Rethinking the possibilities for hegemonic femininity: Exploring a Gramscian framework

In this paper I consider and challenge the ways in which hegemonic femininity has mainly been conceptualised in the gender literature. This approach has several limitations, including being strongly binary, positioning girls and women as Other and frequently essentialised. After suggesting some criteria for a more useful conceptualisation, I consider some of the alternatives, which I critique for their dependence on sexuality and sexual desire. I propose an alternative definition of hegemonic gender performances, avoiding binary distinctions, building on Francis et al.'s (2016) suggestion that a more directly Gramscian conceptualisation may be useful. Having outlined this alternative, I examine how it is played out in the specific context of one English primary


Introduction
This paper has two main parts, and two related functions. First, I consider the ways in which hegemonic femininity has been conceptualised in the literature around gender. I discuss some of the limitations of its conception, including its strongly binary nature, and suggest criteria for an alternative. After examining one previously suggested approach, I go on to propose an alternative, more directly Gramscian (Francis, Archer, Moote, Witt, & Yeomans, 2016), way of conceptualising hegemonic gender forms that deals with some of the problems and goes some way to avoiding an inbuilt gender binary. I then move on to consider my own data about a dominant female group in one primary school class, to examine how such a conceptualisation of hegemonic gender performances can illuminate our examination of hegemonic femininities and masculinities in a particular context. Femininity is underconceptualised in the current literature on gender (Budgeon, 2014;Connell & Messerschmidt, 2005;Schippers, 2007), including that which is related to my own field of children and education (Francis et al., 2016;Paechter, 2012). While there has been considerable investigation into how we understand different masculinities and the relationships between them, corresponding attention has not been given to theorisations of femininities (Connell & Messerschmidt, 2005;Dahl, 2012). It is not altogether clear why this should be the case. However, it is noteworthy that most research into how we might understand gender, and in particular into dominance in gender relations, has not included specific and detailed consideration of femininity as a concept. Connell's (Carrigan, Connell, & Lee, 1985;Connell, 1987Connell, , 1995 foundational work on socially dominant gender forms focused more or less entirely on masculinities, with a strong emphasis on hegemonic forms, with femininities only briefly mentioned and even then only in relation to masculinities. Other major theorists, such as Butler (1990Butler ( , 1993Butler ( , 2004 have concentrated on gender more generally, and while Halberstam (1998) certainly writes about women, the discussion is entirely about those with female bodies who identify with and perform masculinity. While there is a considerable body of work that investigates specific forms of femininity, especially among children (Adams & Bettis, 2003;Allan, 2009;Blaise, 2005;Epstein, Kehily, Mac an Ghaill, & Redman, 2001;Evaldson, 2003;Jackson, 2004;Kehily, Mac an Ghaill, Epstein, & Redman, 2002;Messner, 2000;Paechter, 2007;Reay, 2001;Renold, 2001;Skelton & Francis, 2003) most of this research takes the conceptualisation of femininity itself more or less for granted, or refers to an unexplored notion of 'hegemonic femininity', treated as locally dominant and rooted in Connell's (1987Connell's ( , 1995Connell's ( , 2002 theorising of hegemonic masculinity, but without taking into account either its limitations or Connell's assertion that there is no possibility of hegemonic femininity. The overall effect of this is that researchers reflect wider social relations and treat females and femininity as the Other of males and masculinity. Femininity thereby becomes some sort of counterpart to masculinity, defined entirely in opposition and subordinate to it, such that femininity cannot be conceptualised at all without a masculinity (Dahl, 2012). This is demonstrated particularly strongly in Connell's definition of hegemonic masculinity and her subsequent dismissal of the possibility of a parallel hegemonic femininity, an approach that I have myself followed in previous work. The widespread takeup of Connell's ideas and their application in a wide range of studies has solidified this position. It is almost as if the underlying sexism within society at large has crept into and pervaded feminist theorisations so that we have barely noticed the lack of focus on women and girls. By concentrating on dominant men and their relationship with other men, we have ignored, or taken for granted, the positions of women.
Connell defines hegemonic masculinity as: The configuration of gender practice which embodies the currently accepted answer to the problem of the legitimation of patriarchy, which guarantees (or is taken to guarantee) the dominant position of men and the subordination of women (Connell, 1995: 77) This 'configuration of gender practice' is, however, only a configuration of male gender practice. Connell is clear that it does not apply to women, and, indeed, states that, because all forms of femininity are constructed in the context of male domination, 'there is no femininity that is hegemonic in the sense that the dominant form of masculinity is hegemonic among men' (Connell, 1987: 183). Instead, Connell suggests, we have 'emphasised femininity', which is constructed as a counterpart, or subordinated Other, to hegemonic masculinity, 'performed especially to men' (Connell, 1987: 188) and focused around an internalised subordination and subjugation in relation to dominant masculinities. Furthermore, and in contrast to Connell's nuanced and context-bound understanding of hegemonic masculinity, emphasised femininity is conceived in strongly essentialist and stereotypical terms, leaving no possibility of the local variation that she consistently insists on for hegemonic masculinity (Connell & Messerschmidt, 2005). This does not seem to me to be a good place to be theoretically, especially as some researchers have since used the term 'hegemonic femininity' as if it were both coherent and generally understood. We therefore need to consider possible alternatives.
Before we do so, however, we need to examine what a conception of hegemonic femininity would be like and the work it would have to do. First, it would have to have a clear relationship to hegemonic masculinity which would not simply leave it as Other. Second, such a concept would need to take into account the ways in which hegemonic gender forms are supportive of the status quo, that they perpetuate an unequal gender regime. It would also account for positions of female dominance, at least with respect to other women and girls, but would not be tied to female bodies. Similarly, this account of hegemonic femininity would be related to a revised theorisation of hegemonic masculinity, which, while giving us a way of understanding the dominance of some males over others, and over females, would not assume that masculinity is always attached to male bodies. Finally, this new conception should not be totalising: it would include an understanding that not everyone is in either a hegemonic or some sort of subordinate or otherwise degraded position, but that some people's takeup of gender can be simply non-hegemonic, rather than being caught up in a hegemonic/subordinated relation.
One possible alternative: heterosexuality-based co-construction Schippers (2007) suggests that we can replace the hegemonic masculinity/emphasised femininity pair with a conception of co-construction of hegemonic forms. Following Butler (1990), she proposes that 'heterosexual desire, as a defining feature for both women and men, is what binds the masculine and feminine in a binary, hierarchical relationship' (90). In the context of Western societies, she argues, 'the cultural construction of embodied sexual relations, along with other features of masculinity and femininity, defines a naturalized masculine sexuality as physically dominant in relation to femininity' (90). This naturalisation of male sexual dominance allows us, Schippers suggests, to reconceive of hegemonic masculinity, along with a relational concept of hegemonic femininity. She rewords and elaborates Connell's (1995) definition: Hegemonic masculinity is the qualities defined as manly that establish and legitimate a hierarchical and complementary relationship to femininity and that, by doing so, guarantee the dominant position of men and the subordination of women.
Hegemonic femininity consists of the characteristics defined as womanly that establish and legitimate a hierarchical and complementary relationship to hegemonic masculinity and that, by doing so, guarantee the dominant position of men and the subordination of women. (Schippers, 2007: 94, italics in original) Schippers thus uses a reworked version of Connell's original definition of hegemonic masculinity to claim a space for a complementary hegemonic femininity and to define the latter in terms that reflect this.
Such a redefinition is a laudable enterprise and would allow a more thorough theoretical grounding for concept of hegemonic femininity. It puts masculinity and femininity on a more equal basis and maintains the hierarchical/domination aspects of hegemony. There are, however, several difficulties with this approach. First, there are some clear problems with the foundation of gender in sexual desire. I have explored these elsewhere in relation to children (Paechter, 2017), but should note in addition that the assumption that all humans have sexual desire is also empirically problematic given the existence of people who are asexual. If 'the possession of erotic desire for the feminine object is constructed as masculine and being the object of masculine desire is feminine' (Schippers, 2007: 90), does that mean that asexual people, who do not desire, can only be feminine? Furthermore, Schippers slips in her discussion between seeing hegemonic masculinity and femininity as, on the one hand, monolithic and unitary, and, on the other, locally defined. This slippage is not unusual, and is hard to avoid (Paechter, 2012), but it is problematic.
A further difficulty with Schippers' conceptualisation is that she falls into the same trap as does Connell in treating non-hegemonic forms of masculinity and femininity as always and necessarily problematic positions, rather than just different (Francis & Paechter, 2015;Paechter, 2012Paechter, , 2017. Shippers takes an unusual position here, as, having set up hegemonic masculinities and femininities as complementary, she characterises non-hegemonic forms differently for men and women. These different forms also essentialise masculinity and femininity as tied to male and female bodies (Francis et al., 2016), something which we explicitly want to avoid. Most importantly, Schippers argues that, because masculinity is always superior to femininity, there can be no possibility of subordinate masculinity. What were characterised by Connell as subordinate masculinities, she argues, are 'simply hegemonic femininity embodied or enacted by men' (Schippers, 2007: 96). This is because, for Schippers, gender hegemony is legitimated by preserving the hierarchy between masculinity and femininity. Consequently, no masculine characteristics can be regarded as subordinate, because masculinity itself can never be subordinate. Thus, when people with male bodies exhibit characteristics that are not part of hegemonic masculinity, such as desire for other men, weakness or compliance, they have, instead, to be treated as feminine, and thus automatically inferior and stigmatised (Budgeon, 2014).
When those with female bodies perform or embody the characteristics of hegemonic masculinity, however, such as aggression, promiscuity, sexual inaccessibility, or having sexual desire for other women, this is considered to be not masculinity, because, presumably, female bodies cannot be masculine. Instead it is treated as pariah femininity, because such behaviour is seen as 'contaminating to the relationship between masculinity and femininity' (95) by refusing to conform to the complementary relation of male dominance and female subordination. This idea is extended and developed by Budgeon (2014), who follows Schippers (2007) in her conceptualisation of hegemonic femininity and masculinity. She notes, for example, that 'new femininities' (McRobbie, 2009;Renold & Ringrose, 2012), which valorise assertiveness, individuality and achievement, are not regarded as masculine because they maintain powerful markers of conventional femininity. Budgeon also suggests that, in some contexts, 'pariah femininities' are those perceived as expressing traditional femininity in excessively accentuated ways (327). She notes that young women in some studies exhibit active disdain for those they perceive as insufficiently assertive and too overtly dependent on male approval. At the same time, assertiveness can only be taken so far: those who embrace feminism and thereby 'threaten heterosexual norms of attraction and the loss of approval by men and therefore undermine hierarchical gender complementarity' (327) can tip themselves over into pariah status.
While it is welcome to see femininity as something that can be performed by people with male bodies, this way of conceiving hegemonic masculinities and femininities is problematic because it seems to construct hegemonic femininity, at least, as something that is constant across contexts. It also leaves no space for straightforward difference. Within specific social groups, however, there are femininities that are constructed oppositionally to local dominant (and therefore, in Schippers' conception, hegemonic) forms that are neither pariah nor stereotypically masculine. Again, we need to leave space for non-subordinate difference.
Overall, therefore, Schippers' (2007) attempt to reconfigure Connell's (1995) concept of hegemonic masculinity to include a complementary hegemonic femininity, does not fulfill the criteria I set out earlier, and, indeed, perpetuates some of the more problematic features of Connell's conceptualisation, such as her implied essentialism and lack of attention to straightforward difference as opposed to subordination.
A more fully Gramscian approach to hegemonic femininity An alternative possibility comes from Francis et al. (2016), who suggest in passing that we could have a more faithfully Gramscian approach to the question of hegemonic gender performances. They argue that we could see some productions of both femininity and masculinity as hegemonic in that they maintain social norms rather than because they bear power in their own right. Such an approach would require us to take a step back and consider gender itself as a set of hegemonic forms and ideologies which influence how people identify and behave. Gramsci's (1971) concept of hegemony refers to the ways in which dominant classes are enabled to maintain their ascendancy by convincing oppressed members of society that the established order is in the latter's interests, backed up by the ever-present possibility of coercive force: The 'normal' exercise of hegemony… is characterised by the combination of force and consent, which balance each other reciprocally, without force predominating excessively over consent. Indeed, the attempt is always made to ensure that force will appear to be based on the consent of the majority (Gramsci, 1971: 80) Dominance is therefore perpetuated mainly by consent, without the oppressed being even fully aware that it is happening, because it is part of the taken for granted of everyday life. Wetherell and Edley (1999) argue that Hegemonic ideologies preserve, legitimate and naturalize the interests of the powerfulmarginalizing and subordinating the claims of other groups. Hegemony is not automatic, however, but involves contest and constant struggle. (336) In the context of gender, this 'constant struggle' relates particularly to masculinity, which has to be repeatedly established and validated by other males (Connell, 1987(Connell, , 1989Martino & Pallotta-Chiarolli, 2003;Paechter, 2007). This struggle is also, of course, part of the constant interplay of power and resistance throughout society (Foucault, 1977(Foucault, , 1978(Foucault, , 1988. Because these claims and the associated struggles take place through ideological contestation, they involve compromises, consensus and incorporation. Conceptions of masculinity and femininity change over time and place as interactions and resistances require constant local negotiation and rejustification. Nevertheless, the resulting, however shifting, consensus is one in which those personifying or performing hegemonic forms are able to mobilise power by consent: they are recognised in some ways as having the right to their dominance. As Kenway (2002) argues, The active consent of the dominated groups is mobilized and reproduced because the dominant class is recognized as representing the interests of numerous social groups…. It has become hegemonic because it has articulated to its discourse the overwhelming majority of ideological elements characteristic of the social formation (56) This means that powerful social formations are maintained not by force, but by a consent underpinned by ideologies that present the status quo as the natural order of things (Kenway, 2002). In this sense, hegemonic orders function as regimes of truth (Daldal, 2014;Foucault, 1977): ways of thinking about the world and behaving within it that are hard to escape because of their normalising function. Hegemonic gender relations are thus deeply embedded in people's understanding of themselves, of who they are, and who they ought, or aspire, to be. In the specific context of male dominance, Arnot (1982) explains that a Gramscian concept of hegemony refers to a whole range of structures and activities as well as values, attitudes, beliefs and morality that in various ways support the established order and the… interests which dominate it…. Women in this analysis must offer unconsciously or consciously their "consent" to their subordination before male power is secured. They are encouraged "freely" to choose their inferior status and to accept their exploitation as natural. (66) The normalisation of hegemonic gender forms makes it appear to those oppressed by them that such forms are not just how things are but how they ought to be. Those oppressed by a particular local gender ideology are nevertheless caught up and invested in it.
A strictly Gramscian understanding of hegemony would place the responsibility for constructing such regimes of truth in the hands of the ruling classes: the role of revolutionary leaders is to undermine these and produce alternative hegemonic forms (Gramsci, 1971;Pringle, 2005). In the context of gender, this would usually mean men, or maybe those men who are white, cis-gender, able-bodied and from the global North. However, I think that it is possible to provide a more Foucaultian interpretation in which hegemonic forces work in less hierarchical and deliberate ways, through multiple and networked power relations (Foucault, 1977(Foucault, , 1978(Foucault, , 1980(Foucault, , 1982. A hegemonic gender order would, therefore, be a way in which gender norms are maintained through a complex of power relations and normalising forces which influence people's 'thinking as usual' (Schutz, 1964) coupled with a continued investment in gendered social arrangements. This investment, might, for example, be due to things such as the pleasure adults and children get from gendered behaviour (Blaise, 2005), or the sense of stability that comes from understanding oneself as having a coherent gender identity. While there are some potential difficulties in combining the ideas of Gramsci and Foucault (Pringle, 2005), I think there are sufficient points of contact, including the focus of both writers, in different ways, on resistance and the possibility of alternatives (Kenway, 1990;Smart, 1986), to be able to do this in the context of gender.
What, then, would it mean to think about hegemonic masculinities and femininities in a more directly Gramscian way, as acting together to support social norms? I think that one of the most crucial things is that the two are conceived together, rather than starting with hegemonic (and non-hegemonic) masculinity and then trying to fit femininity around it. In this way we would see masculinity and femininity as complementary in so far as they are different aspects of a gender order that works to maintain a particular (currently male-dominated) status quo. This would not, however, mean that they were necessarily polar opposites: I will argue below, using my own research, that in some circumstances local hegemonic masculinities and femininities can have a considerable amount in common while still supporting the gender order. I think our definition needs also to attempt to avoid essentialism, so that being a man, for example, or behaving in masculine ways, is not tied to a particular bodily form. We need to be able, among other things, to include the possibility of people identifying as men, with bodies recognised as male, acting in significantly, and even, hegemonically, feminine ways, and also to ensure that we do not exclude from hegemonically masculine or feminine performances those people who do not identify as male or female at all. Consequently, I think it is necessary to move away from building into the definition an explicit mention of male domination over women: one aspect of the social norms supported by hegemonic masculinities and femininities is surely the assumption of a gender binary and thus the exclusion and negation of those who do not fit within it. We might thus have an initial definition that runs something like this: Hegemonic gender performances are those which act, within a particular context, to uphold a gender binary and maintain traditional social relations between genders Such a definition will need testing and fleshing out in future work, and I hope others will rise to this challenge. Meanwhile, however, I would like to put down some initial thoughts about how such hegemonic performances are manifested in practice. My intention has been to reflect current, more fully developed, understandings of hegemonic masculinities, so I will focus here on what hegemonic femininities might look like under this formulation. I will start with some general speculations about how we might understand hegemonic femininities in adults, before going on to look at applications of this concept to my own research on girls.
Like hegemonic masculinities, hegemonic femininities have to be thought of as local phenomena. Which forms support the traditional gender order is dependent on local circumstances, including (and reciprocally) local hegemonic masculinities. This means that they must be viewed intersectionally: different ethnicities, social class positions, sexual orientations and so on will be more or less able to make claims to hegemonic positions according to specific local circumstances. One thing that it is important to stress, however, is that hegemonic femininities would not have to conform to the weak and fluttery stereotype of Connell's emphasised femininity, although there will still be some contexts where, particularly when enacted in support of powerful men, such femininities are hegemonic (and, indeed, successful, in the sense of reaping the rewards of the patriarchal dividend (Connell, 1987(Connell, , 1995). It seems to me that there are feminine positions which, while in many ways strong and powerful, continue to support traditional gender relations. One example of this is the figure of former British Prime Minister Margaret Thatcher, who projected simultaneously the image of the strong leader (the Iron Lady) and that of a traditional wife and mother (electoral broadcasts featuring her with a shopping basket or doing the washing up), and who surrounded herself with a cabinet of hegemonically masculine men. Similarly, women who are professionally successful but see themselves, behave, and are treated, as exceptions, and who do not support other women following behind them, also perpetuate the traditional gender order, however much they overtly 'break the mould'. Such women are portrayed, and perform, as successful in their own right (an important contemporary aspiration), while maintaining an aura of conventional femininity. It is also likely that McRobbie's (2009) 'global girls', as ambitious, assertive young professional women, would represent hegemonic femininity in specific local contexts. In many ways, such hegemonic femininities celebrate the results of feminism, particularly through media portrayals of 'girl power', while simultaneously neutralising its more powerful effects, thus allowing women a wider range of roles, lifestyles, and civic participation, without significantly challenging the traditional gender order.
What these forms of femininity have in common, and in common with the operation of hegemonic masculinity in relation to men, is that they both perpetuate the traditional gender order and present themselves as the femininity that all women should aspire to. This does not mean that all women actually do want to be like these, any more than all men want to be like their local hegemonically masculine models (and, indeed, Connell (1995) argues that few men are, in practice, hegemonically masculine). What we have, instead, is cultural models of aspiration against which a particular social group expects women to measure themselves, even if in practice they do not, and which, however much they are presented as feminist, maintain traditional forms of male dominance. In order to look in more detail at how such hegemonic femininities operate, I am now going to focus on my own research on girls in one primary school class during their last two years there, to examine the position of the dominant 'cool girls' in that setting as a local hegemonic femininity and consider how it is related to local hegemonic masculinities.

Gramscian hegemonic femininities in the primary classroom
Conceptualising hegemonic gendered forms in a more strictly Gramscian way allows for a wider, more intersectional and more fluid understanding of hegemonic masculinities and femininities in the local context. It also permits a more nuanced understanding of how different forms of femininity (not just those that are hegemonic) contribute to hegemonic gender relations. In this section I am going to look in detail at the operation of a group of 'cool girls' and their relation to the local dominant boys, in one primary school class. Research on these children took place as part of a study of tomboy identities 1 among children on the cusp of puberty, carried out through case studies in two primary school classes, one in inner London and the other in a leafy outer suburb. A combination of ethnographic observation and in-depth interviews were used in order to capture the lived experience of tomboy girls in primary school and to capture the nuances of their peer relationships (Paechter, 2010;Paechter & Clark, 2007a, 2007b, 2010. We deliberately chose two strongly contrasting schools in order to examine how tomboy identities were taken up and constructed in different settings. In addition to their very different locations, the schools were distinct in other ways. While the inner city school was small, consciously multi-ethnic in composition and with a strong ethos of community and co-operation, the suburban school was much larger, highly competitive, both internally and with other schools, and operated with an overarching assumption of whiteness which rendered other ethnicities invisible, at least in the official discourses and performances of the school. Both boys and girls were interviewed in small friendship groups about relationships within the class and their understanding of what it meant to be a tomboy, and girls identified as tomboys by themselves or by their peers were also interviewed alone twice during the period of the study. Class and sports teachers, and some parents, were also interviewed. The children were observed in all aspects of school life for one or two days of each week over two terms spanning their final two years in primary school, and field notes taken of their interactions. Additional observations took place at after-school clubs and related sporting events. All interviews were transcribed and the data were thematically analysed in parallel with the fieldwork, so that preliminary results could be used to inform data collection and support further development of the themes of the inquiry. As a result of this, we developed an additional focus: the inter-group relations between girls at the suburban school, Holly Bank. 2 1 Tomboy Identities Study, supported by the Economic and Social Research Council under Grant RES-00-22-1032. I was the grantholder and Sheryl Clark the researcher on this project. 2 All names as pseudonyms. In the case of the children, these were self-given, so some of them are not conventional names and they do not always reflect either gender or Paechter Women's Studies International Forum 68 (2018) 121-128 In this class there were two clear hegemonic groups of 'cool' boys and girls, coalescing in the case of the boys around Humphrey, the most powerful child in the class, and in the case of the girls around Kelly. Each led a small, tight-knit and exclusive group of same-gendered children who collectively personified the hegemonic masculine and feminine norms in this context. Both also worked hard to maintain the dominance of their group, and of themselves at the core of it, and both were skilled at mobilising power in the social world of the classroom, dining hall and playground. Both groups operated through a combination of consent and coercion. They exercised 'intellectual and moral leadership' (Gramsci, 1971: 57) by defining themselves as 'cool' and exclusive and, in particular, by controlling dominant local discourses around what it meant to be cool. In this way they were able to manipulate local taken-for-granted understandings of masculinity and femininity, allowing their hegemonic positions to be accepted almost without question. At the same time, they exerted coercive dominance by overtly or covertly bullying other children in the class (and, on occasion, each other, particularly among the 'cool' girls). As will be discussed below, these local hegemonic masculinities and femininities had a considerable amount in common, unlike Connell's (1987) characterisations of hegemonic masculinity and emphasised femininity as polar opposites. They were not, however, equal: Humphrey, in particular was assiduous in maintaining the overall superiority of himself and his friends, and Kelly and her group made little attempt to challenge this, despite their powerful positioning in relation to other girls and to the rest of the boys in the class. Traditional binary gendered power relations were therefore maintained, despite the similarities between the two groups.
Although many children in the class (and Humphrey, in particular) were strongly invested in gender difference, hegemonic masculinities and femininities in this context were in many ways similar, suggesting that only small differences between groups, particularly in ability to mobilise power, are necessary to maintain the traditional gender order. Both groups projected a consistent image of good looks and heterosexual desirability, achieved by the boys through physical fitness and muscularity, and by the girls through a combination of thinness and physical artifice such as makeup, clothes and hair. This was enhanced by the relative affluence of their families, which allowed the boys to participate in out of school sports and the girls to wear expensive and fashionable clothing when not at school, and have beauty treatments with their mothers at weekends. Both Kelly and Humphrey were also white, though whiteness was not in itself a requirement for belonging to the 'cool' groups, which both included children from other ethnic backgrounds at one time or another. Humphrey, who was extremely sporty, playing for three football teams, including the juniors in a national side, was taller than most of the other boys, blond and muscular. Kelly and two of her friends, Bridget and Pippa, had highlights or similar alterations to their hair at points during the fieldwork period, and Chelsea's hair was long and naturally blonde. All the hegemonic children were acutely aware of local codes requiring the subtle alteration of school uniform (such as the pulling up of girls' skirts to be short but not too short, or the pushing down of boys' trousers toward the hips), and scorned other children who were less aware of these norms. Both groups also projected a sense of the superiority of their own 'perfect' bodies, either by looking pointedly and mockingly at other children's 'less than perfect' ones, or by insisting on careful grooming for members of their own group. For example, when changing for physical education, Humphrey drew attention to the comparative lack of fitness of a low-status boy, Foxbat: Foxbat's soft, unmuscled thighs, too short shorts and big black running shoes provoke ridicule from Humphrey and Donald, who whisper, 'look at Foxbat' and laugh. (field notes) Kelly's group, similarly, expected near-perfect bodies from members, with any imperfections corrected swiftly in order to avoid teasing, exclusion and censure: Kelly puts her hand beside her mouth in order to whisper to Pippa that Bridget's hair is sticking up (there is a bubble of hair in her otherwise perfect ponytail). Pippa passes the whisper on and Bridget, quickly realising that the whispering is about her, demands to know what they're saying. 'What is she talking about, tell me!' Eventually they tell her and she tries to straighten out her hair, embarrassed. (Field notes.) Despite this sexualised 'beautiful people' image and their strong investment in heterosexuality, the hegemonic girls and boys in this class did not form into 'celebrity couples' as was found in Renold's (2005) research. This was possibly not for lack of desire on the part of the 'cool girls', who teased each other constantly about liking Humphrey, and, indeed, there were several fallings-out between girls in the class that related to 'fancying' him. Humphrey and Frederick also claimed in their interview that Chelsea flirted with them. Humphrey himself, however, spoke of a girlfriend in another class and they generally kept themselves aloof from Kelly's group when at school, though they did chat on social media in the evenings. This public distancing of the 'cool' boys from their female counterparts was part of what allowed them to maintain their overall dominance: while the girls attempted to position themselves as the boys' equals, this refusal to take part in romantic relationships within the class underlined the boys' self-positioning as superior even to the 'cool' girls. Although two of the hegemonic girls had previously had relationships with middle status boys, the separation of the hegemonic boys, and Kelly, the central figure of the hegemonic girls, from the majority of the class in terms of 'going out' reflected a general and overt exclusivity that was assiduously practiced by, and granted to, hegemonic children of both genders.
In addition to this exclusive separation regarding romantic liaisons, neither group joined in with playground games involving the whole of the rest of the class. Humphrey and his friends played football every playtime, so were away from their classmates then for that reason, but they also often sat apart from the others, including Kelly's group, at lunchtime. This was explained by Bridget, a hegemonic girl, as being because 'they think they're in a better league than us' (field notes). Kelly and her friends also maintained both a spatial and a social exclusivity during playtimes, which was tacitly accepted by the other children: everyone seems to know that you can't just go and sit with [Kelly's group] and this requires a set invitation and is considered a privilege (field notes) Similarly, at the Christmas disco, the 'cool' girls did not join the other children in dancing, but remained aloof: The 'cool girls' hang out near the wall and refrain from any dancing. Even Pippa doesn't dance, even though I know she is a talented dancer and have seen her dance on the playground. (field notes) This acceptance of the hegemonic status quo by the rest of the class reflects the largely passive nature of counter-hegemonic resistance in this particular setting. This mainly took the form of not aspiring to become part of the 'cool' group, which meant that one could generally avoid trouble, rather than, for example, trying to replace them in their hegemonic position, which would be more likely to result in overt bullying and victimisation (Paechter & Clark, 2016).
The hegemonic girls spent most of playtime on a ramp leading up to an outside storeroom. This ramp had two advantages. First, it had a barrier down one side, which meant that they could easily prevent other children accessing 'their' area uninvited. Second, because it was slightly raised, it allowed them to maintain a panoptic observation of activities across the rest of the playground. From this space, Kelly and her friends would send messages to other children in order to request (footnote continued) ethnicity. their presence or set traps to provoke foolishness that might cause others to be laughed at. This spatial separation of the hegemonic girls in the class also meant that the periodic oustings from the group of one or another girl were visible to all as they involved exclusion from this area. This supported the aura of exclusivity surrounding the group, as it made it very obvious how easy it was for a girl to lose her place there. It also acted as a form of internal and external coercion: any disagreement with or resistance to Kelly's position, in particular, was punished by a public exclusion which operated as a demonstration of her power. The physical aloofness of both male and female hegemonic groups reflected their strong assumption that everyone else in the class aspired to join them and so had to be kept at a distance. Other children were therefore constructed by the dominant children's oral and spatial discourses as having failed masculinities and femininities. This reflects the ways in which hegemonic groups in general privilege their own position and project it as exclusive. The very nature of hegemonic masculinities and femininities includes their comparative rarity as fully enacted, partly because the mobilisation of power from such positionings is made possible by precisely that exclusivity. Like masculinity in general, these identities have to be defended as well as constantly projected as highly desirable.
Maintaining both physical and social spaces into which only the elite were admitted was a significant aspect of Kelly and Humphrey's ability, in particular, to dominate other children in the class and maintain their hegemonic positions. This was augmented by their preparedness to bully and torment weaker peers, either themselves or via other children. This is an example of the force that reciprocally balances consent in Gramscian domination. It operated differently, however, for the two groups. As well as the constant threat of exclusion brought to bear on insiders, Kelly and her friends used subtle or not-sosubtle forms of manipulation and ridicule to keep other children in their place. For example, much of the daily activity of Kelly's group involved gossiping and undermining others, including finding ways of making other children look ridiculous in public: Britney warns the others 'Mia's 3 talking to Kelly now' and sure enough Mia comes back to ask Hedgehog [a boy] what a chav is. Before answering, Chelsea [at the time ousted from the 'cool' group] warns that it's 'a spam from Kelly' and he agrees with her. They tell me that a 'spam' is a question from Kelly via someone else that is not a legitimate question but simply designed to make fun of the answerer. (field notes) Humphrey's coercive methods were generally more straightforward, including overt mocking of weaker boys, by imitating their speech, personal mannerisms or dress. However, his ability to dominate and control other children was generally stronger than Kelly's, and many of the class were frightened of him. Unlike the hegemonic girls, who completely ignored those who were lowest status, Humphrey and his friends overtly asserted and reinforced their dominance by systematically bullying the weakest boys, as well as the weakest girls.
Power and control were also maintained by both 'cool' girl and boy groups through a determined effort to keep their disputes to themselves, and, in particular, not to involve teachers. This latter was known as 'grassing', and the taboo against this was a particular feature of hegemonic femininities and masculinities in this class. Indeed, the extent to which it operated was one feature of the dominant children's ability to manufacture consent (Gramsci, 1971), through their control of local discourses, to practices that were manifestly not in other children's interest. This reflects Litowitz's (2000) argument that 'domination is increasingly a matter of colonizing the internal world of the dominated classes' (524). In Kelly's group it was so important that when Chelsea and Joanna were temporarily ousted without explanation, they suffered in silence for several days without telling any adults. The taboo on 'grassing' also meant that it was easier for both Kelly and Humphrey to maintain an outward appearance of being well socialised, or 'good' children in the eyes of teachers. For example, on one occasion when Humphrey repeatedly subjected a low status girl to unwanted sexualised attention during preparation for a class performance, he presented her passive resistance (being reluctant to touch his hands) as non-cooperation, reporting it as such to the teacher. Her seeming inability to complain about Humphrey's behaviour on this or other occasions, reflecting both the dominant ethos of 'not grassing' and her own group's commitment to being 'nice ' (compliant and uncomplaining), allowed him and his friends to maintaim a regime of sexual harassment of weaker girls without compromising his reputation for good behaviour (Paechter & Clark, 2016). It is notable, also, that Humphrey's hegemonic position allowed him to break the strong taboo on 'grassing' without loss of status.
Despite the fact that both Kelly and Humphrey cultivated a reputation for being 'naughty' among the other children, and that this was a part of their high status, it was also important for this to be masked from teachers. Such findings echo George's (2007) research into girls friendship groups. She found that girls with 'queen bee' status, who controlled their friendship groups in similar ways to Kelly, were seen as exceptionally kind and helpful by their teachers. This reflects wider conceptions of hegemonic feminine behaviour and is another example of how these femininities support traditional gender relations, while preserving dominance within the child group, and especially over other femininities. In the case of the hegemonic girls in our study, this could extend to appearing to be helpful to other children, while at the same time acting unkindly. The following example also shows the way that individual girls in the hegemonic 'cool' group worked to defend and maintain their status within that group by undermining the position of more vulnerable others, in this case, Joanna, who was eventually permanently ousted from the group. In this way, the coercive aspect of hegemony operated even within the dominant group: here it is used to maintain allegiance to Kelly, at the group's apex. The consistent panoptic self-surveillance required for hegemonic group membership is itself part of the mechanism for projecting the desirability of hegemonic positions, and also illustrates once again the control of internal worlds (Litowitz, 2000). Here, Chelsea makes a series of moves (including offering her gluestick to the low status Melissa having implicitly refused Joanna the loan of it) which work to unsettle Joanna while allowing Chelsea to end up appearing to be generous: They have to glue something in their book and Chelsea lends Bridget her glue stick. When Joanna asks to use it Chelsea says there's not much left and shows her (it's almost the whole stick). Then she asks both Melissa and Pippa if they want to use it. Melissa says yes and says softly to Joanna that she will let her use it afterwards. Bridget, sensing something amiss, asks Chelsea if she likes Joanna. I think she says no. Then when Joanna begins to use the gluestick Bridget asks her if she asked Chelsea and looks to Chelsea to answer. Joanna apologises and says she thought Chelsea said yes, and she passes her back the gluestick before she's done with it. Chelsea changes her mind and says she can use it, she's not even done. (Field notes.) The importance given to appearing to be pleasant and helpful, while actually bullying or undermining others, contrasted strongly with the position of the 'nice girls' at the other end of the class hierarchy, for whom actually being kind was a unifying value. By maintaining such appearances, however, the hegemonic girls were able to call on wider discourses of femininity to support their dominance, mobilising power while maintaining traditional gender relations (Litowitz, 2000;McRobbie, 2009).
Much of this covert bullying of weaker children depended on amassing a considerable amount of knowledge about others, alongside maintaining control of what counted as 'important' knowledge within the class. Being at the apex of class knowledge hierarchies was a central aspect of hegemonic masculinities and femininities and was a key element in preserving these children's hegemonic status. Both Kelly and Humphrey, even more than their friends, were adept at amassing knowledge of all kinds (Paechter & Clark, 2010), and in hoarding it until it could be used to best advantage. While most of this information was unimportant in itself, being mainly gossip about who liked whom and other details about the shifting friendships in the class, it was used by the children as a form of currency and a means of buying temporary favour and prestige from the hegemonic groups in general and these two children in particular. As a result, any new titbit of gossip or other information was brought to them. For example, when Dave looked up the word 'masturbation' in his dictionary, he reported the definition to a select group of boys around him, and to Kelly. This reflects the Gramscian nature of hegemonic gender relations: non-hegemonic children maintained their own subordination by offering up information to Kelly and Humphrey in this way and allowing them to judge its importance (Litowitz, 2000;Smart, 1983).
Both Kelly and Humphrey also actively sought out knowledge, either by moving physically through the classroom when it was permitted or by openly asking for it. The knowledge hierarchy in the class also meant that they were able to find out a considerable amount about others while withholding information about themselves. Humphrey was clearly at the apex in this regard, however: when he was called out of class to be reprimanded, Kelly was entirely unable to find out what this was for, despite clearly badly wanting to know, and remarking that 'Humphrey never tells us anything' (field notes). Once again, although hegemonic masculinities and femininities shared the ability to amass and manipulate classroom gossip, information about the most dominant boy was withheld from even the most dominant girl, maintaining traditional binary gender hierarchies.
A particularly desirable sub-set of knowledge was that about sex and sexuality. Related to this was the way in which taking up and defending hegemonic masculinity or femininity also involved a strong investment in heterosexuality. A considerable proportion of classroom gossip related to this, including couple relationships between boys and girls inside and outside of the class, and, in the case of the hegemonic girls, about the fine line between being 'cool' and being 'tarty' (the latter seen as a marker of working-class status in this almost entirely middle-class setting), which was used to cement ideas about inclusion and exclusion from their group. Humphrey and his friend Glazer also used overt sexuality as a way of bullying and harassing weaker girls in the class, by pretending to 'fancy' them and then repeatedly invading their personal space, causing considerable distress.
Although not as able to mobilise power as much as the hegemonic boys, the hegemonic girls were far from being the subservient handmaidens portrayed by Connell (1987) in her depictions of emphasised femininity. In particular, hegemonic femininity in this setting was not, as Connell suggests, 'performed especially to [boys] (188), nor did they reflect any form of 'fragility in mating scenes' or compliance with [boys]' desire for titilation and ego-stroking' (187). On the contrary, they took pride in their assertiveness, reflecting the sassy and go-getting image of McRobbie's (2009) adult 'global girls', and they repeatedly (though unsuccessfully) claimed equality with Humphrey and his friends. They joined in with the dominant boys in deriding the 'niceness' of the lowest status girls, though they did not go so far as to bully them (Paechter & Clark, 2016). They were often seen to stand up to the boys, though even Kelly admitted that Humphrey's derision was frequently more than she felt able to tackle: Kelly says, 'the boys laugh at you, especially Humphrey.' I'm surprised at Kelly's feelings since she seems to have so much power in the class and I suggest that surely she can stand up to him. Kelly says only sometimes. (field notes) Sometimes this challenge to the hegemonic boys took symbolic form, as when the 'cool' girls marched onto the football pitch in a line in order to disrupt the game, or seized the boys' ball and ran off with it. This was something that girls also did in Renold's (2005) study of sexualities among similar aged children, so it is possible that actively resisting the dominance of playground football is a feature of many primary school-based hegemonic femininities. Despite these girls' assertiveness and power relative to other girls, however, they were not able to compete with the hegemonic boys. Their femininities, while paralleling in many ways the local hegemonic masculinities and mobilising similar amounts of power with respect to their own gender, continued to maintain traditional binary gender hierarchies. That this was not for want of trying reflects the social dominance of men in their local area and in wider society.

Conclusion
The relative lack of attention to a theorising of femininity is a serious problem which I have attempted to address in this paper. I have argued that Connell's (1987Connell's ( , 1995Connell's ( , 2002 insistence that there can be no possibility of hegemonic femininity is dependent on the way in which she sets up the concept of hegemonic masculinity so that hegemony can only be accorded to men. This resulted in her characterisation of the counterpart of hegemonic masculinity, 'emphasised femininity', as an essentialised and largely outdated form of femininity that bears little relation to women's and girls' actual lives. I have suggested that an alternative approach might be to start from a gender-neutral and more strongly Gramscian conception of hegemonic gender performances. This reconceptualisation allows us to consider forms of hegemonic femininity, which, like hegemonic masculinities, are locally produced and reflect the dominant and aspirational femininity in a particular context. Exploring this through an example from my own research, it has become clear that hegemonic forms of femininity do not have to conform to Connell's (1987) model of emphasised femininity, and can, instead be more like McRobbie's (2009) assertive and ambitious 'global girls'. Indeed, in some contexts, including the one discussed here, hegemonic masculinities and femininities, rather than being opposites, may actually have a considerable number of characteristics in common. While both uphold traditional gender binaries and preserve a gender order dominated by men, they can operate socially in very similar ways. Although the 'cool' girls in my study were subordinate to the 'cool' boys, they in no way performed the subervient and male-focused emphasised femininity described by Connell (1987). On the contrary, they positioned themselves as strong and independent, maintained a constant resistance to this positioning, and mobilised considerable power in relation to the remaining boys in the class as well as to other girls. That this should be the case is unsurprising once we understand that hegemonic femininities parallel hegemonic masculinities in being constructed in relation to other femininities, not, as with emphasised femininity, as a subservient Other to a particular form of hegemonic masculinity. Recognising the possibility of hegemonic femininities, and conceptualising them as working with hegemonic masculinities to preserve a still patriarchal and binary gender order, should enable us to gain further and more nuanced insights into the mutual relationship of masculinities and femininities and how power moves and is mobilised between them.