When Does »Too Early« Become »Too Late«? Reflections of Croatian Secondary School Educators on the Persistence of LGBT Taboos in the Education System

• The article presents one of the topics generated by a thematic analysis of data collected through a focus group process within the scientific research project LGBT (In)Visibility in School: The Educators’ Perspective, which focused on the taboo position of sexual and gender diversity within the education system. In four focus groups conducted with secondary school educators in Zagreb, the participants identified certain key factors in perpetuating such a position: the understanding of the child/student as an innocent being whose sexual education should begin at a time that educators themselves are unable to determine, the perception of parents as barriers to the inclusion of topics of sexual and gender diversity in the curriculum, and the absence of a systematic, LGBT-in-clusive approach to teaching about sexuality. The concluding part of the article discusses the limitations of the research conducted and makes recommendations for future empirical and practical coverage of this topic.


Introduction
The topic of LGBT inclusivity within the education system -by which we mean the measure and quality of the responsiveness of the education system to the needs of LGBT individuals, including the representation of LGBT topics in the curriculum and the way in which they are explicitly and implicitly addressed in the school's culture -is insufficiently represented in Croatian empirical research (Pađen & Huić, 2021). The few empirical studies that do exist describe the heteronormativity and hostility of the Croatian social context in general and the education system in particular. Applying Meyer's model of minority stress to the Croatian context, Kamenov et al. (2015) summarise the empirical studies conducted to date related to the position of homosexual and bisexual people in Croatia and conclude that, despite a favourable legal framework, the propagated tolerance of the Catholic Church, the official comprehension of homosexuality as a variant of sexual orientation and the absence of extremely negative public attitudes towards the LGB community, the daily existence of LGB people is not entirely positive: " [t] he legal framework is not fully implemented in practice, the Church continues to condemn and does not condone the enjoyment of sexual relations, LGB people do not feel free to seek professional help and continue to experience violence and discrimination due to their sexual orientation (Bosnić et al., Jelić, 2013;Jugović & Ančić, 2013;Milković, 2013)" (Kamenov et al., 2015, p. 28). Using predominantly EU LGBT Survey data from 2012, supplemented by domestic empirical research, Vučković Juroš (2015) provides an overview of the dimensions of the social exclusion of sexual minorities in Croatia, and, following Šućur (2004, quoted in Vučković Juroš, 2015, analyses their position within a four-dimensional context: civic, economic, social and interpersonal. For the context of the present article, we are particularly interested in the social dimension, within which the author analyses the heteronormativity of the education and health systems, referring to studies that suggest neglect or inadequate treatment of LGBT topics in textbooks and the controversy over the introduction of the Health Education Curriculum (Brumen, 2012;Juras, 2013, both quoted in Vučkovic Juroš, 2015. It is in public affairs, i.e., in the polarisation of the public around the Health Education Curriculum, that the persistence of taboos on sexuality in society/education, as well as the closely related negative attitudes towards sexual and gender diversity, can be detected. Although the Curriculum itself consists of four modules, the point of contention in the discussions surrounding it was only the fourth module, Gender Equality and Responsible Sexual Behaviour, 3 whose content was perceived by a part of the public as an attempt at hypersexualisation and gender indoctrination of children (see Bekić, 2013). This resulted in the whole project being relatively unsuccessful. The most recent attempt to introduce the topic of sexuality into the Croatian education system is the adoption of the cross-curricular theme Health in 2019 (Narodne novine, 2019). In relation to the aforementioned Health Education Curriculum, this represents a step backwards: apart from intensely represented and, from a pedagogical perspective, highly reductionist medical discourse of sexuality, LGBT topics are completely omitted (Đaković & Novosel, 2020;Tomac, 2021), which for (LGBT) students is likely to have various negative consequences. 4 Reports on the experiences of Croatian LGBT students indicate that "[t] he consequences of the absence of such content can also be seen in the problems that LGBT students face during education. According to a 2012 EU LGBT study, 31% of Croatian respondents often and always experienced negative comments or negative behaviour due to their sexual orientation or gender identity at the time of their schooling when they were under the age of 18. Likewise, 24% of respondents felt discriminated against by school or university employees in the past 12 months" (Vučković Juroš, 2015, p. 209). Based on available empirical data (Hodžić & Bijelić, 2012;Milković, 2013;Kovačić & Horvat, 2016; data from Zagreb Pride and CESI; all quoted in Jurčić, 2018), Jurčić (2018) considers schools to be unsafe places for LGBT students due to the presence of various forms of abuse and discrimination of sexual and gender minorities. The fact that educators also have a role in the production of this lack of safety is shown by the recent report Human Rights in Croatia: Review of the Situation for 2019, which addresses the problem of peer homophobic violence and the lack of adequate psychosocial support for LGBT students in schools; specifically, their fear of teachers' negative reactions in the case of reporting violence and consequently possible outing to parents without the student's consent (Đaković & Novosel, 2020). The previously described resistance to the introduction of sexuality education, that is, to the inclusion of LGBT issues in the education system, predominantly relied on the argument that parents possess a constitutional right to the value consistency of content to which students are exposed at home and at school (Bekić, 2013). In view of the considerable media presence of such discourse of 'parental rights' , this may have contributed to the impression of overall parental resistance to any form of sexuality education for students. However, the results of research conducted on the basis of data collected in 2013 and 2014 on parents' attitudes to the introduction of the Health Education Curriculum, which the authors supplement with an overview of related research conducted in Croatia (CRO Demoskop, 2013;Janković et al. 2013;Turčin, 2013;all quoted in Kuštreba et al., 2015), show that approximately three quarters of the surveyed parents actually support the curriculum. Although such results might initially seem surprising, a similar discrepancy between intense parental resistance to sexuality education represented in public and the lack of empirical evidence to confirm the massive scale of this resistance has been observed in various contexts (Depauli & Plaute, 2018;Peter et al., 2015;Robinson et al., 2017;Ullman & Ferfolja, 2016), although it should be noted that support for sexuality education is higher at the lower levels of the education system, which generally do not include explicit discussions of sexuality or sexual and gender diversity (Kuštreba et al., 2015;Ullman & Ferfolja, 2016). In view of all of the above, and in particular given the insight into whether parents differentiate between supporting content related to sexuality in general and content related to sexual and gender diversity specifically, we support Ullman and Ferfolja's (2016) suggestion regarding the need to conduct explicitly focused research on LGBT inclusiveness on nationally representative samples of parents before making educational-political decisions and intervening in the curriculum. However, for an in-depth understanding of the educators' perspective and the complexity of the need to find a balance between their personal and professional positioning on a daily basis, we consider qualitative studies conducted with educators equally significant. Educators possess different levels of competence and act in different micro-contexts. Moreover, they have diverse understandings of the rights of (LGBT) students and the various parental groups with which they collaborate. The aim of the present research is to analyse the different discourses that educators use when discussing appropriate modalities (content, methodical approach and timing) of addressing LGBT topics in education. More specifically, we wanted to address the following research questions: 1.
Are LGBT topics presented in the participants' everyday pedagogical practice? 2.
Which arguments dominate the participants' discourses when explaining the representation of LGBT topics in their everyday pedagogical practice? 3. Which period of the child's education do the participants recognise as optimal for addressing LGBT topics?
It is evident from the research questions that we approached the research topic without distinguishing the identity variations encompassed by the acronym LGBT, respecting the strong intertwining of sexuality and gender, i.e., sexual and gender diversity, which is why the aim was not broken down into specific research questions dealing with the representation and perception of the identity groups associated with the acronym (e.g., the representation of transgender topics in the curriculum). Such an approach has been the subject of elaborate criticism aimed at its role in perpetuating the invisibility of transgender students in the educational process (see Dugan et al., 2012;Greytak et al., 2009;Meyer, 2022). However, our research was carried out in a research context in which the empirical approach to sexual and gender diversity is in its infancy, so we consider this approach a justified methodological choice.

Method
The data presented in the present article were collected as part of the scientific research project LGBT (In)Visibility in School: The Educators' Perspective, which employed a focus group procedure. Four focus groups were conducted with a total of 27 staff members from secondary schools in Zagreb. 5 The four schools included in the research responded to an invitation to participate in the research sent to all secondary schools in the capital in three rounds. Due to financial limitations and the assumption that schools in smaller communities could be even more reluctant to participate in the research, we decided to conduct it in the capital city, where all three researchers work. The collected data were analysed using thematic analysis (Braun & Clarke, 2006), differentiating a total of five themes. 6 In this article, we present a theme that covers the polarity of the premature or belated addressing of sexual and gender diversity in education with 5 "In October and November 2019, we conducted four focus groups with secondary education staff (teachers, principals, librarians) from four secondary schools in the City of Zagreb, which included a total of 27 participants (6 in the first, 6 in the second, 7 in the third and 8 in the fourth focus group). This was a convenient sample, formed in secondary schools which accepted our invitation to participate in the research. We moderated the focus groups ourselves (in pairs or in all together) according to a protocol developed for the purposes of the study. The shortest focus group lasted about 60 minutes, while the longest went on for about 90 minutes, depending on the time available to the participants in the study, to which we adjusted the study protocol and the range of topics that were addressed in individual focus groups. " (Bartulović et al., 2023, 6) 6 "All focus groups were fully transcribed by students of pedagogy, fully acquainted with the ethical aspects of dealing with recordings and transcripts. The transcripts were then processed using thematic analysis (Braun & Clarke 2006), in the following sequence: familiarizing with data, generating initial codes, searching for themes, reviewing themes, defining and naming themes and producing the report (for more details on thematic analysis, see Braun &Clarke, 2006, andBraun et al., 2019). Initially, each of us carried out the first two phases independently, after which we agreed upon codes on which to perform further data processing. The search for topics, their revision, definition, and naming, as well as the preparation of study reports, is the result of joint work by all three authors. Finally, we generated a total of five thematic units: discourse of 'professionality' vs. teacher responsibility; micro-context vs. macro-context of school; homeroom teacher as a homophobe or a haven; teacher as a technician or an intellectual; and taboo themes -sooner or later. (…) All phases of the study were conducted in accordance with the ethical principles of conducting research in the social sciences, taking into account the prompt obtaining of permits, voluntary participation, securing participants' informed consent, deleting the recordings after transcription and storing the anonymized transcripts separately from the research consents. " (Bartulović et al., 2023, pp. 6-7) regard to the chronological and developmental age of the child/student. A detailed description of the methodology is available in Bartulović et al. (2023). We want to emphasise the fact that, due to the specific subject areas that some of our participants teach, providing information about them could jeopardise their (and their schools') anonymity, possibly resulting in negative personal and professional consequences. In the interest of adhering to high ethical standards, we therefore decided to exclude demographic data on the participants. We consider this decision appropriate given that the emphasis in the present paper is on the detection of different lines of argumentation that appeared in group discussions, and not on the analysis of how each identified line of argumentation is connected to certain identity markers of the participants.

Results and discussion
We have dealt with the taboo of sexuality in educational institutions before (Bartulović & Kušević, 2020), having analysed it within the context of "the prevailing discourse of sexuality, characterized by the developmental line of nonsexual gender conformed child -unstable hypersexual sometimes gender nonconformed adolescent -stable and discrete gender conformed heterosexual adult" (Bartulović & Kušević, 2020, p. 133), whereby we discussed the importance of parental resistance to explicit LGBT inclusiveness at the lower levels of the formal education system, as well as structural barriers in the form of insufficient teacher preparedness to address a topic that is recognised as the exclusive domain of parental intervention (Bartulović & Kušević, 2020). Continuing with the line of thinking about sexuality within the education system that was established in the previous study, in the present article we analyse three arguments that repeatedly appear within the focus groups as explanations for the persistence of LGBT taboos in education: 'child innocence' , 'parental resistance' and the lack of a systematic approach to LGBT inclusivity in education. We see these subthemes as part of one main theme focused on the optimal timing for addressing LGBT topics in education: each of the subthemes thoroughly describes one line of resistance/constraints to the introduction of LGBT topics in education, which is why the optimal timing for addressing them becomes never, thus enabling the persistence of LGBT taboos in the education system.

The argument of 'child innocence'
The argument of 'child innocence' often appears in studies that place sexuality in an educational context, as a lens for understanding the complex relationship between social attitudes about sexuality and the way in which it is (or is not) institutionally mediated. Different authors note that the predominantly represented approach to sexuality education stems from corresponding social discourses that view sexuality in the context of childhood as unimportant, developmentally inappropriate and threatening to children (Renold, 2005;Davies & Robinson, 2010;Egan & Hawkes, 2010;Egan, 2013;Robinson, 2013; all quoted in Robinson et al., 2017). A qualitative study conducted with Australian parents, which sought to explore their understanding of sexuality education at home and at school (Dyson & Smith, 2012), found that some parents, in an effort to protect their child's 'innocence' , delay talking to their children about sexuality until puberty, which is recognised in the literature as the problem of the myth of 'the sex talk' (Walker, 2004, p. 246). The described problem of perpetuating the taboo of sexuality is also recognised by the participants in each of the focus groups conducted:

It's a taboo, 7 it still is… (…) But we're all going to pretend that's all… it's all OK. But we don't really have to talk about it.
We all pretend that they are not… as if this is not happening, as if there is no sex life, as if they really are on their wedding night, when they swear at the altar that they will be together as long as they both shall live, that before that there is nothing, but there is, although… This taboo status was confirmed in each of the four conducted focus groups. Although the research participants were critical of this taboo, as can be seen from the quotations above, the data reveal a lack of consensus on whether the taboo should be destabilised as soon as possible or as late as possible. The participants believe that the beginning of sexuality education, including education on sexual and gender diversity, is delayed in secondary school. They identify a possible correction of this in the introduction of comprehensive sexuality education from an early age: Although the positive effects of a comprehensive approach to sexuality education that teachers have consistently advocated are abundantly documented in the scientific literature (Depauli & Plaute, 2018), it would be interesting, given that we are dealing with secondary school educators, to see whether they would be advocated by the educators at lower levels of the system. We asked ourselves this question having noticed an interesting paradox: although the teachers criticised the taboo nature of the topic, which prevents them from talking about it openly with students and advocated the necessity of addressing it as early as in the preschool context, in some of the focus groups a secondary school student was, without any discussion, presented as a highly sensitive being with whom sexual and gender differences should be discussed with caution: Where it's actually somehow made known that he [a writer discussed in class] and his companion weren't just friends, that they were more than that. Now, do we go into it further than that? We don't. Why? They are children. Some things very often, when you let them go, end up going in a significantly different direction. Third grade, for example, is much better for me.
It's still a major, major taboo in society, and especially for children at that... I mean, they are very sensitive at that age...
(…) they deal with it at the age of fifteen or sixteen and they're not ready to face such things at all because it creates an aversion for them. Because at that age they're not ready to talk about such things through work at all. Shannon and Smith (2017) see this as a construction of an oppositional narrative based on pairing the image of an innocent child with a threatening concept of sexual and gender diversity.
We consider the understanding of students described in the quotes above by some focus group participants to be significant from a pedagogical perspective, as each image of the child is productive: we position ourselves in a pedagogical relationship according to how we see the child and how we assess his/her educational capacities (Bašić, 2011). When it comes to the topic of sexuality, the described tendency to hyperinfantilise students and postpone formal sexuality education due to their alleged unpreparedness and 'sensitivity' leads to a situation in which too early suddenly becomes too late, that is, in which the daily life of secondary school students is not accompanied by adequate educational support:

So… We don't even know what homophobia is. It's the fear of people…
In the third grade of secondary school… Therefore… I touch on such things here, but do I have something prescribed in the curriculum in particular? No.
Then you tell them why gay pride is even held at all, which is also something I have to explain.
In all of the focus groups, parents were presented as particularly ardent guardians of the image of the 'innocent' child. In the analysis of parental resistance to LGBT-inclusive teaching, however, we also recognise certain polarities.

The argument of 'parental resistance'
The participants in all of the focus groups agree that parental influence on how a child's attitude about sexual and gender diversity will be shaped is crucial: Well, I' d say, I've been employed in education for six years now, so I' d say that in general the most important factor, not just about the LGBT issues but on the level of the whole society, actually the biggest pressure and the biggest responsibility is on the parents. When you talk to a child in, say, the second or third grade of secondary school and they expresses some of their attitudes, then their parent comes to you and it becomes clear who shaped those attitudes. Or the parent doesn't come to you, or they never show up, you never meet them, and then you understand why the child is in such a direction. So I think that parents have the greatest influence on the child, especially in early childhood, perhaps so that… that the first question of empathy arises here. Therefore, someone who was raised to develop empathy will certainly not condemn members of the LGBT community, if they were raised in that way from the very beginning.
Recognising this crucial importance of parental influence, it is not surprising that the participants' statements are saturated with the argument of parental resistance, which was present in each of the focus groups: Some parents would even come to us… (…) …and say: "My child won't listen to this because it's not something my child does…", there.

Yes, but again, if there are parents… I mean, in the end it all comes down to the parents. If someone shows up again who disagrees and applies pressure and then it ends up in the media and then it's all finished. I mean, there's always things like that, you know: "A scandal in a reputable institution", or something of the sort.
This argument is very common in the literature as well (Flores, 2014;Eisenberg et al., 2012;Reis & Seidl, 1989;UNESCO, 2009;all quoted in Peter et al., 2015;Martino & Cumming-Potvin, 2016;Duffy et al., 2013, Milton, 2010, both quoted in Ullman & Ferfolja, 2015. From the findings of our research, however, it cannot be concluded that parental reactions were on a large scale or strong. This corresponds to the findings of authors who speak of a vocal minority (Eisenberg et al., 2012;Horn et al., 2013), that is, on the three-parent syndrome (Martino & Pallotta-Chiarolli, 2003;quoted in Ferfolja & Ullman, 2017, p. 350), which warns that teachers may perceive parental resistance as stronger or more large scale than it actually is. This is understandable because it carries a risk for both the teacher individually and the institution as a whole, as can be seen from the last quote above. In our research, however, a teacher who integrates topics of sexual and gender diversity into her teaching offered the opposite perspective: I've never had parents report that I did something… maybe they have, but I've not heard about it… that I did something wrong or that I hurt anyone and so on (...).
Empirical studies addressing parental attitudes about sexuality education in general, and education about sexual and gender diversity in particular, in very different social contexts suggest that the majority of parents want sexuality education of high quality for their children (Ferfolja & Ullman, 2017;Peter et al., 2015;Robinson et al., 2017), although some expressed fear of too much information, that is, the need to protect the aforementioned 'innocence' of the child (Dyson & Smith, 2012). As noted in the Introduction, the topic of parental attitudes towards sexuality education has been an area of study in the Republic of Croatia as well (Kuštreba et al., 2015). In a hitherto unmentioned study, a sample of 1,673 parents of primary school sixth-graders were surveyed on their attitudes towards sexuality education and the factors associated with those attitudes. The results show that most parents consider sexuality education important and necessary, including the topic of homosexuality, which, at the level of the upper grades of primary school, is considered important by 57.9% of surveyed mothers and 59.1% of surveyed fathers (Janković, 2011).
Although parental support, as we shall discuss in the Conclusion, is important for the implementation of this part of the curriculum, Kuštreba et al. (2015) stress the importance of focusing on the students themselves, that is, on the right of young people to relevant, scientifically based information on sexuality (Breaken & Cardinal, 2008, as quoted in Kuštreba et al., 2015). This brings us to the question of the relevance of parental resistance in thinking about the right of young people to sexuality education. At no time did the participants in our focus groups invoke the parental right to decide whether a student should be exposed to topics of sexual and gender diversity in school; in fact, as will be shown in the following section, the dominant position was that there is a need for systematic access to these topics at the lowest levels of the education system. In one of the focus groups, it was even stated that parents, in order to better support the school's endeavours, should be forced into such education. However, studies often consider the valid legal right of parents to exclude their children from classes involving mediating content about sexuality that the parents deem inappropriate (Bialystok, 2018;Robinson et al., 2017;Ullman & Ferfolja, 2016). This option is based on the perception of parents as the exclusive, or at least primary, holders of the right to sexual education of their 'own' children. The questioning of this right is discussed in the works of some Croatian authors (Jakovac-Lozić, 2020;Rukavina Kovačević, 2014), which emphasise the validity of parental rights in various Croatian and international documents 8 (where, we believe, a more extensive elaboration of the relationship between parental rights and the rights of the child is lacking, as is a more elaborate discussion on the complex relationship between the legislative and pedagogical perspectives on this issue).
However, certain authors question the existence of something that could be called the fundamental rights of parents (Bialystok, 2018;Howe, 2001;Montague, 2000;Westman, 1999), pointing out: "My position is that it is children who have fundamental rights and it is parents who have duties and responsibilities to provide for the rights and best interests of their children. While we may assume that parents have certain rights, they do not have fundamental rights qua parents. The rights they do have are delegated to them by society and are dependent on the fulfilment of their obligations to their children" (Howe, 2001, pp. 61-62). In their discussion of students' rights to sexuality education, Ferfolja and Ullman (2017) focus precisely on the child as an individual being and his/her right to education about human sexuality (Clarke, 2011, as quoted in Ferfolja & Ullman, 2017), which they do not perceive as a question of value, but as part of the right to education. The right of children to sexual education is indirectly manifested in the relational character of sexuality, that is, in the fact that it is, among other things, lived through peer relationships. Bialystok (2018) therefore observes that parents cannot employ their own value system to decide on the type of content, as the absence of this content from the formal curriculum could have detrimental consequences for other people's children as well. Nothing that has been mentioned so far leads to the marginalisation of the parental position in the process of implementing education for sexual and gender diversity; quite the contrary, especially bearing in mind earlier results showing that the more informed parents are about sexuality education, the more they generally want and support its implementation for their children (Kuštreba et al., 2015). This points to the necessity of finding an empirically articulated optimal strategy for the collaborative involvement of parents in the process (Bialystok, 2018;Eisenberg et al., 2012;Eisenberg et al., 2008), which would recognise them as the most committed advocates of their children's interests, but not as the holders of parental rights independent of the children.

The argument of the 'top-down solution'
The paradoxical moments identified in the first two subthemes (the preschool child as a being who should be introduced to sexual education vs. the secondary school student as a particularly sensitive being with whom certain topics should not be addressed; real vs. assumed resistance of parents to the introduction of LGBT topics in education) also pervade the third subtheme, which shows the participants' recognition of their own responsibility for the implementation of LGBT themes, but with a simultaneous focus on solutions from above, i.e., the decision of the ministry. Such a decision would, after all, solve the problems related to polarities identified in the first two subthemes. The three lines of argumentation of the subthemes in their synergy describe the subtle mechanisms by which the LGBT taboo in education is maintained.
Concluding that LGBT topics are not visible at all in the official curriculum and the materials they use, as well as that broaching such topics would depend solely on the will and creativity of each individual educator (recognising that some subjects and some types of school culture are more suitable for this), the participants in all of the focus groups emphasised the importance of political will in the process of formalising LGBT content in the curriculum as early as possible, that is, the responsibility of the institutions in charge to provide a formal framework for teaching about sexual and gender diversity: Because it all starts at a slightly higher functional level, right, the level of state politics, in order to get to such things at all, because that's the thing, everyone should change these things from the top.
The participants view the presence of LGBT content in the curriculum as a form of removing the taboo status of the topic: Through, yes, very, very small steps. (…) You know when you're actually more or less unaware of all the commercials you come across during the day? In fact, you're actually unconsciously absorbing something. That's how a society as a society will unconsciously absorb the fact that a certain question is normally asked, that it's legal in a textbook: "What do you think about this and this…?", that this will be equal to such approaches. This unconsciously, simply unconsciously moves in small steps towards some surely better position in all this.
This serves as a form of protecting the educator, who would be given the legitimacy to broach LGBT topics in their teaching through an official curriculum that explicitly addressed such topics, but also as a form of corrective for the attitudes of educators whose personal position is not LGBT inclusive: As soon as such an issue is in one of the textbooks, then it's -aha, this is a textbook that has been prescribed, that is here for us, that the government has sanctioned, or the ministry (…) so I'm talking about some kind of authority that needs to approve something in order for it to be in the textbook, in the schoolbook that is prescribed to them there. Now, would a teacher, who agrees or disagrees with it… she should give a lecture within a specific situation that explains that it's the same as a different religion, different skin colour, that too is the same. We need to simply get to that level as a society. There, that's what I think. And that can certainly be done in small steps. Because when it's imposed on someone, when it's in someone's curriculum, that is, when it's part of someone's teaching plan, then this somehow either equalises things, positions them in a more tolerant way, and that's it.
Emphasising the importance of including LGBT content in textbooks is in line with the results of a survey of teachers' and parents' attitudes towards sexuality education in Austrian primary schools, where teachers specifically stressed high-quality didactic materials that can support teaching about topics they perceive as challenging as the factor with the strongest influence on the successful implementation of sexuality education (Depauli & Plaute, 2018). In the same study, parents stressed the quality of teachers' preparedness to teach about sexuality as the strongest factor in the success of sexuality education (Depauli & Plaute, 2018). In focus groups conducted in different contexts in Australia, parents also stressed the importance of institutional and managerial support, that is, the unacceptability of leaving the decision on whether to include LGBT content in teaching to the individual teacher (Ullman & Ferfolja, 2016). In addition to the support shown by umbrella institutions, Flores (2014) and Payne and Smith (2018) emphasise the important role of school leadership, which can function as a gatekeeper to its employees' professional development through its misunderstanding of the relationship between LGBT stigma and academic success, believing that such a form of professional development is unnecessary in their context, that it would encounter various forms of resistance, or that the teachers would not even be interested in it (Payne & Smith, 2018). Eisenberg et al. (2012) place such support from principals at the institutional level, using an ecological model in their analysis of sources of support and barriers to sexuality education. Furthermore, in all reform efforts to increase the level of LGBT inclusion of schools, it is also necessary to consider the interpersonal level, which primarily refers to the potential of parental support, i.e., resistance, which was discussed in the previous section, as well as the social level, which addresses the issue of educational policies and broader societal attitudes to LGBT issues (Eisenberg et al., 2012). Hence, the responsibility for shaping the LGBT inclusive culture (school) cannot be located within one actor. This was also recognised by some of the participants in our focus groups: So I think, in principle, that it should be everyone, since everyone is [responsible] for children… The whole society. (…) So, starting with the teacher, the professional service, the textbook, the teaching plan.
If we all have the well-being of the students in mind, then we all need to have that in mind as well. I think the responsibility lies with everyone.
Such an approach requires teachers to assume an engaged position, which, in exceptional situations, may even be more progressive than any centralised policy. In a study by Eisenberg et al. (2012), some teachers, when asked about the need for legal regulation of sexuality education, stated that they oppose such an initiative, believing that it jeopardises the progressive practices that engaged teachers, guided by their intrinsic motivation and the principle of justice, already implement in their schools (Eisenberg et al., 2012). Along these specific lines, we consider particularly productive the position expressed in one of the focus groups, which calls into question the hierarchical relationship between the ministry in charge and the school in situations where ministry support to raise the level of inclusive education is inadequate: I mean, it should all actually start from the umbrella institution, it should start from the ministry, and I even somehow think that it's our duty as educators, as people involved in the educational process, that if we see omissions coming from the ministry, then somehow, from that human side, we need to actually point out these things if we see that they are missing from the curriculum, from the teaching plan and so on, we nonetheless, in a way that is close to children, acceptable to children, that does not exclude, that, I mean, works hard on acceptance, tolerance, as well as understanding, I think that it somehow must be our duty to include this as well.
Such a position recognises that waiting for a solution from above can be an alibi strategy for avoiding pedagogical responsibility, which we consider extremely important in contexts such as Croatia, where, as is clear from the data presented in the introductory section, such interventions from above are unlikely to occur in the near future. Discussing the optimal way to implement reforms in education, Fullan (1994) points out that systems do not change by waiting for change at a certain level of the system, but by the activity of the individuals who make up the system, 9 by their forming alliances and synergistically insisting on change (Fullan, in print, as quoted in Fullan, 1994). The more interconnected and harmonised the different levels are, the more certain achieving change becomes, which is why the synergy of top-down and bottomup approaches is considered optimal (Fullan, 1994).

Conclusion
In this article, we have presented three arguments that strongly moderate the performance of our focus group participants in the domain of LGBTinclusive teaching, making them susceptible to maintaining the status quo, more so than is perhaps required by the circumstances described in this study. Regarding the argument of children's innocence and the top-down solution, the participants expressed different opinions about the optimal modalities for overcoming LGBT taboos in education. Considering that the research was carried out in secondary schools, it would be interesting to investigate the attitudes of educational workers at the lower levels of the education system on the optimal timing for the introduction of systematic sexual education. The views of early childhood educators, professionals who work with children at an age when they are perceived as asexual in conservative discourse, seem particularly significant in this regard. Their views would contribute to the understanding of the issues discussed in this paper. This is particularly true due to the different construction of the autonomy of education workers at different levels of the Croatian education system, whereby the system of early childhood education is strongly characterised by the positioning of the educator as a reflexive practitioner who is minimally reliant on top-down instructions. Their views on the second subtheme, the role of parental resistance in preserving LGBT taboos in education, could also be valuable due to their daily interactions with parents.
The data show that in order to overcome the taboos of sexual and gender diversity in education, it is especially important to address the issue of the relationship between parents and school. Some of the possibilities for collaboration with parents in this process described in the literature include workshops for parents, organising debates, school-parent dialogue with relevant institutions and 9 An example of such teacher engagement in a school context that is not yet LGBT inclusive is described by Martino and Cumming-Potvin (2016) in a case study of a Canadian teacher, who, similar to the attitude of one focus group participant in our research described above on the importance of letting the LGBT topics in 'through small pores' , chose a strategy of educational moments made up of consciously using every recognised opportunity to include LGBT topics in teaching, most often encouraged by student interest. The function of this approach was to normalise LGBT topics, but also to protect the teacher from excessive parental supervision and potential professional sanctions.
NGOs, parent participation in curriculum evaluation and revision, and building a trust-based school culture between teachers and parents (Bialystok, 2018;Eisenberg et al., 2008). In our focus groups, none of these forms of initiating parent-school collaboration was mentioned, which we see as a potent field for further research and practical interventions based on such research. Although we recognise the complexity of all of the challenges that may arise, which result from attempting to find a balance between recognising parents as the child's primary caregivers and the needs and rights of children/students or society as a whole, we believe that parental resistance should never call into question LGBT inclusiveness in education. Our reasoning follows the arguments of Bialystok (2018, p. 22), who considers making parental homophobia part of public education unacceptable, because "views that contradict liberal equality should not translate into public policy and are not deserving of special accommodation. In our context, it likewise seems important to reach an expert consensus as soon as possible on whether parents should be able to influence parts of the curriculum concerning the rights of their own children, and always also of other people's children, so that by perpetuating discussions about the acceptability of LGBT-inclusive education we would not also perpetuate sexual and gender diversity as controversial topics (Ferfolja & Ullman, 2017), especially in a situation where, as described in the present article, the silent majority of parents do not share such an attitude.
As the authors of this article, we share the view that parts of the curriculum corresponding to the principles of equality and non-discrimination, which are woven into the foundations of our society, should not be the subject of parental questioning. This does not mean that collaboration with parents in the implementation of this curriculum is unimportant (the importance of a high standard of collaboration between the family and school in a multilevel approach to the implementation of sexuality education has been empirically confirmed by, for example, Eisenberg et al., 2012). However, in a context where parental influence is necessarily limited, insisting on collaboration leads to the danger of reducing it to a kind of coerced and at least illusory parental support for decisions already made by the school, which we consider to be an inappropriate positioning of family-school relations. The relationship between family and school in such situations should therefore be considered carefully: the school should take responsibility for shaping its LGBT-inclusive culture, but it should also take responsibility for dialogically addressing mutual fears, dissatisfactions, risks and possible courses of action. These responsibilities are the domain of both parents and educators, and it is the interconnectedness of these two groups, configured in various ways, that creates micro-contexts with specific needs and corresponding modalities of possible courses of action. It seems particularly important to keep in mind that the dyadic nature of this relationship is actually illusory, and that the primary subjects of LGBT-inclusive education are actually students, although the importance of their attitudes, interests and needs is rarely mentioned even in scientific sources (Shannon & Smith, 2017). The empirical coverage of their perspective, for which, according to established standards of research in the social sciences, it is generally necessary to obtain parents' active consent, brings us back to the necessity of a systematic approach to removing the taboo from the topic of sexual and gender diversity in education.
The conducted study has several limitations. As described in the paper that presents the second theme generated by the conducted research, we recognise its limitations as the insufficient heterogeneity of the focus group participants, their self-selection and the consequent supporting attitudes towards sexual and gender diversity; conducting the research exclusively in the capital city; and the extensiveness of the focus group protocol, which made it difficult to explore certain topics that opened up in the conversation more deeply (Bartulović at al., accepted for publication). We also view the lack of the conceptual demarcation of sexuality and gender addressed in the Introduction as a further limitation.