The soul freed of cares? Islamic prayer, subjectivity, and the contradictions of moral selfhood in Minangkabau, Indonesia

ABSTRACT

In this article, I explore conceptions and individual experiences of Islamic prayer among Minangkabau people in Indonesia, describing their role in the constitution of selfhood and moral subjectivity. I show how prayer serves as an arena in which the contradictions between conflicting visions of moral selfhood may be engaged, ideally promising reconciliation and transcendence but also inviting experiences of failure. I argue that a rich ethnographic account of Islam's role in constituting selves and subjectivities must address not only the logic and coherence of particular discursive traditions and practices but also the experiential tensions that are central to their application and embodiment. [Islam, prayer, subjectivity, self, morality, Indonesia, Minangkabau]

With Islam playing such a prominent role as the epitome of Otherness for many in the West, it has become especially important for ethnographers to present the lives and experiences of Muslim people in ways that undermine the false assumption that “Islam” is monolithic, a fixed set of ideologies and practices that stands apart from the complexities of human lives. Most people in West Sumatra, Indonesia, however, might argue that a more abstract and less flexible understanding of Islam is, in some sense, correct. Although they would not deny that a broad range of social, cultural, and individual differences exists among Muslims, they generally understand “Islam” itself to be a distinct entity, undivided, perfect. It is the Truth, something that can never be altered by any human activity. This understanding allows people in West Sumatra to make a kind of division between the messiness of the human world and the perfection of Islam, but it also presents them with a problem, one more personal than the task of the anthropologist set on undermining facile essentialisms: One must strive to constitute a properly “Muslim” self while also existing in the human world.

In this article, I focus on one particularly significant arena through which people in West Sumatra engage this challenge: the practice of shalat, the daily Islamic prayers. My discussion of shalat serves as an entry into a broader field of ethnographic inquiry regarding the role Islam plays as a source of moral selfhood. It contributes to a growing conversation concerning the ways Islamic conceptions and practices participate in constituting the subjectivities through which people imagine and experience the nature and value of the self and orient themselves in moral space (e.g., Ewing 1997; Hirschkind 2006; Lambek 1993; Mahmood 2005; Marsden 2005). I outline Minangkabau ideas about the practice of prayer and the constitution of moral selves and also examine the difficulties some individuals have in realizing them. I then place this discussion within a larger social and cultural context, describing conceptions of moral selfhood in West Sumatra and the role Islam plays in these conceptions. I argue that in West Sumatra, Islamic prayer serves as an arena in which multiple conceptions of moral selfhood, and the contradictions between them, may be engaged by individual actors. Although the practice of prayer may thus serve as a vehicle through which such contradictions can be challenged, and sometimes even transcended, it may also be a vehicle through which the intransigence of these contradictions is more directly experienced. I conclude that the practice of prayer does not simply fashion a particular self or moral subjectivity, even when interpreted within a particular discursive tradition. Rather, prayer becomes a compelling practice and a key site for moral experience in part because it provokes individuals to engage the contradictions of selfhood. I make the argument that an ethnographically rich understanding of Islam's role in forming selves and moral subjectivities necessitates examining how Islamic discourses and practices are entangled in larger social and cultural systems, their contradictions, and the experiences of individuals employing and embodying them.

The ethnographic material presented below is taken from 21 months of fieldwork I conducted in the small city of Bukittinggi in West Sumatra, Indonesia, in 2002–03. Most of the population of West Sumatra is Minangkabau (“Minang”), a society known in anthropology for its dual dedication to a matrilineal kinship system and to Islam and recognized among scholars of Indonesia for its history of considerable influence in Indonesian political, religious, and intellectual life (Kahin 1999). By the prevailing local understanding, to be Minang is to be Muslim and to be Muslim is to be moral. My fieldwork was aimed at getting a picture of moral life and selfhood as it emerged in the everyday lives and experiences of people in Bukittinggi. I employed person-centered methods (Levy and Hollan 1998), engaging in extended and repeated conversations with a small number of men and women over a period of months, asking open-ended questions that encouraged them to explore not merely the shape of social and cultural structures but also their subjective engagements with those structures.

The realization of moral selfhood through prayer

The field of the “moral” can be defined in a variety of ways in ethnographic work (Howell 1997). I use the term moral here in reference to the idea of a person, or self, that not only acts good but also is good, that realizes its purpose and achieves its proper mode of being.¹ When people evaluate the success or failure of such realization, they engage with morality. This sense of the term necessarily links moral life with broader questions about the nature of the self in the manner Charles Taylor (1989) has described in tracing the development of formal conceptions of selfhood in the West. Taylor examines the various capacities that have been assigned to the self in different philosophical traditions. One of his key insights is that conceptions of the self always implicate moral questions: Given particular capacities, what should one do to achieve a valuable existence? Which capacities should be developed and which should be suppressed to achieve human value, and how should one go about doing this?²

The arena of the moral in this sense goes well beyond the narrower usage described by James Laidlaw (1995:17–21, 2002), who draws from Bernard Williams and Michel Foucault in limiting “morality” to codes of conduct.³ The broader conception of the moral includes (as does Laidlaw's “ethics”) the processes through which actors work to fashion themselves as particular kinds of persons. Morality may be an arena of conscious practice (Lambek 2000a) in addition to habitus, of the reproduction of established social structures as well as the exercise of free choice in pushing against social structures or navigating their contradictions (Robbins 2007). Beyond the issue of freedom, morality becomes a significant dimension of life whenever people are concerned with “the ultimate terms of their existence” (Parish 1994:290) and how these terms define one's personal value and the value of others. People can experience themselves as inhabiting moral states, or failing to inhabit them, whether or not they have exercised freedom in realizing those states. Such a conception invites ethnographic exploration of the capacities of self imagined within particular cultural traditions, the practices through which they are cultivated, and the experiences of actors attempting to endow them with human value (e.g., Marsden 2005; Parish 1994; Simon 2007).

In West Sumatra, the performance of shalat, the five obligatory daily prayers, is an especially rich point of engagement with self-consciously “Islamic” conceptions of the self and its moral orientations and with Islamic practices of self-formation. Shalat is one of the five fundamental practices required of every Muslim, and the only one required multiple times every day.⁴ In Bukittinggi, it is one of the most frequent topics of sermons and religious lectures, which regularly emphasize the obligation to pray, the benefits of prayer, the proper way to approach prayer, and the gravity of the sin of neglected prayer. Posters sold in the marketplace and subsequently hung on the walls of some shops and living rooms detail the tortures one will face in the afterlife for each prayer that goes unfulfilled. Learning how to recite the Koran and learning to perform shalat are the most prominent elements of a Minangkabau child's religious education. When my conversations with people turned to matters of religion, or even to the idea of being a good person, prayer usually found its way to the center of the discussion. This was often the case whether or not the person in question prayed regularly.⁵ It is through the performance of shalat that the capacities of a moral self are said to be realized. Some people claimed to me that in the absence of shalat such realization was impossible, an idea that no one discussed in connection with any other Islamic practice.

Shalat consists of a series of prescribed movements of the body carried out as an embodiment of submission to God. These movements are performed in concert with the recitation of the first sura (chapter) of the Koran, praising and declaring one's faith in and submission to God, followed by the recitation of another, usually brief Koranic passage of the worshipper's choosing.⁶ Each performance of shalat involves a series of these movements and recitations.⁷ When a person has completed the ritual, he or she may also elect to take advantage of the closeness to God achieved during prayer and ask God for more specific forgiveness or guidance in facing personal problems, but this is not considered obligatory.⁸ Shalat is sometimes performed collectively, most notably by men in mosques during the Friday noon prayer (a standard Islamic obligation for men but not for women). Going to the mosque or prayer house to perform shalat is said to strengthen its value, as it constitutes a more involved performance of piety. Most of the time, people in West Sumatra perform shalat at home or in some other private space or, perhaps, in the marketplace in a public space set up for this purpose (musholla). Either way, people usually perform shalat as an individual act. An optional midnight prayer, almost always performed alone, is said to be the most intensely experienced kind of shalat and is practiced especially by those faced with a difficult problem.

The ways in which people understand a moral self to be cultivated through shalat—and the ways in which they discuss, in an experience-near fashion (Geertz 1984), the concept I have more abstractly termed “moral selfhood”—can best be appreciated by turning to what people told me about its significance and effects. I begin with Da Dan, a struggling petty trader in his mid-thirties.⁹ Da Dan and I sat frequently in the small, cluttered main room of his house. Recently erected in the neighborhood of Bukittinggi that Da Dan considered to be his home “village”(kampuang), the house was built in part from reused wooden boards and aluminum sheets. Da Dan typically spoke in a controlled, self-confident register, offering strong, thoughtful opinions punctuated by drags on a cigarette. “When Muslims pray,” Da Dan told me, “pray the right way, understanding what is said, understanding prayer … after praying they feel strong. Very strong. Without any problems. All problems disappear.” For example, he elaborated, “if we pray, we won't ever be sick. Never any kind of illness at all.” He explained that, when the prayer is “received” by God, “the pain disappears. Why?” he asked, rhetorically, and then answered,

The soul is freed of cares [plong]. Without any burden. No burden in one's life. As for myself, if I don't pray enough, [I think,]“What's going to happen? What about tomorrow? What about the future? What about the kids? What about the household? What about the debt? What about this?” It's a lot of “What about? What about? What about?”—those are the questions. If a person prays, it's never like that. None of that. If it's true [betul] prayer. Not just any prayer—no. If it's true prayer. Sometimes, you know, people pray but they still have problems in their life. It's not God's fault. Ask them, “How are you praying?”

There is a great deal to consider in Da Dan's comments, including the notion that prayer must be true to be effective. I concentrate here on the claim that shalat can eliminate all problems from one's life. This idea was echoed by many of my subjects, who discussed the ways that prayer allows one to realize his or her most valuable mode of being. Below, I outline three main dimensions of this realization: the mechanical realization of a clean and well-ordered self, the realization of a self that acts morally, and the realization of a self that does not suffer.

First, shalat transforms a self mechanically into one that is clean and well ordered. When I first talked about shalat with Da Palo, an underemployed skilled laborer in his fifties, he focused entirely on this kind of realization. “The benefit of prayer is so great,” he began. He then referenced the required ablutions that precede the prayers: “Muslims are obligated to pray five times, right? In terms of the face alone, it gets washed every day, five times. Just to pick an example, if the face is washed all the time, it's going to be clean.”

After expanding on the idea of cleanliness, he turned to the way prayer structures a person's use of time. Shalat takes place at specific times of the day, signaled by the call to prayer that echoes throughout the city. Da Palo traced out the course of an entire day for me, as structured by prayer: rising early for the dawn prayer to drink coffee and then go to work, working until the noon prayer signals a break and time to eat, working again until the afternoon prayer signals that the workday is through, going home tired and dirty from work, washing and praying again, and being with one's family until the evening prayer signals the time to eat dinner and then go to bed. He concluded, “At eight o’clock in the evening we rest, sleep, and tomorrow morning go to work again, at full strength, right? If you sleep at the appropriate time, the next morning your body will be fresh when you wake up. And that's how prayer educates human beings.”

People's claims about the temporal discipline of prayer often went further than this. By embodying the regularity of prayer, people explained, they came to realize the disciplined use of time in their everyday lives. They came to know time as divisible in an orderly, productive fashion that allows human beings to fully take advantage of the opportunities provided by God, leading to greater prosperity. This notion resonates broadly in a context in which individual endeavor in a competitive marketplace is central to economic life, as I discuss below.

Of even more concern to most of the people I spoke with, however, was a second kind of realization effected by prayer: A self that prays will always act morally. Da Palo offered the common example that a person who prayed would have no desire to steal a large amount of money sitting within easy reach. This will be true, he said, as long as the prayer is not just performed, but is truly mandirikan, established as a genuine act of the autonomous self. Mandirikan conventionally translates as “establish,” but it has more significant connotations. The root of the term is diri, meaning “self,” but many forms of the word take on the meaning of “standing.” Further, to be mandiri is to be “autonomous,” to be as a freestanding self. Da Palo is thus expressing the idea that true prayer must be an autonomous expression of the self. The self must not just do prayer but become a state of prayer, a state of moral consciousness that is then carried out into the everyday world. People frequently repeated this central point to me: A person who truly prays will inhabit a moral state, always aware of the difference between right and wrong, and will never become driven by a desire to do wrong.

The realization of a self that acts morally is closely connected to a third kind of realization: that of a self that does not suffer. Ni Galeh, a widow in her mid-forties who owns a small clothing shop, described those who do not pray: “They punish themselves. Their lives are aimless. It's senseless, their talk is senseless. They can't do this or that. Yes, their thoughts are chaotic. Their souls aren't calm, their spirits aren't calm.”¹⁰“They get punished,” she explained. “If we don't pray, we’ll disturb people, we’ll steal other people's things. If we pray properly … we won't feel like getting angry, won't want to steal, won't want to disturb people. Those are sins.” Notice how Ni Galeh weaves together the idea that a person who does not pray will act immorally and the idea that this person will suffer. People who do not pray do bad things “and it seems right,” she said. “But it's still wrong. They don't get any guidance. Their spirits aren't calm.” In this state of turmoil, their immoral acts are, in her words, “out of their consciousness”[diluar kesadaran].

Virtually everyone I spoke with explained that praying resulted in a feeling of tanang (calm), a positive state associated with a lack of anxiety and a general feeling of well-being.¹¹ Despite the apparent cliché, people spoke convincingly and with emotion about their experiences of tanang. Interpreted within a particular religious framework, the experience can be understood as “a spiritual state, but also at least partially a purely physical consequence of the prescribed regulated breathing, chanting, and rhythmic prostrations of the supplicants, which have a potent physiological effect” (Lindholm 1996:143). This state of well-being may be said to extend well beyond the moment of prayer itself, until—as Da Dan claimed—a person who prays regularly literally has no problems in life at all.

The basis for all of these realizations of the moral self is the idea that, in praying, a person directly embodies submission to God. The term Islam itself refers to the submission to God that is the defining characteristic of moral selfhood. In this sense, the reason a person should pray is simply that God commands prayer, and, therefore, practicing prayer constitutes a moral self. However, as should be clear at this point, the way people in West Sumatra engage with prayer reaches far beyond this abstract ideology. Even if shalat is ideologically defined as a form of submission, in contrast to forms of prayer that constitute supplication (Headley 2000; Parkin 2000b), shalat is not experienced purely as a “non-reciprocal” affirmation of faith (Parkin 2000a). It is crucial to take into account people's concern with how the moral selfhood promised by shalat is manifested in the experience of living. People's emotional investment in shalat is rooted most deeply in the promise that a moral self can be realized through the living of a physically pure, orderly, and prosperous life; experiencing an absence of problematic immoral desires and behaviors; achieving a sense of tranquility; and attaining freedom from suffering.

I return now to Da Dan, who extolled at length the wonders of prayer in creating such an ideal mode of being. My conversation with him underwent a distinct shift when we moved from discussing Islamic practice in the abstract to discussing his personal experience. When I asked him how often he performed the prayers, he responded, “Right now, I’m prayer-free [perai shalat]. It's been four months—it's been five months. For reasons that aren't clear. I don't understand. There is no reason.” Over the course of the conversation that followed, Da Dan's description of his prayer-free period grew from four or five months to “almost a year” and, eventually, “more than a year,” but no matter how I approached the issue, he only repeated that he was confused and surprised by his own lack of prayer and could not understand it. It troubled him. The desire to pray would sometimes arise, he said, but then it would fail to materialize into practice. At the same time, he insisted that the powers of prayer were not simply theoretical: “I’ve experienced it myself,” he explained. “Praying—if it's real prayer, I’ve been released from every problem and achieved happiness. I have.”

Da Dan was one of several respondents who told me of such problems with prayer, who seemed to have internalized the need to pray deeply enough that it was a matter of emotional significance for them, but who still found themselves failing (through commission or omission) in their prayers. Considering that an individual's own practice or lack of practice of prayer is widely considered to be an inappropriate topic for open inquiry and discussion, this acknowledgment by even a small number of respondents is significant. Clearly, believing in the ideologies of self and subjectivity associated with prayer does not eliminate friction from their practice. To understand how prayer holds the promise of constituting a moral self, and the subjective tensions inherent in realizing this promise, one must frame prayer within the broader social and cultural context of selfhood in which it is interpreted and experienced.

Outlines of moral selfhood in contemporary Minangkabau society

I offer the following ethnographic sketch to illustrate the connections between the moral selves constituted through prayer and a larger, more complex cultural and moral web. The full significance of Islamic discourses, and the subjectivities tied to them, can only be appreciated within this larger context. The central argument of this sketch is that Islam in West Sumatra has participated in the elaboration of contradictory conceptions and experiences of authentic moral selfhood that can be understood schematically as follows: Selves are sometimes imagined as most properly constituted by their deference to, and integration into, larger social and spiritual powers, yet selves may also be understood as most properly autonomous. The shape of moral selfhood in West Sumatra can be found in the way these conflicting modes of moral selfhood are each valued and are each located in particular forms of practice and expression.

By the early 19th century, Muslim traders from India and the Middle East had been active along Sumatra's west coast for centuries, and Islam had come to the highlands of west Sumatra and been embraced, at least rhetorically, by some Minangkabau royalty 300 years earlier (Marsden 1966). Islam had spread widely, if not deeply and evenly, through the region. Some entire villages still remained unconverted, and mosques were generally marginal to villages in which they were located (Dobbin 1983). Although Minangkabau royalty controlled most of the trade in gold that had driven regional commerce for generations, wealth being generated by the expanding trade in village cash crops allowed an increasing number of Minangkabau Muslims to make the pilgrimage to Mecca. In 1803 Minangkabau pilgrims witnessed the conquest of Mecca by Wahhabis, representatives of a movement stressing the need for society to return to a pure form of Islam and advocating the use of force, if necessary, to achieve this end. Impressed, some of these pilgrims, who came to be known as Padris,¹² returned to western Sumatra determined to establish Islamic purity and supremacy in their homeland.

The Padri movement appears to have gained traction in large part because the region was in desperate need of a new framework for social and moral order, one that included a clear place for independent endeavor in the marketplace. This need was due to a combination of factors detailed in Christine Dobbin's 1983 historical study. The authority of Minangkabau royalty, who had worked to monopolize regional trade, had eroded along with the dwindling supply of the gold on which royal commercial power was based. Meanwhile, opportunities for trade were greatly expanding because of the European presence, and individuals were increasingly pursuing their own economic fortunes by planting cash crops on land that lay outside the boundaries of both royal and kinship controls. The growing marketplaces that resulted—also outside of village control—were plagued by bandits, commercial disputes, gambling, drinking, and opium. This context accounts for the particular appeal of Islam's framework for egalitarian and individualistic endeavor within a united moral community, one originally fashioned out of the social and ecological conditions of the Middle East (Lindholm 1996). Dutch colonial forces helped to dull the edge of the Padri movement and defeated it militarily, but the movement had nevertheless succeeded in firmly establishing Islam at the center of Minangkabau moral discourse (Ricklefs 2001). As Minangkabau society became integrated into the Indonesian nation-state in the 20th century, it continued to imagine itself in terms of an “Islamic trading society” (Kahin 1999:71).

The dual moral emphasis on an integrative social order and the preservation of individual autonomy that drew Minangkabau traders to the Padri movement still resonates culturally, as it also does with the realities of social and economic survival in contemporary Bukittinggi. Many residents of Bukittinggi depend on their village-based kinship ties, for, as members of a lineage (or, for men, through marriage to a member of a lineage) in a particular kampuang (village, neighborhood), they gain use rights to property for cultivation or habitation. The kampuang—its coffee houses, shops, and washing sites, its prayer houses and ritual celebrations—remains a key arena for daily social interaction on an intimate scale. However, to make a living, most residents must venture out into the region's larger economy, which is dominated by petty trade and independent skilled and unskilled laborers.¹³ For most people in Bukittinggi, making a living is a competitive and often very individualistic pursuit (cf. Kahn 1980). This is particularly true for men, as it is women who inherit use rights for lineage property. Young men are traditionally expected to leave the cradle of village life for a time to pursue independent financial success and only then to return to marry, after which they live in their wives’ homes. The economic realities of West Sumatra make achieving such independent economic success difficult or impossible for many men, although it remains a masculine ideal (cf. Whalley 1993).

For the last 200 years, Minangkabau society has worked to reconcile Islam with the conceptually traditional, village-based system of social order, known as adat (or, in Minang, adaik). Adat defines properly “Minangkabau” ways of interacting, including kinship and property relations, ritual life, and etiquette. The notion of conflict between Islam and adat, usually in regard to the matrilineal kinship system at the heart of adat, is frequently discussed in West Sumatran ethnographic and historical literature (e.g., Abdullah 1966, 1971; Blackwood 2000; Hadler 1998; Hamka 1984; Sanday 2002; Whalley 1993), but this literature often points to the significance of merging and accommodation as well as to conflict between the two frameworks. To a large degree, the idea that adat is a self-contained, well-defined set of social codes and structures, and that it can be understood as something indigenous and in tension with “Islam,” was a product of Dutch colonial authorities and scholars (Kahn 1993; cf. Hefner 2000). Many “traditional” adat forms were, in fact, shaped by the colonial era (Benda-Beckmann 1979; Benda-Beckmann and Benda-Beckmann 1985; Kahn 1993; Kato 1982). However, many Minangkabau thinkers adopted this “traditional” conception of adat, attempting to define its precise outlines or considering how it should be reconciled with Islam (e.g., Amir M.S. 2003; Hamka 1984; Navis 1984).¹⁴ Although the notion of friction between the two has not disappeared, the relationship has evolved in such a way that in contemporary Bukittinggi, Islam has—at least rhetorically—become the dominant moral term, co-opting adat. Public discourse on moral life in West Sumatra today inevitably involves reference to the phrase “adaik basandi syarak, syarak basandi kitabullah,” meaning “adat is based on Islamic law, Islamic law is based on the Koran” and, thus, asserting that adat itself has its most fundamental basis in Islam. (The phrase is so pervasive that it is most often simply abbreviated to “ABS SBK.”) Earlier versions of the saying posited that adat was based on what was true or right, with no mention of Islam; or that adat and Islamic law were mutually constituting; or that adat was based on Islamic law, which was based on the Koran, which, in turn, was based on what was true or right (Abdullah 1985; Kato 1982). These versions appear to be totally absent from discourse in contemporary Bukittinggi. To act properly according to adat, one must, by definition, also act in an Islamic fashion.

Many elements of adat—that is, many elements of social interaction that are now identified self-consciously as characteristically “Minangkabau”—have not been erased in favor of “Islamic” social life but have, instead, been reframed by or interwoven with Islamic discourses (cf. Abdullah 1985). People use Islamic conceptions to make sense of or work through the conflicts of moral selfhood that arise in the course of everyday interactions. What I am particularly interested in here is the way that everyday interactions—especially properly Minangkabau ones—consistently communicate the notion that individual will and desire must be displaced so that the good of social integration can be realized.

One can see this ritualized in etiquette, the component of adat known as baso-basi, in which strong emotions and individual endeavor and consumption are displaced in favor of conventional, ritualized interactions. Merely by carrying out this etiquette, Minangkabau people demonstrate that they do not act according to inner, individual impulses but, instead, according to conventions that they have accepted as morally superior. The content of baso-basi reinforces the message of social integration. The foundation of baso-basi is simply the act of bagaua, socializing by being out in the community, greeting, being together with and talking to other people. This is often considered the single most telling sign of a person's overall moral status. As one man put it to me—addressing what he saw as foreign misconceptions of West Sumatra's Islamic society and the way they obviously contrasted with the social interaction we saw around us every day—“How can people be terrorists if they are able to bagaua with others?” To bagaua according to baso-basi means to properly greet others, to offer food and drink to others before consuming them, and to avoid direct expressions of disapproval, anger, longing, sorrow, or other sentiments that may cause social conflict. These sentiments are, instead, suppressed or expressed obliquely, for example, through sayings, metaphors, or allusions.

In contrast, perhaps the worst thing a person can be in Minangkabau society is sombong (cf. Heider 1991:86; Wikan 1990:64). To be sombong is to exhibit a kind of arrogance, to appear to be separate from, as if superior to, others. Failure to socialize with others is sombong, showing signs of one's own good fortune is sombong, wearing expensive clothing is sombong, and insisting one's own opinion is correct or forcing it on others is sombong.

It would be a mistake to underestimate the moral weight invested in proper socializing, to dismiss it as mere etiquette. Frederick Errington (1984) went so far as to argue that baso-basi is the fundamental basis of Minangkabau personhood, its performance the very goal of social interaction itself. He concluded that, whereas he, like other Americans, “was accustomed to measuring individual worth primarily in terms of a personal self,” in contrast, “the Minangkabau were accustomed to measuring individual worth primarily in terms of the extent to which an individual observed baso-basi” (Errington 1984:20). In this view, a moral self is created by displacing internal, individual impulses—inherently antisocial and immoral—with proper behavior as defined by the community.

This view resonates with an Islamic discourse that is in increasingly widespread circulation in West Sumatra: the interaction of napasu and aka (or nafsu and akal in Indonesian). The various uses of these terms in West Sumatra and in other Islamic societies are too complex to cover in detail here.¹⁵ Schematically, following the most common conceptions employed in West Sumatra, napasu consists of innate, bodily appetites that motivate people to act in an animalistic manner—and, thus, amorally. Aka is the power of reason and can be directed at constraining napasu in the interest of acting morally. Aka, however, is innate only as a potential and must be cultivated through socialization and ritual practice.¹⁶

The moral gravity of sombong (arrogance), and its entwinement with Islamic discourses, can be appreciated by considering that, in West Sumatra, sombong is frequently cited as the key quality of Ibilih, the Devil. The Koranic story of the origin of evil and suffering in the world describes Ibilih as an angel who was cast out of heaven, and into the lives of humans, after refusing God's command to bow down in deference to Adam, the first human. Ibilih is described in Minangkabau terms as sombong for elevating his own position and refusing to submit to God's commands. Sombong, in this sense, is understood as the root of evil: a refusal to displace one's individual will with submission to God.

Errington's claims about the fundamental importance of baso-basi and its emphasis on social integration have foundation. I saw a great deal of evidence that this particular moral orientation is not only pervasive but also deeply felt in everyday experience.¹⁷ However, many ethnographic accounts of the nature of “persons” or “selves” that base what the self is, or how it is imagined, on a particular moral orientation (such as “social integration”) have been persuasively critiqued for being overly essentializing or ethnographically partial (Gregg 2007; Hollan 1992; Kusserow 2004; Lamb 1997; Lindholm 1997; McHugh 1989; Murray 1993; Spiro 1993), and any analysis of Minangkabau selfhood anchored in social integration would be subject to this same critique. The social nature of Minangkabau persons that emerges in Errington's analysis is a partial story, one that stems from an ethnographic focus on interactions that are self-consciously social in nature, aimed at publicly communicating what is properly communicated in public.

For all of the emphasis in the public performance of baso-basi on displacing individual will and desire, Minangkabau culture, in fact, also greatly celebrates individual autonomy. This is seen clearly in celebrations of Minangkabau ethnic identity. For example, people repeatedly returned to the same saying to describe what they saw as Minangkabau character: “Wanting to be on top when pressed down, wanting to be outside when caged in.”¹⁸ It refers most obviously to a desire to be free of burdens and restrictions, and people cited it often when talking to me about the proclivity of Minangkabau people for independent trade, their tendency to be particularly clever or crafty, and their refusal to passively submit to others (as, they would often say, the hierarchical Javanese do). Perhaps more tellingly, people connected the phrase, with pride, to personal stories that they told me of their own refusals to be pushed around by authorities, whether by direct resistance or, more usually, stealthy maneuverings.

“Wanting to be on top when pressed down, wanting to be outside when caged in,” not only refers to a desire to be free of burdens and restrictions, however. It also refers to a desire to simultaneously bear those burdens and honor those restrictions while being free of them: that is, wanting to actually be outside while still, apparently, caged in. It references an attempt to do the seemingly impossible, managing to successfully realize the public, socially integrated part of oneself while still cultivating oneself as an autonomous actor.

The kinds of careful and indirect communication people engage in with others do not at all mean that individual desires and emotions, criticisms, and so on are devalued or dismissed as inauthentic. To the contrary, such impulses are kept out of public circulation because of their great but particular value: They make up the realm of the paribadi. The paribadi is the “personal”: those attributes and behaviors that adhere to the individual person and should be kept apart from those parts of one's self that integrate with others. One's individual character traits are one's paribadi, but so are one's relationship with a spouse, one's mystical powers, and one's financial pursuits.¹⁹ When my respondents told me stories of their lives, they often stressed to me that such stories were paribadi: I was not to circulate them inside the community. These stories were not only embarrassing accounts of pain, failure, and moral lapses but also of personal success that would be sombong to talk about openly. It was these stories that people referred back to when claiming that, after months of conversations, I now understood who they really were. Far from being suppressed or devalued, the paribadi is the site of an alternate locus of the authentic self, albeit one that is, by definition, inappropriate to display openly.

Discussing daily interactions with my subjects, I came to see clearly that they felt that these interactions needed to be tightly controlled not only so that smooth integration could be maintained on a social level but also so that their own, and others’, autonomy could be preserved in the process. Integration with others involves risks: that one will violate the boundaries or autonomy of another or that one will lose control of oneself if one's own boundaries are not protected. Thus, just as part of the self must be realized in relationships with others, another part of the self must be kept from being exposed too directly to such relationships (cf. Wikan 1990). For all of the concern in West Sumatra with the moral value of socializing, it is striking that people almost invariably attributed immoral behavior to the consequences of this socializing. Rather than blame immoral behaviors on the innate immoral impulses of human beings, people almost always described such behaviors as the result of interpersonal contact, regrettable social norms, or supernatural entities that corrupted what had been the innate good intentions of the individual.

All of this becomes intertwined with Islamic discourses as well. When Ibilih refused to bow down to Adam, God condemned him to eternity in hell but then gave him a reprieve: He could inhabit the world of humans until judgment day, exploiting their napasu (appetites) to draw them toward evil. There is no concept of original sin in Islam: Instead, human beings are understood to be born pure, and only through interacting with a corrupt world do they become corrupted. As human beings face the temptations and corruptions of the world, they experience the presence of Ibilih. In this view, the goal for human beings is not so much to resist innate appetites as it is to keep the world from drawing them away from the pure core of themselves: the ati, or emotional heart in which there exist those things that people truly feel, no matter what they may express outwardly, and their true intentions, no matter how they actually behave. In particular, the ati is where compassion (ibo) and faith (iman) reside. The ati is, at its core, good. This concern with the truth of inner being is entwined in a larger discourse that distinguishes between the laia, or external reality, and the batin, or internal, metaphysical reality (cf. Bowen 1993; Geertz 1984). This distinction is associated with Sufi influences, especially strong in Java, that have sometimes been underappreciated in accounts of contemporary Islam in Indonesia (Howell 2001; cf. Woodward 1989).

In the West Sumatran context, Islamic discourses of moral selfhood are thus intertwined with Minangkabau cultural concerns about social order and integration, as realized through the conventional performance of baso-basi, but also with concerns about the value of individual autonomy and internal states. Minangkabau moral selfhood cannot be usefully essentialized as based on a particular concept of or locus of the self or on one particular value or moral orientation. Rather, it is located within the particular ways that multiple and contradictory concepts of the moral self are managed, each given its own forms of expression and arenas of meaningful experience. These contradictory concepts do not form alternate models of selfhood that circulate separately in West Sumatra, creating different kinds of Minangkabau selves. Although one moral orientation may be dominant within a particular cultural sphere, such neat correlations do not always exist (cf. Robbins 2007). Successful socializing, for instance, often both realizes integration on a public level and defines and protects the realm of the paribadi (personal). The challenges this balancing poses are an integral part of moral selfhood for all Minangkabau actors and are, thus, an integral part of the nature of Islam in Minangkabau society. This observation is not surprising if one understands cultural configurations of selfhood to grow out of, rather than simply create, the processes through which people reflect on and try to make sense of the worlds in which they find themselves (cf. Cohen 1994), worlds that are often contradictory.

In his detailed intellectual history of selfhood in the West, Jerrold Seigel (2005) has argued that theorists have had to confront three major dimensions of the human self: bodily, relational (the integration of the self into larger social structures that help to constitute it), and reflective (a self-conscious agency capable of reacting to, elaborating on, or even moving against the other dimensions in creative ways).²⁰ He also persuasively argues that theories of the self that deny or minimize one or more of these dimensions are necessarily less satisfying than those that attempt to understand the interrelationships and tensions that exist between them. My analysis of the culture of selfhood in Minangkabau society supports this idea (although I have given less attention here to the body) from an ethnographic perspective: Like the theories of individual philosophers—if less coherently—cultural systems grow through people's attempts to think through and elaborate on these different dimensions of self and the relationships between them. Broadly speaking, the dual orientations toward integration and autonomy in Minangkabau selfhood are elaborations of a basic existential paradox in the experience of the human self: The self is undeniably constituted by its integration with the world yet undeniably autonomous as well. This tension emerges through a particular history and in particular cultural patterns of moral selfhood in West Sumatra, revolving around ways of managing, expressing, and even celebrating these conflicting dimensions of the experience of self, although never completely reconciling them.

The promise and problems of transcendence through prayer

The possibility and promise of such a reconciliation are central to shalat's power. The very idea of Islam is moral perfection, and Islam is most directly engaged, experienced, and embodied by people in West Sumatra through the practice of shalat. In Minangkabau, shalat's key position as an Islamic practice allows it to serve as a site of crystallization for the tensions that pervade moral selfhood in everyday life. The ritual offers the possibility of experiencing their resolution—or of being faced with their stubborn endurance. It may push people toward at least momentary transcendence, but the elusiveness of this promised transcendence may also be a source of anxiety and frustration. The moral self that is crystallized in prayer is, ideally, at once both completely integrated and completely autonomous.

Shalat can be understood as an act of total submission of the individual self to a larger power—to God but also to the community of believers and its conventional practices. In the practice of prayer, people must turn their bodies over to the prescribed motions of the ritual and their attentions completely to their submission to God. They must do this at the same time and in the same way that everyone else is doing it, five times every day. If not performed in the proper way, at the proper time, in the proper state of purity and the proper state of mind, the prayer is said to be invalid and the obligation to God unfulfilled.

At the same time, shalat can be understood as an act of individual will that must arise from within the self. The connection with God occurs in the ati (heart) and must constitute a genuine desire to submit—it cannot be faked or forced. For prayers to be effective, and for them to be valid, they must represent genuine intentions and be offered in a state of khusuak, a total, sincere concentration on God. God must completely saturate the consciousness of people while they pray: They cannot think about what is on TV, what they are doing later, what they want from God, or whether or not they will be able to feed their children the next day. All of these are elements of daily life in the corrupting world but not part of the transcendent truth of God that is the sole focus of prayer. If any of these other concerns spontaneously arise from within the self during prayer, the prayer is not genuine. As a man named Da Eri emphasized to me, “Prayer is a matter of the heart.” He explained, “Sometimes people do the daily prayers just great, but in fact their heart isn't khusuak. … Concentration. A direct connection with God. So, even though their movements are good, you don't know if their prayer is good.”

Notably, each person I have quoted above regarding prayer also stresses that prayer only creates a moral self if it is true (or real, or proper) prayer. The state of khusuak must be achieved for this to be the case. Khusuak itself is often described by the very same term used to describe the effects of prayer: It is a state of tanang, calm. The cause and the effect of moral transformation through prayer merge. Ultimately, such a transformation both works on and is carried out by the self—it must be a matter of individual will. Prayer promises a realization of moral selfhood through complete submission of the self to God; prayer also requires that this realization genuinely reflect and arise from the will of that very self.

That conceptions of Islamic prayer can reach out in these different directions is apparent in ethnographic accounts of Islam in Sumatra and beyond. Although shalat takes the form of an embodied practice, its social, political, and experiential meanings are not contained within its physical practice (Starrett 1995). Addressing developments in the 1930s in lowland Aceh, at the northern tip of Sumatra, Siegel (1969) writes that shalat was understood to be a triumph of akal over nafsu, as the worshipper suppressed spontaneous inner appetites through the imposition of reason. This was exemplified in the discipline involved in carrying out God's commands precisely. “Prayer [was] not an outpouring of inner feeling but the believer's conformation of himself to God's commands” (Siegel 1969:114). The importance of inner states of prayer, and the worshipper's connection with God, was played down in favor of prayer's formal and mechanical elements of conformity and discipline that led to a strongly united community.

Bowen contrasts this situation with that of Gayo society in the Acehnese highlands, where prayer was most often understood in the villages “as an act of communication between the worshippers and God” (1989:604), similar in form to communications that took place between people and spirits or ancestors. Bowen (1993) writes that Gayo society in the 1930s witnessed a battle among Islamic leaders over how to place intention in relation to prayer. Approaching shalat as a form of communication with God, Gayo traditionalists practiced ritual recitations of the intention to pray before beginning shalat itself.²¹ Modernists vigorously fought this practice, insisting that prayer was essentially about obedience to God's commands and that, because the recitation of intent was not founded in any authoritative Islamic source, it was a distortion of proper Islamic practice and thus disobedient to God. Significantly, Bowen writes, modernists were also concerned that announcing intent undermined the notion that intention was already inherent in any form of practice, manifest in the doing of a ritual act.

Islamic reformers elsewhere have also shown a particular concern for emphasizing the entirely conventional states of subjectivity that must be manifested in prayer through entirely conventional practice. James Wilce (2002) writes of just such a shift in Bangladesh, where “tuneful” prayers are now regularly condemned as a personalized and, thus, narcissistic form of expression inappropriate to the worship of (i.e., submission to) God. Wilce's ethnography does not focus on shalat specifically but, rather, on a broader category of Islamic prayers that more readily lend themselves to variation in performance. His larger insights, however, still apply: “Prayer balances individuated and corporate voices” (Wilce 2002:289) that are not always easily reconciled, leading to struggles over the proper nature of the balance.

As evident above, people in West Sumatra are concerned not only with the external conventions of shalat but also with the internal states associated with its practice, making a mismatch between the two ripe for consideration. Although the moral emphasis on matters of batin, or inner realities, is significant, such concerns have not coalesced around organized discourses that favor the inner truth of worship over conventional external forms, as they have in Java (Headley 2000). In West Sumatra, shalat is like baso-basi (etiquette) in that it is carefully packaged into a conventional form, an embodiment of the deference of the individual to larger powers as a means of creating a wholly moral self. To not pray, or to pray in any but the prescribed manner, is a moral failure. Yet the true value of prayer is said to be a highly personal, or paribadi, matter, not to be made a matter of public discussion or spectacle. Unlike baso-basi, prayer must be both wholly conventional and wholly personal, an act of complete submission and an act of genuine intention. If everyday life in West Sumatra is a constant balancing act of realizing and expressing different, even contradictory, elements of moral selfhood, prayer offers the possibility of transcending those contradictions. As the most direct engagement people have with Islam, it promises the complete moral perfection of Islam itself, in which all contradictions vanish and a moral self emerges without contradiction.

Saba Mahmood's argument in a 2001 article and 2005 ethnographic monograph is particularly relevant here and important to engage, as it is among the most widely read and influential contemporary anthropological work addressing Islam's role in constituting selves and moral subjectivities. Mahmood's work speaks to the ways that prayer can act as a site of transcendence beyond the tensions between different dimensions of self, an attempt to reconcile what shapes the self from without and what arises from within. She suggests that this promise of transcendence calls into question the notion that tension necessarily exists between what is social and what is individual within the subjectivity of selfhood.

Mahmood's concern is with the way people inhabit moral states through bodily practice such as prayer. She describes members of a women's piety movement in Cairo who are concerned that, in Egypt, prayer has become an empty ritual of Islamic identity and who insist that prayer must be performed in the proper subjective state: It must arise spontaneously from a genuine desire to pray that exists in the heart. She argues that the women in the piety movement understand the daily prayers to be “a key site for purposefully molding their intentions, emotions, and desires in accord with orthodox standards of Islamic piety” (Mahmood 2001:828), thus blurring the distinction between any kind of spontaneous or genuine states of selfhood and any imposed or conventional states. The ritual of prayer, she says, is seen by these women as a practice that can inculcate a self that spontaneously conforms to moral conventions at all moments of everyday life.²² Pray genuinely, and conventional moral behavior will arise spontaneously from the self. The prayer, a form of conventional moral behavior, must also be genuine, and the women strive at all moments of daily life to realize themselves as moral so that, when the time comes to pray, the desire and concentration necessary for prayer will also arise spontaneously.

Mahmood argues that this conception of prayer calls into question the nature of the oppositions assumed in many theories of ritual. She points out that rituals may be understood (as in Turner 1969) as taking spontaneous emotions that arise in everyday life, but that cannot normally be expressed because of the conventions of social interaction, and channeling them into an acceptable, orderly form of public expression. Alternately, they have been analyzed (as in Tambiah 1985) as practices that push aside the chaotic and disorderly emotions that arise spontaneously in everyday life and impose orderly, conventional states in their place. Both of these approaches, she writes, assume an opposition between what arises spontaneously and genuinely from within the self and what is imposed on the self from outside. The women in the piety movement understand shalat to constitute selves that spontaneously practice conventional Islamic morality, and this, writes Mahmood, “problematizes the ‘naturalness’ of emotions as well as the ‘conventionality’ of ritual action, calling into question any a priori distinction between formal (conventional) behavior and spontaneous (intentional) conduct” (2001:828).

Although Mahmood and I diverge on what conclusions to draw from this challenge to the analytical distinction between imposed convention and spontaneous will, clearly when individuals engage in conventional ritual practices, the possibility is always present that both of these factors are simultaneously in play. The point is certainly relevant to prayer in West Sumatra. In West Sumatra, prayer is, indeed, a practice in which what genuinely arises from the self and what is imposed on the self are supposed to converge: A state of prayer must be completely authentic to the self and spontaneous, and simultaneously it must be entirely conventional. It is this point of intersection between experiencing these two dimensions of self—self as an autonomous actor and self as constituted through its relationships with and submission to something larger—that I find most compelling about prayer. Mahmood shows that elements of these dimensions are entangled in prayer and are not easily separable—and, thus, prayer may become the locus of an ideology of transcendence. My own work emphasizes that the experience of prayer and the ideological demands of prayer may also help to realize the enduring tensions between spontaneity and convention, in which the two lay claims on being the source of the authentic self but cannot be neatly fused or reconciled.

Each of the individual accounts of prayer or failure to pray offered to me by Minangkabau people follows its own contours—an important point in and of itself—but some significant patterns can be noted. Some people appear to have not internalized (Spiro 1987) the need to pray deeply enough that they go beyond publicly professing the importance of prayer.²³ Yet, as the petty trader Da Dan did, others truly appear to wrestle with the importance of prayer and their failure to practice it, and these people seemed to me to struggle with their ability to successfully embody the ideology of prayer and its transcendence—they could not realize the state of prayer and its complete submission to God as a genuine and spontaneous expression of their own will. All of them connected the troubles they faced in their lives, the way their consciousness and identity was grounded in these troubles, and their inability to pray.

Ni Yas, a middle-aged woman, expressed this connection clearly. After having first told me that she prayed regularly (following the typical pattern of these conversations), she later acknowledged that she did not, and that this was a sin. She said she prayed

when my heart wants to do it. If I’m troubled … in fact I’m prone to neglect my prayers. Other people aren't like that. People are troubled and they’re diligent in their prayers. But as for me, I am like that. If I’m troubled I don't concentrate when I pray. For example, if I have a lot of thoughts, sometimes I’m praying and my mind is over there. That's like not doing it at all: praying without khusuak, right? Because my thoughts are, for example, … “Where am I going to get money later to buy rice?”

How, she asked me, could she possibly be khusuak while praying?

Da Dan abruptly made the following statement as we discussed his failure to pray: “It's not ‘disappointment’—there isn't any. Because all human beings have no reason to be disappointed in God. None. Supposing a human being says that he is disappointed in God, it means that he's crazy. Yes, according to my way of thinking, there is nothing that God does that is disappointing.” We had earlier discussed emotion terms, including the one translated here as “disappointment”(kecewa), but nothing in that conversation had suggested that Da Dan's lack of prayer had anything to do with disappointment, with God or anything else. Something in the web of meanings attached to prayer in Da Dan's own mind had brought this out.

His words resonated with those of another respondent, Ni Gan, a woman of about thirty. Ni Gan regularly used our conversations to agonize to me over her troubles, which were extensive. Her finances were precarious, her marriage unhappy, and her relations with family and neighbors strained. She commented to me that it was remarkable that she often neglected to pray, despite her troubles. She pointed out that, logically, she should be especially attentive to prayer when troubled, but she found that just the opposite was true. “I should get close to God, right? I don’t, I keep my distance. I do believe—I have conviction … but I have a lot of problems to think about, to the point that I don't pray.” After again recounting the details of her troubles, she continued, “It's as if I’m angry [bangih] with God, I feel.²⁴ Because, why is it that it comes down on me only? I have so many problems. I should, I should draw myself nearer to God, right, so that God provides relief, offers a path. But it's not like this. To the contrary, I don't draw closer. Yes, it's wrong, I know it's wrong. I’m certain I’m … wrong, but what can I do? This heart in here truly isn't tanang.”

Recall that tanang is the word used to describe the experience of moral soundness that results from prayer—and also, paradoxically, to describe the preconditions for genuine prayer. The problem, Ni Gan explained to me, was that her mind was constantly occupied with her troubles. She tried to forget them, but they remained inside the very core of her ati, her heart. She found it impossible to pray in the proper subjective state in which all of her troubles would be pushed out of consciousness, leaving room only for God and perfect integration with God—an ideologically more authentic state of subjectivity in which there is no suffering. Instead, those troubled thoughts continued to well up inside her—her pain and her anger seem genuine to her. They matter to her and define who she is in a directly embodied way, but they run up against the ideology of the practice of prayer. “I’m going to pray,” she said. “I think I’m going to pray, then I think of things again. So it feels like it's a waste to pray. I’m praying, but my mind isn't on praying. I’m just wearing myself out bobbing up and down.” Shalat often fails to reconcile Ni Gan's spontaneous subjective state with the conventional state that is supposed to constitute her moral self. Prayer could form her into a moral subject, but, for her to truly pray, moral subjectivity must first arise from within her, and often it simply does not.

Ideology and experience often clash in regard to shalat. Recall Da Palo's account of a day ordered by ritual prayer: Prayer structured getting up in the morning for work, working, taking breaks from working, going home from work, and going to sleep in time to be well-rested for the following day's work. This was not only an account of a pious self but also an account of a self that has realized its proper potential in the marketplace. His account demonstrates a merging of religious and economic elements of ideology. Nevertheless, the day Da Palo decribes does not match his own typical day at all, and not merely because he does not pray consistently. A skilled laborer who is rarely employed, Da Palo, like many men, has no job to attend to in the morning, no need to take a break or turn in early. About an hour before he described this structured day to me, I had woken him from his night's sleep when I arrived at his house in the late morning.

Such gaps between the ideology and the experience of practice fall outside of Mahmood's gaze, but her ethnographic data does invite consideration of their significance. She writes of a young woman who complains that she has trouble getting herself to pray (Mahmood 2005:124–126). A member of the women's piety movement tells her that she must put her self in a state of piety at every single moment of her life and that, if she does this, she will find that she naturally engages in the practice whenever she should. Mahmood does not recount the woman's reaction to this impossible (if logical) advice, nor does she address whether the women in the piety movement find themselves able to realize it. The problem and the attempt to transcend it are clear, but the way such solutions play out in experiences is less so.

Drawing on an approach inspired by the work of Foucault, Mahmood builds from the premise that people's selves and subjectivities are constituted entirely by the relations of power inherent in a particular set of discourses and practices. In the piety movement, through embodied practices such as prayer, selves and subjectivities are constituted without the divides between convention and spontaneity, inner authenticity and imposed restrictions, that are assumed in most Western theory. Ultimately, she concludes that the way Islam has been elaborated within the piety movement poses a challenge to the very concepts of “freedom” and “autonomy” central to Western liberal ideologies: What does freedom or autonomy mean when moral selves are those that embody a wholly conventional will?

Mahmood acknowledges that the discursive tradition of the piety movement is not the only one circulating in Egypt and that different discourses and differently constituted subjects may exist within a single cultural context (2005:131–133). She contrasts the movement's ideology of prayer with one that situates piety's role in terms of modern citizenship, emphasizing that, although the piety movement is not overtly political in a traditional sense, it is deeply political in offering a particular argument for how persons become proper subjects (cf. Hirschkind 2006).²⁵ Her concern is thus with the coherence of the ideology expressed in answer to the woman's complaint: Through embodied practices such as prayer, a person constitutes a moral self. Mahmood justifiably maintains that “understanding the coherence of a discursive tradition” is “a necessary step toward explaining the force that a discourse commands” (2005:17).

Recently in West Sumatra there have been somewhat sporadic but growing efforts to mandate certain forms of Islamic piety, for example, a working knowledge of prayer, participation by students in Islamic education or group prayer, or an ability to read the Koran. The shift toward greater regional autonomy in Indonesia since 2000 has enabled regional and local governments in West Sumatra to institute a patchwork of laws making such demonstrations of piety prerequisites for school graduation, marriage, or civil service employment. In 2007, the provincial legislature passed a regulation to spread similar requirements throughout West Sumatra as a whole.²⁶ Girls and women in school or working in government offices have been told, under similar laws, that they must wear head scarves. The arguments supporting these requirements are coherent: Such embodied practices cultivate moral selves and should therefore be required of citizens (at least Muslim ones) in order to create a moral population. However, compelling individuals to be pious risks undermining piety altogether by making pious practices the kind of empty rituals of identity that bothered the women Mahmood studied in Cairo, turning piety into a tool and undercutting its value as a genuine engagement with the sacred (cf. Hooker and Huda 2007).

This idea also troubled many of the people I talked to in Bukittinggi, who constantly complained that so much of public religious life was hypocritical, a mere show. Ni Yas, the middle-aged woman who had difficulty praying when she felt troubled, argued to me that a woman should wear a veil only because her heart is clean: Women should veil on the outside, she said, only if they are “veiled” on the inside, seeing this as the central issue rather than the effect veiling might have on creating moral selves.²⁷ The frictions inherent in mandating moral practice were expressed starkly by Da Palo, who had his own struggles with skipping prayers, drinking, and other moral lapses. In a statement that points at the wider political landscape to which these issues are relevant, when his wife nagged him to reform his behavior, he argued that in Islam it was immoral to force morality on others. He equated her approach to that of Jemaah Islamiyah, the radical Islamic organization in Southeast Asia that has been blamed for many of the bombings in Indonesia over the past decade: Just as that group erred in trying to “purify the world” through violence, his wife was attempting the self-defeating task of forcing him to behave morally.

Discussion: Islam, subjectivity, and the conflicting demands of moral selfhood

As several recent anthropological meditations have argued, the ethnographic utility of the concept of “subjectivity” derives from its situation at the intersection of sociocultural structures and individual experience and agency (Biehl et al. 2007; Luhrmann 2006; Ortner 2005). By invoking both the idea that individuals are subject to the shaping and controlling powers of social and cultural forces and the idea that the world is subject to the powers of individuals to interpret it and act on it, “subjectivity” can transcend the opposition between culture or structures of power and human consciousness and agency—or, at least, invite an examination of the entanglement between them.²⁸

This means that any investigation of subjectivity requires us, as ethnographers, to look through a “double lens” (Linger 2005), examining individual experience as well as the social and cultural structures in which it is embedded. It is possible to significantly engage with such experience by taking seriously how individuals move through their social worlds and what they tell us about what it is like for them to do so. Such ethnographic studies attempt to “represent human behavior and subjective experience from the point of view of the acting, intending, and attentive subject, to actively explore the emotional saliency and motivational force of cultural beliefs and symbols (rather than to assume such saliency and force)” (Hollan 2001:48–49).²⁹ To read subjectivity directly from public discourse is not to investigate subjectivity at all. A small effort to inform ethnographic investigation with attention to individual experience, such as I have offered above, reveals it in significantly different light.

For one thing, seeing cultural discourses as grounded in the individuals that engage with them prevents us from easily dividing them into separate, abstract lines of logic that clash only in a disembodied way. Multiple lines of discourse in West Sumatra define moral selves in Islamic (as well as non-Islamic) terms. Understanding the coherence of these discourses is a necessary step toward understanding Islamic subjectivity, allowing us to see how discourses clash and compete politically and ideologically. However, this is not a final step. It does not yet tell us what subjective force a discourse commands or the tensions that arise when it is embodied or applied. Conflicting discourses do not necessarily create alternate subjectivities and alternate selves that circulate separately within Islamic communities but may become intertwined in the active processes of management that mark selfhood (cf. Marsden 2005:163–164 n. 14). Selves, subjectivities, and the politics that flow from them may be constituted through efforts to grapple with irresolvable tensions. This is the difference between analyzing “subject positions” and analyzing “the formation of subjectivities, complex structures of thought, feeling, reflection, and the like, that make social beings always more than the occupants of particular positions and the holders of particular identities” (Ortner 2005:37).

The tensions that mark subjectivity are, Sherry Ortner notes, often central to discussions of the fractured “postmodern” condition, the analysis of which often fits into a larger pattern of cultural analysis found in Max Weber's work and particularly in the Geertzian interpretive anthropology that was influenced by it: Subjectivities form through engagement with “issues of anxiety over meaning and order” that arise in the course of living (2005:41). Such anxieties may also be particularly acute for members of dominated groups, whose “complex subjectivity … arises from the emotional experience of negotiating contradictory emotional codes” (Luhrmann 2006:356) to interact with members of dominating and dominated social groups. As T. M. Luhrmann explains (using homeless Americans as her example), such tension is not merely a matter of multiple moral discourses circulating in a single society: “The emotional vise arises not only out of there being two codes [but also from] a recognition that in being responsive to each, one fails both” (2006:356).

These insights regarding the way subjectivities become structured around contradiction and uncertainty can be extended beyond their application to cases of postmodernity or oppression. Cultural and moral systems are often built around contradictions and attempts to manage them (e.g., Laidlaw 1995; Nuckolls 1996, 1998; Parish 1996). The analysis of moral selfhood in West Sumatra that I have presented above suggests that social experience—and even “Islam” itself—presents people with conflicting demands that call for simultaneously divergent appraisals of the experience of self and different reactions to and representations of those experiences. Moral selfhood comes in managing the conflicting demands so that one finds ways to succeed in satisfying each without failing the other.

The ritual of shalat, I have argued, crystallizes moral tensions that pervade everyday life in West Sumatra, making those tensions “more thinkable” (Jackson 2005:95). This does not mean that each individual comes to understand or experience the tensions in the same way. For some, shalat provides an arena in which a coherent moral self can settle and tensions can be transcended. For others, in Michael Jackson's (2005:xxvii) terms, the “sense” of shalat overflows its socially accepted “significance,” refusing a secure place within prescribed discourses of moral selfhood. In such instances, Jackson notes, echoing Da Dan's comments on his own failure to pray, people often remark that things simply “don't make sense” (2005:xxviii). Jackson is primarily interested in attending to such overflow as it plays out in an “event”: “an occasion, a happening, where something vital is at play and at risk, when something memorable or momentous is undergone, and where questions of right and wrongful conduct are felt to be matters of life and death” (2005:xxix–xxx). Shalat may be too routinized to be seen as an event, yet, for many people I spoke with in West Sumatra, in any given instance of shalat, something vital was at play, and at risk, and the questions raised were matters of the deepest and most far-reaching moral significance. Everyday subjectivity may struggle with social meaning too.

Understood as part of a dynamic engagement with experiencing subjects, it becomes more difficult to read the ideologies of prayer in West Sumatra or among the women in the Egyptian piety movement as more than rhetorical challenges to the notion that spontaneity and conventionality exist in tension. Instead, these ideologies of transcendence would appear to gain traction as they are used to address that very tension, one that is never permanently transcended or reconciled. Spontaneity and conventionality do sometimes merge, but they can never remain permanently joined; neither can they ever be wholly untangled. The ideology of the piety movement and the analysis used by Mahmood mirror each other in the way they attempt to deal with this dilemma: They attempt to transcend the opposition by making one term subordinate to the other. The reflective self is completely swallowed by the relational self, which becomes the sole locus of authenticity and morality—or the sole locus of authentic social analysis.

The ethnographic material from West Sumatra presented above suggests that moral selfhood remains a compelling struggle for people because of the complex and contradictory sources from which selfhood is built and the contradictory demands placed on the self in the course of living. It further suggests that Islam, or any particular set of Islamic discourses and practices, does not manifest itself in isolation, adopted by individuals in pure form to fashion the self in a particular way, but gets pulled into this same complex web. Islamic discourses and practices themselves seem to speak to such contradictions, their endurance, and the constant struggle to transcend them. Although it is possible and important to see prayer as a node of practice situated within particular sets of relations of power, employed to cultivate particular subjectivities of moral selfhood, it is also important to see it as a node in the experience of uncertainty, tension, and anxiety that accompany them. As Magnus Marsden argues, ethnographic accounts of Islamic societies that focus narrowly on the moral logic of self-discipline are characterized most strongly by the quality of “coherence,” and so “are unable to confront the ways in which Muslims are called upon to face, explain and contend with inconsistencies and complexities in their attempts to live virtuous lives” (2005:260–261). Engaging in Islam's ritual practices does not lead in a direct way to the constitution of a particular moral self, even if a discourse suggesting that it does is accepted as logical and true, but such practice does encourage an embodied engagement with such complexities and tensions.

The way Islamic ideas and practices participate in constituting selfhood, and the politics that flow from this, take particular forms in each Islamic society. In Indonesia, such questions of subjectivity necessarily intersect with political battles over the proper institutionalization of Islam's role in the public sphere and its relation to democratic practices. The battles were shaped in large part by the manipulations of the Suharto regime but have now burst into more open and vital form (Hefner 2000). The perspective I have offered above is important because, if scholars are going to understand Islam not as an abstract thing but as something indivisible from the lives of Muslim people, we need to examine not only the logic and coherence of particular Islamic discourses, and these political battles, but also how and why these discourses become embodied and experienced by the people who engage with them.

Notes

Acknowledgments

Acknowledgments.  The research on which this article is based was supported by the National Science Foundation under Grant No. 0130317; the Department of Anthropology at the University of California, San Diego; and the United States–Indonesia Society. I also wish to extend my thanks to Steven Parish and Jenny Simon, both of whom kindly read drafts of this article and provided insights that helped me to improve it. I am most grateful to all of the many people in Indonesia whose support and patience made my research possible. My special appreciation goes to the School of Social and Political Sciences at Andalas University in Padang, Indonesia, and to the Department of Anthropology, in particular, for inviting me to present and discuss a version of this article in June 2008. Any opinions, findings, and conclusions or recommendations expressed in this article are my own and do not necessarily reflect the views of the National Science Foundation or any other person or institution that provided me with support or advice.

1. The terms person and self are not defined consistently in anthropological literature (Harris 1989; Lindholm 1997; Spiro 1993) and their usage often overlaps, making it difficult to engage with a range of literature that uses these terms without furthering the confusion or being diverted into long exposition on definitions. Broadly, person is most fruitfully used to indicate the outlines of individuals’ particular social and moral relationships with others and with larger social and metaphysical collectives. A “self” can better be understood as a person's living of his or her own particular personhood. This would imply that a self is “a locus of experience, including experience of that human's own someoneness” (Harris 1989:601; cf. Hollan 1992), and also that this organizing of experience comes about as an intrinsic part of the process of living as a social actor. In Steven Parish's (n.d.) terms, the self is a “vehicle for living.” I am concerned in this article with conceptions of personhood but also with the way people subjectively experience their own personhood. For the sake of simplicity, I favor the use of selfhood throughout the article.

2. To take one typical example: Does human existence cultivate value through the practice of rationality and containment of passions or through overcoming the restrictions of rationality to let loose creative passions?

3. Also drawing from Friedrich Nietzsche, he more particularly identifies morality with “ethical systems where self-denying values inform law-like obligations” (Laidlaw 2002:317). It should become clear in the ethnographic material in this article that Islamic prayer does involve moral practice in this narrower sense of the term but also that an investigation of the moral or ethical context in which prayer is embedded takes one beyond the fact of prayer as a self-denying obligation. The distinction between “morality” and “ethics” is not consistent in philosophical, or anthropological, literature.

4. The other four “pillars” of Islam are the profession of faith, fasting during the month of Ramadan, the giving of alms, and the pilgrimage to Mecca for those who are financially and physically able.

5. When people in West Sumatra wanted to inquire about my own, non-Islamic religious practices, they virtually always began by asking me, “How do you pray?”

6. The words recited are, of course, taken directly from the Koran and are, therefore, in Arabic. The general meaning of the first sura is commonly known to Minangkabau people, but very few Minangkabau people understand Arabic. It is considered more important to recite the words properly, and to recite them with the proper intention, than it is to understand the words themselves. Similarly, when Minangkabau children learn to recite the Koran, they learn the proper pronunciation and technique but do not learn how to comprehend Arabic.

7. Shalat varies slightly according to which of the daily prayers is being performed.

8. This latter sort of prayer is classified as a kind of doa, a broad term for prayer, rather than as shalat (cf. Parkin 2000b). John Bowen (1989) highlights this difference by referring to forms of doa as “prayer” and to shalat as “worship,” a reasonable and useful distinction also noted by Stephen C. Headley (2000:215). I have fallen back on the use of prayer here as this is the more common usage in the literature. Throughout this article, prayer refers specifically to shalat and not to other forms of prayer, unless otherwise noted. In Minangkabau, shalat is also referred to as sumbayang.

9. All names used are pseudonyms. Interview materials are taken from audio recordings, and the translations are my own. I generally conducted interviews one-on-one in the Minangkabau language or in a mixture of Minangkabau and the Indonesian national language. The two languages are closely related and the boundary between them is porous. In West Sumatra, many elements of public discourse, including that pertaining to religion, are frequently discussed in Indonesian or in a mixture of the two languages.

10. The distinction between “soul” and “spirit” here is not significant, and Ni Galeh is not referring to two different parts of a person but, rather, she is referring in two different ways to the same, vaguely defined dimension of an individual. Batin, the term translated as “soul,” is derived from Islamic discourses that more broadly refer to an “inner,” or metaphysical, realm of existence. However, people often use the term as a shorthand for the metaphysical dimension of an individual person's being—that is, something like a “soul” (cf. Bowen 1993; Geertz 1984). Jiwa, the term translated as “spirit,” is a word with Sanskrit roots that is used to refer to the noncorporeal entities that exist inside human beings and give them life and agency.

11. People interpreted this state of tanang for me using a variety of descriptions, including “there's no feeling of anxiety,”“we feel close to God,”“my life is pleasant,” and “my heart is happy.” Ni Tasi, who I quote again below, explained it to me in the following way: “That is the wonder of prayer—that's it. As tangled up as my heart is, hey, I go pray—God willing, my spirit will be tanang”[Sakusuik-kusuik ati awak, eh, pai sumbayang—insyalah tanang jiwa awak]. Charles Hirschkind (2006:ch. 3) discusses the states of “calm,”“tranquility,”“relaxation,” and “closeness to God” that are said to result from listening to audio recordings of sermons in Cairo, Egypt, and the claims that such states must be differentiated from the calm or relaxation that results from engaging with secular material such as popular music.

12. The name Padri is apparently derived from the port city of Pedir, located at the northern tip of Sumatra in what is now the province of Aceh. The port served as the point of embarkation for Minang pilgrims to Mecca (Ricklefs 2001).

13. Official city statistics in 2002 recorded over 40 percent of the working population in Bukittinggi to be involved in trade (perdagangan), by far the largest category represented (Badan Pusat Statistik Kota Bukittinggi and Badan Pembangunan Daerah Kota Bukittinggi 2002:38).

14. With the particular language and practices of a seventh-century Arabian society playing a central role in Islam's definition, its “universal” or “global” elements, how they are made local, and how the particulars of local social life are to be reshaped according to these elements are questions that are central to issues of Islamic identity and practice around the world (Bowen 1998; Lambek 2000b).

15. For more detailed Indonesian and Malay examples, see Siegel 1969, Woodward 1989, Brenner 1995, Peletz 1996, and Simon 2007. Rosen 1984 discusses a Moroccan context and Anderson 1985 an Afghan one. Tritton 1971 offers general conceptual background.

16. Although aka and napasu were regularly discussed by my subjects, they did not act as the master terms that they appear to be in Michael Peletz's (1995, 1996) accounts of Malays in Negeri Sembilan or in James Siegel's (1969) account of Aceh. Most ethnographers of Minangkabau society say little or nothing about napasu and aka. For example, the terms are nearly absent in two recent ethnographies focused on Minangkabau village women, penned by Evelyn Blackwood (2000) and Peggy Sanday (2002). Blackwood writes that the villagers she knew did not use these terms with her and notes that other, although not all, ethnographers of Minangkabau society also report their absence (2000:49; cf. Kahn 1993:126). Residents of Bukittinggi may be somewhat more likely to employ these concepts (perhaps by virtue of their greater exposure to national and international Islamic discourses) than are the residents of villages where most West Sumatran ethnographies are set, although this is only speculation.

17. For example, shame (malu) and the threat of shame appear to play a large role in people's attempts to present themselves as biaso, or normal, rather than stand out as individuals (cf. Fessler 1995).

18.  Taimpik nak di ateh, takuruang nak di lua.

19.  Paribadi, in Minangkabau, can be used as a noun (e.g., as in paribadinyo, to refer to one's “personality”) or as an adjective (e.g., to indicate that something is “personal”).

20. Transformed into modes of being, parallel dimensions emerge in an existential approach that “distinguishes three modes of being-in-the-world: eigenwelt (the mode of being in relation to oneself), mitwelt (the mode of being in relation with others), and umwelt (the mode of being in relation to one's biological or physical environment)” (Jackson 2005:32).

21. In the context of Indonesian Islam, “traditionalist” is a label applied to those seen as more accommodating of indigenous cultural influences within religious practice. “Modernists,” in contrast, are those who have emphasized the need to reform cultural practices to more faithfully and purely adhere to Islamic ideals.

22.  Hirschkind (2001, 2006) offers a parallel argument about how the process of listening to audio recordings of sermons shapes particular kinds of pious dispositions among Muslims in Cairo.

23. These are people who did not appear greatly bothered by their failure to pray. I never heard anyone attempt to justify not praying (certainly no one ever suggested that his or her troubles with prayer constituted an acceptable excuse for not praying), and no one I spoke with rejected the notion that prayer is essential. At the same time, it seemed to me that some of the people I spoke with who did not pray regularly were only troubled by their failure to pray because I was asking them to discuss it. They had no way to publicly justify their lack of practice. It is probably significant that those who—by my own estimation and the general consensus of people in Bukittinggi—are least likely to pray or to agonize over lack of prayer are younger men. These are the same people who are encouraged by Minangkabau social norms and economic realities to break away from the bounds of village and kinship and most overtly assert themselves as autonomous actors, as individuals who can succeed in the competitive marketplace through individual initiative.

24. When I later asked her about this remark, she denied that she had any anger toward God but did acknowledge being kesal, annoyed or fed up, with God.

25. This political dimension is the reason, Mahmood asserts, that the Egyptian state has shown discomfort with the movement, although she does not elaborate the details of the tension between the state and the piety movement in her 2005 monograph.

26. The extent of enforcement of these laws is not yet clear.

27. Discussions about veiling in West Sumatra follow some of the same patterns as discussions of prayer, pointing toward the larger issues of morality and Islamic practice that they share. It is generally considered morally proper for women to cover their heads. However, only something like half of the women in Bukittinggi did so on an everyday basis in 2002 and 2003 (although the percentage had probably risen a bit higher by 2008, when I returned to Bukittinggi for a short visit). Further, this represents a far greater portion of the population than had veiled as recently as the 1990s. Despite the laws mentioned in the text, there is widespread sentiment that veiling or not veiling must be an individual choice and cannot be forced (cf. Brenner 1996; Smith-Hefner 2007). People frequently told me that they were disturbed by young women who veiled despite behaving improperly in other ways. The hypocrisy in these cases is generally thought to be far worse than the failure to veil. As this article suggests, the failure to pray is more problematic, as—far more than veiling—it is a practice considered absolutely fundamental to being a Muslim.

28. This is no doubt why attention to “subjectivity” has focused especially on cases of suffering and oppression, as mentioned subsequently in the text. Such cases cry out for appreciation of the structures of power from which suffering flows. They may tempt scholars to focus on these structures at the exclusion of all else. Employing “subjectivity” as a framework allows us to take questions of experience and agency seriously without placing them in opposition to questions of subjugation. Attending to subjectivity is necessary for rich ethnographic accounts of human social life in any circumstance, and we should be careful not to reify the distinction between dominated groups and others, as if members of what we identify as dominated groups somehow have a different order of, or even “more,” subjectivity than others.

29.  Hollan (2001:48) points out that the established term for such studies, person centered, is misleading in that such ethnographies are not generally concerned with “the person” and could be more accurately described as “individual” or “subject” centered.