A discourse of virtue: how poor women entrepreneurs justify their activities in the context of moderate Islam

ABSTRACT Entrepreneurship has been both celebrated and critiqued in terms of its ability to assist women in developing countries to overcome the constraints of patriarchy, with recent views acknowledging its potential for incremental, dialectical change. This is particularly true for women entrepreneurs in contexts where Islamic gender relations are practiced, which can pose certain limits to women’s entrepreneurship. We contribute to an emerging stream of research that highlights how women entrepreneurs in such contexts leverage diverse interpretative repertoires to describe and justify their work. In particular, we shed light on the understudied but populous group of women entrepreneurs of lower social class in contexts of moderate Islam. We identify virtuous repertoires as a key discursive element that assists women in these contexts to present their entrepreneurial activities and discuss implications for theory and practice.


Introduction
Entrepreneurship could be a valuable activity for women entrepreneurs in developing countries, who can benefit from greater financial independence and improved status within their household and community (Datta and Gailey 2012;Haugh and Talwar 2016). Nevertheless, the patriarchal gender relations that frequently characterize such contexts can potentially inhibit the purported benefits of entrepreneurship, as entrepreneurship does not change the root cause of women's poverty (Murthy et al. 2008;Venkatesh et al. 2017). An emerging stream of work on women's entrepreneurship in non-Western, patriarchal contexts (e.g. Alkhaled and Berglund 2018;Sigalla and Carney 2012) suggests that gains for women might be incremental, as the status quo and women's entrepreneurial activities adjust to each other in a dialectical way. The key component in this dialectical process is the interpretative repertoires that women use in their discourse (i.e. the discursive tools for characterizing and evaluating actions and events, Potter and Wetherell 2010). Studies revealing the interpretative repertoires used by local women entrepreneurs in different regional contexts (e.g. Braches and Elliott 2017;Díaz-García and Welter 2013) suggest that these vary substantively depending on a number of factors.
In Islamic contexts in particular, religious prescriptions markedly affect how women entrepreneurs can describe, legitimize and justify their right to become entrepreneurs and their entrepreneurial activities. Prior studies reveal that in contexts of conservative Islam and for women entrepreneurs of lower social class, interpretative repertoires used to justify entrepreneurship focus on concealing entrepreneurial activities, as well as limiting them to feminized sectors (Al-Dajani and Marlow 2010). In comparison, middle to higher class women entrepreneurs in both conservative and moderate Islamic contexts tend to use more sophisticated repertoires that reinterpret Islamic scriptures to cast entrepreneurship as highly compatible with them (e.g. Althalathini, Al-Dajani, and Apostolopoulos 2021;Sakai 2019;Tlaiss and McAdam 2021b). In contrast, the study of how women entrepreneurs of lower social class justify their entrepreneurial activities in a context of moderate Islam has received far less attention, in spite of such contexts being increasingly influential in the Muslim global community (Religious Freedom Institute 2020). In this study, we therefore focus on the following questions: What are the types of interpretative repertoires that lower social class women entrepreneurs employ in a context of moderate Islam to justify their entrepreneurial activity? How do these repertoires compare to those employed by women entrepreneurs in other Islamic contexts?
To answer these questions, we adopted a qualitative approach that focused on the discourse of lower social class women in rural Indonesia, specifically in three villages of Central Java. In this context, a moderate interpretation of Islam and patriarchy limit the role of women, yet women's entrepreneurship is thriving. The data was collected in two rounds of field visits and comprises of interviews with women entrepreneurs, supplemented by observations and informal discussions. In our analysis, we employed Wetherell and Potter (1988) and Potter and Wetherell (2010) discourse analysis approach to identify and examine (a) the types of interpretative repertoires that women entrepreneurs used to uphold or challenge expectations, based on local religion and culture and (b) how they combined interpretative repertoires that varied in their conformance towards religious prescriptions.
Our findings indicate that lower social class women entrepreneurs in the context of moderate Islam justify their entrepreneurial engagement amid the prevalence of conservative gender relations with two main groups of interpretative repertoires: compliant and virtuous. Certain conservative views on gender relations persist within a moderate Islamic community, namely the positioning of women as submissive dependents of their husbands and the limiting of women's roles to the domestic sphere (Ida 2001). Compliant repertoires are aligned with these conservative gender relations yet provide room for women to be entrepreneurs as they present entrepreneurship as conforming to the status quo. In contrast, virtuous repertoires downplayed the incremental shifts in gender relations due to women's entrepreneurial engagement by shifting attention towards nongendered, non-religious values that are highly esteemed within the local culture. Importantly, all women interviewed used a combination of compliant and virtuous repertoires to present their entrepreneurial activities as aligned with the prevalent conservative gender relations. These findings suggest that women of lower social class in a moderate Islamic context can be skilful in integrating both religious and cultural values in order to cast entrepreneurship as a legitimate economic activity for themselves without being overly threatening to the status quo.
This study informs prior literature in three ways. First, it extends prior studies on lower social class women entrepreneurs in conservative Islamic contexts (e.g. Al-Dajani and Marlow 2013;Ritchie 2016) by showing that interpretative repertoires that downplay the contribution and scope of women's entrepreneurship are also relevant to women entrepreneurs in similarly lower class but moderate Islamic contexts. In contrast, repertoires found to be employed by higher social class women entrepreneurs in both conservative and moderate Islamic contexts (e.g. Tlaiss 2015b; Verduijn and Essers 2013), which rely on feminist interpretations of Islamic scriptures, are absent in the discourse of lower social class women in moderate Islamic contexts. Second, our study contributes to the stream of literature on women's entrepreneurship in Islamic contexts by identifying virtuous repertoires, a group of interpretative repertoires that are employed specifically by lower class women entrepreneurs in moderate Islamic contexts. This group of repertoires has not, to our knowledge, been discussed in prior literature, and reveals a means for women entrepreneurs in such contexts to justify their activities while mitigating the constraints of patriarchy and of religious prescriptions. Third, our study complements literature on women entrepreneurs in Islamic contexts (cited above, also Roomi 2013; Sakai and Fauzia 2016; Tlaiss 2015a), by additionally illustrating the ways that interpretative repertoires of different compliance towards the status quo can be combined in order to mitigate opposition, in contexts similar to ours. As a whole, this study highlights the value of considering both social class and types of Islamic interpretation alongside contextually relevant values when studying the nexus of women's entrepreneurship, Islam and discourse. Finally, our study could be useful to policymakers and practitioners that are interested in leveraging contextually appropriate discourse to promote women's entrepreneurship in communities where the latter is not a widely legitimate cause.

Entrepreneurship, patriarchy and the role of discourse
Entrepreneurship has been frequently celebrated in practice and academia as an instrument for enabling poor and lower social class women in developing countries to overcome the multiple constraints they face (Alvarez and Barney 2014;Bruton, Ketchen, and Duane Ireland 2013). In particular, it can be useful in strengthening women's financial independence (Inayatullah and Birley 1997;Kulb et al. 2016;Yunus 1998), which enable them to develop self-reliance and gain decision-making power within their household  and be recognized as productive members of the local community (Sanyal 2009). Yet, entrepreneurship is not a panacea and might even sustain gender-based inequality, as its embeddedness in local contexts can contribute to the reproduction of patriarchal structures and practices (Jennings, Devereaux Jennings, and Sharifian 2016) that are the root cause of women's poverty (Al-Dajani and Marlow 2013). Patriarchy, which is defined as 'a system of social structures and practices in which men dominate, oppress and exploit women' (Walby 1991, 16), can cripple even well intended efforts at promoting entrepreneurship. This is because women entrepreneurs can continue to lack control over their mobility, education, and social engagement (Murthy et al. 2008) and even suffer increased domestic violence (Bulte and Lensink 2019;Green et al. 2015;Schuler et al. 1996).
Beyond these contrasting views, an emerging literature suggests that entrepreneurship contributes to incremental changes in the status quo of patriarchal structures and practices. Verduijn et al. (2014) observe the dual side of entrepreneurship, whereby empowerment and oppression not only co-exist, but are also in constant tension with each other. This is reflected in Sigalla and Carney (2012), according to which entrepreneurship transforms women into efficient market actors but at the cost of increased inequality. Their studies show that women who exhibit pre-defined, normative characteristics of successful entrepreneurs (e.g. credit worthiness, discipline, strategic mindset, in control attitude) are able to acquire the necessary resources to grow their business and improve their social status. At the same time, women who are constrained by conservative gender relations become excluded from necessary resources. As a result, entrepreneurship ends up perpetuating the social divide between these two groups.
An emerging stream of research focusing on the role of discourse further clarifies how women entrepreneurs navigate the coexistence and tension between entrepreneurship and patriarchy. Discourse is important to this relationship, as it is not only an instrument to produce, transmit, and reinforce patriarchy, but also a medium to expose inequalities and to counter the effects of patriarchy (Ahl and Marlow 2012;Mills 2003). Research within this stream emphasizes the individual microresistance processes employed by women, through which alternative interpretations and new gendered subjectivities emerge (Calás, Smircich, and Bourne 2009;Thomas and Davies 2005). In contexts wherein patriarchal structures and practices markedly limit women's roles and activities, micro-resistance is a particularly relevant element for justifying and legitimizing the expanding limits of womanhood (Alkhaled and Berglund 2018;Goss et al. 2011), including women's entrepreneurial activities. Microresistance in discourse can be isolated through the identification of relevant interpretative repertoires (Potter and Wetherell 2010;Wetherell and Potter 1988) that women draw upon when talking about their entrepreneurial activities. Interpretative repertoires, which are 'recurrently used systems of terms used for characterizing and evaluating actions, events and other phenomena' (Potter and Wetherell 2010, 149), act as linguistic building blocks in the presentation of entrepreneurship as a legitimate economic activity for women amid patriarchy.
Such interpretative repertoires that women employ to legitimize their entrepreneurial activities, and the way they use them, can vary greatly across geographical contexts. The discourse of women entrepreneurs in Western contexts has been found to draw mainly on a masculine interpretation of entrepreneurship, which is implicitly in dissonance with womanhood (Ahl 2002;Bruni, Gherardi, and Poggio 2004;Duberley and Carrigan 2013). For instance, the 'anti-child, anti-woman' repertoire encapsulates the expectation for women in Germany to separate the public and private spheres of their lives in order to be considered as legitimate entrepreneurs (Braches and Elliott 2017). In turn, women entrepreneurs in Spain confront the same expectation with the 'clearing the hurdles' repertoire to accentuate their ability to match men's business performance, and the 'juggling act' repertoire that presents them as profitable business owners who also conform to traditional feminine values (Díaz-García and Welter 2013).
Although masculine interpretations of entrepreneurship are also present in non-Western contexts, their influence is weaker, in particular within highly patriarchal cultures. Scholars have therefore called for transcending the limits of the Western archetype of entrepreneurship to highlight the voices of non-Western women entrepreneurs who remain underrepresented within entrepreneurship discourse (Verduijn and Essers 2013) and for studying them vis-à-vis their local context (Alkhaled 2021). Women in such contexts have been found to re-interpret religious and cultural discourses to recalibrate their role and social position before casting entrepreneurship as compatible with their womanhood (e.g. Tlaiss 2015a, 2015b). Re-interpretation is nevertheless a challenging feat as cultural discourses are highly institutionalized and guarded by various gatekeepers at local and state levels (Claus and Tracey 2019). As observed by Mair, Wolf, and Seelos (2016), context-sensitivity is essential in the re-interpretation of cultural discourses, to ensure acceptance and support from the local community. It seems, therefore, that the social and cultural context in which women are embedded partly determines the availability and appropriateness of interpretative repertoires that they can use to justify their entrepreneurial activities.

Women entrepreneurs' justification of their activities vis-à-vis local culture and religion
Particularly challenging for women entrepreneurs are contexts where cultural and religious patriarchal prescriptions coincide with poverty, limiting their ability to engage in entrepreneurship, and to benefit from it. Research on women's entrepreneurship in poverty contexts has identified several such prescriptions that inhibit women from being entrepreneurial. The cultural practice of secluding unmarried women from public observation, such as the purdah (in India, Bangladesh, and Pakistan), restrains women from taking part in social and economic activities outside the bounds of their home or without the supervision of a male member of the family (Marta, Patel, and Wincent 2017). These practices are more common in rural areas, which is where the majority of the poor are located (Ritchie 2016). Local interpretations of Islam also vary across contexts and affect the types of entrepreneurship and related discourses that are possible for women to employ. Prior research at the intersection of gender and Islam refutes the notion that Islam instils gender discrimination and inequality, and threatens women's entrepreneurial activity (Essers et al. 2009;Tlaiss 2015b;Tlaiss and McAdam 2021b). Islam as a religion of peace and social justice emphasizes that men and women are equal in God's eyes; yet, patriarchal interpretations of sacred scriptures have asserted men's superiority and control over women (Brenner 2011).
A nascent stream of literature examining women's entrepreneurship in Islamic contexts has shed light on how the types of interpretative repertoires they use vary across Islamic contexts. As illustrated in Figure 1, we make sense of this literature by mapping findings across two contextual dimensions, women's social class and type of Islamic interpretation, based on the information provided by the authors of the studies, particularly in their Methods and Findings sections 1 Studies on women entrepreneurs of upper/ middle social class in Islamic contexts demonstrate that women employ sophisticated interpretative repertoires to present their entrepreneurial engagement as largely aligned with Islam. Women entrepreneurs in these social classes are knowledgeable of feminist interpretations of Islam and thus able to expose patriarchal interpretations of sacred scriptures and find evidence for gender equality (Essers and Benschop 2009). The women are therefore well-versed in quoting sacred texts and referring to appropriate Islamic teachings when resisting mobility restrictions, gender segregation (Althalathini, Al-Dajani, and Apostolopoulos 2021;Roomi 2013) and the social stigma placed on working women (Essers and Benschop 2009;Sakai 2010;Sakai and Fauzia 2016;Tlaiss 2015a;Tlaiss and McAdam 2021a). They legitimize their entrepreneurial engagement by grounding their motivation on religious justifications, such as emulating the lives of their entrepreneurial religious role models, namely Prophet Muhammad and his wife Khadijah, and answering a divine calling through entrepreneurial endeavours (Althalathini, Al-Dajani, and Apostolopoulos 2021;Essers and Benschop 2009;Tlaiss 2015a;Tlaiss and McAdam 2021b). Furthermore, they leverage repertoires from Islamic morality and work values to display adherence to Islamic teachings through their business, there by legitimizing it (Essers and Benschop 2009;Sakai 2010;Sakai and Fauzia 2016;Tlaiss 2015a;McAdam 2021a, 2021b).
Studies on women entrepreneurs of lower social class in conservative Islamic contexts demonstrate that women employ less sophisticated interpretative repertoires and make minimal reference to sacred scriptures and specific Islamic teachings. The study by Al-Dajani and Marlow (2010) sheds light on how women present their entrepreneurial engagement as aligned with established boundaries of female activities. In addition to operating in feminine sectors, the legitimacy of women's activities depends on their ability to display adherence to traditional gender scripts, such as submission to men's control and the prioritization of domestic responsibilities. While these repertoires allowed some women in this study to operate their businesses overtly, others still faced opposition from male patriarchs and had to operate covertly. Similarly, Ritchie (2016) demonstrates that in rural areas wherein strict forms of segregation, such as the purdah, are widely practiced, men have total control over women's social movement, access to resources and economic engagement. Here, it is important for women to draw on interpretative repertoires which signal close adherence to local social protocols despite their somewhat increased mobility. However, in areas where the purdah is practiced in conjunction with stricter cultural values, such as the ashtunwali (tribal code of honour and shame), these interpretative repertoires are viewed as radical and thus cannot be used to break barriers and overcome women's exclusion from entrepreneurial activities.

Islamic Interpretation
Taken together, these studies establish valuable insights on the use of discourse by women entrepreneurs in non-Western, particularly Islamic, contexts. Additional studies on the nexus of women's entrepreneurship and religion could further extend this body of work and the scholarly understanding of women entrepreneurs' experience and discourse. Furthermore, while discourse can be profoundly influenced by regional differences in terms of the type of Islamic interpretation and in terms of women's social status, we are not aware of any studies that have focused on women entrepreneurs of lower social class in a context of moderate Islam, even though this group is increasingly influential in the Muslim global community (Religious Freedom Institute 2020). With this study we focus on this rather overlooked group of women entrepreneurs and attempt to generate answers to the following questions: What are the types of interpretative repertoires that lower social class women entrepreneurs employ in a context of moderate Islam to justify their entrepreneurial activity? How do these repertoires compare to those employed by women entrepreneurs in other Islamic contexts?

Context of the study
The context that allowed us insights into the aforementioned questions was rural Indonesia. Microenterprises are essential for the Indonesian economy as they constitute 99% (58 million) of enterprises in the country (Naples 2018b), 60% of which are owned by women (The World Bank 2016). Authorities have placed consistent emphasis on strengthening and increasing women-owned microenterprises (Tambunan 2011), hence various non-governmental organizations and enterprises have supported this agenda by integrating microcredit and enterprise development into their programmes (OECD 2018a). With various forms of microcredit institutions, such as governmentowned (Naples 2018b; The World Bank 2016), religious-based (Afoukane, Utami, and Nugroho 2021;Sakai 2010) and trade-based (Torri 2012), serving the Indonesian poor in both urban and rural areas, women-owned microenterprises are thought to be growing although statistical data at regional-and country-level are unavailable. In the case of our research location, the presence of a village-owned microcredit cooperative has facilitated the increase in women's entrepreneurship in the area. As stated by the board members of the villages' microcredit cooperatives, the increase in membership (by 75% in two cooperatives and 300% in one cooperative) and the large number of active loans and demand for new loans attest to the growing interest in starting and expanding microenterprise activities among women in the respective villages.
Over 207 million of the Indonesian population (87.2%) are Muslims. Islamic women's social movements are active and vocal in their discussion of gender and sexuality, which policymakers take into account in policy formulations and amendments (Qibtiyah 2009). Nevertheless, women's voice is not unified due to the diverse Islamic interpretations. One of the dominant views on gender relations was established by Nahdlatul Ulama (NU), which is Indonesia's largest Muslim organization with over 90 million followers (Religious Freedom Institute 2020). Its moderate Islamic interpretation on gender relations (Wahyuni and Wafiroh 2013) is disseminated through various channels, including its own women's division (i.e. Fatayat NU, Muslimat NU) (Rinaldo 2014;Sajaroh and Mahmudah 2018). This moderate Islamic interpretation accepts kodrat (the innate nature or biological characteristics) as the basis for gender relations, but re-interprets them as equal and complementary rather than hierarchical (Qibtiyah 2018). Women and men are expected to support each other; hence, women may work to support the family's finances provided that their domestic duties are prioritized (Wahyuni and Wafiroh 2013). The moderate Islamic interpretation of gender relations is however facing opposition due to the rise of religious conservatism in the country. Conservative Islam follows a literal interpretation of sacred scriptures in which men and women's kodrat determine their roles and positions (Qibtiyah 2018). Therefore, the conservative interpretation of Islam positions women as subordinates of men and delimits their role to fulfiling motherhood and domestic tasks, while excluding them from economic activities (Brenner 2011;Ida 2001). Our selected context is one of moderate Islam as NU is the dominant Islamic organization in our research site and in Central Java overall. Furthermore, all of the interviewed women were members of Fatayat NU and Muslimat NU, and some of them were educated in NU's boarding schools. Nevertheless, conservative views of Islam co-exist even within overall moderate Islamic contexts.
Furthermore, the view on gender relations according to the local culture in our context is similar to that of conservative Islamic interpretations. The Javanese ethnic group, to which the study participants belong, is the largest ethnic group with 40% of Indonesian population belonging to it (Indonesia.go.id 2017). According to Ida (2001), Javanese culture is hierarchical with men positioned as the head of the family and women as subordinates and companions. Women's main role is to serve their husbands by providing free domestic labour and engaging in traditionally feminine activities (e.g. batik making, traditional medicine) at home. They are expected to be financially dependent on the husband, who is the breadwinner of the family. Nevertheless, entrepreneurship among Javanese women of lower social class is common and thriving. This is especially true in our selected research location where entrepreneurship is one of the few employment options for women.

Method of the study
We employed an inductive approach (Potter and Wetherell 2010) to understand women's entrepreneurial experience amid patriarchy. We collected data from multiple data sources (i.e. observation, archives, informal interviews), with semi-structured interviews comprising the core of the data. Considering the culture and the generally low level of education within our chosen local context, we chose qualitative interviewing due to the flexibility it offered for capturing women's lived experience as compared to surveys and other quantitative approaches. Moreover, interviewing was a less intrusive method compared to participant observation given that the authors were not full insiders to the local context and that women's businesses were located mostly within their houses. In analysing the data, we took a poststructuralist feminist perspective according to which gender is socially and discursively constructed, and examined how gender and power relations were (re)produced in the women's account of their entrepreneurial experience (Ahl 2002(Ahl , 2006. Accordingly, we drew on discourse analysis stemming from the works of Potter and Wetherell (2010) and Wetherell and Potter (1988), who have developed an approach for identifying interpretative repertoires.
Our team brought both an outsider and a (partial) insider standpoint to bear on this study, which has certain benefits (Langley and Klag 2019;Naples 1996). The first author's Indonesian nationality ensured ease of communication with the research participants and other stakeholders. Yet, our outsider standpoint (in terms of religion, for the first author, and religion and nationality for the second) strengthened our ability to assess critically the phenomenon unconstrained by takenfor-granted local cultural structures and practices. As we were aware that gender inequalities can sometimes be perpetuated within mainstream entrepreneurship research (Ahl 2004(Ahl , 2006, we also tried to be actively reflexive about our own taken-for-granted attitudes concerning gender and feminism to avoid becoming judgemental of local culture and practices. To balance objectivity and empathetic understanding of the local context, we put the women's own voices at the centre of our methodological approach.

Sampling and data collection
To better understand the local sociocultural context of the study and the key issues faced by poor women entrepreneurs in contexts of moderate Islam, the first author travelled to Batang District (Central Java, Indonesia) in July 2017. She gained access to the research participants with the help of the founder and officers of Bina Swadaya, an Indonesian social enterprise specializing in community development. She spent two weeks shadowing Bina Swadaya's officers and visited eight villages which were part of the community development programme. She conducted informal interviews with Bina Swadaya officers and various stakeholders, such as government and village officials, activists, and local women. These interviews were complemented by observation of community meetings and social gatherings. The first-round interview and observation data shed light on the various religious and cultural discourses that shape women entrepreneurs' lives. The data also indicated noteworthy contradictions, such as women's entrepreneurship thriving despite the highly gendered division of labour. We therefore oriented our study towards a better understanding of how local women navigate these contradictions.
The primary empirical material was collected during the second round of data collection, which took place in March 2019. At that time, the first author visited three of the previously visited villages to conduct one-on-one interviews with 34 women entrepreneurs. 2 All of them were married and had at least one child, while their age ranged from 26 to 55 years, with an average of 37.5 years. Half of the women had completed the obligatory nine years of education, one quarter of the women had completed six years of education, and one quarter had at least twelve years of education. The majority of women entrepreneurs (29 persons) operated their businesses at home. Two women had microbusinesses outside the home, while three women had microbusinesses both at home and in a business space outside the home. Twenty women operated a single microbusiness and fifteen women operated several microbusinesses. The types of microbusinesses were clothing (six women), grocery stalls (ten women), prepared food (eight women), baked goods and snacks (twelve women), and other businesses (fifteen women) such as laundry service, seamstress service, or tea shop. The businesses also varied in terms of their lifespan; nearly 40% were new businesses (less than five years), 30% were established businesses (five to ten years), and 30% were mature businesses (more than fifteen years). All women were managers of their household funds since their marriage, as is typical in these contexts, and indicated that their business income was essential for improving the household finances. Their income was mainly used for groceries, children's tuition and pocket money, women's personal needs, and community-related social activities (e.g. regular prayer and community meetings, marriage and funeral offerings, alms). These are considered as secondary and feminized household expenditures; hence, could not be used as proof of their increased decisionmaking power (Alam 2012).
Interviews took place in a neutral place outside the women's home, either in the office of the village-owned cooperatives or in the residence of a board member of the cooperative. Interviews were semi-structured, conducted in Indonesian, and lasted an average of 40 minutes. As the interviews were audio recorded, all of the participants were assured of confidentiality and signed a consent form. The interviews were focused on a number of issues, including microbusiness activities and gender roles, to provide an understanding of the women's entrepreneurial experience as well as discourse.

Data analysis
Discourse analysis (Potter and Wetherell 2010;Wetherell and Potter 1988) guided our analysis of the ways that women entrepreneurs use language to represent and justify their entrepreneurial experience. This approach suggests that people's accounting of their lived experience is built upon preexisting linguistic resources, such as interpretative repertoires (Wetherell and Potter 1988). With numerous variations and types of interpretative repertoires available, the selection of repertoires depends on the function of the account, such as to request, to justify, and to legitimize, among others (Potter and Wetherell 2010). We accordingly coded for the types of interpretative repertoires that women entrepreneurs employ.
The data analysis followed an inductive process, whereby we moved iteratively between the following three stages of analysis (Potter and Wetherell 2010). At the first stage, we prepared the data for analysis by transcribing the audio recordings of the interviews, reading the 34 original transcripts repeatedly and coding them with the help of the Atlas.ti software. Coding was conducted on the original transcripts in Indonesian, in order to retain nuanced meanings shaped by linguistics and culture-specific expressions (Charlebois 2010;Xian 2008). A second round of coding focused specifically on the research questions and the use of interpretative repertoires in the women's discourse, also examining the similarities and differences in the types of repertoires within and across accounts (Wetherell and Potter 1988). We identified six interpretative repertoires -wife's income as supplementary, permission makes activity possible, tenacious, financially autonomous, productive, and collaborative -which will be discussed in detail in the following section. In the second stage, the analysis followed a continuous iteration process as each repertoire was examined in relation to the specific excerpt containing it (e.g. the topic being discussed, the presence of and interaction with other interpretative repertoires) and the entire interview transcripts (Potter and Wetherell 2010). Based on this stage of the analysis, we were able to abstract the six interpretative repertoires into two broader categories, which we termed compliant repertoires and virtuous repertoires. During the third stage, we examined in detail the combination and sequence of the repertoires in relation to their category membership (Potter and Wetherell 2010;Wetherell and Potter 1988) and identified one consistent pattern of interpretative repertoire combination.
The results of the analysis are described in the following sections, illustrated by excerpts from the interviews. Additional illustrative examples can be found in Tables A1, A2 in the Appendix. These excerpts have been translated to English focusing on meaning rather than word-to-word accuracy in order to remain faithful to the original connotations present in the women's discourse (Xian 2008). All of the women's names used in this article are pseudonyms.

Findings
Our analysis revealed that the lower social class women entrepreneurs in our context of moderate Islam were subjected to constraints posed by conservative views of gender relations, in line with findings of studies on women entrepreneurs in contexts of conservative Islam. However, in a moderate Islamic context, women entrepreneurs were able to downplay and re-interpret some of these constraints by employing compliant repertoires that exposed alternative, moderate interpretations of gender relations. Beyond these compliant repertoires, which have been identified in prior research (Sakai 2019;Sakai and Fauzia 2016), our analysis also highlighted a divergent group of repertoires, which we termed virtuous repertoires, that have not been reported previously, and which were central to the women's talk about their entrepreneurial activity. Below we describe how women entrepreneurs referenced the conservative gender relations and how the women attenuated them with the use of compliant and virtuous repertoires, respectively.

Conservative gender relations
Conservative views on gender relations were manifested in descriptions of an unequal marital relationship and the gendered division of labour. According to the interviewees, Islamic teaching asserts that men and women are equal in God's eyes, but married men are bestowed the role of the family's Imam (i.e. the prayer leader in Islamic religious service): Women and men are equal in God's eyes. But the Imam is the husband. It means that we have to be more obedient towards the husband. -Fajarina Men as the Imams were responsible for leading and ensuring the spiritual well-being of the family members. Married couples could discuss and argue over family matters, but men had the final say and women were expected to obediently follow their husband's decision. As leaders, men also had control over women's activities as wives were obliged to acquire their husband's ridho (permission) prior to performing any activity beyond the boundaries of the domestic space: It means that when the husband [gives] ridho, then everything is enjoyable . . . The husband's permission needs to be followed as a wife . . . When husband is not [giving] ridho then we are not even allowed to take a step . . . If I want to go anywhere, I have to ask the husband . . . If we do not ask for permission, we are sinful. The husband does not punish, but the punishment is from Above, from our God. Meaning that even to take the first step, whatever steps, without the husband's permission is sinful. -Erina A wife had to acquire her husband's ridho when engaging in non-domestic activities, such as participating in the village's organizational activities or pursuing an economic opportunity. Consequently, women could not act without the husband's permission -even for activities deemed as good: [I am expected] to be devoted to the husband . . . The husband's decision must be respected . . . If my husband does not permit, even though it is to do something good, I am sinful. Not allowed. Without the husband's permission, we will have difficulties in life, difficulties. It has to be with the husband's blessings. For everything we have to ask for husband's permission. If the husband permits then it is allowed, we are comfortable, happy. -Ginanita Additionally, women's submission to men's leadership was in line with the division of labour based on kodrat, which is based on biological characteristics of women and men. With productive labour being assigned to men, husbands became the obligatory income providers for their wife and children. In turn, women's kodrat as mothers determined their childrearing and caretaking roles and tasks. Additionally, women were entrusted with domestic responsibilities. Their principal duty was to serve the husband by preparing his meals and clothes. Women's other duties involved managing a tidy and clean home while ensuring that the children were well behaved and educated: When it comes to sweeping the floor, do washing, do the laundry, it is all me . . . The main responsibility is to serve the husband. For instance, I must prepare food or prepare the clothes, cook. . . . Whichever way we manage our time so that everything is ready when needed, whether it is clothes or food, whatever. -Batari The gendered division of labour fortified women's dependence on their husbands' leadership in terms of their ability to engage in productive work. However, the limited availability of work opportunities for the husbands and their unstable flow of income created a window for women's entrepreneurship. In response, the women acknowledged the importance of ensuring their family's livelihood through their economic participation, but not at the expense of violating religious teachings or being disrespectful of their husbands. Hence, the women's accounts of their entrepreneurial experience were infused with various displays of adherence towards the established, conservative gender relations. These conservative gender relations were nevertheless attenuated by drawing on compliant as well as virtuous repertoires.

Compliant repertoires
Conversations with the women on their entrepreneurial activities indicated various instances where entrepreneurship was not fully in line with the conservative view on gender relations. Men's role as the family leaders and providers was particularly affected as women became economically productive through entrepreneurship. We observed two types of interpretative repertoires which emerged as women cast entrepreneurship as being within the scope of men's leadership, which we group under the theme of compliant repertoires. These repertoires emerged when the women addressed the issue of their right to earn an income through entrepreneurship. The first such repertoire, which we termed 'wife's income as supplementary', referred to women's duty to help their husband while positioning their earnings as lesser than -instead of equal to or a substitute for -the husband's income. As illustrated in Fajarina's account below, women's position as helpers implied the responsibility to ease men's burden rather than the obligation to contribute to the household finances: It is very much allowed as additional [ Accordingly, the 'wife's income as supplementary' repertoire entailed the downplaying of women's financial contribution to safeguard men's role as the family provider. As in Fajarina's account, the women employed the word 'only' to ensure that their income is in no way perceived as equivalent or competitive to the husband's.
Additionally, the women minimized the gravity of their financial contribution by clarifying that they allocated their earnings to what they described as small expenses, such as children's pocket money, daily groceries, social expenses, and personal need items. Concurrently, by claiming that their husband's earnings were allocated to larger expenditures, such as to purchase household assets and appliances, the women positioned their contribution as less important than the husband's: Well, the money from the husband is maybe for bigger needs because it can be saved beforehand. Then the money from me is for groceries. And like how it is with people in the village, Ma'am, there are a lot of invitations. And for the children's pocket money, and school . . . Bigger needs . . . something that requires large funding like, maybe, the child asks for a motorcycle -Dhatu The second interpretative repertoire, which we termed 'permission makes activity possible', emerged when the women talked about their individual right to become entrepreneurs. The women's account indirectly addressed the concern that men's authority could be undermined due to women's newly established economic contribution. Hence, the 'permission makes activity possible' repertoire emphasized the importance of acquiring the husband's blessing for starting a business. Such permission was sometimes essential for entrepreneurship, as the case of Kanista, who was denied other types of employment, indicates: 'I am actually not allowed to do any work. Since a long time ago . . . my husband is angry if I work. But, with selling [clothes] like this, maybe he is giving me the opportunity'. As the husband's permission could be given under certain conditions, the 'permission makes activity possible' repertoire also entailed adapting the business operation according to the husband's demands: My husband is like that. If we are ikhlas (sincere) then everything will be made easier . . . So, my husband supports . . . As long as the business is not consuming too much of my time and the husband could be taken care of, why not? -Erina We grouped the 'wife's income as supplementary' and 'permission makes activity possible' repertoires under the broader category we termed compliant repertoire as they demonstrate women's compliance towards conservative gender relations. The compliant repertoires were legitimate justifications for the women's entrepreneurial engagement since entrepreneurship is presented as reinforcing men's status as the leaders and providers of the family, and women's role as obedient and helpful wives.

Virtuous repertoires
Misalignment between entrepreneurship and conservative gender relations was also managed through the use of interpretative repertoires that we have grouped under the theme of virtuous repertoires. With virtuous repertoires, the women justified their entrepreneurial activities by highlighting four admirable qualities that were rooted in the local culture (i.e. tenacious, financially autonomous, productive, and collaborative).
The majority of villagers were struggling to make ends meet due to the lack of opportunities in the declining village's economy. The women were open about their financial struggles with little reference to their husband's irregular and/ or meagre income: In the beginning, when one has nothing -there must be something to do. Want this, do not have money; want that, do not have money. Anxious. I Have many children but I cannot do anything. The little money was only for food. -Ishwari Although the women did not dispute their husband's responsibility to provide for the family, the women did not emphasize their right to be provided for but rather painted themselves as pragmatic problem-solvers in turbulent times. Such discourse was in line with the local culture which favours tenacity and perseverance over resignation when facing life's challenges. Accordingly, as illustrated by Lova, the women employed the repertoire we termed 'tenacious' as they highlighted the various struggles they faced due to poverty (e.g. insufficient funds for children's education and daily consumption, lack of emergency savings) and their commitment to overcome the problems through their entrepreneurial activities: For me, as a woman, yes I understand that according to religion the one earning an income is the husband. But my condition is actually lacking . . . The husband's [income] is not enough for the family. It is better than being at home not doing anything, it is better to help the husband. Yes, and for exercise, for activities. -Lova Furthermore, the local culture encouraged individuals to cope with adversities without having to rely heavily on others, in spite of broader Javanese culture being collectivistic, and expecting women to be reliant on men (Adamson 2007). In light of their financial struggles, the women advocated for financial independence through the use of the repertoire we termed 'financially autonomous'. As illustrated by Kencana, the women used the 'financially autonomous' repertoire to describe the inconvenience of being overly reliant on their husbands and the satisfaction resulting from having a personal income from their entrepreneurial activities: [With the business] I am not dependent on others as my mother taught me from a long time ago to never be dependent on others, I have to be independent . . . The difference with having my own business is that if I want to buy something, I do not have to ask the husband for money, so I can do it myself. It is so different. In the past, if I wanted to buy this and that, I had to tell the husband, 'Are you willing to buy it? Can you buy it?' Now after having my own business . . . it is up to me when I want to buy something. -Kencana Furthermore, the 'financially autonomous' repertoire was presented as a way for women to lighten their husband's financial burden: Both the 'tenacious' and the 'financially autonomous' repertoires were centered on the urgency of overcoming financial struggle while presenting entrepreneurship as the solution to the problem of poverty. The focus on their financial struggles also diverted attention away from the possibility of women's entrepreneurial income resulting in increased independence and unwanted shifts in the established gender order.
Additionally, the women avoided drawing attention on whether their dependence on the husband was weakened through entrepreneurship by highlighting how idleness, a specific quality which was strongly frowned upon in the local culture, was something they were trying to avoid. They concurred that a woman should not sit around 'doing nothing' but should instead fill her time with meaningful and constructive activities. For instance, Fajarina stated that, 'as much as possible, a woman should not be idle' and Ginanita maintained that entrepreneurship is allowed 'if our intention is to help the husband, to have an activity rather than being idle, gossiping, then it is allowed'. The women argued that household chores could be completed in the morning while childcare could be juggled alongside other domestic activities. Consequently, they were left with a considerable portion of free time and saw entrepreneurship as a virtuous activity with which to fill this idle time. Like Kamala, the women drew on the repertoire, which we termed 'productive', as they emphasized their dislike of inactivity and their view of entrepreneurship as a productive and positive activity to fill their free time with: With the 'productive' repertoire, women's shifting role from purely domestic to progressively entrepreneurial was downplayed. Similar to the 'tenacious' and 'financially autonomous' repertoires, the 'productive' repertoire presented entrepreneurship as the solution to the problem at hand, which in this case was inactivity after fulfiling their domestic responsibilities.
We observed a final notable virtuous repertoire. Naturally, the women had to reallocate portions of their time and energy for operating their microbusinesses. Even with diligent time managementfor instance doing food preparations at dawn or late in the evening, shopping for supplies and opening the business during school hours -they at times found themselves unable to complete household chores or felt drained due to the mounting tasks. The women revealed that their husbands were willing to lighten their burden by aiding them in performing some domestic or business tasks. Although the women were proud of their husbands' assistance, they did not suggest that some of their domestic responsibilities were taken over by the husband. Instead, the women employed the repertoire we termed 'collaborative' as they drew on the local value of cooperation and helping others in the family and the community. As illustrated by Ayudisa, the central part of the 'collaborative' repertoire was the husband being understanding of her busy schedule and his willingness to do domestic chores in her stead: Well, when he is at home, when I am busy, he understands. He will be the one cooking. Yes, he only does not want to do laundry, he only washes his own clothes. Well, if I am [busy], he does everything. He is the one at home, well, sweeping the floor, everything. He is understanding. -Ayudisa The cooperative relationship also meant that some husbands were understanding when the women were unable to fulfil their principal duty of serving the husband. Hence, as illustrated with Ika's example below, the women employed the 'collaborative' repertoire to emphasize that the husband himself withdrew his right to be served and encouraged the woman to continue operating their business tasks: Well, actually, according to religious teachings, when the husband comes home, [we] have to make him a drink, have to prepare some food. But not my husband. For instance, [he will say] 'you are still working on that, it is okay, just work on it. I can make my own food. Take it easy'. -Ika The cooperative relationship was extended to the business as well. Some husbands were willing to assist their wife with minor tasks related to the business (e.g. shopping for supplies, tending the shop, helping with food preparations) when she had to complete some household chores or to attend community gatherings. For example, Limar employed the 'collaborative' repertoire to accentuate her husband's willingness to help her with minor business tasks whenever she was occupied with her children or other responsibilities: The 'collaborative' repertoire therefore drew focus on being cooperative, which is praiseworthy according to local values, while deflecting attention from the incremental change in marital gender relations due to women's entrepreneurial activities.
Taken together, the four virtuous repertoires -tenacious, financially autonomous, productive and collaborative -downplayed the shifting gender relations due to women's entrepreneurial engagement. In contrast with compliant repertoires, the virtuous repertoires were grounded on local cultural values rather than religious teachings. Due to this, virtuous repertoires were as difficult to contest as compliant repertoires since the respective cultural values were considered admirable for both male and female members of the community.

Use of types of interpretative repertoires in women entrepreneurs' account
While our main objective in this study was to identify the types of interpretative repertoires that women use to justify their entrepreneurial activity, an additional relevant finding that emerged from our analysis relates to the ways that these repertoires were combined. We stumbled upon this insight when we looked for differences in the use of interpretative repertoires among the women in our sample. Rather than differences, what we observed was great consistency, with most women employing various of the aforementioned interpretative repertoires in parallel when talking about their entrepreneurial activities (see Table 1). Such a combination of repertoires with varying degrees of compliance to conservative gender relations is important to note, as the sequence and cooccurrence of interpretative repertoires is likely a strategic choice in discourse, rather than a coincidence, and equally important to the choice of types of repertoires to employ (Potter and Wetherell 2010). Specifically, we observed that the women employed the virtuous repertoires in the following way: they first paid tribute to conservative gender relations, and then used a compliant repertoire alongside a virtuous repertoire to progressively deflect attention away from constraints and towards justifications that were related to undisputed values. Danastri, for instance, was faced with the misalignment between women's entrepreneurship and the conservative positioning of husbands as financial providers. In an effort to reconcile these contrasting dimensions, she recounted: Yes, actually, the husband's responsibility is to provide for the wife. Maybe for me, yes, it is true, but, if I do nothing at home, not working or anything, it feels uncomfortable to only wait for the husband. Spacing out. What to do? Women only cook, do laundry, mop the floor, that is all. Then, what else to do? Then I tried to talk to the husband, 'let's do like this', [he replied] 'okay then, let's try'. -Danastri It is notable that after acknowledging the husband's role, she employed one of the virtuous repertoires (i.e. 'productive') as the core of her argument and presented the discomfort of idleness as the justification for her entrepreneurial activity. The 'productive' repertoire was combined with a compliant repertoire -'permission makes activity possible' -which highlighted her submission to the husband's leadership through the process of acquiring his permission for starting a business. The same pattern can be seen in Jayanti's account on women's right to entrepreneurship: Yes, it is allowed as long as we ask for the husband's permission. For instance, 'I want [to do] this, can I? To help out, to help out'. For instance, for soap, we do not have to ask the husband all the time. We need something, want this, have to go to a social event . . . We cannot ask all the time . . . As women sometimes we feel discomfort towards the husband. So, if [the woman is] capable [and] as long as the husband allows, then it is okay. We can have a business. -Jayanti In this quote, Jayanti leveraged the virtuous repertoire of 'financially autonomous' as the centre of her account as she talked about the freedom of having her own money. That repertoire was combined with the compliant repertoire of 'permission makes activity possible' with which she emphasized the importance of acquiring the husband's permission prior to starting a business. Through the 'permission makes activity possible' repertoire, Jayanti paid tribute to the expectations of the conservative gender order in terms of women's obedience to their husband, while presenting entrepreneurship as the outcome of her husband's decision. These and other examples indicated that women entrepreneurs were astute and strategic not only in their selection of appropriate interpretative repertoires, but also in their combination alongside each other, when discussing and justifying their capacity to work as entrepreneurs in the local context.

Discussion
Studies at the intersection of gender and religion within entrepreneurship in Islamic contexts reveal a range of interpretative repertoires that women employ to justify their entrepreneurial motivation and activity against restrictive cultural prescriptions (e.g. Roomi 2013;Tlaiss 2015aTlaiss , 2015b. The variation depends to a great extent on the social class and Islamic context to which the women belong, which determines the stringency of the constraints imposed on them and the types of linguistic resources available to them to navigate these constraints. Studies on women entrepreneurs of middle to upper social class in conservative and moderate Islamic contexts (Althalathini, Al-Dajani, and Apostolopoulos 2021;Essers et al. 2009;Roomi 2013;Sakai 2010;Sakai and Fauzia 2016;Tlaiss 2015a;McAdam 2021a, 2021b) demonstrate that women entrepreneurs are wellequipped to use a feminist interpretation of Islam to downplay patriarchal constraints, such as mobility restrictions, gender segregations, and social stigma towards working women. These women are therefore able to employ more radical yet sophisticated interpretative repertoires to cast their entrepreneurship as largely aligned with Islam. In contrast, lower social class women entrepreneurs in conservative Islamic contexts are generally unable to employ a feminist interpretation of Islam, focusing instead on the cultural boundaries of women's activities. The resulting interpretative repertoires are either highly accommodating (Al-Dajani and Marlow 2010) or provocative (Ritchie 2016) towards the established order; hence, sharpening the misalignment between women's entrepreneurship and Islam. Our study complements this emerging stream of research as we study an overlooked yet populous group of women entrepreneurs of lower social class in a moderate Islamic context. We discuss the theoretical implications of our findings in the following sections.

Interpretative repertoires' variation as gender intersects with religion and class
Consistent with our observation of the interpretative repertoires employed by women entrepreneurs of lower social class in conservative Islamic contexts (Al-Dajani and Marlow 2010; Ritchie 2016), the repertoires employed by women in our study emphasized adherence to conservative gender relations without any reference to feminist interpretations of Islam. Yet, in contrast to lower social class women entrepreneurs in conservative contexts (Al-Dajani and Marlow 2010; Ritchie 2016), our findings suggest that lower social class women entrepreneurs in moderate Islamic contexts typically avoid repertoires that signal either full compliance or full challenge of conservative gender relations. Instead, they deftly navigate them to justify their right to engage in entrepreneurship. The women in our study employed interpretative repertoires (i.e. 'wife's income as supplementary' and 'permission makes activity possible'), which are widely accepted by the local community as they exhibit compliance to the status quo, to pay tribute to conservative gender relations and signal that they will not threaten men's position as the leaders and providers of the family (Al-Dajani and Marlow 2013). In that way, they position entrepreneurship as aligned with the existing gender order (Jennings, Devereaux Jennings, and Sharifian 2016;Verduijn et al. 2014) as they cast their business engagement as a part of performing their role of devoted and helpful wives. Such compliant repertoires have in the past been found to be useful to women of middle/ upper social class (Sakai 2019;Sakai and Fauzia 2016). By noting these similarities and differences, our study highlights that some of previously reported interpretative repertoires can be of relevance to lower social class women entrepreneurs in moderate Islamic contexts.
Furthermore, our findings extend prior literature through the identification of a new set of interpretative repertoires that we termed virtuous repertoires, which women of lower social class in moderate Islamic contexts draw on to justify their engagement in entrepreneurship. Women entrepreneurs in such contexts might not have the requisite education and sophistication to interpret scriptures in self-serving and self-justifying ways (e.g. Althalathini, Al-Dajani, and Apostolopoulos 2021;Tlaiss and McAdam 2021b). In the absence of such an option, they turn to locally available, favoured qualities celebrated in the local culture that apply to both women and men. In our case the qualities are tenacity, financial autonomy, productivity and collaboration; but in other contexts, the specific values may vary. What is important from a theoretical point of view is that local values that are difficult to contest can constitute a very useful resource for women entrepreneurs of lower social class who lack other discursive options (Goss et al. 2011). These values enable them to avoid a direct confrontation with gatekeepers of conservative gender relations (Al-Dajani and Marlow 2013), which could possibly severely damage their legitimacy and that of their businesses (Jennings, Devereaux Jennings, and Sharifian 2016). Instead, they enable them to draw attention away from confrontation and towards cultural dimensions on which consensus exists between women and men. Our study cannot definitively conclude why this group of repertoires might be less prevalent for women entrepreneurs that belong to higher social classes; yet, we speculate that this is because the privileges that higher social class women enjoy weaken the importance and ability to use hard work or financial need as the justification for their activities (Khurshid 2015). For instance, for women entrepreneurs of higher social class whose family's financial stability is not dependent on their earnings from entrepreneurship (Tlaiss 2015a(Tlaiss , 2015b, virtuous repertoires could be less convincing or acceptable. Likewise, it is possibly less likely for women entrepreneurs belonging to the middle/ upper class to be at the receiving end of domestic help from their husbands, due to the weaker perceived benefits from their entrepreneurial income and the greater availability of hired help to assist them with domestic responsibilities (Sakai 2019;Tlaiss 2015a). It is therefore likely that virtuous repertoires can be highly useful to lower social class women entrepreneurs in contexts of moderate Islam, but less so to women entrepreneurs of higher social class.
A third issue that merits discussion vis-à-vis prior literature is the concurrent use of virtuous alongside compliant repertoires, after women entrepreneurs pay symbolic tribute to conservative gender relations. This emergent finding sheds light on an aspect that has not been examined in prior literature and which is notable, given that the sequencing and combination of repertoires can be equally important to the selection of the repertoires themselves (Potter and Wetherell 2010). The combination we detected allows for a delicate balance to be preserved when justifying the activities of women entrepreneurs in moderate Islamic contexts; when they use it, they are seen as neither overly challenging nor overly complying to the status quo. By paying tribute to conservative gender relations first, these women entrepreneurs try to pre-empt criticism that they are not fulfiling their roles mandated by local religious and cultural prescriptions. Nevertheless, right after this tribute has been paid to religious and cultural prescriptions, the women carefully weaken their iron cage by emphasizing the compatibility of their activity with those prescriptions and diverting attention towards alternative values that are non-religious and gender-neutral. By revealing the microprocess of recombining repertoires of varying degrees of adherence to conservative gender relations, we add to literature that has focused on the ways that incremental, dialectical change is enacted through women's entrepreneurship (e.g. Al-Dajani and Marlow 2013; Alkhaled 2021; Verduijn and Essers 2013). Without such a process, the use of virtuous repertoires on its own might not be sufficiently convincing or might be perceived as erring towards a disrespect towards religious and cultural prescriptions.

Conclusion
In this study, we focus on the role of discourse as women in an Islamic context navigate restrictive patriarchal structures and practices to cast entrepreneurship as a legitimate activity for themselves. By focusing on an understudied group of women entrepreneurs who are of lower social class and live in a context of moderate Islam, we offer three contributions to prior literature on the intersection of gender and Islam within entrepreneurship (e.g. Essers et al. 2009;Tlaiss 2015b;Tlaiss and McAdam 2021b). First, we extend findings of prior studies by revealing whether interpretative repertoires that women entrepreneurs use in other contexts (e.g. more affluent, more conservative) are applicable for women entrepreneurs in this group. We find that the social circumstance of Muslim women entrepreneurs in our context can impede them from accessing feminist interpretations of Islam to position entrepreneurship as aligned with Islam (e.g. Althalathini, Al-Dajani, and Apostolopoulos 2021; Tlaiss 2015a; McAdam 2021a, 2021b); yet, these women are able to draw on compliant repertoires that downplay the role of women's entrepreneurship to make it more palatable to the status quo, as has been reported in moderate Islamic contexts (Sakai 2019;Sakai and Fauzia 2016). Second, we identify a set of interpretative repertoires -virtuous repertoires -that seems to be novel and unique to a setting of lower social class women entrepreneurs in a moderate Islamic context. These repertoires, rather than drawing on or reinterpreting religious values, focus instead on local non-religious and gender-neutral cultural values in order to help women entrepreneurs justify their entrepreneurial activities. Third, we highlight that the identification of interpretative repertoires that incrementally challenge prevalent gender relations can be further informed by an understanding of their use alongside compliance to conservative gender relations. The combination of these discursive elements allow women entrepreneurs to perform a balancing act between adhering to local religious and cultural prescriptions and subtly yet strategically challenging these, which unfortunately seems not to be an option for Muslim women of lower social class in more conservative Islamic contexts (Al-Dajani and Marlow 2010; Ritchie 2016).
These findings also have relevance for policymakers and organizations that aim to promote women's entrepreneurship and self-sufficiency, specifically among the poor, in Indonesia and other regions that ascribe to moderate interpretations of Islam. The World Bank's (2016) report on women's entrepreneurship in Indonesia, for example, promotes the elimination of extreme poverty through tailored support for necessity-driven women entrepreneurs, but without any reference to building a positive discourse on women's entrepreneurship. Accordingly, our findings call for solutions that integrate a holistic understanding of the relationship between discourse and local religious and cultural values, which can complement the tailoring of financial services and capacity building schemes. We have seen that even women who manage to become entrepreneurs in these contexts and have access to supportive financial services are to an extent bound by religious and cultural prescriptions, and need to actively and strategically navigate them to present their entrepreneurial activity in a way that is acceptable to their family and community. Policymakers and organizations aiming to help these women entrepreneurs can tailor the communication and framing of their programmes in ways that mimic the skilful use of repertoires by the women in our study. They can proactively but symbolically 'pay tribute' to conservative gender relations to deter direct opposition, while quickly shifting the focus on compliant repertoires and those that draw on non-religious, gender-neutral cultural values that they have identified as appropriate to each region. If such discourse is incorporated into their programmes, it could accelerate the process of local women entrepreneurs convincing their family and community that entrepreneurship is legitimate and non-threatening to the established order, thus opening up the way for greater changes over time.

Limitations and future research
Our study has certain limitations. While it has shown how certain types of interpretative repertoires can help women in Islamic contexts justify their entrepreneurial activities against restrictive cultural prescriptions, we cannot establish conclusions on the individual factors (e.g. age, education level) influencing the types of repertoires used or their consequences on business outcomes (e.g. survival, sales or profitability of businesses). For reference, we present in Table A3 of the Appendix the occurrence of repertoires across women of different age groups and education levels. The majority of the women employed one of the compliant repertoires and one of the virtuous repertoires, yet any variations within the sample should be interpreted with caution due to the nature of the sample (purposeful rather than random) and resulting dataset. Future research that examines these relationships from a quantitative point of view could therefore add to our understanding of both the antecedents and the outcomes of any variations in repertoire use. While this is a challenging goal in this context, methods such as longitudinal diaries could potentially capture causality between individual and other factors and repertoire use, as well as between repertoire use and business outcomes.
This study also limits itself to data collected in one local context in Indonesia, where Islam is very influential for the lives of local women. Nevertheless, Islamic interpretation and its stringency towards gender relations can vary from context to context. Future research can therefore corroborate our results in similar contexts or establish differences between our findings and those generated in other Islamic contexts. Additionally, in this study we mainly focus on women who have already managed to operate a business, as we had less access to women who are explicitly not allowed to become entrepreneurs. Therefore, while we account for the ways that interpretative repertoires can be leveraged to justify business engagement for some women in an Islamic context, we refrain from suggesting that such justification is possible for all local women. We speculate that for those women who find themselves under the dominance of a particularly conservative patriarch, even subtle skilful employment of compliant and virtuous repertoires might be an extremely risky strategy.
A productive yet particularly challenging direction for future research could be to find access also to those women who are explicitly not allowed any business activity and compare their discourse to those who have managed to make the leap to business ownership. Overall, we hope that the addition of our study to the emerging body of work on women's entrepreneurship in Islamic contexts will provide impetus for further work on how women can reap positive outcomes from aligning their Islamic identity and their right to entrepreneurship. Notes 1. To proxy social class, we considered descriptions of education attainment, level of business sophistication, and number of employees in the women's businesses. We follow Qibtiyah (2018) to assess the type of gender-based Islamic interpretation: conservative Islam, at one end of the spectrum, tends towards a literal interpretation of scriptures and a complete rejection of feminist ideas, while moderate Islam at the other end allows for contextual, interdisciplinary, and equitable understandings of scriptures. 2. We defined women entrepreneurs as self-employed women who start, manage and own one or multiple businesses (Tlaiss and McAdam 2021b).