaning, not the reflector of reality (Foucault, 1972). He viewed discourse "as a social and political entity, the means by which what we know of the world can be created (rather than simply represented)" (McHoul & Grace, 1993, p.13).

A 'discourse' from Foucault's perspective is whatever constrains or enables writing, speaking and thinking within specific historical limits. Thus for Foucault, discourse analysis became a powerful strategy for change. From Foucault's perspective discourse analysis should examine how access to forms of language and how ways of using language have become 'normal' and dominant through complex historical processes. He highlighted that in any given historical period we can only speak or think about a given social practice (such as domestic violence, masculinity and femininity) in certain specific ways and not others. A gendered discourse can therefore be defined as:

 the production and interpretations of sets of related utterances
 ... which effect and sustain the different categorisations and
 positions of women and men (Black & Coward, 1998, p.111).

Most analyses of power portray it as negative and repressive, however Foucault stressed the productive nature of power in modern times (Foucault, 1977). For Foucault, knowledge is power (he wrote 'power/knowledge' as one word), and he saw one's identity or sense-of-self (which he called 'subjectivity') as being constructed through discourse, language or discursive practices (Foucault, 1972, 1980a). He argued that 'power/knowledge' subjects people to normalising 'truths' which shape their lives and realities, for example through notions of what constitutes a 'healthy' body or personality, or a 'normal' or 'functional' male or female.

Verbal abuse: marking the boundaries of gender

In my recent research with women who have been subjected to domestic violence (Bagshaw & Chung, 2000; Bagshaw, Chung, Couch, Lilburn, & Wadham, 2000) and with adolescents involved in conflict in schools (Bagshaw, 1998; Rigby & Bagshaw, 2001) the majority of subjects surveyed and interviewed reported that verbal abuse can be far more damaging to their self-esteem in the long-term than other forms if abuse, including physical abuse, posing a challenge to the prevailing 'sticks and stones' myth. In fact if one views the 'sticks and stones' playground taunt in its entirety, there is a suggestion that words may be more hurtful than the first two lines of the rhyme suggests:

 Sticks and stones will break my bones
 But names will never hurt me.
 When I'm dead and in my grave,
 Think of the names you called me (Turner, 1969).

For my doctoral thesis: Verbal abuse and adolescent gendered identities (Bagshaw, 2004), I investigated the role and function verbal abuse plays in the social construction of masculinity and femininity in a wide range of secondary schools in South Australia, with a particular focus on two State secondary schools. From the findings of a discourse analysis of qualitative data gathered from 832 South Australian year nine adolescent males and females (13-15 years) in a large survey in 7 schools (n = 652), 17 focus groups in a further 7 schools (n = 146) and 32 in-depth interviews in the two State schools, it was clear that language, in particular verbal abuse, played a powerful role in defining the gendered identities of the adolescent boys and girls involved. The total study included teachers and Year 9 students from State and private, mixed-sex and single-sex schools in South Australia and the majority of participants reported that teachers and counsellors paid far more serious attention to physical abuse than verbal abuse, in spite of emerging evidence that the latter is potentially more harmful, especially in the longer term (Rigby & Bagshaw, 2001).

A discourse analysis of the in-depth interviews with 32 male and female Year 9 students in two State secondary schools in the Northern suburbs of Adelaide demonstrated how language and abuse played a powerful role in determining and policing the boundaries of 'normal' or 'appropriate' masculinities and femininities in their peer group, by elevating some forms of masculinity or femininity and denigrating or marginalising other forms. An analysis of repeated statements in the interview texts demonstrated discursive constructions of hierarchies of masculinities and femininities in the dominant discourses of the peer group (see Table 1, Appendix). These constructions and the related abuses were largely unchallenged by the teachers or by the students, who were often fearful of being denigrated or abused themselves (Bagshaw, 2004).

The overall findings of this study illustrated how hegemonic masculinist discourses in schools can construct the boundaries of gender. The adolescents in the study described how the dominant discourses among their peers prescribed 'appropriate' or 'cool' ways to be 'masculine' or 'feminine', with boys and their female followers more often than not dictating and policing the norms. The girls and boys who were categorised as 'cool', 'powerful' and/or 'popular' were discursively placed in what students described as the "upper level" of the peer group, along with some girls who they described as "butch" because they fought and won physical fights. In my analysis I described another small group of boys and girls in the upper level as 'smart'. Notably, they were not named by their peers when asked who was 'popular' or 'powerful' in their year, but when I asked students who they admired the most they were prominent. These 'smart' students managed to resist the dominant gendered discourses, which defined the boundaries of behaviour for many others, and at the same time were held in high esteem. They were generally described as attractive, good 'all-rounders' and were often high achievers in both sport and school work. The students in the 'middle level' of the hierarchy mostly complied with the dominant discourses, although some managed to resist or rebel and avoided being abused by gaining support from others in their group. Those who were discursively designated to the 'lower level' were regularly abused as they fell outside of the norms constructed by the discourse ('fat', 'ugly', 'sluts' etc) and were marginalised or isolated. These included students who were structurally subordinated due to poverty, disability or ethnicity ('povo', 'Minda', 'retard', 'crippo', ching chong' etc).

To give an example of the effect of one dominant gendered discourse, the discourse on body image, a number of girls who could be categorised (from my perspective as a much older, Western, middle-class, white, female social worker) as being of 'normal' or 'healthy' weight in relation to their height and to their peer group (between size 10 and size 12) described themselves as "fat" and said they were regularly insulted about their weight by their peers, in particular by males in the 'cool' group. They reported that only "very thin" (size 8), "pretty" girls who dressed in "skimpy, sexy, fashionable" clothes were seen as sexually desirable by their male peers in the 'upper level' of the peer group. All of the girls interviewed were concerned about their weight and size and many reported that they felt depressed or anxious about their weight most of the time, in spite of being relatively thin. Some said they wore 'baggy' tops to cover up their size and were preoccupied with diet. A few reported having episodes of bulimia and anorexia, one since she was eight years of age. Some young boys also reported being anxious about their size, either seeing themselves as too short or too heavy, but not in such a defining way as for girls. Students stated that if a boy is "overweight and confident" his size can stand him in good stead in physical fights.

Boys and girls tended to be conscious of the importance of "attitude" and appearance and avoided, if possible, dressing and behaving in ways that made them vulnerable to taunts and insults from their peers. To be accepted by the 'popular' and 'powerful' group (these two words were used synonymously), or to avoid being abused, boys and girls dressed in the 'in' clothes specified by that group. For example, the boys in each school wore specific American brand-name clothes and shoes and fashioned their hair in the particular way prescribed by their peers.

Many boys, including the 'smart' boys, stressed they did not want to be seen as being studious or getting high grades for fear of being classed as a "girl", "geek", "nerd" or "loser". Both boys and girls admitted to insulting others in order to be accepted by their peers. The most common insults were linked to gendered appearances or performances. Boys admitted to feeling pressured into abusing others to avoid being called a "wuss" or a "poofter". The most insulting words for boys were those that feminised them ("girl", "wuss", "poofter") and the most insulting word for most girls was "slut" which was commonly used by girls toward each other. It was much more insulting for boys to be called words which feminised them than for girls to be called "butch" or some other name which denoted them as being masculine.

Foucault saw power operating through discourses such as these in which knowledge, meaning and truth, as well as identity are produced and perpetuated. He did not see power as the domination over others by an individual or a group, which is often how it is conceived. Rather he believed that discourses always function in relation to power.

 Power must be analysed as something which circulates, or rather as
 something that only functions in the form of a chain. Power is
 employed and exercised through a net-like organisation....
 individuals are the vehicles of power, not its points of
 application (Foucault 1980, p. 98).

Foucault believed that events, such as gendered conflicts between peers at school, happen according to certain constraints, rules or conditions of possibility, which are also defined by specific historical conditions (McHoul & Grace, 1993). He also noted that language and procedures that harness knowledge all involve some form of unequal relationship. Judgements about so called 'normal' levels of functioning are made by one person or group of another, based on the knowledge possessed by the former, who can be an 'expert' such as a school counsellor or teacher, or a member of the 'popular/powerful' or 'cool' group in a school who is given the power to define, categorise and label their peers and thereby to shape their gendered identities.

Foucault was intent on exposing the operations of power and was particularly interested in the way that hegemonic or global forms of power, such as patriarchy, rely on practices that exist in micro-levels of society such as in families and schools (Foucault, 1980). He noted that the modern system of power is not hierarchical and centralised, but decentred and localised. He traced the history of institutions through which these practices were developed (such as schools) and highlighted how these systems actually recruited people into willingly collaborating in the disciplining and policing of their own lives (Foucault, 1988). He noted that this collaboration is rarely conscious but operates in relation to norms that are seen as 'truths' in the dominant culture. This has implications for the way that social workers intervene in the lives of their clients.

Language, gender and male hegemony

Language is laden with social values and both carries ideas and shapes ideas (Whorf, 1976). Symbols, or representations of our 'reality,' are also circumscribed by the limitations of language. Black and Coward assert that "language has a material existence. It defines our possibilities and limitations, it constitutes our subjectivities" (Black & Coward, 1998, p. 114).

Kaplan believes that language is the most important of all forms of human communication and that it is through the acquisition of language that we become social beings (Kaplan, 1998). In a society structured along a series of unequal divisions, there are a number of groups who have power in relation to other groups (whites, males, managers, professionals) and the forms of domination and subordination are not always identical (Black & Coward, 1998). It is therefore crucial to identify the relationship between what is said and who said it when analysing discourses.

Most contemporary feminist linguistic scholars view language and gender as social and cultural constructs (Cameron, 1998). They have made us aware that sexism is located in the way meaning is constructed in particular discourses. Their critical discourse analyses have exposed sexist assumptions in communication and sexist meanings that are viewed as legitimate and normal in dominant discourses. For example in Western culture the statement 'you think like a woman' can be construed as an insult, especially if directed to a man, based on a widely held belief in our culture that women think illogically. On the other hand, 'you think like a man' is rarely viewed as an insult and can be seen by some as a compliment.

Language influences our attitudes and behaviour and can be used to reinforce harmful or hurtful stereotypes, such as those that are ageist, sexist, racist and so forth. Feminist activity over the past 25 years has sensitised many people to the non-neutral or sexist nature of language, reflecting notions of what is 'normal or deviant', 'masculine or feminine', 'central or peripheral'. Feminist scholars have highlighted that women are frequently the victims of male oppression in both communication and interpretation in discourse, for example they are frequently interrupted or their contributions are trivialised or ignored (McConnell-Ginet, 1998). Many languages have an underlying pattern whereby 'male' is positive and 'female' is negative--a social construction which is often taken for granted as reality.

In examining the connotations of language, Doyle notes that feminine terms are often negative or derogatory, whereas parallel masculine terms are often positive or non-judgemental (for example ladies' man versus man-eater). Doyle also notes that many negative terms for women have no masculine counterpart (for example nymphomania), and even when counterparts exist they are seldom used (Doyle, 1998). Black and Coward add that attributes of the male can disappear into a non-gendered subject, whereas women are often precisely defined as specifically feminine, and frequently as sexual categories such as "whore, slag, mother, virgin, housewife" (Black & Coward, 1998). This is evident in dominant discourses around domestic violence and sexual harassment, which allow males to deny the effect of their gendered subjectivity on females and constrain them from taking responsibility for their actions. For example, in spite of research evidence to the contrary, Pease noted in the early 1990's that there was still a conceptual tendency to misdirect the causes of domestic violence toward women (Pease, 1991). Judges, magistrates, service providers and others in the Australian community have often misleadingly "redirected responsibility for the violent behaviour from the perpetrator to [other] factors such as alcohol abuse, stress or the fact that women may have somehow "asked for it""(Pease, 1991, p. 26). These suggestions were affirmed by the findings of our domestic violence study in South Australia: Reshaping Responses to Domestic Violence (Bagshaw et al., 2000). Most women in this study were so influenced by this discourse that they blamed themselves for causing the abuse directed against them. Victim-blaming discourses serve to deny women individual agency and to authorise and naturalise men's violent behaviour (Lane, 1999; Stapleton & Lane, 1999; Walker, 1995).

In her analysis of the Sun's reporting of crimes of sexual violence, Clark noted omissions and commissions in language that subtly shifted the blame from the perpetrator to the victim. For example in reports of violence against women, details were given about the victim in a way that labelled her as wife, unmarried mum, mother of two, prostitute, and the attacker (male) was often excused or made invisible, except where the victim was a mother, a child, a girl or old woman (Clark, 1998). She concluded that:

 Naming is a powerful ideological tool. It is also an accurate
 pointer to the ideology of the namer (Clark, 1998, p. 184).

Feminist scholars have also challenged structuralist and masculinist research methodologies, theories and discourses in the field of victimology which have focused on the individual and thereby implied that victims contribute to violence against them (Kelly, 1987; Smart, 1989; Stanko, 1985). Since the late 1950s, some victimologists have implied that victims precipitate their own victimisation and have ignored or distorted gendered analyses of violence. Attempts have been made to identify different types of victims and/or to explain how victims have precipitated violence against them with the implication that the victim needs to change or is in some way responsible for, or contributes to the violence. Feminists have pointed out that the effect is to ignore those responsible for the violence and to leave the violence unchallenged (McCarthy, 1997).

In my research with the Year 9 adolescents in schools (Bagshaw, 2004), students reported that there is a double-standard--one for boys and one for girls--and boys tend to set the scene. It was not uncommon for girls to abuse each other when a male had a sexual relationship with both of them at the same time, calling each other "sluts", "whores", "tarts" and so forth, not considering that they could or should take their anger out on the culprit (the boy). On the one hand, a girl was called a "slut" or a "tart" if she dressed or behaved in a sexually provocative way, was relatively powerless (Madonna, who is perceived as powerful, can behave as a 'slut') and/or had sex with a male outside of a monogamous relationship. On the other hand, students reported that young males tended to brag about their sexual exploits with girls, which aided their popularity, and were labelled positively, for example as "legends". There was no equivalent to the words 'whore', 'slut' or 'tart' for males in the language used.

'Popular' boys (who were also seen as powerful) tended not to pay attention to girls unless they wore facial make-up, tight-fitting, 'sexy', fashionable clothes and were 'thin' and 'attractive' according to male-defined norms. They tended to insult girls who were they saw as overweight ("fatty", "whale"), or were considered to be 'unfeminine' and powerless ("butch", "leso") or wore unfashionable clothes and hairstyles ("geek"), as did the 'popular/powerful' girls who were their followers. Hegemonic patriarchal discourse in schools affected the gendered performances of boys as well as girls. If boys did not fit in with the normalising discourse, for example by appearing to be disinterested in girls as sex-objects, studying hard, achieving academically at school, or if they dressed in inappropriate ways, they were taunted and called names such as a "poofter", "wuss", "girl", "loser" or "geek" (Bagshaw, 2004).

Many male and female adolescents I interviewed worried constantly about their appearance, what to wear and how to behave at school for fear of being marginalised and abused if they did not fit with the discursively defined 'appropriate' ways to be masculine or feminine. The norms were in part established and maintained by dominant individuals and groups amongst their peers, mainly males, but also reflected the dominant discourses in the wider patriarchal society. Those who rebelled against the discourses were either extremely confident or secure in themselves and/or their relationships (those I designated as 'smart') or did so at their peril.

Implications for social work intervention

The idea that the postmodernist view of the self, or identity, is a social construct that occurs through discursive relations with others is an important idea for social workers to grasp and understand. This view implies firstly, that social workers need to be careful in their choice of language, interpretations they make and the meanings they ascribe to a person's behaviour or identity in a particular situation. Essentialism can contribute to categorising and labelling clients and their problems in a way that impedes opportunities for client-centred practice and reifies and reinforces the power/knowledge of the worker. Traits such as those linked to ethnicity, age, sexuality, ability or gender, should not be automatically assigned to a person's identity, as any one of these factors may not be seen by that person as relevant or important in a particular context or at a particular point in time. From a poststructuralist perspective it is the clients who supply the interpretive context for determining the meanings of events, which requires the social worker to take a self-reflexive approach to practice (Gergen & Gergen, 1991; Jones, 1992; Mauthner & Doucet, 1998; Probyn, 1993; Steier, 1991).

Reflexivity is based on social constructionist approaches to inquiry; that is on the idea that "worlds are constructed, or even autonomously invented, by 'scientific' inquirers who are simultaneously participants in their worlds" (Steier, 1991, p. 1). It implies acceptance of the idea that knowledge is embedded in the constructing process. Steier (1991, p. 2) variously describes reflexivity as: "turning-back of one's experience upon oneself" and "being conscious of ourselves as we see ourselves" (p. 5). Self-reflexivity recognises that our practices are culturally specific (Fraser & Nicholson, 1990) and involves "being explicit about the operation of power" (Ribbens, 1989, p.162). Alldred stresses that reflexivity should involve:

 critical scrutiny ... that acknowledges that analysis is an
 artefact, produced in a particular moment by a person occupying
 particular subject positions, and within the particular power
 relations described (Alldred, 1998, p.147).

In self-reflexive practice it is recognised that it is impossible to be 'neutral' and the influences of characteristics such as gender, race, class, ability, age and sexuality on the worker's relationship with their clients are critically examined.

In social work we are provided with an opportunity to understand the stories and themes that have shaped a person's life and the degree to which the normalising power of dominant discourses have included, excluded or marginalised people. By focusing on the meaningful areas of the lives of people, and by assisting them to be reflexive and to analyse or deconstruct the influence of dominant discourses on their lives, we will be in a better position to empower our clients. Narrative ideas, such as separating the person from the problem and encouraging the person to provide an account of the effects of the problem on their lives and on their interactions with others, leaves people free to explore alternative and preferred ways of knowing and being (White, 1989; Winslade & Monk, 2000).

Understanding poststructuralist ideas about language as discourse and the close alignment between power, knowledge and discourse enables the social worker to ask new questions and see new things about the nature of social relations, in particular from the perspective of those who are marginalised and/or relatively disempowered. From this perspective social workers should strive to:

* understand how power relationships both produce and are produced by discourses;

* know how to challenge dominant discourses and practices through discourse analysis (disclosing ideas, beliefs, norms) and not collude with these discourses and their outcomes for individuals;

* be interested in the lived experience of people and discursive effects on the subjectivities of people;

* develop communication competence in the knowledge that clients supply the interpretive framework that is necessary for determining appropriate intervention;

* use enabling, empowering strategies whilst resisting the imposition of top-down agendas;

* be inclusive, especially ensuring the inclusion of previously excluded or marginalised voices such as those of indigenous people, migrants, people with disability, gays, lesbians, women or children;

* value the transformative power of the process, not just the outcome of intervention.

Conclusions

Social workers should emphasise the importance of reflexivity in their practice by developing awareness of the effects of dominant cultural and professional discourses on their lives in order to more fully understand the standpoint of the 'other', and by empowering their clients to identify, resist and challenge discourses and their effects on their lives. Deconstructing and reconstructing binary gendered discourses may enable males and females to relate to each other in more respectful and affirming ways as they progress through their lives.

The study of verbal abuse in secondary schools in South Australia, described in this paper, has demonstrated how dominant, binary, gendered discourses can be destructive to adolescents as individuals, and to their ongoing relationships with others. The findings suggest a need to introduce strategies to enable staff and students in these schools to identify, analyse and challenge the discourses and abuses that underpin, justify and promote abusive behaviours, such as those that divide people into categories of inclusion and exclusion (male/female, 'cool'/'geek', 'slut'/'legend' and so forth). These strategies will need to involve the wider community and will require a variety of changes including a revision of school policies, changes to the education of teachers, counsellors and students and the implementation of a variety of creative programs that are accessible to the different peer cultures.

The interviews that I conducted with young people in schools demonstrated the various ways in which adolescent gendered subjectivities are constituted through discursively defined social structures or categories, and how they continue to speak into existence those same categories and structures through the same discourses, demonstrating Foucault's concept of the power of discourse (Foucault, 1988). Discourse analysis understands language as more than a mode for transmitting information in the process of communication. Language, or the way we talk about things, actively constructs our reality and constructs and reproduces social systems, and therefore has political and social implications (Potter & Wetherell, 1989). The power of discourse lies in its invisibility--it is taken for granted as the status quo. Once a discourse is made visible, it loses its power and opens up possibilities for other options to emerge. By introducing young people to philosophy (for example Foucault's ideas about discourse, power and subjectivity) and to techniques for analysing discourses, such as through deconstructing gendered themes and stories in advertisements, films and plays, they will be provided with an opportunity to interrupt the apparent inevitability of the male-female dualism, thereby opening up the possibility of more flexible and multiple ways of being gendered (Davies, 1989, 1993, 1994). By conceiving of masculinities and femininities as discursive phenomena, offering a range of possible subject positions, there is greater potential for provoking change than if masculinity or femininity is conceived as an essence.

Appendix

Table 1: The discursive positioning of masculinities and femininities
in the peer group hierarchy by Year 9 students in Mitchell High and
Newland High

Assigned Degree of Discursive
power/status masculinity/femininity of boys effect: style of
level of boys in as defined by the dominant masculinity
the peer group discourse
hierarchy

UPPER LEVEL

Upper level 1 Ideal masculinity Hegemonic
 (unattainable) masculinity

Upper level 2 Exaggerated masculinity 'Cool', 'macho'

Upper level 3 Flexible, confident masculinity 'Smart'

Upper level 4

MIDDLE LEVEL

Middle level 5 Rebellious masculinity 'Resistant'

Middle level 6 Conformist, conservative 'Complicit'
 masculinity

Middle level 7

Middle level 8

LOWER LEVEL

Lower level 9

Lower level 10 Failed, feminine or weak 'Subordinated'
 masculinity: 'loser', 'geek',
 loner', 'wuss', 'poofter'

Lower level 11 Structurally marginalised 'Marginalised'
 masculinity--poor, disabled,
 ethnically different: 'povo',
 Minda, retard, 'crippo', 'ching
 chong'.

Assigned Degree of Discursive
power/status masculinity/femininity of girls effect: style
level of girls in as defined by the dominant of femininity
the peer group discourse
hierarchy

UPPER LEVEL

Upper level 1 Ideal femininity Hegemonic
 (unattainable) masculinity

Upper level 2

Upper level 3 Masculinised femininity-- 'Cool', 'macho'
 tough', 'butch', 'macho'

Upper level 4 Emphasised femininity-- 'Cool',
 'sluttish' 'complicit'

MIDDLE LEVEL

Middle level 5

Middle level 6 Rebellious femininity 'Resistant'

Middle level 7 Conformist, conservative 'Complicit'
 femininity

Middle level 8 'Copycats' or 'Wannabees' 'Complicit'

LOWER LEVEL

Lower level 9 Masculinised femininity: Subordinated
 'butch, 'lesso'

Lower level 10 Failed or weak femininity: 'try 'Subordinated'
 hard', 'loser', 'slut', 'loner',
 'geek'

Lower level 11 Structurally marginalised 'Marginalised'
 femininity: poor, disabled,
 ethnically different: 'povo',
 Minda, retard, 'crippo', 'ching
 chong'.

References

Alldred, P. (1998). Ethnography and Discourse Analysis: Dilemmas in Representing the Voices of Children. In J. Ribbens & R. Edwards (eds.), Feminist Dilemmas in Qualitative Research: Public Knowledge and Private Living (pp. 147-170). London: SAGE Publications.

Bagshaw, D. (1998). What Adolescents Say About Conflict in Schools. Children Australia, 23(3), 17-22.

Bagshaw, D. (2000). The Three M's-Mediation, Postmodernism, and the New Millennium. Mediation Quarterly, 18(3), 205-220.

Bagshaw, D. (2004). Verbal abuse and adolescent identities. Marking the boundaries of gender. Unpublished thesis for Doctor of Philosophy, University of Melbourne, Melbourne.

Bagshaw, D., & Chung, D. (2000). Gender Politics in Research: Male and Female Violence in Intimate Relationships. Women Against Violence (8), 4-23.

Bagshaw, D., Chung, D., Couch, M., Lilburn, S., & Wadham, B. (2000). Reshaping Responses to Domestic Violence. Canberra: Partnerships Against Domestic Violence and Department of Human Services.

Black, M., & Coward, R. (1998). Linguistic, Social and Sexual relations: A Review of Dale Spender's Man Made Language. The Feminist Critique of Language: A Reader, 100-118.

Burr, V. (2003). Social Constructionism. London: Routledge.

Cameron, D. (1998). The Feminist Critique of Language: A Reader (Second ed.). New York and London: Routledge.

Clark, K. (1998). The Linguistics of Blame: Representations of Women in the Sun's Reporting of Crimes of Sexual Violence. In D. Cameron (Ed.), The Feminist Critique of Language: a Reader (Second ed., pp. 183-197). London and New York: Routledge.

Davies, B. (1989). Frogs and Snails and Feminist Tales. Sydney: Allen and Unwin.

Davies, B. (1993). Shards of Glass. Sydney: Allen & Unwin.

Davies, B. (1994). Poststructuralist theory and classroom practice. Geelong: Deakin University Press.

Doyle, M. (1998). Introduction to the A-Z of Non-Sexist Language. In D. Cameron (Ed.), The Feminist Critique of Language: A Reader (Second ed., pp. 149-154). London and New York: Routledge.

Fine, M. (1994). Working the Hyphens: Reinventing Self and Other in Qualitative Research. In N. Denzin & Y. Lincoln (Eds.), Handbook of Qualitative Research (pp. 70-82). Thousand Oaks, California: SAGE Publications.

Foucault, M. (1970). The Order of Things. London: Tavistock.

Foucault, M. (1972). The Archaeology of Knowledge. London: Tavistock.

Foucault, M. (1977). Discipline and Punish: The Birth of the Prison. London: Allen Lane.

Foucault, M. (1980). Power/Knowledge: Selected Interviews and Other Writings 1972-1977. London: Harvester Press.

Foucault, M. (1988). Technologies of the self. In M. L. Gutman & P. Hutton (Eds.), Technologies of the self. Amherst: University of Massachusetts Press.

Fraser, N., & Nicholson, L. J. (Eds.). (1990). Feminism/Postmodernism. New York, London: Routledge.

Game, A. (1991). Undoing the Social: Towards a Deconstructive Sociology. Milton Keynes: Open University Press.

Gergen, K., & Gergen, M. (1991). Research and Reflexivity. London: SAGE Publications.

Haraway, D. J. (1988). Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective. Feminist Studies, 14, 575-599.

Hartman, A. (1991). Words Create Words. Social Work. Journal of the National Association of Social Workers, 36(4), 275-276.

Hartsock, N. (1996). Postmodernism and Political Change: Issues for Feminist Theory. In S. Hekman (ed.). Feminist interpretations of Michel Foucault. University Park, Pennsylvania: The Pennsylvania State University Press.

Hoff, L. (1996). Collaborative Feminist Research and the Myth of Objectivity. In B. Yllo & M. Bograd (Eds.), Feminist Perspectives in Wife Abuse. Newbury Park: SAGE Publications.

Jessup, H., & Rogerson, S. (1999). Postmodernism and the teaching and practice of interpersonal skills. In B. Pease & J. Fook (Eds.), Transforming Social Work Practice. Postmodern critical perspectives. St Leonards: Allen & Unwin.

Jones, A. (1992). Writing feminist Educational Research: Am 'I' in the Text? In S. Middleton & A. Jones (Eds.), Women and Education in Aotearoa 2 (pp. 18-32). Wellington, NZ: Bridgit Williams Books Ltd.

Kaplan, C. (1998). Language and Gender. In D. Cameron (Ed.), The Feminist Critique of Language: A Reader (Second ed., pp. 54-64). London and New York: Routledge.

Kelly, L. (1987). The Continuum of Sexual Violence. In J. Hanmer & M. Maynard (Eds.), Women, Violence and Social Control (pp. 46-60). London: Macmillan Press.

Lane, B. (1999, August 31). She made me do it. The Age, pp. 13.

Mauthner, N., & Doucet, A. (1998). Reflections on a Voice-centred Relational Method: Analysing Maternal and Domestic Voices. In J. Ribbens & R. Edwards (eds.), Feminist Dilemmas in Qualitative Research: Public Knowledge and Private Lives (pp. 119-146). London: SAGE Publications.

McCarthy, T. (Ed.). (1997). Rethinking Theories of Victimology. Men's Violence Against Women. Thousand Oaks, California: SAGE publications.

McConnell-Ginet, S. (1998). The Sexual (Re)Production of Meaning: A Discourse-Based Approach (Second ed., pp. 198-210). London and New York: Routledge.

McHoul, A., & Grace, W. (1993). The Foucault Primer: Discourse, power and the subject. Melbourne: Melbourne University Press.

Payne, M. (2005). Modern Social Work Theory (Third ed.). New York: Palgrave Macmillan.

Pease, B. (1991). Evaluating Treatment Programs for Violent Men: Dilemmas and Priorities. Medlowbath: The NSW Local Domestic Violence Committees.

Penna, S., & O'Brien, M. (1996). Postmodernism and Social Policy: A Small Step Forwards? Journal of Social Policy, 25(1), 39-61.

Potter, J., & Wetherell, M. (1989). Discourse and Social Policy: Beyond Attitudes and Behaviour (Second ed.). London: Sage.

Probyn, E. (1993). Sexing the Self: Gendered Positions in Cultural Studies. London: Routledge.

Ribbens, J. (1989). Interviewing: an "unnatural situation"? Women's Studies International Forum, 12(6), 579-592.

Rigby, K., & Bagshaw, D. (2001). The reported consequences of negative interactions with peers among Australian school children. Children Australia, 26(4), 36-41.

Smart, C. (1989). Feminism and the Power of Law. London: Routledge.

Stanko, E. (1985). Intimate Intrusions: Women's experience of male violence. London: Routledge & Kegan Paul.

Stapleton, J., & Lane, B. (1999, 31 August). "Dinosaur" magistrate rapped. The Age, pp. 4.

Steier, F. (1991). Research as Self-Reflexivity, Self-Reflexivity as Social Process. In F. Steier (Ed.), Research and Reflexivity (pp. 1-11). London: SAGE Publications.

Turner, I. (1969). Cindarella dressed in yella. Australian children's play-rhymes. Melbourne Australia: Heinemann Educational.

Walker, M. (1995). Interpreting the Figures: Increase in women's violence or just more masculinist legal tactics? The Australian Feminist Law Journal, 5(August), 123-125.

Weedon, C. (1987). Feminist Practice and Post Structuralist Theory. Oxford: Basil Blackwell.

White, M. (1989). The externalising of the problem and the re-authoring of lives and relationships. Dulwich Centre Newsletter, Summer.

Whorf, B. L. (1976). Language, Thought and Reality. Cambridge, MA: MIT Press.

Winslade, J., & Monk, G. (2000). Narrative Mediation. A New Approach to Conflict Resolution. San Francisco: Jossey-Bass.

Worell, J. (1996). Opening Doors to Feminist Research. Psychology of Women Quarterly, 20, 469-485.

Dale Bagshaw, Associate Professor, School of Social Work & Social Policy, University of South Australia. Email: dale.bagshaw@unisa.edu.au