Social work and gender: An argument for practical accounts.

This article contributes to the debate on gender and social work by examining dominant approaches within the field. Anti-discriminatory, woman-centered and intersectional accounts are critiqued for reliance upon both reification and isolation of gender. Via examination of poststructural, queer and trans theories within social work, the author then presents accounts based upon structural/materialist, ethnomethodological and discursive theories, in order to open up debates about conceptualization of gender. These are used to suggest that social work should adopt a focus on gender as a practical accomplishment that occurs within various settings or contexts.

This article is a contribution to opening up the conversation on gender and social work. It is concerned, more precisely, with the conceptualization and usage of “gender” within social work theory, research, and practice. Although a key feature of everyday life, within social work, gender has what sociologists sometimes call a “seen-but-unnoticed” quality. It is frequently overlooked and, perhaps more importantly, where it is considered, gender is theorized in a number of rather limited ways.

For example, social work is often described as a female-dominated profession, but one in which men disproportionately occupy senior roles. Yet, McPhail has argued that “social work is more correctly described as a female majority, male-dominated profession” (McPhail, 2004b: 325), because, although there are many more women than men in the field, they do not necessarily dominate. This is an important argument since, to describe social work as “female-dominated” suggests that, merely because they are far greater in number, women hold more power. Yet, this disregards some vital points. First, the smaller number of men in the profession may actually hold more institutional power, and, second, a profession like social work is, as with many fields involving the care of others, devalued. Third, the question of how power works within social work institutions, and how this relates to gender, is likely to be a lot more complicated.

Discussions about challenging oppression and discrimination within social work theory and practice are some of the few occasions on which gender is openly acknowledged (Dominelli, 2002a; Mullaly, 2007; Thompson, 2012). Yet these, too, often rely upon limited accounts. Thompson’s text, for example, describes gender as a “fundamental dimension of human experience, revealing an ever-present set of differences between men and women” (Thompson, 2012: 55). While he does go on to point out that social, rather than biological, processes produce gender, it is largely at the level of attitudes that his suggestions for change are leveled. This tends to individualize gender, to see it as a personal characteristic, and to see gender oppression merely as a form of personal behavior or values.

In part, these points relate to the ways in which gender is defined. Second-wave feminism, for example, separated the concept of “sex” from “gender,” in order to show that “gender” refers to a set of social expectations that may be challenged (Oakley, 1972; Unger, 1979). However, sometimes this notion of gender as a set of cultural practices has been reduced to role or identity, so that gender is treated as a preexisting characteristic or property of the individual. Later feminist theories remind us, rather, that gender is a social relationship, based upon the promotion of hierarchy, and one that is reiterated through interactions in everyday life.

This article pays considerable attention to this notion of gender as a form of practice, since it is my contention that much of social work theory actually treats gender as a rather static characteristic. After having reviewed some of the more familiar approaches to gender within social work, I will go on to open up debates via consideration of materialist, interactionist, and discursive accounts, before finally considering what social work theory, research, and practice might learn from these.

How does social work think about gender?

Where social work theory or research does think about gender, we see the influence of feminist and/or sociological theories. Orme’s book, Gender and Community Care, argues that the “gender politics of social work has to include the relationship between the helper and those who require help, and … between the individual and the state” (Orme, 2001: 14). She highlights the disproportionate representation of women in mental health services, elder abuse, and those cared for in the community, pointing out that these are all areas in which gender is usually ignored or invisible or that, when it is noticed, the response is usually to suggest that men and women should be treated differently. Orme argues this “categorisation of femaleness and maleness, femininity and masculinity as dichotomous opposites does not reflect the lived experience of users of community care services” (Orme, 2001: 239).

Scourfield points out that assumptions about gender difference “permeate interventions” in social work (Scourfield, 2010: 2), and he links these with heteronormativity. He makes a case for the analysis of gender as a social category, since the category relates to questions of social inequality (Scourfield, 2002). Christie similarly argues that, within discourses of welfare, persons are gendered, “offering them specific gendered identities and subject positions” (Christie, 2001: 9). In relation to men in social work, he notes that they are often seen as either good (e.g. “male role models”) or bad (e.g. “dangerous/abusers”).

Sociological social work texts see gender as referring to a social or cultural set of ideas reflecting normative assumptions but, although such texts make reference to gender as a practice, they often work at the level of attitudes or values, encouraging social workers to reflect upon their own assumptions about gender (Llewellyn et al., 2008; Sheach-Leith et al., 2011). Treating gender concepts at the level of attitudes is a rather individualized approach, in which it seems to be an interpersonal characteristic only, although there are other texts that consider gender as a practice and insist on its contextualization within late or reflexive modernity (Dunk-West and Verity, 2013).

There are attempts within social work to think about how gender relates to questions of race, class, disability, age, or sexuality, but more often gender is treated as a stand-alone issue. An example of this would be some feminist work on care, which argues that women need to be released from the burden of caring for dependents. Although this point about the effects of state and family reliance upon unpaid care is an important one, work by disabled feminists has pointed out that the category “women” includes those being cared for, and that these arguments position disabled women and men as a “burden” (Morris, 1991). Others have noted the heteronormativity of such arguments, based, as they often are, on an assumed heterosexual couple (Manthorpe, 2003).

However, by far the most regular usage of “gender” within social work is where it is treated as “already given” (Smith, 1990: 159); that is, used as a label referring to an assumed characteristic. Here, the formula runs, “gender causes x.” An interesting example of this would be Failure to Protect: Moving Beyond Gendered Responses (Strega et al., 2013), which examines why, in professional responses to child sexual abuse, mothers are often held responsible via “failure to protect.” In one sense, this is vitally important: why does some social work practice tend to blame mothers and ignore fathers? Why are mothers often held accountable for men’s abuse of children? But, in another sense, the book never really asks how gender works, or is made to matter, in these contexts, and instead frequently treats it as a mono-causal explanation.

This kind of usage of gender is limited for a number of reasons: first, gender may take on a thing-like quality and appear to have agency—“gender causes x.” Second, it treats a group (e.g. men) as homogeneous. But this doesn’t ask if all men are therefore more likely to abuse children, for example. And it doesn’t ask whether all men are equally powerful. Third, it doesn’t really get to grips with just how gender works in a given situation. Fourth, it may lead to simplistic explanations. Of course, it is important to think about why men overwhelmingly commit most forms of sexual violence, but this does not mean “gender causes abuse.” And, lastly, this is a rather interior view of gender. The gender of the person seems to be some kind of characteristic that causes a problem or outcome.

Woman-centered practice?

Much of the feminist social work literature treats gender as a basis for similarity and shared purpose. Hanmer and Statham’s text, Women and Social Work, for example, develops what they term a “woman-centred practice,” and makes the case that, since women are the majority of social workers and service users, a commonality of gendered experience, along the lines of “being female, their relationships with men, children, living within the family, employment and working conditions” (Hanmer and Statham, 1999: 18), forms the basis of social change through social work.

Although the book does acknowledge differences along lines of race, age, disability, class, and sexuality, this notion of commonality, or what Dominelli and McLeod term “non-hierarchical relationships between the social worker and the woman/women she is working with” (Dominelli and McLeod, 1989: 38), has been critiqued for assuming that feminist social work means working with women; that empowerment is the only purpose of such work; that empowerment resolves power dynamics within relationships; and that women’s shared experience means automatic rapport (Baines, 1997; Orme, 2003; White, 2006; Wise, 1990). Hanmer and Statham’s text mentions lesbian, black, and ethnic minority women in relation to forms of diversity, but their description of women’s commonalities relies upon the normative assumptions of whiteness and heterosexuality.

This “sameness” problem has been the target of other social work writings. Lewis’ research argues that both race and gender are mutually constituted, yet within social work they are often treated as separate spheres. She argues that gender and race are experienced differently according to context, and so may have different meanings and effects, even for the same person. So, just as the category of gender must be one that allows for differences, so race, too, must not be treated as already given, as referring to some kind of essential black or white “culture.” In relation to the black, female social workers in her study, Lewis suggests that “‘racial’ and ethnic categories are simultaneously occupied and resisted as a way of mediating a set of working lives which are overdetermined by ‘race’ and gender” (Lewis, 2000: 205–206).

Indeed, if gender is to be seen in its complexity, then this must not be taken solely to refer to women. For some theorists in social work, it is important to think about work with men and fathers, the complexity of men’s position within social work, notions of “masculinity” and the category “men” (Christie, 2001; Cree, 1996; Featherstone et al., 2007; Scourfield, 2003). This also relates to how social work thinks about trans issues and transgender people, a point to which I shall return.

Social work, gender, and intersectionality

One response to this assumed gender sameness, and the treatment of gender in isolation, is to consider intersectionality theory (Mehrotra, 2010; Murphy et al., 2009; Wahab et al., forthcoming). Crenshaw’s argument proposes that the consideration of subordination within single categories, like gender, prevents analysis of race and gender for black women, since the claims of sex discrimination within law are largely based upon experiences of white women in relation to gender only (Crenshaw, 1991). This has been taken up in Incorporating Intersectionality in Social Work Practice, Research, Policy, and Education (Murphy et al., 2009), which argues that social work should consider how oppressions intersect to form interlocking patterns of injustice. This means that attention to gender alone is insufficient, since race and class make a difference, and it also means that any individual might experience both oppression and privilege.

While this goes some way to challenging supposed gender sameness, the authors accept Andersen’s (2005) claim that sexuality does not occupy the same place as race, class, and gender, since it has largely to do with identity–cultural issues rather than political–structural ones. Andersen argues:

sexuality has never been formally used to deny sexual groups the right to vote, nor has it been used in the formal and legal definition of personhood as is historically true of African Americans and other groups. Gays and lesbians have never been formally segregated in the labor market nor denied citizenship because of the labor they provide. (Andersen, 2005: 451)

Murphy et al., while pointing out the need to consider questions of sexuality, accept this view and suggest that sexuality cannot be treated as equivalent to race, class, and gender. Here, then, is an obvious problem with some intersectionality theory. An argument against a hierarchy of oppressions is contradicted by establishment of another. And, as Schilt notes, this separate treatment of sexuality ignores ways in which citizenship is denied to lesbian, gay, bisexual, or transgender people, and also that “gay men and lesbians who have nonnormative gender presentations, who are working-class, and/or who are racial/ethnic minorities are often those who end up being most excluded from legitimate avenues of employment” (Schilt, 2008: 112). Given that authors, such as Collins, argue that “what is needed is a framework that not only analyzes heterosexism as a system of oppression, but also conceptualizes its links to race, class, and gender as comparable systems of oppression” (Collins, 2000: 128–129), this suppression of sexuality analysis in a social work text seems misguided.

Poststructural and postmodern feminist social work

Poststructural and postmodern theories have questioned the notion of identity or experience-based knowledge that features in some feminist work, because poststructuralist theories do not treat language as a reflector of reality, but rather a powerful way of constructing knowledge. Thus, any claims that feminist social work should be based upon validating the experiences of women are thrown into question because those experiences are not merely authentic, they are motivated, linguistic accounts, which aim to achieve certain effects, and they are open to different interpretations.

Feminist poststructuralists also challenge the notion of women’s shared experience, since the category “woman” is itself experienced differently and fractured along race, class, sexuality, disability, age, and other lines (Featherstone, 2001; Morley and Macfarlane, 2011; Pollack and Rossiter, 2010; Rossiter, 2000; Sands and Nuccio, 1992). Of course, this is not merely a poststructural claim. Earlier feminist debates also centered on potential exclusions of the category “woman” by race, sexuality, and so on, but here the concern is more with the powerful effects of language use. So, while Sands and Nuccio’s (1992) arguments for a postmodern feminist social work, based upon difference, diversity, and recognizing the marginalized do not sound particularly challenging, their questions about the potentially oppressive nature of gendered or racialized categories used by social workers raise important concerns regarding the nature of social work knowledge.

Dominelli has argued strongly against “individualistic” postmodern theory, which, she says, does not consider systematic patterns of discrimination along gender lines (Dominelli, 2002b: 34). She also claims that postmodern feminism assumes that power “subsumes any form of opposition” (Dominelli, 2002a: 169). This seems a rather limited reading of feminist postmodern theories, which are not based on notions of the individual subject at all, but are rather concerned with how subjectivity is produced through powerful discourses, interested in how dominant knowledge forms arise, and in how these may be opposed via various forms of subjugated, but not silenced, knowledge. Dominelli, however, argues for woman-centered practice, which seeks equality based on empowerment, listening to the stories, and validating the experiences of women, a point that postmodern theories would reject as both naive and asserting a powerful claim about what kinds of knowledge count.

What such debates demonstrate, of course, is that what constitutes feminist social work is not agreed. White’s study argues, “women social workers’ anecdotal accounts of their experiences were of feminist identities that were fluid, sometimes fragile or even non-existent” (White, 2006: 3). She is also critical of woman-centered practice because this seems largely based upon community and voluntary models that exist outside of state social work. While she is not critical of such feminist work per se, White argues that the woman-centered model of practice is largely “isolated from an analysis of the features of the organisational regime of social work that are associated with its location in the state” (White, 2006: 31).

Postmodern feminist social work theories reject the notion of egalitarian power relations as a fantasy that does not engage with the power dynamics that always exist between social workers and clients, a point also made in earlier work (Wise, 1990). Power is not seen as a one-way street; that is, something always held by social workers over service users. There is no space outside of power relations, and so postmodern thinkers call for reflexivity about power within all practices. The feminist model of empowerment, for example, may be criticized because it sees power as somehow given to the (always) powerless service user by the (always) powerful social worker, but also because the notion of “empowerment” has been co-opted by neoliberal state welfare, so that it replaces any concern for wider structural change with individualized notions of “choice.”

Queer and trans theory

The influence of queer and trans theories on social work has been more limited to date, but where this has been addressed, then the notion of “gender” itself is challenged (Burdge, 2007; McPhail, 2004a; Nagoshi and Brzuzy, 2010; Wahab et al., forthcoming). The dichotomous view of gender is brought into question, as this is a powerful technology for the regulation of persons. Social work writings on trans people generally caution against the reification of gender categories, with phrases such as “gender variant” or “gender nonconforming” also being used (Davis, 2009; Hartley and Whittle, 2003; Kahn, 2014; Martin and Yonkin, 2006). Yet, at the same time, there may be a tendency, in some accounts, to theorize “transgender identity” based upon developmental stages, or gender as something fixed by the age of 3 (Mallon and DeCrescenzo, 2009). Spade, however, argues that the vulnerabilities of trans people, especially those marginalized due to poverty, are the result of “legal and administrative systems of domination … that employ rigid gender binaries” (Spade, 2011: 13). Queer and trans theories thus argue that the category “gender” should be questioned, and it is to this that I now turn.

Opening up the debate on “gender”

… ‘enough already with gender!’ The reason for such exasperation has to do with the way gender has become operationalized in ‘gender research projects’ … In many of these instances, gender is taken for granted as the point of departure for a set of descriptions of social practices, understood as an adjective that qualifies established objects of social science: gendered work, gendered performance, gendered play. In fact, there is little inquiry on the production of difference… (Butler, 2011b: 21)

Collins’ (2000) Black Feminist Thought argues that feminist work on gender has largely reflected the experiences of white, middle-class women. Writing mainly about African American women’s experiences, Collins argues that many arguments within feminist theory, such as the role of women as carers in the home or the oppressive nature of family life, do not consider black women’s experiences of (often poorly paid domestic) work or of the positive role that black families might play in helping to challenge racism. This is not to valorize “the black family” or to deny the significance of sexism, but rather to insist that feminism, and any account of gender relations, must take questions about race on board. As well as this absence of race, black feminist writers also identify the construction of racial stereotypes (such as, “more oppressed/in need of feminist help” or “strong, black women/who don’t need feminism”) within some theories.

In relation to questions of sexuality, too, feminist theories have been criticized for their heteronormativity. Lorde’s work, for example, has asked not only why race but also why sexuality, and lesbianism in particular, has been missing from some feminist accounts (Lorde, 1996). Rubin, too, argues that feminism is not necessarily the preferred theory of sexual oppression and that, in some cases, feminists have proposed “a very conservative sexual morality” (Rubin, 1984: 302). Of course, this is a complicated picture, since Rubin’s objections are, in some cases, toward forms of lesbian feminism that she found to be restrictive or hierarchical, but she is also making a case, not against feminism, but against theories that see sexuality merely as a derivative of gender.

Material and structural accounts of gender

Materialist or structuralist accounts focus on institutions, such as the family or the workplace, in order to examine how gender inequality is produced and reproduced within such settings. Connell’s work, for example, describes gender as “the structure of social relations that centres on the reproductive arena, and the set of practices that bring reproductive distinctions between bodies into social processes” (Connell, 2009: 11). This is because she views gender as a pattern within wider social relations, and so is critical of any gender theory that does not consider issues such as education, domestic violence, or health, all of which are “gendered.” For Connell, then, societies exhibit a “gender order” (Connell, 2009: 73).

Another example of structural theory is Risman’s work on family relations (Risman, 1998). Risman argues that institutions, such as workplaces or the family, produce inequality between women and men. She makes a case for a focus on material constraints, which she sees as lacking from other theories. For Risman, gender is a structure that has consequences for people at individual, interactional, and institutional levels. Her study of single fathers is particularly interesting in this respect, as they were engaged in homemaking and caring for children. Indeed, Risman refers to single fathers’ work as “mothering” (Risman, 1998: 52), since she found that responsibility for home and care is better explained by parental role rather than gender. Risman also says that single fathers “described themselves as more feminine than did other men” (Risman, 1998: 65). Thus, for Risman, a family structure in which there is one, male parent determines “gender,” in the sense that this results in a particular sense of self (“more feminine”) and in work usually associated with women. In heterosexual couple families, women were far more likely to do this caring work.

It is possible to raise some questions about this perspective, not least in terms of methodology, because Risman largely tests for gender as a measurable variable (e.g. see “Measurement of Parenting Variables” or “Gendered selves” (Risman, 1998: 59 and 76)). This does not allow much space for the negotiation of gender within an interactional context or the role of language in that process. Indeed, Risman is rather dismissive of in-depth interviews, due to the distortions and failures of memory that she sees in such methods. However, it is also important to acknowledge that Risman’s view of gender as a structure does not see this as determinative, since, in some cases, those structures and their consequences may be challenged. However, Risman’s point is that institutional forms constrain ways of behaving; or, they have certain gendered consequences, such as inequalities between women and men.

This approach to gender is often taken up in work on stratification of social work organizations. Here, it is argued that the gendered structure of social work, with a disproportionate number of men in senior and management positions, results in gendered inequality for women in terms of treatment and career prospects (Dominelli, 2002b; Harlow, 2004; Kirwan, 1994). Yet it would also be possible to argue that such explanations tell us little about how gender works in these settings. Are “men” and “women” treated differently, regardless of race, sexuality, disability, class, or other issues? If the explanation for inequality is merely “gender difference,” then how exactly do gendered ideas about persons arise within social work in the first place? How are dominant or oppressive ideas about gender resisted within social work teams or settings? Is gender the primary factor or point of identification for social workers? These kinds of questions, which structural explanations often avoid, bring us on to the question of how gender is produced through practices.

The practice of gender

For ethnomethodologists, a problem with structural accounts is that these assume an institutional form results in gendered consequences, but this does not ask how gender is achieved. What practices, for example, produce a gendered institution or society, and how are these, in fact, constitutive of something called “gender?” Instead, ethnomethodological accounts are concerned with how gender is achieved in everyday life; that is, with how all people ordinarily achieve a gender status.

Garfinkel’s study of Agnes, a person who presented as intersex but later revealed herself to be a transsexual woman, was undertaken not to demonstrate the special features of intersex persons or transsexualism, but rather to show that, for all people, “sex status” is an ordinary social achievement. Garfinkel argued that social life is “rigorously dichotomized into the ‘natural,’ i.e., moral, entities of male and female” (Garfinkel, 1984: 116), and so, in order to be taken for a “normal” person, one has to be taken for a man or woman. But this process involves various cues, to do with appearance, speech, biography, and so on, that each person (or “member”) gives. So, for Garfinkel, “members’ practices alone produce the observable-tellable normal sexuality of persons” (Garfinkel, 1984: 181).

This work was developed further in Kessler and McKenna’s study, which argued that the attribution of gender is a primary feature of everyday life, and that what they term “gender role” refers to a set of prescriptive characteristics or expectations (Kessler and McKenna, 1985: 11). Kessler and McKenna argue that this process of gendering persons into just one of the two categories (female or male) is fundamental to social life, and yet unremarkable. This allows, for example, for the presentation of gender as a social “fact,” in which some theorists or researchers account for certain behaviors as caused by gender (“gender causes x”).

These arguments influenced the “doing gender” perspective of West and Zimmerman, which states that gender “is the activity of managing situated conduct in light of normative conceptions of attitudes and activities appropriate for one’s sex category” (West and Zimmerman, 1987: 127). Crucially, this emphasizes the concept of accountability, because:

a person engaged in virtually any activity may be held accountable for performance of that activity as a woman or a man … to ‘do’ gender is not always to live up to normative conceptions of femininity or masculinity; it is to engage in behavior at the risk of gender assessment. (West and Zimmerman, 1987: 136)

In later work on “doing difference,” West and Fenstermaker have shown that similar processes apply to race and class (West and Fenstermaker, 2002). West and Zimmerman have also been critical of structural perspectives, which assume that gender may be undone in order to undo inequality. They argue that gender is not so easily abandoned, since all of everyday life is accountable in gendered terms (West and Zimmerman, 2009).

Risman has suggested that the doing gender perspective is in danger simply of labelling any activity as masculinity or femininity and, along with others, argues that this may give the impression that nothing can change (Deutsch, 2007; Risman, 2009). In the sense, identified by Butler, of gender being treated as a given explanation for phenomena, Risman’s point is important, but this would be a misreading of ethnomethodological claims. Ethnomethodologists explore what ordinary people count as examples of “masculinity” or “femininity,” and are interested in transformational possibilities. After all, they see gender as a moral, not merely practical, order. Thus, Deutsch’s proposal to “reserve the phrase ‘doing gender’ to refer to social interactions that reproduce gender difference and [to] use the phrase ‘undoing gender’ to refer to social interactions that reduce gender difference” (Deutsch, 2007: 122) seems simplistic: how do we know when gender is being either reproduced or reduced? And isn’t it possible that both are occurring within any interaction that appears to involve gender?

Within social work, ethnomethodological perspectives on gender are rare, but there is research that considers gender as practice. Pösö’s work, in which probation officers attempted to identify whether speakers in transcripts were female or male, demonstrates contradictory views of, and methods for identifying, gender. Generally, talk about emotions, relationships, or children were associated with women, and objectivity and reticence in speech associated with men. Pösö argues that gender is “situational and … case-specific” (Pösö, 2003: 175), and that more attention should be given to the ways in which it is practised.

Scourfield’s ethnographic study of a childcare social work team examines constructions of gender, and suggests “an underlying dichotomy of men as abusers, and women as carers” (Scourfield, 2003: 60). Women were primarily seen as responsible for children’s welfare and they were expected to protect children from abusive men, with the “failure to protect” discourse a feature. Men were often described as dangerous, threatening, or absent/irrelevant, something that Scourfield sees as part of the continued overlooking of men, and blaming of women, within child protection. Thus, while there are “multiple gendered discourses in the culture of the social work office that constitute the knowledge available to social workers,” these are, at the same time, both powerfully limiting and open to challenge (Scourfield, 2003: 151).

Butler and performativity

Butler’s work on gender echoes aspects of ethnomethodology and doing gender, since it is concerned with gender as “a set of repeated acts within a highly rigid regulatory frame that congeal over time to produce the appearance of substance” (Butler, 1990: 33). However, Butler’s work also demonstrates the influence of poststructural theories and a concern with the heteronormative aspects of gendered practices, noting that the:

heterosexualization of desire requires and institutes the production of discrete and asymmetrical oppositions between ‘feminine’ and ‘masculine,’ where these are understood as expressive attributes of ‘male’ and ‘female.’ The cultural matrix through which gender identity has become intelligible requires that certain kinds of ‘identities’ cannot ‘exist’—that is, those in which gender does not follow from sex and those in which the practices of desire do not ‘follow’ from either sex or gender. (Butler, 1990: 17)

Of course, this does not mean that other kinds of “gender” do not exist, and Butler uses the example of drag to show how gender is practised, but also, that it is always imitative. By this she means that drag is no mere copy of an original gender, but rather that in “imitating gender, drag implicitly reveals the imitative structure of gender itself—as well as its contingency” (Butler, 1990: 137).

In Bodies That Matter, Butler clarifies this performative sense of gender, arguing that this is not about gender as an individual choice or mere play, since “performativity must be understood not as a singular or deliberate ‘act,’ but, rather, as the reiterative and citational practice by which discourse produces the effects that it names” (Butler, 2011a: xii). This is an important point because, while Butler’s presentation of drag in Gender Trouble tends to suggest a challenge to traditional versions of gender, work by others, such as Bridges, argues that some forms of “drag” may be used as a temporary joke, actually to reinforce “normal” gender (Bridges, 2010). Indeed, Butler herself later noted that drag is not necessarily subversive (Butler, 2011a).

Butler’s argument is that gender precedes the individual; that is, that subjectivity must be taken up through gender, so one comes to be a person through being taken for a woman or man. When an individual does not appear to be gendered in a “normal” way, then it is that individual, rather than the gender order, that is questioned. In relation to social work, Green and Featherstone have analyzed Butler’s potential, and have suggested that her work helps to challenge dogmatic and morally certain positions within anti-oppressive theory, which they describe as a “project that believes in its own innocence and construct[s] social workers as disembodied carriers of a ‘pure’ project” (Green and Featherstone, 2014: 32).

Gender as discourse

The emphasis in Butler’s work on the question of discourse is taken up in a range of theories, influenced in part by the poststructuralist turn to language, which consider gender as discourse. These theories see gender as produced via social and textual practices, which regulate the ways in which we may think about men, women, and others. One important implication of this is that gender is not fixed, nor is it simply attached to individuals. Instead, people contest gendered meanings and subject positions, although, in order to be taken seriously, they may well have to use familiar and expected ways of expressing themselves. Further, as Kessler and McKenna argued, and Butler acknowledged in her later work, the reception of a gendered claim, by audience or perceiver, matters.

Smith’s discussion of femininity as discourse suggests that the very concept “femininity” is produced through practices and their embeddedness in texts. So, gender is not merely a structure or ideology imposed upon un/willing subjects, but rather it is a “complex of actual relations vested in texts” (Smith, 1990: 163). This is an interesting point, as we hear here Smith’s joint adherence to both a materialist and discursive account of gender, which she sees as mutually dependent, since gender is produced within both local and wider social relations. That is, a discourse of gender relates to people’s actions within localized settings and the organization of their ways of thinking and talking.

Like Garfinkel, Smith insists that gender is a moral order, which means that it is coordinated with wider social and economic relations, so that femininity is “a textual discourse vested in women’s magazines and television, advertisements,” and so on (Smith, 1990: 163). The moral order attempts to position women and femininity only in relation to the, more valued, men and masculinity, and for women this implies the need to be considered “attractive” or “desirable,” “a condition of participation in circles organized heterosexually” (Smith, 1990: 194). Smith refers to play and interplay within gendered discourse, in order to argue that it does not prescribe action, and yet she also reminds us that social texts establish recognizable concepts and categories, so that what is done may (or may not) be recognized as an instance of what is authorized. Thus, to take up gender within discourse is to be recognized as demonstrating a proper instance of such, that is, a “proper” man or woman.

Returning to social work and gender

In my research, I have argued for an analysis of gender as a practical achievement within everyday social work contexts. Drawing upon the ethnomethodological and discursive theories discussed earlier, I have suggested that gender is neither a characteristic merely acquired and passed on through socialization or reproduction of structural forms nor something inherent in the person. Rather, social work processes involve the production of gender through practical means, which relate both to immediate, local, and wider, institutional contexts. An example of this would be my analysis of the ways in which notions about “gender role” are used within the assessment of lesbian or gay foster care or adoption applicants (Hicks, 2011, 2013). Here, I have demonstrated how social workers and applicants draw upon and produce ideas about gender in order to categorize “identities” or “lifestyles,” and I have noted that, in most cases, the issue of “gender role models” has to be:

addressed in relation to gay and lesbian applicants, and those applicants, as well as some social workers, who, in other contexts, are opposed to notions of gender role, must conform since they are held accountable. And while there is resistance to gender norms here, a standard and institutional discourse dominates, one in which adherence to a moral order that upholds expected gender roles is required. (Hicks, 2013: 158)

This is confirmed in other research (Wood, 2013), and reminds us of the ethnomethodological point that, where any person is perceived to question standard gender in some way, then it is usually that individual or group category, rather than the gender hierarchy, that is held to account, since gender functions as a moral order. This approach to the theorization of gender within social work emphasizes its reliance on other categories, such as race or sexuality, and its active production via interactions involving powerful linguistic claims, moving us away from essentialist, functionalist and, to some extent, structuralist accounts.

In using this article to review various theorizations of gender, my point has been to highlight ways in which social work may be limited in the versions that it prioritizes. The tendency to treat gender in isolation, critiqued in some accounts (Brown, 1992; Shah, 1989), or to take up a solely structural view indicates a reification of gender and an ignorance of its production through practice. My argument has been that, bar a few examples (Pösö, 2003; Scourfield, 2003), social work rarely connects with gender as practice, ironic for a discipline so concerned with practical dynamics. This, then, is also an argument for attention to the ways in which gender is produced through social work, something that draws upon both the practical and the discursive, rather than starting with something termed “gender” and then looking for its effects. This may prove controversial in a field somewhat dominated by anti-discriminatory approaches; that is, where gender is considered at all; yet it is my argument that taking up Butler’s “inquiry on the production of difference” (Butler, 2011b: 21) may open up possibilities for less restrictive accounts of gender within social work’s various fields.