Download this publication

The #MeToo movement has helped create a global spotlight on campaigns to end sexual harassment and assault, human trafficking and child marriage. These movements add to a growing emphasis on girls’ rights, highlighted by a number of celebrities recently speaking against child marriage and in this way bringing attention to children’s welfare.[1] Beyoncé allows her song “Freedom” to be used to create awareness about child marriage and Oprah Winfrey together with numerous other celebrities, have publicly demanded action on barriers to girls’ education, including child marriage. But we find in our research that human rights campaigns can actually backfire and decrease support for banning child marriage. 

Female traditional leaders’ effectiveness as policy advocates

The emerging literature on governance and traditional leadership suggests that traditional leaders may have a greater capacity to organize and broker responses to rural problems than elected politicians (Cammack et al., 2009; Díaz-Cayeros et al., 2014; Baldwin, 2016). To unpack how and why this is the case, we need more research. Few studies examine traditional leaders’ effectiveness as public opinion influencers. In our project, we tested whether traditional leaders have a more positive effect on policy advocacy for human rights reform than elected officials. This is the first study to examine the impact of female traditional leaders specifically.

The gender and politics literature places substantial emphasis on how female leaders shape agendas and influence political outcomes (Franceschet et al., 2012). Yet, this research focuses almost entirely on state actors, such as parliamentarians and cabinet ministers, and does not examine the effect of a leaders’ gender in their governance effectiveness. Analyses of how female parliamentarians shape attitudes and a focus on gendered aspects of traditional leadership is missing. This gap exists in part because scholars and practitioners assume that traditional leaders are men, and they expect traditional leaders to promote patriarchal practices (Bauer, 2016; Htun and Weldon, 2010). Yet, female traditional leaders are almost as common as female elected representatives in countries such as Malawi and rural Zambia, where matrilineal cultural practices exist. In Malawi, 10 percent of traditional leaders are women. Female traditional authorities exist where matrilineality is practiced, a custom in which property and inheritance pass through the mother’s line; in patrilineal societies, in contrast, property and inheritance pass through the father’s family. Other African and Asian ethnic groups practice matrilineality as well.

Some female traditional leaders have forwarded feminist goals. For instance, female senior chief Theresa Kachindamoto of Dedza District in Malawi figured prominently in the world media in April 2016, when she annulled 850 child marriages and suspended all village heads that refused to ban the practice of child marriage. Among the Chewa people of Malawi, Mozambique, and Zambia, women played a key role in the selection of their paramount Chief, the Kalonga, who rules on behalf of his mother, called Nyangu. In countries like Botswana, while no woman has served as president, a woman has also been paramount chief. Despite the importance of female traditional leaders in Africa and Asia, their effects as policy advocates are unexplored.

Child Marriage in Malawi

Inclusion of traditional leaders in programs to promote development and women’s rights, is linked to their proximity to the issues and respondents, and the high trust people have in them. Traditional institutions in Malawi play a key role in local communities, and are present across the country. For most Malawians, traditional leaders combine traditional, cultural and governmental functions when they promote development activities and advance the government’s agenda.

Malawi has undergone several attempts to implement reforms that address long-standing cultural traditions considered discriminatory and obstructive to effective development. A recent legal reform in this respect is the Marriage, Divorce, and Family Relations Act (The Marriage Act) of 2015, which banned child marriage. Child marriage is commonly defined as a formal marriage or informal union entered into before the age of 18. The importance of this reform is underscored by the fact that Malawi has one of the highest rates of child marriage in the world. The country is ranked 9th internationally in terms of child marriage prevalence (UNICEF, 2015), which is common across all three regions of the country, but highest in the central region.

To curb the practice of child marriage, international development agents and the Malawi government emphasized the importance of including traditional leaders in the campaign. This makes sense since most child marriages are conducted under customary law and presided over by village headmen, who are widely considered in Malawian society to be custodians of culture. Traditional leaders across the country have been urged by the government to develop action plans against child marriage, and to attend workshops presenting different aspects of the law.

How we did our research

We conducted a survey experiment in Malawi in 2016 with a team of researchers at the Program on Governance and Local Development (GLD), which allows us to examine the impact of authority type (whether traditional or elected) and authority gender (male versus female) on citizens’ support for reform to stop child marriage. The experiment was embedded within the larger Local Governance Performance Index (LGPI). While 7,500 Malawians participated in the LGPI, the experiment included 1,381 Malawians who were randomly assigned to one of five groups. Each of the first four groups heard a different statement against child marriage. The fifth group heard no such message.

Each of the four messages had a different messenger advocating against child marriage:

  • A male traditional authority, also sometimes referred to as a chief.
  • A female traditional authority.
  • A male member of parliament.
  • A female member of parliament.


To understand the role of different authorities in increasing opposition to child marriage, we asked respondents to rate on a four-point scale the extent to which they agree that the minimum age of marriage should be raised to 18 years. A control group received no frame; respondents from this group were simply asked the extent to which they would support the law to increase the marriage age to 18. Here’s an example of the statement and subsequent question participants in the Malawi study heard:

“A female traditional authority is supporting the new law that sets the minimum age of marriage at 18. What do you think of the law? Do you strongly support it, support it, not support it or strongly not support it?”

The control mentions that the new law was passed by Parliament, while the treatments stated that the new law was supported by different actors. We carefully considered different ways of formulating the control and the treatment conditions in order to make them as similar as possible. Some might consider the control to be a stronger statement because it mentions that the new law was passed by parliament. However, we see no reason that the new law being passed as opposed to being supported connotes a stronger treatment. This is because we expect respondents would have interpreted this question through the lens of the Marriage, Divorce and Family Relations Law of 2015. At the time of the survey in 2016 there was still continuing public debate over the provisions of the new law and also ongoing initiatives by government, legislators, and civil-society organizations to sensitize the public about the new law. The term law, as opposed to bill, in the Malawian setting denotes a rule or regulation that has been passed by Parliament (and assented to by the President), and is legally enforceable.

The messenger mattered, in this study

In our study, we found most Malawians (76 percent) strongly supported a law restricting marriage for individuals under 18 years of age. 10 percent supported the concept. Only 15 percent did not support it. Support for the law was not significantly higher among women than men — but was higher in matrilineal areas and among those who support gender equality.

Our survey experiment allowed us to see whether certain messengers were more effective at promoting campaigns to end child marriage. We found that messages from a female traditional authority were the most effective overall. Support for the reform was slightly higher when endorsed by one of the female messengers than by either of the male messengers: 75 percent strongly supported the reform when advocated by a female MP, while 78 percent strongly supported the reform when advocated by a female traditional authority. If the advocate was male, only 74 percent (male MP) and 72 percent (male traditional authority), respectively, strongly supported the reform.

But, the female traditional authority endorsement was not more effective than having no endorsement at all — the control group. We found that 80 percent supported the law when there was no endorsement. When the message came from a male authority figure — either a male traditional authority or an MP — there was a backfire effect relative to the control. Female MPs also produced a backfire effect relative to the control.

Why are women chiefs most effective in advocating against child marriage?

We believe that Malawians’ high trust in traditional authorities — coupled with their perceptions that women are competent on gender-based issues — make female traditional authorities the most effective endorsers. According to a 2014 Afrobarometer survey, Malawians are more likely to approve of the performance of traditional authorities than MPs.

The Afrobarometer survey revealed 47 percent of men and 53 percent of women trust “parliament,” while 66 percent of men and 73 percent of women trust “traditional leaders.” Another question on whether respondents approved of how Malawi leaders had performed their jobs over the past 12 months, only 30 percent of men and 35 percent of women approved of “Your MP” — but 79 percent of respondents of both genders approved of their “traditional leader.”


Created with Highcharts 6.1.1 The Afrobarometer survey 79 79 30 30 47 47 66 66 79 79 35 35 53 53 73 73 men woman Approve performance of traditional leaders Approve performance of parliamentarians Trust parliament Trust in traditional leaders 0 10 20 30 40 50 60 70 80 90

 Trust and performance approval for traditional and elected leaders in Malawi (Afrobarometer, 2014). Source: Afrobarometer R6 2014.Question wording: “How much do you trust each of the following, or haven’t you heard enough about them to say?” “Parliament,” “Traditional leaders.” "Do you approve or disapprove of the way that the following people have performed their jobs over the past twelve months, or haven’t you heard enough about them to say?” “Your MP" and "Your traditional leader.”

In in the LGPI survey as well, Malawians have higher trust in their traditional authorities than their parliamentarians: 53 percent of respondents trusted their traditional authority very much, compared to 28 percent expressing high levels of trust in their MPs.

Cultural backgrounds can make some messages backfire

All four messengers caused backfire effects among some sub-segments of the population — actually decreasing support for banning child marriage. The backlash we saw was related to kinship systems, an important cultural practice. In Malawi, different groups have different kinship systems: in matrilineal groups such as the Yao, kinship follows the mother’s line; in patrilineal groups such as the Tumbuka, kinship follows the father’s line.

Among people in matrilineal groups, we found that male messengers caused a backfire effect — lower support for banning child marriage if a male traditional authority or male MP advocated the reform than if there was no endorsement (i.e., the control group). Among people in patrilineal groups, the most effective endorser was the male MP, who was more effective than the control. This raises concerns human rights campaigns might have limited or even negative impacts, but also shows that the most effective endorser depends on the group that is hearing the message.

What do these findings mean for human rights campaigners?

Our study is not the first to recognize the possibility that human rights campaigns might backfire and cause those who oppose a change to become even more resistant. To be sure, this research was conducted in Malawi, and the findings may not apply in other countries in the same way. But our research shows that it is important for policymakers developing programs to curb child marriage and other human rights campaigns to consider messenger effects and the potential for backlash.

More specifically, campaigners may need to look to cultural backgrounds and other population characteristics when identifying the most effective messenger for any campaign. Tailoring programs based on this type of research may well be key to doing no harm.


Baldwin, K. (2016). The paradox of traditional chiefs in democratic Africa. New York, NY: Cambridge University Press.

Bauer. G. (2016). ‘What is wrong with a woman being chief?’ Women Chiefs and Symbolic and Substantive Representation in Botswana. Journal of Asian and African Studies, 51(2): 222-237. doi: 10.1177/0021909614545700

Cammack, D., Kanyongolo, E., & O’Neill, T. (2009). “Town chiefs” in Malawi. Working paper no. 3, Africa power and politics, London: Overseas Development Institute.

Díaz-Cayeros, A., Magaloni, B., & Alexander, R. E. (2014). Traditional governance, citizen engagement and local public goods: Evidence from Mexico. World Development, 53, 80-93. doi:

Franceschet, S., Krook, M.L., & Piscopo, J.M. (2012). The impact of gender quotas. Oxford: Oxford University Press.

Htun, M., & Weldon, L. (2010). When do governments promote women’s rights? A framework for the comparative analysis of sex equality policy: A comparative politics of gender. Perspectives on Politics, 8(1), 207–216.doi:10.1017/S1537592709992787.

UNICEF. (2015). The state of the world’s children 2015. New York: United Nation Children’s Fund. Retrieved from:



[1] See for instance:;

Lise Rakner

Professor at University of Bergen and Affiliated Research Professor