Social movements, democratization, and civil wars
Donatella della Porta
Episodes of democratization sometimes turn violent. For instance, violence was present in the mobilization for democracy that started in Eastern Europe in 1989 or across the Middle East and North Africa (MENA) region in 2011. On rarer occasions, conflicts have escalated into civil wars. Focusing on a specific path of violent escalation, our research bridges the literatures in the fields of social movement studies, civil wars, nonviolent resistance, and democratization, which have usually proceeded quite apart from each other.
The social science literature on civil wars has singled out various potential causes for their onset, duration, and brutality, mainly drawing on large-N studies. Among these causes are economic modernization, ethnic nationalism, absence of political democracy, rough terrain, cross-border sanctuaries, state weakness (political instability and large country population), distant areas, diasporas or foreign government support, and type of production (high value, low weight). This work has brought about an accumulation of variables defined as root causes for civil wars’ onset, intensity, and duration, with some contradictory results. After a focus on grievance theories, addressing human rights deficits and discrimination as well as protracted social conflicts over deprivation, another wave of analysis addressed greed as motivation, highlighting natural resources but also the corruption of governments engaging in rent-seeking and predation or, more generally, neo-patrimonial and personalist regimes (Collier and Hoeffler 2004).
However, large-N studies have often been criticized as inconclusive, as findings are sensitive to coding and measurement procedures of the dependent variable as well as the considerable distance between theoretical constructs and proxies (Kalyvas 2008). On the other hand, detailed case studies have not produced much accumulation of knowledge. We have suggested in this volume that especially social movement studies, but also studies of failed democratization and nonviolent revolutions, could usefully integrate research on civil war through a focus on causal mechanisms. In what follows, we will summarize the results of the empirical chapters.
In our research, we have focused on democratization episodes that failed, bringing about the onset of civil wars. While democratization is admittedly only one path towards civil war, its analysis allows us to single out some main processes that facilitate conflict escalation at the political level. Social movement studies have pointed to sudden shifts in opportunities, which allow activists to challenge elites. Looking at the exogenous determinants of protest repertoires, social movement scholars have stressed the roles of both stable institutions and contingent developments. Regarding the former, more centralized political power is seen as making political institutions less accessible ‘from below’ and, thus, fuelling more violent protest (Goodwin 2001). Additionally, conflict escalation has been considered to be more likely when historically rooted ways to deal with opponents orient authorities towards exclusive strategies. Less durable political contingencies, such as the lack of availability and influence of political allies have also been mentioned as closing windows of opportunity for protestors, often producing escalation (see della Porta and Diani 2006, Chapter 8 for a review). Low levels of freedom and democracy have been linked to political violence at the national level, even in Western Europe (Engene 2004), and less proportional electoral systems have been linked to ethnic violence (Crenshaw 2011). The weakness of democracy, civil liberties, human rights, rule of law, and so on are often considered to be root causes for radicalization. While moderation and power-sharing are seen as facilitating peaceful democratization, research on revolutions states that, when normal channels of access to the political system are blocked, violence might be perceived as necessary, as there is ‘no other way out’ (Goodwin 2001).
Our research confirmed political destabilization as a causal mechanism in episodes of civil war. The perceived appearance of a possibility for regime change is followed by strong reactions from the regime, which, rather than bringing about a defeat of the opposition, lead to its radicalization and to conflict escalation. Actors from below appropriate opportunities to mobilize, but are not strong enough to produce the breakdown of the regime which, in turn, is shattered but not fully defeated. Emerging internal anarchy pushes citizens to take responsibility for their own security (Toft 2003). Our case studies show how opportunities open up during the conflicts through dynamic interactions.
Libya is a case of failed democratization in a weak state, which has been destabilized. In Libya, Colonel Qaddafi, who took power in 1969 through a coup d’état, shifted power to members of relatively disadvantaged (small) Libyan tribes and urban middle-class professionals. Qaddafi’s political project then moved towards a totalitarian-style state in which popular rhetoric combined with brutal repression in the control of all elements of civil and political life. State institutions remained underdeveloped, viewed as potential hotspots of opposition, and were left underdeveloped and disempowered as Qaddafi disbanded the (tribally based) National Congressional Council. The defection by the elites developed quickly, as early as February 2011, with several of the most important government officials resigning from their posts and joining the opposition. During the revolt, representatives of economic interests – such as the Arabian Gulf Oil Company, the second-largest state-owned oil company in Libya – also defected to the rebels, as did Islamic leaders and clerics, among them the Network of Free Ulema (the clerics) and some key tribes such as the Warfalla, Tuareg, and Magarha.
The state was stronger in Syria, where ‘the dense links between Bashar al-Asad and the regime core prevented elites’ abandonment of the president; at the same time, thicker state-society relations in Syria prevented a Libya-style social isolation in the top leadership’ (Hinnebusch, Imady, and Zintl 2016, 225). First, the regime enjoyed a nationalist legitimacy, given its stance against Israel; second, neoliberal reforms had enriched the well-connected middle classes in cities like Aleppo and Damascus. Bashar al-Asad followed his father as Syrian president in June 2000, promising political liberalization; however, the president and his close family remained at the core of the regime, occupying leading positions in the army (al-Asad’s brother), military intelligence (his brother-in-law and one cousin), and the economy. In general, clan loyalty also played a role, as Alawi elites continued support for the regime, while defectors were Sunnis. However, the wave of protests that spread from Tunisia and Egypt and also to Syria contributed to destabilize the regime by producing cleavages within the elites.
In Yemen as well, the peaceful uprising, inspired by events in Tunisia and Egypt, unsettled the authoritarian power of the country’s leader, Ali Abdullah Saleh. Yemen, born out of the unification of North and South Yemen in 1990, was a weak state plagued by poverty, rampant corruption of the ruling elites, and a security void. With the localized wars in the North, calls for secession in the South and presence of al-Qaeda, Yemen’s unity was fragile, and national identifications were undermined by regional and tribal affiliations. The regime of Ali Abdullah Saleh relied on repression but also on patronage networks that further fragmented society and weakened the state (Alley 2010). The core of the regime was built around Saleh’s family, who occupied the most important military and state security positions, and other elites allied with the president through personal relationships and exchange of favours rather than principles or ideology (Durac 2013; Khosrokhavar 2012). The fluid nature of these alliances was confirmed when, in March 2011, following the massacre of peaceful protestors, core members of the regime defected and joined the opposition. The rupture of the regime led to armed confrontations between Saleh and the defectors, changing the course of the previously peaceful uprising. At the same time, as Saleh and his family directed all their energies into the struggle for survival, the security situation in the whole country started to deteriorate dramatically.
In the 1980s, Yugoslavia was certainly not a weak state and initially seemed to enjoy the best conditions for a peaceful and straightforward democratization process, including a relatively high degree of liberalization, reflected in a comparatively large number of civil society organizations. A looming economic crisis was exacerbated by Tito’s death, and the ensuing succession struggle helped usher in a new generation of party leaders in the various republics. The breakdown of Titoism as a political and economic system (Vasilevski 2007, 5) developed with increased internal competition among the republics. By 1988, ‘the great majority of Tito’s chosen coterie of republics’ leaders had been replaced with leaders who had no common loyalties’, and ‘both the Yugoslav federation and the Yugoslav Communist party – the League of Communists – which nominally ruled the country had lost much of their legitimacy’ (Pavkovic ´ 2000, 76). Socioeconomic and cultural inequalities then fuelled internal conflicts, as political elites at the level of the republics started playing on nationalism as a way of maintaining their own power (Licht 2000, 113). In fact, this happened through the activation of multiple cleavages:
Old and new divisions in the political class came to the fore, such as between promoters of greater control of Serbia’s central government over its autonomous provinces and their foes; between advocates of a stronger federal centre and protectors of the status quo; between proponents of change in the party’s Kosovo policy and their opponents; between conservative and liberally minded politicians; between members of various political generations; and between high- and low-ranking officials. Since the divisions often cut across one another and high officials engaged in complex political maneuvering, relations within the political class became rather complicated.
(Vladisavljević 2008, 126)
In sum, both weak and less weak states were destabilized through waves of protest for democracy, which challenged regimes that had already shown some signs of stress. We thus observed that in Libya, Syria, Yemen, and Yugoslavia alike, political destabilization was a key mechanism in the overall process of mobilization for democracy radicalizing into civil war. At the same time, our cases show the different roles the mechanism can take in the overall process. In Libya, it intersected with economic conflicts around the oil industry and was exacerbated by social fragmentation around tribes and traditional religious elites. In Syria, social fragmentation instead countered political instability due to the stronger social underpinnings of the regime around President Bashar al-Assad. In Yemen, the destabilization meant the rupture of the regime, accompanied by security deterioration and fragmentation, with new alliances and divisions built around the warring elite factions. In Yugoslavia, finally, political instability also intersected with social fragmentation but within a more nationalist framework and formalized political structure.
In research on social movements, the policing of protest is considered a central factor in radicalization. Research on the topic has identified a tendency to use harsher styles of protest policing against social and political groups that are perceived as greater threats to political elites, given that they are more ideologically driven or more radical in their aims (see Davenport 2000; della Porta and Fillieule 2004; Earl 2003). Additionally, police repression is more likely to be directed against groups that are poorer in material resources as well as in political connections (della Porta 1998; Earl, Soule, and McCarthy 2003). The form state power takes has a clear impact on the policing of protest. If repression is always much more brutal in authoritarian than in democratic regimes (e.g. Uysal 2005, on Turkey), even authoritarian regimes vary in the amount, forms, and actors of protest they are willing to tolerate, as well as in the forms in which they police the opposition (Boudreau 2004). Violent uprisings seem to develop when very repressive regimes have long thwarted any development of an autonomous associational life – let alone, of social movements (Goodwin 2001). While repression might reduce the individual’s availability to participate by increasing the costs of protesting, in some cases a sense of injustice as well as the creation of intense feelings of identification and solidarity can strengthen the motivation to oppose an unjust and brutal regime (Davenport 2005; della Porta and Piazza 2008; Francisco 2005). So, especially when the protest is widespread and well supported, repression can backfire due to outrage about police disrespect for citizens’ rights at the national as well as the transnational level (Davenport 2005; Francisco 2005). In general, repression seems able to control low levels of discontent but fuels armed opposition in cases of broader unrest (Regan and Norton 2005).
If repression is a cost, resistance can become a reward in itself. Our research indicated that this tended to happen especially in situations of indiscriminate and inconsistent repression, which left some free space out of the reach of the regime. Past and present brutal repression strongly influenced the activists’ choices, reducing faith in a peaceful surrender of incumbents. Similarly, in our cases, memories of regime brutality worked as a constraint to mobilization in some moments, but in others fuelled protest as the only way to express a call for dignity. The costs of compliance with the regime became unbearable. At the same time, brutal repression fuelled counter-violence, with spiralling escalation.
Never a liberalized autocracy, Qaddafi’s regime had used extremely high levels of repression of any political dissent (Joffé 2016). At the onset of the conflict, the protest in fact politicized on the issue of repression, with demonstrations organized by the families of prisoners killed in the Abu Salim prison massacre as well as against the growing repression of those very protests, primarily but not only in Benghazi. In February 2011, violence escalated in various parts of the country, as the regime warned citizens that the state security would use force to disperse any collective actions perceived as promoting public sedition (Jacinto 2011). On 27 February, Qaddafi blamed foreign powers for the disorder and called for loyalists to ‘cleanse Libya house by house’ (Joffé 2016, 132). Indiscriminate police attacks against protestors immediately spiralled into violent confrontations as the regime proved unable to impose order, especially in some regions.
Also in Syria, where repression had thwarted the development of a civil society, the dynamics of the protests dramatically changed during the upheavals when protest was met with brutal repression, but repressed cleavages re-emerged during the uprising. Mobilization started in March 2011 in the southern city of Dara’, moving
From largely peaceful protests (that were immediately repressed using live ammunition by police) to a full out civil war in which an increasing amount of heavily armed and well trained cadre of army defectors, regular Syrian and foreign fighters are staging an armed insurrection against the regime of Bashar al-Asad.
(Gelvin 2011, 108–9)
Escalation followed the torture of children, accused of having spray-painted slogans against Assad. Repression then produced further escalation, as ‘The regime’s forces, lacking training and experience in crowd or riot control, continued to respond with excessive violence, multiplying regime enemies, and making funerals occasions for more confrontations’ (Hinnebusch, Imady, and Zintl 2016, 231). Police brutality was in fact taken as proof of the ineffectiveness of the alternative, nonviolent option – especially after the killing of 138 during the first day of Ramadan, and the shelling and takeover of Hama and Deir ez-Zor as well as Idlib and Latakya in August. While the funerals of the victims of regime repression reignited protest, on 19 July 2012 the assassinations of members of the Syrian security elite in Damascus confirmed the vulnerability of the regime to armed attacks.
In Yemen, the rulers of the new republic in the early 1990s had initially allowed for some liberalization, making the country’s vibrant civil society and public sphere open to dissent and debate, uniquely in the region (Carapico 2007; Wedeen 2008). The democratic façade started to crumble following the 1994 civil war between the South and the North, with Saleh’s rule becoming more overtly authoritarian. Overt political contention was nevertheless widespread in Yemen and protests, organized in particular around social claims, were held regularly throughout the 1990s and the 2000s. They were usually met with a mix of repression and concessions. A similar strategy was adopted by Saleh from January 2011 on, when arrests and harassment of protestors were accompanied by promises of better salaries and other improvements. Gradually, however, more violence was deployed, until on 18 March 2011, snipers fired upon peaceful protestors in Sana’a, killing at least 52 (HRW 2013a; 2013b).
The brutality of the regime outraged Yemenis, leading to the radicalization of people’s demands. Over the next days, despite the violence and the announcement of the state of emergency on 23 March 2011, the numbers of protestors grew and demonstrations spread across the country, reaching even the most remote areas (Carapico 2011). While the youth remained committed to peaceful methods, the regime defectors with their army units and tribal militias engaged in armed clashes with forces loyal to Saleh. By mid-June, cities like Sana’a were divided by checkpoints and some neighbourhoods were turned into military zones (Fattah 2011). The escalation of violence was eventually quelled by the transitional deal brokered by the Gulf Cooperation Council in November 2011. The agreement, which forced Saleh to resign but granted him immunity, was rejected by the youth protestors, who saw it as a betrayal of the revolution’s objectives. Thus, despite the declaration of the so-called ‘transitional period’, neither the popular mobilization of the youth nor the brutal repression of resistance ended (Alwazir 2016). At the same time, with the fragmentation of the movement for change, some of its participants, like the Houthis, returned to their narrow interests and more violent repertoires of action.
Repression had at first been weaker in Yugoslavia where, even if purges had occasionally followed calls for reforms, there had been a gradual liberalization. The situation precipitated, however, after the 1990 elections brought Miloševic ´ into power as president of Serbia, and the army sided with him in his attempt to conquer part of Croatia and Bosnia. Although no war was declared, a military draft was initiated in May 1991. When violence erupted in the Serbian-inhabited areas of Croatia, the Yugoslav people’s army intervened, the Serbian government declared, in order to protect them. This triggered spirals of protest and repression, with the federal army nevertheless unable to impose order in all the republics. The intervention of the army served to quickly radicalize all sides of the conflict, with violence soon spiralling out of control.
Summarizing, while the degree of previous brutality of the regimes we analysed varied, in all cases, initially peaceful protests escalated in response to repression that was considered indiscriminate, but also ineffective in re-establishing order in the national territories.
Social movement studies have stressed the importance of coalition building. As in revolutions, the emergence of waves of protest is related to the capacity to activate large networks by bridging already existing connections and creating new ones. As mobilizing networks always involve cooperation but also (potential) competition, mobilization is likely to subside when (especially in declining phases of protest cycles) splits multiply around strategies and ideological divergences (della Porta and Diani 2006). Multi-cleavage environments present even more challenges as they provide increased opportunities for divisions (della Porta 2013). Research on democratization has often suggested that the distribution of power among ethnic groups could jeopardize the search for peaceful solutions. Studies on nonviolent revolutions have considered the unity of the challengers as an important factor to success (Schock 2005). Civil war studies have often mentioned ethnic belonging as a resilient source of conflict, especially when mixed with a perception of discrimination as socioeconomic grievances are triggered by regional political marginalization (Ylönen 2005). Civil wars stem from horizontal inequality and ethno-political discrimination, as ‘countries with one or more ethnic group(s) radically poorer than the national average and countries with large groups discriminated from national politics have a significantly higher risk of armed anti-governmental opposition’ (Buhaug, Gates and Lujala 2013, 429). In particular, ‘large discriminated groups boost the probability of governmental civil wars, in part because of the evident disconnect between demographic power and political privileges’ (ibid., 429). Confirming in part these views, one can note that in fact, in our cases as well, the weakness of civil society interacted with (the potential for) ethnic divisions, which in turn fragmented the opposition.
In Libya, in 2011 and 2012, a popular uprising against long-term authoritarianism evolved into a civil war that was ultimately interrupted through the direct military involvement of a coalition of foreign states. The existent tribal divisions and traditional values were utilized by the regime in order to divide and control. Qaddafi, in fact, played the tribes against one another, even while trying to accommodate some of their requests for power resource sharing, privileging some over others. Together with patron-client networks, inter-tribal and inter-ethnic distrust therefore contributed to the dictator’s power. A strict form of patrimonialism and clientelism developed, but the country remained politically and administratively fragmented. Latent tribal conflicts had involved Berbers, Warfalla, and Cyrenaica groups in particular, and violent Islamism also played a role. As Anderson (2011: 6) explains:
Libyan society has been fractured, and every national institution, including the military, is divided by the cleavages of kinship and region. As opposed to Tunisia and Egypt, Libya has no system of political alliances, network of economic associations, or national organizations of any kind. Thus, what seemed to begin as nonviolent protests similar to those staged in Tunisia and Egypt soon became an all-out secession – or multiple separate secessions – from a failed state.
In 2011, protests grew in size and frequency, with one early slogan reading ‘Libya free, nation united’. Uprisings started in Eastern Libya, spreading after a few days to Tripoli; Benghazi was the Eastern base of the anti-regime protests, which also grew in the western cities of Zawiya, Misrata, and even in the capital, Tripoli. While the level of violence rose, peaceful unarmed demonstrations also continued: ‘as the chaos of war progressed, disrupting communal and distribution services, spontaneous committees began to emerge that took over municipal and communal responsibilities’ (Joffé 2016, 132). However, the opposition remained fragmented, with low organizational capacity and a proliferation of heavily armed militias on regional, ethnic, tribal, and religious bases.
The opposition also fragmented in Syria, where the heterogeneity of civil society – with ethnic minorities, including Kurds, Palestinians, and Armenians, as well as religious minorities such as Christians, Druze, Sunnis, Shiites, Alawis, and Ismae-lis (as specific Shiite denominations) and small communities of Jews –‘had long undermined collective action among the opposition’ (Hinnebusch, Imady, and Zintl 2016, 225). As President Hafez al-Asad took power in 1970, he consolidated his control over the armed forces through tribal and family connections, resulting in Alawi domination of sensitive positions in the army (Perthes 1992). Here as well, the regime addressed the various groups through patronage, as ‘The army, intelligence services and Ba’ath party organizations became an instrument for institutionalized corruption in which powerful patrons secured support from lower ranking members, via a highly developed clientelist system’ (Donker 2012). When the uprising emerged, efforts at mobilizing brought about the creation of an umbrella organization for the local committees – the Local Coordinating Committees of Syria (LCCS) – which especially involved young people without previous experience of political socialization (Hinnebusch 1993). However, the uprising also mobilized long-standing opponents of the regime, including Kurdish parties, the Committees for the Revival of Civil Society (which emerged from the 2001 ‘Damascus Spring’), as well as elite oppositional leaders who joined the mobilization effort soon after it begun. Even if Kurds, Sunnis, or other minority groups often dominated in terms of numbers, coordination committees aimed at being inclusive. Trans-sectarian appeals were also spread by the movement: ‘we Syrians are one people’, declared the singer and activist, Fadwa Soliman (Khos-rokhavar 2012, 218), and the words of a song read: ‘Come on, Bashar, leave . . . You, the US agent! The Syrian people will not accept humiliation any more’ (ibid., 241).
But as in Libya, Syrian civil society was weak and its fragmented nature shaped, despite protestors’ statements to the contrary, how the uprising developed. As the strong repression of protests had thwarted political organizations, leaving religious, tribal, clan, and family ties as the main bases of identification and networking, protests spread especially in the periphery, based on strong tribal links. Divisions grew in a fragmented Syrian civil society, with a multitude of minorities and experiences of sectarian violence already in the 1980s (with the Muslim Brotherhood’s revolt culminating in the massacre of Hama in 1982). As the uprising became more violent, ethnic and sectarian divisions grew ever more polarized – culminating in the eventual emergence of Jihadist movements and the founding of an ‘Islamic’ State.
In Yemen, the early stages of mobilization brought surprising levels of unity, giving rise to an unlikely alliance among Zaydi rebels, Islamists, socialists, tribal men, and urban youth. The regional and tribal interests seemed to have been suspended and replaced by the shared goal of toppling the regime. Occupied squares across Yemen became ‘physical and social spaces where people from all these different groups met and exchanged views, shared living conditions, and generally interacted in a largely unanticipated manner’ (Lackner 2016, 164). The unity started to break up, however, when armed factions joined the struggle. Many abandoned the squares, as they felt uncomfortable with the regime defectors becoming the new leaders of the revolution (Manea 2015). The fragmentation further increased with the (internationally reached) power transfer deal. While the youth of the revolution rejected the deal and kept protesting, other actors agreed to it and moved towards institutional politics, or returned to their narrow interests and agendas. During the period of the National Dialogue Conference between 2013 and 2014, despite the talks of unity, the fragmentation in the country deepened, with the Houthis expanding their territorial control, Southerners calling for secession, and tribes engaging in localized struggles for resources or autonomy. Finally, the alignment between the former protestors (the Houthis) and the former president (Saleh) broke up the clearly demarcated camps of the uprising (opposition versus regime), putting forth new alliances and divisions that marked the phase of the civil war.
In Yugoslavia, civil society organizations had developed in various contentious moments. Especially in Slovenia, new social movement claims had spread on issues of reproductive rights, gay and lesbian rights, and nuclear energy (Licht 2000, 120). In the relatively liberal climate of the late 1980s, feminists, environmentalists, and pacifists coordinated their protests (Figa 1997, 169; Benderly 1997, 184), consisting of peaceful, direct actions (Figa 1997, 169). Mainly composed of intellectuals, these groups developed a strong anti-nationalist stance (Bilic 2012, 94). Some characteristics of the Yugoslavian state building, and its position during the Cold War, formed however the preconditions for inter-ethnic conflicts that the federalist asset was not able to address. The weak power of the federation, with the domination of the republics over the centre (Bunce 1999, 111), facilitated fragmentation into multiple states. Additionally, the historical enmities between the Catholic Croats and the Orthodox Serbs – and between the Croatian fascists, the Ustasha, and the Serbian resistance, the Chetniks – were then revived. The effects of economic and cultural inequalities became all the more dramatic, given another characteristic of Tito’s Yugoslavia. As he had linked socialism to the nation,
[This] contained a critical weakness that Tito and his colleagues could never have imagined. As long as the Communist movement remained strong, Yugoslavism was not in danger. If nationalism reared its head the party could and did push it back under the surface. If the League of Communists of Yugoslavia should disintegrate, however, then the Yugoslavism it championed would disintegrate too.
(Stokes 1993, 223)
In the years to come, in the face of brutal escalation, Milošević continued to present himself as a reformist communist and a nationalist: he was described, in fact, as
the proverbial example (though rare in actual fact in the region) of the communist who, recognizing the threats imposed by the dissolution of the communist order, opted nearly overnight (and helped along by audience reactions to a speech given in Kosovo) to redefine himself as a nationalist.
(ibid., 92)
As such, social fragmentation can be observed in each of our four cases. But whereas, in Yugoslavia, this fragmentation was linked mostly to nationalism, in Syria, it also translated through sectarian and ethnic differences. In Libya, though also present in the Syrian case, social fragmentation was expressed was specifically through the tribes. In Yemen, the fragmentation was even more multi-layered, with tribal, regional, and sectarian lines of division. This meant that in all cases, social fragmentation intersected with other political and economic mechanisms, but it did so through country-specific institutions.
Political instability sometimes brings about a redefinition of (formal and informal) borders, which can also imply a fluidification of previously existing borders and the setting of new ones. Social movement studies have looked at the creation of free space. In authoritarian regimes, but also in democracies, challengers might constitute spaces in which to prefigure alternative ways of living and social orders. Squatted youth centres or protest camps might be seen as engaged in fluidifying power relations, order, and borders. A fluidification of borders becomes increasingly visible in civil wars as rebels gain control of territories, where an emergent order has to be established. In fact, research on civil wars has focused on territorial contestation, as at their onset there is a fluidification of borders, with some capacity by the rebels to control specific territories. Typically during insurrections, especially in authoritarian regimes, spaces are freed up by the opposition. Our research also pointed to the fluidification of borders as a key mechanism in civil wars dynamics, constituted by the contested reformation of borders and order.
In Libya, the process of fluidification of borders is visible in the focusing of protests in certain geographic areas, given the regime’s inability to keep control over parts of its territory. The collapse of the regime was triggered by a loss of support from the key Libyan tribes as well as key individuals within the government and the army, especially those from the eastern region of Cyrenaica (such as the former Minister of Justice and the former Minister of Internal Affairs), while the police and the army withdrew from Benghazi. A quick loss of territorial control by the regime’s security forces has been linked to the geographic structure of the country, with the population concentrated in Tripoli and Benghazi and a desert in between. Even after the military conflict ended with the regime’s breakdown on 23 October 2011, tensions between tribal elites from east and west (Cyrenaica and Tripolitania), with their respective militia alliances, continued to fuel insecurity under still strained economic conditions.
In Syria, rapid escalation of the uprising was also related to increasing territorial contestation as rebels began to declare cities ‘liberated’, which the regime would subsequently try (often successfully) to re-occupy. As a reaction, more residents would start to arm themselves and train in order to resist state repression. The most active regions at the time were those around Idlib, Homs, and Hama: a conservative Sunni majority region. Retreating from the latter areas, the regime focused repression on the two main cities of Aleppo and Damascus. In this context, peaceful opposition survived, notwithstanding repression, but increasingly transformed itself into governance-related institutions. Because of
government services and security break down in many areas, local neighbourhoods organized an alternative local governance set-up, including self-defence groups, an underground clinic systems, schools, media, and transportation services. With the intensification of the armed conflict, local communities became active in human relief.
(Hinnebusch, Imady, and Zintl 2016, 242)
This process of redrawing of borders, with some areas freed by the rebels, intensified following the summer of 2012, after the rebel attack on both Damascus and Aleppo.
With the escalation, internationalization also increased. Inside the country, Palestinian refugees had to take positions in the conflict, and Kurds contacted foreign Kurdish parties (mainly the PKK and Iraqi KDP). Some countries provided arms to the regime (Russia and Iran), while others supported rebel factions (from countries such as Saudi Arabia, Qatar, Turkey, and the United States). The growing tensions in the region were reflected in international interventions, with Turkey, Saudi Arabia, and Qatar supporting the opposition, but Iran, Hezbollah, and Iraq supporting the regime. There was, moreover, local control and supply of services by al-Qaeda or the Kurdish Democratic Unionist Party. With this internationalization, rebel groups proved able to control areas south of Damascus, around Hama, and all across the northern parts of the country. As we saw, the enduring loss of regime control over these regions necessitated a further development of governance organizations – a process that turned highly contentious as civil opposition councils, rebel groups, and Islamist movements increasingly struggled among themselves over control and boundaries of the areas they controlled. Added to this situation was an increasing sectarian rhetoric within which IS emerged. Its rise was a response to the militarism of the regime, but also to the vacuum produced by state failure in both Iraq and Syria, with the spreading of sectarian identities, the war economy, and plundering of state resources. Once polarization developed and radical Islamists increased their influence, they were able to bring an Islamic alternative to governance in practice.
In Yemen, where the popular mobilization of 2011 broke from the usual scripts of political engagements in the country, occupied squares, in Sana’a and other cities, gained the status of ‘free spaces’. They were not necessarily safe or free of the regime’s violence, but nevertheless turned into zones of creativity and encounters among diverse actors, against the rigid social boundaries typical in Yemen (Lackner 2016; Alwazir 2016). Occupation of squares continued well into 2013, to protest – among other things – the outcomes of the transitional agreement that subverted the goals of the popular uprising. For the youth, the interference of international actors clearly revealed that Yemen’s future political landscape had to meet the interests of regional powers and not popular aspirations. The decisive role of the neighbour-ing Gulf monarchies was confirmed when, after the Houthis reached and overran parts of the South in March 2015, the Arab coalition responded with airstrikes and blockades of Yemen (Salisbury 2016). The fluidification of borders took place also at the local level, as during the transitional period, beginning in late 2011, the void in security and state was filled by various groups taking over in particular localities, such as the Houthis in the North, al-Qaeda in some southern cities, and various tribes in areas across the country. In 2013–14, the Houthi rebels managed to move beyond the highlands, contesting new provinces until they reached and seized Sana’a in September 2014. While they initially earned some local support by providing basic services and removing corrupt officials, their expansion southwards gave rise to growing opposition and acts of resistance. With the latter being brutally repressed, the armed confrontations over contested territories followed, resulting in a full-fledged civil war.
In Yugoslavia, as well, borders fluidified quickly. In September 1988, ‘high officials of Serbia now effectively certified specific protest groups and their demands and claims as fully legitimate. They openly embraced popular participation in politics, albeit on populist terms’ (Vladisavljevic ´ 2008, 150). So-called rallies of solidarity in support of the Kosovo Serbs multiplied in Vojvodina, Serbia, and Montenegro, mobilizing tens of thousands – up to 700,000 in Belgrade on 19 October 1988. While the protests were initially mainly nonviolent and moderate in their discourse, they radicalized later on. Initially justifying intervention as the defence of oppressed (Serbian) minorities, during the mobilization, the aims broadened to bringing both Vojvodina and Kosovo under Serbian control. Fearing the increasing power of the Serbian republic, most other federal states declared their independence (Vasilevski 2007). The former communist elites played the most important role in all the new countries, even though, with the exception of Slovenia, democratization was delayed by the internal wars. After 1989 followed the years of dramatic events, such as the wars in Croatia (with about 25,000 people killed) and Bosnia-Herzegovina (with about 200,000 deaths and a million people displaced), plus the 78 days of the NATO bombing campaign, allegedly against violations of human rights in Kosovo, with a rampant economic crisis and mounting corruption.
The fluidification of borders is very pronounced in all our cases. In both Libya and Syria, we see that the loss of regime control over (part of) its territory necessitated the creation of local governance organizations. With their emergence, governance fragments and fluidification set in, as boundaries and authority over territory become contested. But whereas in Libya this mechanism was mostly internal – as no state institutions remained after Gaddafi’s removal, tribes and rival governments took over – in Syria and Yemen, it was much more closely linked to sectarian identification and the intervention of foreign actors. Practical governance fell apart along ethnic and sectarian lines, a process that was exacerbated by its resonance among the foreign actors that intervened. In the Yugoslavian case, instead, we observed that the fluidification of borders emerged through questioning the relationship of specific regions to a Serbian nation.
In both social movement studies and civil war studies, researchers have mainly addressed socioeconomic characteristics as potential sources of grievances. In work on social movements, however, grievances have tended to be dismissed as causal explanations and considered a constant rather than a variable. Moreover, interest in social stratification has been marginal, as social movements have been considered increasingly as post-materialist. Only recently has attention re-emerged for the structural cleavages that might be mobilized during conflicts (della Porta 2013). In research on contentious politics, the type of economic arrangement has been considered relevant in defining specific paths of democratization. In this direction, Elisabeth Wood (2000) pointed to the intertwining of political and economic power in cases of oligarchic regimes in which democratization often involved (at least partially) violent insurgencies. In research on civil wars, grievances related to economic conditions were initially at the centre of attention, especially those operationalized as poverty, inequality, and ethnic discrimination. In fact, social polarization and horizontal social inequality emerged as positively correlated with the onset of the conflict. Economic polarization adds up to ethnic polarization in increasing the risk of violence. We have suggested here, as well, that violence should be considered as affecting economic conditions as much as it is affected by them. Hence, grievances related to inequality can themselves be produced through the act of mobilization, as political destabilization triggers spirals of impoverishment.
In Libya, the economic recession of the mid-1970s, with the related economic liberalization and privatization of state companies, had already affected the state capacity for patronage, increasing the power and revenues of a few protected individuals (Perthes 1992, 124) while social inequalities increased. Since the late 1990s, neoliberal and administrative reforms have unsettled previous social structures. Qaddafi’s attempt to normalize relations with the United States and the European Union brought about some, although very limited, liberalization, but also the dismissal of more than a third of those employed in the public administration. In reaction, protesting the delayed completion of housing projects in the coastal regions, people squatted in several hundred uninhabited, half-finished buildings. Local demonstrations also targeted inefficient distribution of government services. Corruption apparently also involved the protection of drug cartels, whose militias were tolerated (Salih 2014). In a country with nearly 70 per cent unemployment among young people, unemployed young men constituted the largest part of the protestors on the streets. In fact, concerns initially addressed material needs. As in Tunisia and Egypt, calls for dignity, freedom, and justice were frequent, sometimes accompanied by claims for more job opportunities, improved economic conditions, and better education (Michael 2011). Civil war clearly contributed to further economic difficulties by increasing social insecurity, with heavy consequences for an already deprived economy.
Similarly, in Syria, neoliberal reforms had brought about a withdrawal from the previous social contract between the regime and the population, with 30 per cent of the population below the poverty line and 11.4 per cent below indigence. Moreover, the neoliberal reforms had weakened the Ba’ath Party’s capacity for co-optation, resulting in a small political-economic elite reaping the benefits of economic liberalization, while traditional methods of state-led social support broke down. As the uprising erupted, protestors stigmatized figures like Rami Makhlouf, Assad’s cousin and one of the wealthiest people in Syria, for their corrupt business dealings. Even though many of the initial activists were more highly educated Syrians from the middle classes, this spread the image of a disenfranchised mass opposing a regime-linked economic elite.
In Yemen, as well, protests in early January 2011 started as calls for the end of corruption and for better salaries. Concentration of power and money in the hands of the regime’s elite seemed all the more outrageous, given the impoverishment of many citizens. Protests about material claims and labour rights had been indeed frequent in the country. When hopes for better living conditions and policies focused on social justice did not materialize during the transitional period, discontent grew. Attuned to the anger of people, in the summer of 2014, the Houthis returned to protests when President Hadi announced cuts in fuel subsidies. Once protests turned into armed clashes and the Houthis took over the capital, the country started its descent into civil war. With the escalation of violence, particularly after Saudi Arabia’s involvement, destitution in the poorest Arab country spiralled, fuelling further radicalization but also pushing children and men to join the armed struggle for the pursuit of financial rewards.
In Yugoslavia, the specific version of a socialist economy – with tolerance for small private business and a ‘self-management’ system that gave workers control of their factories and workplaces by electing their management – initially seemed to benefit the country’s development. The situation started to deteriorate, however, in the 1970s, as the oil crisis helped put an abrupt end to the so-called Yugoslav miracle (Stokes 1993, 229), forcing the country to adopt structural adjustment policies within various austerity programmes in order to repay the large accumulated foreign debt (Schaeffer 2000, 51; Stokes 1993, 229–30). Increases in unemployment and prices fuelled discontent among the population (Stokes 1993, 238–41), and workers’ strikes multiplied in the 1980s (Vladisavljevic ´ 2008, 111). Territorial inequalities also became increasingly dramatic between the relatively rich Slovenia and Croatia, on the one hand, and the much poorer Kosovo and Macedonia, on the other. In this context, the civil war further contributed to impoverishment and inequalities, with a few enriching themselves at the expense of the large majority of the population.
In sum, albeit to different degrees in the different areas, social inequality played a role by spreading grievances that fuelled protest. These critiques involved inequalities among social groups, with the development of kleptocratic regimes, particularly in the Middle East cases, where neoliberal reforms jeopardized previous patronage but also increased the richness of the corrupt few at the expense of the many. In Yugoslavia, while the inequalities were comparatively smaller, socioeconomic differences among the republics fuelled discontent, especially as they interacted with declining socioeconomic standards and increasing political instability. In all cases, the civil war further increased poverty and inequality, with small groups able to enrich themselves in the war economy, often in the informal or even criminal markets.
Social movement studies that focused on the dynamics of mobilization have systematically addressed the importance of networks. In fact, the format of these networks but also their content have been studied as generating relevant consequences for social movements. Research on political violence has pointed to the creation of military bodies for self-defence within social movement organizations as a tipping point in the emergence of clandestine political organizations (della Porta 2013). Research on civil wars but also on nonviolence has looked especially at the availability of armaments as triggering conflict militarization. In fact, insurgency has been defined as a specific technology of military conflict, with small and lightly armed bands using guerrilla strategies especially in rural areas (Kalyvas 2008). In particular, research has singled out the role of the military – or, often, multiple military bodies – in terms of strength, cohesion, and loyalty.
In the literature on nonviolent revolutions, Nepstad (2011) noted that while those that win the support of the military tend to succeed, when the military is composed of different ethnic groups, endowed with varying power, it is difficult for the army as a whole to side with the opposition. Rather, there will be defectors, which can cause violent developments. In civil war research, with its emphasis on individual rationalities, scholars have focused on military networks and the reasons to participate in them. In particular, attention has been paid to the agents who do the killing (Mitchell 2004) and to pre-existing networks (Mampilly 2011). The specific dynamics within rebel groups also emerge as crucial in our analysis. As violence generates violence, the armed forces split between those loyal to the regime and those who oppose it. Those who abandon the regime bring with them arms and military skills, which heavily affect the development of the conflicts, although with varying outcomes. As guns become audible, the civil society is silenced. Foreign military intervention challenges the dictators, but also cost many lives among the civil population.
In Libya, the conflict had moved into a pre-civil war phase by March 2011, assisted by an external military intervention. In the distribution of arms and equipment to the army, the dictator had privileged the regions inhabited by loyal tribes and, mistrusting the army, financed his own militia (Khosrokhavar 2012). The regime thus developed an apparently loyal security system: ‘The most important characteristic of these security organizations is that they are neither subject to institutional political control nor to control by the public but have been controlled exclusively by the Revolutionary Leadership led by Colonel Qadhafi’ (Cerone 2011, 790). By the end of the 1990s, a special elite-military the ‘People’s Guard’, composed of select loyalists, was formed with the task of protecting Qaddafi’s family and the regime infrastructure, especially from ‘internal enemies’ (Mattes 2004). During the protest wave, those who defected from the regime contributed the military leadership of the opposition. The rebels also accumulated arms through raids of local security forces’ headquarters. As the brutality of the fighting increased, the UN Security Council approved a resolution imposing a no-fly zone over Libya; soon thereafter, US and French warships in the Mediterranean started bombing Qaddafi’s security forces on the outskirts of Benghazi, allowing the rebels to rearm, regroup, and advance towards the west. The opposition then received aid in terms of communications equipment, training, and above all military support from a substantial number of countries and NATO. NATO’s aerial forces helped the rebels advance to the east, culminating in the fall of Tripoli on 20 August and the killing of Mu’ammar Qaddafi and one of his sons in Sirte on 20 October.
Similarly in Syria, defecting soldiers, especially among Sunni conscripts, brought armaments to the opposition while hundreds of soldiers were in fact executed for refusing to fire on the population. At the same time, notwithstanding some regime concessions, continued repression inhibited negotiations between the opposition and the regime, and protestors increasingly started to believe that armed protection of nonviolent demonstrations was both justified and necessary in the face of deadly repression. Despite the initial understanding that violence by the protestors was to be avoided as it would fuel violence by the regime, violent defensive reactions to violent regime repression started to spread. On 6 June 2011, at Jisr as-Shourough, one of the first liberations of a town through local residents’ ambush of the army and government buildings apparently took place. Notwithstanding heavy army intervention, which allowed the regime to retake control, the example was followed in other cities. The regime, on the other hand, felt justified in using lethal force (including airstrikes and artillery) against what it now, increasingly justifiably, described as an armed insurrection. The violence then became more organized, as Syrian soldiers started to defect. While defecting soldiers never endangered the stability of the Syrian army itself, they did bring military expertise and a belief in the possibility of armed opposition to the uprising.
A turning point in the activation of military networks’ spiral of escalation was in fact, on 29 July 2011, the foundation of the Free Syrian Army (FSA), which aimed at promoting desertion from the army and protecting civilians from repression. Though initially intended to become an opposition alternative to the regular Syrian army, it never lived up to its expectations. Initially, militias formed in various parts of the country based on friendships, clan allegiances, and neighbourhood solidarity. As militias became better structured and trained, larger rebel organizations emerged: Liwa al-Tawhid and Nour al-Din al-Zenki are two examples. Jihadists, including Arabs from other countries in the region, also joined militias; al-Qaeda formed their own group, and so the list went on. Many of these groups were largely autonomous, or completely independent, from the FSA. Predatory practices, criminal activities, and warlordism spread, as an activist noted, ‘Now everyone in Syria is armed, and weapons bring out the worst in people . . . Weapons and power are addictive’ (Hinnebusch, Imady, and Zintl 2016, 243).
In Yemen, as in Syria and Libya, an uprising that was designed as peaceful and popular quickly gained insurgent dimensions. After the March 2011 massacre of over 50 protestors in Sana’a, core members of the regime started to defect, bringing military units, skills, and equipment to the side of the opposition. Their presence, although first welcomed by the protestors, soon changed the course of the revolution, as popular protests were overshadowed by armed confrontations between Saleh and regime defectors. Four years later, when the transitional period reached a dead end, the contest between the Houthis and President Hadi was accompanied by the emergence of militarized groups on both sides. The Houthis, now allied with Saleh, relied on army units and tribes loyal to the former president, while the groups of the anti-Houthi camp received military assistance from Saudi Arabia. With the escalating violence and abundance of weapons, many previously peaceful protestors, like members of the southern Hirak movement, were transformed into soldiers in the course of the war.
In Yugoslavia, military networks were eventually activated as a result of rising uncertainty. As nationalist Croatian politicians began to discriminate against ethnic Serbs living within the borders of that country in the early 1990s, the minority Serbs perceived themselves under siege. When the spectre of Croatian independence became a reality in June 1991, the worst fears of the Serbian minority seemed to have been realized. Unable to rely on ‘their’ government for protection against regional militias, the Serbs armed themselves and soon received the backing of the Yugoslav army – by now controlled by politicians, mainly Serbian by nationality, defiant about the prospect of the country’s break-up. However, faced with the disintegration of the Yugoslav army, the civil war was fought mainly by militias. In Croatia and Bosnia,
[conflicts] were spawned not so much by the convulsive surging of ancient hatreds or by frenzies whipped up by demagogic politicians and the media as by the administrations of small – sometimes very small – bands of opportunistic marauders recruited by political leaders and operating under their general guidance. Many of these participants were drawn from street gangs or from bands of soccer hooligans. Others were criminals specifically released from prison for the purpose. Their participation was required because the Yugoslav army, despite years of supposedly influential nationalist propaganda and centuries of supposedly pent-up ethnic hatreds, substantially disintegrated early in the war and refused to fight.
(Mueller 2000, 42)
While agreement on separatism remained weak in the population, support was manipulated through selective incentives (including food and liquor) as high rates of desertion from the armed forces served to arm the locals, who formed paramilitary units. Paramilitary gangs and foreign mercenaries, among them convicted criminals, were responsible for some of the most violent episodes during the Yugoslav wars.
Thus, we can conclude from the case studies that the activation of military networks is an indispensable part of the transition to civil war. At the same time, we were able to observe that in each case this mechanism intersected differently with other mechanisms constituting the overall process towards civil war. In Libya, the initial militias were tribally based and quickly turned to offensive as an international coalition provided support. In Syria, militias emerged around much smaller units but eventually institutionalized along ethnic and sectarian boundaries, weakening the overall military capabilities of the opposition. In Yemen, with the splits in the military and the sponsorship of Saudi Arabia, armed militias emerged on both sides of the conflict. Ironically, the rebels relied on the army men of Saleh, while the anti-Houthi camp was composed of various groups armed by Saudi Arabia. Yugoslav militias were often just as fragmented as in Libya and Syria, but would align along nationalist defined identities.
Social movement studies have sometimes pointed at the vicious circles that emerge during protest cycles, bringing about ideological and strategic transformation in social movement repertoires through the activation of some specific emotional tensions. In particular, motives and feelings change in action, during interactions with allies and opponents. Escalation might happen in spirals of reciprocal revenge, which bring about an intensification of emotions (della Porta 2013). Approaches to civil war also stressed these dynamics, pointing in particular to the combination of various types of emotions in intense time as enhancing the propensity to violence. Once again, while focusing especially on the onset of civil wars, our case studies pointed at its persistence through aroused emotions, with spirals of reciprocal revenge.
In Libya, about 300–400 people were killed in the first half of February (Schemm 2011), while violent clashes spread quickly. The killing of protestors fuelled a narrative of revenge. At the same time, however, there was also a quick development of NGOs to support and protect the communities. In March, a Coalition of Libyan NGOs was created, grouping 240 NGOs (Quesnay 2013, 124). With the spread of the armed conflicts, brutality increased on the side of the regimes, but also in the emerging militias. Even after the end of the regime, violence remained high, as episodes of brutality among different groups of the population spiralled into reciprocal revenge.
In Syria, too, the regime continued to use live ammunition, although it was still unable to suppress the opposition. The situation thus escalated, notwithstanding LCCS calls for peaceful protest. At the start of Ramadan of 2011, after about five months of mobilization, 2,000 had been killed, with another thousand slain by the end of the month. Violence then increased on the part of the protestors as well, initially mainly in little-organized attacks against regime officials, security services, army personnel, and the Ba’ath Party buildings. In fact, the situation became increasingly difficult to address, given the spiral of reciprocal brutalization, as:
Red lines around the use of particular weapons systems were overstepped one by one: a spiral of violence leading from bullets to bombs and from snipers to sarin was set in motion. As civilians on both sides were targeted, the security dilemma gave the conflict an additional dimension: each side, to feel secure, created armed groups that made all less secure. Proponents of nonviolent struggle were not only marginalized but also, like vast parts of the initially uninvolved civilian population, driven out of their homes, injured or killed, with many going into exile to escape the violence, leaving the field to the armed factions.
(Hinnebusch, Imady, and Zintl 2016, 235–6)
On 18 July 2012, the FSA and the Jihadi group Liwa al-Islam (Brigade of Islam) claimed responsibility for the killing of Asef Shawkat (the Minister of Defence) and three other elite security leaders. This would prove to be the signal the opposition needed to start an armed assault on Damascus and Aleppo that proved to be the turning point in the transformation of a peaceful Syrian uprising to a civil war. In mid-June 2012, more than 8,000 people had already died; by 18 July the civil war had claimed 20,000 victims and twice as many in November of the same year. As the opposition became increasingly fragmented, brutality (and ensuing fears as well as revenge) also spiralled among its ranks.
In Yemen, the feelings of retribution have been a particularly important force in strengthening divisions and leading to the realignments that changed the course of the popular mobilization. During the transitional period, which was meant to lead to Yemen’s democratization, the Houthis and the ousted President Saleh united against the common political enemy – Hadi and al-Islah – in what many Yemenis saw as an ‘alliance of revenge’. Their rapprochement annulled the clearly demarcated camps that characterized the popular uprising (opposition versus the regime) and put forward new divisions and coalitions built around the warring elite factions. Later, once the violence radicalized and spread – particularly after the Arab coalition airstrikes – sentiments of retaliation intensified, with the Houthi and anti-Houthi camps accusing each other of war crimes.
In collapsing Yugoslavia, insecurity was fuelled by historically grounded expectations about bad intentions. Croats and Serbs escalated demands on one another, as Croats denied Serbs equal status and Serbians reacted belligerently. Similarly, Kosovo Albanians reacted to Serbian provocations. So, as in these examples, groups negotiate less and less, reaching fewer and fewer agreements and instead escalating their demands. In this process, groups might become more secretive about their military plans in emerging anarchy, which can also cause them to underestimate the adversary’s strength (Rose 2000). As Kaldor (2006) noted, in Bosnia-Herzegovina, the most ethnically mixed state, ethnic cleansing was performed by militias armed from outside by Milošević and Tuđman. The area was characterized by both tolerance and fear, in a most militarized country with paramilitary, foreign mercenaries, and local police. With the formal economic collapse, the siege of Bosnian cities followed a typical path:
First, the regular forces would shell the area and issue frightening propaganda so as to instill a mood of panic. Then the paramilitary forces would close in and terrorize the non-Serb residents with random killing, rape and looting. Control over local administration would then be established. In the more extreme cases, non-Serb men were separated from the women and killed or taken to detention centres. Women were robbed and/or raped and allowed to go or taken to rape detention centres. Houses and cultural buildings such as mosques were looted, burned or blown up. The paramilitary groups also seemed to have lists of prominent people – community leaders, intellectuals, SDA members, wealthy people – who were separated from the rest and executed.
(ibid., 54)
Similar patterns developed in areas controlled by the Croatians. Paramilitaries with economic motivations were hired to do the dirty jobs, and ethnic cleansing followed the logic of guerrillas, who rely on support by the population.
In sum, civil wars develop upon emotional mechanisms that are linked, in particular, to spirals of revenge. As fear spread in all our cases, based on feelings of increasing insecurity, people within the different groups became outraged by the deeds and narratives of extreme brutality against fellow group members. Spirals of reciprocal revenge caused that brutality to escalate.
Social movement studies and civil war studies have long shared a focus on (often distant) causes. More recently, in both areas, researchers have noted that motivations as well as conditions emerge in action – they are that is, at least to a certain extent, endogenous and processual. Social movement studies have pointed to identification processes as a relevant activity of social movement organizations. Promoting collective action implies first of all a process of identification with the others. That is, individual identities need to be transformed into collective identities and then politicized. Also in studies on civil wars, a constructivist approach contrasted primordialist visions of ethnicity. In fact, the very concept of ethnic conflict has been criticized as addressing phenomena that are too heterogeneous (Gilley 2004). In our cases, religious and ethnic narratives of enmities were, in fact, mobilized, fuelling the continuation of civil wars.
In Libya, we find a different type of religiously fuelled tension from the other cases, as the population is overwhelmingly Sunni. The rising popularity and public expressions of Salafism, a strict and often militant form of Sunni Islam, had caused tensions between Salafis and Sufis. In the immediate aftermath of the regime collapse there have been numerous instances of the destruction of Sufi shrines, all of which indicate a rise in sectarian violence (UNHR 2012). It can also be argued that with the emergence of ISIS in 2014, intra-Sunni sectarian tensions have intensified.
In Syria, even though Friday prayers were the occasion for people to gather, the claims were initially overwhelmingly non-religious, targeting the (power of) the security services and corruption, and calling for political liberalization. Later, sectarian rhetoric increased as the conflict radicalized. With the passing of time, ‘The uprising has taken an increasingly sectarian tone, pitting the Sunni majority against the Alawi minority, and has shown the lengths to which the regime and its supporters will go to defend their own survival’ (Donker 2012). Sectarian-based attacks became more frequent among opposition and regime-linked militias alike. In combination with the vacuum produced by state failure in both Iraq and Syria, with the spreading of a war economy and plundering of state resources, this provided the context in which IS could emerge. By mid-2015, 4 million out of 21 million inhabitants had left the country. So, ‘the processes that led to and fostered sectarian strife fed into an apparent drive for establishment of “sectarian-cleansed” regions secure from the threat of the “other,” raising the possibility of post-war fragmentation of Syrian territory along confessional lines’ (Hinnebusch, Imady, and Zintl 2016, 239). This was all the more the case as Sunni Islamists consolidated control of the borders with Sunni-dominated Iraq, Jordan, and Turkey, thus ‘raising the specter of state boundaries being redrawn along sectarian and/or ethnic lines. Thus, militarization and sectarianization proved to be mutually reinforcing’ (ibid., 239). Similarly, narratives of oppression and resistance later supported the institution of a Kurdish state at the northern border.
In Yemen, where national loyalties had traditionally been weaker than tribal or regional allegiances, the early stages of popular mobilization brought surprising levels of unity, with divergent actors emphasizing the pan-Yemeni character of the uprising (Lackner 2016; Yadav 2015). Gradually, however, old divisions came back into play, with many groups – such as the Houthis, tribes, or southern secessionists – turning to their narrow identifications and localized interests. Once the competition between the Houthis/Saleh and Hadi/Islah took the shape of a war, both sides resorted to a sectarian rhetoric, which was previously unusual in Yemeni politics. Sectarianism has enabled all parts of the conflict to recruit fighters, delegitimize enemies, and justify violence. With the divisions exacerbated by sectarian hatred, the calls for revenge spiralled even further.
In Yugoslavia, historical conflicts between Serbs and Croats – as well as (in part overlapping) – between Catholic, Protestant, and Muslim groups – represented historical memories, sleeping but still there to be revived. During the conflict, exclusive nationalist narratives were mobilized. Serbs were praised as the brave soldiers who had protected Austria from the Turks. Even the Chetniks, who had collaborated with the Nazis, were seen in a positive light, as freedom lovers. In contrast, the Croats were the murderers, traitors, and Serb-haters, as were (even if less central) the Turks and the Austro-Hungarians (Maleševic ´ 2002). The mentioned variety of autonomous civil society organizations notwithstanding, nationalist groups sponsored from above came to dominate the public sphere by the end of the 1980s. This was the case especially in Serbia, where an aggressive framing about ‘great Serbia’ developed, activating past memories. Similarly, in Croatia, former party members came to lead the nationalist movement. Feeling threatened by Miloševic´’s nationalist rhetoric, they tried to increase popular support by linking up with nationalist dissidents (Pavkovic ´ 2000, 112).
Using nationalist propaganda that stressed the past and present victimization of the Serbs, Milošević was able to exploit the support for the Serb minority in Kosovo, which had been protesting since 1985 against the perceived discrimination by Kosovo’s Albanian majority. Initially, Serbian nationalism pushed for defensive frames. The relations between Kosovo Albanians and Kosovo Serbs then degenerated, as the Serbs staged rallies in Belgrade: ‘The organisers of these protests were Serb and Montenegrin farmers, skilled workers, teachers and low-ranking communist officials. This gave the movement the look of an anti-elite, grassroots movement of harassed Serb and Montenegrin minorities in Kosovo’ (ibid., 83). As a turning point, on the night and early morning of 24 and 25 April 1987, Milošević addressed a gathering of 15,000 Kosovo Serbs and Montenegrins with inflammatory words. As other republican leaders started to fear Milošević as a proponent of a Greater Serbia, they reacted by mobilizing for their own sovereignty. In this way, ‘nationalism became a dominant political force largely as an unintended outcome of high levels of mobilization and spiralling social, economic and political conflicts in a complex, authoritarian multi-national state which experienced a severe economic crisis’ (Vladisavljević 2008, 6).
In sum, in our four cases, emotional and cognitive mechanisms interacted in sustaining civil wars. Exclusive forms of group identification developed during intense interactions that contributed to reactivate and fuel narratives of historical animosity. A Manichean rhetoric was used by leaders to strengthen symbolic incentives towards mobilization against the (ethnic, religious, national) other. Identities emerged as based on ethnicity or religion in all four cases, with tribal identification politicized in the Middle Eastern examples.
The aim of this volume has been to develop some theoretically driven analysis of cases of escalation from peaceful social movements for democracy into civil wars. To this end, building upon studies of other forms of political violence, we have identified some mechanisms that we singled out in part deductively – bridging the analysis of civil wars with social science literature on social movements, nonviolent revolutions, and democratization – but also in part inductively, from some studies on critical cases, looking at the presence and different combinations of these mechanisms at the onset of civil wars and in their development. Our main message is that, beyond root causes, understanding when and how civil wars emerge requires an analysis of the emergent characteristics of violence, as it triggers specific dynamics and constellations of mechanisms that produce and reproduce the conditions for its own survival.
Admittedly, our aim is to contribute to theory building, rather than to test existing theories. Indeed, we have identified the way in which a set of mechanisms emerge and relate, rather than considering them as either necessary or sufficient causes. Additionally, we used some process-tracing analysis of what we considered as relevant cases. This approach resulted in mapping the ways in which a variety of mechanisms can interrelate in constituting a transformative process of mobilization for democracy into civil war.
While focusing on some recent moments of failed attempts at democratization during the Arab Spring, we added an historical case in another area in order to reflect on the robustness of our mechanisms. But our sample was far from representative: much more systematic analysis is necessary to investigate the capacity of the selected mechanisms to travel in space and in time. Finally, while we have elsewhere analysed the involvement of civil society in more successful democratization processes (della Porta 2014; 2016), a systematic comparison of positive and negative cases remains to be undertaken.
Alley, April Longley. 2010. ‘The rules of the game: unpacking patronage politics in Yemen’. The Middle East Journal 64(3): 385–409. doi:10.3751/64.3.13.
Alwazir, Atiaf. 2016. ‘Yemen’s enduring resistance: youth between politics and informal mobilization’. Mediterranean Politics 21(1): 170–91. doi:10.1080/13629395.2015.1081446.
Anderson, Lisa. 2011. ‘Demystifying the Arab Spring: parsing the differences between Tunisia, Egypt, and Libya’. Foreign Affairs (May/June).
Benderly, Jill. 1997. ‘Feminist movements in Yugoslavia, 1978–1992’. In M.K. Bokovoy, J.A. Irvine, and C.S. Lilly (eds), State-Society Relations in Yugoslavia, 1945–1992. New York: St. Martin’s Press, pp. 183–210.
Bilic, Bojan. 2012. We Were Gasping for Air. Baden-Baden: Nomos.
Boudreau, Vincent. 2004. Resisting Dictatorship: Repression and Protest in Southeast Asia. Cambridge: Cambridge University Press.
Buhaug, Halvard, Gates, Scott, and Lujala, Päivi. 2013. ‘Geography, rebel capability, and the duration of civil conflict’. Journal of Confl ict Resolution 53(4): 544–69. doi:10.1177/0022002709336457.
Bunce, Valerie. 1999. Subversive Institutions: The Design and the Destruction of Socialism and the State. Cambridge: Cambridge University Press.
Carapico, Sheila. 2007. Civil Society in Yemen: The Political Economy of Activism in Modern Arabia. Cambridge: Cambridge University Press.
Carapico, Sheila. 2011. ‘No exit: Yemen’s existential crisis’. Middle East Research and Information Project, May 3. Available at:www.merip.org/mero/mero050311–1?ip_login_no_cache=5d2267fcaf95daf21e7829175eadb779
Cerone, John. 2011. ‘Documents on Libya, introductory note’. International Legal Materials 505.
Collier, Paul, and Hoeffler, Anke. 2004. ‘Greed and Grievance in Civil War’. Oxford Economic Papers 56(4): 563–95. doi:10.1093/oep/gpf064.
Crenshaw, Martha. 2011. Explaining Terrorism: Causes, Processes and Consequences. London: Routledge.
Davenport, Christian (ed.). 2000. Paths to State Repression: Human Rights Violations and Contentious Politics. Boulder, CO: Rowman & Littlefield.
Davenport, Christian. 2005. ‘Repression and mobilization: Insights from political science and sociology’. In Christian Davenport, Hank Johnston, and Carol Mueller (eds), Repression and Mobilization: Social Movements, Protest, and Contention. Minneapolis, MN: The University of Minnesota Press.
della Porta, Donatella. 1998. ‘The political discourse on protest policing’. In Mario Giugni, Doug McAdam, and Charles Tilly (eds), How Movements Matter. Minneapolis, MN: The University of Minnesota Press.
della Porta, Donatella. 2013. Clandestine Political Violence. Cambridge: Cambridge University Press. http://dx.doi.org/10.1017/CBO9781139043144.
della Porta, Donatella. 2014. Mobilizing for Democracy: Comparing 1989 and 2011. Oxford: Oxford University Press.
della Porta, Donatella. 2016. Where Did the Revolution Go?, Cambridge: Cambridge University Press.
della Porta, Donatella, and Diani, Mario (2006) Social Movements: An Introduction. Oxford: Blackwell.
della Porta, Donatella, and Fillieule, Olivier. 2004. ‘Policing social movements’. In David A. Snow, Sarah A. Soule, and Hanspeter Kriesi (eds), The Blackwell Companion to Social Movements. Oxford: Blackwell, pp. 217–41.
della Porta, Donatella, and Gbikpi, Bernard. 2012. ‘The riots: A dynamic view’. In Seraphim Seferiades and Hank Johnston (eds), Violent Protest, Contentious Politics and the Neo-liberal State. Farnham: Ashgate, pp. 87–102.
della Porta, Donatella, and Piazza, Gianni. 2008. Voices of the Valley. Voices of the Straits: How Protest Creates Community. New York: Berghahn.
Donker, Teije H. 2012. Mobilizing for Democracy in Syria. Available at: http://cosmos.sns.it.
Durac, Vincent. 2013. ‘Protest movements and political change: an analysis of the “Arab Uprisings” of 2011’. Journal of Contemporary African Studies 31(2): 175–93. doi:10.1080/ 02589001.2013.783754.
Earl, Jennifer. 2003. ‘Tanks, tear gas and taxes’. Sociological Theory 21: 44–68.
Earl, Jennifer, Soule, Sarah A., and McCarthy, John. 2003. ‘Protest under fire? Explaining protest policing’. American Sociological Review 69: 581–606.
Engene, Jan Oscar. 2004. Terrorism in Western Europe. Cheltenham: Edward Elgar.
Fattah, Khaled. 2011. ‘Yemen: A social Intifada in a republic of sheikhs’. Middle East Policy 18 (3): 79–85. doi:10.1111/j.1475–4967.2011.00499.x.
Figa, Jozef. 1997. ‘Socializing the state: civil society and democratization from below in Slovenia’. In M.K. Bokovoy, J.A. Irvine, and C.S. Lilly (eds), State-Society Relations in Yugoslavia, 1945–1992. New York: St. Martin’s Press, pp. 163–82.
Francisco, Ronald A. 2005. ‘The dictator’s dilemma’ . In Christian Davenport, Hank Johnston, and Carol Mueller (eds), Repression and Mobilization. Minneapolis, MN: The University of Minnesota Press.
Gelvin, James L. 2011. Divided Loyalties: Nationalism and Mass Politics in Syria at the Close of Empire. 3rd edn. Berkeley, CA: University of California Press.
Gilley, Bruce. 2004. ‘Against the concept of ethnic conflict’. Third World Quarterly 25(6): 1155–66. doi:10.1080/0143659042000256959.
Goodwin, Jeff. 2001. No Other Way Out: States and Revolutionary Movements, 1945–1991. Cambridge: Cambridge University Press.
Hinnebusch, Raymond A. 1993. ‘State and civil society in Syria’. The Middle East Journal 47 (2): 243–57.
Hinnebusch, Raymond, Imady, Omar, and Zintl, Tina. 2016. ‘Civil resistance in the Syrian uprising: from peaceful protest to sectarian civil war’. In Adam Roberts, Michael Willis, and Timothy Garton Ash (eds), Civil Resistance in the Arab Spring: Triumphs and Disasters. Oxford: Oxford University Press, pp. 223–47.
HRW. 2013a. ‘Unpunished massacre: Yemen’s failed response to the “Friday of Dignity” killings’. Available at: www.hrw.org/report/2013/02/12/unpunished-massacre/yemens-failed-response-friday-dignity-killings
HRW. 2013b. ‘Between a drone and Al-Qaeda. The civilian cost of US targeted killings in Yemen’. 22 October. New York: Human Rights Watch. Available at:www.hrw.org/report/2013/10/22/between-drone-and-al-qaeda/civilian-cost-us-targeted-killings-yemen
Jacinto, Leela. 2011. ‘Benghazi’s Tahrir Square: Times Square style meets revolutionary zeal’. France24, 25 April. Available at: www.france24.com/en/20110425-libya-benghazi-tahrir-square-times-reporters-notebook-leela-jacinto
Joffé, George. 2016. ‘Civil resistance in Libya during the Arab Spring’. In Adam Roberts, Michael Willis, and Timothy Garton Ash (eds), Civil Resistance in the Arab Spring: Triumphs and Disasters. Oxford: Oxford University Press, pp. 116–40.
Kaldor, Mary. 2006. New and Old Wars. Cambridge: Polity.
Kalyvas, S.N. 2008. ‘Promises and pitfalls of an emerging research program: the micrody-namics of civil war’. In Stathis N. Kalyvas, Ian Shapiro, and Tarek Masoud (eds), Order, Conflict, and Violence. Cambridge: Cambridge University Press, pp. 397–421. Available at: http://ebooks.cambridge.org/ref/id/CBO9780511755903A028
Khosrokhavar, Farhad. 2012. The New Arab Revolutions That Shook the World. Boulder, CO: Paradigm Publishers.
Lackner, Helen. 2016. ‘The change squares of Yemen’. In Adam Roberts, Michael J. Willis, Rory McCarthy, and Timothy Garton Ash (eds), Civil Resistance in the Arab Spring, pp. 141–68. Oxford: Oxford University Press. Available at: www.oxfordscholarship.com/view/10.1093/acprof:oso/9780198749028.001.0001/acprof-9780198749028-chapter-6
Licht, Sonja. 2000. ‘Civil society, democracy, and the Yugoslav wars’, in M. Spencer (ed.), The Lesson of Yugoslavia. New York: Elsevier, pp. 111–24.
Maleševic ´, Siniša. 2002. Ideology, Legitimacy and the New State: Yugoslavia, Serbia and Croatia. London: Frank Cass.
Mampilly, Zachariah Cherian. 2011. Rebel Rulers: Insurgent Governance and Civilian Life during War. Ithaca, NY: Cornell University Press.
Manea, Elham. 2015. ‘Yemen’s Arab Spring: outsmarting the cunning state?’ In Larbi Sadiki (ed), Routledge Handbook of the Arab Spring: Rethinking Democratization. New York: Routledge, pp. 160–72.
Mattes, Hanspeter. 2004. ‘Challenges to security sector governance in the Middle East: the Libyan case’. Paper presented at Geneva Centre for the Democratic Control of Armed Forces (DCAF) as a part of Security Governance in the Mediterranean Project, 13 July.
Michael, Maggie. 2011. ‘Protesters in Libya demand Gaddafi ouster and reforms’. The Washington Post, 17 February. Available at: www.washingtonpost.com/wp-dyn/content/article/2011/02/16/AR2011021607292.html
Mitchell, Neil J. 2004. Agents of Atrocity: Leaders, Followers, and the Violation of Human Rights in Civil War. New York: Palgrave Macmillan.
Mueller, John. 2000. ‘The banality of “ethnic war”’. International Security, 25(1): 42–70.
Nepstad, Sharon Erickson. 2011. Nonviolent Revolutions: Civil Resistance in the Late 20th Century. Oxford: Oxford University Press.
Pavkovic ´, Aleksandar. 2000. The Fragmentation of Yugoslavia: Nationalism and War in the Balkans. 2nd edn. Houndmills: Macmillan.
Perthes, Volker. 1992. ‘The Syrian private industrial and commercial sectors and the state’. International Journal of Middle East Studies 24(2): 207–30.
Quesnay, Arthur 2013. ‘L’insurrection libyenne, un movement révolutionnaire decentralize’, in Amin Allal and Thomas Pierret (eds), Au cœur des révolte arabes. Devenir revolutionaries. Paris: Armand Colin, pp. 113–32.
Regan, Patrick, and Norton, Daniel. 2005. ‘Greed, grievance, and mobilization in civil war’. Journal of Conflict Resolution 49(3): 319–36.
Rose, William. 2000. ‘The security dilemma and ethnic conflict: some new hypotheses’. Security Studies 9(4): 1–51. doi:10.1080/09636410008429412.
Salih, Yassin Al-Haj. 2014. ‘The Syrian Shabiha and their state – statehood & participation’. Heinrich Böll Stiftung. Available at: https://lb.boell.org/en/2014/03/03/syrian-shabiha-and-their-state-statehood-participation
Salisbury, Peter. 2016. ‘Yemen: stemming the rise of a chaos state’. Middle East and North Africa Programme. Chatham House. Available at: www.chathamhouse.org/publication/yemen-stemming-rise-chaos-state
Schaeffer, Robert K. 2000. ‘Democratization, division and war in Yugoslavia: a comparative perspective’. In M. Spencer (ed.), The Lesson of Yugoslavia. New York: Elsevier, pp. 47–63.
Schemm, Paul. 2011. ‘Battle at army base broke Gadhafi hold in Benghazi’. The Washington Post, 25 February. Available at:www.washingtonpost.com/wp-dyn/content/article/2011/02/25/AR2011022505021.html
Schock, Kurt. 2005. ‘Unarmed insurrections: people power movements in nondemocracies’, in Social Movements, Protest, and Contention, vol. 22. Minneapolis, MN: University of Minnesota Press.
Stokes, Gale. 1993. The Walls Came Tumbling Down: The Collapse of Communism in Eastern Europe. New York: Oxford University Press.
Toft, Monica. 2003. The Geography of Ethnic Violence: Identity, Interests, and the Indivisibility of Territory. Princeton, NJ: Princeton University Press.
UNHR. 2012. ‘The destruction of cultural and religious sites: a violation of human rights’. UNHR Office of High Commissioner, 24 September. Available at: www.ohchr.org/EN/NewsEvents/Pages/DestructionShrines.aspx
Uysal, Ayshen. 2005. ‘Organisation du maintien de l’ordre et répression policière en Turquie’, in Donatella della Porta and Olivier Fillieule (eds), Maintien de l’ordre et police des foules. Paris: Presses de Science Po.
Vasilevski, Steven. 2007. ‘Diverging paths, diverging outcomes: a comparative analysis of post-communist transition in the successor states of Yugoslavia’. YCISS Post-Communist Studies Programme Research Paper Series. Toronto: York Centre for International and Security Studies.
Vladisavljević, Nebojša. 2008. Serbia’s Antibureaucratic Revolution: Milošević, the Fall of Communism and Nationalist Mobilization. Houndmills: Palgrave Macmillan.
Wedeen, Lisa. 2008. Peripheral Visions: Publics, Power, and Performance in Yemen. Chicago: University of Chicago Press.
Wood, Elisabeth Jean. 2000. Forging Democracy from below: Insurgent Transitions in South Africa and El Salvador. Cambridge: Cambridge University Press.
Yadav, Stacey Philbrick. 2015. ‘The “Yemen Model” as a failure of political imagination’. International Journal of Middle East Studies 47(1): 144–7. doi:10.1017/S0020743814001512.
Ylönen, Aleksi. 2005. ‘Grievances and the roots of insurgencies: Southern Sudan and Darfur’. Peace, Conflict and Development: An Interdisciplinary Journal 7: 99–134.