African Women, Religion and climate

GENDER & SOCIETY, Vol. 29 No. 6, December 2015 982 –1008 DOI: 10.1177/0891243215602106 © 2015 by The Author(s)

Women’s religious authority in a sub-saharan setting:

Dialectics of empowerment and Dependency

Victor AgAdjAniAn University of Kansas, USA

Western scholarship on religion and gender has devoted considerable attention to wom- en’s entry into leadership roles across various religious traditions and denominations. However, very little is known about the dynamics of women’s religious authority and leadership in developing settings, especially in sub-Saharan Africa, a region of powerful and diverse religious expressions. this study employs a combination of uniquely rich and diverse data to examine women’s formal religious authority in a predominantly christian setting in Mozambique. i first use survey data to test hypotheses regarding the prevalence and patterns of women’s formal leadership across different denominational groups. i then support and extend the quantitative results with insights on pathways and consequences of women’s ascent to formal congregation authority drawn from qualitative data. the analysis illustrates how women’s religious authority both defies and reasserts the gen- dered constraints of the religious marketplace and the broader gender ideology in this developing context.

Keywords: religion; women’s leadership; gender ideology; sub-Saharan Africa

Women’s rise to formal leadership roles in religious institutions in Western settings has attracted considerable scholarly attention. In the United States, where this topic has generated the most interest, much

AUThOR’S NOTE: the support of the Eunice Kennedy Shriver national institute of child Health and Human development (nicHd) grant # r01Hd050175 for data collec- tion is gratefully acknowledged. i am thankful to cecilia Menjívar for valuable comments and suggestions. correspondence concerning this article should be addressed to Victor Agadjanian, Foundation distinguished Professor, department of Sociology, University of Kansas, Lawrence, KS 60045, USA; e-mail: [email protected].

602106GASXXX10.1177/0891243215602106Gender & SocietyWomen’s Religious Authority research-article2015

Agadjanian / WOMEN’S RELIGIOUS AUTHORITY 983

research has examined ordination of women in various denominations, which started in earnest in the nineteenth century and has both reflected and ushered in major changes in religious organization and activities (e.g., Bednarowski 1980; Dodson 2002; Lehman 1985; Ruether and McLaughlin 1979). This scholarship has produced research focused on women’s lead- ership in marginal religions (Wessinger 1993), women’s leadership in mainstream denominational traditions (Wessinger 1996)—and other top- ics (e.g., Chaves 1997; Nesbitt 1997; Schmidt 1996; Zikmund, Lummis, and Chang 1998). Using diverse theoretical perspectives and a variety of data sources, this research documents and analyzes women’s growing entry into religious leadership, the broader societal changes that promote this entry, and the consequences of increased women’s leadership for reli- gious congregations and for society at large.

Fewer attempts have been made to examine women’s ascension to for- mal religious leadership in contemporary developing settings, especially for sub-Saharan Africa, a region of massive religious complexity and vigorous and multifaceted social change. The literature on religion and gender in such settings typically focuses on the effects of women’s reli- gious participation on their informal authority within the religious realm (e.g., Bano and Kalmbach 2012; van Doorn-harder 2006) and their status, opportunities, and engagement outside of it (e.g., Agadjanian and Yabiku, forthcoming; Brusco 1995; Coleman 2010; Rinaldo 2013), but glosses over women’s formal involvement in religious organizational leadership and the challenges and barriers surrounding this involvement. Using a combination of unique quantitative and qualitative data, this study helps to fill this important gap by examining patterns and mechanisms of wom- en’s formal religious authority, that is, organizational authority sanctioned through formal appointment or election, in a predominantly Christian sub- Saharan setting. Quantitative analyses demonstrate considerable denomi- national variations in women’s presence among congregation leaders, yet also the limits of that presence regardless of denominational type. Qualitative analyses elucidate women’s pathways to formal religious authority as well as constraints that the patriarchal gender ideology con- tinues to impose on women church leaders.

WOMEN IN FORMAL RELIGIOUS LEAdERSHIp: A REvIEW

My broader theoretical approach is inspired by the cross-national scholarship on how women employ religious doctrinal and organizational tools in their struggles for equality and advancement. This literature uses

984 GENdER & SOCIETY / december 2015

both historical and contemporary cases to argue that religious piety and religious organizational engagement are used by women to transform the religious realm from within, by reinterpreting and repositioning religious teachings and norms so as to enable and promote women’s agency and empowerment (e.g., Avishai 2008; Bartkowski and Read 2003; Chong 2008; Dodson 2002; González 2013; Khurshid 2015; Mahmood 2005; Prickett 2015; Rinaldo 2013; van Doorn-harder 2006).

More specifically, however, my conceptualization draws from the body of interdisciplinary studies that analyze causes, processes, and conse- quences of women’s engagement in formal religious leadership in the U.S. and other Western settings. This research shows that the roads to religious leadership in Western settings have differed across denominations, reflect- ing unique denominational norms, rules, and circumstances. Thus, main- stream Protestant denominations and progressive groups within Judaism became increasingly amenable to women’s formal leadership roles as early as the eighteenth century, with a particularly rapid expansion of these roles in the twentieth century (Charlton 1997; Chaves 1996; Larson 1999; Lehman 1985; Marder 1996; Nadell 1998; Zikmund, Lummis, and Chang 1998). Women’s entry into leadership positions was also histori- cally more common in non-mainstream religious movements, such as Quakerism, Shakerism, Spiritualism, Christian Science, and Theosophy, which are typically characterized by deemphasizing a masculine deity and the doctrine of the Fall, denial of the need for formally ordained clergy, and asserting women’s roles outside the sphere of marriage and mother- hood (Bednarowski 1980; Larson 1999; Plant 2003). In contemporary Pentecostal churches, despite these churches’ generally conservative, patriarchal theological and social narratives, women often rise to positions of considerable influence and authority through charisma invigorated by God’s calling, especially in churches with no fixed ordination rules (Lawless 1993). Finally, the Roman Catholic Church has historically banned ordination of women as priests and this ban has persisted despite a growing critique from both within and outside that denomination. Yet, even the Catholic Church has seen a dramatic expansion of laywomen ministry (Ecklund 2006; Flinn 1996). Catholic women’s rise as de facto congregation leaders has become possible in part due to an increasing shortage of priests (Wallace 1992).

In spite of the expansion in women’s religious leadership, research has shown persistent barriers to women’s ordination even in more pro- gressive denominations (Lehman 1980; Marder 1996; Nesbitt 1993; Tucker 1996). According to Chaves (1996), increased women’s ordination

Agadjanian / WOMEN’S RELIGIOUS AUTHORITY 985

has reflected religious organizations’ adjustment to external political and institutional pressures, in particular, those involving women’s rights and gender equality, but also internal organizational and sociocultural charac- teristics of different denominations. Whereas these external and internal pressures helped pave the way for women’s entry into religious leadership positions, they also often clashed with the religious organizations’ internal goals and priorities. As a result, women’s formal leadership, while no longer opposed in principle, often remained limited to subordinate roles (Chang 1997; Chaves 1997; Nesbitt 1997; Schmidt 1996; Sullins 2000), and women’s leadership posts often revert to men after women leaders leave the church scene (Baer 1993; Wessinger 1993, 2)

Analytic Approach

I formulate several hypotheses about women in congregation leader- ship in the sub-Saharan setting under study. First, I hypothesize that women’s presence in formal leadership positions would be most promi- nent in ideologically more liberal and organizationally more flexible denominations (hypothesis 1). Next, I consider the size and gender com- position of congregation membership, hypothesizing that regardless of denomination women would be more likely to lead smaller congrega- tions (hypothesis 2) and congregations with a higher women-to-men ratio of members (hypothesis 3). Because the wealth of a congregation is an important marker of its leader’s prestige, men should be particularly reluctant to cede leadership to women in more established, affluent con- gregations. Therefore, I expect that wealthier congregations, regardless of other characteristics, would be less likely to have a woman leader (hypothesis 4). With respect to urban–rural differences I propose two competing hypotheses. On the one hand, assuming that the urban envi- ronment connotes greater overall gender equality, urban congregations should be more likely to have a woman leader (hypothesis 5a). On the other hand, however, if urban congregation leadership commands greater prestige, it should be more attractive to men; we therefore would see fewer women among congregation leaders in urban areas, net of other factors (hypothesis 5b).

With regard to differences between men and women congregation lead- ers, I focus on leaders’ formal religious training, typically in a Bible school or another theological program, and their ranks in the church hierarchy. Thus, I hypothesize that women leaders should be less likely to have received any formal religious training net of other characteristics (hypoth- esis 6). Finally, I also expect to find women leaders underrepresented in

986 GENdER & SOCIETY / december 2015

higher-ranking positions (e.g., bishop, superintendent, priest, pastor, or minister) even in churches where such positions are not formally closed to women, regardless of other congregation characteristics and of leader’s formal religious training (hypothesis 7).

I test these hypotheses using survey data and then complement the statistical tests with insights from qualitative data on how women’s formal authority is established, exercised, and constrained across different denominations. Among its other contributions, qualitative analysis allows me to distinguish and compare two subtypes of women’s authority of office: office authority that women achieve on their own and office authority that they gain by virtue of association with husband church lead- ers, that is, by sharing their husband’s authority or inheriting it after their husband’s death.

SETTING

The data come from the district of Chibuto in southern Mozambique, an impoverished nation located in Southeast Africa. With a population of about 220,000, Chibuto is in many respects typical for Mozambique and the rest of the subcontinent. It is largely mono-ethnic Changana-speaking, with patrilineal kinship system and bridewealth-based marriage: Upon the transfer of bridewealth, married women’s productive and reproductive capacities and outputs belong to their husband’s families. The district’s economy is pivoted on low-yield subsistence agriculture: Although women play a major role in agricultural production, they do not inherit land. The area’s proximity to South Africa, Mozambique’s much more prosperous neighbor, and to Maputo, Mozambique’s capital city, has made men’s labor migration a vital source of livelihood for its population (De Vletter 2007). In this setting, mass men’s labor migration and the incomes that it generates overall have helped to maintain the traditional gender ideology by perpetuating women’s economic dependence on men (Loforte 2000).

Although Mozambique has a large Muslim minority, Chibuto district, as much of Mozambique’s south, is overwhelmingly Christian. Before Mozambique’s independence from Portugal in 1975, the area’s religious landscape was largely defined by the Roman Catholic Church and by what I define as traditional Protestant and Evangelical churches (hereafter, tra- ditional Protestant churches). The Catholic Church was the quasi-official church of the Portuguese colonial empire. Protestant missions, such as those of the Presbyterian, Methodist, Anglican, and Baptist churches, and

Agadjanian / WOMEN’S RELIGIOUS AUTHORITY 987

the Church of the Nazarene were also established during the colonial years and played an important role in forming the indigenous educated elites that later led the national liberation struggle (Cruz e Silva 2001). Despite the pre-eminence of the Catholic Church and considerable influ- ence of traditional Protestant churches throughout most of the twentieth century, these churches have lost much of the colonial-era clout and, with it, much of their membership after the nation’s independence (Morier- Genoud and Anouilh 2013).

The Catholic and traditional Protestant churches’ relative decline was paralleled—and largely caused—by the growth of Pentecostal and Charismatic churches. The rising tide of Pentecostal and Charismatic Christianity, reflecting broader processes of postcolonial development, is a subcontinent-wide phenomenon (Garrard 2009; Kalu 2003). Despite (or maybe because of) its magnitude, standard definitions and classifications of this array of churches remain elusive (Anderson 2002; Garrard 2009). here, I distinguish between generally older, African-initiated or heavily Africanized churches on the one hand and more recent, global neo-Pente- costal churches on the other. The churches of the former variety sprang up and propagated in southern Africa since the first half of the twentieth century. Zionist churches and Apostolic churches have been particularly influential among them. These churches first penetrated Mozambique from South Africa, often introduced by returning migrants, but they truly blossomed after the independence, and especially after the onset of neo- liberal economic reforms in the early 1990s, finding a rich pool of poten- tial converts among disenfranchised peasant masses nominally (and often forcibly) affiliated with Catholicism and traditional Protestantism. While formally and vocally distancing themselves from local pre-Christian beliefs and practices, both Apostolic and Zionist churches have de facto embraced the traditional notions of illness and misfortune as caused by evil spirits and correspondingly adapted miracle healing as a weapon to fight those spirits (Pfeiffer 2002).

The appeal of Zionist churches, centered on prophesying, tongue- speaking, and divine cure powered by the holy Spirit, was particularly strong and, accordingly, their numeric rise was especially spectacular. Thus, according to Mozambique’s 1997 general census, in the nation’s south Zionists made up 40 percent of the population, or three times the share of Catholics. Even after a slightly lower percentage (35 percent) recorded in the 2007 census, Zionists remained the largest denominational category (Instituto Nacional de Estatística 1999, 2009). Without delving into the theological and organizational complexity of Zionists, I should

988 GENdER & SOCIETY / december 2015

note that this conglomerate encompasses a large number of churches and is internally diverse and often ridden by fierce, even if veiled, ideological clashes and organizational rivalries (Agadjanian 1999). Apostolic churches, the other denominational block in that category, can be classi- fied as African-initiated Charismatics. While sharing many features with Zionist theology and practices, Apostolics are more institutionally intro- verted and have a distinctively corporate, rigid, and hierarchical organiza- tional structure which stands in stark contrast with generally loosely organized Zionist churches. Although Apostolic churches have not seen a numeric and demographic growth comparable to that of Zionists, they nonetheless have a noticeable presence on the area’s religious scene (Agadjanian 2012).

Finally, in the last few decades, Mozambique, as much of the subconti- nent, has seen an increasing proliferation of newer (neo-)Pentecostal denominations, such as the Assemblies of God and, more recently, transna- tionals like the Brazilian-origin Universal Church of the Kingdom of God (van de Kamp and van Dijk 2010). First implanted in urban areas and catering to the urban working and lower-middle classes, these churches have now been vigorously spreading into the countryside. These churches (to which I hereafter refer simply as Pentecostal) refute what they see as Zionists’ utilization of de facto traditional, ancestor spirit–based healing practices and, in response to the expectations of their ever more demanding audiences, go beyond Zionists’ narrow focus on physical cure by offering a wider array of social and economic wellness-enhancing solutions.

Regardless of denominational affiliation, location, and size, each con- gregation typically has one supreme leader (Jehovah’s Witnesses being an exception). These leaders may or may not have any formal religious train- ing and may occupy different ranks in their respective church hierarchies, for example, bishop, priest, pastor, minister, deacon, evangelist, or coor- dinator (animador in Portuguese).

METHOdS

The study uses a combination of quantitative and qualitative data col- lected between 2008 and 2014. For the statistical analysis of women’s lead- ership in the district’s religious congregations, I use data from a survey of religious congregations of Chibuto district (hereafter Chibuto congregation census) conducted by the collaborative research team of Arizona State University (USA) and Eduardo Mondlane University (Mozambique) mainly

Agadjanian / WOMEN’S RELIGIOUS AUTHORITY 989

in 2008 for the project “Religious Organizations in the Fight against hIV/ AIDS.” As part of the project, all religious congregations located within the district boundaries were enumerated and then in-person interviews were carried out with their formal leaders (i.e., individuals appointed or elected to head the congregation) or their deputies if the leaders were unavailable. In total, representatives of 1,126 congregations were interviewed, with a nearly 100 percent participation rate. An interview lasted approximately 45 minutes, on average, and covered the characteristics of congregation leader- ship and membership, characteristics of congregation facilities, basic doc- trinal and organizational aspects, and community-based activities deployed by the congregation (additional details on the census design and content are available from the authors upon request). I exclude three Muslim communi- ties from the analysis; I also exclude one community of Jehovah’s Witnesses because the nature of leadership in this denomination is vastly different from that in other churches. The basic characteristics of the surveyed con- gregations are provided in Table 1.

As part of the project, to complement the congregation census data, the research team conducted numerous focus group discussions and individ- ual in-depth interviews with both leaders and rank-and-file members of various congregations belonging to different denominations and located

Table 1: Characteristics of Religious Congregations, Chibuto Congregation Census (N=1122)

Characteristics

Denominational category (percent) Catholic 8.2 Traditional Protestant 15.7 Apostolic 10.3 Zionist 47.8 Pentecostal 18.0 Location (percent) Urban (district headquarters) 18.6 Rural 81.4 Type of congregation facility (percent) Built facility with tile or metal sheet roof 16.5 Built facility with reed/grass roof or no roof 43.3 No built facility 40.2 Number of attendees at last regular service (median) 19.0 Women-to-men ratio of last regular service attendees (median) 2.7 Number of congregations 1122

990 GENdER & SOCIETY / december 2015

in different parts of the district. These data covered a wide range of topics paralleling the content of the census. For the current analysis, I rely on data obtained from congregation leaders only: four focus group discussions (one in the district headquarters and three in different rural areas, each including 5-7 participants) and 24 individual interviews with leaders of various rural and urban congregations (some of whom also participated in focus groups). Both focus group discussions and individual interviews used in this analysis dealt with participants’ religious career trajectories, the nature and range of their activities, general and gender-specific challenges that they face in exercising those activities, and their strategies in navigating those chal- lenges. Finally, the quantitative and qualitative data are supported by fre- quent and prolonged field observations of church services and other events in numerous congregations of different denominational types. These obser- vations were conducted throughout the entire span of data collection.

I start the analysis with descriptive statistics from the congregation census data that outline the women’s involvement in church formal lead- ership positions (i.e., positions legitimized through appointment or elec- tion) across different types of denominations. I then fit a multivariate binomial logistic regression model predicting the likelihood of a congre- gation being led by a woman from the denominational type and other characteristics. Next, I fit logistic regression models predicting the char- acteristics of congregation leaders; in these analyses, leader’s gender is the predictor of interest. Finally, I use qualitative data to explore women’s paths to and experience of formal leadership positions. Because of the length limitation, I do not include quotes from the qualitative data and instead summarize participants’ narratives succinctly to illuminate my key arguments.

WOMEN IN CHURCH LEAdERSHIp IN CHIBUTO

Analysis of the Chibuto Congregation Census

Table 2 shows women’s share among congregation top leaders by denominational category from the census. Section A shows the percentage of women among congregation leaders. Section B displays the same per- centage only for congregations headed by a leader who occupies a higher rank in the church hierarchy, such as bishop, superintendent, priest, pastor, or minister. Forty-eight percent of all congregations were headed by such leaders; in the other 52 percent of congregations, the leaders held a low rank in the church hierarchy, for example, deacon, subdeacon, evangelist, or “coor- dinator.” As can be seen in column A, almost one third of all congregations

Agadjanian / WOMEN’S RELIGIOUS AUTHORITY 991

were led by women. The share of women leaders was highest among Catholics and traditional Protestants, but even among Pentecostals and Zionists it was around the overall average. Apostolic churches clearly stand out with a very low share of congregation leaders, under 4 percent. For higher-ranking leaders, women’s share among Catholics, not surprisingly, drops to 0 percent and among Apostolics to 2 percent; it also becomes noticeably lower than the overall share of women among congregation lead- ers in Pentecostal congregations. In traditional Protestant and Zionist con- gregations, however, the difference between the figures in the two columns is much less pronounced.

Column C of Table 2 shows percentages of leaders of the surveyed congregations who have received some formal training (e.g., Bible school, training seminars, etc.) by denominational type.1 Women leaders overall have a somewhat lower share of those who had any formal training com- pared to men leaders, but the gender gap varies noticeably by denomina- tion: It tends to be wider in denominational groups with lower overall share of leaders who had received a formal training, such as Apostolic and Zionist churches. In contrast, in denominations where a relatively large share of leaders received some training, the gap is small (Catholics) or almost nonexistent (traditional Protestants).

To test my hypotheses, I fit multivariate models. Table 3 presents the results of a logistic regression model predicting that congregation leader

Table 2: Percent of Women among Congregation leaders and Men leaders Percent of Received Some Formal Religious Training, Chibuto Congregation Census (N=1122)

Denominational Category

A. Percentage of Women

Leaders in All Congregations

B. Percentage of Women Leaders in Congregations Led

by a Higher- Ranking Office Holder (48% of

Total)

C. Percentage of Women and Men

Leaders with Formal Religious Training

Men Women

Catholic 41.9 0.0 27.8 20.5 Traditional

Protestant 39.2 38.4 22.6 21.7

Apostolic 3.7 1.9 5.9 0.0 Zionist 34.8 31.1 13.4 5.4 Pentecostal 30.4 20.2 14.3 9.7 Total average 32.3 27.5 14.9 10.8

992 GENdER & SOCIETY / december 2015

is a woman. The covariates are the denominational category of the church (traditional Protestants is the reference), urban versus rural location of the congregation, number of attendees at the last regular service, the women- to-men ratio of attendees at that service, and a set of dummies for the type of the congregation facility (a proxy, however imperfect, for congrega- tion’s material and financial health). The results are presented as regres- sion parameter estimates: A positive sign of a statistically significant coefficient means an increase in the likelihood of a woman being at the helm of a congregation associated with the corresponding predictor and relative to the reference category; a negative sign points to a decrease in that likelihood.

Mirroring the bivariate distributions in Table 2A, Apostolic congrega- tions again stand out with a much lower probability of having a woman as a congregation leader, compared to traditional Protestants and, in fact, compared to any other denominational group. The difference between

Table 3. logistic regression predicting the likelihood of the congregation leader being a woman, parameter estimates and standard errors, Chibuto Congregation Census

Covariate Coefficient SE

Denominational category Catholic 0.367 0.272 Apostolic –2.791** 0.550 Zionist –0.233 0.187 Pentecostal –0.392+ 0.224 [Traditional Protestant] Area of congregation location Urban 0.025 0.183 [Rural] Number of people attending last regular service –0.017** 0.004 Women-to-men ratio among last service attendees 0.026** 0.006 Type of congregation facility Sturdy facility (tile or metal sheet roof) –0.176 0.207 Precarious facility (no or reed/grass roof) –0.168 0.150 [No permanent facility] Constant –0.091 0.198 Likelihood ratio chi-square 123** Number of congregations 1118

NoTeS: Reference categories in brackets; Significance level: **p<.01, *p<.05, +p<.1, two-tailed test.

Agadjanian / WOMEN’S RELIGIOUS AUTHORITY 993

Pentecostals and Zionists, on the one hand, and traditional Protestants, on the other, is in the same direction, but only Pentecostals are marginally different from traditional Protestants. The odds of a woman leading a congregation are not statistically distinguishable between traditional Protestants and Catholics. In general, these results, and especially the contrast between traditional Protestant, on the one hand, and Apostolics and Pentecostals, on the other, lend support to hypothesis 1.

With regard to other predictors, women tend to head smaller congregations than do men: The size of a congregation, measured by the number of attend- ees, is negatively associated with the probability of its leader being a woman. The women-to-men ratio of attendees, on the contrary, shows a strong posi- tive association. Both results support hypotheses 2 and 3. In contrast, congre- gation material status (approximated, however imperfectly, by the quality of the roof of the congregation main facility) is not related to the gender of the congregation leader; thus, hypothesis 4 is not confirmed. Finally, no signifi- cant urban–rural differences in the likelihood of women’s leadership can be observed, controlling for other characteristics, possibly suggesting the mutu- ally cancelling tendencies proposed in hypotheses 5a and 5b.

Next, I examine whether gender is associated with formal religious training and with rank of congregation leader. Column A of Table 4 shows the results of a model predicting the likelihood of a congregation leader having received some religious training. The gender of the congregation’s leader is the main predictor. The model controls for the same congregation characteristics as the model presented in Table 3.2 The results show that the coefficient for gender, while negative and nontrivial in magnitude, is not statistically significant, when controlling for other factors. The coef- ficient barely changes when Catholic congregations are excluded from the model (not shown). Thus, although women leaders are significantly less likely to have received at least some formal religious training than their men counterparts, the gender difference is largely explained by other con- gregation characteristics. hypothesis 6 is therefore not confirmed.

The results of a multivariate logistic regression predicting the likeli- hood of a congregation leader being in a higher-ranking position are presented in Table 4, column B (Catholic congregations are excluded). The results support the hypothesis that women congregation leaders are significantly less likely to hold a higher-ranking church office, regardless of denominational category and of congregation characteristics (hypoth- esis 7). Finally, column C adds religious training to the previous model. While having received religious training significantly increases the like- lihood of holding a higher-ranking post, the addition of this covariate

994 GENdER & SOCIETY / december 2015

Table 4: logistic Regressions Predicting the likelihood of the Congregation leader Having Received Some Formal Religious Training (a) and Holding a Higher-Ranking Post in Church (b and C), Parameter estimates and Standard errors, Chibuto Congregation Census

Covariate

A. Leader Received Some Formal Religious

Training

B. Leader Is in Higher-Ranking

Posta

C. Leader Is in Higher-Ranking

Posta

Coef. SE Coef. SE Coef. SE

Gender of congregation leader Woman –0.338 0.213 –0.343* 0.146 –0.323* 0.147 [Man] Denominational category Catholic –0.200 0.327 n/a n/a n/a n/a Apostolic –2.022** 0.488 –0.322 0.267 –0.175 0.271 Zionist –0.787** 0.242 0.337+ 0.182 0.418+ 0.186 Pentecostal –0.698* 0.289 –0.224 0.215 -0.163 0.218 [Traditional Protestant] Area of congregation location Urban 0.045 0.233 0.554** 0.172 0.555** 0.173 [Rural] Number of people attend-

ing last regular service 0.022** 0.004 0.018** 0.004 0.016** 0.004

Women-to-men ratio among last service attendees

–0.005 0.009 0.007+ 0.004 0.008+ 0.004

Type of congregation facility Precarious facility (no

or reed/grass roof) 0.617** 0.224 0.178 0.143 0.135 0.144

Sturdy facility (tile or metal sheet roof)

0.619* 0.275 0.313 0.203 0.284 0.205

[No permanent facility] Leader’s formal religious training Received some training n/a n/a 0.807** 0.217 [Received no training] Constant –2.181** 0.276 –0.621** 0.205 –0.726** 0.209 Likelihood ratio chi-

square 90** 74** 88**

Number of congregations 1118 1026 1026

NoTe: Reference categories in brackets; significance level: **p < .01, *p < .05, †p < .1, two-tailed test. Coef. = coefficient. a. Catholics are excluded.

Agadjanian / WOMEN’S RELIGIOUS AUTHORITY 995

does not affect the association between gender and leader’s rank. Women’s disadvantage in access to top positions in the church hierarchy is therefore not assuaged by obtaining formal religious training.

Among other covariates, Zionist congregations are most likely to have higher-ranking leaders even though the difference from traditional Protestants is only marginally significant (but the differences between Zionists, on the one hand, and Pentecostals or Apostolics, on the other, are highly significant). The urban location of the congregation and the size of its membership have significant positive effects, whereas the positive effect of the gender composition of last service’s attendees is only margin- ally significant. Finally, the quality of the congregation facility is not associated with the rank of the congregation leader.

Women’s pathways to Formal Church Authority: Insights from Qualitative data

In this part of the analysis, I explore women’s rise to and exercise of formal authority in their congregations based on the qualitative data. These data point to two main trajectories for women’s advancement to congregational leadership: appointment or election of women as leaders based mainly on their own merit, and assumption of a leadership post by virtue of association with their husbands who are or were congregation leaders. Although the former trajectory, formal appointment or election of women as congregation leaders, is relatively infrequent, such direct ascent to formal leadership is possible, and women leaders who achieve it typi- cally enjoy considerable respect and legitimacy.

The story of Esmeralda3 offers an example of direct ascension to the pulpit. She had been the pastor of her Zionist congregation since 2001. She was elected to that post by the congregation community and then approved by the church’s higher leadership. Esmeralda was born into a Catholic family, but when she was about ten years old her parents switched to her current Zionist church in search of treatment for her health problems; she has remained in the church ever since, rising through the congregation hierarchy to become a mufundzisi (pastor or minister, in Changana). her office progression was uncommon even for her congrega- tion: All the steps of the ladder she had gone through before reaching the pastor status were occupied by men at the time of the interview. Perhaps even more unusual in her trajectory was that she was married to a Catholic. Women typically marry men of the same church or are expected to switch to their husband’s churches at marriage. What perhaps allowed

996 GENdER & SOCIETY / december 2015

Esmeralda to circumvent that norm was the fact that she was her polygy- nous husband’s junior wife (his senior wife was Catholic).

While Esmeralda’s rise through her church hierarchy was independent of her husband, he did play a crucial, if only symbolic, role: Even though her husband belonged to a different church, Esmeralda had to secure his approval for her assumption of the subsequent leadership posts and to convey that approval to her church superiors. Esmeralda now enjoys wide support in her congregation, among men and women alike. In fact, she insisted in the interview that her promotion to the rank of pastor happened with strong support from the congregation’s men. She also affirmed that men leaders of other congregations of her church as well as men in her own congregation treat her with great respect.

As Table 2 suggests, women’s presence in leadership posts in their congregations is quite common in traditional Protestant churches. Unlike many Zionist churches, where formal church hierarchy and procedural regulations are not well developed and congregation leaders are elected at the meetings of congregation members, in traditional Protestant churches top congregation leadership instatement is typically a more formalized procedure. Yet, like the relative informality of Zionists, the formal but women-friendlier leadership promotion regime of many traditional Protestant churches facilitates women’s ascension to leadership posts. For example, Regina became the pastor in her congregation of the Church of Nazarene after the previous pastor, a man, had retired. She was appointed by the church district council with the official ceremony of investiture carried out by the church’s district male superintendent. her husband is also a pastor of the same church but in a different congregation: Regina affirmed that her appointment was independent of her husband’s. She claimed that her gender was never an issue because in her church it is normal for women to occupy leadership positions.

Another, more common, path to what can be defined as quasi-formal leadership for women is by association with their husbands who have formal leadership posts. Spouses of church leaders often wield consider- able influence (e.g., Detwiler-Breidenbach 2000), and this study site is no exception to this pattern. Wives of congregation leaders are typically referred to by the titles of their husbands: Thus, the wife of a pastor would also be called pastor or, more specifically “women’s pastor” (mufundzisi wa vamamana) or “mother pastor” (mamana mufundzisi). Most com- monly, the role of these women is subordinate to that of their husbands: Although interviewed women leaders-by-association were always quick to stress full harmony in their own and their husband’s decisions and actions, they also typically admitted that “man is the head” (bava i

Agadjanian / WOMEN’S RELIGIOUS AUTHORITY 997

nhloko). These women leaders’ duties are often limited to dealing with specifically women’s issues and guiding congregation women’s meetings.

however, some of these women become fairly influential in their con- gregation’s activities and even de facto head their congregations when their husbands, the primary office holders, are away working or traveling. The case of Ofelia, a “mother pastor” in an urban Zionist congregation, provides an illustration. The congregation elected her husband and her to be the pastor and the women’s pastor, respectively, two years after the previous pastor of their congregation (to which Ofelia had belonged since birth) had passed away. Before that, both were deacons. however, these days Ofelia’s husband, like very many local men, spends much of his time working in South Africa; so, in his absence, Ofelia functions as the sole pastor: She leads the Sunday services and otherwise directs the life of the congregation.

Such joint, even if highly unequal, office-holding is a nearly universal practice if the man leader does not have another wife. Although many churches discourage polygyny among their leaders (usually on the grounds that polygynous men would not have enough energy left after fulfilling their marital duties to attend properly to the congregation’s spir- itual and organizational needs), in reality this discouragement is rarely heeded. In fact, quite the contrary, a church leadership office typically correlates with higher social status and greater economic resources, which, in turn, facilitates marrying additional wives. These new wives are usually integrated in the congregation hierarchy, but their place in that hierarchy commonly reflects the marital order: As junior wives are con- sidered subordinate to the first wife, so they are typically accorded lower- level congregation posts than that of the first wife. The case of Bernardo, a Zionist pastor in a small village who had three wives, provides an illus- tration: his first (senior) wife was a pastor, his second wife a deacon, and his third wife an evangelist in the same congregation.

Because of higher men’s mortality, church leaders’ wives often outlive their husbands. In case of a male leader’s death, it is very common for his widow to remain at the helm of the congregation as the sole leader. Typically, women who inherit congregation leadership from their deceased husbands enjoy full legitimacy, often further reinforced by cha- risma, either their own or “transferred” from their husbands. Such con- gregations may remain under the women’s leadership for years, in fact, often until their own deaths. The story of Arlinda, a Zionist pastor, offers an example of this scenario. She switched to her current Zionist church from another Zionist church when she married a man of the current

998 GENdER & SOCIETY / december 2015

church. At that time, her husband was a deacon, and she became a deacon by association. When he was promoted to the rank of pastor, she became a pastor too. She remained in that rank after her husband’s passing. Or, take the case of Lúcia. her late husband was also a pastor. Lúcia had never gone to school, and her husband taught her how to read and write. When he passed away, some 15 years before the interview, Lúcia became the congregation’s sole pastor and has remained in that capacity ever since. She said that her authority is never questioned, both because of her post and because of her respectable age.

The same office inheritance mechanism functions at the lower levels of the church hierarchy as well. however, an inherited office is often a dead-end one: Usually no independent subsequent ascension happens after the husband’s death. Consider the story of Gilda, a mukokheli (con- gregation warden) in her Zionist congregation, to which she belonged since birth. When her husband was alive, she accompanied his rise through the lower rungs of the church hierarchy. her husband was a mukokheli when he died, and she retained that title after his death. There are two other wardens in the congregation—a married couple—who have held that joint office since before the death of Gilda’s husband. however, she said she could not climb any further on the congregational ladder because her husband is no longer alive. When confronted with the exam- ple of Esmeralda, the pastor who made it to the top of the congregation leadership on her own, Gilda said that Esmeralda’s advancement became possible because her husband, even though not part of her church, was alive to approve it. Yet, widowhood does not always stop women’s rise in the church hierarchy: For example, Julieta, whose husband had died three years before she was interviewed, not only remained the sole leader of her congregation but was also promoted to become her church’s president. Importantly, most women who achieve high leadership posi- tions after their husband’s deaths are older, and their advanced age helps to counterbalance the traditional patriarchal gender ideology in enhanc- ing their leadership credentials.

Despite the legitimacy that comes with a formal leadership status and age, the focus group participants and individual informants acknowledged diffi- culties in dealing with men congregation members, especially those whose bad behavior (e.g., drunkenness, wife abuse, or adultery) call for reprimand. In such cases, women leaders typically enlist the help of their husbands or other influential congregation men to try to bring the debauchers to reason. Dealing with deviant men in church is, however, rather uncommon—not least because men’s abusive behavior within their households is infrequently brought out to the congregation leader’s attention. But even when that

Agadjanian / WOMEN’S RELIGIOUS AUTHORITY 999

happens, rarely do any actions follow. Congregation leaders’ reluctance to rebuke men is not a simple expression of the reigning gender ideology that tends to absolve men of any wrongdoing or at least to equate husband’s and wife’s responsibility for almost any marital or family problem. Because the economic and, especially, monetary resources remain disproportionately in the hands of men, women leaders, not unlike their men counterparts, have to be lenient with congregation men, especially if those are migrants earning substantial incomes in South Africa, because those men might contribute financially to the church. Faith and prayer do not fix a leaking roof, and a new roof does more to enhance the image of a church—and to make it more attractive for potential converts—than the intangible loyalty of its numerous women members. Congregation membership fees, typically an equivalent of some $.30 per adult member per month, are not sufficient to cover a congre- gation’s most basic needs, even when they are indeed paid.

Ironically, therefore, the ideological devotion and organizational com- mitment of the church’s massive, loyal, yet penniless women’s congrega- tion membership are of lesser value to the congregation’s organizational health than potential monetary contributions of the relatively well-off but otherwise uninvolved men. And to muster financial favors of those men, women and men leaders alike must engage in the same gendered dis- course glorifying men’s preeminence and domination. Congregation financial pressures also imply that more affluent individuals are more likely to be promoted through the church leadership ranks: The vast majority of congregation leaders do not receive any salary or allowance from their churches (with the exceptions of high hierarchs in the Catholic and some traditional Protestant churches) and therefore are expected to cover their expenses and at least part of congregational needs through income from other sources. Again, this situation strongly privileges men, who are more likely to have income-generating employment: In the Chibuto congregation census, 28 percent of Men leaders reported gainful employment other than subsistence agriculture, compared to only 9 per- cent of women leaders. Although the congregation census did not include questions on leaders’ personal incomes, the nature of most women’s out- side-the-home nonagricultural work, typically involving petty trade with scanty and episodic earnings, further increases women’s disadvantage.

CONCLUSION

Despite the importance of religion to the social (re)production of gen- der, gender themes have been relatively marginal within religious studies

1000 GENdER & SOCIETY / december 2015

(Avishai, Jafar, and Rinaldo 2015). This marginalization is particularly pronounced in the literature dealing with developing countries. In the foregoing analysis, I used Western scholarship as a point of departure to examine women’s involvement in formal religious authority in a typical rural sub-Saharan context. As this analysis suggests, similarly to Western settings, sub-Saharan women’s rise to religious leadership roles, as well as the limitations of this rise, reflects the broader societal realities and expectations of (in)equality between women and men (Wessinger 1996, 6). As elsewhere, this process is shaped through “loose coupling” between these broader forces and church internal ideological and organizational priorities (Chaves 1996). And, as in the Western world (Zikmund 1986), this process unfolds gradually in sub-Saharan Africa. In this study, women still tended to lead smaller congregations than did men, and women con- gregation leaders were significantly less likely than men leaders to occupy higher-ranking church offices, even after controlling for theological train- ing. Yet, with almost a third of the district’s religious congregations being headed by women, the process of women’s entry into leadership roles in Chibuto has been more advanced than many outsiders might imagine. Women’s ascent through the church leadership ranks increasingly defines the face, identity, and expression of religious organizations in sub-Saharan Africa. however, women leaders must confront and often conform to the gendered world around them. Rarely does their rise to positions of author- ity explicitly clash with deeply entrenched and omnipresent gender inequalities, and with the pervasive stereotypes about women’s subalter- nity—the stereotypes to which, incidentally, most women church leaders still readily subscribe. hence, the expansion of women’s religious leader- ship roles both challenges and reasserts the patriarchal gender ideology and the social hierarchies and relationships that this ideology cements. As observed in Chibuto, formal church leadership does empower women, yet, at the same time, the power that women leaders gain remains mainly limited to “women’s matters” and is exercised largely within the gendered constraints of religious ideologies. This dialectical reality parallels, with appropriate caveats, the findings of a growing number of studies that show how women’s “doing” of religion (Avishai 2008) promotes their empowerment by creatively engaging their faith without fundamentally challenging the ideological and symbolic boundaries of the religious doc- trine (e.g., Chong 2008; González 2013; Mahmood 2005; Prickett 2015; Rinaldo 2013; van Doorn-harder 2006).

Whereas these gendered processes are present in all churches, their scale and specific manifestations vary across denominations. Western lit- erature has long identified the denominational diversity in women’s

Agadjanian / WOMEN’S RELIGIOUS AUTHORITY 1001

access to congregation authority and in the nature of women’s leadership functions. My analysis illustrates this diversity in a developing setting. I could not fully examine the denominational nuances of gendered leader- ship dynamics. My denominational typology is necessarily schematic, and even within each denominational group, as defined in this study, there exists a wide range of women’s leadership patterns and experiences shaped by church theology, organizational structure, demographic compo- sition, and financial standing. Nonetheless, the denominational compari- sons produced informative insights. In general, they showed that traditional Protestant churches generally offer the most favorable environment for women’s formal leadership advancement, paralleling the historical experi- ence of many Western settings. Interestingly, Zionists, who are doctrinally and socially much more conservative than traditional Protestants, do not differ significantly from them in the probability of a congregation leader being a woman. Drawing historical parallels with the Western experience described earlier, I argue that while many traditional Protestant churches facilitate women’s ascension to leadership through inclusive formal ordi- nation mechanisms, among Zionists (as in Western non-mainstream reli- gious movements in years past) this ascension has become possible precisely because such mechanisms are not well formalized.

The Catholic congregations display a remarkable contrast between an overall high share of women congregation leaders and the absence of women among top-ranking leaders. This contrast is reminiscent of the situation in Western Catholic communities, where the share of women lay ministers has been growing despite the continuing ban on women’s priest- hood (Flinn 1996; Wallace 1992) and, in broader terms, reflects the fre- quently noted disjunction between the relatively liberal grassroots and the conservative officialdom of the Roman Catholic Church (Reese 1996). On the opposite end of the denominational spectrum are Apostolics, whose highly patriarchal, rigid organizational structure all but excludes women from formal leadership roles, regardless of the rank.

however, in terms of wider societal impact, the gender leadership dynamics in Zionist and (neo-)Pentecostal churches seem most important. The inexorable advance of Pentecostalism in the subcontinent in general and in Mozambique in particular reflects broader globalization processes and is poised to reshape the local religious landscape, as did the rise of Zionist and other African Independent Churches a few decades earlier (Meyer 2004). It will also likely have vast repercussions for women’s roles in religious organizations. In Mozambique, the Pentecostal wave is currently represented by such relatively well-established international

1002 GENdER & SOCIETY / december 2015

conglomerates as the Assemblies of God as well as comparatively recent neo-Pentecostal arrivals like the Universal Church of the Kingdom of God. These newer denominations started their triumphant march in cities by appealing to the urban middle and working class, but they have lately been making inroads into rural areas as well, tapping into the small but growing better-off peasant and trader segments. In this religious marketplace, the Pentecostal expansion targets primarily Zionists, not only by absorbing individual switchers from Zionist churches but sometimes also by engulfing entire Zionist churches.4 On some counts, Pentecostals are definitely more “modern” than Zionists—in their refutation of traditional religious prac- tices, such as remnants of witch-doctorship in Zionist healing rituals, and their more holistic, Western-style emphasis on broader well-being beyond illness cure. Yet, the rise of Pentecostalism and the concomitant decline of Zionism may also undermine the gains made by women in Zionist churches over the past decades. First, Pentecostals typically have a more rigorous leadership structure, in which congregation leaders are usually appointed by church top hierarchs rather than being elected by congregants, as it is often done in semi-autonomous, organizationally amorphous Zionist congrega- tions. The more formalized organizational structure of Pentecostal churches also tends to preclude women from ascending to leadership positions through their marital relation: A pastor’s wife remains such, and her infor- mal influence is not conflated with her husband’s formal authority. The noticeably higher percentage of women among higher-ranking church office holders in Zionist congregations compared to Pentecostal congrega- tions, shown in Table 2, alludes to these differences.5

Yet, future trends in women’s religious leadership are also contingent on ideological and organizational changes within Pentecostal churches. Women’s involvement is central to the global Pentecostal expansion (Anderson 2013), and the rise of Pentecostal and Charismatic churches elsewhere has been associated with women’s empowerment (Brusco 1995; Machado 1996). In sub-Saharan Africa, it has been argued that neo-Pente- costals tend to encourage more gender-egalitarian romantic and marital relationships (van de Kamp 2013). And in the United States, many Pentecostal congregations are led by women even though Pentecostal doc- trines typically discourage women’s leadership (Lawless 1993). It is pos- sible that the new Pentecostal churches, as they gain territory and influence in sub-Saharan settings like the one examined here, will gradually become more open to women’s leadership. In fact, guided by Western experience, one might think of women’s entry into leadership positions and greater acceptance of that leadership as two mutually reinforcing processes:

Agadjanian / WOMEN’S RELIGIOUS AUTHORITY 1003

Women’s success as church leaders and their more egalitarian leadership styles (e.g., Lawless 1993; Lehman 1993) should help enhance the legiti- macy of women’s leadership, which in turn would help bring more women into leadership positions.

Finally, I should stress again the importance of seeing the dynamics of women’s authority through the prism of local religious economy, which itself functions within a broader socioeconomic context. Women’s leader- ship is constrained by church organizational and financial needs and by ferocious, even if usually hidden, competition among different churches in the religious marketplace (Agadjanian and Menjívar 2011). While wom- en’s increased authority within their congregations largely mirrors the predominance of women among active congregation members, it is also, at least to some extent, indicative of the declining appeal of congregation leadership roles, especially of those at the lower rungs of the leadership ladder, to men, who are reoriented toward more socially and economically rewarding pursuits—either at the very top of the church hierarchy (which I could not examine with the district-level data) or completely outside the religious realm. These dynamics resemble those in the U.S. and other Western settings, where the declining prestige of clergy and resulting retreat of men from church leadership positions have opened up opportuni- ties for women’s advancement (e.g., Nesbitt 1997; Wallace 1991, 1992). My study demonstrates that, as in the West, the frontier of “hidden resist- ance” (Marder 1996) to women’s leadership in the sub-Saharan has moved up to the more prestigious—and more lucrative—higher-ranking positions. Yet, while these dynamics reflect the universal gendered patterns of alloca- tion, redistribution, and assertion of power, women’s church authority has an enormous potential for transforming—and reinvigorating—religious organizations (e.g., Ecklund 2006; Niemelä 2011; Watling 2002) and, by extension, the multiple roles and influences of these organizations in soci- ety at large (e.g., Ingersoll 2003; Olson, Crawford, and Deckman 2005). Future research must examine how women’s ascent to formal religious authority alters gender ideology and reconfigures gender hierarchies beyond the religious realm in developing settings.

NOTES

1. Unfortunately, the congregation census data do not allow for classifying leader’s training by level, duration, frequency, or intensity.

2. The models do not control for leader’s age because a large share of congre- gation census respondents did not know their age (or the age of the congregation

1004 GENdER & SOCIETY / december 2015

leader, in case of proxy respondents). Educational attainment is available only for the census respondents (and therefore not for all congregation leaders). I acknowledge these limitations.

3. All names used in this study are pseudonyms. 4. For example, we interviewed a Zionist (man) pastor who decided to move

his entire church to a new, foreign-based Pentecostal church that had recently entered Chibuto’s religious scene—because that new church preached “the right faith” and because its leaders promised to pay for a new building for the entre- preneurial pastor’s congregation (and hinted at the seriousness of their intentions by presenting him with a brand-new suit).

5. The relatively small number of surveyed congregations does not allow for an analytic separation of congregations belonging to more established Pentecostal churches from those of newer churches that can be classified as neo-Pentecostal.

REFERENCES

Agadjanian, Victor. 1999. Zionist churches in the urban sociocultural space of Mozambique in the 1980s and 1990s. In Lusotopie: Enjeux contemporains dans les Espaces Lusophones. Paris: Karthala.

Agadjanian, Victor. 2012. Joining, switching, and quitting: Denominational dif- ferences and socioeconomic correlates in a sub-Saharan setting. Paper pre- sented at the Society for the Scientific Study of Religion and the Religious Research Association Annual Meeting, Phoenix, AZ.

Agadjanian, Victor, and Cecilia Menjívar. 2011. Fighting down the scourge, building up the church: Organizational constraints in religious involvement with hIV/AIDS in Mozambique. global Public Health Special Issue on Reli- gious Responses to hIV and AIDS: S148-62.

Agadjanian, Victor, and Scott T. Yabiku. Forthcoming. Religious belonging, reli- gious agency and women’s autonomy in sub-Saharan Africa. journal for the Scientific Study of religion.

Anderson, Allan h. 2002. Diversity in the definition of “Pentecostal/Charismatic” and its ecumenical implications. Mission Studies 19:40-55.

Anderson, Allan h. 2013. to the ends of the Earth: Pentecostalism and the trans- formation of world christianity. Oxford, UK: Oxford University Press.

Avishai, Orit. 2008. “Doing religion” in a secular world: Women in conservative religions and the questions of agency. gender & Society 22:409-33.

Avishai, Orit, Afshan Jafar, and Rachel Rinaldo. 2015. A gender lens on religion. gender & Society 29:5-25.

Baer, hans A. 1993. The limited empowerment of women in Black spiritual churches: An alternative vehicle to religious leadership. Sociology of religion 54:65-82.

Bano, Masooda, and hilary Kalmbach. 2012. Women, leadership, and mosques: changes in contemporary islamic authority. Leiden, the Netherlands: Brill.

Agadjanian / WOMEN’S RELIGIOUS AUTHORITY 1005

Bartkowski, John P., and Jen’an Ghazal Read. 2003. Veiled submission: Gen- der, power, and identity among Evangelical and Muslim women in the United States. Qualitative Sociology 26:71-92.

Bednarowski, Mary F. 1980. Outside the mainstream: Women’s religion and women religious leaders in nineteenth-century America. journal of the Ameri- can Academy of religion 48:207-31.

Brusco, Elizabeth E. 1995. the reformation of machismo: Evangelical conversion and gender in colombia. Austin: University of Texas Press.

Chang, Patricia M. Y. 1997. Female clergy in the contemporary Protestant Church: A current assessment. journal for the Scientific Study of religion 36:565-73.

Charlton, Joy. 1997. Clergywomen of the pioneer generation: A longitudinal study. journal for the Scientific Study of religion 36:599-613.

Chaves, Mark. 1996. Ordaining women: The diffusion of an organizational inno- vation. American journal of Sociology 101:840-73.

Chaves, Mark. 1997. ordaining women: culture and conflict in religious organi- zations. Cambridge, MA: harvard University Press.

Chong, Kelly h. 2008. deliverance and submission: Evangelical women and the negotiation of patriarchy in South Korea. Cambridge, MA: harvard Univer- sity Asia Center.

Coleman, Isobel. 2010. Paradise beneath her feet: How women are transforming the Middle East. New York: Random house.

Cruz e Silva, Teresa. 2001. Protestant churches and the formation of political consciousness in southern Mozambique (1930-1974). Basel, Switzerland: P. Schlettwein.

De Vletter, F. 2007. Migration and development in Mozambique: Poverty, inequal- ity and survival. development Southern Africa 24:137-53.

Detwiler-Breidenbach, Ann M. 2000. Language, gender, and context in an immigrant ministry: New spaces for the pastor’s wife. Sociology of religion 61:455-59.

Dodson, Jualynn E. 2002. Engendering church: Women, power, and the AME church. Lanham, MD: Rowman & Littlefield.

Ecklund, Elain h. 2006. Organizational culture and women’s leadership: A study of six Catholic parishes. Sociology of religion 67:81-98.

Flinn, Virginia Sullivan. 1996. Ministerial attitudes and aspirations of Catholic laywomen in the United States. In religious institutions and women’s leader- ship: new roles inside the mainstream, edited by Catherine Wessinger. Colum- bia: University of South Carolina Press.

Garrard, David J. 2009. African Pentecostalism. journal of Beliefs and Values 30:231-44.

González, Alessandra L. 2013. islamic feminism in Kuwait: the politics and par- adoxes. New York: Palgrave Macmillan.

Ingersoll, Julie. 2003. Evangelical christian women: War stories in the gender battles. New York: New York University Press.

1006 GENdER & SOCIETY / december 2015

Instituto Nacional de Estatística. 1999. resultados definitivos do ii recensea- mento geral da População e Habitação. Maputo, Mozambique: Instituto Nacional de Estatística.

Instituto Nacional de Estatística. 2009. resultados definitivos do iii recensea- mento geral da População e Habitação. Maputo, Mozambique: Instituto Nacional de Estatística.

Kalu, Ogbu. 2003. Pentecostal and Charismatic reshaping of the African religious landscape in the 1990s. Mission Studies 20:84-109.

Khurshid, Ayesha. 2015. Islamic traditions of modernity: Gender, class, and Islam in a transnational women’s educational project. gender & Society 29:98-121.

Larson, Rebecca. 1999. daughters of light: Quaker women preaching and proph- esying in the colonies and abroad, 1700-1775. Chapel hill: University of North Carolina Press.

Lawless, Elaine J. 1993. Not so different a story after all: Pentecostal women in the pulpit. In Women’s leadership in marginal religions: Explorations outside the mainstream, edited by Catherine Wessinger. Urbana: University of Illinois Press.

Lehman, Edward C., Jr. 1980. Patterns of lay resistance to women in ministry. Sociological Analysis 41:317-38.

Lehman, Edward C., Jr. 1985. Women clergy: Breaking through gender barriers. New Brunswick, NJ: Transactions Books.

Lehman, Edward C., Jr. 1993. gender and work: the case of the clergy. Albany: State University of New York Press.

Loforte, Ana M. 2000. género e poder entre os tsonga de Moçambique. Maputo, Mozambique: Promédia.

Machado, Maria das D. C. 1996. carismáticos e Pentecostais: Adesão religiosa na esfera familiar. Campinas, Brazil: Autores Associados/ANPOCS.

Mahmood, Saba. 2005. Politics of piety: the islamic revival and the feminist sub- ject. Princeton, NJ: Princeton University Press.

Marder, Janet R. 1996. Are women changing the rabbinate? A reform perspec- tive. In religious institutions and women’s leadership: new roles inside the mainstream, edited by Catherine Wessinger. Columbia: University of South Carolina Press.

Meyer, Birgit. 2004. Christianity in Africa: From African Independent to Pente- costal-Charismatic churches. Annual review of Anthropology 33:447-74.

Morier-Genoud, Eric, and Pierre Anouilh. 2013. The Catholic Church in Mozam- bique under revolution, war, and democracy. In religion and politics in a global society: comparative perspectives from the Portuguese-speaking world, edited by Paul C. Manuel, Alynna Lyon, and Clyde Wilcox. Lanham, MD: Lexington Books.

Nadell, Pamela S. 1998. Women who would be rabbis: A history of women’s ordi- nation 1889-1985. Boston: Beacon Press.

Nesbitt, Paula D. 1993. Dual ordination tracks. Sociology of religion 54:13-30. Nesbitt, Paula D. 1997. Feminization of the clergy and America: occupational

and organizational perspectives. Oxford, UK: Oxford University Press.

Agadjanian / WOMEN’S RELIGIOUS AUTHORITY 1007

Niemelä, Kati. 2011. Female clergy as agents of religious change? religions 2:358-71.

Olson, Laura A., Sue E. S. Crawford, and Melissa M. Deckman. 2005. Women with a mission: religion, gender, and the politics of women clergy. Tusca- loosa: University of Alabama Press.

Pfeiffer, James. 2002. African Independent churches in Mozambique: healing the afflictions of inequality. Medical Anthropology Quarterly 16:176-99.

Plant, helen. 2003. “Subjective testimonies”: Women Quaker ministers and spiri- tual authority in England, 1750-1825. gender & History 15:296-318.

Prickett, Pamela J. 2015. Negotiating gendered religious space: The particulari- ties of patriarchy in an African American mosque. gender & Society 29:51-72.

Reese, Thomas J. 1996. inside the Vatican: the politics and organization of the catholic church. Cambridge, MA: harvard University Press.

Rinaldo, Rachel. 2013. Mobilizing piety: islam and feminism in indonesia. New York: Oxford University Press.

Ruether, Rosemary, and Eleanor McLaughlin, eds. 1979. Women of spirit: Female leadership in the jewish and christian traditions. New York: Simon & Schuster.

Schmidt, Frederick W. 1996. A still small voice: Women, ordination, and the church. Syracuse, NY: Syracuse University Press.

Sullins, Paul. 2000. The stained glass ceiling: Career attainment for women clergy. Sociology of religion 61:243-66.

Tucker, Cynthia Grant. 1996. Women in the Unitarian-Universalist ministry. In religious institutions and women’s leadership: new roles inside the main- stream, edited by Catherine Wessinger. Columbia: University of South Caro- lina Press.

van de Kamp, Linda. 2013. Public counselling: Brazilian Pentecostal intimate performances among urban women in Mozambique. culture, Health & Sexu- ality 15 (Suppl 4): S523-36.

van de Kamp, Linda, and Rijk van Dijk. 2010. Pentecostals moving South-South: Ghanaian and Brazilian transnationalism in Southern Africa. In religion crossing boundaries: transnational dynamics in Africa and the new African diasporic religions, edited by A. Adogame and J. Spickard. Leiden, the Neth- erlands: Brill.

van Doorn-harder, Pieternella. 2006. Women shaping islam: reading the Qur’an in indonesia. Chicago: University of Illinois Press.

Wallace, Ruth A. 1991. Women administrators of priestless parishes: Constraints and opportunities. review of religious research 32:289-304.

Wallace, Ruth A. 1992. they call her pastor: A new role for catholic women. Albany: State University of New York Press.

Watling, Tony. 2002. “Leadership” or “dialogue”? Women, authority, and reli- gious change in a Netherlands community. Sociology of religion 63:515-38.

1008 GENdER & SOCIETY / december 2015

Wessinger, Catherine, ed. 1993. Women’s leadership in marginal religions: Explo- rations outside the mainstream. Urbana: University of Illinois Press.

Wessinger, Catherine, ed. 1996. religious institutions and women’s leadership: new roles inside the mainstream. Columbia: University of South Carolina Press.

Zikmund, Barbara B. 1986. Winning ordination for women in mainstream Protes- tant churches. In Women and religion in America: A documentary history, vol. 3 1900-1968, edited by Rosemary R. Ruether and Rosemary S. Keller. New York: harper and Row.

Zikmund, Barbara B., Adair T. Lummis, and Patricia M. Y. Chang. 1998. clergy women: An uphill calling. Louisville, KY: Westminster John Knox Press.

Victor Agadjanian is a Foundation distinguished Professor of Sociology at the University of Kansas. His research has focused on gender, religion, ethnicity, migration, and sexual and reproductive behavior in sub-Saharan Africa and central Eurasia. in his work, Agadjanian often combines statis- tical analyses of survey data with ethnographic explorations.