Entangled ethics: Piety and agency in Turkey

Submission and resistance

Western theories of liberal selfhood, especially those based on Kantian ethics (discussed below), have presumed that human agency is intrinsically resistant to constraints on freedom because these constraints are thought to inhibit full self-development and the realization of individual autonomy. That is, Western liberal thought tends to identify an individual’s agency with acts that resist domination and go against social norms rather than acts that comply with social norms and disciplinary frameworks (Mahmood, 2005: 5). This assumption of ‘agency as resistance’, Mahmood argues, is especially prominent in Western feminist thought. Feminist theorists and anthropologists, such as Judith Butler and Janice Boddy – as well as theorists of subaltern studies, such as Gayatri Spivak – take for granted the idea that the female actor who operates in an arena of gender oppression would naturally attempt to subvert or re-interpret the norms that undergird that oppressive order. To demonstrate this subtle bias in feminist theory, I will discuss two cases that Mahmood herself critiques in her work.

In her study of gender roles in Hofriyat (Sudanese) culture, Janice Boddy (1988, 1989) argues that Hofriyat women’s lives are ‘over-determined’. That is, through the entire complex of associations and practices that constitute Hofriyat femininity, a woman’s body is transformed into a living vessel of her culture’s moral values, which reinforces the identification of the individual woman with her roles as wife and mother in Hofriyat culture (Boddy, 1988: 16). Women’s identities, conceptualized both by women themselves and by the men in their lives, are almost completely bound up with the culture’s expectations of women. The feminine role is thus highly rigid and is seen as naturally static rather than as a creative process actively engaged in the world and sometimes transformed by it. In such a gender schema, infertile women or women prone to miscarriage – i.e. women who do not fulfill the over-determined feminine role of women as mothers – may fail to attain full personhood or full female status in Hofriyat society.

One way in which Hofriyat women deal with the constraints of gender roles is through the practice of spirit possession and trancing that are the centerpiece of the zar cults common in East Africa. While the trancing activities of the zar cults as practiced in Hofriyat culture serve many therapeutic functions, as Boddy makes clear, the trancing permits the spirit-possessed woman to become something other than her socially defined role (Boddy, 1988: 18–19). Through trance the woman is divested of her feminine self with its constraints and moral limitations, thus allowing her to test the limits of gender roles that are otherwise deemed natural and immutable. Everyday assumptions and categories are questioned by the possessing spirit during zar trance, but because this process is seen as coming from outside the trancer, her social place and selfhood are protected. It is the spirit that tests the limits, not the woman herself. The women thus seem to welcome the spirits since the women’s participation in the zar cults allows for improved health and happiness. This follows the feminist assumption that ‘oppressive’ notions of femininity at work in a culture would engender, through some culturally recognized way, avenues by which to disrupt the gender domination and provide arenas for resistance (Boddy, 1989: 345).

Judith Butler has also tended to attribute greater validity to a notion of agency as naturally subversive to oppressive sexual and gender norms, as Mahmood notes. Influenced by the work of Michel Foucault, Butler argues that the capacity of the individual subject to resist norms arises from workings of power itself and does not reside within the individual actor. That is, Butler accepts Foucault’s notion that ‘the set of capacities inhering in a subject – the abilities that define its modes of agency – are not the residue of an undominated self that existed prior to the operations of power but are themselves the product of those operations’ (Mahmood, 2001: 210). The actions of an individual agent (actions that may be resistant or submissive) are embedded in relations of domination, and therefore are made possible by specific relations of subordination; they do not emerge from some sort of a transcendent or autonomous liberal subject. While Mahmood agrees with Butler’s formulation thus far, she finds problematic Butler’s recurrent suggestion that acts of resistance are largely definitive of agency itself. Certainly, Butler has acknowledged that agency is not always rebellious against systems of power, but her actual theorizations of agency tend to focus exclusively on those acts that subvert norms or attempt to destabilize and rearticulate dominant normative structures (Butler, 1999; see especially Butler, 1993; Mahmood, 2005: 20–22; Mahmood, 2001: 210–12).

Embedded in the works of both Boddy and Butler (and of others) is the idea that oppressive norms are somehow imposed on the oppressed subject from the outside rather than playing a role in the construction of the individual agent. Even though Butler gives a nod to the notion that any actor is embedded in normative systems, she and other feminist theorists still operate on the assumption that the liberal agent is one who exists outside of and is ultimately free from dominant normative structures and can therefore resist structural limits on freedom. Such a position ignores the extent to which normative structures constitute individual subjectivity and give form to an individual’s desires, aspirations, emotions, reasoning processes and bodily dispositions (Mahmood, 2005: 19–23). In reality, there is no reason to presume that the human agent innately seeks freedom from structures of normativity, since the development of selfhood (creation of agency) is shaped by the context in which subjects function. Human agents, even those we might see as ‘oppressed’ according to Western feminist standards, may actively cultivate, through a variety of mechanisms, a compliance with the normative structures within which they are embedded.

Mahmood illustrates her argument with an ethnographic study of the women’s mosque movement in Egypt, noting how the participants in this piety movement promoted practices of female modesty and humility that they believed accord with Muslim requirements of submission to the will of God. For example, members of the mosque movement pursued disciplinary activities, such as regular prayers (salah), veiling, certain marriage practices, and submissive bodily dispositions that were, in the view of the practitioners, embodied ideals of feminine piety. These activities were also a way in which Mahmood’s informants attempted to nurture certain emotional states, such as shyness (al-haya), patience (sabr), and fear (khawf), in order to cultivate a sensibility of submission to God’s will. That is, the women engaged in these pious acts as a way to construct an appropriate subjectivity marked by modesty and deference, with the idea that ‘practicing piety’ through repetitive training creates a self (body, emotions, and reason) in which ‘religious virtues acquired the status of embodied habits’ (Mahmood, 2001: 212). Thus, ritual acts are not only symbolic of one’s inner disposition and emotional state but are pragmatic actions that achieve certain ends in the formation of the properly Muslim self. ¹

The type of ethics that informs the pietistic groups in Egypt diverges from the Kantian ethics that undergirds much Western liberal thought. Emmanuel Kant contended that thought or reason precedes and gives rise to action, so that ethical action is derivative of ethical thought, no matter what the actor’s habits and inclinations may be. But Mahmood (2005: 26–31, 118–52) argues that the relationship between ethical action and reasoning should be understood in more dynamic terms. In many cases moral reasoning can give rise to ethical action, but ethical action can also cultivate moral dispositions in practitioners. ² Just as actual practice (action) can produce the disposition and ability to be a good athlete, musician or writer, so can the practice of ethical behavior create ethical dispositions. In the case of the women’s pietistic groups, the actors consciously submitted themselves to certain types of ethical actions with the goal of fostering a properly pious subjectivity, which would then, in a self-reinforcing process, promote further ethical behavior. In Mahmood’s study, human agents are willingly submissive to religious norms that emphasize humility and modesty, and they seek to enhance their submissiveness, not ‘break free’ from and exist external to those normative structures.

When submission is subversive: Contradictory demands of religion and state

Mahmood is certainly right to critique the narrowness and cognitive bias of Western liberal philosophy and its assumptions of agency as resistance. But I suggest that we could shift focus a bit and consider the fact that many people in modern societies exist in contexts where multiple, layered normative systems function simultaneously and make contradictory demands on the individual actor. What in one normative context may be an act of obedience to the ethical expectations at play may in another context be an act of resistance where the individual consciously defies certain social expectations. Furthermore, most people live in complex societies where some normative systems are more compelling or are enforced with more authority and power than other ethical systems. While religion and family may make one (or more) set of moral demands on an individual woman, the community might make another, while the state may attempt to enforce a third set of demands. Contestations about gender roles and norms are thus configured according to (or against) a range of ethical expectations, which often means that an individual’s cultivation of piety – or creation of desired selfhood – can involve submission and resistance at the same time and in the same act.

Such an ethical conundrum was demonstrated to me in the context of research I conducted among women Koran students in Sincan, a famously religiously conservative city in Turkey. Koran courses in Turkey are markedly different from the women’s mosque movement in Egypt. In Turkey, religious services and education are controlled by the state, so that anyone (especially women) who wanted to study Koran beyond the state mandated curriculum might be drawn to small, private Koran study courses organized locally by religiously reputable leaders rather than to large mosque-based meetings. ³ These courses may promote a strict, conservative type of Islam that focuses largely on issues of proper practice, while others may orient more toward mystical concerns. The Sincan Koran course, which became the centerpiece of my field research in Turkey from 1997 through 1999 – with subsequent research visits in 2006, 2007, and 2010 – was led by an intelligent young woman, Meryem, ⁴ who promoted a relatively moderate interpretation of Islam. Meryem had developed a neighborhood reputation as a skilled preacher and had a following among women in her neighborhood. She had in fact worked as a preacher at a local religious radio station before I knew her. The course she taught was designed for like-minded women in the neighborhood who for various reasons chose not to participate in official – i.e. state-run – Koran courses. The course participants came from varied educational and economic situations: some students only had rudimentary formal education, some had graduated from primary school, some from high school, and occasionally a small number of college students took part as well. The group would gather two hours a day, five days a week, to study Arabic, read the Koran and listen to a sermon given by Meryem based on the day’s Koran passage. After the formal lesson, students would often socialize for a while longer to drink tea and discuss current events.

An abiding theme of these discussions was the problem of secularism in Turkey, especially the way in which secularist policies and discourse complicated and impeded the students’ abilities to develop a lifestyle that they saw as appropriately pious. Since the founding of the current Turkish republic (and even before), secularization of civic institutions and the public sphere has been a fundamental component of Turkey’s modernization project. The Turkish state officially secularized many aspects of Turkish society – from education and governance to the organization of time and space – over a relatively short period of time, especially during the 1920s and 1930s. The Turkish state, and not the elected governments, has been the real locus of power and driving force of secularization throughout the history of the Republic. As Jenny White explained (2013: 33), ‘The state (and its institutions, like the military, judiciary, and education administration) presents itself as the guardian of the laicist system and secular national identity.’ Since the founding of the Turkish Republic, the state has, for example, sought to implement and enforce the Kemalist ⁵ reforms, as they are called. These state-led reforms strove not only to remove religious influence from the political arena and place state controls on the public expression of religion, but the reforms also emphasized the reorganization and the ‘de-Islamization’ of patterns of daily life and gender roles and relations in Turkish society. Turkish people were encouraged (and sometimes forced ⁶ ) to dress like Westerners, eat like Westerners (sit at tables and use utensils), be educated like Westerners, and establish households that allowed for more open gender relationships. The most striking aspect of the Kemalist reforms is how the dispositions, practices, and lifestyles of ordinary Turkish citizens came under the scrutiny of the state, so much so that the places and appearances of the bodies of the Turkish people became the ultimate signifiers of the success or failure of the Kemalist westernization efforts.

These westernizing policies indicate that the state did not adopt a version of secularism based on the idea of separation of religion and state, but rather assumed something closer to French-style laicité, where the state maintains some control over religious practice. In the case of Turkey, in fact, the legal and governmental controls on religion are quite formidable. Mosques are administered by the government under the administration of the Directorate of Religious Affairs (Diyanet), which answers directly to the prime ministry. The Diyanet is responsible not only for the training and hiring of all imams and hatips (preachers), but also controls the content of sermons delivered during Friday services. The Diyanet oversees religious instruction in the schools and mosque-based Koran courses, while independent and private Islamic instruction, such as the Sincan Koran course, is explicitly forbidden. The state also outlawed all Sufi orders and, through the Council of Higher Education, has until recently prohibited the wearing of Muslim headscarves at most schools and public universities (see, for example, Kuru, 2009: 161–8). In sum, the Turkish state – in almost theocratic fashion – has exercised the power to enforce its form of Islam over and against the wishes and popular traditional practices of many Turkish citizens.

One of the most consistent ways that the Kemalist state has tried to control religion and make it safe for laicism has been to insist by law, military action, and rhetoric that religion should remain separate from politics. ‘Exploiting’ religion for political purposes has been legally prohibited in Article 24 of the 1982 Turkish constitution, ⁷ though what constitutes the exploitation of religion is left open to interpretation. Ruling political parties that have overtly accommodated the desires of religious constituents have been subject to military coups – in 1960 and in 1997 ⁸ – and to forced closings by the constitutional court. One of the most consistent complaints that I have heard from Kemalist interlocutors against religiously sympathetic political parties and actors is that they have committed the supposed sin of mixing religion and politics. Such an illegitimate intermingling of social spheres – the ‘religious’ and the ‘political’ – was treated as self-evidently criminal and dangerous.

But as my pious interlocutors in Sincan knew very well, religious action cannot always be clearly separated from politics in a secular society, despite claims to the contrary. For example, when the Koran course participants try to wear the headscarf in public institutions – especially in schools – this is both a political and religious act. It is political because the action violates certain rules established by the state to include some members of society (those who uphold the secularist order) and exclude others (those who appear to critique the secularist order by means of the display of a religious symbol). And it is a religious act in that the covering of the head has important religious significance for the bearers, even if they may wear the Muslim head covering for a variety of reasons, such as tradition, family pressure, or rebellion. Mahmood also notes how politics becomes embedded in everyday ethical religious practices, especially when the institutions of religion are subsumed under state power. As she states (2005: 193–4), ‘it is not that the pietists have “politicized” the spiritual domain… but that conditions of secular-liberal modernity are such that for any world-making project (spiritual or otherwise) to succeed and be effective, it must engage with the all-encompassing institutions and structures of modern governance, whether it aspires to state power or not.’

Furthermore, the creation of a secular society does not require the elimination of religion per se, but rather secularism requires that religious acts are relegated to particular institutions and spheres of interaction. But such a partitioning of space to separate out the religious from the secular relies on the notion that religious acts may be discarded at will and without harm to religion itself. This assumption privileges a specifically Protestant worldview in which the interior development of faith or belief is deemed the essential dimension of religiosity. On the other hand, acts and rituals that have often been central to traditional Catholicism, Judaism, and Islam are subsidiary to faith and thus ultimately dispensable. As Protestant Reformer Martin Luther argued, human sinfulness is so profound that there is basically no ritual or deed a person can do to earn salvation. Rather, it is God’s gift of grace and the individual’s faith that provide the possibility of justification and ultimately salvation. This doctrine of ‘sola fide’ (faith alone) is a signature doctrine of Protestantism, though it was adapted in various ways into other forms of Christianity, as well as into Reformed Judaism. The privileging of the interior, cognitive aspects of religiosity above the social, ritualistic aspects of religion makes it possible to imagine a society in which people may still ‘be religious’ – where being religious means ‘having faith’ – while functioning in a non-religious sphere where religious acts and symbols are policed by laicist regulatory authorities (Shively, 2008).

Certainly there are many devout people of different religious commitments, including Muslims, who would see this secular arrangement as proper and embrace the notion of religion as essentially a private belief. But as in the case of the Koran course students and many others, there may also be those who regard it as a fundamental violation of their religiosity to discard religious action at the borders of secular spheres of interaction. In such cases, the secular public sphere is coercive. It is not an arena where the rational debate of those with competing interests can be carried out to a ‘neutral’ conclusion, as Jürgen Habermas (1989) would have it. It is instead a realm of interaction in which ‘negotiation amounts to the exchange of unequal concessions in situations where the weaker party has no choice’, as articulated by Talal Asad (2003: 6). Until recently, the Kemalist establishment in Turkey has had the power to exclude or extract concessions from practicing Muslims who wish to include certain religious symbols in the public sphere of interaction.

As such, the very act of creating a secular society automatically entails that certain religious acts can become politically and socially disruptive. Such a social reality may be characteristic of secular societies, as Winnifred Fallers Sullivan (2005) has pointed out. But modern secular states are founded in part on the notion of liberal selfhood, where the individual may act freely within reasonable limits. What is especially problematic in the Turkish case, however, is that the laicist state has attempted to limit even popular, widespread religious practices from the public sphere for reasons that large portions of the Turkish population find unreasonable. For example, the practice of veiling ⁹ has been the most often discussed and widely politicized religious issue (though it is certainly not the only one). The early Republican (Kemalist reforms) included an orientalist evaluation of the Muslim women’s headscarf as ‘uncivilized’ and contrary to the westernized society the reformers were trying to engineer. Even though attitudes have changed somewhat in recent years, the identification of veiling with backwardness, ignorance, and lack of development has been common among self-identified Kemalists (see Shively, 2005). And through much of its period of dominance, the Kemalist state has deemed the presence of the modern veil – called tesettür ¹⁰ in Turkish – in public institutions to be a violation of the laicist nature of the state. Women who wore the tesettür (and consider it indispensable to religiosity) were barred from schools, ¹¹ universities, and some other public institutions. ¹² As indicated above, the tesettür thus served as a political and religious symbol, not because of the intentions of the veiled women (whatever those might be and however they might be discerned), but because the states’ efforts to create a ‘secular space’ exclude some symbols and actors but includes others.

People who contradicted the Kemalist configuration of secular space, for instance by wearing the tesettür in certain public institutions, were sometimes subject to state disciplinary action. It is true that punishments for violating the laicist order were inconsistently applied and the appearance of unapproved religious symbols, such as Muslim headscarves, have often appeared in secular public spaces (see Gökarıksel, 2012). Nevertheless, religious actors were aware that state disciplinary action against those who violated the secular public sphere was always a possibility and may have come in the form of public shaming, expulsion, threats, surveillance, incarceration, exile ¹³ or even execution. ¹⁴ In this case, it is a fine line between a theocracy that punishes those who fail to practice religion according to rigid legal standards and a laicist state that punishes those who fail to practice secularism according to its rigid requirements. ¹⁵

This is not to suggest that there have been no conflicts among conscientious Muslims about the issue of women’s head-covering and other expressions of piety. Various religious communities (cemaats) often demand particular types of veiling practices for their female members. Some of the more conservative groups require, for example, that women cover their faces and wear the black chador, while others are satisfied with a simple headscarf and overcoat, with the face exposed. Many young, observant women are rather playful in terms of colors, fabrics, and styles when creating their own, often fashionable forms of tesettür (Gökarıksel and Secor, 2010). Overall, there was considerable debate within the pious community about women’s – and men’s – proper appearance, and this has added another layer of ethical concerns to the process by which religiously observant women cultivate pious selfhood.

However, most conscientious Muslims were united in their opposition to the state-initiated limits on religious headscarves, viewing the limits as fundamentally unjust. During much of the period of my initial field research in 1997–99, women who donned the tesettür were excluded from institutions such as universities and government offices. Headscarved women students who refused to uncover their heads at Istanbul University in 1997 and 1998, for example, were banned from classes and exams, or were not allowed to register without a photo ID in which the student was pictured bareheaded. ¹⁶ In return, these women and their supporters would stage frequent protests demanding the right to education without surrendering their religious beliefs and practices.

The ban on the tesettür and the attitude that many Kemalists took toward veiled women were not abstractions for the young Koran course students who were my research subjects either. Indeed, many of them had also suffered directly as a consequence of their choice to wear Muslim headscarves. Several college students in the course had been barred from classes and exams because of the veil. One such student, Ayşe, had been forced to leave school in the middle of her fourth year in her biology education degree program at a Turkish university. She had been able to get away with covering her head while in class and exams since she attended college in a religiously conservative city. The ban on headscarves in universities depended on officials actually enforcing the rules, and many faculty and administrators looked the other way when covered women appeared in class. But by the time of my field research in the late 1990s, several factors had changed to force Ayşe’s withdrawal.

After the 1997 ‘post-modern coup’ when an Islamic-leaning government was forced out of power by military leaders, the political discourse against headscarves at universities reached such an intensity that even universities that had not enforced the ban before – including Ayşe’s university – felt official pressure to begin policing and expelling women wearing tesettür. During this period of heightened surveillance, Ayşe had actually been questioned by the police, who wanted to determine what subversive Islamist organization was behind her decision to don the tesettür at the university (there was no such organization). At this same time, Ayşe needed to complete her student teaching in order to receive her degree, but wearing tesettür in a public secondary school was completely contrary to educational regulations. She either had to remove her headscarf or discontinue her education. In the end, she had to resign from her degree program in biology education because she refused to uncover. Two other college students who attended the Koran course on a regular basis also were affected by these developments. One, Meyda, was able to immigrate to Australia where she could complete her higher education degree while wearing tesettür. Another, Lale, chose to take off her headscarf in order to complete her degree at an eastern Turkish university but became very depressed because of the stress of appearing in public uncovered – depressed enough that her grades began to slip.

Meryem herself sought to attend an Open High School program in order to get a high school education. While the Open High School was a program that allowed self study (basically a high school equivalency program), proctored exams had to be taken bare-headed, and a student ID required an uncovered photo. In the end, Meryem acquiesced and uncovered in order to have an ID picture taken, but this was a source of great bitterness for her. Those Koran students who did not suffer from discrimination in getting education still joined public protests against the ban on the headscarf in universities on principle, but to no avail. In sum, the veiled students and their supporters insisted on the women’s rights to appear in public as they pleased – that is, they resisted the normative system imposed by the Kemalist state while submitting to the requirements of Islam. In turn, the state branded the students as disruptive and even dangerous, and forced them to remain outside the very educational institutions they sought to engage with.

With the coming to power of the Islamic-leaning Justice and Development Party (JDP) in 2002, many of my pious interlocutors were hopeful that the regulations barring covered women from universities and other public institutions would be eased or lifted. And indeed, the JDP did attempt to introduce a law in 2007 that would allow female university students to wear headscarves on campuses. But supporters of the traditional Kemalist establishment staged sizable demonstrations against the change and the threat to the laicist order it seemed to represent to them. Tensions between Kemalists and supporters of the JDP almost reached a breaking point in 2007, and for reasons beyond the issue of veiling at universities. A national crisis emerged in 2007 when former president (and avowed Kemalist) Ahmet Necdet Sezer ended his tenure as president, and the Turkish parliament put forward an important JDP leader, Abdullah Gül, as a replacement. Gül was not only a longtime member of Islamic-leaning political parties, but one accusation lodged against him was that his wife and daughters wore the tesettür, thus indicating the potentially dangerous Islamic orientation of Gül himself.

Up until 2007, Sezer had acted as a counterweight to the JDP in the national government, but the proposed shift of the presidency to Gül would mean a complete JDP monopoly on government power. In reaction, the opposition parties refused to participate in the election process, bringing parliament to a deadlock, while thousands of Turks who supported the opposition parties took to massive street protests. Many secularists I knew were certain that Gül’s appointment represented a potential Islamist takeover of the Turkish government, and they supported the protests and rallied against the JDP. The situation reached a critical phase when the military posted an e-memorandum (e-muhtıra) on their website, accusing ‘administrative authorities’ of manipulating religion in political discourse and threatening military action if the JDP did not back down. (Some have called this action on the part of the military an attempted ‘e-coup’). Shortly thereafter, JDP retreated from its position on the headscarf question and changed tactics on the election of Gül, who did eventually take the office of president after some complex and contested legal changes. ¹⁷ In was not until 2013 that the ban on veiling in the universities and in civil service jobs was finally lifted, though women in the military or judiciary are still prohibited from wearing the Islamic head-covering.

Yet just like Mahmood’s research subjects, my informants also insisted that wearing the Muslim headscarf was an appropriate submission to the requirements of female pious dress and comportment laid out in religious texts, despite the problems veiling created for them before the prohibitions were relaxed. By following the requirements of Islam, the women were striving to make themselves into what they considered to be proper Muslims. ¹⁸ In so doing, they were making a choice between two different – and in this case competing – authorities: the state or God. If they wanted higher education, they would have to submit to the state but then violate the will of God as they understood it. If they submitted to God’s will, then the state would block the path to advanced education. These women chose to submit to God by veiling and attempted to resist the demands of the state. They interpreted their struggle in Islamic terms. In Meryem’s sermons and in informal discussion, the Kemalist state was frequently equated with the Pharaoh (Firavun) of Egypt who demanded submission from Moses and his followers over and against the will of God. In the end, the followers of God’s will were triumphant and the Pharaoh was destroyed. The community of Moses thus served as an ethical model for Muslims for dealing with an ungodly regime.

In many ways, the aspirations of the women in the Egyptian movement match those of the participants in the Sincan Koran course I studied. Both groups of women extolled the virtues of feminine obedience and modesty, promoting such forms of deference such as wearing Muslim head-coverings and submitting to male leadership in mosques, while developing a deeper understanding of the requirements of devotion through Koran study and ritual participation. Both groups also maintained that modest bodily dispositions and submission to the laws of God can cultivate and give rise to properly devout sensibilities. As Meryem argued in one sermon, ‘Following [the rules of the Koran] causes one to think, to think about one’s own role in relationship to God, what one should do.’ Similarly, a young student, Sema, explained to me, ‘An apple rots faster without its skin.’ She went on to clarify that without the constraints of the headscarf and other reminders of one’s submission to God, a woman is more likely to go morally astray. The veil creates a pious moral disposition as much as piety requires the veil.

Considering the situation for covered women in Mahmood’s discussion of agency and submission, though, the lines between what is submission and what is resistance do not hold in the Turkish case. The politicization of veiling and other religious actions in Turkey means that the very acts that are submissive to Islamic ethical norms are resistant to the Kemalist ethical norms at work in Turkish society, norms to which the women were all exposed on a daily basis. Moreover, the veiled women themselves were acutely aware of this conflict, in that their lives were directly affected by these contradictory demands to cover and uncover.

Conclusion: Turkey and Egypt

Mahmood has rightly argued that we should think of agency as ‘a modality of action’ rather than as ‘a synonym for resistance to social norms’, as is common in Western feminist theory (2005: 157). Mahmood’s conceptualization requires that we consider the relationship between thought and action in dynamic terms: certain actions or disciplines may shape the individual’s subjectivity as much as subjectivity and thought may guide an individual’s action. Mahmood also appropriately criticizes feminist and much Western philosophical assumption that the liberal subject – one that seeks self-realization through autonomy and naturally resisting restraint on that autonomy – is a universal human characteristic. Instead, agency is a socially and culturally mediated capacity to act. It does not spring sui generis from the mind of the individual, but one’s desires, aspirations, and agency are all shaped by the ambient cultural norms and expectations.

This article builds on Mahmood’s analysis by pointing to another factor affecting the processes of self-cultivation and the development of agency: most people are exposed to multiple normative structures, some of which may make competing demands on the individual. Such normative systems may derive from class status, ethnic background and educational level, among other domains of human interaction. In the case of my research informants in Sincan, they found that an individual’s submission to one set of norms – the group’s interpretation of Islam – assumed resistance to another, the requirements of the Kemalist state. Even if these competing demands did not confront the students at all times and places, they still seemed to be always aware of the ‘clash’ and spoke of it frequently and passionately.

Not only is this ethical entanglement true in Turkey, but I would suggest that such contradictions exist in all complex societies. In Egypt, for example, the state has also imposed regulatory practices that may impede an individual’s attempt to cultivate what he or she might consider to be a properly pious self. As Gregory Starrett (1998) has illustrated, the Egyptian government has exerted considerable control over religious education and expressions of religiosity since before the country’s independence. Those Egyptians who disagreed with the government’s control over religious education would have to defy the government for the sake of submitting to the will of God. Until the Arab Spring of 2011, the Egyptian government has placed special emphasis on inculcating the ‘correct’ type of religious knowledge through the educational system – teaching properly moderate Islam that conforms to the needs of the state – with the partial goal of undermining Islamic opposition, most especially the political wing of the Muslim Brotherhood. In general, the Egyptian state has also controlled mosque leadership and such essential normative practices as marriage, access to education, and the right to organize into religious groups – control of religiosity that may very well impede some Egyptians’ abilities to pursue certain religious aspirations. Thus, the Egyptian women discussed in Mahmood’s work must have had to deal with this conflict between (at least) two ethical systems and the regulatory practices they each entail. There may have been times when their efforts to submit to the requirements of Islam resulted simultaneously in resistance to Egyptian state demands, as occurred with the pious, covered women in Turkey.

In this article I have focused on the competing demands of religion and state on the actions of the pious Muslim women. Yet there are other ethical systems and domains of interaction that may also make even further demands on women’s agentive action. Such ethical systems in Egypt might include, for example, class-based gender norms. As Farha Ghannam (2011) has demonstrated, socioeconomic structures in a lower-income Cairo neighborhood shape in often unexpected ways the mobility and embodied practices of the neighborhood’s men and women. Other discursive and social domains that might also play a role in the constitution of selves and bodies of pious women include communities in which the actors interact with ideologically heterogeneous populations. ¹⁹ For my Turkish informants, Sincan was a neighborhood that was relatively accepting of the students’ religiosity, since Sincan itself was comprised of a largely religious and conservative population. But whenever the women moved about in other neighborhoods with a different dominant politics than in Sincan, they would be open to slander and verbal threats and insults. As Meryem complained in a discussion with a group of students:

Wearing the türban [tesettür] in universities and even in ordinary public situations creates a lot of stress for the wearer. In the universities, the women never know if they will be punished, expelled, persecuted, teased, refused entrance, etc. Maybe one day they will encounter no problems, but the next day they might receive some sort of a threat. Women are sometimes yelled at or threatened in public situations. I knew a woman in tesettür who got on a bus. When the young woman sat down, her overcoat brushed the floor of the bus a bit. When that happened, a woman passenger on the bus yelled at the woman, calling her dirty (pis) and other negative things. There is a constant tension that the covered women feel that at some point they may suddenly be attacked or discriminated against in some way.

Furthermore, a popular idea among many Kemalists, including some of my friends and acquaintances, was that women who wore tesettür were paid to do so by shady, presumably fundamentalist organizations. That is, these observers saw the tesettür as insincere and ‘false’ (sahte) rather than a true expression of the wearer’s piety. The covered women were accused of ‘selling out’ to fundamentalist forces – as when Ayşe was questioned by security forces at her university to find out who was behind her efforts to wear the headscarf to class (see above). The implication was that the veiled women were willing to cause trouble (display unauthorized religious symbols) only for personal gain, thus denying the women their claims that they cover for the sake of piety and submission to God. Such accusations and slander surely affected the way the Sincan women interpreted their own efforts to maintain their Muslim piety.

To be sure, the norms and policies constraining the movements of women in tesettür have eased in the last couple years under the sympathetic government of the Justice and Development Party. Nevertheless, there are still conflicts about Islamic veiling and other expressions of piety with Kemalists who still have some sway in the Turkish public sphere. For example, headscarves are still not permitted in high schools (except for the religious high schools) and some other public institutions, and many universities still make it difficult for women to cover their heads while in the classroom or in exams. In pursuing their education, pious women must choose which authority to obey, religion or the state, since they both require submission in order to construct the proper devotee on the one hand, or the proper citizen on the other.

In Egypt as well, Salwa Ismail (2006) has demonstrated that communities in new quarters in Cairo use informal mechanisms to police and control individual members of the community, especially women. Religious practices may be subject to that surveillance and control as well. So, for example, young men in the Cairo neighborhood she studied distrusted the hijab as a sign of the wearer’s morality and humility. Some of Ismail’s male interlocutors were certain that the hijab was a way in which untoward and immodest female behavior could be shielded from the view of critical onlookers, only to be displayed later in other contexts, suggesting such women are double-faced (Ismail, 2006: 108–11). In both the Turkish and Egyptian instances, the very ways in which women attempted to cultivate a pious sensibility – through veiling – became a marker for some community critics of the wearers’ failure to create ‘real’, proper piety.

The entanglement of obedience to God and resistance to the state could sometimes be interpreted positively. Among the Koran course participants in Sincan, at least, the fact that the women’s submission to the requirements of Islam entailed defiance of some state and community demands actually seemed to give even more significance to the women’s cultivation of Muslim piety. That is, the women saw their acts of resistance to the state as a sign of their commitment (submission) to God’s will – they faced risks in their efforts to become good Muslims but their willingness to bear those risks signaled their ever deepening devotion to the religion. Even so, they resented the impositions the state placed on them, such as restrictions in educational opportunities and limits on mobility. While my informants represent a particular case, I would argue that any time an individual is subject to competing normative demands would mean that the development of agency would require resistance to one set of norms while submitting to another. The individual would have to consciously position and reposition herself among various normative systems in a lifelong process that may never be fully settled. This is not a criticism of Mahmood’s thesis. Mahmood is not suggesting that the Egyptian women’s goal of cultivating piety necessarily entails inaction or passivity, but is instead critiquing Western feminism’s explicit or implicit understanding of agency as resistance. I am only introducing some other factors to consider when theorizing the development of human agency in relationship to the variety of social norms to which most human beings are subject. For many people, resistance and submission, autonomy and conformity, are deeply entangled with one another.