“Only What’s Right”: Normalising Children’s Gender Discourses in Kindergarten (The Case of Montenegro)

This article presents results from qualitative research on children’s dominant gender discourses in kindergarten and the influence of the socio-pedagogical aspects of kindergarten culture, transmitted via teachers’ gender discourses and personal epistemologies, on the construction of children’s gender discourses and identities. The main questions guiding our research were: What gender stories are narrated in a group


Introduction
This study aimed to examine children's dominant gender discourses in the context of early childhood education settings in Montenegro, as well as under which sociopedagogical influences those discourses are developed. The understanding of the construct of gender in our article is based on the feminist poststructuralist tradition, which views gender as constituted within discourse and constantly changing through human interactions in a social context (Yelland and Grieshaber 2003). Gender is performative (Butler 2002), meaning that gender identity is always a certain gender behaviour within a certain cultural context. Children are not passive recipients but active agents in constructing their own gender identity. Still, our social reality is a highly controlled "market" of ideas, possibilities and meanings, where some meanings are more powerful than others since they are more accessible, more desirable, carry more pleasure, and are more easily recognised by others (MacNaughton 2000(MacNaughton , 2006Robinson and Jones-Diaz 2006), so not every possible meaning of gender is equally desirable or approved by society.
Much research has shown that education institutions function in such a way as to normalise and regulate certain discourses via "regimes of truth" (Foucault 1977(Foucault , 1978, thus supporting certain gender power relations (Robinson and Jones-Diaz 2006), often affirming binary oppositions of "hegemonic masculinity" and "emphasised femininity" and acknowledging their asymmetrical positions in a patriarchal gender order (Connell 1987). "Hegemonic masculinity" is at the top of the gender hierarchy, with all other forms of masculinities complicit, subordinated, and marginalised (Connell 1987), and with femininity also subordinated to it.
Many studies on children's play, toys, play areas and ways of interacting in kindergarten are consistent in their findings that gender stereotypes are undoubtedly present at all these sites, and are also used by children as a way of maintaining their gender category (Adriany 2019b;Blaise and Taylor 2012;Chapman 2016;Chick, Heilman-Houser, and Hunter 2002;Francis 2010;Hughes 2003;Kane 2006;Paechter 2007). Along with these findings, research on teachers' perspectives on the gender aspect of kindergarten life predominantly shows a binary gender matrix with distinctly opposite perceptions of boys and girls (Browne 2004;Hyun and Tyler 2000;Robinson and Jones-Diaz 2006;Warin and Adriany 2017).
However, studies of gender in early childhood institutions often neglect the "soft" institutional mechanisms by which children's gender discourses and identities are shaped, focusing mostly on gender stereotyping in play, toys, and role-playing. Yet, there is some research indicating that the way teachers engage in children's learning activities and play (which is in accordance with their personal epistemologies) influences, shapes, and even normalises children's gender discourses and identities (Hyun and Tyler 2000;Pavlović-Breneselović and Krnjaja 2016;Woodward 2003). The aim of our study was to understand how children "get their gender right", and what they actually perceive as the "right" way to perform gender in kindergarten. However, since the forming and performance of gender identity do not happen in social isolation, we were particularly interested to understand the ways in which the institutional ethos affirms one way and silences other ways of being a boy and a girl in kindergarten. More specifically, we wanted to understand if and how some aspects regarding teachers' personal epistemologies-their motivational styles and the expectations and values that they set in the room regarding children's behaviour-affect children's gender discourses. It is exactly these aspects of teachers' behaviour that are usually perceived as solely "educational" and teaching-related or learning-related, and which are often neglected in research regarding children's gender discourses and identity formation.
This research was conducted during a two-week period in two kindergartens in Cetinje, a medium-sized city in Montenegro. The participants were 54 children (24 boys and 30 girls) and four teachers. The study included participant observation, video recording, field notes, semi-structured interviews with children and child-led tours of their favourite activities and places in the kindergarten rooms. The main questions guiding our research were: What gender stories are narrated in a group, and under which influences do these stories become established as norms?

Living out Gender in Kindergarten
There are several core aspects in children's kindergarten lives that are most suitable for the performance of gender-which is why they are most commonly studied-and they are all connected to play. Diverse studies are unified in recognising children's highly gender-stereotyped preferences in the domain of play.
In the learning process, play has a critical role in perpetuating heteronormative discourses (Robinson and Jones-Diaz 2006, 142). An analysis of play shows that boys and girls construct and maintain gender order through play itself, impacting each other's gendered social order (Dunby 2003). Blaise and Taylor (2012) insist that play is never "just play", but that it is always charged with a power struggle, as a serious job of "gender category maintenance" (Davies 1989). Simultaneously, the toys themselves instruct and lead children in certain ways of doing gender, since "a key discourse perpetuated by the toys is that of gender difference" (Francis 2010, 334) with a binary gender distinction-typical boys' toys and games convey the ideal of a man as an active agent while girls' toys and games insist on feminine aesthetics and care (Adriany 2019b;Blaise and Taylor 2012;Chapman 2016;Cherney and London 2006;Francis 2010;Chick, Heilman-Houser, and Hunter 2002;Freeman 2007;Hughes 2003;Kane 2006;Paechter 2007).
The role of a teacher in the construction of children's gender identity is very important in the context of early childhood education settings. The teachers' pedagogical gaze is mirrored in their planning, the design of the environment, the design of the play area, the choice and display of resources, the nature of their interaction with children, the explicit or implicit norms they establish in the room, and the learning and behavioural expectations from boys and girls, and this significantly influences the context for the performance and construction of children's gender. Studies have shown that the concrete pedagogical strategies that are most commonly employed in order to rethink and subvert oppressive and stereotyped gender discourses and practices in kindergarten-such as feminisation, separatism, fusion, and policing (MacNaughton 2000)-have no lasting effect since they are based on a simplistic view of boys and girls and their gender construction, and do not tackle the key issue in the construction of gender identity-the issue of power (Robinson and Jones-Diaz 2006). Such teachers' actions are focused mainly on gender stereotyping. Instead, in order for lasting change to happen, kindergarten must become a dialogic learning community, within which many gender storylines can be examined by teachers and by children, in constant open discussion. For this to be realised, teachers must be reflective practitioners, highly aware of the impact of their gender discourses and personal epistemologies on the construction of children's gender identity. MacNaughton (2000MacNaughton ( , 2003 believes that, based on the poststructuralist-feminist paradigm, teachers should experiment with ways to reduce sexism in a group, talk more about traits that are common for boys and girls than to insist on their differences, expand children's understanding of gender in a way that is not based on the existing gender hierarchy, and empower children to investigate many ways by which they can be and become men and women.

Gender Aspects of Kindergarten Socio-Pedagogical Practices
Children do not construct their gender identities in social isolation-it is always a "communal project" sui generis, where all aspects of reality impact gender discourses and practices. In an institution such as kindergarten, the process of gender construction for a child is influenced by many social interactions (with the teacher and with the other children), but also by many more or less explicit "norms of being" that the institution proclaims, simultaneously silencing anything that is on the other side of those norms. This process happens mostly via a firmly established set of expectations and rules, conveyed usually by teachers-thus teachers' gender discourses, gender pedagogies and personal epistemologies become especially pronounced-or by some other ecological elements of organisation, such as space or time (Bronfenbrenner and Morris 2006). Though this aspect of gender construction is often neglected in studies, some research shows a common set of teacher behaviours that are part of the broader socio-pedagogical climate in the institution and that are repeatedly found to influence the way children construct their gender identities, as presented below.
Kindergarten institutions, by preserving the culture of a "good pupil" ("good kid") who is glorified, implicitly strengthen gender dualism and thinking in binary categories where boys and girls are opposed. Quite often, girls take on or have a "quasi-teacher" role imposed on them, which expects them to display reasonable behaviour as the "assistant" or the teacher's helpful "coordinator", and this is in contrast to the unreasonable and "silly" behaviour of boys (Chick, Heilman-Houser, and Hunter 2002;Frances 2000in Martin 2011Pavlović-Breneselović and Krnjaja 2016). This is consistent with the research of , which shows that girls are significantly more encouraged in terms of all behaviours dealing with caring, fixing, cleaning, and helping others, thus becoming "little helpers in kindergarten" (Meland and Kaltvedt 2019, 98).
Many studies show that teachers' discourses also shape children's gender roles. They do it more or less explicitly, and it includes the criteria for rewarding certain behaviours in boys and girls. The research shows that girls are most often commended for their looks-the way they dress, their hairstyle, and their behaviour concerned with caring and helping others; on the other hand, boys' height or size is accentuated the most in a positive context (Chick, Heilman-Houser, and Hunter 2002, 152). Teachers contribute to sustaining traditional gender stereotypes in kindergarten, emphasising children's appearance, the way they are dressed, their size and physical skills, for example, commenting on boys' size and strength, while calling girls "sweet and likeable" (Meland and Kaltvedt 2019, 97). With their different expectations for boys and girls, the teachers also influence the perpetuation of the traditional binary image of gender relations: girls should sit still and be patient, while the same is not expected of boys . Meland, Kaltvedt, and Reikerås (2019) believe that a significant difference between boys and girls in certain aspects of play, especially in pretend play, can be explained partly by the different expectations the teachers have of boys and girls.
The teacher's choice of practices for achieving gender equality in kindergarten lies first and foremost in his/her epistemological theories-the theories about how a child learns-as shown in numerous research works (MacNaughton 2000(MacNaughton , 2003. If their personal epistemologies are more objectivist in nature Pintrich 1997, 2002)-that is, that there is one objective source of knowledge, there is one correct answer, and knowledge is justified by external sources-it could lead to pedagogical practices that promote predetermined and fixed gender values and role models as norms, thus normalising children's gender discourses and restricting them from learning and understanding any alternatives. The field of early childhood education lacks research on this relationship, but in the context of studies of elementary and secondary schools it has been unambiguously shown that the objectivist epistemology of teachers leads to a more controlling motivation style, often praising more feminine features and behaviours, such as working quietly, obedience, listening to the teacher, and so on (Reeve et al. 2014;Roth and Weinstock 2013).

Context of the Research
In 2006, Montenegro was the last ex-Yugoslavia country to gain its independence. The process of Montenegrin society's post-socialist transformation is affected by the traditional and modern value systems (like most of the south-east Balkan countries), through a combination of patriarchal-collectivistic values with modern tendencies characterised by individualism (Lučić 2016).
Regional research within sociology and studies of childhood also support the idea of the paternalism as the dominant modus of a collectivistic patriarchal society (Colić 1997;Tomanović 2003Tomanović , 2004Tomanović and Ignjatović 2006;Trebješanin 2008), which is also characterised by a distinct male-female dichotomy, with the male child enjoying privileges (Trebješanin 2008). The value of men can be found in folk proverbs that are still relevant today, such as "A male that is like a poppy seed is still better than a female one", or "When a boy is born even the tiles on the roof rejoice-and when a girl is born, even the fireplace cries" (Trebješanin 2008, 255). In such a patriarchal macrosystem (Bronfenbrenner and Morris 2006), there exists a rare phenomenon known as virdžinaa cultural phenomenon in Montenegro, parts of Serbia and Albania by which, if there is no male baby born into the family, the youngest girl "becomes" a boy, pledging virginity and to live like a man until the end of her life.
The legislative and strategic framework of Montenegro's preschool education policy does not explicitly delve into children's gender identity (Crna Gora Zavod za Školstvo 2011; Ministarstvo prosvjete Crne Gore 2016)-only under the broader umbrella of individualisation as a general educational principle.
Research into this phenomenon is also almost non-existent, as far as we are aware. Only recently a study on educators' perceptions of children's gender identities affirmed the dominance of the discourse of social determinism in teachers' understandings of the development of children's gender identities, with a predominant focus on gender stereotypes and sporadic activities on gender re-socialisation, when and if needed (Milić et al. 2021).
The required qualification of early childhood educators in Montenegro takes three years to complete (180 ECTS). In the official curriculum for the Department of Early Childhood Education, there is no explicit content regarding gender issues in education. In the current catalogue of programmes for the professional development of in-service educators (Crna Gora Zavod za školstvo 2019), there are no seminars dedicated to the subject of gender in the early childhood education context.

Study Location
We decided to carry out the research in Cetinje, a city that is part of a group of mediumsized cities in the country. Montenegro is geographically a very small country. The socio-economic status of those in its cities is more or less equal and their "ethnicity" is uniform, so Cetinje was a valid representative of the Montenegrin population. Most kindergarten institutions in Montenegro are city-based, so the urban/rural criterion was not applicable to our choice of kindergartens. Cetinje has only one preschool institution that is city-based, so we did our research in two of its kindergartens.

Methods
Since the aim of our study was to identify the most prominent gender discourses of children in kindergarten, we directed the research towards the following questions: Which are the dominant discourses of femininity and masculinity in children's narratives? (How do children categorise themselves and others?) How do the given discourses normalise children's everyday life? (By what practices and patterns of emotional investment?) And, what mechanisms of institutional life possibly shape their gender discourses?
We wanted to give the most direct voice to children in the research (Prout and James 2005) and to reveal the children's perspective (Sommer, Pramling Samuelsson, and Hundeide 2010), so we used participant observation, video recording, field notes, semistructured interviews with children and child-led tours of the children's favourite activities in the kindergarten, as ways of listening to and consulting with children, in accordance with the Mosaic approach (Clark 2005;Clark and Moss 2001). Semistructured (spontaneous) group interviews were especially suitable for our research, since they enabled us to most directly discover some of the causes behind certain gender discourses of children, in a context which is most suitable for them-being surrounded by close friends in kindergartens. We chose to employ child-led tours through which we wanted to discover what kind of impact the play-centres and didactic materials and toys have on children's gender discourses.
When analysing the received data, we applied feminist-poststructuralist discourse analysis (MacNaughton 2000(MacNaughton , 2003 as a conceptual research tool, which was necessary in deconstructing social relations and understanding how the dominant discourses work, the dominant ways of categorising self and others, the patterns of emotional investment, and dominant "regimes of truth" (Foucault 1980) by which the process of (hetero)normalisation in the kindergarten is carried out.

Research Procedure
In the study 54 children and four teachers from two kindergartens participated (24 boys and 30 girls aged 5 or 6; 2 female teachers from each kindergarten). We spent seven days in each group, two on the process of adapting to each other, and five on the research itself. We were in the groups every day from 8 am to 1 pm. We spent a larger amount of time with two of the teachers (Teacher 1 and Teacher 3), since the "morning teachers" stay in the group until 12 am (midday); the "afternoon teachers" are in the group from 11 am onwards. We used a hand-held video camera.
We conducted interviews in groups of four to six children. The groups were almost always gender-based, since we started our interviews during some of their free-play activity time, when typically boys and girls gather in small, same-sex groups. We usually did our interviews sitting with children at the table while they were drawing or painting something. During that time, the other children and the teacher continued with their activities. Ten girls and 14 boys actively participated in the interview discussions.
Tours were initiated on an individual basis, and 13 girls and 11 boys participated in the child-led tours. We asked them if they would like to show us some of their favourite places to play in kindergarten, and as the tour unfolded, we talked about the reasons for their choices. Usually the tours and interviews happened between 10:30am and 11:30am.
About 26 hours of footage were taken (transcribed observations, interviews and childled tours), along with the field notes. In the transcript-coding process, we singled out 23 "raw" codes (Carspecken 1996) of higher and lower orders. Then we chose the analytical and thematic perspective for further analysis, sorting the codes into several categories, and the most dominant were: 1) gender stereotyping, and 2) kindergarten culture as the agent of normalisation of children's gender performance.

Research Ethics
We received research approval from the Bureau for Education Services and the Ministry of Education of Montenegro. In addition to that, all four teachers gave their consent to participate in the research on a voluntary basis, after the topic, goal, and method of the research were introduced to them at an opening meeting, which satisfied the ethical imperative of eliminating deception (Willig and Stainton-Rogers 2017). We explicitly stated to teachers several times during the study period that they were allowed to stop the research at any time, considering their or the children's needs, therefore realising continuous consent of the participants (Adriany 2019a). The teachers introduced the children's parents to the topic and aims of the study, and on that basis we were given a common consent for the research with their children (Hammersley and Atkinson 2007;Pole and Morrison 2003). The subjects' privacy (Hammersley and Atkinson 2007;Pole and Morrison 2003) was protected by putting the subjects' initials, rather than names, on all transcribed statements. Given the sensitivity of the topic, we were especially considerate of our behaviour and utterances in the research field, carefully framing every question that we prepared for the children, and being highly sensitive to their verbal and nonverbal behaviour (e.g., if a child looked bored or refused to answer, we did not proceed with the conversation).

"Only what's right": Gender Stereotypes in Kindergarten
The gender stereotype discourse is strongly present in children's narratives about toys, activities, roles, and professions, even in the boys' and girls' mutual perceptions.
The interview excerpts unambiguously show that both boys and girls prefer playing with toys typical for their gender group. The boys listed cars, blocks, puzzles, figures of superheroes and cartoon characters (Spiderman and Star Wars), lasers, and tools, while the girls' choice is always dolls, and these toys were often tightly linked with the play area that they preferred-the Construction Corner for boys and the Dramatic Area for girls. Typically, when asked which areas are their favourite to play in at kindergarten, the boys would lead us to the corner with blocks and cars (the Construction Corner), and for each of the 13 girls who participated in child-led tours, two favourite play areas in the kindergarten were the Dramatic Play area (family-themed role-playing: "mums", "aunts", "housewives", and "female neighbours", with the obligatory set of toys, such as babies, a baby stroller, a kitchen, and activities such as changing diapers, dressing babies, cooking for the family, and so on) and the Art Centre (Tour 1, 7, 9). Numerous studies confirm the children's gender-stereotyped toy preferences (Chick, Heilman-House, and Hunter 2002;Francis 2010), which reflect the discourse of binary opposition that is supported by the narrative of the male as an active and strong subject and of females as tender, caring, and concerned with aesthetics and appearance (Francis 2010).
What is particularly symptomatic is how toys in our research appear as an important rationale in conversations with children about why they cannot play together, especially for the boys. Occasional mutual discrimination between boys and girls is often based on stereotyping toys, games, or some qualities they personally see as typically boyish or girlish. As in the case of boy O. who says he does not want to be with girls, because boys and girls are simply polar opposites. Boy Đ. joins him, saying he likes nothing about girls, only "what's right"-"and stuff", something little girls cannot possibly do. So, (baby) dolls are unacceptable for these boys simply because they are a part of the girl culture. This is boys' way of affirming their position, by not participating in activities associated with girls. As Davies discusses, through sending signals such as "this is what disgusts me" (Davies 2006, 433), they actually signal that it is not what they are, positioning themselves as heterosexual and dominant. Obviously, toys appear as highly important symbols or icons through which the status of masculine/feminine is established and lived out. In the aforementioned ways, boys and girls mutually exclude themselves from an opposite sex group, using the symbolism of toys or play areas as important gender signifiers to position themselves within a certain society-approved category. This is a strong manifestation of "gender category maintenance" (Davies 1989). In this manner, boys and girls participate in a continuous process of mutual "othering"-learning that the traits of the other gender group are not desirable or appropriate to have or even to "try"-and this learning is manifested at the somatic level as well.
When we asked them about their desired future professions, the answer was also highly gendered. The girls would like to be a ballerina, a queen, a princess, a YouTuber or a gymnast (Interview II, 13, 72-75). The desired professions for boys were mostly ones that involve strength, speed, fighting, and interest in machines, such as a boxer, an electrician, Spiderman, a soldier, and so on (Interview I, 10, 32-46). Boy Đ. was whispering when stating that he is "interested in being a wife, a housewife, cleaning, too" (Field note 10, 16-20), and when asked why he does not try, he said that he is a boy. He is acutely aware of the "right" way of practising his gender identity and the potential risk of being excluded or branded as less masculine or a sissy (Kimmel and Mahler 2003;Kostas 2022) if he perhaps, quietly and with a smile, reveals his interest in housework. As Connell states, "to sustain a given pattern of hegemony requires the policing of men" (Connell and Messerschmidt 2005, 844) and undertaking a variety of strategies to ensure that their manhood is unquestionable (Kimmel and Mahler 2003), which is even more important for boys because of the "particular salience of masculinity for power relations" (Paechter 2007, 62), that is, "they have more to lose" (Kostas 2019, 14).

How the Kindergarten Culture (Hetero)Normalises Gender Discourses
We would like to point out several aspects of kindergarten culture that we found implicitly shape and normalise children's gender discourses, specifically via teacher narratives and the personal epistemologies and practices that are grounded on those beliefs. We categorised these teacher practices as: affirming stereotypical gender roles, subdividing children explicitly by their gender, the use of terms of endearment (for girls) and, most importantly, promoting a gender construct that is based on the concept of behaving. These practices are then established as important criteria for gender subdivision among the children themselves.

"Behaving" as an Important Gender Aspect
When we asked the girls if there is a boy(s) that they like to play with, the strongest criterion for that decision was their (good) behaviour, specifically, girls would like to play with boys who: 1) "are good", which means 2) "they behave well" (they are good and behave), which, furthermore, means that 3) they behave well, "like the girls".  (Interview I,13,(51)(52)(53)(54)(55)(56)(57)(58) In searching for the children's perceptions of the meaning of "behaving well", we identified that it is influenced by what the teacher allows or does not allow, meaning that "good behaviour" is "doing what we are told to" (by the teachers). So, "behaving well" equals "behaving like a girl" and "not behaving like a boy". The teacher explicitly reinforces this discourse by making unambiguous statements, such as "look at" (O I,3,63;O II,4,43;O I,5,23,27; O I, 1, 33) the girls or a specific girl as a role model of good behaviour in kindergarten, or "we do not like" (O I,1,11;O II,2,33;O I,5,43,47) a specific boy's behaviour. Girls perpetuate these values, pointing out that they would not like to be boys because the boys always: 1) mess around, 2) are naughty, and 3) do not behave when they are told to (Interview I,13,. Children's narratives very much mirror the narrative of their teachers. Analysing the transcripts from video-taped observations and well as our field notes, we noted that the teachers' narratives are saturated with statements such as: "You must listen to the teacher" (O I, 1, 42), "Oh, you are all so good today … you listen to everything I say" (O I, 2, 49), "See how the girls are listening to me" (O II, 5, 22) (speaking to boys), "If you're not quiet and don't behave, we won't go outside" (O II, 3, 17), "Boys, can you finally stop making all those noises" (O I, 1, 33) (while boys are role-playing as truck drivers), "You have to listen to me carefully. … If you're not quiet and listen, you won't know anything" (O I, 4, 31, 32), and "My lovely girls, so good and neat, they always listen to their teacher" (O II, 3, 39) (speaking to a teacher from another group about her girls). Thus, being quiet, being good and listening to the teacher (to name a few) become strong criteria for praising the behaviour of children-praising predominantly, or even exclusively, girls. In the same vein, behaviours that do not fit into the category of "quiet, good, listening, sitting, behaving"-for example, running, chasing, talking, walkingare criticised by teachers (boys and their behaviour). So, the marketplace of meaning (MacNaughton 2000) is, in a way, already set up, and the choice for children is somewhat narrow.
Here we can attest to how a specific set of behavioural norms becomes tightly connected to children's gender performance in kindergarten, while a certain ethical value is added to living out gender identities in the room-the "good" ones are those who are quiet, who sit down, listen to the teacher, and do not make noises-which are the practices associated with girls; while everything that is opposite, and associated with boys, is in a way pathologised and labelled as "bad". This is how, in Foucauldian terms, the ethical aspects of a regime of truth become institutionalised (Foucault 1977(Foucault , 1978)-teachers lend their support to a certain gender discourse, promoting a particular system of morality for the institution. Other research from Balkan countries also confirms this finding-that the ideal of a child behaving in kindergarten is very feminine in nature, and is used as a norm regardless of gender (for boys and for girls equally) (Pavlović-Breneselović and Krnjaja 2016).
Moreover, these values of listening, sitting, and not making noises are of an epistemological nature with the underlining premise that children will not be able to know (to know knowledge, or to know how to behave) if they do not listen to the teacher. This is in accordance with our previous research on teachers' epistemologies in Montenegrin kindergartens (Marojević, Todorović, Milić 2020), which are objectivist in nature and praise listening to the teacher, being quiet, neat, sitting down and working at the table, and so on. The teacher is the one who is recognised as the crucial source of knowledge, as well the one who justifies it. It is precisely these two things that are important dimensions of the teachers' objectivist personal epistemologies Pintrich 1997, 2002). What we can conclude here, implicitly, is that if the teacher's pedagogical gaze (MacNaughton 2000) is based on objectivist epistemologies and if the teacher is not a reflective practitioner, the less their teaching supports autonomy in children and the more controlling it is (Reeve et al. 2014;Roth and Weinstock 2013), often constraining children's active construction of meaning in any aspect of their identity, including gender. Kindergarten culture is even more prone to those epistemological dynamics considering the adult-child power relationships that are even more pronounced in early childhood education.
Some researchers have found that the expectations teachers have of boys and girls are different . Our research shows different results: teachers' expectations of all children are somewhat girl-specific, meaning that there is a core set of behaviours that is requested from boys and girls equally and encouraged by teachers in the kindergarten room, a core that is feminine in nature-being quiet, being good, listening to the teacher, sitting, being neat-which means that it actually better suits girls than boys in the group, and consequently girls are often praised for behaving in this manner, while the boys are often criticised for not behaving accordingly. This is in accordance with the idea of kindergartens' "feminine" ethos (Walkerdine 1988), which is also confirmed by numerous studies: teachers encourage both boys and girls to display more "feminine" ways of behaving-playing quietly, helping others, sitting and moving less (Woodward 2003); they are more intimate with the girls, relying on them for help more often (Pavlović-Breneselović and Krnjaja 2016), commending them for such behaviour (Chick, Heilman-Houser, and Hunter 2002, 151), and deeming them "easier" to work with and more "teachable" (Hyun and Tyler 2000, 349), while boys are mentioned in terms associated with punishment and called "bad" significantly more often (Pavlović-Breneselović and Krnjaja 2016). Boys are perceived as less "teachable" (Hyun and Tyler 2000, 349) and more stressful to work with (Hyun and Tyler 2000).

Teachers' Practices that Affirm the Binary Opposition Gender Discourse
We also identified that teachers use a language of endearment specifically with girls, often calling them terms such as "sweetie", "my sweetheart", "my sweet little sunshine", "my honey", and referring to their appearance as well: "Look at her, how beautiful she is today" (O II,5,43), "Oh, you look like a real princess" (O I, 1, 27), "Who did your braids today? They look perfect!" (O I, 3, 39), and so on. The boys were never the subject of such language from the teacher. Not once did we note the teacher referring to boys' looks, wardrobe, or appearance. Yet, the boys themselves in a few instances recognised the appearance of a girl as a very important trait of a girl, almost imitating the vocabulary typical of teachers: Boy B.: I'm in love with N. Teacher 1.: How do you know that? Boy B.: Easy, she's good, she's neat, she wears dresses, that's why I love her. (O I, 4, 74-76) In discussion with the teacher about future husbands, the girls mainly pointed out the way boys "behave" as a desirable feature of a future husband: Teacher 2.: … and how would you like him to be? (Her future husband.) Girl A.: I would like him to be sweet … Teacher 2.: … Anything else …? Girl A.: To always do as I say … When I ask him: "Take me to the market to buy something for the kids", then he immediately takes me … (Interview I,2,(17)(18)(19)(20)(21) We would like to remind the reader that this is the same phrase that the other girl used to describe boys who are not well behaved-ones that do not "do as they're told".
In both cases the influence of the teacher's discourse around what is supposed to constitute a "good boy" and a "good girl" is obvious in the very choice of words that the children made. What the teacher says is extremely important and something to refer to (on one occasion girl A. said to girl H. that she would "have to ask the president of the country as well as the teacher" [Field note 2, 76] if she wants to get married when the day comes).
In Kindergarten 1 (Teachers 1 and 2), on four of the five observation days the children played outdoors after a group activity in the classroom, which usually took place at tables-including drawing, painting, making puzzles or playing with Lego (at the table). On all four days-four times, Teacher 1 explicitly asked the girls (one or more of them) to tidy the tables and the room, while the other children prepared to play outdoors and put on their shoes (O I, 2, 44, 45; O I, 1, 93; O I, 2, 86). The observation transcripts from Kindergarten 2 show that in Kindergarten 2 (Teacher 3 and Teacher 4) every day it happened that Teacher 3 gave napkins specifically to the girls, to help with cleaning the tables after breakfast.
Since the girls did not question the teacher's decisions even once, and since the teachers choose girls for the job of cleaning routinely, it seems that for the girls, as well as for the teachers, it was the most common, "natural", and "normal" behaviour that did not inherently present any problem. Indeed, there is a certain kind of morality associated with it that prevents it from being questioned in any way. This practice, embedded in the emphasised femininity discourse, is institutionalised and, as such, has a "hidden power" that "precludes debate, therefore marginalizing and/or silencing alternatives: everyone 'just knows' that they are right and normal" (MacNaughton 2000, 52). Following this line of thought, girls and teachers in kindergartens "just know" that it is "normal" to give a girl a cleaning job.
All the teachers in our research use gender as a criterion for separating girls from boys on some occasions, thus "benefiting" girls. There is a certain gender-based rule that girls should do some things first, as the excerpts from the observation show: Teacher 2: Are you done girls? (Asks girls that are putting their shoes on to go outside.) Come on boys, sit at the table … Be patient for a minute, the girls are finishing, and then you will have your turn to put your shoes on … T. (boy): But the girls have to be first … Teacher 2: Well, girls have to be first, they take precedence … They are girls … (O I, 4, 65-67) In these three aspects-terms of endearment followed by emphasis on a girl's appearance, reinforcing stereotypical gender roles for girls, as well as this specific rule that the girls do some things first-what is noticeable is that the teacher can very much influence and affirm boys' and girls' own perceptions of what is typically feminine or masculine. All these teacher practices have elements typical of the emphasised femininity discourse-compliance with subordination and compliance to patriarchy (Connell and Messerschmidt 2005;Kostas 2014). We need to emphasise again that Montenegrin teachers do not have formal pre-service or in-service training in gender equity or gender identity formation in children. Teachers' gender discourses are, in such a context, probably mostly shaped by their own personal experiences (Chapman 2022;Robinson and Jones-Diaz 2006) and influenced by the highly patriarchal society and culture in Montenegro, where the value of a male child is still much greater (confirmed by numerous practices, among others, by the cases of selective abortion of female babies, whereby for many years, even decades, the number of newborn baby boys has been higher than the number of newborn baby girls [Statistical Office of Montenegro 2020]). Hence, teachers' educational practices are probably in accordance with stereotypical and heteronormative ideas of boys' and girls' roles and activities.
As we stated earlier, similar findings were found in numerous other research works (Chick, Heilman-Houser, and Hunter 2002;Frances 2000in Martin 2011Pavlović-Breneselović and Krnjaja 2016;Walkerdine 1988), confirming the same idea-that teachers' gender discourses, expectations and the pedagogical praxis based on them significantly shape children's gender discourses. Our main thesis is that the influence of the kindergarten ethos is undoubtedly strong, and more feminine in nature, and though it could seem as if it best suits girls or "benefits" them sometimes, it actually perpetuates a "magic circle" of binary opposition discourses and practices (Pavlović-Breneselović and Krnjaja 2016, 66).

Conclusion
We believe that the still predominantly patriarchal Montenegrin society plays a huge role in shaping gender discourses throughout embodied sites of micro-, mezo-, exo-, macro-, and chrono-systems (Bronfenbrenner and Morris 2006) in institutions, the family and kindergarten being the most influential of these. This is also true for Montenegrin teachers, who do not have any official counterbalancing mechanism to help them prevent, become aware of, or subvert the typical binary gender narrative within categories of hegemonic masculinity and emphasised femininity (Connell 1987) through their pre-service or in-service training, for example.
Yet, the period of early childhood education puts teachers in a "unique position to subvert a continuation of traditionally differentiated gender roles through genderflexible teaching because they have the opportunity to intervene at the very beginning of an individual's educational trajectory, yet this rarely occurs" (Warin and Adriany 2017, 376). Often that crucial period of formation is missed if teachers do not operate as reflective practitioners, as was the case in our study. Many studies have shown that a key tool for revaluing binary gender categorisation in kindergarten is the teacher's gender pedagogies and personal epistemologies (Hofer and Pintrich 2002;MacNaughton 2000). Teachers should be aware of their implicit gender theories, implicit pedagogies, and epistemological theories, and they should rely on them to work on unconditioning and subverting ontoepistemic injustice towards children (Murris 2016) and subverting essentialist thinking (Davies 2003).

Implications for Policy and Practice
The main implication of our research refers to the necessity of studying and rethinking Montenegrin early childhood teachers' gender discourses that still seem to be strongly based on traditional patriarchal values. We strongly agree with MacNaughton (1997, 324) that the "teachers' pedagogical gaze is the key site in the curriculum where the gendered exercise of power between children could be made visible", so the main pedagogical work should be carried out through a feminist reconstruction of that gaze. This could best be done in the context of their own practice, through programmes for their professional development that could include action research or personal histories, with a strong critical, feminist-poststructuralist orientation, to name a few, aimed at understanding and deconstructing their own gender pedagogies and personal epistemologies and the ways the latter influence the construction of children's gender identity.
The findings from our research suggest that Montenegrin early childhood educators may not be aware of how the culture and ethos of a kindergarten may implicitly convey traditional and stereotypical gender values and attitudes and, more importantly, how their objectivist epistemologies (mirrored in highly instructive motivational styles, praising "feminine" values of order, silence, and listening to teachers) may also preserve certain oppressive gender power relations among children. This critical awareness of teachers could be achieved through programmes for professional development that would be highly contextualised within the milieu of their everyday practices within the kindergarten as a learning community. This is in accordance with the findings of many studies that suggest that the continuing professional development of early childhood educators must be deeply connected with their practice and focused on developing their reflective capacities, and that the least effective form of professional development is through short-term theoretical seminars (European Commission 2014; Mitter and Putcha 2018;Peeters et al. 2015;Peeters 2016;Urban et al. 2011). Yet, the latter is the most common, if not the only, professional form of development in Montenegro (Peeters 2016).
We strongly recommend further research focusing narrowly on the relationship between teachers' epistemological theories and the dominant gender discourses in kindergarten. We also recommend further research about the regulation mechanisms in kindergarten (Gore 2005) and how they influence the perpetuation of oppressive gender power relations.

Limitations
Findings from our study should be considered in the light of the small sample size and relatively restricted time frame that the research was based on; yet, considering the chosen methodology, it was not our attention to make any generalisation, but to trace and understand the process of establishing dominant gender discourses in the research field. In addition to that, we think that some other methodological approaches could further explain the link between teachers' epistemological beliefs and their gender practices in kindergarten-for example, using quantitative measures of teachers' personal epistemologies (e.g., Epistemological Beliefs Survey by Kardash and Wood [2000]) and then linking them to specific gender pedagogies that teachers perform.