Ceasefire Rationales: A Comparative Study of Ceasefires in the Moro and Communist Conflicts in the Philippines

ABSTRACT Ceasefires are part of most contemporary peace processes, however empirical insight suggests that the characteristics of ceasefires vary greatly across conflict settings. This paper contributes to filling a research and knowledge gap about how different types of ceasefire come about through a comparative case study of ceasefires in the Moro and communist insurgencies in the Philippines. I argue that to understand differences in the characteristics of the ceasefires in these conflicts, it is important to consider the aims, ideologies and strategies of the conflicting parties and how this shapes their approach to a ceasefire. Following this, I suggest that ceasefires must be analyzed and understood with sensitivity to conflict issues and approaches to violence, as this contributes to an explanation of how ceasefire is used as a political tool and how it shapes the dynamics of conflict. Based on interviews and document studies, the article maps the characteristics of ceasefires in both cases over time and analyses the approaches to ceasefires of the parties to the conflict. Knowledge of how ceasefires come about and what shapes them can help both academia and policymakers draw more informed and accurate conclusions about their outcomes and effects.


Introduction
This paper seeks to explore and analyze how and why ceasefires are a salient part of some peace efforts while not of others. Ceasefires are part of most contemporary peace processes but empirical insight suggests that their characteristics vary greatly across conflict settings. Two cases that illustrate how fundamental these differences can be are the two main armed conflicts in the Philippines: the conflict between the Philippine government and Moro groups in the southernmost Mindanao region; and that between the government and the communist party, which is nationwide. Both conflicts are protracted intrastate conflicts that have experienced prolonged efforts at peace negotiations, however the character and salience of ceasefires as part of these two separate efforts varies greatly. In the Moro conflict in Mindanao, the government and the Moro Islamic Liberation Front (MILF) have been engaged in a ceasefire for most of a 20 year period. The ceasefire is a bilateral formal engagement with an extensive institutional arrangement and has been characterized by high degrees of non-violent interaction and cooperation. On the other hand, in the conflict between the government and the communist insurgency of the Communist Party of the Philippines (CPP) -New People's Army (NPA) -New Democratic Front of the Philippines (NDFP), ceasefires have not been a salient feature of engagement. Those that have occurred have been predominantly unilateral, temporary and short-lived. How can this variation in the characteristics of ceasefires be understood?
For a long time, ceasefires were seldom given theoretical and analytical leverage in the peace and conflict literature. 1 They were generally mentioned in passing or viewed merely as a means to an end. A growing body of literature is devoted to in-depth analysis of ceasefires. Among these, some studies have focused on explaining ceasefire duration, i.e. why some ceasefires last for a longer period of time while others break down, highlighting amongst other things, ceasefire design and the role of external actors as important factors in explaining ceasefire duration. 2 More in-depth case studies have analyzed the effects and consequences of ceasefires, including on shaping the dynamics and structures of conflicts and peace attempts 3 and on influencing local social orders and governance regimes. 4 Taken together, these studies highlight the multifaceted nature of ceasefires and their varying impacts on the dynamics of conflict, including their intended or unintended negative implications on peace and conflict trajectories. However, while the outcome, dynamics and impact of ceasefires have attracted more scholarly attention, there is a lack of systematic research and knowledge on how different types of ceasefire come about.
This paper aims to contribute to filling the research and knowledge gap through a comparative case study of ceasefires in the Moro and communist insurgencies in the Philippines. I argue that to understand differences in the characteristics of the ceasefires in these conflicts, it is important to consider 1 Åkebo, Ceasefire Agreements and Peace Processes; Clayton et al., "Ceasefires in Intra-state Peace Processes"; Kolås, "Naga Militancy and Violent Politics"; Höglund, Violence in the Midst of Peace Negotiations. 2 E.g. Fortna, Peace Time; Karakus and Svensson, "Between the Bombs". 3 E.g. Kolås, "Naga Militancy and Violent Politics"; Åkebo, Ceasefire Agreements and Peace Processes. 4 E.g. Sosnowski, "Violence and Order"; Woods, "Ceasefire Capitalism". the aims, ideologies and strategies of the conflicting parties and how this shapes their approach to a ceasefire. Following this, I suggest that ceasefires must be analyzed and understood with sensitivity to conflict issues and approaches to violence, as this contributes to an explanation of how ceasefire is used as a political tool and how it shapes the dynamics of conflict. The analysis is grounded in a historical and contextual understanding of ceasefires in these two conflict settings. The rationale of the case comparison is not that the Mindanao conflict represents a 'successful' case of ceasefire that the communist conflict should replicate, rather the point is to highlight how differences in these two conflicts shape what type of ceasefire come about. Previous research suggests that different ceasefire characteristics shape the dynamics of violence and conflict resolution. 5 Knowledge about how different types of ceasefire come about is important because it can help both academia and policymakers draw more informed and accurate conclusions about the outcomes and effects of ceasefires.
The remainder of the paper is structured as follows. First, I outline the theoretical framework used to inform and structure the empirical analysis. Secondly, there is a brief background provided on the two conflicts in the Philippines and a mapping of the occurrence and characteristics of ceasefires over time. This is followed by analysis of the differences in the characteristics of ceasefires in the two conflicts through consideration of how sensitivity to conflict issues and approaches to violence contributes to shaping ceasefires. At the end of the paper, I discuss the consequences of these insights for research and policy.

Ceasefire Diversity Across Conflict Settings
There is no widely recognized definition of a ceasefire in the literature. Conventionally, ceasefires are often treated as an outcome of armed conflict, as bargaining events or events of cessation of violence based on the number of battle-related casualties. In this article, I approach the ceasefire as a declaration of intent or agreement between parties to an intrastate armed conflict to change the patterns of interaction by ending or taking a break from the use of violence. 6 There are two key aspects to this definition. First, a ceasefire involves at least two parties and is essentially relational. Second, it is a long-term process prescribed within broader peace and conflict dynamics. This places particular importance on how the characteristics of ceasefires might be shaped by the experiences and legacies of previous ceasefire engagements between the parties. It also underlines the inherently political dimension of the ceasefire process. This broad definition enables analysis of ceasefires across empirical settings and the drawing of comparative insights based on similarities and differences between the cases for theory development purposes.
In this section I outline a theoretical framework for analyzing key characteristics of ceasefires and how variation across conflict settings can be understood. The framework will be used to facilitate the collection of empirical materials and to structure and guide the case studies and the comparative analysis.

Ceasefire Characteristics
Ceasefires can differ in many respects. To enable analysis of variations across settings of protracted violent conflict with experience of multiple peace attempts, it is important to depict the key characteristics of ceasefires in each case. Many ceasefires are bilateral undertakings by two conflicting parties but they can also be unilateral declarations by one party or multilateral undertakings involving several actors. Ceasefires differ in terms of formality. They can be verbal undertakings, as for example several of the ceasefires concluded in Myanmar, 7 or formal written accords with varying degrees of detail regarding e.g. what is defined as a prohibited violent act. While some ceasefires are explicitly temporary in nature, others are framed with an intention to continue indefinitely. Ceasefires also differ in the extent to which they are linked to a broader peace negotiation process. Some ceasefires form part of a comprehensive peace accord or include an explicit aim of reaching such accord. 8 Other ceasefires are separate undertakings. Such ceasefires may be local understandings with no direct link to a process of elite level negotiations, for example as in Syria. 9 To different degrees, ceasefires can include measures for ensuring compliance with the agreement. For example, this can take the form of joint organizational bodies for cooperation on ceasefire-related issues, such as coordination of movements of armed forces, internal or external monitoring mechanisms to oversee compliance and verification and possibly mechanisms to rule and sanction ceasefire violations. 10 Thus, ceasefires vary in the extent to which they include an organizational structure to facilitate and oversee implementation.
Some ceasefires are short-lived. They can be declared as limited in time right from the outset or may break down soon after being entered, for example as a result of political setbacks, lack of political will or incidents on the ground and ceasefire violations. 11  years or even decades. Of course, this does not necessarily mean that the ceasefire is widely respected or that the parties to the agreement move forward in efforts to find a peace settlement. Indeed, a ceasefire might freeze the conflict in a 'no-war-no-peace' situation, for example as witnessed on the Korean peninsula. While duration is often treated as an outcome variable, often with an implicit or explicit assumption that a longer duration indicates a successful ceasefire, I suggest that one of the major advantages of studying the occurrence and duration of ceasefires over time in the same conflict setting, is that such study can reveal important patterns of interaction and facilitate understanding of the role ceasefire plays in this context and why this is so. Following this, in this paper I treat the patterns of occurrence and duration of ceasefires as elements of the ceasefire characteristic for analytical purposes, because it tells us something about the very nature of ceasefires in a conflict setting. Table 1 summarizes the key aspects sought in the case-studies to delineate characteristics of ceasefires over time in each of the two conflicts. Furthermore, I am also interested in characterizing and understanding these efforts and what shapes them, rather than their outcomes and impact.

Understanding Diversity in Ceasefire Characteristics
There can be various motivations behind conflicting parties engaging in a ceasefire. It can be for tactical reasonsto find space to recruit, regroup and rearm. Non-state groups might strive to gain recognition and legitimacy as an actor from engaging in a formal agreement with a state and/or strive to gain recognition of territorial power and control. There can also be external incentives to engage in a ceasefire, for example economic incentives or other resources. Pressure exerted by the international community may prompt parties to engage in a ceasefire. Engaging in a ceasefire may also be motivated by a conviction that engaging in political dialogue is the most fruitful way forward to reach desired goals. 12 In this respect, the issue of trust and  Åkebo (2016, 46). 12 Smith, Stopping Wars. confidence in particular have been raised in the conflict resolution literature. Chounet-Cambas 13 emphasized how the establishment of a ceasefire at the commencement of a peace negotiation process can serve the purpose of easing tension between conflicting parties and signalling goodwill. It has also been suggested that due to deep-seated mistrust, actors may deem an agreement on substantive issues necessary before committing to a break or ending the use of violence, pointing to the importance of trust in the process and what it can potentially deliver. 14 The underlying motivation for initiating a ceasefire at any given moment can be difficult to depict. Motivation can also change over time, for example as a result of experiences from engaging in a ceasefire or due to contextual changes and actors may hold different motivations within an organization. However, by looking at the characteristics and approaches to ceasefires over time, I suggest that it is possible to illustrate important patterns. 15 To understand the characteristics of ceasefires, I draw on the literature on conflict transformation and armed organizations and suggests that the aims, ideologies and strategies of the conflicting parties must be considered by way of shaping conflicting parties' approaches to ceasefires and ceasefire characteristics. In both the Moro and the communist conflicts in the Philippines that are the subject of analysis in this study, the conflicting parties have at least intermittently committed themselves to peace negotiation processes. These insurgencies have been described as essentially political and ideologically-driven struggles. 16 In line with Dudouet, 17 I assume that in such conflicts, armed organizations 'have a long-standing and well-developed political vision for their country, which has at various times led to a variety of strategies (violent, non-violent or a combination of both) to implement the vision'. This suggests that rather than seeing a transition from war to peace as linear, attention should be devoted to the shifts and interplay between military and non-violent forms of interaction. To delineate the rationale of 13 Chounet-Cambas, Negotiating Ceasefires, 19. 14 Ramsbotham et al.,Contemporary Conflict Resolution, Åkebo (Ceasefire Agreements and Peace Processes) provides a holistic framework to analyze the initiation, form and content and implementation of ceasefires in relation to peace processes. The study identifies a number of key factors that can influence change in conflicting parties' attitudes, behaviours and relationships and help to analyze ceasefires in relation to peace processes, including recognition, status and legitimacy; trust and confidence; whether the parties' claims are being met; external incentives and resources; contextual changes; and intraparty dynamics. It also stresses the importance of considering e.g. the main actors' claims and positions and experiences of the conflict and earlier peace attempts as the basis for analyzing ceasefires in relation to peace processes (p. 46). This study, as well as other studies such as Milton-Edwards ('"The "Warriors Break"'), have pointed to the importance of endogenous drivers of original conflict causes as important to understanding ceasefires. Studies like these further underline the importance of more in-depth and systematic analysis of how the aims and conflict issues shape ceasefire characteristics. 16 Rodriguez and Santos, "Introduction". In particular this can be contrasted with more opportunistic, driven conflicts, see e.g. Kaplan (Resisting War). 17 Dudouet,From War to Politics,6. ceasefires along these trajectories with consideration of the parties' aims, ideologies and strategies, I suggest analysis on sensitivity to conflict issues and attachment to violence.
Sensitivity to conflict issues must be considered by way of shaping the characteristics of ceasefires. As Rodriguez and Santos 18 put it, both the Moro and communist insurgencies are 'for the most part ideologically driven, predictable, and supported by a part of the local population'. While power politics is important, this suggests that the struggles are not reduced to a question of winning a war but to the achievement of political objectives. Furthermore, in contemporary civil wars, questions of the state, location of power and identity are closely intertwined. 19 This also suggests that it is primary for a conflict party to deliver to its constituencies in order to maintain and garner support from them for its political struggle. As Werner and Yuen 20 put it, 'agreements determine who gets what and when'. If we understand conflict issues as a main driving force for warring parties, to differing degrees a ceasefire can be seen as a step towards enacting a vision of the future organization of the state and society. In this sense, a ceasefire does not merely imply a halt in the use of violence but can be used to achieve certain political ends. For example, recognition and territorial control can be sought through a ceasefire, which can speak directly to conflict issues. Having such claims met can imply a structural change in the conflict and the actors' relationships. 21 Following this, I expect that the approach to a ceasefire will be shaped by how they are deemed to be used for political purposes to achieve conflict aims. Thus, how do ceasefires play into the conflict issues? To what extent does a ceasefire include elements of or respond to conflict claims?
Protagonists' approaches to violence will also be important to consider by way of shaping ceasefires. Armed organizations are signified as being prepared for and carrying out violent acts, however attachment to violence can vary. 22 For example, the influence of armed units within larger organizations, such as the army or an armed wing of a belligerent group, can vary both between organizations and over time. This can be expected to shape approaches to ceasefires and the substance they are given. Various beliefs and ideologies can also serve to sustain violence, which can be reinforced by outside actors and influences. For example, this was the case in several wars of national liberation and revolutionary change in 1960s, which were strongly influenced by ideas of violence as a necessary component for the 18 Rodriguez and Santos, "Introduction," 2. 19 Kaufmann, "Escaping the Symbolic Politics Trap". 20 Werner and Yuen, "Making and Keeping Peace," 262. 21 Åkebo, Ceasefire Agreements and Peace Processes. 22 Kriesberg and Millar, "Protagonist Strategies"; Kaplan, Resisting War.
success of a revolution. 23 Religious belief has also been used to legitimise the use of violence. Milton-Edwards 24 rightfully cautions against assumptions that Islamist organizations would be particularly alien to ceasefires and shows how Hamas has made several attempts at a ceasefire, while other actors involved in the conflict have been unwilling to extend recognition to the organisation by engaging in a ceasefire. Thus, how does a ceasefire resonate with the struggle, given the parties' approaches to violence?
In summary, I argue that to understand the characteristics of ceasefires in the Moro and communist conflicts, it is important to consider the aims, ideologies and strategies of the conflicting parties and how this shapes their approach to ceasefires, as well as which types of ceasefire come about. In doing this, I place a particular focus on sensitivity to conflict issues and approaches to violence. I suggest that this contributes to explaining how ceasefires are used politically and how they shape conflict dynamics. Thus, what are the main characteristics of ceasefires in the two conflicts and what patterns can be discerned over time? How can we understand the overarching rationale of ceasefires, for instance how do conflict issues play into ceasefires and how does a ceasefire resonate with approaches to violence?

Methods and Materials
The study uses a comparative case study approach. 25 Application of this approach allows for an in-depth exploration of a phenomenon that extends over a longer period of time. 26 It also enables an understanding of how a phenomenon develops in different contexts and why this is so. 27 This approach is particularly suitable given the theory-developing ambitions of this study. It enables me to gain in-depth knowledge of ceasefires in each of the two conflict settings and to use insights from each case to inform the understanding of other cases. The aim is to facilitate theory development from which it is possible to propose how a phenomenon can be interpreted, rather than to test theory, by looking for evidence that indicates a general pattern. 28 The two selected cases of the Moro and communist conflicts in the Philippines are both cases of protracted intrastate armed conflict with enduring experience of attempts at peace. The cases display different patterns in terms of the characteristics of ceasefires and the conflicts differ in terms of aims, ideologies and strategies of the conflict parties. As both conflicts are embedded within the context of the Philippine state, plus the fact that the 23 Kriesberg and Millar,"Protagonist Strategies",17. 24 Milton-Edwards, "The 'Warriors Break'". 25 della Porta, "Comparative Analysis"; Stake, Multiple Case Study Analysis. 26 Ritchie and Lewis, Qualitative Research Practice. 27 Stake, Multiple Case Study Analysis. 28 Eriksson, Partition and Redemption, 28.
Philippine government is a main conflict party in both conflicts, I argue that this renders the cases ideal for comparative analysis of how conflict aims, ideologies and strategies shape ceasefire characteristics.
The study builds on written primary and secondary source documents, including agreement texts, joint statements, newspaper articles, scholarly accounts and reports from NGOs. It also builds on 22 semi-structured interviews with people who have either been directly involved in ceasefire processes in these two conflicts or have good insight into these processes, including academics and civil society actors. 29 Interviews were conducted in Manila and in Davao in Mindanao in 2016 and 2018, in addition to two Skype interviews and a telephone interview. The interviews aimed to capture different perspectives on ceasefires in these contexts, as well as to fill gaps in the historical narrative about events and developments.
The empirical analysis begins with a brief background to the two Philippine conflicts. This is followed by a mapping of the characteristics of ceasefires in each case over time, before a comparative analysis of these patterns is discussed.

Background to the Two Philippine Conflicts
From the time of its independence from Spanish and American colonialism in 1946, the Philippines has experienced a number of protracted violent conflicts rooted in processes of state formation. Between 1946Between -1986, the Philippines was under the authoritarian rule of the Marcos regime. It is in this context that the armed conflict between the government and Moro armed groups in the Mindanao region and the conflict between the government of the Philippines and the communist party emerged in the late 1960s. In 1986, a 'people's power revolution' resulted in the introduction of democracy to the country. However, while this led to greater political and civil freedom, the state has continued to be plagued by problems that are deeply embedded in societal and political structures, including high levels of poverty and inequality, as well as corruption and the concentration of land ownership that is often closely linked to seats of political power.
The background to the conflict in the Philippines southernmost region of Mindanao is commonly referred to as a history of colonial struggle involving government policies that were unfavourable to the Muslim population in Mindanao, also referred to as the Moro people, and included resettlement 29 Several authors have stressed the advantages of complementary drawing on different data collection methods and sources in case study research because they can contribute different insights to the inquiry as well as make triangulation possible (see e.g. Druckman, Doing Research; Yin, Case Study Research; Lamont and Swidler, "Methodological Pluralism"). For semi-structured interviews and advantages and limitations with the method, see e.g., Kvale and Brinkmann, InterViews; Lamont and Swidler, "Methodological Pluralism".
programmes and land reform processes, as well as political and economic marginalization. 30 The separatist claim is based on identity grounds, with reference to the history, institutions, society, cultural and religious identity of the Moros. 31 The claim concerns particular territories in Mindanao inhabited by the Moros and which they have historically ruled, along with the status and government of these areas (see Map 1 for a map of the Philippines 32 ). Moro nationalist groups started organizing themselves in opposition to the government in the late 1960s. In the beginning, the Moro National Liberation Front (MNLF) led the armed struggle for the self-determination. The Philippine government and the MNLF began engaging in peace talks in 1972 and reached a peace accord in 1996, with international backing from the Organisation of Islamic Cooperation (OIC) (formerly the Organisation of Islamic Conference). In 1997, the government began engaging with the Moro Islamic Liberation Front (MILF)a group that splintered from the MNLF in 1977 and did not accept the 1996 accord. While the MNLF emphasized the ethno-nationalist dimension of its claim, the MILF put greater emphasis on Islam. The MILF demand for a separate independent state or autonomy within a Philippine framework shifted over the years, until it eventually reached a peace agreement with the Philippine government in 2014, with final settlement on establishing the Bangsamoro autonomous region.
In the same period in which Moro groups started organizing in the south in the late 1960s, the communist insurgency emerged in the northern part of the Philippine archipelago, on the island of Luzon. The Communist Party of the Philippines (CPP) was formally established in 1968. The CPP joined forces with an active guerrilla movement in Central Luzon in 1969, from which emerged the military wing of the CPP, the New People's Army (NPA). 33 The third element of the communist rebellion is its multisectoral support organization, the National Democratic Front of the Philippines (NDFP), formed in 1973 as an umbrella for its mass organization. 34 The communist insurgency envisions an agrarian revolution in the Philippines based on the Chinese and Cuban revolutions and rooted in Mao's thinking. The CPP-NPA-NDFP emphasizes three basic problems at the root of the conflict including US imperialism, bureaucratic capitalism and feudalism. 35 It aims to establish a 'national democratic' socialist system and to nationalize the economy. 36 When the NPA began its military activities, it started off with 30 McKenna, "Fighting for the Homeland"; McKenna, "Saints, Scholars and the Idealized Past". 31 Buendia, "The GRP-MILF Peace Talks". 32 Moro groups have claimed certain areas in the western-most parts of the Mindanao region and the Sulu archipelago. 33 Mediansky, "The New People's Army". 34 Kowalewski, "Counterinsurgent Paramilitarism". 35 ICG, "The Communist Insurgency in the Philippines," 3. 36 Walch, "Collaboration or Obstruction?," 45-6; Santos, "The Communist Front," 17. only a few hundred cadres but had expanded to several thousand by the early 1970s. An initial blow to the NPA in the 1970s weakened the movement and led the Marcos regime to focus on dealing with the Moro conflict in Mindanao. One consequence described by Mediansky 37 as the communist insurgency, was largely overlooked and 'could develop practically unhampered' throughout its formative years. It changed its strategy and began to disperse and also initiate operations on islands other than Luzon. 38 It was only in the early 1980s that the Philippine government started to recognize the importance of dealing with the communist insurgency seriously, which at the time was growing in size and capacity, as well as its level of popular support. Negotiations began in 1986. 39 The communist party lost some momentum in the 1980s following an ideological split in the movement. 40 One faction signed a peace agreement with the government in 2000. 41 Nevertheless, the NPA has remained strong in some areas and financially resilient through domestic sources of funding, including revolutionary tax. Efforts to resolve the communist conflict have been compounded by various factors, including the shifts in the presidential administrations since the start of the conflict and the parallel peace process in Mindanao, which has garnered more attention from the Philippine government and from the international community. 42 In the following section, I explore and analyze how and to what extent ceasefires have been part of peace efforts in these two conflict settings.

Ceasefires in the Moro Conflict in Mindanao
In Mindanao, the government under President Fidel Ramos and the MILF signed a ceasefire agreement in 1997 as a starting point for engaging in peace negotiations. Thus, the ceasefire was initiated as a separate undertaking but linked to the negotiation process, with the stated goal of finding a settlement for the conflict. The ceasefire has lasted for about 20 years, and the ceasefire organization also remains active after a peace accord was settled in 2014. There were three periods of major ceasefire breakdown. The first two breakdowns in 2000 and 2003, were associated with military operations aimed at taking control of MILF areas. The third breakdown in 2008, was the result of a failure to settle a draft memorandum of understanding that dealt with the core issues of the conflict. However, after each breakdown the ceasefire was restored and the overarching ceasefire arrangement, including the ceasefire bodies continued. 43 38 Ibid., 2. 39 In the on-and-off negotiations between the government and the communist party, the NDF have been negotiating on behalf of the CPP-NPA, led by a panel of leaders who are in exile in Utrecht, the Netherlands. 40 Arcala Hall, "Living in the Shadow of Violence". 41 The group settling an agreement with the Philippine government is the Rebolusyonaryong Partido ng Manggagawa ng Pilipinas (Revolutionary Workers Party of the Philippines) and its Revolutionary Proletarian Army-Alex Boncayao Brigade (RPM-P/RPA-ABB). 42 ICG, "The Communist Insurgency in the Philippines". 43 Åkebo, "'Coexistence Ceasefire'".
When the government and the MILF engaged in ceasefire in the late 1990s the idea of a ceasefire was not new to the Moro conflict. The first declaration of ceasefire dates back to the 1976 Tripoli Agreement and involved the MNLF. 44 However, this agreement lasted only a few months. While it addressed central issues of contention to the conflict, many important aspects of these issues were left open for further discussions. Soon it became clear that the parties had considerable disagreements on the implementation, and Marcos started to unilaterally implement the agreement. 45 An informal ceasefire settled in 1986 between President Aquino and MNLF leader Misuari formed the basis of the 1993 Interim Ceasefire Agreement, aiming to formalize and strengthen the structure and conduct of the informal ceasefire for the duration of formal peace talks. 46 It included the creation of joint committees with external representation of OIC members to oversee compliance, as had been agreed upon earlier in the 1976 Tripoli Agreement. This ceasefire was succeeded by the 1996 Final Agreement, which included a political settlement between the government and the MNLF and the integration of MNLF fighters into the armed forces of the Philippines. 47 Against this background, a comprehensive and detailed ceasefire arrangement served as a basis when the peace negotiation process began between the government and the MILF in the 1990s. The 1997 Agreement for General Cessation of Hostilities and related documents specify the basic rules of the ceasefire, including prohibited hostile and provocative acts. 48 It also stipulates a process of identifying areas to be considered under the control of the MILF. And it includes an organizational structure led by the Coordinating Committees for the Cessation of Hostilities (CCCHs) to facilitate its implementation by overseeing compliance, coordinating observance of its provisions and solving conflicts that may arise. 49 Over the years, the CCCH bodies have enabled the parties to develop routines and have become important forums for the prevention of violent incidents. 50 The parties consider the 1997 agreement as the 'mother reference' of their ceasefire arrangement and reaffirmit in all their subsequent accords. The 44 "Tripoli Agreement," 1976. 45 Abubakar, "Review of the Mindanao Peace Process". The Jeddah Accord of 1987 similarly included a declaration of cessation of hostilities and the creation of provincial committees to observe compliance. However, this agreement saw a similar outcome as the Tripoli Agreement. 46 "Interim GRP-MNLF Ceasefire Agreement," 1993. 47 "Final Agreement," 1996. However, an armed faction led by Misuari remains active and has infrequently been associated with eruptions of violence. 48 "Agreement for a General Cessation of Hostilities" (AGCH), 1997; "Implementing Administrative Guidelines of the GRP-MILF Agreement on the General Cessation of Hostilities," 1997; "Implementing Operational Guidelines of the GRP-MILF Agreement on the General Cessation of Hostilities," 1997. 49  organizational structure of this agreement has also continued over the years, although it has been modified and expanded including in response to the major ceasefire breakdowns. This is in line with the overarching logic of the peace process, which is based on taking small incremental steps before settling the substantive political issues, as well as on the extension and improvement of previous agreements rather than overturning them. 51 For example, in the aftermath of the first major ceasefire collapse in 2002, the parties agreed to establish a body to coordinate military and police operations in or near MILF areas. This mechanismthe Ad Hoc Joint Action Group (AHJAG)was put in place a few years later in 2005. In 2003 after the second major ceasefire breakdown, the parties further enhanced the ceasefire structure by appointing an International Monitoring Team (IMT). 52 The IMT included a small number of monitors from Malaysia, Libya and Brunei deployed in 2004 and was also later completed with monitors from Japan, Indonesia and Norway. An independent civilian grassroots monitoring initiativethe Bantay Ceasefirealso emerged in 2003 outside the formal ceasefire structure to complement efforts around implementation. Lastly, in 2010 after the third ceasefire breakdown, the parties equipped the IMT with a Civilian Protection Component. Although the government and the MILF have observed a ceasefire for the best part of 20 years, it has not been without friction. The initial negotiations and setup of the ceasefire in the late 1990s reveal mistrust between the parties and internal disagreement over the ceasefire. 53 The first years of ceasefire were largely characterized by struggles over territorial control. This is evident in both the major ceasefire collapses that took place in this period, as well as in the modifications to the ceasefire structure. However, over the years of ceasefire engagement the parties have developed patterns of non-violent engagement, cooperation and coordination through the ceasefire bodies. 54 A ceasefire breach in 2015the Mamasapano incidentunderscores the fact that a ceasefire works when it is allowed to work. This incident involved a raid by the Philippine National Police against a wanted terrorist residing near MILF areas. The operation was conducted without coordination with the ceasefire structure and ended up as an encounter between Police and MILF fighters. 55 However, the fact that the ceasefire remained in place and was not jeopardized by this ceasefire breach suggests that parties have changed their way of interacting and their relationships through the ceasefire engagement. 56

Ceasefires in the Communist Conflict
In contrast to the Mindanao conflict and peace negotiations, the government and the communist party have never engaged in comprehensive ceasefire arrangements that include detailed rules of engagement and structures for cooperation and ruling on ceasefire violations. Instead, the parties have generally engaged in unilateral and temporary ceasefires, which have not necessarily been related to the ongoing formal peace talks. Ceasefires over the Christmas holidays have been a recurring feature of the conflict, for example to allow combatants from both sides to travel home and spend time with their families and gather support, by enabling people to celebrate culturally important events. 57 They have continued even in the most challenging periods of negotiations. 58 These ceasefires are always unilateral declarations bound to specific dates, thus the parameters and timeframes are very limited. Short-term ceasefires have also been declared to facilitate the release of captured soldiers and Police 59 or as a result of external factors, such as to allow aid for villages hit by natural disasters or most recently, to fight the spread of the Covid-19 pandemic. 60 The dynamics around formal peace negotiations also reveal interesting trends. In 1986, as part of the first formal peace talks under the Corazon Aquino presidency, the parties signed a 60-day national ceasefire agreement. However, the ceasefire collapsed before the parties were able to agree on the mechanisms for its implementation. The breakdown was the result of a violent incident when military and Police attacked a peaceful protest demonstration. 61 When the talks collapsed, the government identified and arrested several NPA commanders. Experience from this ceasefire has seemingly impacted subsequent ceasefire discussions. Progress at the peace talks table 55 Ferdinandh, "ARMM Officials Mark Mamasapano Tragedy Anniversary". 56 Åkebo, "'Coexistence Ceasefire'". 57 This seems to be specific for the Christmas celebration, with no similar pattern discerned for the celebrations in Ramadan or Eid al-Fitr in the Moro conflict. 58 Anonymous interview, 2018. 59 Interview with Hernani Braganza, member of the Philippine government negotiating team in the communist conflict, OPAPP (Office of the Presidential Adviser on the Peace Process), Manila, 12 November 2018. 60 The government first declared a unilateral ceasefire in relation to the Covid-19 pandemic from 19 March to 15 April 2020. The communist party did not reciprocate the government declaration. However, it heeded the appeal by UN Secretary General Antonio Guterres, for a global ceasefire between warring parties worldwide to fight the pandemic by declaring a unilateral ceasefire from 26 March to 15 April 2020 (Sison, "CPP Orders Nationwide Ceasefire"). 61 Reyes, "The Mapping of Peace Constituency Actors". This is commonly referred to as the Mendiola Massacre of 1987.
remained limited in the following years, as fighting continued between the parties during the negotiation periods and increased significantly after the collapse of each talk. Government practice has been defined as a 'limited suspension of offensive military operation' in this regard. 62 Through several agreements reached in the 1990s, including the Hague Joint Declaration of 1992, 63 the parties set an agenda for the talks consisting of four items for sequential negotiation and settlement: human rights and international humanitarian law, socio-economic reforms, political and constitutional reforms and an end to hostilities and disposition of forces. This framework relegated the reaching of a comprehensive ceasefireincluded in the fourth itemto the bottom of the agenda. 64 At the onset it was anticipated that this would be a fairly fast-moving process but instead it proved to be protracted. In 1998, the parties reached the Comprehensive Agreement on Respect for Human Rights and International Law (CARHRIL), the first item on the agenda. But the process stalled in 1999 and the negotiations were suspended.
A new attempt at formal talks in 2011 under a new presidency also failed. One of the justifications was 'the government's insistent call for a ceasefire and its reservations on the validity of previously signed agreements', in addition to issues concerning the release of NDFP consultants protected by the Joint Agreement on Safety and Immunity Guarantees (JASIG) agreement of 1995. 65 This questioned the practice of validating all previous agreements when talks resume under a new president. 66 When new attempts at formal peace talks were initiated in 2016, the communist party declared a new unilateral ceasefire and the government responded with the same measure. The parties also agreed that the ceasefire would be accompanied by amnesty. In the latest period of formal peace negotiations, the parties agreed to 'fast-track' negotiations and to commence parallel discussions on the three remaining issues on the agenda for the talks. As part of this new strategy before the talks were suspended, they also opened up the opportunity for a bilateral ceasefire. With an ambition to move beyond unilateral short-term ceasefires, the government negotiation team suggest a 'coordinated unilateral ceasefire', which is described by 62 Arcala Hall, "Living in the Shadow of Violence," 233. 63 Other important agreements signed in this period are an agreement on how to ensure safe passage and safety for people involved in the negotiations (the Joint Agreement on Safety and Immunity Guarantees, JASIG, 1995) and an agreement addressing socio-economic issues (the Joint Agreement in Support of Socioeconomic Projects of Private Development Organizations and Institutes, 1998). 64 For instance, this rationale is echoed in a joint agreement settled in 1995, which states that comprehensive agreements on the first three items 'shall prepare the ground for the comprehensive agreement on the end of hostilities and disposition of forces' (Joint Agreement on the Formation, Sequence and Operationalization of the Reciprocal Working Committees (RWCs) of the GRP and the NDFP Negotiating Panels, 1995). 65 Reyes, "The Mapping of Peace Constituency Actors". 66 See joint statements launched by the parties, e.g. the Oslo Communique, 2001 andthe Joint Statement, 2016. one of the members of the team as 'a ceasefire that is built on a simple declaration on both sides but with a level of coordination'. 67 It is suggested that such a ceasefire could develop into a bilateral ceasefire over time, when both parties had agreed on a substantial agenda on the issue of socio-economic reform. 68 The ceasefire lasted six months, becoming the longest ceasefire ever reached between the parties. However it broke down, reportedly as the result of an NPA ambush of a presidential convoy. The government saw this as an indication that the communist party was not living up to its promises and cancelled the peace negotiations. 69

Understanding Ceasefire Diversity Across the Two Conflicts
A review of ceasefires in the two conflicts shows there are fundamental differences in the characteristics of ceasefires. In Mindanao, a comprehensive ceasefire has been a central component of engagement between the government and the MILF. The ceasefire is a formal bilateral engagement and includes an extensive organizational structure that has expanded and evolved over time. The ceasefire has also generated high levels of interaction and cooperation. In the case of the communist conflict, ceasefires have been unilateral declarations of a generally temporary nature and short-lived. This section of the paper discusses how these differences in ceasefires can be understood through consideration of the different aims, ideologies and strategies of the conflicting parties. With regard to this, I focus on sensitivity to conflict issues and approaches to violence as important to shaping approaches to ceasefire and their characteristics. At the core of the Mindanao conflict is an identity-based claim for selfdetermination for Muslim Mindanao and recognition of the Moros' right to their 'homeland', 70 the Bangsamoro. Thus, the Moro identity is closely intertwined with territory. The MILF has been described as a radical Islamic revivalist organization that 'advocates the Islamization of society and its political institutions, in particular the state'. 71 Given its roots within the MNFL, there are nationalistic elements to these aims. In terms 67 Interview with Braganza, 2018. 68 The idea of a 'coordinated unilateral ceasefire' is further described as a ceasefire that is built on a simple declaration on both sides but with a level of coordination, for instance through establishing ceasefire committees at which the parties will meet and 'safety zones', which are areas in which the armed units of the communist party cannot bring their guns out outside of the zones. The measure would constitute an attempt at building 'safety nets' into the ceasefire, in the event of violations (interview with Braganza, 2018). 69 Anonymous interview, 2018. 70 Santos, "Delays in the Peace Negotiations". 71 Santos and Santos,Primed and Purposeful,345 of ceasefires, given the MILF quest for self-determination and its statehood ambitions, entering into a formal bilateral agreement with the Philippine government has implied a degree of recognition as an actor. The ceasefire also recognized that the MILF held territorial control over certain areas that they had claimed, commonly referred to as MILF 'camps'. Recognition of the camps was put forth by the MILF as a precondition for engaging in talks with the government. When first settled in 1997, the ceasefire stipulated that the parties were to identify areas to be considered as MILF areas and that movement of MILF fighters outside of these areas should be coordinated. The first two major ceasefire breakdowns were related to military operations aimed at taking over control of these camps and it has been posited that the objective of the military was to withdraw the recognition of camps given by the government during the ceasefire negotiations, in order to safeguard the territorial integrity of the state. 72 This suggests that these were contentious issues and spoke directly of the core conflict incompatibility. The MILF political and military power was largely located in these areas, and recognized as being in control of them could be seen as a step towards sovereignty in these areas. 73 Thus, there was a rationale for the MILF to engage in a ceasefire that would accept and recognize on-the-ground realities in terms of territorial control, as this would allow them a degree of authority to develop their social and political institutions within these areas.
Although the MILF was severely weakened militarily by the military offensives in 2000 and 2003, it maintained great influence over these areas and was able to continue developing the Bangsamoro society and institutions, which was central to the conflict issues. As part of this and within the conceptual framework of the ceasefire, the parties agreed to facilitate rehabilitation and development in the war affected areas. 74 The MILF was authorised to organize a development agency, the Bangsamoro Development Agency (BDA), to lead and implement development projects. 75 This also help the MILF to maintain and improve support for its governance amongst its constituencies. As the previous MILF negotiating chairman, Mohager Iqbal put it, 'agreeing to cooperate and coordinate with government in implementing development projects in areas affected by the war even before a peace agreement is signed is a way of helping our people '. 76 This reflects the governing ambitions of the MILF, which were partially enacted within the framework of the ceasefire. 72 Coronel Ferrer, Costly Wars, Elusive Peace. 73 Åkebo, "'Coexistence Ceasefire'", 484. 74 Santos, ("War and Peace on the Moro Front," 79-80) describes this as a 'novel concept of a ceasefire not only for peace negotiations but also for rehabilitation and development'. 75 The agency was funded by government and private organizations. 76 Interviewed in Santos, "War and Peace on the Moro Front," 79.
As the conflict issues in the Moro conflict are closely linked to territory, this must be considered by way of shaping the characteristics of ceasefires. It has prompted an extensive organizational structure for cooperation and coordination and shaped interactions between people from different sides of the conflict, both within the joint ceasefire bodies and between people on the ground. The Mindanao conflict is geographically focused on the areas historically inhabited by the Muslim population in Mindanao and claimed by Moro groups. From the perspective of the Philippine government and military, this has posed a threat to the territorial integrity of the Philippines. As previously mentioned, the ceasefire recognized the influence of the MILF over several of its camps early on by stipulating territorial demarcation of these areas. First, these areas are not only military camps but communities with people moving in and out of the areas. This implies that the territorial demarcation did not result in a sharp separation of troops but that the process of coordinating movements around these areas prompted high degrees of interaction between the military and MILF fighters and civilians. 77 This prompted a comprehensive ceasefire arrangement. Second, the conflict landscape in Mindanao is complex with multiple armed elements and sources of violence. As the MILF has not been the only armed group involved in conflict with the government, ceasefire cooperation and coordination with the MILF has been in the interests of the government, as it has aided the government in dealing with other armed elements that were deemed a greater threat than the MILF. 78 For example, as part of this agreement, the MILF have shared intelligence with the Philippine army and allowed them to operate against armed elements within their areas. Another example is that the MILF has expelled leaders of the Abu Sayyaf Group residing and operating within its areas. 79 Thus to facilitate the ceasefire and uphold key principles, including coordination of movement of MILF fighters outside the areas they control, as well as coordination of police and military operations against other groups in or near MILF areas (in accordance with the AHJAG agreement), this has prompted an extensive ceasefire structure and institution and also shaped interactions on the ground. In this respect, the ceasefire resonates with Sosnowski's description of a 'substantive' 77 Åkebo, "'Coexistence Ceasefire'," 484. 78 Ibid. 79 ICG, "Counter-Insurgency vs. Counter-Terrorism," 10-11. The Abu Sayyaf Group is also making Moro separatist demands; however, it aspires to create an Islamic state for the whole of Mindanao, not just the Muslim populated areas, and reflect a more radical version of Islamic revivalism in this quest. It has remained firmly opposed to any compromise on independence. Unlike the MNLF and the MILF, the group has not been negotiating with the government or engaged in ceasefire. It has exclusively invested in violent tactics, with an approach described by Santos and Dinampo as 'extreme prejudice', including by targeting civilians and with a proclaimed 'willingness to die for the cause' (Santos and Dinampo, "Abu Sayyaf Reloaded," 119-21). These aspects are seemingly important to understanding the absence of a ceasefire in this case.
ceasefire, marked by high degrees of detail and with specific security arrangements being a central feature. 80 In contrast to the Moro conflict, the communist insurgency is aimed at overthrowing the central government and replacing it with a 'peoples' power' and a Maoist governance system. The strategy for reaching this goal is through a 'protracted peoples' war' and by winning over the majority of the population to join the struggle. The strategy is to start in the countryside before finally assuming power in the cities, including the capital city of Manila. The communists recognize that this struggle will take time and accordingly reference it as a 'protracted war'. Despite decades of negotiations, the CPP remains committed to its vision of creating a radically new form of state in the Philippines while these demands are deemed unacceptable to the government, which has rejected what they see as demands for power-sharing with the communists. 81 Santos suggests that to both parties negotiations have generally been of secondary importance and used to serve other objectives: the communists have relied on its political and military strategies to win popular support to overthrow the government, while over the years ruling state regimes have predominantly invested in defeating the group through military means. While it was agreed early on that a ceasefire would be the last item on the agenda once agreements had been reached on all other substantive elements, it is evident that the government negotiators have emphasized the issue of ceasefire, although there have also been sectors within government and military that have questioned which changes would come about and whether a ceasefire would be favourable to the government. 82 In an effort to come around the communists' resistance to a bilateral ceasefire agreement, the government has tried to advocate the concept of a 'coordinated unilateral ceasefire', which would contain very similar features to that of a bilateral ceasefire. In this regard, it can be noted that according to Quimpo,83 at the commencement of negotiations in 1986-87 the government had already explicitly stated that engaging in a ceasefire would not grant the negotiating NDF 'belligerency status'. This underlines that there was no rationale for seeking or extending recognition via engagement in a formal ceasefire. Under these conditions, a bilateral ceasefire was interpreted by the communists as 'a surrender to GRP [Government of the Republic of the Philippines] authority'. 84 Thus, instead we have seen a pattern of unilateral declaration of intentions that resonates with descriptions of 'symbolic' ceasefires, with no further details of commitments or of modes for implementation. 85 80 Sosnowski, "Towards a Typology". 81 Santos,"The Communist Front," 18. 82 Interview with Braganza, 2018. 83 Quimpo, "The Use of Human Rights". 84 Ibid. 85 Sosnowski, "Towards a Typology".

Concluding Remarks
This paper aimed to increase the knowledge and understanding of what shapes the characteristics of ceasefires and how variation across conflicts can be understood. On the basis of two case studies within the same country, the paper highlights the diversity amongst different types of ceasefire, variations in approaches to ceasefires and their meaning to conflicting parties. It also demonstrates how one actorthe Philippine governmentinvolved in multiple conflicts can use different strategies and approaches for conflict resolution across conflict contexts. The study shows how in the Moro conflict in Mindanao, there was a rationale for a bilateral ceasefire engagement that included territorial demarcation and an extensive ceasefire structure for security cooperation and coordination. In comparison to the communist conflict, the government recognized the MILF demand for self-determination more and was more open to engaging in this type of ceasefire agreement. This approach was also a partial response to the incompatibility, for example through recognition of some MILF territorial influence. Situated within a geographically confined conflict landscape but also including the presence of multiple armed elements and characterized by fluidity in terms of movement and control, this has urged high levels of ceasefire interaction, cooperation and territorial coexistence. The government side was also aided by this cooperation and coordination, as it helped it to conduct its operations against other armed elements. Ultimately, the ceasefire contributed to not only transforming the behaviour of the conflicting parties but also the structure of their relationships. In the case of the communist conflict, which involves regime change and with geographical spread across the Philippine state, there has not been the same rationale in terms of ceasefires. The pattern has been to talk while fighting, to declare temporary unilateral ceasefires for specific purposes and to settle the core conflict issues before mutually agreeing to cease fire. This suggests that what type of ceasefires conflicting parties are willing to enter into is fundamentally shaped by what their conflict aims are to begin with. Ceasefires can be used to achieve or come closer to the aims of the struggle, but it is not deemed equally useful for all types of insurgency.
The study highlights the advantages of approaching ceasefires from a longterm perspective, as it helps illuminate important patterns in types of ceasefire, the salience of ceasefires in engagement between conflict actors and changes in relationships. It allows the recognition of the dynamism of the ceasefire process and an understanding of the changing patterns of interaction through ceasefire engagement, as well as the legacies of previous ceasefires and peace attempts. The study improves our understanding of the variation in ceasefire types through a contextual analysis of the overarching rationales for engaging in ceasefires across two conflict contexts. Of course, these differences do not imply that the conflicts are bound to follow a certain trajectory in terms of ceasefires. Conflict actors, structures and issues are dynamic and can change over time, as result of interactions between the warring parties, internal political dynamics and the external context. Furthermore, events in social life have multiple causes that are often mutually reinforcing. In this respect, the findings from this study prompt questions for exploration in future research regarding how this is significant to the influence of other factors, for example that of external actors, timing and sequencing of peace initiatives, the consequences of intraparty rivalry, as well as of elections, change of regime and changes in political agendas.
Findings from the study may also be relevant to other cases of intrastate conflicts with experience of attempts at peace with armed insurgencies, such as for example Colombia, Myanmar and Sudan. Previous research suggests that both the type and design of a ceasefire are important to understanding the outcome and duration. Knowledge about how different types of ceasefire come about can avert the risk of misleadingly interpreting causes of ceasefires as outcomes. That is, the analysis may miss the fact that the very reason why we see a certain type of ceasefire in a conflict is also what explains the outcome of the ceasefire. Such insights are important for both qualitative and quantitative research that aims to explain the implementation, outcome and duration of ceasefires. The knowledge can also be useful for policy and inform conflict resolution efforts. In this respect, a better understanding of the different rationales for ceasefires cautions against searching for a template for ceasefire design and the expectation that it will operate in a similar way across conflict settings.