Social Cleavages and Armed Group Consolidation: The Case of Khalifa Haftar’s Libyan Arab Armed Forces

Abstract

In Libya’s fragmented military landscape, Khalifa Haftar’s Libyan Arab Armed Forces stands out as the only armed group that concentrated power over entire regions. The interplay of relative power and social cleavages conditioned Haftar’s choices of consolidation strategies. As Haftar grew stronger, he progressively moved from cooperative and competitive approaches toward coercion. Social cleavages wrought by conflict – as opposed to seemingly objective group boundaries – contributed to shaping Haftar’s choices between coercive and cooperative strategies. The significance of these cleavages depended on how violence had shaped local social cohesion and armed groups’ links with communities.

In 2014, Khalifa Haftar mobilized a loose alliance of armed groups formed from Libya’s fragmented military landscape. Over the following years, he warded off successive challenges from within his coalition and gradually centralized control. Haftar’s forces have to date been the only Libyan faction to consolidate authority over an entire region – eastern Libya – and become powerful enough to obtain a chance at seizing overall power. Despite being defeated in Tripoli in 2020, his Libyan Arab Armed Forces (LAAF) has remained cohesive in stark contrast to its divided adversaries.

Haftar formed a core of loyal units out of an initially fractious coalition. He used his military campaigns against his declared enemies as an instrument for consolidation among his allies. Within his alliance, a progressive shift from cooperation toward outcompeting his allies eventually resulted in his wielding of coercive measures against disgruntled former supporters. In this way, Haftar slowly transformed opportunistic support into a centralized power structure.

What caused Haftar to shift between coercive, cooperative and competitive approaches to consolidation? The argument presented here centers on the interplay of relative power with the social cleavages wrought by conflict. Where his coercive power was insufficient or where it proved difficult to mobilize allies against a common enemy, Haftar opted for cooperation, integrating local forces into the LAAF while granting them substantial autonomy. As foreign support strengthened his position, his approach to these same forces shifted toward competition. Haftar’s allies became dependent on his decisions to provide or withhold support; in parallel, he gained autonomy by building forces under the direct control of close relatives. Eventually, he was in a position to use coercion toward holdovers. Foreign backing, therefore, was a key component of Haftar’s consolidation of power.

But relative power alone cannot account for the constraints Haftar faced in his consolidation drive. Theories of militant infighting and alliance-building that stress the role of threats and opportunities emanating from changing power relations have tended to downplay the role of social cleavages, such as ethnic or sectarian boundaries. I argue that the social cleavages conditioning militant choices are those shaped by violent conflict, rather than seemingly objective boundaries such as those of tribes or ethnic groups. Conflict not only renders such divides more or less salient, it can also produce new cleavages and identities and reinforce or erode social cohesion among ingroups, thereby altering barriers to defection. Haftar helped draw a socio-political divide that allowed him to concentrate power among his allies. Later, he sought to redraw the divides in order to woo new allies, but these attempts met with mixed success, and ultimately collided with the lasting identities and loyalties formed during the 2011 war. Social divides wrought by violent conflict limit actors’ flexibility in reacting to changing power relations – particularly where armed groups are closely linked to their local social base.

This article first develops hypotheses on the role of relative power, social cleavages, and social cohesion in conditioning choices between different consolidation strategies. After briefly sketching the background of Libya’s military fragmentation, it then provides an account of Haftar’s rise, leveraging in-case comparisons to show how and why he alternated between different strategies to concentrate power. The article draws on research I conducted for a book on political and military fragmentation in Libya, which involved hundreds of interviews with actors and observers in the country since 2011.¹ It also makes extensive use of Arabic sources, including from local media and social media, as well as video documentation. Preparatory work conducted specifically for this article included a Working Paper that provides more details on actors and events during the LAAF’s evolution. The article refers to the Working Paper wherever its space constraints require greater brevity.²

Relative Power, Social Cleavages, and Social Cohesion in Consolidation Processes

Calculations of relative power are widely recognized as key to armed group behavior, especially in fragmented conflicts characterized by defections and infighting. As leaders of armed groups ponder whether alliance, defection or splintering could improve their survival chances, they react to changing threats and opportunities emanating from enemies and allies.³ Depending on their evolving power vis-à-vis rivals, commanders may see the necessity or opportunity to forcefully consolidate control within their own groups or their alliance. Opportunities consist in changes in relative power that decrease the costs of forcible consolidation; necessity beckons if a rival’s growing power threatens the primacy of another faction. The reasons for ongoing or anticipated changes in relative power can vary and include, for example, a recent or expected boost in foreign support; the temporary weakness of third parties in the conflict (such as the government); or recruitment patterns that over time favor some commanders over others.⁴ Several studies have found that fighting between rebel groups is more likely between factions of disparate strength, and less likely between comparably sized groups.⁵ We can therefore hypothesize that power asymmetries between rivals or reluctant allies will incentivize the stronger armed groups to consolidate power by force. The costlier such a forceful strategy is – that is, the more resistance to consolidation a rival can muster – the more likely factions are to cooperate with or outcompete their rivals in their quest for consolidation.

Studies of alliances, splits and infighting among armed groups have tended to emphasize their quest for survival and power, and to downplay the role of ethnic, sectarian and ideological identities.⁶ Approaches that do accord weight to social cleavages disagree over their impact on fragmentation and consolidation. Hafez, Gabbay, and Gade argue that a common identity and interconnecting social networks within ethnic, tribal or sectarian groups can facilitate consolidation by cooperation.⁷ The boundaries of such groups are barriers to defection; consolidation across such boundaries is therefore more likely to be coercive. By contrast, Pischedda asserts that co-ethnicity increases the likelihood of fighting between rebel groups, since they aspire to control the same constituencies and territories.⁸

I suggest that the analytical value of social cleavages lies in their political meaning in a conflict, and depends on their relation to social cohesion as well as to the social basis of armed groups. Objective boundaries between ethnic or tribal groups have little explanatory power; what matters are the boundaries that the conflict creates or renders salient.⁹ Contrary to influential theories of “ethnic conflict,” the transformative effects of violence go beyond the hardening of existing cleavages, and they do not necessarily reinforce the social cohesion of ingroups.¹⁰ Evidence from Libya supports arguments that emphasize the role of violence in rendering some identity cleavages salient but not others, and creating new ones.¹¹ Such cleavages truncate individuals’ social networks – often lastingly, due to forced displacement or the rise of actors with a vested interest in deepening the rifts.¹²

The tenacity of conflict-induced societal cleavages, and their effectiveness as barriers to defection, rest on the role of social cohesion, defined as the density and quality of social ties.¹³ Research on political violence is according growing attention to social cohesion, but mostly considers it as an independent variable that influences communities’ capacity to respond to violence.¹⁴ There has been little scrutiny of how violence alters social cohesion, and how cohesion relates to the social embeddedness of armed groups.¹⁵ Violent conflict that hardens social cleavages can reinforce or erode social cohesion among ingroups, depending on whether the struggle enjoys broad community support or causes rifts within communities.¹⁶ Social cohesion (or its weakness) can be observed empirically during fieldwork. While quantifiable data on the number and multiplex nature of social ties in a community is generally hard to come by, participant observation can offer insights into network density and quality, with the latter evident in the extent to which social ties sustain interpersonal trust and norms such as reciprocity and solidarity.¹⁷

As I demonstrated elsewhere, social cohesion in local communities makes political and military fragmentation within such communities more durable, since factions are linked to their local rivals by a dense network of social ties and cannot move against them ruthlessly.¹⁸ Social cohesion therefore counteracts the incentives for coercive consolidation among factions of the same ingroup highlighted by Pischedda. It can also thwart cooperative and competitive consolidation. Competing factions in cohesive communities may coordinate in the face of external threats, but will otherwise resist attempts at centralization.¹⁹ Armed groups that are less reliant on a local social base, and operate in less cohesive communities, will be more flexible in their responses to threats and opportunities. We may therefore hypothesize that variations of social cohesion and the social embeddedness of armed groups across different communities help explain successes and failures in consolidation.

Fragmentation, Social Cleavages, and Social Cohesion in Libya

Fragmentation has marked Libya’s political and military landscape since the demise of Muammar Qadhafi’s regime. By the end of the 2011 civil war, 236 revolutionary armed groups had formed in the coastal city of Misrata alone.²⁰ After the war, the revolutionary factions grew further and countless new armed groups formed across the country, as factional leaders nurtured them with state funding, and furnished official cover through newly created security institutions.²¹ Rivalries over the control of security institutions were a key driver behind the escalation into the second civil war (2014-15), during which state institutions split between eastern and western Libya. Governments in Tripoli have since continued to juggle multiple competing factions, while in the east, Khalifa Haftar progressively centralized authority over the coalition of armed groups he mobilized in 2014 – a process of consolidation that has to date remained the exception in Libya.

Even with Haftar’s gradual expansion, military fragmentation in Libya has been such that it is difficult to narrow down the number of armed groups even approximately. In Misrata, at least several dozen armed groups have continued to operate in post-2011 conflicts. In Tripoli, a process of consolidation reduced the plethora of post-revolutionary factions to around fifteen main militias by 2018, of which four dominated central Tripoli and its institutions.²² One study from 2018 identified 122 armed groups across Libya, but that list was far from comprehensive.²³ The UN Panel of Experts on Libya in 2019 counted 49 groups fighting for the GNA and 61 groups fighting for the LAAF.²⁴ But the list was incomplete even when taking into account only those groups that participated in the conflict, and it did not include the many militias that were not fighting.

In a fragmented scene of small militias, most factions have limited ambitions. Few Libyan factions explicitly raise their weapons against the government. The vast majority claim official status as units of the interior or defense ministries, and many receive salaries from one of the two rival governments, despite not being under effective government oversight.²⁵ Local-level concerns often drive recruitment, such as community protection or financial incentives, whether in the form of state salaries or predation. By contrast, Haftar’s faction from the outset aspired to overall power.

Social Cleavages and Cohesion

Behind the big picture of Libya’s fragmented military landscape, there is much variation between armed groups in their relationships with local communities. The formative period of these variations was the 2011 war, during which some western Libyan cities rapidly gained a reputation as revolutionary strongholds, while other communities became known as regime loyalists. These labels often obscured significant intra-communal divides. But they developed a power of their own as crimes committed by members of one community against another provoked deep inter-communal rifts. The war thereby created lasting identities of western Libyan communities as either revolutionary or loyalist, and drew deep rifts between them that truncated pre-war social networks. Violent conflict promoted social cohesion in places where communities united to fight against external threats. Elsewhere, it provoked internal cleavages, distrust and violent local power struggles.

By contrast, because eastern Libya was shielded from regime forces early on in the conflict, the revolutionary-loyalist divide there was not associated with intercommunal rifts, and the absence of collective struggle meant that eastern communities did not become more cohesive. In southern Libya, intercommunal conflicts erupted only when the regime fell; their association with revolutionary versus loyalist divides was superficial rather than lasting; and because armed groups were less strongly rooted in local communities, conflicts did not promote social cohesion within these communities.²⁶

Communities that emerged cohesively from collective struggle in 2011 – such as Misrata, Zintan, and the Amazigh towns in the Nafusa mountains – host powerful local forces, many of which are not standing armed groups. Their leaders and members are demobilized most of the time, mobilizing in the face of acute external threats. This type of armed group generally relies on tight-knit networks of friends, neighbors and relatives, and new recruits come from such social networks. Commanders do not take unilateral decisions to mobilize, but are part of complex, informal negotiations and consultations with influential businessmen, politicians as well as their immediate social surroundings.²⁷ Multiple groups co-exist in these communities, but clashes between them are extremely rare. These groups are socially embedded in the sense that their boundaries within society are highly permeable.²⁸ But far from all of Libyan armed groups correspond to this type of a socially embedded force. Many are militias defending more parochial interests, and therefore less bound by communal definitions of friend and enemy. As shown below, such differences in the politicization of identity and in local social cohesion help explain the divergent outcomes of Haftar’s attempts to establish LAAF units across social divides.

Militant Consolidation: The Case of Khalifa Haftar’s LAAF

The rise of the LAAF to become the only Libyan organization since 2011 to centralize power over entire regions was meteoric. Prior to May 2014, Haftar did not control any forces; by 2018, he was at the head of Libya’s dominant military coalition. With a controversial past²⁹ and after almost two decades in U.S. exile, Haftar returned to Benghazi – then the seat of the newly formed rebel leadership, the National Transitional Council (NTC) – in March 2011, two weeks after the uprising against Qadhafi had erupted. He refused to submit to the command of the NTC’s Chief of Staff, instead becoming “commander of land forces” and formally reporting to the NTC’s Defense Minister.³⁰ In reality, none of these positions meant meaningful control over the armed groups that made up the bulk of rebel forces. Haftar’s role in the 2011 war was unremarkable. Contrary to other commanders, he did not build up significant forces of his own.

Following the demise of the Qadhafi regime, Haftar unsuccessfully promoted himself as candidate for a leading military position. But his obvious personal ambitions provoked distrust, and he went empty-handed. Haftar then tried to mobilize supporters among army officers who were disgruntled over their marginalization by the revolutionary armed groups. In autumn 2013, there were persistent allegations that Haftar was plotting the overthrow of the parliament that succeeded the NTC, the General National Congress (GNC), and he was retired along with hundreds of other officers.³¹

During this period, Haftar was one of many armed actors who were using or threatening force, thereby pushing Libya’s transitional process to the edge. Two camps emerged. One portrayed itself as defending the 2011 revolution against counter-revolutionary plots, and gained dominance over the GNC; the opposing coalition sought to bring about the GNC’s downfall. It included proponents of eastern autonomy; the biggest political party in the GNC, Mahmoud Jibril’s National Forces Alliance (NFA); as well as armed groups from Zintan that were allied with the NFA.³² Haftar’s attempts at building alliances contributed to the formation of this camp, but contrary to others, he lacked a military force of his own.

On February 14, 2014, Haftar appeared on TV declaring Libya’s transitional institutions dissolved in the name of a “general command of the Libyan army.”³³ But no forces moved in support of his declared coup, and Haftar fled Tripoli to Benghazi as the government issued an arrest warrant for him.³⁴

Cooperating for Mobilization

Until February 2014, Haftar had no organization he could rely on. But after the debacle of his coup declaration, he focused on mobilizing support in Benghazi and eastern Libya, where three constituencies offered fertile ground for his efforts. First, military officers and their relatives were furious at a spate of assassinations that targeted former and active officers in Benghazi and Darna. The Tripoli government proved unable to track down the perpetrators, who remained unknown – though jihadist groups such as Ansar al-Sharia were widely suspected of being responsible for many of the killings.³⁵ Second, the eastern autonomy movement and its armed proponents were challenging the legitimacy of the transitional institutions increasingly openly. Third, an escalating conflict over territory opposed revolutionary and Islamist-leaning armed groups from Benghazi to armed groups recruited from the Awagir and other eastern tribes in the Benghazi periphery, as well as to the Saeqa Special Forces in Benghazi.³⁶

From March to early May 2014, Haftar received leaders of armed groups in a house on Benghazi’s outskirts, and toured eastern Libya to meet with army officers.³⁷ A number of officers from the wider Benghazi area joined Haftar to plan an offensive against revolutionary and jihadist groups in Benghazi.³⁸ Several of them had personal reasons for joining Haftar, having been retired or dismissed from military service by the authorities. Air force officers at Benina and Tobruk put themselves and their ageing aircraft at Haftar’s disposal.³⁹ Most importantly, however, Haftar obtained support from members of the Saeqa Special Forces and from Awagir militia leaders in Benghazi’s outskirts and environs.⁴⁰

As he launched his “Operation Dignity” in Benghazi on May 16, 2014, Haftar was at the head of a loose alliance of armed groups, each of which had their own leadership, weapons and interests. Some saw in him a savior from jihadist groups, whom they suspected behind the assassinations of military officers. But many had a more opportunistic relationship with Haftar. The Benghazi militias sought to use his operation to prevail over their local rivals. The eastern autonomy movement wanted to advance its own agenda, in which Haftar could feature only in an interim capacity. As a member of the Firjan tribe, Haftar was not considered an easterner. His relationship with prospective allies in western Libya was even more tenuous.

Haftar’s operation was an open rebellion against the government in Tripoli. The day after launching his operation, he issued a statement in the name of the “leadership of the Libyan National Army,” and called on all soldiers to report to service.⁴¹ Operation Dignity furnished an overall narrative for the disparate struggles that were escalating across Libya, by announcing a merciless fight against all “Islamists” – a category that could be applied at will to political opponents. The political landscape split in two as numerous politicians and armed groups across the country declared their support for Haftar’s operation. His declared intent to eradicate the Muslim Brotherhood also resonated strongly in Egypt and the United Arab Emirates, whose support would be decisive in lifting Haftar out of the mass of Libyan militia leaders.⁴²

Operation Dignity opened hostilities against a wide range of groups, including revolutionary armed groups that enjoyed the status of official units.⁴³ This approach triggered two reactions that would play out in Haftar’s favor. First, because the threat from Haftar united them, several of Benghazi’s Islamist-leaning and revolutionary armed groups banded together with Ansar al-Sharia to form the Benghazi Revolutionaries Shura Council (BRSC). Although many BRSC fighters were not Islamists and several revolutionary armed groups continued to fight outside the BRSC umbrella, this allowed Haftar to depict his adversaries as terrorists. It would also create an opening for returnees from Syria and former members of Ansar al-Sharia to establish an Islamic State affiliate in Benghazi, which emerged in late 2014 to fight against Haftar’s forces separately from the BRSC. It became common for Haftar’s supporters to refer to his enemies in Benghazi as Dawa’esh – members of Islamic State. Instead of isolating the extremists, Haftar’s purported counterterrorist operation significantly boosted them.⁴⁴ But as a means of demonizing a wide range of political opponents and mobilizing foreign support, it was highly effective.

Second, in response to Haftar’s offensive, his adversaries began to attack armed groups, military officers and activists they suspected of supporting Haftar. During July and August 2014, the BRSC took control of much of Benghazi, and assassinations accelerated in September.⁴⁵ The component groups of Operation Dignity now had little choice but to stick to their alliance with Haftar to defend themselves, and the jihadist threat drove vast parts of eastern society into Haftar’s arms. Haftar also gained the support of a Salafist armed group whose leader designated Haftar’s adversaries as “apostates.”⁴⁶ This was the beginning of Haftar’s alliance with followers of an ultraconservative strand of Salafism led by the Saudi preacher Rabi’ al-Madkhali. The Madkhalis were radically opposed to everything from democracy to the Muslim Brotherhood and al-Qaeda, and would become a key component of Haftar’s forces.

Haftar’s branding of his forces as the “Libyan National Army” – which would gradually give way to the LAAF – was a masterstroke. It offered any armed group that joined his coalition the incentive to become part of “the” army. In reality, the vast majority of Haftar’s forces were made up of civilians, as his officers themselves admitted.⁴⁷ Even army units that had survived the revolution, such as the Saeqa Special Forces, were by this point closer to militias than to regular forces.⁴⁸ Moreover, several prominent eastern army officers openly rejected Haftar’s operation. Still, to some Libyan officers as well as to foreign patrons, Haftar’s branding conveyed an aspiration to rebuilding an army.

Haftar’s own military power in the early days of his operation was insignificant, which paradoxically helped to mobilize support for his ploy. His chances of emerging as the leading figure from the demise of the transitional institutions appeared slim. Actors across the country looked at his rebellion as an opportunity rather than a threat. Many declared their support for the operation of the “Libyan army” without explicitly backing Haftar – among them the National Forces Alliance and the Zintani-led Qa’qa’ and Sawaeq Battalions, which attacked the GNC in Tripoli two days after the start of Haftar’s operation in Benghazi.⁴⁹ While loosely allying with Haftar, these actors competed with him for the helm of the rebellion. But as a result of their moves, the notion of a parallel army leadership opposed to the transitional institutions took root.

After a brief respite for the elections to the GNC’s successor parliament, the House of Representatives (HoR), war erupted in Tripoli. On July 13, 2014, a coalition of revolutionary and Islamist-leaning armed groups calling itself “Dawn of Libya” attacked the positions of the Zintani-led armed groups that had loosely allied with Haftar. Dawn of Libya claimed to be a reaction to counter-revolutionary movements, by which the operation’s architects meant both Haftar’s power grab and the Zintani-led armed groups.⁵⁰

In early August, the HoR was due to meet for the first time in Benghazi, but fighting raged both in Benghazi and in Tripoli. The question where the parliament should convene became entangled with the escalating civil war. Influential allies of Haftar engineered the parliament’s move to Tobruk, where army units had aligned with Haftar. Thirty HoR members opposed to Haftar boycotted the Tobruk sessions.⁵¹ A rump GNC reconvened in Tripoli in late August, exploiting the controversy over the legality of the HoR’s Tobruk secessions, and formed its own government.⁵²

Although Haftar still had no official position, his operation now enjoyed the backing of the internationally recognized parliament and its government.⁵³ Haftar supporters in the HoR succeeded in appointing a close associate of Haftar, Abderrazeq al-Nadhuri, as Chief of Staff. Shortly before Nadhuri’s appointment, Emirati warplanes had already struck Dawn of Libya positions in Tripoli.⁵⁴ But official legitimacy made it easier for Egypt and the UAE to provide support to Haftar.⁵⁵ In September, both states began supplying Haftar with weapons and ammunition.⁵⁶ Over time, that support would prove crucial in establishing Haftar as the uncontested leader of his coalition.

In the first year of the Benghazi operation, however, Haftar primarily harnessed the self-motivation of armed groups. Key to mobilization in his coalition was agitation against “Misratans” – families of Misratan origin who had moved to Benghazi dozens or hundreds of years ago, and formed a sizeable part of the city’s commercial and educated elite. Members of such families were indeed among the leaders and rank-and-file of the BRSC, but so were many members of eastern tribes, and families of Misratan origin were far from broadly supportive of the BRSC. The rifts of the Benghazi conflict often split local families in two. Incitement against “Misratans” or “Turks” – since some of these families were of mixed Ottoman-Libyan descent – exploited socioeconomic rancor among recruits for Haftar’s operation.⁵⁷ The leaders of armed groups from Benghazi’s outskirts groups coveted the wealth and administrative positions members of “Misratan” families held in the city.

To gain a foothold in Benghazi, Haftar covertly armed cells of civilian fighters in the city that surfaced when Dignity forces moved into Benghazi’s eastern suburbs in October 2014.⁵⁸ As they took control, these groups kidnapped and killed perceived opponents, and burned the homes of families who had members fighting with the BRSC – or in some cases simply because they were of Misratan origin.⁵⁹ Leaders of armed groups in Haftar’s coalition openly vowed to kill or expel “Misratans” and “Turks.”⁶⁰ One commander publicly explained that to satisfy the victims of terrorism, it was justified that “anyone who was a suspect” had their houses razed.⁶¹ Anyone whom Haftar’s armed groups accused of links to their enemies risked being disappeared, and bodies bearing torture marks became a common sight on Benghazi’s outskirts.⁶² Thousands of families fled Benghazi to Misrata, Tripoli and other cities in the west; fighters in Haftar’s operation seized many of their properties.⁶³

Some of the groups involved in such acts appeared to be outside Haftar’s direct control. But their acts served his interests, while the absence of formal ties accorded Haftar deniability. They spread terror that intimidated Haftar’s political opponents. They also socialized a new generation of young men from greater Benghazi to wield ruthless violence, without regard for commonly accepted moral codes or social cohesion. And they implicated perpetrators in crimes for which they wanted to avoid accountability – making them interested in Haftar’s continued leadership and averse to any outreach toward his opponents.

To recapitulate, in 2014, Haftar mobilized disparate and often tenuous allies. Since he lacked forces of his own, his approach was cooperative by necessity. Social divides were a product of the conflict more than they were constitutive of it. Haftar and his allies provoked a deep, quasi-ethnic rift through eastern Libya by equating their “terrorist” enemies with “Misratans” and “Turks.” Killings and forced displacement cemented this divide, turning it into a serious barrier to defection for Haftar’s allies and a precondition for consolidation by Haftar. Crucially, his declaration of war against all “Islamists” also proved instrumental in mobilizing foreign support, thereby laying the groundwork for the concentration of power.

Consolidation in the East

Over the first year of Haftar’s operation, his narrative of restoring the army to its rightful place gradually became dominant across eastern Libya, and Haftar gained genuine popularity in the region. When a particularly deadly bombing struck the eastern town of al-Qubba in February 2015, he exploited the groundswell in support for his counterterrorist agenda to obtain the top position he had long coveted.⁶⁴ On March 2, 2015, the HoR established the post of “General Commander of the Armed Forces,” which was superior to the Chief of Staff and also included the competencies of the defense minister. The same day, HoR speaker Agilah Saleh, to whom Haftar nominally answered under that law, appointed Haftar to the post.⁶⁵

As General Commander appointed by the internationally recognized parliament, Haftar formally towered above the armed groups that were fighting for him in Benghazi. But his own forces were still insignificant compared to these groups. To strengthen his authority, he used the military supplies he received from his foreign backers, privileging loyalists and penalizing commanders he considered unreliable. This generated serious tensions.⁶⁶ Key commanders became increasingly vocal in their criticism.⁶⁷ Haftar tried to remove one such commander, Col. Faraj al-Barassi, but failed because the latter had the backing of forces under his own command and fellow commanders.⁶⁸

Meanwhile, Haftar’s allies in western Libya entered into local ceasefire agreements with their enemies, with remaining Haftar loyalists in the region too weak to continue the war.⁶⁹ As UN-led negotiations over the reestablishment of a unified government progressed during autumn 2015, Haftar faced increasing difficulties in holding his alliance together. Several of the unruly eastern commanders in Haftar’s coalition threw their support behind the UN-led negotiations, raising the prospect that the agreement would allow Haftar’s eastern allies to sideline him. They could count on the support of Ibrahim al-Jadhran, a militia leader who controlled most oil ports in eastern Libya. When the Libyan Political Agreement (LPA) was signed in December 2015, Haftar and HoR President Agilah Saleh colluded to ensure the HoR neither ratified the LPA nor endorsed the newly created Government of National Accord (GNA).⁷⁰

During the first half of 2016, Haftar’s hold over eastern Libya appeared increasingly fragile. Defectors from his coalition tried to leverage the weight of the new, internationally recognized government against him. In January 2016, the GNA’s Presidency Council designated one of the disgruntled Benghazi commanders, al-Mahdi al-Barghathi, as Defense Minister. Throughout spring and summer 2016, Barghathi and Jadhran competed with Haftar in staging displays of support by notables in their tribes, the Awagir and Magharba.⁷¹ Barghathi’s ally, the Awagir militia leader Faraj Ga’im, accused Haftar of responsibility for assassinations and bombings in Benghazi.⁷² The challenge to Haftar’s authority was unprecedented.

Haftar, however, transformed this precarious situation into an opportunity to sideline his challengers and consolidate control. Central to his success was the growing foreign backing he enjoyed. In January 2016, French special forces began covertly supporting Haftar’s operation in Benghazi.⁷³ French assistance helped Haftar’s coalition break a long stalemate and make significant advances in Benghazi in February 2016. It also signaled that Haftar’s backers had not abandoned him, despite their declarations of support for the GNA. In spring 2016, Emirati backing made a qualitative leap with the installation of a covert UAE airbase in eastern Libya.⁷⁴ At the same time, the rival east-based Central Bank secured Russian support to print its own dinar banknotes, which it used over the following years to underwrite dozens of billions of dinars in debt to commercial banks in the east.⁷⁵ Around a third of eastern expenditure financed in this manner went straight to Haftar’s forces.⁷⁶

Such foreign support helped Haftar to expand his coalition in ways that reduced his dependence on unruly eastern commanders and increased his ability to clamp down on challengers. He promoted three groups in particular. First, he reinstated former regime loyalists in military and intelligence positions. These regime loyalists often sat uneasily with Haftar allies who were at least rhetorically committed to the 2011 revolution. But they were more dependent on Haftar’s support, since they had been unable to establish their own armed groups after 2011. Many saw Haftar’s forces as an opportunity to regain influence and take their revenge on the constituencies that had supported the revolution – particularly Misrata.

Second, Haftar strengthened the hardline Salafists who had begun joining his alliance as early as June 2014. Armed groups imbued with a doctrine that emphasized political obedience were more willing to carry out arrests that were bound to provoke negative reactions in local communities.

Third, in late 2015 Haftar began recruiting mercenary fighters from Darfur for the Benghazi war.⁷⁷ Over the next years, he would come to deploy thousands of such mercenaries to remote areas of central and southern Libya, thereby projecting force in ways he could not have done had he continued to depend on his original allies in Benghazi.

The process of strengthening more docile elements over the unruly components of Haftar’s coalition was slow, but it progressively granted Haftar a measure of autonomy from the constituencies that had supported him since 2014. Haftar used that autonomy to ratchet up repression against disobedient former allies, and signs of dissent in the east more broadly.⁷⁸ In Benghazi, bodies of victims of extrajudicial executions by Haftar’s forces began turning up by the dozens in rubbish dumps.⁷⁹ Meanwhile, Haftar televised meetings at which tribal politicians declared their loyalty to him, in spectacles that closely resembled those staged by Qadhafi.⁸⁰

Haftar increasingly revealed his authoritarian tendencies because his popularity in much of the east was unbroken. It rose to new heights after he seized control of the oil crescent. Following repeated skirmishes with Jadhran’s forces in Ajdabiya from June 2016 onwards, Haftar eventually succeeded in taking over the oil crescent without major fighting, by coopting and empowering rivals of Jadhran in the Magharba tribe. Haftar thereby expanded his control across eastern Libya, with the exception of Darna and remaining pockets of resistance in Benghazi. With Jadhran, he had ejected a key eastern ally of the unity government. The prospect that eastern GNA supporters could sideline Haftar vanished. GNA Defense Minister al-Mahdi al-Barghathi fled Benghazi, and in November, Haftar’s forces captured Barghathi’s base.⁸¹

Once Haftar had warded off the challenge from within his former coalition, the path was clear for him to consolidate control over the east. During autumn 2016, he replaced elected mayors with his appointees in eight eastern cities.⁸² Beginning in January 2017, a Haftar protégé in the Saeqa Special Forces, Mahmoud al-Warfalli, spread terror across the east with a series of videos that showed him executing captives in Benghazi.⁸³ In February 2017, Salafist fighters loyal to Haftar twice shot their way into the home of Faraj al-Barassi, one of the officers who had defied Haftar’s authority.⁸⁴ This showed that Haftar now had unprecedented autonomy from his erstwhile allies. In May 2017, a car bomb that killed a tribal politician who had supported al-Mahdi al-Barghathi against Haftar reinforced that message.⁸⁵ In October, 36 bodies of executed prisoners were found near al-Abyar, on Benghazi’s outskirts, further underlining the impunity Haftar’s forces enjoyed.⁸⁶

Western states began courting Haftar, even more so after Haftar took control over the airbases of al-Jufra and Tamanhant in June 2017, thereby expanding toward central and southern Libya. In July 2017, France elevated Haftar by inviting him to a summit meeting with President Emmanuel Macron and GNA Prime Minister Fayez al-Serraj. An official visit to Italy followed soon after. International courtship of Haftar meant that Western governments would not exert pressure on his foreign supporters to respect the UN arms embargo. The weapons flow to Haftar continued unabated, while the GNA in Tripoli was unable to procure arms from abroad.⁸⁷

Haftar now used foreign support to promote loyalists in his coalition and establish new units over which he exerted complete control through his sons, relatives, or close confidants. The first of these units to emerge, in late 2016, was the 106^th Battalion, which was led by Haftar’s son Saddam, whose brother Khaled succeeded him as the head of the battalion. Over 2017-2018, other units led by Haftar’s nephews, in-laws, and close associates of his sons emerged.⁸⁸ They received the most advanced equipment Haftar obtained from his foreign backers.⁸⁹

In addition to ensuring personal loyalty to Haftar, several of the new units included Salafists⁹⁰ and local constituencies that harbored resentments against the former revolutionary strongholds.⁹¹ They were also intertwined with economic interests. A network of business and financial ties emerged around Haftar’s sons and relatives, as well as a handful of close associates. This group made large profits through predation, seizing public and private land, or monopolizing markets such as subsidized fuel or the export of scrap metal.⁹² Along with these developments came an even stronger promotion of a personality cult surrounding Haftar.⁹³

The rise of Haftar’s cronies and the centralization of power within a narrow circle of relatives disappointed those LAAF officers who had harbored hopes that they were helping rebuild a national army.⁹⁴ It also fueled resentment among Haftar’s eastern allies, who often saw themselves as the real owners of a project that Haftar had usurped. But it made mobilization against Haftar increasingly difficult. In August 2017, the Awagir militia leader Faraj Ga’im made another attempt. He had himself appointed Deputy Interior Minister in the GNA, thereby openly challenging Haftar’s authority. After several months during which both sides avoided confrontation, two attempts targeted Ga’im in November 2017. Ga’im publicly accused Haftar, calling on Saeqa commander Wanis Bukhamada to “save the Libyan army” and sideline Haftar.⁹⁵ In response, Haftar’s sons Saddam and Khaled attacked Ga’im’s family home and captured him.⁹⁶ To other commanders in eastern Libya, this demonstrated Haftar’s ability to dispose of former allies at will. Haftar only released Ga’im in August 2018, following continued petitioning from Ga’im’s relatives. A year later, he eventually allowed a diminished Ga’im to regain a role as the head of a small force in Benghazi.⁹⁷

Haftar’s suppression of the last remaining challenge to his authority in the east, combined with his increasingly blatant cronyism, left many of his former allies out in the cold. Even formerly close Haftar associates, if deemed unreliable, now risked being abducted and held incommunicado.⁹⁸ Several politicians and commanders who had supported his Benghazi war found refuge in Tripoli or abroad. There, many reestablished contacts with their former enemies, although the rift wrought in 2014 remained deep, inhibiting alliance-building. Some vocal Haftar supporters in the Libyan media slowly turned into his outspoken opponents.⁹⁹

In sum, Haftar concentrated power in the east through a phased approach. As his allies began challenging his preeminence, he increasingly outcompeted them by mobilizing additional supporters, while also building his own, loyal forces. Initially, Haftar was cautious in confrontations with challengers from within his coalition. But his ability to use coercion gradually increased. Ultimately, he cemented his power by demonstrating that he could wield unlimited repression even against his former support base. The fuel for that process was foreign assistance; its by-product disgruntled but powerless former allies.

Cautious Expansion amid Ambiguity

In the east, Haftar had designated a clear enemy, and anyone who challenged him risked accusations of colluding with terrorists. But such all-out war against mortal enemies was unsuitable to gain footholds in the south and west of the country. In western Libya, fighting had ended in summer 2015, in the south several months later. Most municipalities reported to the GNA in Tripoli, hoping to mobilize resources from it. As an assemblage of rival factions, the GNA exerted no central authority, and local armed groups continued to operate largely independently, with rivalries over turf common in Tripoli. But escalations of violence were localized and transitory. Public opinion was averse to renewed conflict. Moreover, with the campaigns against the Islamic State in Sirte and Sabratha, western Libyan armed groups in 2016 drew clearer lines between themselves and the extremists.

In this context, starting a war would have been highly unpopular. To expand beyond the east, Haftar had to change strategies. He had to appear as bringing stability and liberation from abusive militias. His expansion into the oil crescent in September 2016 set the tone: following a largely bloodless take-over, Haftar surprised local and foreign observers alike by handing control over the oil facilities back to the Tripoli-based National Oil Corporation (NOC), allowing revenues to accrue to the Central Bank in Tripoli.

By contrast, Haftar’s opponents increasingly acquired a reputation as spoilers. With covert backing from GNA Defense Minister al-Mahdi al-Barghathi, the Benghazi Defense Battalions (BDB) allied with Jadhran and some individuals with extremist ties, attempting to capture the oil crescent from Haftar in December 2016, and again in March 2017.¹⁰⁰ In June 2018, some BDB members joined a third attempt led by Jadhran. Each time, the GNA and NOC condemned their actions, and production resumed once Haftar’s forces had taken back control (though in 2018, Haftar blocked exports for several weeks).¹⁰¹

In May 2017, fighters affiliated with the BDB and a Misratan armed group executed dozens of prisoners they had taken after overrunning one of Haftar’s bases in southern Libya.¹⁰² The incident provoked an outcry that, along with a brief campaign of Egyptian airstrikes, led the Misratans and the BDB to vacate southern and central Libya. Haftar’s forces took over airbases in those regions without encountering resistance.¹⁰³ In the oil crescent and central Libya, Haftar deployed his recently formed loyalist units, thereby displaying his new ability to project power.¹⁰⁴

Beyond central Libya, Haftar had to tread carefully. Western and southern Libya were fragmented along communal lines, with many communities also suffering internal divides. Allying too closely with armed groups from one community risked alienating those groups’ local rivals. Towns that had suffered exactions and marginalization at the hands of neighboring revolutionary strongholds offered fertile terrain for Haftar, but his forces had to avoid threatening their neighbors or risk being taken out. The collective struggle in 2011 had welded together revolutionary strongholds; in places like Misrata or the Amazigh towns, social cohesion and commitment to the revolutionary ideals prevented Haftar from gaining a foothold.¹⁰⁵

In Zintan, the necessity to maintain social peace despite an internal political rift did much to thwart Haftar’s ambitions. Zintan had emerged united from its revolutionary struggle. In 2014, the town’s forces allied with Haftar to defend themselves. But the following year, former revolutionaries under Osama Juwaili negotiated ceasefires with their adversaries. Without Juwaili’s forces, Haftar’s allies in Zintan were unable to continue the war. In July 2017, Juwaili accepted a senior military position in the GNA, further weakening the Haftar faction in Zintan. But Juwaili never confronted Haftar’s Zintani allies. A Zintani Salafist group aligned with Haftar used the al-Wutiya airbase to supply Haftar loyalists in western Libya, fueling internal tensions in Zintan. But the dense social networks that linked both factions kept them from attacking each other, right up until Haftar’s April 2019 Tripoli offensive.¹⁰⁶

Elsewhere, Haftar enticed local officers to nominally join the LAAF while keeping a low profile. This applied to Bani Walid, where regime loyalism remained strong and social pressure discouraged Haftar’s local allies from operating in the open. Many in Bani Walid saw Haftar as a traitor due to his support for the revolution in 2011.¹⁰⁷ It also applied to Tarhuna, where the militia of the Kani brothers had exclusive control, including over the LAAF unit whose establishment they allowed to attract some salary payments from Haftar.¹⁰⁸ Similarly, in the southern town of Ubari, some Tuareg soldiers joined the LAAF to receive a second salary, in addition to what Tripoli paid them. But their primary loyalties continued to lie with their local friends and relatives whose allegiance to the GNA was equally nominal.¹⁰⁹

A first attempt to take a more forceful approach in southern Libya proved unsuccessful. In February 2018, Haftar tried to expand in the southern city of Sabha by extending the LAAF franchise to armed groups from the Awlad Suleiman.¹¹⁰ But the largest Awlad Suleiman militia, the 6^th Brigade, split over its alignment with Haftar. Moreover, given that these groups had fought against the Tubu in preceding intercommunal conflicts, Haftar’s approach alienated large parts of the Tubu community in Sabha.¹¹¹ Several Tubu armed groups that had previously supported Haftar – while Awlad Suleiman militias had allied with Haftar’s Misratan opponents – now turned against him.

In January 2019, Haftar was finally able to tilt the balance of power in Sabha. He tasked a force of some 350 vehicles with the declared aim of “liberating the south from terrorists and Chadian gangs.”¹¹² The core units involved were the Tareq ben Ziyad and 128^th Battalions, whose commanders both had close ties to Haftar’s sons, were themselves from southern Libyan tribes, and employed a substantial number of Sudanese mercenaries as auxiliaries. They encountered little resistance. Across the south, anger had been on the rise for months due to rampant insecurity, the collapse of public services, and the disinterest of the government in Tripoli for the region’s predicament. Many hoped Haftar’s take-over would bring improvement.

The force was too small to take southern Libya through a military offensive. Instead, using promises of cash and the expectation that he would soon rule in Tripoli, Haftar enticed southern armed groups into declaring their loyalty to him. In Sabha, most armed groups from the Awlad Suleiman did exactly that, helped by the fact that a fellow tribesman, Hassan al-Zadma, led the 128^th Battalion. Those who refused to align with Haftar kept their weapons, but lay low. Tubu armed groups in Sabha negotiated the handover of their bases, then withdrew toward Murzuq.

Several Tuareg armed groups in Ubari also adhered to Haftar’s forces, allowing Haftar to take control of the key oilfield of al-Sharara. A last-ditch attempt by the GNA to reverse Haftar’s advance by appointing the senior Tuareg officer Ali Kanna as the GNA commander for the southern region backfired. In a situation where Haftar promised to bring stability, the GNA appeared to sow division. Local opinion was overwhelmingly in favor of allowing Haftar to take-over, to prevent a rift through the local community.¹¹³

Resistance to Haftar’s southern expansion was limited to parts of the Tubu community. Haftar’s alignment with the Awlad Suleiman against the Tubu in Sabha was accompanied by rhetoric against “Chadian gangs” – code language for Tubu armed groups. When Haftar’s forces moved toward Murzuq, local Tubu forces opposed them.¹¹⁴ But that opposition soon faltered, and even violent reprisals by Haftar’s forces against the Tubu community in Murzuq did not provoke wider resistance.¹¹⁵ The Tubu had no allies to support them against Haftar, whether in the south or in the GNA. Moreover, Haftar retained some Tubu allies – including a recently established Salafist armed group in Umm al-Araneb.¹¹⁶ Tubu armed groups in other southern towns therefore recognized Haftar’s authority while otherwise staying put – much like what had happened before in Sabha and Ubari.

Haftar acquired dominance rather than effective territorial control over southern Libya. After an initial groundswell of support as the eastern government supplied goods and cash to southern cities, it became clear that the region’s predicament persisted. Armed groups and foreign mercenaries largely remained in their positions. The militias that had joined Haftar retained much of their autonomy.¹¹⁷ Several armed groups in the region refused to join the LAAF, but kept a low profile to avoid conflict.

Among the Tuareg in Ubari and the Awlad Suleiman in Sabha, social ties certainly helped to prevent open conflict between pro- and anti-Haftar factions, as did the uncertainty over the balance of power. Haftar’s successful and largely bloodless southern expansion had fundamentally altered the calculations of both local and international actors. The expectation that Haftar would seize power in Tripoli became increasingly widespread. Given such expectations, Haftar’s opponents in the south chose to lay low.

In a context of ambiguity, and where Haftar lacked the overwhelming force needed to take control, he expanded by cultivating local allies and encouraging them to cooperate. Objective ethnic cleavages had little impact on his choices: Amazigh communities in western Libya resisted Haftar’s courtship, while in southern Libya, some Tubu and Tuareg armed groups did join the LAAF despite Haftar’s Arab nationalist rhetoric, while others opposed him. Where the cleavages of past conflicts were relatively superficial, such as in Ubari and Sabha, social permeability facilitated a cooperative approach, but a modicum of social cohesion discouraged Haftar’s local allies from moving against his opponents within their own community. In Zintan and Bani Walid, social cohesion constrained Haftar’s local allies in their room for maneuver. And it prevented Haftar from making inroads into communities that defined themselves through their role in the 2011 war, such as Misrata and the Amazigh towns.

Enduring Social Cleavages and Stalled Consolidation

As expectations grew in early 2019 that Haftar would expand in western Libya, UN-led efforts to broker a deal between Serraj and Haftar accelerated, and militia leaders from the region covertly intensified their contacts with him. Haftar gained promises from a number of actors in western cities that they would side with his forces, or at least not oppose them. In exchange, these actors hoped to safeguard their local fiefdoms or feature in new political alignments dominated by Haftar.¹¹⁸ Few western militia leaders openly switched sides. Chief among them was Adel Da’ab in Gharyan, whose group had fought against Haftar’s Zintani allies in 2014-15, but in early 2019 became the LAAF’s 111^th Battalion.¹¹⁹ In Gharyan, Da’ab’s defection caused tensions with Haftar’s opponents. Meanwhile, there were growing suspicions that the Kani brothers’ militia in Tarhuna was colluding with Haftar. In August 2918, the Kaniyat had allied with Misratan armed groups in an attempt to win territory in Tripoli, but in January 2019, after being abandoned by their Misratan allies, they turned to Haftar for support.

On 3 April 2019, Haftar launched a large-scale offensive on Tripoli, forcing actors in western Libya to take sides. Attacking western Libyan armed groups head-on was a gamble on his superior firepower and the willingness of western Libyan forces to accept his take-over. Da’ab and the Kaniyat allowed Haftar to use Gharyan and Tarhuna as forward bases for his offensive, but the reactions of other armed groups confounded Haftar’s expectations. Zawiyan forces on 5 April captured over a hundred of Haftar’s troops who had apparently counted on cooperation from Zawiya to reach Tripoli.¹²⁰ Misratan forces mobilized to stop Haftar’s offensive. The Tripoli militias that were widely suspected of being opportunistic enough to collude with Haftar were among the first to stand in his way.

It was not simply that most of western Libya’s standing militias united against the common threat that Haftar’s forces posed to them. Haftar’s offensive also caused dormant forces to mobilize in revolutionary strongholds like Misrata and the Amazigh towns that had either fought in 2011 or had relatives that did. Their cohesion and social embeddedness helped explain the fierce resistance these forces offered to Haftar’s offensive despite receiving little or nothing in the way of supplies and salaries from the Tripoli government.¹²¹ The social cleavages wrought by conflicts since 2011 therefore dissuaded potential defectors. Haftar’s propaganda machine openly cultivated a desire for revenge against the former revolutionary strongholds in general, and Misrata in particular. Such revanchist sentiment was rampant across the LAAF, and all the more threatening as Haftar’s forces had been responsible for war crimes in Benghazi and Darna that were unparalleled in Libya’s post-2011 conflicts.

Even for more opportunistically-minded groups, side-switching made little sense. After all, Haftar could give his enemies no credible assurances that he would not move against them once he had won the war. Western Libyan militia leaders had watched and learned how Haftar had sidelined former allies in Benghazi once he no longer needed them. After the first three days of the conflict, virtually no defections occurred for the remainder of the war in western Libya. The Salafist-leaning 604^th Battalion in Sirte was a notable exception that proved the rule. In January 2020, it flipped, allowing Haftar to capture the city without a fight. Sirte had been held by Misratan forces who were widely unpopular in the city, a former regime stronghold at the margins of western Libya, and the 604^th Battalion was heavily recruited from the Firjan – Haftar’s tribe.¹²²

Initial expectations of Haftar’s rapid victory having been disappointed, the war settled into a stalemate. This still appeared to play to Haftar’s advantage, since he enjoyed UAE military backing and political support from the Trump administration and France. In addition, from September 2019 onwards, he deployed Russian mercenaries who slowly tilted the balance in his favor.¹²³ The GNA, by contrast, received only limited support from Turkey in the first months of the war, and during the autumn this support stopped, making the situation of Haftar’s opponents increasingly desperate.¹²⁴ Turkey only resumed its support in December 2019, after forcing the GNA to sign an agreement on maritime boundaries.¹²⁵

The fact that Haftar had multiple foreign backers granted him substantial room for maneuver. French and Egyptian officials claimed that they had warned Haftar against an offensive on Tripoli – but they still supported his war politically once it had started.¹²⁶ In January 2020, Russia and Turkey tried to coax their respective clients into a ceasefire. Serraj, entirely dependent on immediate Turkish support to ensure the GNA’s survival, had no choice but to accept.¹²⁷ But Haftar, encouraged by his UAE backers, refused.¹²⁸ Still, Russian military support to Haftar increased over the following months.¹²⁹

With foreign intervention in the war heavily tilted in his favor, Haftar could have been expected to replicate the successful consolidation of power over his alliance that he had previously achieved in Benghazi. In addition to his core loyalists, Haftar’s forces in Tripoli relied heavily on more opportunistic allies.¹³⁰ As in Benghazi, many hoped to use Haftar to win the war, then defend their independence from his authority or even turn on him. But as in Benghazi, his allies had little choice but to stick with him once the war had started, and they grew dependent on the foreign support Haftar alone was able to mobilize. Haftar gradually strengthened his authority over opportunistic allies as the war continued. Many among the unrulier commanders were killed in the war – some under dubious circumstances – which allowed Haftar to integrate their foot soldiers into tighter command structures.¹³¹

During spring 2020, direct Turkish intervention upturned the balance of power, and Haftar’s forces were routed from western Libya. Defeat posed a grave threat to Haftar, given that his coalition was built on the assumption that he would prevail. Foreign support that prevented Haftar’s enemies from advancing toward eastern Libya was crucial in stabilizing his authority after the defeat in Tripoli.¹³² However, foreign backing alone could not explain the fact that Haftar’s rule over the east survived intact. The core of the LAAF proved cohesive. Economic predation by Haftar’s inner circle and their LAAF units increased further during and after the Tripoli war, as other LAAF funding streams dried up.¹³³ Haftar’s sons provoked growing popular resentment with conspicuous displays of wealth and propaganda aimed at their glorification.¹³⁴ Haftar’s power structure made up for his declining popularity with intensified repression. As the LAAF faced growing financial difficulties and eastern politicians gained positions in a new Government of National Unity (GNU) during spring 2021, some observers saw Haftar’s forces as “falling apart.”¹³⁵ Confounding such expectations, Haftar and his close relatives have maintained firm control of eastern and central Libya, as well as dominance over the south. Haftar’s ability to survive what could have been fatal blows underlines how successfully he has transformed an initially loose alliance in eastern Libya into a centralized power structure – albeit one highly dependent on Haftar and his close relatives.¹³⁶

Conclusion

Haftar’s power relative to allies and enemies shaped his approach to consolidation. Where his own power was insufficient to prevail against enemies, he cooperated with allies, leaving open what role they would play after victory. In the ambivalent situation that prevailed in western and southern Libya between 2016 and 2018, he quietly cultivated allies, while avoiding confrontations in areas where his coercive power was limited. As his power increased, Haftar gradually imposed his authority over allies by granting or withholding the support he obtained from foreign backers. Once he had established forces of his own, he punished disloyal members of his coalition by arresting, killing or expelling them, pour encourager les autres. Ultimately, he wielded coercive force to cow his erstwhile allies and their social support base into submission. His access to multiple foreign state sponsors accorded Haftar wide leeway in adapting his strategy.

After this approach proved successful in Benghazi, Haftar attempted to replicate it in Tripoli. This time, his gamble failed, and the Turkish intervention was not the only reason. Haftar undoubtedly overestimated the opportunism of armed groups in western Libya, just as he underestimated the social embeddedness and cohesion of the forces that mobilized to resist his advance. Theorists who see armed groups as perfectly flexible in their reactions to the changing balance of power might have made the same mistake.

The barriers to defection were not ethnic or tribal barriers as such, but social cleavages that received their political significance above all from the role they had played in conflicts since 2011, and the social transformation of communities they reflected. Armed groups’ capacity for opportunistic behavior varied strongly depending on how they related to local communities, and how cohesive these communities were. Armed groups with a weak social base in southern and western Libya were flexible in deciding whether or not to join Haftar. The LAAF, which gradually gained autonomy from its support base, was increasingly free to use coercion against former allies in eastern Libya, while treading carefully in cohesive western Libyan communities. In former revolutionary strongholds such as Misrata and the Amazigh towns, social cohesion and the enduring rifts of past conflicts inhibited Haftar’s courtship of local allies, thereby encouraging him to take a coercive approach.

The argument presented here adds to the growing number of studies that emphasize the transformative effects of violent conflict on social structures, networks, boundaries and identities.¹³⁷ The effects of ethnic, tribal or religious boundaries cannot be compared independently of these boundaries’ political significance in a given conflict.¹³⁸ Network density, social cohesion, solidarity, and the truncation of social networks differ widely between groups that superficially fall into a common category such as “tribe.” Violent conflict often deeply transforms such aspects of communities, and their relations with armed groups.

A challenge for future research is how to reconcile the causal effects of conflict-induced cleavages with rational action in response to threats and opportunities. Arguing that such cleavages are a consequence rather than a cause of violent conflict does not yet explain how they might interact with strategic behavior.¹³⁹ Whereas seemingly objective social boundaries have served to predict behavior, the effects of conflict-induced cleavages can only be determined ex post. A key reason for this is that the constraints they impose often contradict the imperatives that drive strategic action.¹⁴⁰ Fully accounting for such conflicting incentives could offer fresh perspectives not only on militant fragmentation and consolidation, but also on other aspects of violent conflict.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Notes

1 Wolfram Lacher, Libya’s Fragmentation (London: I.B. Tauris, 2020).

2 Wolfram Lacher, “A Most Irregular Army: The Rise of Khalifa Haftar’s Libyan Arab Armed Forces”, SWP Working Paper, November 2020, doi: 10.18449/2020WP12.

3 Fotini Christia, Alliance Formation in Civil Wars (New York: Cambridge University Press, 2012); Michael Woldemariam, Insurgent Fragmentation in the Horn of Africa: Rebellion and Its Discontents (Cambridge, UK: Cambridge University Press, 2018).

4 Costantino Pischedda, “Wars within Wars: Why Windows of Opportunity and Vulnerability Cause Inter-Rebel Fighting in Internal Conflicts,” International Security 43, no. 1 (2018): 138–76; Jonah Schulhofer-Wohl, “On-Side Fighting in Civil War: The Logic of Mortal Alignment in Syria,” Rationality and Society 32, no. 4 (2020): 402–60; Eric Mosinger, “Balance of Loyalties: Explaining Rebel Factional Struggles in the Nicaraguan Revolution,” Security Studies 28, no. 5 (2019): 935–75.

5 Hanne Fjelde and Desirée Nilsson, “Rebels against Rebels: Explaining Violence between Rebel Groups,” Journal of Conflict Resolution 56, no. 4 (2012): 604–28; Emily Kalah Gade, Mohammed Hafez, and Michael Gabbay, “Fratricide in Rebel Movements: A Network Analysis of Syrian Militant Infighting,” Journal of Peace Research 56, no. 3 (2019): 321–35; Pischedda, “Wars within Wars”.

6 Stathis Kalyvas, “Ethnic Defection in Civil War,” Comparative Political Studies 41, no. 8 (2008): 1043–68; Paul Staniland, “Between a Rock and a Hard Place: Insurgent Fratricide, Ethnic Defection, and the Rise of Pro-State Paramilitaries,” Journal of Conflict Resolution 56, no. 1 (2012): 16–40; Lee Seymour, “Why Factions Switch Sides in Civil Wars,” International Security 39, no. 2 (2014): 92–131; Fjelde and Nilsson, “Rebels against Rebels”; Christia, Alliance Formation in Civil Wars; Woldemariam, Insurgent Fragmentation in the Horn of Africa.

7 See their introduction to this special issue.

8 Pischedda, “Wars within Wars”.

9 Daniel Posner, “The Political Salience of Cultural Difference: Why Chewas and Tumbukas are Allies in Zambia and Adversaries in Malawi,” American Political Science Review 98, no. 4 (2004): 529–45; Kanchan Chandra and Steven Wilkinson, “Measuring the Effect of ‘Ethnicity’,” Comparative Political Studies 41, no. 4–5 (2008): 515–63.

10 Barry Posen, “The Security Dilemma and Ethnic Conflict”, Survival 35, no. 1 (1995): 27–47; Chaim Kaufmann, “Possible and Impossible Solutions to Ethnic Civil Wars,” International Security 20, no. 4: 136–75.

11 Lacher, Libya’s Fragmentation, 60–4; Charles Tilly, The Politics of Collective Violence, (Cambridge, UK: Cambridge University Press, 2003); Randall Collins, “C-Escalation and D-Escalation: A Theory of the Time-Dynamics of Conflict,” American Sociological Review 77, no. 1 (2012): 1–20; Max Bergholz, Violence as a Generative Force: Identity, Nationalism, and Memory in a Balkan Community (Ithaca, NY: Cornell University Press, 2016); Jana Krause, Resilient Communities: Non-Violence and Civilian Agency in Communal War, (Cambridge, UK: Cambridge University Press, 2018).

12 Elisabeth Jean Wood, “The Social Processes of Civil War: The Wartime Transformation of Social Networks,” Annual Review of Political Science 11, no. 1 (2008): 539–61.

13 Noah Friedkin, “Social Cohesion,” Annual Review of Sociology 30 (2004): 409–25.

14 Ana Arjona, Rebelocracy (Cambridge, UK: Cambridge University Press, 2016); Oliver Kaplan, Resisting War: How Communities Protect Themselves (Cambridge, UK: Cambridge University Press, 2017); Jori Breslawski, “The Social Terrain of Rebel Held Territory,” Journal of Conflict Resolution 65, no. 2–3 (2021), 453–79.

15 But see Krause, Resilient Communities.

16 Lacher, Libya’s Fragmentation, 59–101.

17 For similar conceptualizations, see Petersen’s concept of communities or Mazur’s ‘dense local networks’: Roger Petersen, Resistance and Rebellion: Lessons from Eastern Europe (Cambridge, UK: Cambridge University Press, 2001); Kevin Mazur, Revolution in Syria: Identity, Networks, and Repression (Cambridge, UK: Cambridge University Press, 2021).

18 Lacher, Libya’s Fragmentation, 151–72.

19 Lacher, Libya’s Fragmentation, 104–33; 151–72. For similar findings from northern Mali, see Denia Chebli, “La révolte en heritage: Militantisme en famille et fragmentation au Nord-Mali (MNLA),” Cahiers d’Études africaines 59, no. 2 (2019) : 453–81.

20 Brian McQuinn, “After the Fall: Libya’s Evolving Armed Groups,” Small Arms Survey Working Paper 12, 2012.

21 Wolfram Lacher and Peter Cole, “Politics by Other Means: Conflicting Interests in Libya’s Security Sector,” Small Arms Survey Working Paper 20, 2014.

22 Wolfram Lacher and Alaa al-Idrissi, “Capital of Militias: Tripoli’s Armed Groups Capture the State,” Small Arms Survey Briefing Paper, June 2018).

23 Andrea Carboni and James Moody, “Between the Cracks: Actor Fragmentation and Local Conflict Systems in the Libyan Civil War,” Small Wars and Insurgencies 29, no. 3 (2018): 456–90.

24 UN Security Council 2019, Final Report of the Panel of Experts on Libya Established Pursuant to Security Council Resolution 1973 (2011), S/2019/914, December 9, 2019.

25 Emadeddin Badi, Exploring Armed Groups in Libya: Perspectives on Security Sector Reform in a Hybrid Environment, (Geneva: Geneva Centre for Security Sector Governance, 2020).

26 Badi, Exploring Armed Groups in Libya, 8–13, 43–56; Lacher, Libya’s Fragmentation.

27 Lacher, Libya’s Fragmentation, 104–33. For parallels in the Abkhaz army, see Anastasia Shesterinina, “In and Out of the Unit: Social Ties and Insurgent Cohesion in Civil War,” Households in Conflict Network Working Paper 311, August 2019.

28 Lacher, Libya’s Fragmentation, 104–9.

29 Lacher, “A Most Irregular Army,” 7.

30 Khalifa Haftar, interview with al-Naba’TV, October 2013, https://www.youtube.com/watch?v=Ipn8r098TVA; Khalifa Haftar, interview with Libya International Channel, June 2013, https://www.youtube.com/watch?v=ABRGhuKKjCg (all links accessed September 6, 2021).

31 Haftar, interview with al-Naba TV; Ashour al-Shamis, interview with al-Tanasuh TV, May 18, 2019, https://www.youtube.com/watch?v=GH8Jde5EJ3s.

32 Lacher and Cole, “Politics by Other Means,” 43–50.

33 Khalifa Haftar, televised statement, February 14, 2014, https://www.youtube.com/watch?v=R5uXd3oGDTw.

34 David Kirkpatrick, “In Libya, a Coup. Or Perhaps Not,” New York Times, 14 February 2014.

35 Hanan Salah, “Counting the Dead in Benghazi,” Foreign Policy, 6 June 2014.

36 Frederic Wehrey, “The Battle for Benghazi,” The Atlantic, 28 February 2014. For the meaning and political significance of tribes in post-2011 Libya, see Lacher, Libya’s Fragmentation, 67–73, 178–90.

37 Salem al-Obeidi, “حفتر يزور المرج ويلتقي بأهالي المدينة وضباط الجيش” [Haftar visits al-Marj and meets with people of the city and army officers], al-Wasat, 6 March 2014, http://alwasat.ly/news/libya/6381; Khalifa Haftar, speech in Qasr Libya, 8 March 2014, https://www.youtube.com/watch?v=2i2W8KyDs8M.

38 Mohamed al-Hejazi, interview on al-Shahed, episode 5, Libya al-Ahrar TV, 10 July 2019, https://libyaalahrar.tv/2019/07/10/5-/; Al-Hejazi, Interview on al-Shahed, Episode 6, Libya al-Ahrar TV, 11 July 2019, https://libyaalahrar.tv/2019/07/11/الشاهد-مع-القيادي-المنشق-عن-عملية-الكر6-

39 Lacher and Cole, “Politics by Other Means,” 26–7, 60.

40 Lacher, “A Most Irregular Army,” 9–10.

41 Khalifa Haftar, Press Conference, 17 May 2014, https://www.youtube.com/watch?v=C4z2d40CK9M; Al-Wasat, “حفتر يعلن استمرار عملية كرامة ليبيا في بنغازي” [Haftar announces the continuation of Operation Dignity Libya in Benghazi], 17 May 2014, http://alwasat.ly/news/libya/16619.

42 Al-Sharq al-Awsat, “حفتر: هدفنا تنظيف ليبيا من الإخوان المسلمين” [Haftar: Our Goal is to Purge the Muslim Brotherhood from Libya], 20 May 2014, https://aawsat.com/home/article/100236.

43 BBC News, “Libya Clashes between Rival Militias in Benghazi,” 16 May 2014; Mary Fitzgerald, “Libyan Renegade General Khalifa Haftar Claims He is Winning his War,” Guardian, 24 June 2014; Al-Hejazi, Episode 6.

44 Frederic Wehrey, “Ending Libya’s Civil War: Reconciling Politics, Rebuilding Security,” Carnegie Paper, Carnegie Endowment for International Peace, September 2014.

45 Amnesty International, Benghazi’s Descent into Chaos. Abductions, Summary Killings and Other Abuses (London: Amnesty International, January 2015), 35–6.

46 Ashraf al-Mayyar, Televised Statement, 10 June 2014, https://www.youtube.com/watch?v=r4v_sKiYDNs.

47 Faraj al-Barassi, public Comments Recorded on Video in March or April 2015, https://www.youtube.com/watch?v=F4bpy_GkW40; Mohamed al-Hejazi, Interview on al-Shahed, Episode 9, Libya al-Ahrar TV, 23 July 2019, https://libyaalahrar.tv/2019/07/23/9-/ ; Frederic Wehrey, “‘Whoever Controls Benghazi Controls Libya’,” The Atlantic, 1 July 2017.

48 Badi, Exploring Armed Groups in Libya, 24–6; Lacher and Cole, “Politics by Other Means,” 55–7.

49 Ahmed Elumami and Ulf Laessing, “Gunmen Loyal to Ex-General Storm Libyan Parliament, Demand Suspension”, Reuters, May 18, 2014; Al-Wasat, “تحالف القوى الوطنية يعلن تأييده كرامة ليبيا”, 21 May 2014, http://alwasat.ly/news/libya/17295.

50 Lacher and Cole, “Politics by Other Means”; Lacher, Libya’s Fragmentation, 34–9.

51 David Kirkpatrick, “Strife in Libya Could Presage Long Civil War,” New York Times, 24 August 2014.

52 Reuters, “Libya’s Ex-Parliament Reconvenes, Appoints Omar al-Hasi as PM,” 25 August 2014.

53 Salem al-Obeidi, “مجلس النواب: فجر ليبيا وأنصار الشريعة جماعات إرهابية” [Parliament: Dawn of Libya and Ansar al-Sharia are terrorist groups], al-Wasat, 24 August 2014, http://alwasat.ly/news/libya/30200.

54 David Kirkpatrick and Eric Schmitt, “Arab Nations Strike in Libya, Surprising U.S.,” New York Times, 25 August 2014.

55 Jum’a Hamdallah, “رئيس الأركان الليبي: السيسي وعدنا بدعم الجيش” [Libyan Chief of Staff: al-Sisi Promised Us to Support the army], al-Masry al-Youm, 26 August 2014, https://www.almasryalyoum.com/news/details/509942.

56 UN Security Council, Final report of the Panel of Experts established pursuant to resolution 1973 (2011), S/2015/128, 23 February 2015.

57 Wehrey, “Whoever Controls Benghazi”.

58 Mohamed al-Hejazi, interview on al-Shahed, episode 7, Libya al-Ahrar TV, 16 July 2019, https://libyaalahrar.tv/2019/07/16/7-الشاهد-مع-القيادي-المنشق-عن-عملية-الكر/; Amnesty International, Benghazi’s Descent into Chaos, 9.

59 Amnesty International, Benghazi’s Descent into Chaos, 16–20; Al-Hejazi, Episode 7; Isabel Debre, “US Court Hears Torture Case against Top Libyan Commander”, AP, 12 June 2020.

60 Khaled Bulghib, Statement on Video Posted Online on October 26, 2014, https://www.youtube.com/watch?v=yrAxhaS3Xys; Ayad al-Fsay, statement on video posted online in February 2015, https://www.youtube.com/watch?v=LRx_lMxH9QI; also see Office of the High Commissioner for Human Rights, Investigation by the Office of the United Nations High Commissioner for Human Rights on Libya: detailed findings, A/HRC/31/CRP.3, 15 February 2016, 33.

61 Faraj al-Barassi, statement on video posted online on November 2, 2014, https://www.facebook.com/watch/?v=1628079944086081.

62 Amnesty International, Benghazi’s Descent into Chaos, 30–4.

63 Human Rights Watch, “Libya: Displaced Benghazi Families Prevented From Return,” 1 February 2018, https://www.hrw.org/news/2018/02/01/libya-displaced-benghazi-families-prevented-return.

64 Voice of America, “IS Militants Claim Deadly Bombings in Libya,” 20 February 2015, https://www.voanews.com/africa/militants-claim-deadly-bombings-libya; Al-Hejazi, episode 10.

65 House of Representatives, Law No. 1 (2015), Tobruk, 2 March 2015; House of Representatives, Decree of the Supreme Commander of the Armed Forces No. 20 (2015), Tobruk, 2 March 2015.

66 Lacher, “A Most Irregular Army,” 15–16.

67 See, for example, Khaled Bulghib, statements on video posted online in May 2015, https://www.youtube.com/watch?v=PKfvt1XONTY.

68 Albayda Media Center, , 22 January 2015, https://www.facebook.com/1605781436315932/photos/-1665839100310165/العقيد-فرج-البرعصي-م/

69 Lacher, Libya’s Fragmentation, 129–32.

70 International Crisis Group, The Libyan Political Agreement: Time for a Reset (Brussels: International Crisis Group, 2016).

71 Lacher, Libya’s Fragmentation, 187–8.

72 Libya al-Mostakbal, “اقعيم يتّهم: لدينا ملفات كبيرة عن تصفيات بنغازي” [Ga’im Accuses: We Have Large Files on the Benghazi Assassinations], 5 June 2016, http://archive2.libya-al-mostakbal.org/news/clicked/97932.

73 Nathalie Guibert, “La France Mène des Opérations Secrètes en Libye,” Le Monde, 23 February 2016.

74 UN Security Council, Final Report of the Panel of Experts in Accordance with Paragraph 13 of Resolution 2278 (2016), S/2017/466, 1 June 2017; Frederic Wehrey, “‘This War Is Out of Our Hands’: The Internationalization of Libya’s Post-2011 Conflicts From Proxies to Boots on the Ground,” New America, 20 September 2020, https://www.newamerica.org/international-security/reports/this-war-is-out-of-our-hands/.

75 Rami Musa, “Defying Critics, Libya’s Eastern Bank Prints its Own Money,” AP, 27 May 2016; International Crisis Group, Of Tanks and Banks: Stopping a Dangerous Escalation in Libya (Brussels: International Crisis Group, May 2019).

76 Al-Wasat, “الحبري: ثلث ميزانية الإنفاق خلال السنوات الثلاث الماضية جرى تسييلها لفائدة الجيش” [Al-Hibri: A Third of the Budget during the Past Three Years Was Spent for the Benefit of the Army], 19 February 2019, http://alwasat.ly/news/libya/236465.

77 UN Security Council, Final Report of the Panel of Experts on the Sudan Established Pursuant to Resolution 1591 (2005), S/2017/22, 9 January 2017.

78 Lacher, “A Most Irregular Army,” 18.

79 UN Security Council, Report of the Secretary-General on the United Nations Support Mission in Libya, S/2016/1011, 1 December 2016.

80 Lacher, Libya’s Fragmentation, 188.

81 Libya al-Mostakbal, “بنغازي.. قتل ضابطين وإيقاف آخرين من الكتيبة 204 دبابات” [Benghazi: Two Officers of the Tank Battalion 204 Killed and Others Arrested], 6 November 2016, http://www.libya-al-mostakbal.org/top/9590/--ضابطين-وإيقاف-آخرين-من-الكتيبة-204-دبابات.html.

82 Lacher, Libya’s Fragmentation, 189.

83 Human Rights Watch, “Libya: Videos Capture Summary Executions,” 16 August 2017, https://www.hrw.org/news/2017/08/16/libya-videos-capture-summary-executions.

84 Al-Wasat, “مقتل «سلفي» وإصابة آخرين في ثاني عملية اقتحام لمنزل العقيد فرج البرعصي“[A “Salafi” Was Killed and Others Injured in the Second Storming of Colonel Faraj al-Barassi’s Residence], 15 February, 2017, http://alwasat.ly/news/libya/124775.

85 Al-Wasat, “مقتل أحد شيوخ قبيلة العواقير ابريك اللواطي في تفجير سيارة مفخخة بسلوق” [One of the Elders of the Awagir Tribe, Brik al-Lawati, Killed in a Car Bomb Explosion in Suluq], 19 May 2017, http://alwasat.ly/news/libya/133300.

86 Human Rights Watch, “Libya: Mass Extra-Judicial Execution”, 29 November 2017, https://www.hrw.org/news/2017/11/29/libya-mass-extra-judicial-execution.

87 Wehrey, “‘This War Is Out of Our Hands’,” 23–5.

88 Lacher, “A Most Irregular Army,” 20–1.

89 Wehrey, “‘This War Is Out of Our Hands’,” 21.

90 Frederic Wehrey and Anouar Boukhars, Salafism in the Maghreb: Politics, Piety, and Militancy, (Oxford: Oxford University Press, 2019), 125; Video Statement by Battalion 128 in Ubari, 9 June 2020, https://www.facebook.com/watch/?v=1643149595835613; Khalifa Haftar, speech to the 10^th Battalion of the Tareq ben Ziyad Brigade, 2 August 2020, https://www.youtube.com/watch?v=-d8Z2gkoRnQ.

91 Lacher, “A Most Irregular Army,” 20–1.

92 Tim Eaton, The Libyan Arab Armed Forces: A Network Analysis of Haftar’s Military Alliance (London: Chatham House, 2021), 33–5; Tim Eaton et al., The Development of Armed Groups in Libya since 2014: Community Dynamics and Economic Interests, (London: Chatham House, 2020), 29–34; Noria Research, Predatory Economies in Eastern Libya: The Dominant Role of the Libyan National Army, Global Initiative Against Transnational Organized Crime, June 2019, https://globalinitiative.net/wp-content/uploads/2019/06/GITOC-Predatory-Economies-Eastern-Libya-WEB.pdf.

93 Mirko Keilberth and Fritz Schaap, “Der Warlord“, Der Spiegel, 30 August 2019; David Kirkpatrick, “A Police State with an Islamist Twist: Inside Hifter’s Libya,” New York Times, 20 February 2020.

94 Wehrey, “‘This War Is Out of Our Hands’,” 21.

95 Televised interview with Faraj Ga’im, 218 TV, 10 November 2017, https://www.facebook.com/watch/?v=910522915765216.

96 Seraj al-Din Abdelhamid, “فرج قعيم. ما الذي يخشاه حفتر من النقيب الشاب في ليبيا؟” [Faraj Ga’im: What Does Haftar Fear from the Young Captain in Libya?], Arabi21, 14 November 2017, https://arabi21.com/story/1048952/فرج-قعيم-ما-الذي-يخشاه-حفتر-من-النقيب-الشاب-في-ليبيا.

97 218TV, “إطلاق سراح فرج قعيم” [The Release of Faraj Ga’im], 19 August 2018, https://www.218tv.net/إطلاق-سراح-فرج-قعيم/; Salem al-Obeidi, “جهاز مكافحة الظواهر السلبية يبدأ حملته الأمنية في بنغازي” [The Anti Negative Phenomena Agency Begins Its Security Campaign in Benghazi], al-Wasat, 18 August 2019, http://alwasat.ly/news/libya/254540.

98 Lacher, “A Most Irregular Army,” 22.

99 Ibid.

100 International Crisis Group, “New Libyan Militia’s Oil Strike Risks Wider Conflagration,” International Crisis Group Commentary, 10 March 2017, https://www.crisisgroup.org/middle-east-north-africa/north-africa/libya/new-libyan-militias-oil-strike-risks-wider-conflagration.

101 International Crisis Group, After the Showdown in Libya’s Oil Crescent (Brussels: International Crisis Group, August 2018).

102 Human Rights Watch, “Libya: Mass Executions Alleged at Military Base”, 21 May 2017, https://www.hrw.org/news/2017/05/21/libya-mass-executions-alleged-military-base.

103 Reuters, “East Libyan forces Take Desert Air Base As They Push West,” 3 June 2017.

104 Afrigatenews, “تسريبات عن تحركات الجيش داخل خليج سرت” [Leaks about the Movements of the Army in the Gulf of Sirte], 5 March 2017, https://www.afrigatenews.net/article/تسريبات-عن-تحركات-الجيش-داخل-خليج-سرت/; Mohamed Mansour, “ليبيا… الهلال النفطي من جديد” [Libya: the oil crescent again], Almayadeen, 21 June 2018, https://www.almayadeen.net/news/politics/887495/ليبيا–-الهلال-النفطي-من-جديد

105 Lacher, Libya’s Fragmentation, 159–63, 171–2.

106 Lacher, Libya’s Fragmentation, 167–70.

107 Author interviews, local observers from Bani Walid, Bani Walid and Tripoli, April 2018 and February 2019.

108 Author interviews, local observer from Tarhuna, Tripoli, February 2019; Afrigatenews, “إكتشف 4 معلومات عن اللواء 22 مشاة بترهونة” [Discover four facts about the 22nd Infantry Brigade in Tarhuna], 6 September 2018, https://www.afrigatenews.net/article/إكتشف- 4 -معلومات-عن-اللواء-22-مشاة-بترهونة/

109 Author interviews, local observer from Ubari and army officer active in the area, Tripoli, March/April 2018.

110 Eaton et al, “The Development of Libyan Armed Groups Since 2014,” 41; Badi, Exploring Armed Groups in Libya, 35–6.

111 Reuters, “Civilians Killed as Armed Groups Clash in Libyan South”, 28 February 2018.

112 The Libyan Address, “القوات المسلحة تلبي نداء الجنوب لتخليصه من القوى الإرهابية والإجرامية”, 15 January 2019, https://www.addresslibya.co/ar/archives/52273; author interviews, observers from southern Libya, Tripoli, February 2019.

113 Author interviews, observers from southern Libya, Tripoli, February 2019.

114 Reuters, “Eastern Libyan Forces Fight Tribesmen for Southern City,” 21 February 2019.

115 UN Security Council, United Nations Support Mission in Libya: Report of the Secretary General, S/2019/682, 26 August 2019, 4.

116 Jérôme Tubiana and Claudio Gramizzi, Lost in Trans-Nation: Tubu and Other Armed Groups and Smugglers along Libya’s Southern Border (Geneva: Small Arms Survey, 2018), 34.

117 Eaton, Libyan Arab Armed Forces, 20–3.

118 Wolfram Lacher, “Who Is Fighting Whom in Tripoli? How the 2019 Civil War is Transforming Libya’s Military Landscape,” Small Arms Survey Briefing Paper, August 2019, 4–5.

119 Afrigatenews, “منتسبو اللواء 106 مجحفل بغريان يباركون عمليات الجيش في الجنوب” [Members of the 106th Brigade in Gharyan Praise the Operations of the Army in the South], 21 February 2019, https://www.afrigatenews.net/article/منتسبو-اللواء-106-مجحفل-بغريان-يباركون-عمليات-الجيش-في-الجنوب/.

120 Frederic Wehrey, “A Libyan Revenant,” Newlines Magazine, 14 July 2021, https://newlinesmag.com/first-person/a-libyan-revenant/.

121 Lacher, “Who Is Fighting Whom,” 10.

122 Wehrey and Boukhars, Salafism, 127–8.

123 David Kirkpatrick, “The White House Blessed a War in Libya, but Russia Won It,” New York Times, 14 April 2020.

124 Author interviews, GNA commanders and officials, Tripoli, October 2019.

125 International Crisis Group, Turkey Wades into Libya’s Troubled Waters (Brussels: International Crisis Group, 2020).

126 Samer al-Atrush, “What Lies Behind the Nine Years of Turmoil in Libya,” Bloomberg, 26 August 2020; Patrick Haimzadeh, “En Libye, le maréchal Haftar accumule les revers, la France aussi,” Orient XXI, 4 June 2020, https://orientxxi.info/magazine/en-libye-le-marechal-haftar-accumule-les-revers-la-france-aussi,3913.

127 Author telephone interviews, senior GNA officials, January 2020; see also Wehrey, “‘This War Is Out of Our Hands’,” 32.

128 Bloomberg, “How Putin Was Thrown Off Course by a Furious Libyan General,” 15 January 2020.

129 Declan Walsh, “Waves of Russian and Emirati Flights Fuel Libyan War, U.N. Finds,” New York Times, 3 September 2020.

130 Lacher, “Who Is Fighting Whom,” 12–15.

131 Ibid., 15; Jalel Harchaoui, “Tarhuna, Mass Graves, and Libya’s Internationalized Civil War,” War on the Rocks, 30 July 2020, https://warontherocks.com/2020/07/tarhuna-mass-graves-and-libyas-internationalized-civil-war/.

132 Wehrey, “‘This War Is Out of Our Hands’,” 34–6.

133 Lacher, “A Most Irregular Army,” 31; Eaton, Libyan Arab Armed Forces, 30–7.

134 Ibid.

135 Sarah Vernhes, “Violences à Benghazi: Haftar perd-il le contrôle de ses forces?,” Jeune Afrique, 4 April 2021; Tarek Megerisi, “Plot Twist: How Europe Should Deal with Libya’s New Government,” ECFR Commentary, 12 February 2021, https://ecfr.eu/article/plot-twist-how-europe-should-deal-with-libyas-new-government/.

136 Eaton, Libyan Arab Armed Forces, 38–9.

137 Tilly, The Politics of Collective Violence; Wood, “Social Processes”; Bergholz, Violence as a Generative Force; Collins, “C-Escalation”.

138 For critiques, see Posner, “Political Salience”; Chandra and Wilkinson, “Measuring the Effect”.

139 Stathis Kalyvas and Matthew Kocher, “Ethnic Cleavages and Irregular War: Iraq and Vietnam”, Politics & Society 35, no. 2 (2007): 183–223; Seymour, “Why Factions Switch Sides,” 129.

140 Nicolas Desgrais, Yvan Guichaoua, and Andrew Lebovich, “Unity is the Exception: Alliance Formation and De-Formation among Armed Actors in Northern Mali,” Small Wars & Insurgencies 29, no. 4 (2018): 654–79; Lacher, Libya’s Fragmentation.