Literature and method: Identifying the essence of peace processes

In recent decades, there has been an intense discussion of the effectiveness of intrastate conflict resolution by means of soft power versus brute force. Although mass media and political discourses tend to praise peace agreements achieved through dialogue, numerous studies have found that these agreements are frequently unstable and likely to result in the recurrence of conflict (^{DeRouen, Lea, & Wallensteen, 2009}; ^{Lounsbery & DeRouen, 2016}; ^{Martinez & Diaz, 2005}; ^{Paris, 2004}; ^{Walter, 2004}). These findings received increased attention from policymakers and negotiators, compelling them to use innovative formulas in new peace processes and driving them to include appropriate provisions into peace agreements to encourage successful conflict resolution and, subsequently, achieve effective peacebuilding³.

Academic literature on conflict analysis and resolution has also increased considerably; it closely examines peace agreements, their dynamics, and provisions included to characterize their contents, address grievances and prevent recurrences of violence (^{Álvarez Calderón & Rodriguez Beltrán, 2018}; ^{Cabrera Cabrera & Corcione Nieto, 2018}; ^{Fernandez-Osorio, 2017}; ^{Gonzalez Martinez, Quintero Cordero, & Ripoll De Castro, 2018}). For instance, ^{Bruch, Muffett, and Nichols (2016)} discuss grievances originating from natural resources management; ^{Caspersen (2017)} explores grievances derived from inter-ethnic/state relations and security sector restructuring (a military and police reform); and ^{Castro-Nunez et al. (2017)} examine economic and social development-related grievances and their connection to land seizures.

One of the most widely used methodologies to characterize peace agreements is identifying its provisions. Provisions are understood here as "a goal-oriented reform or stipulation that (...) falls under a relatively discrete policy domain (e.g., executive branch reform, demobilization, minority rights, police reform)" based on ^{Joshi and Quinn (2015}, p. 15).

A significant body of literature, including works by ^{Hartzell, Hoddie, and Rothchild (2001)}; ^{Mattes and Savun (2010)}; ^{Pettersson and Lotta (2012)}; ^{Paffenholz (2014)}; ^{Ansorg, Haass, and Strasheim (2016)}; ^{Backer, Bhavnani, and Huth (2016)}; ^{Kovacs and Hatz (2016)}; ^{DeRouen and Chowdhury (2016)}; and ^{Jeffery (2017)} have examined the essence of peace agreements using their provisions as common themes to establish priorities that should be resolved between parties. However, the disparity in the grouping of provisions has hindered the compilation of these assessments. The UCDP, for instance, organizes these provisions into four groups: military, political, territorial, and judicial (^{Högbladh, 2012}, p. 44), disregarding some important provisions, such as a military reform, reparations, and human rights, which could limit more indepth categorizations. Meanwhile, the PAM organizes them into five groups: institutions, security, rights, external arrangements, and miscellaneous (^{Joshi & Quinn, 2015}, pp. 15-16).

To date, none of the published research on the 2016 CPAF has used either the PAM or UCDP' methodology to categorize its contents. However, in August 2016, the Colombian government and the FARC decided to assign the Kroc Institute for International Peace Studies at the University of Notre Dame with the "principal responsibility for technical verification and monitoring of the implementation of the accord through the Peace Accords Matrix (PAM) Barometer initiative" (^{LaReau, 2016}). With this in mind, we established that the best methodology for this research is the PAM approach to provide some innovative and useful input for present or future studies on the 2016 CPAF that will contribute to the assessment of the agreement implementation.

This article uses the set of peace agreement provisions and the definitions provided by the PAM, which has characterized the peace accords of 31 countries between 1989 and 2012 (Table 1).

However, it reorganizes them into a new model, which is explained in Table 2. Provisions are organized into five groups: socioeconomic, political, security and defense, justice, and implementation and verification. These are the provisions that best match the six major topics of the official agenda discussed by the Colombian government and the ^{FARC (2016)} during the 2012-2016 peace process, which include: (i) a comprehensive rural reform, (ii) political participation, (iii) ending the conflict, (iv) resolving the illicit drugs issue, (v) victims of the conflict, and (vi) the agreement implementation, verification, and endorsement. A comprehensive rural reform and resolving the drug issues were merged into the first group (socioeconomic provisions) as both topics share most of the provisions established.

Characterization of the 2016 comprehensive peace agreement with the FARC

Negotiations towards the 2016 CPAF officially began in September 2012, with a defined six-point framework, and lasted until a final agreement was announced on August 24, 2016. While the peace process had many supporters in Congress, in major social movements, and international organizations, disapproval among the population grew significantly, resulting from the lack of clear information on the peace talks in Havana, Cuba, and the agreement's partial accords. This uncertainty generated trepidation about the future of the country and certain socioeconomic, political, security, and justice aspects. The ratification referendum of the final agreement in October 2016 achieved the highest degree of uncertainty, resulting in 50.2 % of voters rejecting the accord (^{BBC, 2016}). This outcome made it clear to the Colombian government, the FARC and the international community that the majority of the population had misunderstood the whole process and the established provisions.

Despite the strong international support, including the awarding of the 2016 Nobel Peace Prize to former Colombian President Juan Manuel Santos Calderon, the referendum results forced a renegotiation of the accord, which was finally ratified by the Colombian Congress on November 30, 2016. This reviewed version of the final accord added 13 pages to the original 297-page document. These pages included several explanations regarding the definitions and procedures used to reach the established provisions; it also included some modifications thereto required by the opposition and social movements. Some of these key changes contained the protection of private property and legal landowners; the establishment of a statute for political opposition; and the inclusion of guarantees for peaceful protests. The new pages also included regulations regarding the assignment of public funding to FARO'S political party; the inclusion of FARC militias into the DDR process; stipulations obliging the FARC to end its illegal economic activities and provide accurate information on its finances and assets; the funds deemed illegal to be used for victim reparations and address any damages caused; some procedures to guarantee the security of FARC leaders and combatants; guidelines regarding the substitution of illicit crops, including aerial aspersion; actions to ensure the political participation of FARC'S representatives; the creation of dedicated media and radio stations for FARC'S political party; and finally, guidelines and procedures for a transitional justice system (Fundacion Ideas para la Paz, 2017b; High Commissioner for Peace, 2016e; ^{Lopez, 2016}).

The 2012-2016 peace process with the FARC yielded an exploratory agreement called The General Agreement for Ending the Conflict and Building Stable and Enduring Peace, five joint reports, 13 drafts of supporting texts, 107 joint declarations, and a final agreement called The Final Agreement to End the Conflict and Build Stable and Durable Peace (the 2016 CPAF), a total of 136 documents. After the final agreement, the exploratory document was perhaps the most important one. It established the agenda and the topics to be discussed, as well as the general guidelines for the peace process. Based on the descriptions by the pam, 38 provisions can be identified to assess the 2016 CPAF final agreement (Table 2) from the five-group model of provisions formulated by this article. These provisions can be organized as follows:

Socioeconomic provisions. This first group, consisting of twelve provisions, establishes the way that the 2016 CPAF deals with socioeconomic grievances and describes its strategy in solving the first and fourth topics of the peace process agenda (i.e., a comprehensive rural reform and resolving the illicit drug issue). These provisions are children's rights, cultural protection, economic and social development, an education reform, human rights, indigenous and minority rights, a media reform, natural resource management, rights of self-determination, women's rights, and inter-ethnic and state relations.

Figure 1 shows that Colombia has the most socioeconomic provisions in its peace agreement, namely, twelve occurrences, while some countries such as Guinea-Bissau and Côte d'lvoire have none. In comparison to other peace agreements, this number is well above the median (3) for this group (Table 3).

To address the required comprehensive rural reform, this group of provisions covers historical problems to be resolved at a national level but with a territorial approach that includes access to and use of land; formalization of rural property for untitled proprietors; improvement of the rural cadaster; protection of special environmental interest areas; rural jurisdiction to resolve conflicts; infrastructure; adequate housing; access to drinking water and sanitation, social security; education; health, food and nutrition; and a solidary and cooperative rural economy (^{Común Acuerdo, 2017}; High Commissioner for Peace, 2016d). Conversely, these provisions related to the illicit drug issue, that is, the eradication and substitution of illicit crops, the prevention and treatment of drug use with a public health approach, and the fight against drug trafficking and organized crime (High Commissioner for Peace, 2016f).

Of the 31 peace agreements analyzed, 94 % include at least one provision from this group, economic and social development (69 %) and human rights (63 %), these are the most employed provisions; the right to self-determination is the least used provision (16 %) (Figure 2). Notably, the provisions less employed internationally were children's rights (19 %), indigenous minority rights (19 %), minority rights (19 %), and rights of self-determination (16 %); all of which are explicitly addressed by the 2016 CPAF. Moreover, the provision on women's rights (25 %) was expanded to cover other gender identities as well, incorporating the rights of the lesbian, gay, bisexual, and transgender community (^{Colombian Government & FARC, 2016}).

The 2016 CPAF'S objective is to boost the rural reform and to reduce rural poverty by 50 % within ten years, through the creation of a land bank to assign rural parcels to farmers, victims, and minorities. Means of production and infrastructure are to be procured through popular participation in regional decisions addressing uneven land distribution to develop local and familiar economies, narrow the rural-urban gap, recover the zones affected by the armed conflict, and protect regional multiculturalism (High Commissioner for Peace, 2016d). The implementation of this group of provisions is possibly one of the most critical challenges of this agreement, as even the most optimistic economic assessments forecast a tough economic future for Colombia (Figure 3).

Political provisions. The twelve provisions of this second group involve how the 2016 CPAF addresses equal political participation and administrative issues that have caused or aggravated the armed conflict. It also provides solutions to the second topic of the peace process agenda, that is, political participation. Such provisions consist of boundary demarcation and citizenship, civil administration, a constitutional reform, decentralization/federalism, an electoral/political party reform, and executive branch reforms, as well as an independence referendum, a legislative branch reform, official language and symbols, power sharing during the transitional government, and territorial power sharing.

Figure 4 shows that the 2016 CPAF comprised only four provisions related to this group (a civil administration reform, a constitutional reform, an electoral/political party reform, and a legislative branch reform); this number is the median (4) for this group (Table 4). However, it is low compared to Sudan (ten provisions), Bosnia Herzegovina, the Philippines, and South Africa (eight provisions each).

To resolve grievances related to this group of provisions, the 2016 CPAF encourages forming new political parties, strengthening transparency in electoral processes, increasing the population's democratic involvement, promoting the integration of regions affected by the armed conflict into politics, guaranteeing the political representation of minorities, social movements, victims and stigmatized regions, and establish security protocols aiming to protect FARC members (High Commissioner for Peace, 2016c; Común Acuerdo, 2017).

This agreement was innovative in terms of the measures used to tackle some of the most common challenges in the implementation of peace accords, such as corruption and lack of transparency, violent detractors, and the absence of appropriate conditions for political opposition (^{Colombian Government & FARC, 2016}). Corruption affects the adequate use of resources and precludes the accountability of public organizations and officials, especially when dealing with a post-accord scenario, in which funding is often scarce. To foster transparency, the 2016 CPAF aimed to secure the completion of projects adopted by creating a national council for reconciliation and coexistence and a comprehensive security system for the exercise of politics. The agreement protects opposition leaders, social movements and human rights activists from violent detractors, while promoting a democratic environment for political participation and discussion.

It is possible that other provisions were not included in Colombia's peace agreement because of its reliance on the exploratory document in which strict limits were imposed as conditions for the peace talks. For example, the issues of an executive branch reform, a military/police reform, decentralization/federalism, a power sharing transitional government, and territorial power sharing were excluded, as the government's intention was not to modify the state's structure or administration. Provisions involving areas such as boundary demarcation, a citizenship reform, an independence referendum, and official language and symbols were also excluded in the exploratory document and external to discussion by the parties (^{Colombian Government & FARC, 2012}). In general, 97 % of the peace agreements assessed included at least one provision from this group. An electoral/political party reform (75 %) and a constitutional reform (63 %) were the provisions most used, while territorial power sharing (9 %) was the least used (Figure 5).

Defense and security provisions. The nine provisions in this third group show how the 2016 CPAF addresses defense and security challenges for the post-accord scenario and provide options for the third topic of the peace process agenda (end of armed conflict). These provisions are arms embargo, ceasefire, demobilization, disarmament, a military reform, paramilitary groups, police reform, reintegration, and withdrawal of troops.

The 2016 CPAF involves only six security and defense provisions (ceasefire, demobilization, disarmament, paramilitary groups, reintegration, and withdrawal of troops). In comparison to other peace agreements, this number is the median for this group (6) (Table 5).

According to Figure 6, Sudan had nine occurrences, while Angola, Burundi, Cambodia, and Mozambique had eight, being the countries that included the most provisions of this type. Conversely, Croatia and Guinea-Bissau had the fewest provisions (two occurrences).

The implementation of this group is perhaps the most successful. As of July 2017, after a phase of ceasefire and a bilateral end to hostilities, a structure for the monitoring and verification of the DDR mechanism (formed by the United Nations, the Colombian Government and FARC) was established. Approximately 7,000 FARC members were demobilized and distributed among 19 zones (El Tiempo, 2017b). The United Nations mission in Colombia oversaw and verified FARC'S disarmament. FARC'S decommissioned armament, estimated to include 7,132 large and small guns, was stored in containers under the UN'S control (Semana, 2017a). Humanitarian demining programs led by the National Army of Colombia have also been successful (Accion contra minas, 2016), 660 weapon caches in remote areas were collected and destroyed by the un and the National Army of Colombia (United Nations, 2017). The National Army of Colombia and the National Police have provided security to FARC'S demobilized personnel. A new strategy has even been implemented to manage FARC dissidents who chose to ignore the peace agreement and characterized them as armed groups in order to guarantee social, economic, and political reintegration of FARC members (High Commissioner for Peace, 2016a).

Figure 7 shows that all the 31 peace agreements analyzed included at least one security and defense provision. The most used provisions were those related to the DDR process: demobilization (84 %), disarmament (81 %), and reintegration (78 %). The least used was arms embargo (9 %). Two of the key provisions, a military reform (72 %) and a police reform (69 %), were not included in the 2016 CPAF, as they had been deemed prohibited issues by the Colombian government in the exploratory document (^{Colombian Government & FARC, 2012}). Similarly, the arms embargo provision was not discussed by the parties based on FARC'S commitment to voluntarily relinquish its weaponry and refrain from using violent means.

Justice provisions. This fourth group comprises eight provisions and defines how the 2016 CPAF tackles equality and impartiality while addressing the fifth topic of the peace process agenda (victims of the conflict). These provisions are an amnesty, a commission to addressing damage/loss, internally displaced persons, a judiciary reform, prisoner release, refugees, reparations, and a truth or reconciliation mechanism.

The 2016 CPAF agreement includes all of these justice provisions (eight occurrences), a number that is well above the median (4) for this group (Table 6), in comparison to the other 31 peace agreements analyzed, such as East Timor and Philippines, which include none (Figure 8).

The 2016 CPAF incorporates a novel scheme to solve justice grievances and victims' rights, through a transitional justice system that is built upon a comprehensive structure of truth, justice, reparation and non-repetition; the creation of a commission for the clarification of truth, coexistence and non-repetition; the formation of a unit for the search of missing persons; the development of comprehensive reparation measures for peacebuilding; a special jurisdiction for peace; non-repetition guarantees; and giving historical memory a central role within the reconciliation process (High Commissioner for Peace, 2016g).

This group of provisions is probably one of the most controversial in Colombia, as the conflict and its complex origins have produced numerous victims, countless damages and a number of human rights violations (Pizarro & Moncayo, 2015). Striking an appropriate balance between justice and peace is the task for the transitional justice system approved by the parties, and endorsed by international political leaders. However, it is still under review of the International Criminal Court and human rights organizations (International Criminal Court, 2016).

Figure 9 depicts how 94 % of the agreements included at least one justice provision. The most used were internally displaced persons (72 %) and refugees (69 %), while the ones used the least involved a commission to addressing damage/loss (6 %) and reparations (25 %).

Implementation and verification provisions. This fifth group, consisting of ten provisions, explains how the 2016 CPAF addresses the sixth topic of the agenda (implementation, verification, and endorsement). These provisions are a detailed implementation timeline, a dispute resolution committee, donor support, international arbitration, a ratification mechanism, regional peacekeeping force, review of agreement, a UN peacekeeping force, a UN transitional authority, and a verification/monitoring mechanism.

The 2016 CPAF includes eight implementation and verification provisions (detailed implementation timeline, a dispute resolution committee, donor support, international arbitration, a ratification mechanism, review of agreement, a UN transitional authority, and a verification/monitoring mechanism), a number well above the median of this group (4) in comparison to the other 31 peace agreements analyzed (Table 7).

This agreement has developed a unique approach to ensure compliance with peace initiatives and strategies through the continous verification of the implementation status to identify delays and adjustments needed. To this end, a follow-up and verification commission for the final peace agreement, as well as a mechanism to verify the agreement, were created (High Commissioner for Peace, 2016b). Similarly, the exploratory agreement of the 2016 CPAF promoted considerably constant assistance during the peace talks, both internationally with Cuba and Norway as guarantors and nationally with Chile and Venezuela. Also included were a citizen participation mechanism, a rapporteur system, and a promulgation mechanism. The agreement created the possibility for multilateral participation by academia, social movements, and citizens, expanding the scope of the topics discussed in Havana. The rapporteur system is led by the Universidad Nacional, and the promulgation mechanism, via a dedicated web page, as well as pedagogical campaigns that fostered transparency and enabled constant and necessary feedback.

The 2016 CPAF does not contain the following two provisions: regional peacekeeping force and un peacekeeping force as the parties agreed not to employ foreign troops in its implementation.

Colombia and Burundi, with eight occurrences, are the countries that have the most provisions of this type, in contrast to Djibouti, India, and Senegal that had no occurrences (Figure 10). Figure 11 depicts that 91 % of the agreements included at least one implementation and verification provision. The most utilized is the verification mechanism (78 %) and the detailed implementation timeline (69 %), while the least employed is international arbitration (6 %) and un transitional authority (13 %).

Extension as another source of criticism

Another source of criticism for the 2016 CPAF is its extension and complexity, which promote conflicting and sometimes irreconcilable interpretations of its essence and implementation among the government, the opposition, the FARC, social movements, and international organizations. These misunderstandings involve many critical provisions, such as an amnesty (^{Sanchez, 2016}), an electoral/political party reform (El Pais, 2016), human rights (International Committee of the Red Cross, 2017), a judiciary reform (The New York Times, 2016), transitional justice (^{Brodzinsky, 2016}), reintegration (^{Aidi, 2016}), a detailed implementation timeline (The Economist, 2017), economic and social development (^{Vivanco, 2016b}), demobilization (^{Gomerzano, 2016}), an education reform and children's rights (^{Oppenheimer, 2016}), and the truth or reconciliation mechanism (Vivanco, 2016a).

Published studies on peace agreements have previously reported skepticism and grievances resulting from the lack of understanding of its contents. For instance, ^{Spears (2000)} demonstrated the instability of peace agreements in Africa due to persisting unfamiliarity with power-sharing provisions. Stanley and Holiday described in ^{Stedman, Rothchild, and Cousens (2002}, p. 442) the complications of the un mission in Guatemala (MINU-GUA) because of "a lack of understanding of most Guatemalans regarding the mission's mandate and functions." ^{Mac Ginty, Muldoon, and Ferguson (2007)} explained how political and psychological stubbornness destabilized the post-peace accord environment in Northern Ireland. As a result, mass media in Colombia has been instrumental in creating an adverse perception on the part of the accord's main critics, political opponents, and independent groups using the idea that the extensive number of pages of the 2016 CPAF has negatively impacted the reliability of the topics established, and that the high number of topics adversely affected the agreement implementation (CNN, 2016; El Tiempo, 2017c; ^{Negrete, 2017}; RCN, 2016a; ^{Rendon, 2017}; Semana, 2016).

Given such criticism, could there be a possible correlation between the extent of a peace accord and the number of provisions agreed upon, or between the number of pages or provisions established and the implementation level of the agreement? If such a hypothesis is accurate, in comparison to other countries, Colombia's 2016 CPAF and its implementation may have an uncertain future with agreements covering only 38 provisions within a 310-page document.

Regarding the first part of the question, although some peace agreements have a high number of established provisions vis-a-vis the extent (in pages) of the final accord, such as Indonesia with 25 provisions on seven pages, and India with 17 provisions on six pages, there are also peace agreements such as Mali's, with 16 provisions on 15 pages, and Sierra Leone's with 25 provisions on 28 pages. By running a regression of pam's data, it was possible to determine that there is a significant potential correlation between the extent (in pages) of a peace agreement and the number of provisions agreed upon, according to Pearson's r (30) = .73, p < .001 (Table 8). This suggests that the more pages a peace agreement has, the more provisions it will have.

As to the second part of the question, there are also some short peace agreements with a high percentage of implementation level. Guinea-Bissau, for example, has a 96 % implementation level, ten years after the signing of a three-page agreement. Croatia has a 73 % implementation level of a five-page agreement. There are also multifaceted peace agreements, such as El Salvador's, with a 96 % of implementation level ten years after signing a 106-page agreement, and Bosnia and Herzegovina, with a 93 % implementation level of a 149-page agreement. The regression results of PAM'S data did not show any significant correlation between the extent (in pages) of a peace agreement and its implementation level, according to Pearson's r (29) = .20, p > .001 (Table 9). The results suggest that the implementation level of a peace agreement is not directly influenced by its extension; in other words, the success or failure in the implementation of a peace agreement does not depend on the complexity of the vetted documents of a peace process.

Similarly, whereas there are some peace agreements with few agreed provisions and a high percentage of implementation level, such as Djibouti's with a 89 % implementation a ten years after the signing of a nine-provision agreement, and East Timor's with a 94 % implementation level of an 11-provision agreement, there are also intricate peace agreements, such as Sudan's, with a 73 % implementation level ten years after the signing of a 41-provision agreement, and Burundi's with a 78 % implementation level of a 34-provision agreement. The regression results of pam's data did not show a significant correlation between the number of agreed provisions and its implementation level, Pearson's r (29) = .17,p > .001 (Table 10). Therefore, it is possible to confirm that the implementation level of a peace agreement is not directly influenced by its complexity, in terms of the number of provisions agreed upon.