The significant difference in feelings of safety before and after the revolution begs the question of how the security vacuum was filled when the former regime collapsed, and by whom. Analysis of the evolution of local security governance since the revolution and the survey data show three interlinked characteristics that define security provision in Libya today: it is largely in the hands of local (and non-state) actors; it follows several patterns or modalities across the country; and it displays a significant measure of plurality. This section explores each of the three characteristics.
Local
During the revolution against Qadhafi, local councils arose in the ‘liberated’ areas and took care of city affairs. Libyans welcomed the change, because for the first time they could handle basic bureaucratic services in their own city or district – services that has been centralised in Tripoli under the Qadhafi regime.[29] In 2012 Libya’s interim government (the National Transitional Council, NTC) passed law 59 on public administration in response to popular demand for further decentralisation and better provision of public services. Approximately 100 municipal councils were elected to serve their communities.
However, in the following years, a combination of the further breakdown of security, political contestation, and woeful underfunding left municipal councils struggling to fulfil their obligations and to maintain their political legitimacy.[30] At the same time, Libyans placed high expectations on their municipality to provide services and economic opportunities. As a rentier state, Libya relies almost entirely on revenues from oil, and in turn local councils depend fully on central government to cover their expenses (instead of, for example, on tax collection). Such dependency quickly drained the state budget, which was largely used to pay salaries instead of improving public services and funding municipalities.[31] At present, municipal councils lack the mandate and resources to (fully) govern territories formally under their control. Many councils have little to no budget available, and often no jurisdiction to handle governance affairs even though they are often expected to do so by local residents. A respondent from Sabratha answered in the survey: ‘I trust the municipal council [to provide for the local population] because it is an elected body, but it is weak.’[32]
Attempts to improve local governance and the jurisdiction of municipal councils were also undermined by the burgeoning number of militias following the 2011 revolution and the government’s failed attempts to integrate them into a collapsed security sector. The fateful decision by Libya’s new leaders to place militiamen on the government payroll as interim security providers and use them to guard key infrastructure[33] not only depleted the state budget (recruitment into armed groups skyrocketed due to government payrolling), it also further nested armed groups in local political economies and increased armed groups’ relevance in the daily management of the municipality. Moreover, militias that were not able to legitimise themselves through central government turned to their municipalities for money – for example, by demanding payment for local protection services.
Survey findings demonstrate the sheer variety of local governance providers as well as the often restricted role of municipal councils compared to other local actors. When asked who is the main governance provider in the municipality, the majority of respondents say that local forces (38%) control the municipality, directly followed by the security directorate (31%, see figure 5). Clearly, national actors (GNA and LNA) have little control locally compared to local power brokers. As figure 6 shows, respondents say that the protection of assets, infrastructure and individuals is in the hands of a mix of stakeholders and they see the municipal council as only one among many. When broken down per municipality, we note that the survey findings support the notion that the GNA is a dominant security provider only in Tripoli (see figure 7) but that governance in other municipalities is far more diverse.
Patterns
Because the local power dynamics that affect the way in which governance is organised are different in every locality, security provision in Libya is patterned: however, several ‘modalities’ in security provision co-exist and intertwine. Across the country we find a mix of militia rule, tribal authority and military governance. These are not mutually exclusive or, in the case of militia and tribes, confined to specific regions, but are key to understanding the complex, layered and network-based reality of security provision in Libya.
After the revolution, the fragmentation of territory and power was most intense in Libya’s northwest, in the large urban areas of Tripoli and Misrata, and in the surrounding Warshafana and Nafusa Mountains regions. There, deepening rifts between local armed elites, tribes, clans and families have fragmented territory and consolidated the hold of these groups over their area through the barrel of the gun and forced displacements. It is in this (most densely populated) part of Libya where militia rule created a patchwork of ‘city-states’ like Misrata, Tripoli, Gharyan, Zintan and Zawiya where (coalitions of) local armed actors and militias – some with a tribal character – hold sway in a hard-fought power equilibrium. As a respondent from Misrata pointed out: ‘In my area, money and arms create the most powerful actors.’
Libya’s militia ‘culture’ is entangled with tribalism: a historical social phenomenon and age-old way of structuring society.[34] Libya has around 140 clans and families that extend beyond geographical borders, and around 30 influential tribes.[35] At the top are the Warfalla, Tarhouna, Qadhadfa and Magarha tribes. A number of tribes that were less relevant under the previous regime, like the Tebu and Zintan tribes, have acquired and displayed particular power in post-revolution Libya. In much of Libya’s modern history, political power ascribed to tribes has been restricted by subsequent leaderships. Qadhafi was clever in undoing the power of notable tribal leaders and exploiting tribal differences as a means to keep tribes in control, but he also mixed the importance of tribal identity with nationalist discourse in favour of his own regime.
Tribalism is present across the entire country, but the extent to which tribal identity retains the ability to mobilise groups socially and politically in Libya today is debatable and connected to local power dynamics. Some observers point to the re-emergence of tribal leaders in the public sphere since the revolution, particularly in the field of conflict resolution,[36] whereas others argue against overstating the influence of tribalism in present-day Libya due to counterforces such as urbanisation, population growth, globalisation, and regional and religious movements.[37] Nonetheless, tribal actors were among those who filled the governance and security vacuum that followed the Qadhafi era, and asserted themselves within the fragmented security and justice space. Many Libyans appreciate tribes for their peace-making efforts and for restoring local relations. At the same time, however, in post-revolution power struggles, almost all Libyan families armed themselves and tribes became militarised. With the collapse of central state security and political polarisation after 2014, the armed capacities of tribes became even more pronounced and tribes inserted themselves into local power disputes and armed conflict. [38]
Tribalism is also a strong characteristic of east Libya (Cyrenaica province), but there the influence of tribes and militias is overlaid with military government. In August 2016 the Libyan National Army’s Chief of Staff Abdelraziq al-Nazouri (working under the direct command of Khalifa Haftar) started a process of replacing democratically elected mayors with military commanders in several towns and cities in east Libya. The replacement of elected councillors with appointed military governors raised much concern about a permanent military takeover. Both in Libya and by the international community, the replacements were widely understood as regression on the path of democratisation and a step away from achieving a unified civilian government.
Although internationally criticised for creating a military junta, many citizens in the east seemingly welcomed the LNA’s move, as the elected municipal councils had failed to deliver basic services, including the provision of security. Under de facto military rule, citizens expected a more secure and responsive system of local governance.[39] Indeed, for example in Benghazi, the elected governor established effective cooperation with LNA security forces. Haftar also carefully and effectively crafted a web of tribal alliances because he realised he could only advance with the weight of tribal notables behind the LNA. In Tobruk and elsewhere, this had the effect of tribal elders themselves asking the LNA for a military commander – although they had appreciated the work of the municipal council, they argued that the situation in the city called for a military leader to restore safety.[40]
Yet Haftar and Al-Nazouri’s efforts to bring east Libya under military rule are more complex than they appear. First, because the LNA itself suffers from tribal and political divisions, Haftar’s move to concentrate power in the hands of a loyal few caused resentment with other groups who had participated in his military campaign. Haftar crafted a web of tribal alliances to expand his control in the east, but tribal leaders allegedly are increasingly questioning Haftar’s tribal legitimacy, as he is a member of the western Farjani tribe. The Awaqir tribe from Benghazi, for example, exercises significant power in the city and has overruled Haftar before. For instance, when former GNA Deputy Minister Faraj Qaim from the Awaqire tribe was arrested by the LNA, brewing tensions between the LNA and the Awaqir over his arrest, including a bloody attack on the LNA’s Benghazi base on 23 January, led to his release. Similarly, Madkhali Salafi groups contributed to Haftar’s military campaign but have now started to claim a Salafi footprint in the public domain in return – which fuels tensions with other factions within the LNA who thought they had ousted, not unleashed, Islamists from the east.[41] In any case, as local sources also suggest, armed groups that operate on the LNA’s side are not necessarily tied to Haftar’s cause for ideological reasons, but rather out of pragmatism or convenience.
Second, the overnight appointment of military governors at municipal level did not suddenly change local power dynamics. Local armed groups, including jihadists and tribal militias, would not give up their territory, interests and constituencies to a uniformed LNA commander. Testimony to this was the battle for Benghazi, which Haftar claimed would only last ‘weeks’, but which became a three-year campaign because an array of local armed groups fiercely fought Haftar’s forces. The majority of those were local militias who still retain power in the city despite efforts to disband them.[42] It is probably due to the LNA’s internal strife and its inability to put more remote areas under its control through local forces that Haftar enlisted Chadian and Darfuri militias as mercenaries.[43] As a local source suggests: ‘South Libya is a recruitment base for all of the armed groups. Everybody who has seed funding – even if it is just one vehicle – tries to consolidate its power by enlisting southern mercenaries.’[44]
Plurality
In recent years, debates on how to engage in fragile and conflict-affected states have challenged a conventional, state-centric take on security provision. Many conflict-ridden contexts display modes of security provision that took shape because the state was absent, failing or harmful to its citizens. Instead, in fragile and conflict-affected settings, security providers directly authorised by the state (such as the police and the regular army) tend to co-exist with a multitude of other coercive actors. This is the case in Libya as well, which – due to the state’s inability to fully meet citizens’ needs for security and that alternative actors have stepped in to fulfil that need – is best understood as a context dominated by plural security actors – ‘actors characterised by the ability and willingness to deploy coercive force, lack of integration into formal state institutions, and an organisational structure that persists over time, that seek to ensure the maintenance of communal order, security and peace through elements of prevention, deterrence, investigation of breaches, and punishment’.
Non-state security providers may acquire legitimacy by proving to be more effective and efficient, proximate and relevant to local populations. However, plural security actors are frequently associated with human rights violations, perverse interface with the state, difficulty in providing security equitably in diverse contexts, and an almost ineluctable tendency towards net production of insecurity over time.[45] ‘The panorama of plural security providers can include militias, warlords, customary authorities, political authorities, organized criminal groups, neighborhood gangs, improvised community watch patrols, etc.’[46]
In Libya, such an array of actors such as militias and tribal armed factions also operates simultaneously and with varying relationships with the state and with each other. As discussed in more detail below, the security realm is controlled by multiple (categories of) armed actors, of which only few are affiliated with the government in Tripoli and the administration in Tobruk. Unable to call upon the state as the sole guarantor of their security, Libyans say they have to ‘hustle for protection’[47] from local state and non-state groups, tapping into their own networks and connecting to the armed groups that control their area. The situation of plural security is typical in Libya today, but as we have seen above it does not contribute to perceptions of safety. Survey findings show that respondents generally do not feel safe (see figure 2) but that in municipalities with the least plural security context (such as those with only one security provider) respondents feel safer (see figure 4).