Introduction
Across the world, every year nearly 15 million girls get married while they are still children (UNICEF, 2014). Child marriage (marriage before the age of 18) is a form of violence against children and is internationally recognized as a violation of their human rights. It clearly has a gender dimension since girls are disproportionately affected by it (Yasukawa & Gough, Reference Yasukawa and Gough2016). Girls married as children face increased risk of severe harmful physical and mental health consequences, including unwanted pregnancy, maternal and child morbidity and mortality, poor reproductive health conditions, depression and domestic violence, among other things (Bruce, Reference Bruce2003; Gökçe et al., Reference Gökçe, Özşahin and Zencir2007; Raj et al., Reference Raj, Saggurti, Balaiah and Silverman2009; Raj, Reference Raj2010; Warria, 2017). About half of the girls affected by child marriage globally are from the South Asian countries of India, Bangladesh, Nepal and Afghanistan, with the highest prevalence being in Bangladesh (Malhotra et al., Reference Malhotra, Warner, McGonagle and Lee-Rife2011).
Despite child marriage being prohibited by national law and a punishable offence, 59% of girls in Bangladesh marry before they reach the age of 18 (NIPORT et al., 2016). Although this has significantly reduced from 73% in 1989 (NIPORT et al., 2013), the change mainly involves a decrease in the proportion of girls getting married before 14, while for girls getting married at 15–17 years of age the change has been insignificant (Verma et al., Reference Verma, Sinha and Khanna2013). Evidence links the high prevalence of child marriage in Bangladesh and in other South Asian countries to a number of structural and normative drivers, including poverty, dowry, lack of awareness and access to education and traditional gender norms (Chowdhury, Reference Chowdhury2004; Parsons et al., Reference Parsons, Edmeades, Kes, Petroni, Sexton and Wodon2015; Hossain et al., Reference Hossain, Mahumud and Saw2016; Amin et al., Reference Amin, Asadullah, Hossain and Wahhaj2017a; Wahhaj, Reference Wahhaj2018; Asadullah & Wahhaj, Reference Asadullah and Wahhaj2019). Yet, especially in Bangladesh, despite steady economic growth, mass female education and concerted efforts to eliminate child marriage, its prevalence continues to remain very high.
Marriage practices in a society are strongly influenced by the gender norms prevailing in that society (Desai & Andrist, Reference Desai and Andrist2010). As such, the timing of women’s first marriage reveals valuable information regarding their status and expected role in the public and private life of that particular society (Chowdhury & Trovato, Reference Chowdhury and Trovato1994). In Bangladesh, as in some other South Asian countries, where nearly 90% of all marriages are family-arranged and marriage choices are largely governed by parental preferences and obligations (Peterson et al., Reference Peterson, Kim, McCarthy, Park and Plamondon2011; Rubio, Reference Rubio2014), the high prevalence of girl child marriage is a manifestation of a deeply ingrained structure of inequality that seeks to perpetuate women’s subservient status in marriage.
The prevalence and causes of child marriage in South Asian countries are well documented in existing research. Poverty is often cited as one of the major drivers of child marriage in the region, similar to other lower-income and less-industrialized regions of the world, where girls from poorer households are more likely to get married at an earlier age (Jain & Kurz, Reference Jain and Kurz2007; UNICEF, 2009; Barrientos et al., Reference Barrientos, Byrne, Peña and Villa2014; Solotaroff & Pande, Reference Solotaroff and Pande2014). Lack of access to education and absence of economic opportunities for women also act as significant drivers of child marriage (Islam et al., Reference Islam, Haque and Hossain2016). Except for Sri Lanka, marriage is universal in South Asian countries and viewed as central to social life, especially for women – an obligation rather than a choice – and the high value placed on marriage dictates that every girl has to be married (Verma et al., Reference Verma, Sinha and Khanna2013; De Silva, Reference De Silva2014). In these countries, as soon as a girl reaches puberty it becomes the family’s primary concern to protect her sexuality (virginity) until she is married, and thus also protect the family’s reputation, by preventing her from engaging in premarital sex either by choice or through sexual assault (Mathur et al., Reference Mathur, Greene and Malhotra2003; Yarrow et al., Reference Yarrow, Apland, Anderson and Hamilton2015; Pandey, Reference Pandey2017). Concern over safety and security of young and adolescent girls, especially in poorer households, often compels parents to get them married at an early age (Chowdhury, Reference Chowdhury2004; Raj, Reference Raj2010). Another strong driver of child marriage in South Asia is dowry, which involves the bride’s family paying money and/or assets to the bridegroom’s family, usually at the time of the wedding. Parents have to pay a smaller amount of dowry for younger girls and the amount increases with the age of the girl (Amin & Huq, Reference Amin and Huq2008; Loaiza & Wong, Reference Loaiza and Wong2012).
During the past few decades socioeconomic development in Asian countries in general has resulted in a continual increase in the age of first marriage for both girls and boys (Caldwell, Reference Caldwell2005; Jones & Yeung, Reference Jones and Yeung2014). Yet unlike other Asian countries, steady economic growth and improvement in the Human Development Index in Bangladesh have not been accompanied by an equivalent increase in the age of first marriage for girls. Moreover, since the 1990s government agencies, international development organizations and non-government organizations (NGOs) have carried out numerous interventions, including school stipends and life skills projects for adolescent girls, conditional cash transfer programmes, community engagement, media campaigns and advocacy initiatives in order to prevent child marriage in Bangladesh but with only limited success (Solotaroff & Pande, Reference Solotaroff and Pande2014; Asadullah & Wahnaj, Reference Asadullah and Wahhaj2016).
Furthermore, while across the developing world free or low-cost mass education programmes have resulted in the immediate effect in raising the age of marriage for girls and delaying the initiation of childbearing, it has not been the case for Bangladesh despite a significant increase in girls’ education in the country (Raj et al., Reference Raj, Gomez and Silverman2014a; Streatfield et al., Reference Streatfield, Kamal, Ahsan and Nahar2015). On the other hand, research shows a high level of awareness and unsupportive attitude towards child marriage among rural populations in Bangladesh, even though its prevalence remains very high (Caldwell, Reference Caldwell2005; Yarrow et al., Reference Yarrow, Apland, Anderson and Hamilton2015). While existing research provides valuable insight into the problem, it does not fully explain the stubborn persistence of child marriage in Bangladesh notwithstanding the concerted efforts to address its causes. Neither does it explain why child marriage is so high in the country despite the high level of awareness on its negative implications and unsupportive attitude towards it. The present study sought to address this gap by drawing on empirical data from a study carried out by BRAC, a non-government organization (NGO) based in Bangladesh, as part of an intervention to develop an effective mechanism to prevent child marriage. The study aimed to explore the community-level perceptions and practices of child marriage in a rural setting in Bangladesh and re-examine how those might be influenced by factors such as poverty, dowry and societal expectations about an ideal match. Moreover, it aimed to interrogate the established drivers of child marriage and document the various discourses that work to perpetuate it.
Methods
Study design
The study used a mixed-method approach incorporating both qualitative and quantitative tools for data collection and analysis. Drawing on Creswell and Clark (Reference Creswell and Clark2017), an exploratory sequential mixed-method design (QUAL→quan) was adopted with an initial phase of qualitative data collection and analysis, followed by quantitative data collection and analysis in the second phase. The qualitative and quantitative components were sequential-dependent (Teddlie & Tashakkori, Reference Teddlie and Tashakkori2009), where development of survey questions relied upon outcomes of the analysis of qualitative data. The qualitative component mainly focused on exploring community perceptions and attitudes towards child marriage and provided in-depth contextual understanding. Informed by the qualitative findings, the quantitative component examined the marriage practices in the community and also investigated participants’ knowledge and attitude regarding child marriage. This enabled triangulation, clarification of the findings and interrogation of the findings of one component with findings from the other component. This allowed a broader and more nuanced understanding of the depth and breadth of the issue of child marriage in rural Bangladesh.
In the first phase of the study, facilitated group discussions and semi-structured interviews were carried out with a range of purposively selected participants who were identified as key stakeholders in child marriage prevention. The definition of ‘child’ was adopted from the United Nations Convention on the Rights of Children (UNCRC) and accordingly any marriage that involved a bride and/or bridegroom aged below 18 years was considered a ‘child marriage’. The discussions were centred around questions that sought to explore community perceptions on child marriage, such as attitudes towards it, awareness of its prevalence in the locality, understanding of its various negative implications, particularly for adolescent mothers and their children, and awareness of the causes of child marriage. Findings from group discussions and interviews were used to develop a structured survey questionnaire to collect data on individuals’ perceptions of, and attitudes towards, child marriage, as well as on actual marriage practices from a larger and randomly selected sample during the second phase of the study.
The survey questionnaire was divided in four sections. The first section was used to collect socio-demographic data about the participants. The second section explored their marriage practices including, among others, age at marriage, amount of dowry exchanged, and in the case of child marriage, the underlying motive. The third section looked at the socioeconomic condition of girls who were reported to have been married at any time during the previous 3 years while still a child. A 3-year reference point was chosen as several reports, including the 2011 Demographic and Health Survey report of Bangladesh (NIPORT et al., 2013), indicated a high prevalence of child marriage in the study areas prior to this time frame. The fourth section included a set of statements concerning perceptions of, and attitudes towards, child marriage, drawn from the findings of qualitative data. Responses to these questions were gathered using a three-point Likert scale, which asked the participants whether they agreed, partially agreed or disagreed with the given statements. The questionnaire also included a brief set of questions aimed at assessing participants’ knowledge regarding child marriage prevention law in Bangladesh.
Study site
The study was carried out in the rural areas of the Badarganj sub-district of Rangpur district located in the northern region of Bangladesh. Census data from Bangladesh Bureau of Statistics (BBS, 2015) show Rangpur to have a significantly high prevalence of child marriage, particularly in the age group 15–17 years. At the time of the study, Badarganj sub-district had the highest prevalence of child marriage within Rangpur district, with an average of 25 such marriages taking place each month (Hossen, Reference Hossen2013). Local district administration also identified Badarganj as a child marriage hot-spot, and it is also frequently reported as such in local and national newspapers, even today. The sub-district has a population of 287,742 (50.1% male, 49.9% female) with a poverty head count ratio of 48.3% (World Bank, 2016). Badarganj and the surrounding region are characterized by a high level of poverty and low urbanization, with agriculture being the primary means of livelihood. Similar to most rural areas in the country, women in the region often face multiple layers of poverty-induced vulnerability as well as structural barriers imposed by gendered social norms, which limit their physical mobility and access to opportunities, including to education and employment (Das et al., Reference Das, Yasmin, Ara, Kamruzzaman, Davis, Behrman, Roy and Quisumbing2013).
Sample size, data collection and analysis
A total of 64 purposively selected participants from the Vishnupur, Kalupara and Modhupur Unions in Badarganj took part in the qualitative phase of the study, which involved six group discussions, each facilitated by a three-member team of data collectors including the first author. The Unions were selected based on the prevalence of child marriage as identified by local community-based organizations, NGOs and newspaper reports. The participants comprised of 26 females and 38 males aged between 15 and 60 years. They were from mixed socioeconomic backgrounds and included poor, middle class and local elites with varying levels of education and different occupations. The first group consisted of members of local community-based organizations (CBO) (ten), the second group local informal leaders including elders and social elites (ten), the third and fourth groups adolescent (aged between 15 and 19) girls (ten) and boys (ten) respectively, the fifth group local government officials including Union council members (twelve), and the sixth group consisted of imam (Islamic priests), qazi (marriage registrars for conducting Muslim marriages and divorces), teachers, journalists and local NGO officials (twelve). These were all contacted and reached using the field network of BRAC. The average duration of the group discussions was one hour. Six participants from the group discussions, including local leaders and NGO fieldworkers, were also individually interviewed. These interviews aimed at further probing of issues identified during group discussions.
The qualitative phase continued for one month during June 2014. An inductive thematic analysis approach was used to analyse the data obtained from group discussions and interviews, and to identify key themes. The qualitative phase was followed by the second phase, which involved quantitative data collection through a household survey administered for 4 months during August to November 2014. The survey participants comprised a randomly sampled group of males and females aged between 15 and 65 from the study area. Initially the sample size was determined to be 3102 (95% confidence interval, 1.75% margin of error), but this was increased by 15% in order to take into account missing and non-response cases. The final sample size was estimated as 3560. Survey participants’ socio-demographic profiles, including age, marital status, educational level and self-stated household economic status, are shown in Table 1. Once the survey had been administered a total of 3344 survey responses were recorded. The descriptive analysis of quantitative data was made using IBM SPSS Version 22.
Table 1. Socio-demographic profile of survey participants (N = 3344)
Results
Awareness of, and attitude towards, child marriage
The qualitative and quantitative data showed significant awareness among participants of the causes and negative implications of marriage for child brides, particularly the harmful impact on maternal and child health. During group discussions, participants identified parents’ and family guardians’ concerns about safety and security of adolescent girls outside of home, fear of pre-marital relationships, family reputation, poverty and considerations of dowry amount as the main reasons behind child marriage in their area. The majority of survey participants did not support getting a girl married before she was 18 years of age on the basis of these reasons. According to survey data, 79.1% of the participants did not support getting a girl married before 18 as a way of ensuring her safety and security outside of home; 84.1% to protect family reputation; and 78.7% in order to prevent her from having affairs or running away with boys. Moreover, 88.2% of the participants did not support getting a girl married at an earlier age in order to keep the dowry small. Furthermore, 89.5% of the participants agreed that instead of saving for dowry the money should be invested in educating a girl so that she can become self-reliant.
Knowledge and awareness of the national law on the prevention of child marriage (the Child Marriage Restraint Act) varied among the participants. Most (91.8%) were able to cite the legally defined minimum age of marriage for women, but only 41.4% were able to cite the legally defined minimum age of marriage for men. Only 1.4% were able to fully describe the legal consequences of, and punishments for, child marriage as stipulated in national law, while 19.1% were able to describe it only partially and 79.5% reported that they did not know anything about it.
Child marriage practice
Findings from the group discussions and interviews suggested that, despite people being aware of the negative consequences of child marriage, girls are married off before they reach 18 years and usually even before completing secondary level education. The survey data showed that 1712 marriages were reported to have taken place in the 3 years prior to the study in the 1596 households in the study area. The mean age of marriage was 16.84 for all females and 22.37 for all males. A total of 698 marriages (40.8% of all marriages) were identified as child marriages that involved either a bride or both a bride and a bridegroom who were below 18 years. Within the child marriage cases, 87.2% involved a female child and 12.8% involved both a female and a male child as bride and bridegroom. The mean age of marriage was 15.37 for female children and 15.85 for male children. The lowest reported age of marriage was of a girl who was only 8 years old. Within the total 1712 cases, the highest number of females were married at the age of 17 (158 cases, 9.2% of all marriages) and the highest number of males were married at the age of 20 (102 cases, 5.9% of all marriages). Only 4% of all reported marriages involved a bride or bridegroom aged above the age of 28. Figure 1 shows the distribution of age at marriage for females and males between the ages of 8 and 28 years.
Figure 1. Distribution of age at marriage for females and males (N = 1642).
Poverty and child marriage
During group discussions and interviews, participants blamed poverty for girls being marryed off early. They thought that child marriages mostly take place among poorer households, where parents tend to get their daughters married at an early age to ease financial constraint on the family. This was consistent throughout discussion with almost all the participants, including local government officials and local informal leaders.
In our area, child marriage takes place mostly among families who live off government safety net schemes…[among] people who are destitute. (Local government official, male)
According to the participants, in poor households parents are usually not interested in spending money on a girl’s education because it does not bring any immediate benefit. Instead, they take the easy way of reducing the financial burden on the household by getting daughters married as early as possible. They thought issues such as ensuring the physical safety of girls outside the home, as well as the amount of dowry to be given to the bridegroom at the time of marriage, were of major concern to poor households with one or more unmarried daughters. Several participants pointed out that though parents from poorer households are not interested in spending money on their daughter’s education, they save the money or take loans at high interest to pay for dowry.
The survey data questioned participants’ view from qualitative data that poverty is the main reason behind child marriage by showing that nearly 40% of the marriages took place in middle income and wealthy families. Child marriage comprised 44.9% of all marriages among the extremely poor households, 51.5% among poor households, 31.3% among middle income households and 32% among wealthy households. The composition of household economic status of the families that reported marriage of a female or male child is shown in Table 2.
Table 2. Household economic status of families that reported a child marriage (N = 698)
Table 3. Reason for marrying before the age of 18 (N = 698), multiple responses
a For males, the prospect of getting a dowry.
Safety and security of girls, family honour and child marriage
In addition to poverty, the safety and security of young and adolescent girls outside the home emerged as a key concern among parents, local leaders and social elites, which according to them was the main reason behind the high prevalence of child marriage. According to the survey, 36.6% of participants fully or partially supported this view. During interviews and group discussions, participants reported that it is highly unsafe for young girls on the streets and in other places outside of home, especially if they are good looking. As soon as girls turn 12 or 13, teenage boys and young men start stalking and harassing them sexually.
No matter what her age is, if a girl looks physically matured then she starts to draw a lot of attention…if a girl is good looking then everyone, regardless of being a son of a rich or poor family, wants to get their hands on her. (Parent and local informal leader, female)
Adolescent girls reported sexual harassment and stalking on their way to and from school as a routine experience. Parents, local leaders and local government officials claimed that such things bring bad reputation to a girl’s family. According to several participants, because of these concerns, many parents stop sending their daughters to school when they start to look physically mature. They claimed that a girl who is considered ‘good looking’ could also get abducted by a suitor unwilling to accept refusal from the girl or her family. Such girls end up in a forced marriage with the abductor, or even if they return home, they are never able to get married in a good family.
Many of the parents and local leaders expressed concern about adolescent and young girls getting involved in love affairs (with boys) and then running away. According to them, this usually results in the girls becoming victims of sexual abuse and/or other forms of violence. However, this view was challenged by adolescent girls who reported that such incidents were rare as the community does not approve of premarital romance and there is always a strict surveillance of girls, both in and outside the home. Even a casual conversation with a male classmate in a public place would be interpreted by parents and community members as indication of having an affair.
If I even talk about studies or examinations with any boy from our class on the way to school, someone would inform my home [parents] saying that I was having an affair with that boy. (Adolescent female, student)
When asked about any girls getting abducted or running away with a boyfriend during the last 3 years, the parents and local leaders, however, could not provide any example or agree on a definitive answer.
Dowry and child marriage
Dowry was also identified as an important reason for child marriage by the participants. Participants in group discussions and interviews revealed that dowry is exchanged in almost all marriages, with the dowry amount (cash or equivalent amount of assets) depending on two factors – physical attributes of the bride (looks and age) and qualifications of the bridegroom. The amount goes up with the bride’s age, meaning that the older the bride the larger the dowry. At the same time, if a girl is considered as not good looking – for instance, if she has a darker skin colour – then the amount of dowry goes up as well. It works in the same way with the bridegroom’s qualifications such as his level of educational attainment, income or nature of employment and family status. The amount of dowry demanded by the bridegroom or his family increases with his qualifications. Participants emphasized that a girl’s/woman’s qualifications cannot be equal to, let alone higher than, that of her potential husband.
If a girl is a secondary school graduate then her husband will have to be at least a higher secondary school graduate. Otherwise it will not be a good match…If the girl has a job then the to be husband must have an even better job. (Community-based organization leader, female)
Accordingly, the more qualified a girl is (in terms of educational achievement and earnings) the larger dowry her parents will have to pay because she will require a bridegroom with even higher qualifications. According to most participants, this encourages parents to get their daughters married at an early age instead of sending them to school. The dowry amount was reported to range between 70,000 takas (equivalent to 875 $USD at the conversion rate of 1 $US = 80 Bangladeshi takas) to 500,000 takas (6250 $US).
Survey results show that among the total 1712 cases of marriage, exchange of dowry was reported in 37.8% of the cases involving brides aged between 8 and 28 at the time of marriage. However, among the 698 cases of child marriage, dowry was reported to have been exchanged in 74.9% of cases. Figure 2 shows the amount of dowry paid by the bride’s family to the bridegroom/bridegroom’s family by bride’s age.
Figure 2. Dowry paid for brides aged between 8 and 28 (in $US) (N = 648).
Dowry amount ranged from a minimum of 5000 takas (62.5 $US), paid to the bridegroom’s family in two cases by the families of two 14-year-old girls (one extremely poor, one poor), to a maximum of 3,100,000 takas (38,750 $US) paid in the case of a family of a 17-year-old girl (middle income). In only 1.8% of cases was the amount higher than 400,000 takas (5000 $US). Pearson correlation analysis showed a positive and significant correlation between the bride’s age and the amount of dowry paid by the bride’s family (r = 0.14, p<0.01, 2-tailed). However, the dowry amounts also showed a significantly positive correlation with the household economic status of the bride’s family (r = 0.40, p<0.01, 2-tailed). When taking into consideration the individual categories within the household economic status variable, no significant correlation was observed between the bride’s age and the amount of dowry paid by her family within the extremely poor (r = –0.09, p = not significant), poor (r = 0.02, p = not significant) or middle income (r = 0.11, p = not significant) categories. A significant correlation between the bride’s age and the amount of dowry paid was observed only within the wealthy category (r = 0.62, p<0.05, 2-tailed).
A good match
During group discussions and interviews several participants, particularly parents and local leaders, talked about the difficulty many parents face in finding a suitable (good) match for their daughters. According to them, when parents or guardians consider a marriage proposal for their adolescent daughter to be a good one they seldom let the opportunity to have a good match pass. Fearing that they might not get such a good proposal again, parents or guardians arrange the marriage whether or not the daughter has reached 18 years of age. During the survey, participants who reported a marriage in their family (that took place in the last 3 years before the study period) where the bride or the bridegroom was below 18 years were asked to specify the reason for the marriage. As shown in Table 3, more than half of the participants (55.8%) mentioned finding a good match as the reason behind arranging marriage of an adolescent girl.
Several participants mentioned that since a girl has to get married sooner or later many parents and guardians believe it is better to do it as early as possible. There was common agreement among the participants that a husband should be at least 5–8 years older than his wife in order to ensure a harmonious marriage. However, if parents wait for their daughter to turn 18 before she can be married, all the eligible men in their network might already be married. It would then become very difficult for them to find a good match for the girl. This view contrasts sharply with the findings from survey data, which showed that the vast majority of the participants (89.8%) did not support the notion that a girl could be married before she was 18 if her parents decided to do it upon finding a good match for her, even if the girl agreed to it. Participants in group discussions and interviews claimed that most parents and guardians in the locality were well aware of the negative consequences of child marriage, yet if a suitable match becomes available, they would accept it and arrange marriage for their daughter, even if she is still below 18 years of age. According to the participants, often this was done in the hope of securing the daughter’s future while her father is still alive. Some participants claimed that even a number of local government officials and local leaders had arranged child marriage in their families when they found a good match for a daughter (or for any other dependent, unmarried, female family member). It was commonly reported that it is usually the father, or another male guardian in the case of absence of the father, who decides about the suitability of a match and the timing of marriage, and makes the arrangements.
Discussion
The existing literature often emphasizes gender inequity as the primary contextual factor that increases the risk of child marriage (for example, Jain & Kurz, Reference Jain and Kurz2007; Raj et al., Reference Raj, McDougal, Silverman and Rusch2014b; Yarrow et al., Reference Yarrow, Apland, Anderson and Hamilton2015). However, as a driver of child marriage, gender inequality is often considered only as far as it involves poverty, girl’s lower access to education and lack of employment opportunities. As a result, instead of changing the socio-cultural norms that perpetuate gender inequity, emphasis is often given to interventions aimed at changing girls’ and women’s conditions through their increased access to education and livelihood opportunities. The results of the present study suggest that such interventions may not be adequate to prevent child marriage in Bangladesh unless specific focus is given to addressing the socio-cultural norms that perpetuate gender inequity. This is because the relationship between child marriage and its drivers, such as poverty or dowry, is multifaceted and complex.
The fact that during the observation period two out of five child marriages in the study area reportedly took place in middle income or wealthy households, and only one in ten were reported to have been arranged in order to ease financial constraints on the household, indicates that the relation between child marriage and poverty is not as linear and monotonic as commonly believed. In fact, evidence from another South Asian country, Sri Lanka, shows a significant increase in women’s age at first marriage due to poverty and unemployment during the latter part of the 20th century (Caldwell, Reference Caldwell1992). Furthermore, even though in the international development discourse child marriage is usually associated with the economically poor countries of the global south, evidence shows that child marriage also takes place in the more developed countries, albeit in a much smaller number (Raj & Boehmer, Reference Raj and Boehmer2013). For example, several studies have indicated that in certain areas and communities in the United States every year a considerable number of girls get married even before the age of 16 and in most cases the cause is linked to adherence to gender norms and cultural beliefs (McFarlane et al., Reference McFarlane, Nava, Gilroy and Maddoux2016; Koski & Heymann, Reference Koski and Heymann2018). The global discourse of linking child marriage with poverty and stigmatizing the poor is echoed in the claims of the local elites and affluent members of the community in the study area, who strongly associated child marriage with extreme poverty, even though many of them practised it too.
In a similar vein, the widely recognized linear association between girl’s age and dowry amount is also likely to be more complex and variable than it appears at first glance. Findings from this study clearly indicate that dowry is determined not by a girl’s age alone but also by her household financial status, educational attainment and physical appearance. If dowry is to be considered as a strategy for capital accumulation by men (Chowdhury, Reference Chowdhury2010) then it is logical to assume that men from poorer households would seek an older bride in the hope of getting a larger amount of dowry. The high prevalence of child marriage suggests that this is not the case. The fact that men will marry younger girls and in doing so forego potential monetary benefit indicates that financial gain may not be a deciding factor in child marriage. As White (Reference White2013) argued, dowry thus needs be understood not in terms of material gains of the bridegroom and his family but as a socio-cultural and economic investment in patriarchy that supports male provision and ensure male supremacy over women. Moreover, while the present study’s findings related to perceptions and attitudes towards child marriage closely correspond to current understanding of another significant driver of child marriage that involves parents’ and guardians’ anxiety with ensuring physical safety of unmarried girls, practice-related data show that during the observation period in the study area only one in ten child marriages took place because of such concerns. These findings call for a re-evaluation of the relationships between child marriage and its commonly established drivers while at the same time re-focusing on certain gendered discourses produced and maintained by the socio-cultural structures of patriarchy.
In Bangladesh, as well as in many other countries, the idea that a wife should be subservient to her husband in order to create more harmony in the marriage is widespread (Raj, Reference Raj2010; Yarrow et al., Reference Yarrow, Apland, Anderson and Hamilton2015). The practice of child marriage preserves women’s subservience in marriage by ensuring that girls are married young, thus retaining an inequality in marriage that safeguards the husband’s ability to control and dominate his wife due to his greater authority resulting from higher age and higher achievements in terms of education and income. As noted by Kassan and Mahery (Reference Kassan, Mahery and Boezaart2009), child marriage is society’s mechanism to ensure that ‘girls do not develop an independent sense of self but instead are placed firmly under male control where they are expected to be submissive and work for their in-laws’ household’ (p. 195). This paper argues that poverty, dowry and concerns with physical safety of unmarried girls are not only the causes of child marriage in rural Bangladesh but, more importantly, are also core elements of a set of gendered discourses produced deliberately by the structures of patriarchy to ensure retention of a power relation between the intimate partners that advantages men over women. The discourse of ‘child marriage is practised only by the poor’ allows the non-poor, who are apparently aware of the negative implications of child marriage, to stay in denial and shift the blame and stigma onto the poor while at the same time practising it themselves. Similarly, the discourse of ‘smaller dowry for younger brides’ works to produce a perceived fear about potential financial ‘penalty’ for not marrying girls ‘on time’. This is complemented by the discourse of ‘public spaces not being safe for unmarried girls’, which also uses fear as a mechanism to limit young and adolescent girls’ physical mobility outside of home by designating the public sphere as a dangerous space where men (boys, young adults, older men) would readily sexually harass or assault an unmarried girl or try to seduce her for sexual enjoyment. Such discourses compel parents and guardians to be on the lookout for a ‘good match’ for their daughter, often soon after she reaches puberty.
Contrary to common understanding, such discourses are produced not only at the village level or just in rural areas, and the state plays an active role in producing and reproducing them. The concern that is voiced at the rural and grassroots level about protecting family honour and the reputation of unmarried girls by protecting their sexuality is echoed by the state machinery in Bangladesh through the recent addition of a special provision in the Child Marriage Restraint Act 2017 that allows a girl to get married at any age in special cases. According to the Government of Bangladesh, such provisions are necessary because if an unmarried girl gets pregnant it might be ‘disastrous for the society’ and she must be allowed to get married regardless of her age (Asadullah & Wahhaj, Reference Asadullah and Wahhaj2016). It is the same logic that allows child marriage even in one of the most economically developed countries, the United States (Koski & Heymann, Reference Koski and Heymann2018). State authorities in Bangladesh play their part in perpetuating the subservience of women in marriage in other ways as well: for example, by maintaining an age gap between marriage partners through legal provisions that allow women to marry at the age of 18 but require men to be at least 21. Furthermore, even though the exchange of dowry is prohibited by Bangladeshi law as a criminal offence, it is widespread and practised in many forms in the country, while there is seldom any legal action against any parties involved in paying or receiving dowry. Despite dowry being a major cause of domestic violence against women in Bangladesh (Hasan et al., Reference Hasan, Muhaddes, Camellia, Selim and Rashid2014; Young & Hassan, Reference Young and Hassan2018), the state’s neglect of the problem and lack of action to prevent it makes it complicit in perpetuating the discourse that dowry leads to many girls getting married as children.
There is clearly a disjuncture between how people view child marriage and what they practise in real life. The high level of awareness on child marriage and unsupportive attitudes towards it, as expressed by community members during the study, are likely to be the result of interventions by NGOs and international development organizations working in the area. However, the high prevalence despite the awareness shows that merely creating awareness on child marriage does not ensure that it will translate into practice. Similar arguments can be made about girls’ education as well. Girls’ education is usually considered to be the most important protective factor in delaying age of marriage. Poverty is commonly believed to be the major cause of girls dropping out of school (Mahmud & Amin, Reference Mahmud, Amin, Hannum and Fuller2006). Hence, emphasis is given to preventing girls from dropping out of school in order to prevent child marriage (Jain & Kurz, Reference Jain and Kurz2007; Hossain et al., Reference Hossain, Mahumud and Saw2016). However, a recent study has shown that, in the Bangladeshi context, one of the key reasons why adolescent girls from wealthier families drop out of school is to get married (Streatfield et al., Reference Streatfield, Kamal, Ahsan and Nahar2015). The present study also found that girls are taken away from school often on the pretence of ensuring their safety outside of the home, only to get them married. This means many girls get married not because they have dropped out of school, but that they are deliberately taken out of school for marriage purposes. Amin et al. (Reference Amin, Asadullah, Hossain and Wahhaj2017b) observed that success in changing girls’ schooling norms does not necessarily result in a reduction in child marriage unless other factors are simultaneously addressed. This supports the argument made in this paper that investing only in awareness raising or girls’ education to prevent child marriage may not be as effective as desired unless the broader structures of gender inequality, and the particular form of patriarchal control that promotes child marriage, are also changed.
It is important to note that more than half of the total number of child marriages in the study area reportedly took place because the parents found what they considered to be a ‘good match’. Similar findings were also reported by Asadullah and Wahhaj (Reference Asadullah and Wahhaj2019), who noted parents’ willingness to take girls out of school for marriage when the proposal was ‘too good to refuse’. Getting a daughter married while she is still a child, even when it is prohibited by law to do so, simply because a good match has been found results from an anxiety to conform to societal requirements of an ideal marriage, where the husband must play the role of decision-maker and be the primary bread-earner, and the wife will not be seen as capable of challenging the authority and domination of her husband. As a result, even when parents recognize the advantages of investing in a girl’s education and marrying them at a later age, concern over not being able to find a good match compels them towards child marriage.
The study has its limitations. The findings presented in this paper were based on data from one rural sub-district in northern Bangladesh. Due to the regional variation in the prevalence of child marriage in the country, as noted by Islam et al. (Reference Islam, Haque and Hossain2016), these findings may not reflect the overall scenario in Bangladesh. That would require collecting data from a more representative sample from various other regions of the country, including urban and rural settings, and a rigorous regression analysis, which was beyond the scope of this study. The study was carried out as part of BRAC’s Community Empowerment Programme initiative to develop an appropriate child marriage prevention intervention in rural areas of Bangladesh. Hence, the views of the organization’s field workers were taken into account during study site selection. Local communities were also accessed using BRAC’s field networks. However, this did not have any implications on the findings, analysis or discussions presented in this paper.
In conclusion, the study findings offer new insights into the problem of child marriage in rural areas of Bangladesh that could help to build a more in-depth understanding of the issue. In the Bangladeshi context, the discourses of poverty, dowry, and safety and security of unmarried adolescent girls that are commonly used to rationalize child marriage indicate a deeper rooted problem, which involves using marriage as an institution of control over young and adolescent girls’ sexuality and ensuring women’s subservience to men. Child marriage plays an instrumental role in sustaining gender inequalities and women’s subordinate position in the family and thus in society. Investing only in girl’s education and poverty eradication might not be a complete solution to the persistent problem of child marriage in Bangladesh. In a society like Bangladesh, where for women across all classes marriage is inevitable and central to societal life, and their ideal role is that of a caregiver of the family, girls’ education and work outside of home is allowed only as long as this does not undermine their marriage prospects. Hence, the notion of empowerment of girls through education might not always gain traction in people’s everyday life. This is particularly true for rural communities where the sense of community and collective well-being is stronger, and fear of stigma surrounding girls’ virginity and chastity remains high across all classes. People are aware of the negative consequences of child marriage, but at the same time they are also aware of the consequences of transgressing social norms. This results in mismatches between how people perceive child marriage, and what they actually do in their real lives. Hence, across all classes finding a ‘good match’ remains a key impetus for getting girls married at an early age. While there are likely to be differences in how a good match is perceived by people from different class backgrounds, nonetheless it reflects the deep-seated societal gendered expectations of man as the provider and woman as the caregiver. An in-depth qualitative exploration of the issue is needed to fully understand how the meaning of a good match is constructed at different levels (i.e. individual, community and society), and what can be done by different actors, such as the community, state, media and development organizations, to shift the notion of a ‘good match’ to an ‘equal match’, and thereby delay the age of first marriage of girls in Bangladesh.
Acknowledgments
The authors express their gratitude to Ms Anna Minj, Director of BRAC’s Community Empowerment and Integrated Development programmes in Bangladesh, for support in carrying out the study, and to Mr Abu Nasar Mahfuzur Rahman, former Coordinator, Data Management Unit of BRAC’s Research and Evaluation Division, for guidance and support in managing the survey data. The authors also acknowledge the support of several BRAC officials, namely, Mr Palash Ghosh and Mr Kallol Barua, for field logistics and data collection, and Mr Intaz Ali for data entry. Finally, the authors are thankful to the anonymous reviewer for their comments on an earlier version of the manuscript.
Funding
This study was carried out with support from the Community Empowerment Programme of BRAC in Bangladesh. The statements made, and views expressed, in this paper are solely those of the authors.
Conflicts of Interest
The authors have no conflicts of interest to declare.
Ethical Approval
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008. Approval for the study design was granted by relevant BRAC authorities. Study-related information was clearly explained to each participant before data collection and informed verbal consent was obtained. Parental consent was obtained for participants who were below the age of 18. Strict confidentiality was maintained regarding any personal information gathered during data collection and all information was de-identified during data entry.
