Abstract
The presence of gender provisions in peace agreements affects women’s participation in post-conflict societies as well as the chances that a post-conflict society will move towards gender equality. While there is an overall upward trend in the number of references to women’s rights and gender equality in peace agreements, gender-sensitive agreements are not a given. Why and how are peace agreements with gender provisions adopted? We use statistical analysis to explain why some peace agreements adopt gender provisions while others have no such provisions. Based on an analysis of 98 peace agreements across 55 countries between 2000 and 2016, we find that peace agreements are significantly more likely to have gender provisions when women participate in elite peace processes. Our study also shows that the likelihood of achieving a peace agreement with gender provisions increases when women’s representation in national parliaments increases and when women’s civil society participation is significant.
Keywords
Introduction
The record of women’s participation and representation in peace talks is extremely poor. Until the adoption of United Nations Security Council Resolution (UNSCR) 1325 on Women, Peace and Security (WPS), few peace agreements mentioned women’s rights or contained commitments to achieving gender equality in post-conflict settlements. Until recently, peace processes have been almost wholly the preserve of security providers understood to be men and armed groups as opposed to security stakeholders understood to be civilian women, girls, boys and men.
Almost two decades after the introduction of the WPS agenda, there has been some progress in state commitments to ensuring the full, equal, and meaningful participation of women in peace processes. However, the implementation of these commitments is judged by experts, non-governmental organisations (NGOs), states and the United Nations itself, to have been extremely slow (UN Women, 2015). Women are not consistently engaged in negotiating peace and are still frequently excluded from many peace processes (Aggestam and Svensson, 2018). The United States Council on Foreign Relations (2018) has found that in all major peace processes between 1990 and 2017 just 8 per cent of mediators, 8 per cent of negotiators and 5 per cent of witnesses and signatories to peace agreements were women. 1 These trends in gender inclusion in peace processes take place in the context of two global trends: the intensification of armed violence and conflict across the world over the last decade, in which nearly 90 per cent of those killed are civilians, many of them women and children; and the contemporary failure of half of all peace processes, where most ongoing civil wars are the result of previous, recurring conflicts. 2
Given the intensifying, gendered impacts of conflict as well as the recurrence of civil wars, women’s participation in peace processes and the attentiveness of peace agreements to gender inequality issues is imperative. In particular, the presence or absence of gender provisions in peace agreements affects women’s post-conflict participation and the chances that a post-conflict society will move towards gender equality. Gender inequalities within states are known to affect state and non-state armed groups behaviour including the instigation of conflict, the escalation and severity of violence during crises and the prevalence of conflict-related sexual and gender-based violence (Caprioli, 2005; Davies and True, 2015; Melander, 2005). Moreover, harnessing both women and men’s capacities is crucial to efforts to promote recovery from conflict (True and Hewitt, 2018).
While there is an overall upward trend in the number of references to women’s rights and gender equality in peace agreements since 2000, gender-sensitive agreements are not a given. To give a sense of the mixed progress in this area, the most recent annual report of the UN Secretary-General on WPS observed a decline in all areas of establishing gender-sensitive peace processes including the participation of women, the use of gender technical advisors, and the inclusion of gender provisions in peace agreements. 3 Why then are peace agreements with gender equality and women’s rights provisions adopted? What are the factors that give rise to agreements with explicit gender provisions?
This article aims to shed light on these important questions by adopting a theoretically informed, quantitative approach. 4 All peace agreements are different, as case-specific analysis reveals, with respect to their process design, the levels of public transparency (Jeong, 2005), and the nature of participation in negotiating agreements and the opportunities for women or any other social group to participate (Paffenholz, 2014h). The conditions leading to adoption of peace agreements with gender provisions may also differ widely across cases. Given such diversity, in this study we seek to establish whether there are patterns across peace agreements (and their processes) that might differentiate those with gender provisions from those without such provisions. It is crucial to know whether there are common factors in peace processes where gender-sensitive agreements are negotiated, in order to further understanding of the pathways to more gender-inclusive agreements and of the strategies to achieving gender-equal and lasting peace.
Gender inclusion in peace processes
The 2000 preamble to the landmark UNSCR 1325 stresses the importance of women’s ‘equal participation and full involvement in all efforts for the maintenance and promotion of peace and security’. The rationale for women’s inclusion in peace processes is clear: equal participation is a right, women’s participation contributes to the prevention and resolution of conflicts while the lack of this participation undermines it, and women as well as men play important roles in rebuilding the fabric of recovering societies.
Scholarship on peace processes and transitions has established a substantial relationship between women’s involvement in efforts to prevent, resolve and rebuild from conflict and the effectiveness of those efforts with respect to their impact on the outcomes and durability of peace (Krause et al., 2018). The presence of women in peace processes as witnesses, signatories, mediators and/or negotiators ostensibly makes it 20 per cent more likely that a peace agreement will last at least two years and 35 per cent more likely that it will endure (O’Reilly et al., 2015: 1).
Whereas this research has established that women’s participation matters to peace processes, it does not assess or reveal women’s influence on peace negotiations (Paffenholz et al., 2016). Currently for instance, there is no substantial body of research that examines the provisions within peace agreements that address gender equality and women’s rights and what gives rise to them. Why and how are peace agreements adopted with provisions relating to women’s political participation, gender-specific transitional justice, post-conflict land rights and economic development, measures to redress violence against women and girls, and so on? While we know women’s presence in agenda-setting forums can be politically important, to date no research has explored the relationship between different types of women’s participation and gender or women’s rights provisions in the resulting peace agreements (cf. Goetz and Jenkins, 2016: 220; Jenkins and Goetz 2010). As Kara Ellerby (2016) has argued, we need to extend that focus on the descriptive representation of women to analysing the substantive representation of women’s perspectives in peace processes.
In their investigation of specific provisions that refer to women from January 1990 to May 2010, Bell and O’Rourke (2010: 955–956) found that prior to the adoption of UNSCR 1325, just 11 per cent of peace agreements referenced women, while after its adoption 27 per cent of signed agreements made reference to women. However, this research does not examine the factors that differentiate gender-sensitive peace agreements from those agreements with no reference to women or gender equality. What might be the key conditions for gender equality provisions to be included in a peace process and resulting agreement?
Scholarship on gender and politics suggests two major enabling conditions for gender equality agenda-setting or policy change that may also affect peace processes. First, women’s voice and presence in positions of political power is a crucial condition for introducing new issues informed by women’s experiences to peace agendas. Second, women’s movements and civil society organising are also essential for democratising agendas and achieving policy change that benefits women as a group.
With respect to the first condition, Tripp (2015) argues that the disruptions wrought by conflict provide new opportunities for women to be political and economic agents. With Melanie Hughes (Hughes and Tripp, 2015) she finds that the longer and more intense a conflict is, the greater the rupture to gender relations and more likely women will actively contest and win election as representatives. In turn, these newfound legislative positions enable women to demand inclusion in determining the nature of peace. This finding builds on earlier research by Hughes (2009) that shows how women in conflict-affected states may be able to advance their political position and rights despite gendered inequalities. The reality of some protracted conflicts catapulting women into politics is an enabling condition for the emergence of gender provisions. Women representatives may have a unique ability to agenda-set in peace processes and bring women’s rights or gender equality perspectives to the table.
Conversely, Andersen and Swiss (2014) explore the impact of gender equality and women’s rights provisions in peace accords on the post-conflict promotion and adoption of electoral gender quotas. Their analysis reveals the significant impact of gender provisions in post-conflict settings but it does not analyse the factors that lead to agreements with gender provisions in the first place. Do you need a certain number of women in formal political positions during a peace process in order to adopt a provision relating to women’s electoral participation, for instance? Overall, there is little research across global regions on women’s engagement in democratic decision-making in conflict-affected states. There is a need to connect research on women’s political representation, with peace and security scholarship on women’s presence and gender equality agendas in peace processes.
A second enabling condition for gender-sensitive peace agreements highlighted in scholarship on gender policy change is the inclusion and role of women’s civil society movements (Htun and Weldon, 2010, 2012; Sawer and Grey 2009). For instance, when women’s groups engage in community-level conflict prevention activities, such as the Women’s Situation Room replicated across African states, networking with the media, partnering with traditional gatekeepers including political parties, diplomatic officials and the like, they shape new models of social organisation, and prepare communities to accept new ideas (Onyesoh, 2018).
Tripp discusses how during the 2007–2008 Kenyan election violence, women’s civil society groups persisted in their calls for dialogue, monitoring the violence, and attempting to de-escalate tensions (Tripp, 2015: 37). That grassroots mobilisation process influenced the African Union to appoint a woman mediator in the three-person conflict mediation team, Graca Machel from Mozambique. They liaised with the Women’s Consultation Group, one of four civil society organisations active during the negotiations, and supported the Group’s advocacy for key gender equality provisions in the 2008 National Dialogue and Reconciliation Accord.
In this article, we explore these relationships among women’s elite participation in peace processes, women’s political representation and women’s civil society participation and their impact on the adoption of gender provisions in peace accords. Our study sets out the multifaceted elements expected to affect whether or not a peace agreement includes gender equality and women’s rights provisions. With statistical inference, we are able to show (based on the universe of cases 2000–2016) which factors (and in which combination) are most significant in enabling the inclusion of gender provisions in peace agreements.
Explaining the inclusion of gender provisions in peace agreements
Why do some peace agreements adopt gender provisions and others have no such provisions? This research question is important given that at least half of all peace processes in our dataset contained such provisions. To answer it we constructed an original dataset of 98 peace agreements across 55 countries between 2000 and 2016. The significance of this period is the development of the WPS agenda – beginning with the adoption of UNSCR 1325 in 2000 and currently consisting of eight UNSCRs and associated policies addressing the gendered impacts of conflict. This agenda calls on states to promote women’s equal participation and integration of gender perspectives in peace and security, as well as to uphold women’s rights during and after conflict. To date, half of all the states in the world that have national action plans (NAPs) to achieve these WPS goals are conflict-affected (True, 2016), and their commitments and resources for implementation vary greatly.
Defining peace agreements and gender provisions
To define ‘peace agreements’ (with or without gender provisions), our dependent variable, we initially drew on the Political Settlements Program, the Women and Peace Agreements Database (PA-X Women) released in 2015 (Bell, 2015). 5 This is a comprehensive database on all available peace agreements and their associated gender provisions. However, most studies devise criteria to limit the full range of agreements studied because many agreements included in the database are temporary, or are seen to be not fully peace agreements when reviewed closely. Thus, we refined the PA-X Women database into a dataset of peace agreements 2000–2016 regardless of whether they are lasting or have been implemented yet. We cross checked the dataset with the PA-X Women database and with the UN Peacemaker data set 6 as well as consulting country expertise. Five types of peace agreements are included in our analysis: (i) accords and non-binding agreements, (ii) preliminary agreements, (iii) ceasefire agreements, (iv) final, comprehensive and/or implementation plans, and (v) constitutions.
We included up to three agreements per conflict based on the following four criteria, where the agreement:
relates to an actual conflict (by contrast with land deals or cooperation agreements);
is the most recent pertaining to the conflict;
is the most comprehensive in terms of its substantive provisions; and
is the most operational – there are provisions that can be practically implemented, and it is not just a statement in principle.
For example, for the conflict between the Colombian Government and the Revolutionary Armed Forces of Colombia (FARC), we included two peace agreements texts that meet these four criteria and reflect different peace processes pertaining to the same conflict: The Colombian Agreement to End the Armed Conflict and Build a Stable and Lasting Peace, 2016; and A General Agreement for the Termination of the Conflict and the Construction of a Stable and Lasting Peace, 2012.
To define ‘gender provisions’ of peace agreements we considered references that acknowledge the gendered experiences of conflict and the need to address such experiences as the country moves towards peace by mentioning gender, woman/women, girl(s), sex (es), female(s), mother, widow(s), daughter(s), pregnancy, maternity, and marriage, and and/or issues uniquely, predominantly or specifically affecting women in conflict and peace. Our dataset includes five main types of gender provisions related to
the participation of women (i.e. gender quotas, national women’s machineries, in governmental bodies and post-conflict governance);
post-conflict issues and transitional justice (accountability, consideration of the special needs of women during repatriation, rehabilitation, reintegration);
violence against women (e.g. sexual violence as a violation of ceasefire);
women’s economic empowerment and development (i.e. land and inheritance rights, women’s roles in implementation, education, capacity-building, special attention to female heads of households); and lastly,
any reference to international women-specific legal or human rights mechanisms (such as Convention on the Elimination of All Forms of Discrimination against Women (CEDAW or UNSCR 1325). 7
These gender provisions may be broad commitments to gender equality or non-discrimination on the basis of sex, for instance. However, a generic human rights or anti-discrimination clause such as in the Lesotho 2014 agreement, which does not specifically reference sex or gender and the peace and security context, is not considered a gender provision per se. Provisions might be commitments to address specific gender-based harms, such as, the two provisions prohibiting sexual violence in conflict against civilians by military/armed actors in the 2014 Central African Republic agreement. Gender provisions, moreover, may detail the provision of dedicated services and infrastructure to support women’s recovery from conflict such as article 5 in the Central African Republic 2015 Agreement between the Transitional Government and the Armed Groups on the Principles of Disarmament, Demobilisation, Reintegration and Repatriation, which specified that women combatants will be demobilised separately from men but will have the same opportunities.
Table 1 shows the number and proportion of peace agreements meeting the criteria above that contain gender provisions or not. Roughly half of all peace agreements between 2000 and 2016 contained one or more of the five types of gender provisions.
Peace agreements with gender provisions 2000–2016.
To develop our analysis of the factors affecting the likelihood of a peace agreement including gender provisions, we have used peace agreement as the unit of analysis. 8 We employ a Logit Model to examine why and how peace agreements with gender provisions are adopted. The logit model is a nonlinear methodology that models the probability of a binary data outcome as a function of covariates. The model helps us to understand which variables are associated with the occurrence of an event of interest in a particular country (the dependent variable: peace agreement signing with or without gender provisions). This is variable Y, which takes the value ‘1’ when the event is realised, and ‘0’ otherwise. 9
Our statistical analysis explores the relationship between the peace agreements with gender provisions and the independent variables, considering the year in which the peace agreement was signed. Below we discuss the independent variables we included in the analysis. These variables or factors are drawn from existing studies analysing gender, politics and peace processes which give us a strong reason to expect that they would affect the likelihood of peace agreements including gender provisions.
Women’s participation in elite peace processes
Women’s participation in elite peace processes is a factor with strong potential to affect whether or not peace agreements include gender provisions, given the existing research on the effectiveness of women’s participation in peace talks and its linkages to sustainable peace outcomes (Krause et al., 2018; Paffenholz et al., 2016). Using UN sources (UN Women, 2012), media reports and peace monitoring data, we searched for the names of women as signatories or the presence of women as core participants in Track 1 (official and high-level talks) and Track 2 (unofficial dialogues that inform or support the main talks) elements of peace processes. Women’s participation in elite peace processes (i.e. Track 1 or 2) was coded as ‘1’ when the peace agreement includes women’s participation in the process, ‘0’ otherwise.
State governance and women
The presence of women in positions of political power as parliamentary representatives and their participation in civil society are significant factors considered by feminist political scientists to affect gender-responsive policy (Bashevkin, 2014; Childs and Krook, 2007; Htun and Weldon, 2012; Stetson and Mazur, 2010). We expected these factors to have a positive impact on the adoption of gender provisions in peace accords. To test our expectation, we analysed three variables.
The first variable is proportion of seats held by women in national parliaments, that is, the number of seats held by women members in single or lower chambers of national legislatures, expressed as a percentage of all occupied seats.
The second variable we considered is a women’s civil society participation index. This index ranges from 0 to 1 and measures women’s ability to engage in civil society organisations and public debate freely. It is composed of three dimensions based on over 2500 qualitative local and cross-national experts’ judgments aggregated in a Bayesian item response theory model (Pemstein et al., 2015): women’s freedom of discussion, women’s participation in civil society organisations, and women’s representation in the ranks of journalists (Sundstrom et al., 2015: 13–14).
The third variable relating to state governance we examined is the presence or absence of a NAP on WPS by the government party to a peace agreement. Previous research has shown that NAPs are influential and associated with women’s leadership and engagement in political institutions (see True, 2016). Given their focus on WPS issues, where NAPs have been adopted in conflict-affected countries we would expect them to have a very strong impact on the commitments contained within peace agreements. This variable is coded as ‘1’ in the year when the country signed a plan and for the duration of the plan, ‘0’ otherwise. 10
International factors
We created a variable measuring the number of Women’s International Non-Governmental Organisations (WINGOs) per peace agreement country. We expected WINGOs to have a significant impact on the adoption of gender provisions because they diffuse knowledge, resources and strategies for promoting gender perspectives in peace and security (see Onyesoh, 2018). It is difficult to capture all the organisations, networks and activism in the transnational and international realms, however. The variable was constructed with a sample of civil society organisations including those that participated in the 2015 UN Global Study on the Implementation of UNSCR 1325 (2000), NGOs belonging to the 1325 working group, and the original WINGOs list of organisations updated from the measure used in True and Mintrom (2001) consisting of WINGOs present at the four UN global women’s conferences, 1975–1995. Further, we cross-referenced the WINGOs registered at June of 2017 in the Yearbook of International Organisations under the keywords ‘women’ or ‘gender’ as NGOs and included only those organisations currently registered in our variable. 11
Our sample was also limited to the WINGOs with membership branches in more than one country in the research and with UN NGO consultative status. 12 Because of the low variability in the number of WINGOs between each year for the period 2000 to 2016, the presence of the WINGO by country was verified in five intervals (2000, 2002, 2007, 2010 and 2017).
State socioeconomic factors
With respect to state-level socioeconomic factors that affect the status of women and the capacity for addressing gender equality in peace processes, we considered three factors. First, we analysed the fertility rate (calculated based on the total number of births per woman). This represents the number of children that would be born to a woman if she were to live to the end of her childbearing years and bear children in accordance with age-specific fertility rates of the specified year. According to gender and development scholars and the World Bank, fertility rates are an important indicator of women’s empowerment both in terms of reproductive health and gender, because they indicate the status of women within households and their ability to make autonomous decisions about the number and spacing of children. Fertility rates are frequently used in political science research as a proxy indicator for gender equality with lower fertility rates often coterminous with higher levels of gender equality (e.g. Caprioli, 2005). We expected lower fertility rates to be associated with gender equality provisions in peace agreements.
Second, we analysed the female to male ratio of gross enrolment in tertiary education. Tertiary education, whether or not leading to an advanced research qualification, normally requires, as a minimum condition of admission, the successful completion of education at the secondary level. The gender ratios in primary and secondary education are now less useful because parity has been achieved at primary level, and nearly with respect to secondary enrolment. Here we sought to test the effect of higher-education skills and capacities of women/girls on the propensity for gender provisions to be put forward in peace agreements. Lower relative female education levels, for instance, might explain mobilisation for greater attention to women’s rights and equality, whereas higher relative female education levels might explain the adoption of gender provisions within a peace agreement in terms of the presence of highly skilled, educated women able to advocate for such provisions.
Third, we included a measure of women’s civil liberties judged in terms of personal rights in society and access to property rights and justice, which we expected would have a positive effect on the inclusion of gender provisions. This variable is part of the women’s political empowerment composite index, which ranges from 0 to 1, where ‘1’ is the highest level of empowerment. Empowerment here is defined in terms of women’s civil liberties and combined with the indexes on women civil society participation and women’s political participation (Sundstrom et al., 2015).
National income and government spending, including military expenditures
Three variables are included to reflect the relative income wealth of a country as well as its government spending and specifically its military spending using the World Bank’s online databank source. First, the Gross National Income (GNI) per capita based on purchasing power parity variable is used to measure the wealth of a country which we might expect to affect the likelihood of gender provisions in the peace agreement being adopted as the income resources available in a country have an impact on state capacity to adopt as well as implement innovative policies and is typically included in models seeking to explain political outcomes.
Second, general government final consumption expenditure as a percentage of gross domestic product (GDP) is analysed. This variable includes all government current expenditures for purchases of goods and services. It also includes most expenditures on national defence and security, but excludes government military expenditures that are part of government capital formation. Again, we expected higher levels of government spending to be associated with the adoption of gender provisions, given that the implementation of such provisions generally requires government expenditure.
Third, we incorporated a variable on military expenditure as a percentage of GDP. Military expenditures data is taken from the Stockholm International Peace Research Institute and is derived from the North Atlantic Treaty Organisation definition, which includes all current and capital expenditures on the armed forces, including peacekeeping forces; defence ministries and other government agencies engaged in defence projects; paramilitary forces, if these are judged to be trained and equipped for military operations; and military space activities. We expected increased military expenditures to decrease the chances that gender provisions will be present in a peace accord given that they are generally non-military remedies for conflict, and if greater resources are expended on militaries then fewer resources will be available for non-military initiatives such as implementing gender equality and women’s rights (e.g. Cockburn, 2010).
Living standards
As measures of standard of living and vulnerability in a country, we included improved sanitation facilities from the World Bank database. This indicator refers to the percentage of the population using improved sanitation facilities. These facilities are likely to ensure hygienic separation of human excreta from human contact and may be particularly important for women’s health and participation. 13 Water, sanitation and hygiene (WASH) are known in the international development community to be major issues for women in particular (menstruation, reproductive health, maternal health) and attention to them can help achieve greater gender equality. As such, WASH programmes are a significant feature of gender equality development programmes. In conflict-affected situations, water and sanitation facilities may be negatively affected or relatively poor, such as in internally displaced person and refugee camps. These living standards have particular impacts on women and gender equality and thus have relevance to our study of the conditions under which gender and women’s rights provisions might be adopted in a peace agreement. We might expect higher living standards and sanitation to be pre-conditions for the adoption of gender provisions.
International women’s rights norms
This factor measures compliance with international gender norms. The variable is coded as ‘1’ for the years in which countries in the sample have ratified the CEDAW without reservations on the articles 2 and 16 considered core to the Treaty or ‘0’ otherwise (given that ratification including with reservations is virtually universal). 14 We expected that states that have already made strong commitments to women’s rights, as demonstrated by their unreserved compliance with CEDAW, would be the most receptive to adopting gender provisions in peace agreements (True, 2016: 314).
Net official development assistance (ODA) consists of disbursements of loans made on concessional terms (net of repayments of principal) and grants by official agencies of the members of the Development Assistance Committee (DAC), by multilateral institutions, and by non-DAC countries to promote economic development and welfare in countries and territories in the DAC list of ODA recipients. It includes loans with a grant element of at least 25 per cent (calculated at a rate of discount of 10 per cent). Data are in current US dollars compiled from the World Bank databank. We expected higher levels of ODA to be associated with the adoption of gender provisions as donor states and multilateral institutions often provide aid specifically to promote gender equality and women’s empowerment in conflict-affected and developing countries. 15
Model and analysis
The summary statistics for the independent variables discussed in the previous section are shown in Table 2.
Overview of variables.
Obs.: observations; SD: standard deviation; NAP: national action plan; NGO: non-governmental organisations; GNI: gross national income; PPP: purchasing power parity; GDP: gross domestic product.
The results for the logit model are presented in Table 3. These results indicate that the three types of women’s participation taken together – women’s participation in elite peace processes; their political representation in parliaments; and women’s civil society mobilisation – are major conditions for the adoption of gender provisions in a peace agreement. The cases of negotiating peace agreements with gender provisions within our study help to make sense of this finding. For instance, The 2014 Comprehensive Agreement on Bangsamoro was internationally recognised for its relatively high level of meaningful participation by women in both the formal negotiations and parallel peacebuilding efforts. This participation was supported by a strong level of formal political participation for women in the Philippines national parliament and in the peace process itself (OPAPP, 2016).
Factors influencing adoption of gender provisions in peace agreements.
Note: Standard errors are in brackets. Pseudo R2 = 0.4555; Probability > χ2.
NAP: national action plan; NGO: non-governmental organisations; GNI: gross national income; PPP: purchasing power parity; GDP: gross domestic product.
Significant at 0.1; **significant at 0.05; ***significant at 0.01.
The finding that a peace agreement is far more likely to have gender provisions when women participate in elite peace processes is consistent with political science research that shows a demonstrable connection between the presence of executive decision-making leadership roles and the inclusion of pro-women policies on government agendas (Bashevkin, 2014). Peace processes are high-level fora and women’s presence in these fora increases the chances that pro-women’s rights and gender equality provisions are adopted in agreements designed to bring an end to conflict and promote the security and recovery of the communities affected. In the Colombia case explored in the next section, women members of the Colombia Government and the FARC negotiating teams lobbied for a gender sub-commission to enable greater access to the peace process by women’s civil society groups. As a result, delegations of women victims of conflict-related sexual and gender-based violence were able to provide formal representations to the conflict parties during the negotiating process (see Carvajal and Ãlvarez-Vanegas, 2018).
Women in parliament do not necessarily serve as voices for women as a group in society. However, the positive finding here suggests that as the numbers of women in parliament increase, the diversity of women’s experiences of war, conflict and insecurity can be articulated and translated into claims vis-à-vis the state, in this case in peace agreements. We can see this process emerging in conflict-affected countries like Kenya and Nepal where women’s political representation was critical to the strong gender provisions adopted in both countries’ constitutional peace processes respectively in 2008 and 2015 (see Namako and O’Reilly, 2018).
Women’s civil society participation is also shown to be one of the most important factors in achieving peace agreements with gender provisions. That participation includes women’s freedom to engage in public debate and discussion, their participation in civil society organisations and representation in the ranks of journalists. This finding concurs with other research on women’s organising (Htun and Weldon, 2012), in particular how local women’s civil society activists lobbying for inclusion in peace processes deploy the UNSCR 1325 framework (see Onyesoh, 2018). Women’s groups have been more active than any other groups in collectively organising in favour of peace agreements. In Liberia in 2003, for example, the Women in Peacebuilding Network arranged demonstrations at important venues and during key moments of the negotiations that were instrumental in reaching a final agreement. In Colombia, women’s organisations mobilised through two major national summits in 2013 and 2016 to push for women to be represented in the elite peace process between the Colombian Government and the FARC. Once women’s representation within the negotiations was secured, women’s civil society organisations had a mechanism for substantively influencing the peace agreement.
A strong relationship was found between being party to adoption of a NAP on WPS and the inclusion of gender and women’s rights references in the peace accord to which they are also party. This is not surprising, given the consonant nature of WPS NAPs and peace agreement gender provisions. The processes for negotiating these may be overlapping, as we have seen in the case of Iraq (Parry and Aymerich, 2018). However, given the donor-driven nature of NAP adoption in many conflict-affected countries and the under-resourcing of their implementation – in 2014 only 11 out of 47 NAPs had a specific budget (UN Women, 2015) – we interpret this factor as less significant than the synergy among the types of women’s participation.
Other factors that reduce the likelihood of achieving a peace agreement with gender provisions include the total fertility rate, female ratio of gross enrolment in tertiary education, general government final consumption expenditure, military expenditure and better sanitation facilities. If the total fertility rate increases, peace agreements with gender provisions are less likely. We would expect this, given that fertility rates are considered a proxy for gender equality, with lower fertility rates associated with greater gender equality in the public sphere and the economy.
Contrary to what might be expected, higher female tertiary school enrolment reduces the likelihood of adopting gender provisions. Lower relative female education levels may drive mobilisation for greater attention to women’s rights and equality among various stakeholders to peace agreements including women themselves. At the same time, a greater proportion of highly skilled, educated women may be associated with professionalisation rather than mass mobilisation for social and political change during peace processes.
In terms of GNI per capita, or the wealth of a country, the effect on the adoption of gender provisions is positive but close to zero. Seemingly counterintuitively, an increase in the women’s political empowerment index makes it less likely that a peace agreement will contain gender provisions. This may be because of the civil liberties variables on access to property rights, which in conflict-affected countries are typically skewed towards men. An increment in the general government final consumption expenditure as percentage of GDP also reduces the likelihood of securing peace agreements with gender provisions. This negative effect on gender provisions can be explained by factors such as inefficiency in the administration of public budgets, and the corruption that affects conflict-affected or transitional governments.
Of significance, however, an increase in just 1 per cent in military expenditure as a percentage of GDP makes it less likely that peace agreements will have gender provisions. This is an interesting finding, which has implications for the operational focus of WPS in conflict-affected areas by military actors. Defence forces are now among the most active implementers of WPS NAPs. This finding, however, supports the concerns of scholars and advocates that the military focus of the agenda may undermine its core purpose to promote women’s peace and security participation (Shepherd, 2016; WILPF, 2015). Domingo et al. (2015: 24) found that one of the major obstacles to women’s participation in peace processes is the militarised nature of the peace process, with armed parties dominating negotiations. Moreover, the resources used to build up the military may lead to reductions in other expenditures, such as, in social and economic areas directly or indirectly supporting women’s participation in peace processes or in civil society, both of which are strongly associated with gender-sensitive peace agreements. We further explored this result on military expenditure by analysing the discrete change in probabilities with a given change in this independent variable, holding all other variables at their mean. Our analysis showed that when the military expenditure variable goes from its minimum value (0.07%) to its maximum (32.66%) in our dataset, the probability of an agreement including gender provisions decreases from 0.72 to 0.00. These are powerful findings supporting the importance of women’s participation and presence in delivering gender-sensitive peace agreements and the negative effects of military spending.
With respect to the effects of better sanitation facilities on the population, this has a negative effect on the probability that peace agreements will be signed with gender provisions. An increase in the proportion of a country’s population with access to sanitation, reduces the probability of agreements with gender provisions. This result can be explained by Elinor Ostrom’s (reissued 2015) collective action and public goods theory. Ostrom argues that people are more likely to engage in collective action when they need to work together to provide key public goods and services, often when the state is fragile or not present. In the context of conflict-affected societies where women have had to sustain households and communities by providing public services themselves, they are primed for collective action. Women as a group, thus, have a strong interest in advocating for gender provisions in any peace settlement where their access to sanitation, water and other services and resources is poor. The implication is that international actors and development agencies should support not only women’s basic needs but their strategic interests in advancing peace to be able to themselves address these needs. The case of women peacebuilders and their role in the negotiation of the Bougainville (Papua New Guinea) peace agreement in 2001 illustrates this point (see George, 2016).
We further analysed the probability that a peace agreement will contain gender provisions by changing the value of some of our most statistically significant variables to reveal their full effects. Looking across our dataset, the baseline expectation of a peace agreement including gender provisions is close to 50 per cent. What is theoretically interesting to us is how that probability changes with changes in the variables capturing women’s participation and representation. To begin with, taking all other variables at their mean values in our dataset, we examined the effect of having no women’s participation in an elite peace process (the lowest value in our dataset), the minimum proportion of women representatives in our dataset, and the lowest level of women’s civil society participation in our dataset. With this combination, we found that there is zero chance that a peace agreement with gender provisions will be adopted. By contrast, taking all other variables at their mean values in our dataset, we examined the effect of having the maximum levels of women’s participation in an elite peace process, their parliamentary representation and civil society participation in our dataset. With this combination, we found that the probability of adopting a peace agreement with gender provisions is 100 per cent.
Implications for country-level gender-inclusive peace processes
To closely understand the results of our study at the country level we analysed two cases in more detail, Iraq and Colombia. In Iraq, we can observe the possible impact of military expenditure on the likelihood of gender-sensitive peace agreements. We analysed the three peace agreements for Iraq in our dataset: 2004, 2005 and 2010. Women participated only in the 2005 peace process. Women’s parliamentary representation increased from 2004 to 2005 by 0.71 per cent and from 2005 to 2010 by a further 6.25 per cent. Women’s civil society participation shows reduced levels between 2005 and 2010, and the fertility rate increases from 2004 to 2010. Finally, overall government expenditure was reduced during the period under study while military expenditure as a percentage of GDP increased over the time. We find that while women’s parliamentary representation increases the likelihood of a gender-sensitive peace agreement, by itself it is not a guarantee for achieving this outcome. Effects such as low women’s civil society participation, increased military spending, increases in the rate of fertility, health and water, taken together reduce the likelihood of adopting gender equality provisions in post-war Iraq.
The case of Colombia supports our findings on the positive impact of women’s participation on the chances of having a peace agreement with gender provisions. The 2016 Colombian Final Peace Agreement contains over 100 strong gender provisions recognising the unique experiences of women during the conflict and placing a clear emphasis on women’s participation in all areas of post-conflict recovery including land rights, use and rural governance. A common element in this case reveals a pattern we observed in the other cases: when increases in women’s civil society participation are accompanied by increases in the women’s parliamentary representation and women’s participation in negotiation processes, the likelihood of a gender-sensitive peace agreement is higher year upon year. Based on qualitative research, we know that the gender-based approach of Colombia’s 2016 accord is an outcome of women’s civil society mobilisation during the peace process through two National Summits and that this mobilisation contributed to the strong representation of women inside the peace process, including rural, indigenous, Afro-descendent women in the gender sub-commission designed to mainstream gender throughout the peace process (Salvesen and Nylander, 2017). The results in the state governance indicators (negative effects of increases in military expenditure and of reductions in government expenditure) are also supported in the case of Colombia.
Conclusion
UNSCR 1325 has engendered heightened global awareness of the importance of women’s participation in peace and security. The resolution was in part a response to the lack of attention to the negative impacts of conflict on women and girls in interstate and civil wars before 2000, which was manifest in entirely gender-blind peace accords (see, for example, Mlinarević et al., 2015). This article finds that peace agreements since 2000 are more likely to be gender-sensitive when women participate at all levels in peace processes. Gender provisions to ensure women’s participation after conflict and to protect and uphold women’s rights are far more assured if women are able to participate in elite peace processes, if they are represented in national parliaments, and if women mobilise together in civil society – which may be more frequently the case when they lack access to public goods such as sanitation and/or tertiary education.
Strikingly, increases in military expenditures undermine the chances that a gender-sensitive peace agreement will be adopted. This is also a salient finding for the WPS agenda as many states adopt military approaches to securing women’s rights through interventions by security forces. Overall, however, the interaction and engagement among civil society, states and international organisations is a key dynamic supporting the efforts to integrate gender perspectives into peace agreements and international security policymaking.
Overall, our findings suggest that the democratisation of a peace process, defined as including both the participation and the representation of women in formal peace processes and politics and in civil society mobilisation, matters most in ensuring more gender-sensitive peace agreements. While our findings in this study relate to the adoption of peace agreements with gender provisions, they are potentially relevant to peace implementation as well. In subsequent research, we will consider the relative strength of gender provisions in terms of their mechanisms to institutionalise and monitor women’s rights and gender equality. Further research on the ways in which women are participating in implementation processes and decision-making will also be important to build on the insights of this study. We might expect the same factors analysed here to influence post-conflict processes of implementing gender provisions. The inclusion of women representatives in government modalities for implementing peace agreements may be necessary, if not sufficient, conditions for successful implementation of gender provisions; while women’s civil society participation to monitor and hold states accountable for this implementation and women’s representation in post-conflict politics are likely to be crucial factors in supporting gender-responsive post-conflict implementation. Above all, the promotion of an open, transparent and participatory implementation process – as well as gender-inclusive peace agreements – are vital pathways to gender-equal, sustainable peace.
Supplemental Material
IPS808608_supplemetary_note – Supplemental material for Towards inclusive peace: Analysing gender-sensitive peace agreements 2000–2016
Supplemental material, IPS808608_supplemetary_note for Towards inclusive peace: Analysing gender-sensitive peace agreements 2000–2016 by Jacqui True and Yolanda Riveros-Morales in International Political Science Review
Footnotes
Acknowledgements
We wish to thank Sara E Davies, Nicole George, Amy Haddad, Katrina Lee Koo, Sarah Hewitt, Yasmin Chilmeran, Elliot Dolan-Evans and Alexandra Phelan for their critical contributions to the development of this article as presented at several workshops under the joint project ‘Towards Inclusive Peace: Mapping the Gender in Peace Processes and Implementation’ co-funded by the Australian Research Council and the Australian Department of Foreign Affairs and Trade.
Funding
This work was supported by the Australian Research Council Linkage Scheme and its partner organisation, the Australian Department of Foreign Affairs and Trade (grant number LP160100085).
Notes
Author biographies
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
