Regimes of patriarchy and faith: reflections on challenges in interviewing women and religious minorities in Pakistan

Purpose – Thispaperdiscussesthechallengesthattwodoctoralresearchersfacedwhileresearchingreligious minorities and women in a culturally sensitive society such as Pakistan. Their shared interest in sensitive topicsrelatedtogenderandminoritiesinPakistanledbothresearcherstocollaborateinthisstudytoprovideabetterunderstandingofissuesinqualitativeresearchinthesameresearchcontext.Theydiscussthechallengesofinterviewingparticipantswithintheeducationalcontext.Theyalsosuggestsomewaystoovercomesuch challenges. Design/methodology/approach – Based on Foucualt ’ s writingson regimes oftruth, discourse and systems ofexclusion,theauthorsinthisstudyanalyzehowpatriarchalandfaith-basedregimesoftruthconstrainsome discoursesthataffectparticipants ’ willingness and insights to reflect on the issues freely. Findings – While reflectingon their experiencesin data collection, authors report that qualitative researchers struggle to access participants to investigate issues related to gender subjectivities and minority faiths in educational contexts in developing societies like Pakistan. Researchers face a variety of problems, from their own positionality to participants ’ access to their responses. The reason for this is patriarchal and religious regimes and also their intersecting relations that restrict participants ’ ability to reflect on their issues. Minorities in Pakistan are often prevented from expressing their views freely by blasphemy fears. The discourses of gender are also sensitive. Therefore, the study suggests that in societies such as Pakistan, where religion and gender are emotive terms, the problem can be handled by counter-discourses that challenge truth regimes by conceiving research as a transformative practice. Moreover, such societies require a policy for protecting researchers and participants in the interest of knowledge production and dissemination. Originality/value – This study is originally based on the primary data used in two doctoral studies.


Introduction
Researching sensitive topics in education is often a challenge for researchers, whether they are related to religion (Stern, 2020) or gender (Lenz, 2010). A context such as Pakistan, where faith-based and patriarchal regimes dominate academic and social spaces, makes it difficult for researchers and participants to engage in a research activity without fear. The reason for Researching religious minorities and women this is that several incidents have occurred in the country, including the stabbing of a college professor to death by his student for mixing male and female students in the class discussion (see Imran, 2019) and the life imprisonment of a Hindu teacher accused of blasphemy (see Shamsi, 2019) who was recently sentenced to life imprisonment in February 2022. Further, several incidents of religious minorities at the intersection of faith and gender, such as forced conversions of Hindu girls to Islam (Schaflechner, 2017), restricted religious freedom (Din and Jacob, 2019), violations of rights (Mehfooz, 2021) and Otherisation (Ali, 2015) have caused minority faith and gendered identity groups to observe silence in addition to many others with privileged status in Pakistani society for several reasons. In the backdrop of this socially sensitive environment, the authors of this paper in their doctoral studies aimed to understand how school textbooks that teach English as a foreign language in Pakistan construct discourses of otherness (Study 1) and how the discourses of gendered identity of women academics in Pakistan are constructed as well as their participation in leadership roles in higher education in Pakistan (Study 2). Two doctoral studies shared similar methodological and theoretical foundations, e.g. critical and post-structural discourse theories, and semi-structured interviews as a data collection tool for examining identity in educational contexts, such as schools and universities in Pakistan. In the present paper, we discuss challenges that we encountered while interviewing our participants. We discuss these challenges based on Foucault's writings (see Foucault, 1975Foucault, , 1976Foucault, , 1978 on regimes of truth, exclusionary systems and discourse. In doing so, we first elaborate Foucauldian theoretical lens adopted in the present study followed by a brief account of our own studies, research ethical guidelines we followed, information about the participants, context, sampling and interview procedure adopted in our respective doctoral studies. Lastly, we present the challenges we faced in the entire process of interview in Pakistan. We conclude our study by proposing some suggestions to overcome these challenges.

Theoretical underpinnings
For Foucault (1978, p. 92), power is the multiplicity of force relations that exist within the various social, economic and political spheres in which it operates. Within a certain power structure, these power relations exist in a strategic form, where individuals and groups are in constant conflict. It is sometimes difficult to suppress states of dominance, while in other cases resistance can effectively confront these power structures (see Rabinow, 1994, p. 299). Foucault uses the term "regime of truth" in several writings (for example, Foucault, 1975Foucault, , 1976Foucault, , 1978 to refer to how some truth games are imposed in relation to power. In his interview published in 1976 on the political function of the intellectual, Foucault contends that "truth isn't outside power or deprived of power. On the contrary, truth is produced by virtue of multiple constraints, and it induces the regulated effects of power." In other words, each society has its own regime of truth (Foucault, 1976, p. 13). As Lorenzini (2015) explains, if we understand this term as Foucault does, politics has a relationship not only with institutions but also with the complexity and constitutive field of power relations we inhabit, and truth as such reinforces and induces the power effects we observe. The regime of truth therefore serves as a strategic field within which truth is produced and becomes an element of a number of power relationships. In addition, Weir (2008, p. 368) mentions that Foucault sketched several criteria for truth regimes: how true and false statements are distinguishable; how true and false statements are sanctioned; and the status that truth-tellers are granted.
It is worth noting that discourse operates according to a particular regime of truth in a given society. Mills (1997) writes that discourse does not exist in a vacuum. In a particular context, it explains how ideas, opinions, concepts, ways of thinking and behaviors are framed or structured discursively. In Foucauldian terms, these patterns of thinking and behaving bear the effects of power regulated through truth regimes. As a result, these truth regimes are governed by discursive rules that restrict human activity and thought in specific ways. It is these discursive rules and categories that constitute discourse, and, therefore, knowledge, that Foucault is interested in (see Foucault, 1975). These discursive rules and categories limit human thought to the point that they are unable to think outside of them. The idea of thinking outside of them is considered insane and unreasonable. This is how discursive rules are connected to the exercise of power which reproduces discourse and social systems through forms of selection, exclusion and dominance. In Foucault's (1971) view, power constrains discourse by using systems of exclusion. A few examples of these systems of exclusion include prohibition, divide/rejection and truth versus falsehood distinction. Foucault further explains that these changing criteria for truth and rejection of discourses are based on institutional support, which includes pedagogy, books, publishing, libraries, learned societies and laboratories now in the scientific age (Foucault, 1971).
The Foucauldian regime of truth and systems of exclusion help our discussion in the present study to understand how institutionalized and culturally established regimes of truth in Pakistan constrain study participants' thinking and responses through prevailing discourses of patriarchy and dominant faith. In cases where there are intersecting power relations in our data, we also draw on "intersectionality" perspective (see Crenshaw, 1989) to explain how different social categories, e.g. gender and religion intersect as sites of marginalization in Pakistani society. Intersectionality perspective has proved to be very useful in several disciplines, such as history, sociology, political science, feministic studies, ethnicity studies and so forth in explaining marginalization at its deepest level. As Crenshaw (1989) argues that intersectionality exposes a single-axis thinking that undermines knowledge production and struggles for social justice. We thus use Foucault's notions to shed light on how discourses restrain socially underprivileged populations in socially diverse countries from participating in interview studies thereby leaving their "voices" completely unheard, or partially recorded combined with intersectionality perspective. According to Foucault, regimes of truth in a society entitle people who have authority to speak and what to speak. In this study, we highlight how patriarchal regimes of truth discourage researching gender-related topics and faith-based regimes of truth hinder researching about the issues of religious minorities. Positioning our understanding of Foucault's notion of regimes of truth in a society, we explain how researching gender and people of minority faith is challenging in Pakistan.
Doctoral studies: some contextual background Study 1 on minorities is in its middle phase of progress in Finland whereas Study 2 on gender is in its final stage of completion in the UK. The purpose of Study 1 is to examine the discursive and ideological construction of national identity and marginalization of the Other through English Language Teaching in Pakistan currently being carried out at a Finnish university. To do so, the author has analyzed the national curriculum for English language and English language textbooks (grades 1-12) to examine what dominant ideological messages are encoded, and how they contribute to the Otherisation and marginalization process in Pakistani society. Moreover, interviews with teachers were conducted in order to learn how they engage with textbooks and learners in relation to the multiplicity of their identities in Pakistan. Study 2 investigates the discourses of gendered identity of women academics and their participation in leadership role in higher education in Pakistan. The study focuses on how women academic leaders perceive and construct their leadership identity and how others around them perceive and construct their leadership identity. The study also focuses on unraveling dominant discourses in the construction of their leadership identity. Given their shared interest in sensitive topics related to gender and minorities in Researching religious minorities and women Pakistan, both researchers found it useful to collaborate in this study to provide a better understanding of the issues in conducting interviews in the same context. Our collaborative discussion has implications for wider global contexts, where researcher encounters challenges in qualitative research focusing on gender and religious minorities.

Researcher positionality
The researcher's positionality reveals their relationship with the participants. Chavez (2015) categorizes insider and outsider positionality, the former of which can be further categorized into "total insider" and "partial insider" based on whether the researchers share a number of identities or a few common ones with their participants. Accordingly, both doctoral studies in Finland and the United Kingdom focus on the same research context, namely Pakistan as a country of origin of the researchers. Therefore, they are not complete outsiders, but have assumed positionalities with shared identities, such as country of origin, common or familiar language, province (Sindh) and also cultural similarities. This partial insider positionality was therefore both helpful and impeding in terms of data collection, as further elaborated in our section on challenges.

Context, participants and sampling
The two studies share a common context, i.e. Sindh, which is the second most populous province in Pakistan with 47.85 million people (see Table 1 for population diversity in Sindh by gender and religion). Both doctoral studies, however, differ in their research setting. Study 1 focuses on English language teachers at school level, whereas Study 2 on gender focuses on women leaders in higher education, e.g. universities in Sindh province. Data were collected from ten cities in the Sindh province of Pakistan for Study 1. The teachers were selected from public and private schools who taught English language textbooks from grade 1 to 12. This study used purposive sampling to recruit 40 participants. Of these, six belonged to a minority group, e.g. Hindus, which is the focus of discussion in the current article. In Study 1 relating to minorities, the first author selected teachers who had at least three years of teaching experience in teaching English language textbooks as they were more capable of reflecting on textbook content and classroom environment as compared to novice teachers. The experience of teachers in the sample ranged from 3 to 22 years.
Study 2 relating to gender was conducted at one of the biggest and oldest universities in Sindh province, Pakistan. The university has over 65 teaching departments and many other chairs and research institutes. Out of 65 teaching departments, women held the leadership and administrative roles in only 20% of these departments. Women's participation in other administrative and decision-making bodies is even worse. The study used purposive sampling and interviewed five women leaders (working in different leadership positions, such as head of department, dean, member syndicate, etc.  academics were also interviewed who were working or had already worked under a woman head of the department. Besides, two focus group interviews were also conducted with male and female students separately. Interview procedure: some ethical considerations The interview procedure followed three stages (see Figure 1) starting from preparation of the interview protocol to ethical approval to accessing participants. In both studies, the ethical codes proposed by Sarangi (2015) were strictly adhered to during data collection and analysis. These include (1) ethics of access, (2) ethics of participation, (3) ethics of interpretation and (4) ethics of dissemination/intervention. In this article, we discuss the ethical protocols followed at the first two stages, namely access and participation.
To design the interview protocols for the participants, the authors followed the EU General Data Protection Regulation (GDPR). In accordance with these guidelines, which were subsequently approved by the supervisors of the current doctoral studies as well as the universities' research ethics committee, the interview protocol form contained the following information for the interviewees: (1) Information about the doctoral study, its title and objectives (2) Voluntariness and the rights of the research subjects (3) Protection of the personal data, e.g. data will be processed in Finland and the UK, respectively (4) Data Archiving, e.g. data will be disposed five years after the research has ended. The participants' responses will be archived anonymously in a suitable FAIR-complaint data repository for later research if the participants give permission to the researchers to do so.
(5) Publication of research results, e.g. articles, conference papers, workshops and dissertations (6) Rights of the research subjects, e.g. right to withdraw at any phase, right to check their responses used in the publication, etc.

Topic related interview questions and ethical considerations for participants
Ethical formalities e.g., signing consent form and familiariing with the research privacy notice  Furthermore, participants were also provided with a research privacy notice that assured them about the confidentiality and anonymity of their identities by informing them of the following rights (see Table 2). As shown in Table 2, the participants were informed to have the right to obtain information about whether their personal data is processed, and which personal data is processed. If required, they can request a copy of the personal data processed. If there are any inaccuracies or errors in the processing of their personal data, they have the right to request their personal data to be rectified or supplemented. In addition, they have the right to request their personal data to be erased in certain situations. However, the right to have data erased does not exist if the erasure prevents the purpose of processing from being fulfilled for scientific research purposes or makes it much more difficult. They were ensured to have the right to restrict the processing of their personal data in certain situations, such as if they deny the accuracy of their personal data. In both studies, interviews were recorded with the permission and consent of the participants for transcription and analysis. On average, the interview lasted between 30 and 35 min in Study 1 while in Study 2, interviews lasted between 45 and 90 min. In the present discussion, the responses of the participants have been reported with their consent.

Research challenges under faith-based and patriarchal regimes of truth
In Study 1 on religious minorities, a broader set of research questions are explored, including those relating to religious minorities, the subject of the present discussion. The study used a semi-structured interview to gather insights from English language teachers in Pakistan about English language textbooks in terms of their content focus. Of these participants, six were Hindu teachers, a minority religion in Pakistan. Study 2 on gender, on the other hand, examined the discourses surrounding leadership identity and the participation of female academics in higher education. The study explored the dominant discourses around gendered leadership identity of women academics in higher education in Pakistan. The following sections describe the challenges we encountered as young doctoral researchers during the interview process and how participants responded.
Researcher positionality and accessing participants Positionality in research is determined by where the researcher stands in relation to the participants (Greene, 2014). In Holmes's (2020) view, the social-historical-political location of a researcher influences the way they view social processes, that is, they are not separate from the processes they study. Our discussion draws on Chavez's (2015, p. 475) definition of insider positionality that contrasts with outsider positionality favored in the positivist tradition. According to Chavez (2015), insider positionality can be divided into two categories: partial Research subjects 1 Right to access data (Article 15, GDPR) 2 Right to have data rectified (Article 16, GDPR) 3 Right to have data erased (Article 17, GDRP) 4 Right to the restriction of processing (Article 18, GDPR) Source(s): GDPR, chapter 3. Rights of the data subject retrieved from https://gdpr-info.eu/chapter-3/ Table 2. Research subject rights and total. The term total insider positionality refers to researchers who share multiple identities (e.g. race, ethnicity and class) or profound experiences (e.g. family and membership) with participants, whereas partial insider positionality refers to sharing a single or a few identities with a degree of distance from the group. We as doctoral researchers understood our positionality and how it might affect and/or facilitate the data collection process regarding two sensitive topics, namely gender and faith, in two patriarchal and faithdominated regimes.
Study 1 on religious minorities positioned the researcher as a "partial insider" due to the fact that he studied religious minorities -Hindu and Christian teachersas Muslims. Researchers and participants shared several characteristics in common, including living in the same region (Sindh), speaking similar languages (Sindhi and Urdu) and having a common country of origin (Pakistan). However, faith differences (e.g. Muslim vs. Hindus/Muslim vs. Christians) created a gap between the researcher and the participants. Consequently, the participants perceived fear, largely of blasphemy, given the country's sociopolitical and religious context. As a result of partial insider positionality, accessing participants was challenging for the researcher in Study 1 because of perceived fears by participants. These fears emanate from what Foucault (1978) calls states of domination in a society operating through institutionalized discourses. In Pakistan, for example, these discourses dominantly pertain to specific version of Islam as an official religion that is manifested through various means, including education. This challenge of access to participants was realized several times during the data collection phase in Pakistan. This became more apparent when the researcher tried to contact the "Christian teachers" in his contact and provided them with information about the topic he was researching, its objectives and the type of questions likely to be asked during the interview process. Despite assurances of confidentiality and anonymity formulated in accordance with GDPR guidelines and approved by the university ethical committee in Finland, none of the Christian teachers that the researcher in the current study knew and contacted participated in the study. Sikes (2004, p. 15) argues that it is important for all researchers to think about how they are paradigmatically and philosophically positioned and to understand how their positioning and the assumptions they hold might influence how they approach their research. The goal here is to be a reflexive and reflective researcher who can present their findings and interpretations with confidence, knowing they have considered, acknowledged and been explicit about their stance and how it influenced their work. Being a Pakistani, the researcher understood their unwillingness to participate due to discrimination and severe consequences that the Christian Community in Pakistan has recently experienced. For instance, several incidents of attacking religious places, individual and collective lynchings of Christians, and home burnings (see Dawn, 2013;Dawn, 2014) reside in their memories. Consequently, the first study could not include any Christian participants, which could be possible due to the researcher's "total insider positionality". However, in the case of the Hindu teachers, specifically a female Hindu teacher, the "partial insider positionality" seemed to widen at the intersection of faith and gender (see Crenshaw, 1989). A female Hindu teacher was less comfortable with the researcher coming from a Muslim-dominant faith in Pakistan where there have already been several instances of Hindu girls being forced into Islam and married (see Schaflechner, 2017).
As opposed to "faith" as a differentiating factor, Study 2 on gender placed the researcher within a partial insider positionality based on gender, for example, a male researcher investigating female leadership roles. Our identities/positionalities are a part of the tool kit we carry with us to create theories and analyze data in research process (Reyes, 2020;Massoud, 2022). The researcher shared several common identities with the participants, including language (Sindhi/Urdu), ethnic and regional identity as Sindhi, and religious identity as a Muslim. The gender difference between males and females in the study led many participants Researching religious minorities and women to be curious about how a male researcher could be interested in studying female leadership roles at a public university. As most of the researcher's colleagues in Pakistani academia felt that it was an intrusion into their territory, their responses were bizarre. Some of them said that women researchers work on women-centric topics; therefore, it was surprising for them to learn that a male researcher was researching a woman-centric topic. Conversely, common questions that the researcher in Study 2 received from doctoral colleagues and professors in his school were (1) As a male why was he interested in research on women-centric issues; (2) Has he considered the issue of male privilege in his research? (3) Why does he expect that women would share their personal and professional secrets or experiences with a male researcher especially in a Pakistani context where gender segregation is a common phenomenon. In Foucauldian terms, some discourses are constrained by setting limits over what can be said or avoided. These discourses, in turn, are deployed to exclude others from meaning-making processes (Foucault, 1971). In this case, discourses of gender appear to be a hindrance to researcher while labeling him as an "outsider". All of these questions raised by participants are valid and raise legitimate concerns, but they also indicate that normalizing research into a different gender is still far away. Despite extensive research conducted on women by male researchers and vice versa around the globe, the assumption regarding men's and women's territory of research is contestable in Pakistan. Several studies support the rationale underlying Study 2 for conducting feminist research on a women-centric topic (see Hearn, 1998;Flood, 2013;Duriesmith, 2016;Schulz, 2020). The existing bodywork highlights methodological and practical issues and need for men engaging in feminist research. Gender is considered a sensitive topic around the world and being a male researcher, studying gender requires extra care. Notably, Pakistan is a country where gender segregation is observed at every level and interviewing female participant by a male might raise ethical and cultural issues. Before approaching the participants, the researcher had reflected on several issues that might arise, such as anonymization of the participants' identity and trustworthiness, from getting consent of female participants to expecting them to talk about issues of abuse, harassment, power relations in an organizational setting, and more importantly, where would the interviews take place.

Willingness of the participants
The willingness of participants is crucial to every research study. As mentioned earlier, gender and religious minorities are considered sensitive in societies like Pakistan due to the religious and cultural connotations attached to both social categories. The studies focused on the sensitive aspects of female academic leaders' personal and professional lives and religious minorities' views about dominant textbook knowledge and marginalization of diverse identities through school as a discursive regime that might cause an upsetting and distressing situation during interviews. For this reason, researchers informed them of the sensitive nature of the topic and their right to withdraw from the study.
In both studies, participants' responses varied significantly within and across sample categories. The participants we approached were enthusiastic and positive because "religious minorities" and "gender" were under-researched and rarely discussed topics in Pakistan. In their opinion, academic research on such topics is crucial because it provides first-hand information regarding the causes of the underrepresentation of religious minorities in school textbooks, and the challenges women face in performing leadership positions in higher education. Nevertheless, both researchers agree that some participants showed discomfort and unwillingness when they were told about the project. They all had different reasons for not participating in respective studies. It has two implications. First, it highlights the sociocultural standing of religious minorities and gender in Pakistan. Secondly, it highlights the difficulties associated with conducting research on religious minorities and gender. In Foucault's view (1982, p. 788), power is neither violence nor consent. This is an overall structure of actions that incites, constrains or prohibits. The purpose is always to control the behavior and activities of the subjects. In the case of the present study, both religious minorities and women exist under larger faith-based and patriarchal regimes, respectively, which control their practices of freedom. As a result, participants (minority teachers and women or men talking about women leadership) do not feel comfortable talking about their identities and realities explicitly.
One woman leader in Study 2 opted not to participate in the research because she felt the interview questionnaire was too personal, and she would not like to share her personal experiences. Herod (1993) describes gender relations as a crucial dynamic impacting the interview process and influencing the type of data gathered. As a point of clarification, all the questions were approved by the candidate's university ethics committee, and participants were free not to answer any specific question if they felt uncomfortable. The woman leader who refused to participate was herself a senior professor and a prolific researcher. Likewise, another woman leader agreed to participate without being recorded because of familial restrictions. Although she was assured of anonymity and confidentiality of the research, she politely declined to participate, citing family restrictions on audio recording. A family here refers to a male member. South Asia has been a patriarchal society traditionally. Although globalization has led to significant changes, women remain largely trapped in traditional roles while males remain dominant (Agha and Shaikh, 2022;Verma and Larson, 2001). Though some women are independent and hold prestigious positions in Pakistani society, their life decisions are still made by the male head of the family. The woman who refused was well into her 50s with ample professional experience, but her leadership identity was undermined by her gender identity. It was also true in case Study 1 on religious minorities, where an interview with a Hindu female teacher was conducted in front of the male owner of a coaching center whose permission was requested by both the researcher and a female. However, the male owner preferred to sit in the interview to make sure the female was safe and not harassed.
In Study 2 on gender, the researcher also had difficulty recruiting some male academics. Since the study focused on women academics leading in higher education, the discourses of men academics were important in uncovering how women academics constructed their leadership identities. One male assistant professor refused to take part in the study, as he had applied for a promotion and did not want to take any risks during the promotion process. Another male assistant professor refused on the same grounds, but he also said women should not be given leadership positions. According to him, women are incapable of leadership and should focus on domestic chores, as there is no place for them in social life. Moreover, he said that he was not only against women's leadership but also against their teaching in higher education. In his opinion, women should be allowed to teach in primary and kindergarten schools at the very least. Patriarchal regimes of truth (Foucault, 1972) reduce women's identity to soft skills and domestic responsibilities, deeming them unfit for intellectual pursuits. The unwillingness to participate in women-centric research indicates the gendered nature of Pakistani universities, where talking about gender is far from normal. Foucault (1972) asserts that regimes of truth normalize some discourses while marking others as taboo. In this vein, the unwillingness of participants to engage in religious minorities' identity and women-centric topics indicates the exclusion of certain discourses at large.

Finding a culturally appropriate research setting
Research settings influence how participants respond to the questions in the interview. In Study 1, when it came to interviewing religious minority teachers, e.g. Hindus, the Researching religious minorities and women researcher did not conduct the interview in a formal academic setting or in a public venue where the interview could be more evident to others. As discourse constrains what can be said (Foucault, 1975), the researcher used purposive sampling to identify potential participants in his contacts in several cities of the Sindh province of Pakistan followed by snowball sampling where necessary and selected an interview site that was convenient for the participants based on their suggestions. The researcher used snowball sampling to reach female Hindu teachers, for instance, because of a "trust deficit" at the intersection of gender and religion, two dominant factors involving the researcher's identity. Despite that, he could have access to only one female Hindu teacher. Arguably, a participant, who was interviewed at home because it was his preferred location, mentioned that he felt comfortable speaking about the questions he had already read. There was, however, an interesting incident that occurred before the interview. His father, a retired government schoolteacher, was also present and was informed about the interview and its purpose by his son prior to the interview. Before he left the interview place, he remarked that I understand that you are here for research purpose, but I tell you that this research is futile and lacks any potential impact since I have been living in Pakistan for more than six decades and I have seen the discrimination and biases growing.
Other participants were also contacted at their convenient locations, mostly their homes. In Study 2, however, all interviews were to be conducted on university premises. In such case, it was difficult to find a space without intrusion. Women leaders and men academics were interviewed in their respective offices on campus. Researcher encountered issues when interviewing women academics and conducting focus groups with female students. Although, the university where the research took place was a coeducational institution with many female academics, students and other support staff, yet gender segregation is maintained wherever necessary. It is, for example, not common for men and women to socialize together, either as faculty or students. Seeing a woman sitting with another man in public will always bring a bad reputation for both or for the female specifically. In Foucault's (1975, pp. 43-46) view, "enclosed spaces" such as universities as in the case of Study 2 are necessary for the exercise of power that discipline individuals. It is in such a space that norms and standards are established that are used to judge individuals. Those who do not follow the set norms are referred to as "deviants" and are therefore excluded from society (see also Foucault, 1971).
Notably, the interviews with women leaders happened in their offices smoothly because leadership roles provide them the liberty to interact with men on frequent basis, the same cannot be said of women academics and students. Interviews with women academics and focus groups with female students were conducted in common areas such as computer rooms, laboratories and libraries where others could easily come and go. As women did not feel comfortable giving interviews behind the closed doors of their offices, conducting interviews in presence of constant intrusion might compromise the secrecy and depth of the data, but considering the cultural environment, it was necessary. The act of maintaining gender segregation in public places and avoiding socializing with the opposite gender indicates adherence to a system of surveillance that exercises power (Foucault, 1980). The socialcultural regimes of honor in Pakistani society require women to maintain non-socialized relationships with men. Therefore, the fear of being seen sitting with a male had adverse effect on women's socioculturally accepted behavior.
Issues during the interviews Qualitative researchers have reported difficulty obtaining rich data via interviews. In many cases, researchers do not obtain the intended data. Saunders et al. (2015) have also outlined the difficulties qualitative researchers face in conducting interviews and anonymizing data relating to participants' demographic information, such as ethnicity and religion. They argue that such challenges are overcome through context-specific strategies. When researching culturally sensitive issues, such as gender and faith, there is a risk of not getting rich data. There were many instances in the interviews in our studies where participants felt reluctant to answer some questions. In Study 1 on religious minorities, during the interview with the Hindu English language teachers, the researcher felt that the nature of the questions determined the pace of the participants' conversation, which Foucault (cited in Burchell et al., 1991, p. 59) describes as "limits and forms of the sayable", such as what is possible to speak of? What is constituted in the domain of discourse? Teachers could articulate their responses very confidently when asked neutral questions, such as the purpose of teaching, their professional choice as English language teachers, or their methods for teaching English as a foreign language. However, the questions that pertained to their responses to learner identity, dominant meanings and values in textbooks, their reactions and engagement with such content, learners' reactions, and their own opinions about textbooks while considering the diversity in Pakistan, the teachers seemed a little reluctant or hesitant to give detailed responses or to express themselves confidently. This was evident in their facial expressions and body language (e.g. nervousness when articulating their voices).
In Study 1, one participant requested that the researcher pause the recording for a while since he felt insecure speaking about dominant meanings in textbooks and representations of the Other. Upon stopping the recording, he said he had a lot to say, but he showed a desire not to proceed with the interview further. This resulted in the participant being excluded from the interview data. He told the researcher there were several incidents in Sindh Province against Hindus that made him fear expressing his political or religious views or speaking freely about his identity. To ensure his safety, the researcher in Study 1 erased his recordings from the device in his presence upon his request as an ethical requirement. This perceived fear by the participants was due to the increased number of incidents in recent times relating to mob lynchings based on blasphemy allegations against teachers in schools and colleges across the country. For example, in 2019, an English teacher (head of the department) in Punjab province of Pakistan was stabbed to death by his student over gender mix in the college that the student thought as "un-Islamic" (see Imran, 2019). Moreover, a recent incident of a Hindu physics teacher at a government college in Sindh province of Pakistan is a notable example who was sentenced to life imprisonment on February 8, 2022, based on the student's accusation of his derogatory remarks about the prophet Muhammad who is thought to be the last prophet of Islam (see Shamsi, 2019).
Among the five women leaders interviewed in Study 2 on gender, only two spoke extensively about workplace harassment. Similarly, out of four women academics interviewed in the study, only two women academics shed light on workplace harassment. Others refused to acknowledge workplace harassment despite the prevailing statistics collected by the researcher in his study. The harassment of women at work is extremely widespread, and it is shown to negatively impact a woman's work productivity, emotional health and relationships with her family (Celik and Çelik, 2007;Pryor, 1995). Many women leaders and academics seemed reluctant to address issues such as sexual harassment, male privilege, and sociocultural and religious discourses on women. Yet, some participants managed to talk intensively about many sensitive gender-related topics within the scope of the doctoral study. The disinterest of women in discussing taboo topics is better understood by considering Foucault's rules of exclusion explaining the rejection of some discourses. Each society has its own regime of truth (Foucault, 1972(Foucault, , 1976(Foucault, , 1980) that controls the production and distribution of discourses. Truth regimes in a society decide what is right and what is wrong, and who is allowed to speak. The religious and cultural regimes of truth in Pakistani Researching religious minorities and women society discourage women from discussing sexual harassment and religious minorities from expressing their views and beliefs.

Conclusion
Reflections from two doctoral studies in Pakistan related to research on religious minorities and gender issues were presented in this study. Researchers discussed the challenges they encountered when interviewing their selected samples during the study. In both studies, the interview questions and consent and research privacy notice forms were prepared using GDPR guidelines which were approved by universities in Finland and the UK. As this paper shows, young doctoral researchers face several challenges when researching sensitive topics, such as religious minorities and gender in a patriarchal society. Among these problems, we found that the researcher's positionality as a partial insider influences the process of accessing participants and gathering data, alongside participants' unwillingness, appropriate research locations and difficulties during the interview. These issues can be attributed to participants' perceived fears (for example, blasphemy allegations) when researching religious minorities, as well as labeling people (male or female) when sitting with the opposite gender in public places and limiting access to the study for female participants. Moreover, participants' refusal to speak about such issues and withdrawal from the interview indicate their adherence to these cultural (gender-based) and religious regimes of truth. Consequently, participants' discomfort in disclosing their lived experiences and talking about social taboos has serious repercussions on the field of gender and religious minority research in education and their overall standing in society. In Foucault's (1975) view, this process is seen as docility resulting from societal power relations. As a result of truth regimes, some knowledge forms are entrenched in the social web while others are excluded. Thus, such faith-based and patriarchal regimes restrict the discourses of research in education. As a result, the study holds that educational research is directly linked to societal structures and vice versa. The structural shift can have emancipatory effects on research on such sensitive topics and likewise, the normalization of researching gender and religious minorities can affect structural transformation at the macro level. Therefore, the study suggests that in societies such as Pakistan, where religion and gender are emotive terms with connotative meanings, the problem can be handled by counter-discourses that challenge truth regimes by conceiving research as activism rather than mere an intellectual activity. Moreover, such societies like Pakistan where research and dialogue are still not normalized practices require a policy for ensuring security and legal protection to researchers and participants in the interest of knowledge-making in their respective fields. Future research in similar contexts where gender and religious minorities find a little freedom to articulate their views related to their identities can be helpful to better understand how regimes of truth work in specific societies and what possible remedies can be developed to counter these regimes.