International Recognition Meets Areas of Limited Statehood: Practices and Effects on Hybrid Actors in Post-2011 Libya

ABSTRACT

This article examines the impact of international recognition on intrastate conflict contexts and areas of limited statehood. We conceptualise international recognition-through-interaction in social-relational, process-oriented, non-dualistic and performative (practice) terms. We theorise plural effects beyond the government vs. rebels and conflict causation vs. transformation binaries. Based on two case studies on post-2011 Libya’s security/armed and migration governance actors, including original interviews, we show that the most distinctive power of international recognition-through-interaction lies in drawing (sovereignty) lines. Material empowerment effects are prominent, though only contextually subject to formal international recognition. Identity transformation remains partial and political legitimacy is influenced in complex ways.

KEYWORDS:

• International recognition
• recognition theory
• limited statehood
• civil wars
• Libya

View correction statement:

Correction

Introduction

Several years into the intermittent post-2011 civil war in Libya, international actors of all sorts expressed a shared puzzlement, if not hopelessness, about the amorphous condition of the conflict and the lack of clarity of the narrative(s) and the players involved. ‘The country was not divided into two or three parties, but it imploded into dozens of autonomous local entities’, regretted Ghassan Salamé before resigning as the head of the United Nations Support Mission in Libya (UNSMIL) (Le Monde, 23 August 2019). At the same time, the fluidity of the conflict context increased its malleability, and the potential role of international interaction and recognition in shaping structure and agency therein – for better or worse. Awareness of Libya’s century-long pattern of external interventions building (up) local non-state actors in terms of identity, material power or legitimacy (Anderson 2017, 231) coincided with the recent mainstreaming, among practitioners in the humanitarian and peacebuilding communities, of the notion of ‘conflict sensitivity’ (interviews 22, 4, 9, 10, 20), defined as ‘the ability of an organisation to understand how it interacts with a context and to use this to maximise potential positive and minimise negative impacts on peace’ (Midgley et al. 2022, 1).

This article presents selected results of a research project that examined how various forms of international recognition – understood in a broad, social-theoretical sense – have affected conflict dynamics in Libya since 2011. More specifically, it is driven by the combination of a theoretical and an empirical motivation. Theoretically, based on our reading of the recognition literature in International Relations (IR) and conflict and peace studies, the gap and puzzle that arises is how to make sense of the multiple unit-level effects of (mis)recognition that go beyond causing either (violent) conflict or conflict transformation and reconciliation – a binary in need of overcoming. In other words, we remain agnostic in not assuming the precedence of either the ‘labour of the negative’ (Epstein, Lindemann, and Sending 2018) or positive teleological processes of mutual recognition (Wendt 2003), and we instead shift the focus to more diverse and contingent (micro) outcomes which have thus far received little theoretical attention.

Empirically, our data suggested that, in the Libyan context, the politics and problems of international recognition were not limited to dealings with the state’s governments and their rebel contenders, that is, to the predicaments created by the 2011 revolution and regime change as well as the successive authority splits leading to the coexistence of parallel governments and parliaments in 2014–2015, 2015–2021 and more recently since 2022. In fact, recognition contests at the governmental and parliamentary levels have spilled over to and interacted with international engagement issues concerning a broader variety of domestic actors, many of which furthermore fall through the cracks of the state vs. non-state categorisation. These hurdles pushed us to recalibrate our analysis in two ways: first, approaching the Libyan scene through the lenses of the concept of areas of limited statehood (ALS), rather than mere (violent) conflict, in order to bring to the fore the critical intricacies of statehood, its manifold shades or absence; and second, digging deeper into the international recognition, in particular, of hard-to-classify or hybrid Libyan state-non-state actors.

In short, this article addresses the question of the impact of international recognition on intrastate conflict contexts and ALS, seeking to pluralise the picture of both effects and recipients. It proceeds in four steps. Theoretically, we bring into conversation the literature on international recognition and on ALS (section 2), proposing the notion of international recognition-through-interaction to underline the indissoluble and necessary connection between social intercourse and recognition, and expounding the potential effects we may expect this to have on actors in ALS. We argue that international engagement/recognition practices in ALS and violent intrastate conflict contexts matter insofar as they differentially shape identity formation and change, material empowerment, political legitimation and (sovereignty) line-drawing among domestic (conflict) actors (section 3).

Empirically, these effects are fleshed out through two case studies concerning Libyan domestic security (section 4) and migration governance (section 5). These are the two spheres that have attracted most international attention in this country since 2011, aside from the recurrently contested international recognition of governments (Fernández-Molina under review), and two governance areas where both the structure and the actors have been generally described as hybrid and typical of ALS. While the general applicability of the concept of ALS to post-2011 Libya has already been established (Melcangi and Mezran 2022), our focus on these cases underscores the importance of studying limited statehood in its sectoral dimension, i.e. concerning specific policy areas, besides the more usual territorial and social (population) perspectives (Börzel and Risse 2021, 35–36).

The case studies cover the 2011–2019 period, which includes the three distinct episodes of civil war that Libya went through during this decade (February-October 2011, May 2014–December 2015, April 2019–October 2020)¹ – the first of them overlapping with the processes of revolution and international military intervention – as well as non-war phases devoted to stabilisation, political transition, security sector reform (SSR) and state-building efforts (October 2011–May 2014, December 2015–April 2019). In terms of methodology and data, we primarily rely on 22 semi-structured elite interviews with diplomats, international officials and major NGO staff working on Libya, which we conducted between January and March 2019 during fieldwork in Tunis, where most embassies and international organisation representations to this country were then relocated, as well as in a visit to Brussels and by phone. Interview questions addressed what these practitioners’ governments or organisations were doing in and with Libya(ns), how, why and so what (activities, timeline, interlocutors and partners, procedures, justifications, challenges, effects). Interview material is presented here in a fully anonymised manner in compliance with the majority’s preference. In addition, we triangulated factual information with a dataset of reported acts of engagement/recognition between external and Libyan actors (2011–2018) which we built drawing on a systematic search of news articles from the LexisNexis database. Regarding limitations, due to the difficulties of access and the lack of response from part of the institutions that were contacted, the range of interviewees was not exclusively but predominantly Western and UN-related, which limits the generalisability of the views collected. Also, the project did not include interviews with Libyan actors because of risk assessment and ethical constraints, particularly at the time of the outbreak of the 2019–2020 civil war. As an alternative, the discussion of domestic responses to and effects of international engagement/recognition draws on secondary, fieldwork-based literature on the Libyan conflict.

(International) recognition in violent conflict and areas of limited statehood

Social recognition is one of ‘those human needs that are common to all and whose pursuit is an ontological drive in all’ to such an extent that their denial constitutes the ‘real source of conflict’, wrote Azar in a classic in peace and conflict studies (Azar and Burton 1986, 29). This may be seen as the fundamental common denominator concealed in Holsti’s (1996, 21) contrast between typical inter- and intrastate wars in the late twentieth century: the struggle resulting from the vital necessity for every political actor, at any level, to be ‘constituted’ by others ‘as a subject with a legitimate social standing’ (Wendt 2003, 511), be it in terms of the acceptance of its very existence, respect for universal rights or esteem for a particular identity feature. This intersubjective social relation essential to the ‘formation of the practical self’ (Honneth 1995, 68) is what we understand by recognition. Hegelian-inspired thought on this matter flourished in social and political theory in the 1990s (Honneth 1995; Taylor 1994) but took time to travel to and fertilise the discipline of IR, where a narrower notion of recognition had long prevailed, dominated by legal(istic) doctrine and debates on the recognition of statehood.

In its application to conflict and peace research, broadly understood, over the past two decades, recognition scholarship has yielded important contributions in three main areas. The first of these is the study of the causes of interstate armed conflict. Lindemann (2010, 27–43) has argued (mis)recognition to be a major factor, at least as important as utilitarian motivations, driving wars ‘for a state’s universal dignity’ (in response to violations of state sovereignty), ‘for a state’s particular dignity’ (provoked by attacks on specific political or cultural identities), ‘for prestige’ (in pursuance of hubristic self-identities) and ‘by antipathy’ (facilitated by the absence of shared norms and identities). Similarly, though not necessarily with violent outcomes, Murray (2019, 6) has shown that rising power revisionism during hegemonic transitions is a social construct resulting from the emerging state’s ‘social interactions with other states as it attempts to gain recognition of its identity as a major power’. Secondly, some works on peacemaking and peacebuilding with a predominantly intrastate focus have examined the role of mutual recognition between conflict parties and communities in the transformation of intractable identity-based conflicts (Aggestam and Björkdahl 2013; O’Neill 2012; Strömbom 2014). A common distinction here has been the one between ‘thin’ and ‘thick’ recognition (Allan and Keller 2012, 76–77; Wendt 2003, 511–512). While the former typically amounts to ‘accept[ing] each other as legitimate partners to peace negotiations’ (Aggestam 2015, 510), the latter requires a more profound understanding of the other’s identity or difference which may unsettle the images of the self, the other and the boundaries between them (Strömbom 2014, 175–176). The contrast between these two strands of literature is to a great extent reflective of the opposition between markedly positive (Wendt 2003) and negative (Epstein, Lindemann, and Sending 2018) theoretical readings of the Hegelian ‘struggle for recognition’ in IR.

Thirdly, the picture gets more complicated when we straddle the international-domestic divide. Concern with the impact of external recognition on intrastate armed conflicts goes back to old legal categories such as the nineteenth-century recognition of belligerency – for violent non-state actors (VNSA) such as rebel groups controlling a significant portion of a state’s territory – and the recognition of national liberation movements during the post-World War II decolonisation era (Coggins 2015, 102–103; Peterson 2020, 209–2011). Regarding de facto governments in civil war contexts, legal doctrinal debates have long pitted the principle of non-intervention against so-called ‘premature recognition’ (Lauterpacht 1947/2013, 94–95). More recently, political science scholars of rebel governance have started to study what is arguably the reverse side of this interaction, namely rebel diplomacy (Coggins 2015; Huang 2016). However, none of the strands of literature in this third group have specifically engaged with recognition theory. The connection with the latter has only been made by a handful of recent works on the politics of recognition and VNSA. These have addressed the problem of inclusion/exclusion and the motivations for conflict parties to participate in peace mediation and negotiation processes (Aggestam 2015; Bell 2014) as well as the question as to whether, in a similar way as posited for interstate conflict, recognition of VNSA ‘may incentivise moderation in means, that is, restraint in violence’, while ‘misrecognition may be a factor accounting for the escalation of violence’ (Biene and Daase 2015, 222; see Geis, Clément, and Pfeifer 2021, 7).

A further limitation shared by all the contributions in the third category, and the empirical gap that motivates the conduct of our study, is the neglect of international engagement with, and recognition of, other domestic actors in intrastate conflict contexts that are not necessarily conflict parties or combatants, namely actors beyond the governments vs. rebels/VNSA dichotomy. In order to redress this blind spot and expand the scope of domestic actorness, we propose zooming out the picture of the domestic context in question, from the narrower lens of violent conflict or civil war to the broader phenomenon captured by the concept of ALS. ALS have been defined as ‘those aspects of a country where state authorities (such as national, regional, or local governments) lack the ability to implement and enforce rules and decisions and/or in which they do not control the use of force’ (Börzel and Risse 2021, 34). Such shortcomings can affect either of the three basic dimensions of statehood: territory, population and government/governance. Post-2011 Libya combines limited statehood in all of them: authority fractures and parallel governments have entrenched an apparent territorial East–West fragmentation, a diversity of local non-state political actors and VNSA exercise effective authority over various population groups, and economic and sectoral governance remains patchy, driven by different institutions and logics depending on the policy area (Melcangi and Mezran 2022, 123–124). At the same time, while ALS are commonly associated with violent conflict as both causes and consequences, the relationship between the two phenomena is far from being one of equivalence, as demonstrated by global empirical data (Börzel and Risse 2021, 147) as well as the non-war phases of relative stability in the recent Libyan trajectory. Finally, the widespread evidence that ALS are ‘neither ungoverned nor ungovernable’ invites to shift the analytical focus from combatant to governing actors (Börzel and Risse 2021, 3).

Theorising international recognition-through-interaction and its effects in ALS

How does international recognition then meet ALS? This section expounds on our theoretical understanding of international recognition – or what we call international recognition-through-interaction – as well as the effects we may expect it to have on actors in ALS. Our approach to international recognition is ontologically attuned with processual relationalism (Jackson and Nexon 1999) and specifically builds on the scholarship that has endeavoured to bring Hegelian recognition theory into IR (e.g. Agné et al. 2013; Daase et al. 2015; Epstein, Lindemann, and Sending 2018; Lindemann and Ringmar 2012). This social theory perspective has the advantage of overcoming the narrow, binary view of recognition inherited from international law. The longstanding focus of the latter’s doctrine and debates has been on the criteria for and the effects of the recognition of states, pitting declaratory against constitutive theories (Crawford 2006), and to a much lesser extent on the recognition of governments (Talmon 1998). While distinct in many aspects, the recognition of states and the recognition of governments are inherently linked to each other as the two sides of the coin of ‘international legal sovereignty’ (Krasner 1999, 14–16) and share an either/or nature and zero-sum effects as far as law is concerned (see Peterson 2020; Roth 2015).

By contrast, the social theory approach allows to theorise international recognition in social-relational, process-oriented, non-dualistic and performative terms (see Bouris and Fernández-Molina 2018; Fernández-Molina 2019). Firstly, following Honneth’s reading of Hegel’s ‘struggle for recognition’, recognition is inherently social-relational. Departing from the atomist conception of the self and individual-society relations prevailing in modern social philosophy (Fritzman 2014, 39–40; McBride 2013, 10–12), recognition is viewed as the fundamental mechanism that enables the ‘original intersubjectivity of human life’ and the ‘interpenetration of socialisation and individuation’ on all levels: psychological, sociological, legal and political (Honneth 1995, 29, 16). As put by Taylor (1994, 32–33), the ‘dialogical character’ of human life lies in that ‘[w]e become full human agents’ and ‘[w]e define our identity always in dialogue with, sometimes in struggle against, the things our significant others want to see in us’. This implies, we argue, that social interaction and recognition are inextricably linked to the point that the former cannot happen without some measure of the latter, be it positive or negative. Applying this reasoning to international relations, the policies of ‘engagement without recognition’ (Berg and Ker-Lindsay 2019; Ker-Lindsay 2015) that are sometimes purportedly pursued vis-à-vis contested actors, as a matter of fact, can never be totally deprived of (mis)recognition effects. Likewise, in situations of violent conflict, ‘[w]henever there is some form of engagement with [a VNSA] that goes beyond fighting it, […] there is some degree of recognition involved’ (Geis, Clément, and Pfeifer 2021, 15). Hence our use of the term recognition-through-interaction.

Secondly, going back to Honneth’s theory, recognition consistently emerges as process-oriented rather than amounting to single, specific events. This applies to its threefold incarnation in the form of ‘love’ (care), ‘respect’ (equal rights) and ‘esteem’ (difference) (Honneth 1995, 107, 114–118, 125–126). Thirdly, the fact that all these phenomena occur as gradual processes is consistent with the conception of recognition as non-dualistic or a continuum instead of an either/or alternative. In the sphere of international relations, variation in terms of substance or meaning has been nicely captured by Bartelson’s (2013) three concepts of recognition, i.e. legal, political and moral. When it comes to practices, objects and effects, the international recognition continuum runs ‘from highly formalised to extremely informal modes of recognition, and from the recognition of non-state actors and other political collectives as legitimate negotiating partners to the recognition of entities as sovereign states and as states with specific entitlements’ (Daase et al. 2015, 16; see Biene and Daase 2015).

Fourthly, international recognition is performative insofar as it is based on international practices, that is, ‘socially meaningful patterns of action which, in being performed more or less competently, simultaneously embody, act out, and possibly reify background knowledge and discourse in and on the material world’ (Adler and Pouliot 2011, 6). We identify at least five categories. Declaratory practices are speech acts expressing/performing a formal acknowledgement of the legitimacy and/or sovereign authority of a given actor. Diplomatic practices consist in the contacts and communications inherent to officially ‘representing a polity vis-à-vis a recognised other’ (Sending, Pouliot, and Neumann 2011, 528). Informal engagement practices are other contacts and communications that by contrast deliberately avoid political legitimation and/or sovereignty line-drawing implications, e.g. those portrayed as ‘engagement without recognition’ (Ker-Lindsay 2015) or informal talks in a conflict resolution process (Biene and Daase 2015, 224). Cooperation practices rely on two-way bilateral or multilateral sectoral governance arrangements entailing mutual obligations and benefits for all the parties. Support practices involve one-way assistance to the recognised actor and the wider community within which it is embedded (see Tholens and Al-Jabassini 2023 on security assistance and particularly ‘vertical practices’ between providers and recipients).

So what? Understood in these social-relational, process-oriented, non-dualistic and performative terms, we contend that international recognition-through-interaction has four main interrelated effects in ALS and violent intrastate conflict contexts characterised by the presence thereof. Following an actor-centred approach, the first of these effects is a constitutive one related to identity formation and change in the domestic (governance) actors operating in ALS with which external actors engage. Besides Hegelian recognition theory, the expectation that the collective sense of self and of purpose guiding these local or national entities be shaped and reshaped through interaction with significant others including those outside their state’s borders is also in line with social-psychological theories of identity formation (e.g. symbolic interactionism, role theory) (Greenhill 2008, 354–356) as well as ontological security theory in IR (Mitzen 2006; Steele 2008). What is particular about ALS is that the challenge of uncertainty that intersubjective identity formation faces there at an intrastate level is no different from, if not greater to, that posed by the conditions of anarchy at the interstate level (Murray 2019, 29–52). Indeed, as argued by Börzel and Risse (2021, 17, 276–277), the ‘anarchy problematique’ is a major shared feature of the international system and ALS. Reversing common assumptions, in distinctly fluid ALS contexts one might expect that ‘in contrast to the more stable collective identities of states, the group identity of [a VNSA] can change far more quickly’ (Geis, Clément, and Pfeifer 2021, 7–8, emphasis added). Empirically, the evidence that would instead disprove this effect would be that the identity of domestic actors (e.g the Libyan security/armed and migration governance actors discussed below) does not change substantially following sustained international engagement.

The second effect of international recognition-through-interaction is one of material empowerment which influences the concerned ALS actors’ strategic calculations. Not only their identity but also many of their resources are ‘inhering in the relations that they have with others’ (Sending, Pouliot, and Neumann 2015, 12; see Adler-Nissen and Pouliot 2014). Examples of potential material gains include foreign funding to groups categorised as national liberation movements or rightful rebels (‘legitimate representatives of the people’) and access to a state’s assets abroad and international financing resulting from the formal/legal recognition of an aspiring government (Peterson 2020, 213–214; Roth 2015, 143), as in the case of Libya’s National Transitional Council (NTC) in 2011. They also comprise military build-up (e.g. arms and skills transfers) through international security cooperation practices, and the improvement through support practices (e.g. aid) of the humanitarian and socioeconomic situation in a given actor’s community, which positively impacts the effectiveness and legitimacy of its governance. As a result, the quest for recognition ‘is often quite strategic and reputation is a resource in the struggle for power’ (Lindemann and Ringmar 2012, 221). The counter-evidence in this case would be that international recognition, with a greater or lesser degree of formality, does not constitute a necessary condition for resource provision.

Thirdly, international recognition-through-interaction affects the political legitimacy of ALS actors in the eyes of third actors or audiences – i.e. other significant others, domestic as well as foreign – in both empirical and normative terms. In its basic meaning, legitimacy relates the normative status of a governing institution to the voluntary compliance of the population governed, and therefore the primary locus of the recognitional (social-relational) process that enables it is domestic. Yet, we argue that recognition from external actors may influence it in two ways. On one hand, empirical legitimacy – the social acceptance of a specific authority as rightful – has performance (output legitimacy) as a core component and is thus entangled in a feedback loop with the perceived effectiveness of governance (Börzel and Risse 2021, 70–72, 98; see Schmelzle and Stollenwerk 2018), which is in turn affected by the material (dis)empowerment effects of international (non)recognition discussed above. On the other hand, normative legitimacy (based on general principles) is often constructed in an interactive discursive manner, through a series of ‘reflected appraisals’ or ‘mirroring’ (Wendt 1999, 327) between claims and representations from actors at all levels. Chief among these are international actors, whose declaratory and engagement practices vis-à-vis ALS actors have relatively direct legitimation effects when ‘taken up and used in local or domestic discourses’ (Börzel and Risse 2021, 91), and more indirect ones when diffused to other international players. That being said, the international recognition and legitimacy of ALS actors do not always go hand in hand with each other, and the causal relationship between them is not necessarily a straightforward, positive one (Geis, Clément, and Pfeifer 2021, 10). The counter-evidence of these effects would be no change in the political legitimacy of domestic actors involved in international interactions, which would remain attached to purely domestic sources.

The fourth effect of international recognition-through-interaction in ALS is one of (re)drawing lines, and most particularly the sovereignty lines marking the distinction between state and non-state actors in the context in question. Establishing categories and divisions among local interlocutors or partners is the common purpose of, among others, international declaratory practices such as labelling or naming, e.g. listing armed groups as terrorists (Haspeslagh 2021), as well as the vetting procedures followed in some cooperation and support practices. Yet, we consider that the most consequential type of external boundary-work in ALS concerns the state vs. non-state and public vs. private binaries associated to Western political modernity, which are typically ‘blurred and problematic’ in these contexts (Börzel and Risse 2021, 44, 51–53). For instance, in the realms of ALS security or migration discussed below, ‘the state is here often only one actor among many’; governance is ‘provided more often than not by complex actor constellations that involve different groups of actors at different levels of society with varying relationships to the state’ (Schröder 2018, 379; see Lavenex 2018, 520–521). This poses a tricky categorisation paradox: while the state vs. non-state dichotomy is routinely unsettled or outgrown in everyday governance, it still remains hard to avoid and ‘deeply ingrained as a reference framework’, both analytically and normatively, with the state enduring as an ‘ideal conception that provides an aspirational model of political ordering’ (Schröder 2018, 380).

Upon this background, the role of international recognition is to provide certain ALS actors with what Krasner (1999, 11–25) calls ‘international legal sovereignty’, namely to establish their status in the international system, irrespective of their degree of exercise of ‘domestic sovereignty’ (effective political authority within the relevant borders), ‘interdependence sovereignty’ (control of transborder flows of all sorts) and ‘Westphalian sovereignty’ (capability to exclude external actors from their territory). When the other forms of sovereignty are weak and unstable, and artificial as this might be, international recognition and sovereignty is what ultimately (re)draws the line as to what is and what is not the state. This effect would be refuted by evidence that state vs. non-state lines and other categories have been shaped strictly domestically.

Effects on Libyan domestic security/armed actors

The four theorised effects of international recognition-through-interaction have been prominent in post-2011 Libya’s domestic security, an area in which both structure and actors have been repeatedly described as hybrid. In-betweenness in relation to the state vs. non-state binary became the main feature of the security apparatus that took shape during the years of political transition and attempted state-building that followed the overthrow of the regime of Muammar Gaddafi (October 2011–May 2014). Not having immediately seized control of the state, the victorious revolutionary brigades (thuwwar) moved on to either form their own military/security units – consolidating themselves as autonomous VNSA – or join state security institutions or swing between these two roles. Moreover, a state-VNSA amalgam and symbiosis materialised in hybrid institutions such as the Supreme Security Committee (parallel police force) and the Libya Shield Forces (temporary substitute for the army), which ‘allowed a diverse range of armed groups to operate under official cover’ (Lacher and Cole 2014, 30) with the implicit blessing of the state (Constantini 2018, 105; see Wehrey 2018, 99–100, 122–123). The situation would be further complicated by the authority split and civil war that broke out in 2014–2015, which led to even the formal state structure being duplicated as parallel institutions were set in the east of the country. Limited statehood in the security sphere became thus twofold. A new government recognition controversy and contestation as to whose was Libya’s ‘legitimate army’ was superimposed on the persisting, underlying ‘hybrid reality of military power: most armed groups claim ties with a state institution as they continue to operate as militias’ (International Crisis Group 2016, ii; see Eaton 2018, 6).

Alongside diplomatic or informal engagement practices, international interaction with Libya’s hybrid security/armed actors during the 2011–2019 period fell to a great extent under the umbrella of so-called security assistance, i.e. support practices involving ‘training and equipping’ security forces in a foreign country (Tholens and Al-Jabassini 2023, 5). Yet, there was a significant bifurcation between the ‘informal processes’ that tended to prevail during wartime – with non-state actors lacking formal recognition as recipients – and the ‘formal’ or ‘semi-formal processes’ prioritised in the non-war transition stages, at least by international actors aligned with the UN position – with greater concern for legal sovereignty lines (Al-Jabassini and Badi 2023, 4–5). The foreign actors involved in the former case ranged from NATO and EU member states (US, UK, France) to regional and global powers (Qatar, United Arab Emirates [UAE], Saudi Arabia, Egypt, Turkey, Russia), while the leading players in more formal practices related to SSR were the US, UK, Italy and Turkey.

Our interviews with (mostly Western and UN-related) international diplomats and practitioners in early 2019 demonstrated general awareness of the hybridity of the Libyan (in)security landscape: ‘Unofficial actors are actually more important than the official ones. Furthermore, the lines between these two categories are often blurred. For example, many armed groups or militias actually change their hats from official to unofficial actors as it suits them. That is the case of the four militias that control Tripoli – on which the Government of National Accord (GNA) [operating between 2015 and 2021] relies for security purposes – which sometimes act as state actors and sometimes claim independence from the GNA’ (interview 3; see interview 13; Lacher and al-Idrissi 2018). Yet, arguments in favour of international engagement or ‘constructive contact’ with Libyan VNSA were not short in supply: from effectiveness-based ‘realpolitik’ in the sense of ‘acknowledging the reality and the balance of power on the ground’ to practical needs such as safe access to various parts of the country (e.g. foreign embassies dealing with the militia that controlled the Tripoli airport); from contributing to local security (e.g. UNSMIL’s role in negotiating a new security arrangement and ceasefire among the capital city’s militias in September 2018) to inclusive mediation between armed groups at the wider state level as well as national self-interest related to the concerned states’ own domestic security issues (e.g. the UK’s push for the extradition of the brother of the suicide bomber responsible for the Manchester terrorist attack of 2017, who was held by a militia in Libya) (interviews 3, 5, 13).

At the same time, and maybe guided by shared expectations about conflict sensitivity, our interviewees emphasised the need to tread carefully in interacting with Libyan security/armed actors in general and VNSA in particular. Some of them explained that their engagement with the latter, while not secret, was ‘usually discreet and low-profile’ (interview 3), or only indirect via mediation NGOs (interview 13). Such caution was motivated by anticipated or demonstrated effects related to identity formation and change, material empowerment, political legitimation and (sovereignty) line-drawing among these Libyan actors.

Firstly, international recognition-through-interaction played a part in the identity formation and change of Libyan security/armed actors by shaping the shifting framing of the conflict and its parties over these years. Most VNSA’s larger-scale self-identification and positioning within the broader game of the conflict owed much to recognition from external supporters at different points in time. This applies to the revolutionary framing encouraged during the 2011 civil war by NATO members as well as regional powers such as Qatar. The display of the Qatari flag in some of the most iconic sites of the revolution was a graphic proof of Doha’s association with the new revolutionary identity, starting from the positive recognition (coverage and amplification) of the 17th February uprising by the pan-Arab television broadcaster Al Jazeera (interview 7). A similar thing happened with the most widespread discursive demarcation of the two sides during the 2014–2015 authority split and civil war, which pitted the supposed ‘Islamists’ of the General National Congress (GNC) and Libyan Dawn, especially Misratan militias, against the self-styled ‘anti-Islamists’ aligned with the House of Representatives (HoR) and the Operation Dignity led by the military strongman Khalifa Haftar. Rather than reflecting the actual makeup of the two armed alliances, both of which comprised an assorted range of non-Islamist and Islamist forces including Salafists (Lacher 2020, 119, 127), such a framing was primarily attuned with and reinforced by the ideological leanings of the regional backers of each side, i.e. Turkey and Qatar for the GNC/Libyan Dawn, and Egypt, UAE and Saudi Arabia in the case the HoD/Operation Dignity. That being said, the depth of the identity (trans)formation effect on Libyan VNSA must be qualified in view of the counter-evidence that most of these actors’ collective sense of self and of purpose was forged and deepened throughout the conflict as primarily local or communal, as demonstrated by Lacher’s (2020, 8, 103–109) research on their ‘social embeddedness’ (see also Collombier and Lacher 2023, 7–9).

Secondly, international recognition-through-interaction significantly contributed to the material empowerment of various Libyan security/armed actors from 2011 to the post-2016 stage. In 2011, in the midst of the first civil war and international military intervention, foreign support for the thuwwar was partly direct and informal, as in the case of some NATO states’ dispatch of intelligence officers, special forces and arm deliveries to Misrata, Benghazi and other revolutionary centres (Mundy 2018, 77–78), and partly indirect and semi-formal, mediated by the NTC that the revolutionary local councils had set up as a tool of rebel diplomacy. The NTC itself accumulated wider and increasingly formal international recognition in several steps, through various state declaratory practices, which in turn upgraded the material advantages afforded to its armed affiliates: its designation as ‘the legitimate representative of the Libyan people’ facilitated foreign funding for the rebels and, even more crucially, its later recognition as ‘holding governmental authority’ de jure to the exclusion of the Gaddafi government granted it access to the Libyan state’s sovereign assets abroad (Bartu 2015, 45–48; Talmon 2011). Later on, from 2012 onwards, Libya would see a swing between non-war phases in which formal security assistance practices such as SSR came to the fore and two new civil wars where international engagement largely reverted to informality.

In 2012–2013, international interaction with Libyan security/armed actors displayed a focus on SSR in line with the liberal peace- and state-building paradigm. This involved tangibly supporting the formation and empowerment of a centralised security structure and a brand-new national army for the Libyan state. The US, the UK, Italy and Turkey agreed to train abroad troops of the latter military, baptised as the General Purpose Force or General Protection Force, while other Western and Arab states made commitments about police training (Constantini 2018, 105; Laessing 2020, 46, 49; Pack 2021, 91–95). Distinctively in such SSR efforts, the material empowerment effect on security/armed actors was, at least formally, connected to and conditional on sovereignty line-drawing, that is, these actors’ transitioning from non-state to state actors. Similarly, following the establishment of the internationally-recognised GNA in late 2015, while SSR remained off the table as impracticable, Western interaction with Libyan security/armed actors generally would seek to preserve a façade of the sovereignty lines set by the new international government recognition consensus. Accordingly, the deployment of US-UK special forces, intelligence, training, arms shipments and airstrikes during the so-called Battle of Sirte to dislodge the Islamic State (IS) group from central Libya in 2016 prioritised support for Misratan militias aligned with the GNA, though far from constituting a proper national army (interview 3; International Crisis Group 2016, 23; Mundy 2018, 194–197).

By contrast, this type of security assistance subject to sovereignty line-drawing became very secondary during the 2014–2015 and 2019–2020 civil wars, when we saw a return to more informal if not covert forms of engagement between the Libyan VNSA integrated into either Operation Dignity/Libyan Arab Armed Forces (LAAF) or Libyan Dawn/pro-GNA forces and their respective foreign patrons. The latter contributed enormous material support through funding (UAE, Qatar), arms shipments (UAE, Egypt, Sudan, Russia, Turkey), equipment refurbishment (Russia) and eventually military personnel, private military companies and transfers of Syrian mercenaries to the Libyan battlefield (Turkey, Russia) (Al-Jabassini and Badi 2023, 12–18; Lacher 2020, 41, 153). As an exception, only Turkey formalised its military intervention and emphasised international legal sovereignty in December 2019–January 2020, when its parliament passed a memorandum of understanding on security cooperation with the GNA as well as the dispatch of national armed forces to Libya (International Crisis Group 2020, 5). Altogether, a certain level of formal international recognition significantly contributed to some Libyan security/armed actors’ material empowerment in non-war phases, yet its impact on resource provision was almost negligible during wartime.

Thirdly, when it comes to the political legitimacy effects of international recognition-through-interaction, our interviewees assumed their engagement with Libyan security/armed actors to have normative legitimation effects inasmuch it was filtered through and leveraged in domestic and local politics. This seemed to be the purpose, for example, of the French foreign minister’s visit to Benghazi and public demonstration of ‘respect’ for Haftar just before the latter’s launch of his offensive on Tripoli, which led to the 2019–2020 civil war (Le Figaro, 20 March 2019). By contrast, other foreign embassies expressed concern about the ‘choice of who we send’ and ‘who meets whom in order to avoid conferring legitimacy to the wrong people’, especially among VNSA. For example, a junior female diplomat was able to ‘meet Libyan actors across the country’ with ‘nobody […] interested in taking and posting pictures online as it would happen in the case of the ambassador’ (interview 3). Overall, though, these external legitimation effects were partial and mixed. As noted above, the first and foremost source of legitimacy for Libyan security/armed groups remained their ‘social embeddedness’ and web of ties within their closer communities, from which they could not ultimately be extricated as discrete organisations and units of analysis (Lacher 2020, 8). In addition, the aforementioned identity formation and material empowerment effects of international interaction, e.g. the influence of the external framing of the conflict on alliance-building dynamics and strategic conditions, interacted with the local legitimacy of Libyan security/armed groups in complex and sometimes conflicting ways. Tensions between the armed actors’ strategic and social constraints have indeed been considered the main driver of the process of Libya’s political fragmentation (Lacher 2020, 9).

Fourthly, the most distinctive and consequential impact of international recognition-through-interaction emerging from our interviews concerned (re)drawing lines, including sovereignty lines, among Libyan security/armed actors. To start with, in a conflict context such as Libya’s, international actors reporting to or funded by US, European and UN institutions were required to follow specific vetting procedures in the selection of local security/armed partners for various cooperation and support practices. In principle, the most constraining framework was the so-called Leahy Law, which bans US military and security assistance to foreign forces that have committed gross human rights violations with impunity (interviews 8, 12). The UK’s 1998 Human Rights Act and Overseas Security and Justice Assistance Guidance established more wide-ranging normative red lines and procedures, including a risk assessment of potential human rights violations by local partners (interview 3). The EU relied on its member states for the vetting of agents of the Libyan security forces benefitting from training programmes, e.g. in the framework of the EU Naval Force Mediterranean (EUNAVFOR MED) Operation Sophia and EU Border Assistance Mission in Libya (EUBAM) (interviews 20, 21). In the case of UN agencies and services, the human rights vetting of security/armed actors was run by UNSMIL (interviews 16, 17). In practice, though, all these individual background checks based on various existing databases and sanction lists were ‘only really effective for high-profile war criminals’ (interview 16), and generally ‘difficult to implement with absolute certainty in a context such as Libya, due to the blurred lines between official and unofficial armed actors […] and the dark origins of many of the former’ (interview 8). Another red line involved in sorting out Libyan security/armed actors was the one preventing foreign engagement with what various governments designated as terrorist organisations. This was unsurprisingly a deeply contentious label – one often weaponised and surrounded by vast grey areas in this context as elsewhere – but the common denominator was to emphasise the exclusion of any contact with IS, Al Qaeda in the Islamic Maghreb and other ‘really extremist actors’ (interviews 5, 15).

More fundamentally, international actors had to routinely draw a line to separate state security forces from VNSA that they considered as ‘militias’. This is the type of boundary-work that constructs ‘international legal sovereignty’ (Krasner 1999, 14). Besides the SSR attempts in 2012–2013, the clearest evidence of the artificiality and malleability of sovereignty lines in this hybrid security context can be found in post-2016 arguments about the singling out of Haftar’s LAAF from other militia alliances: ‘We have nothing to do with militias. [But] if the international community acknowledges [Haftar] as the leader of the army, then he is not a militia for me. Whether that is the truth or not, that is a perception of reality’ (interview 4). The LAAF’s distinction as the sovereign state army, or nearly so, resulted from a tenuous two-step recognition process that was initially domestic, based on its 2014 designation by the HoR as the ‘Libyan National Army’, and then underwritten internationally. The consequence of this was a contradictory situation of ‘dual international recognition’ of rival Libyan security forces (interview 13), including both GNA-aligned armed groups and the LAAF/‘Libyan National Army’. At the same time, in order to meet practical needs such as access and safety on the ground, foreign actors tended to avoid engaging with the aspiring national militaries and instead preferred that ‘security is managed at the local level’, in a decentralised and ad hoc fashion, by the main local security or police actor in each city or area (interview 8). In the latter case, rather than drawing or reinforcing any sovereignty lines, international recognition-through-interaction purposefully avoided them.

Effects on Libyan migration governance actors

Another area where the effects of international recognition-through-interaction have been conspicuous in the post-2011 Libyan conflict context is migration governance. As with domestic security, agency and structure in this sphere have been characterised by hybridity in relation to the state vs. non-state binary. Furthermore, the limitations of state capacity and the reduced scope of domestic migration policy that are typical of ALS, combined with the unprecedentedly high salience of international migration containment for foreign actors such as Libya’s European neighbours, have led to the rise of ‘informal modes of governance by default, with a prominent role of private actors and international institutions’, and to the country becoming increasingly subject to ‘patterns of external migration governance’ (Lavenex 2018, 526–527; for a broader critical take on EU border externalisation see Stierl 2020). In general, migration was not viewed as a primary concern for Libyan actors, but an ‘externally imposed agenda which creates political and economic opportunities’ (interview 9). More specifically, the range of key foreign players here was much more limited than in the field of security governance, and essentially comprised the EU, its southern member states led by Italy, UN humanitarian agencies and international NGOs. On the domestic front, the leading Libyan state institutions involved since 2012 under successive Tripoli governments were the Ministry of Interior’s Department for Combating Illegal Migration (DCIM), which was responsible for both inland and at-sea migration management, and the Libyan Coast Guard, which implemented the latter. Both of them attracted sustained criticism because of their perceived collusion with, or practical indistinction from, VNSA and criminal actors (local militias, organised crime and smuggling networks) (Pacciardi and Berndtsson 2022, 4017–4022), as well as the extreme and systemic human rights violations associated to migrant detention and interceptions at sea (Amnesty International 2017; Human Rights Watch 2019).

This situation posed two dilemmas between greater and lesser evils for international engagement. The first of them stemmed from the paradoxical circle whereby the greater the capacity of Libyan border management and the reduction in departures from the coast and deaths at sea – an area in which some European governments prided themselves in having ‘achieved incredible results together with the Libyans’ (interview 19) – the higher the number of intercepted migrants held in arbitrary detention in inhuman and degrading conditions and subject to the gravest abuses in the country’s proliferating detention centres (interview 20; Human Rights Watch 2019; Phillips 2020, 5). The second predicament concerned the selection of Libyan state and non-state partners. The advocated conflict-sensitive approach required weighing the consequences of international interaction favouring ‘migrant smugglers turning into security enforcers, gaining recognition from the GNA and incentives from external actors’, that is, potentially opening a pathway to SSR: ‘We can think about the counterfactuals: What would they be doing instead? Could this be an incentive for security sector reform?’ (interview 22). Alternatively, foreign actors were able to bypass the ineffective Tripoli government by making direct arrangements with local VNSA that controlled the migration business in coastal areas (Lacher 2018, 20). Whatever the choice, the consequences included effects of identity formation and change, material empowerment, political legitimation and (sovereignty) line-drawing among Libyan migration governance actors.

Firstly, regarding identity formation and change, international recognition-through-interaction came hand in hand with new framings of the nature and activity of Libyan migration governance actors which either reinforced or challenged their primarily local self-identification. For example, groups who claimed that their activity was simply their way of economically supporting their communities were externally criminalised as migrant ‘smugglers’ or even ‘traffickers’ (interview 3; see Eaton 2018, 10–11). On the other hand, the encouragement through international engagement of the incorporation into state roles of non-state actors, including VNSA, could arguably represent a ‘potentially positive recognition process leading to tying these groups to stability’ (interview 22) and reshaping their identities accordingly. This was supposed to be practically reinforced by foreign capacity-building and training programmes, which were the most direct and individualised form of interaction, in particular, with members of the Libyan Coast Guard. However, the available evidence suggests that the training of Libyan Coast Guard personnel, many of which were former revolutionary brigade fighters, did not generally succeed in replacing aggressive outlooks by a prevailing concern for the protection of migrants (Human Rights Watch 2019, 21–22, 27). A confidential EU military report from January 2022 acknowledged that EU training was ‘no longer fully followed’, which entailed a persistent deficit in ‘proper behavioural standards […] compliant with human rights’ (Associated Press, 25 January 2022). In short, identity transformation in this respect was at best partial.

Secondly, and largely in contrast, international recognition-through-interaction did have a definite impact on the material empowerment of Libyan migration governance actors. The Libyan Coast Guard’s capacity was technically beefed up through support practices such as training in the framework of the EU’s Seahorse maritime surveillance programme and the EU Naval Force Mediterranean (EUNAVFOR MED) Operation Sophia. The latter signed a memorandum of understanding with the Libyan Ministry of Defence to this purpose in 2016. Allocations from the migration-focused EU Emergency Trust Fund (EUTF) for Africa enabled the delivery of three training packages, including one in naval bases of member states such as Italy, Greece and Spain (interviews 11, 20, 21). Yet, rather than training, what the Libyan Coast Guard saw as its most urgent need and most demanded from these international partners was equipment (Loschi, Raineri, and Strazzari 2018, 6–7). In response, Italy in particular provided material assistance in the form of donations of patrol boats and logistical support for vessel maintenance, following the conclusion of a bilateral memorandum of understanding with the GNA in 2017 (Human Rights Watch 2019, 21–22). Memoranda of understanding suggest that the formalisation of the interaction, which involved the recognition of the concerned Libyan state actors, was viewed as a necessary condition for resource provision. By contrast, other EU member states preferred to abstain from bilaterally supporting or funding the Libyan Coast Guard because of the ‘politically sensitive’ nature of its activity (interview 1).

The interaction dynamics were very different in the case of detention centres, where the material empowerment of the Libyan actors in control was largely an unintended, though not unnoticed, consequence of a different type of support practices, i.e. the humanitarian aid delivered to the migrants held there. In an apparent attempt to avoid formal recognition, the EU and EU member states dodged directly funding the DCIM, which run the centres often just on paper, and instead channelled emergency assistance (e.g. food, material, clothing, hygiene kits) through UN agencies and NGOs that worked to mitigate the dire conditions inside them (Human Rights Watch 2019, 27–28). However, and indirect as it might be, interaction with the centres was still in full swing. Widespread reports of embezzlement and corruption (interviews 9, 17; Phillips 2020, 3, 5) suggested that international aid was in fact ‘fuelling a criminal economy of exploitation and traffic’ as well as ‘further [empowering] non-state armed actors and militias’ (Loschi, Raineri, and Strazzari 2018, 22). Importantly for the analysis here, the rent-seeking behaviour of Libyan VNSA and other migration governance actors was often intertwined with domestic and international recognition-seeking, with various degrees of formality, as in the case of militias that set up their own detention centre after seizing a group of migrants and then sought the DCIM’s recognition to turn this into an official state facility paving the way for international interaction and material assistance (interview 9). Ultimately, the European recognition avoidance did not prevent material empowerment.

Thirdly, international recognition-through-interaction influenced the political legitimacy of Libyan migration governance actors in multifaceted ways. To start with, ostensibly state actors such as the GNA owed much of their very creation and ex ante international recognition to the European urgency to find ‘regular’ Libyan institutional partners with whom migration control cooperation could be effectively and legally boosted (International Crisis Group 2016, 6, 23; Lacher 2018, 26). This exogenous and instrumental origin entailed double-edged dependence and effects in terms of legitimacy: international recognition actively contributed to building the empirical legitimacy – especially output legitimacy – of the GNA and migration governance actors formally subordinate to it such as the Libyan Coast Guard (Human Rights Watch 2019, 22), yet at the same time such a government was ‘domestically criticised as a puppet for submitting to these external pressures’ (interview 9). At a lower level of analysis and regarding the Libyan Coast Guard, critics also voiced concern that, absent proper SSR, European capacity-building entail the ‘unwarranted legitimisation, co-option and institutionalisation of highly controversial security actors’ (Loschi, Raineri, and Strazzari 2018, 8).

Another, more normative and indirect form of legitimation resulting from international interaction was the one that primarily benefitted specific migration governance practices, and by extension the actors performing them. This was contentious, in particular, when it came to legally dubious and human rights-violating practices such as the return of intercepted migrants to the Libyan coast and their warehousing in detention centres. UN humanitarian agencies such as International Organisation for Migration (IOM) and the UN High Commissioner for Refugees (UNHCR) were shamed for their presence at disembarkation ports and inside detention centres, which their detractors saw as legitimising the behaviour of the Libyan Coast Guard and the centre owners (Foreign Policy, 10 October 2019), yet they defended as being ultimately in the interest of migrant protection and assistance. For instance, a UNHCR official argued that ‘the UNHCR’s work at the detention centres did not legitimise them – as they would exist anyway – and by registering jailed migrants it prevented migrants from being sold on to traffickers’ (BBC News, 31 July 2019).

Fourthly, the effects of international recognition-through-interaction in terms of (re)drawing lines among Libyan migration governance actors ranged from those of simple vetting to others with deeper sovereignty implications. The artificiality of the lines at stake here was particularly paradoxical in view of the European discourse’s emphasis on the stark criminalisation of non-state actors categorised as smugglers, traffickers or ‘mafias’. Vetting affected EU and EU member states’ support practices such as the training of the Libyan Coast Guard, and followed the same procedures discussed above for the selection of local security/armed partners (interview 13). However, yet again, the robustness of the system was questionable in practice, as demonstrated by evidence of EU trainings having benefitted individuals that the UN Security Council’s Panel of Experts on Libya identified as oil smugglers and human traffickers (Loschi, Raineri, and Strazzari 2018, 7). In addition, some VNSA leaders turned into key members of the Libyan Coast Guard were involved in international informal engagement and diplomatic practices. Intentional or not, their invitation e.g. to meetings with Italian government representatives (Avvenire, 4 October 2019; The Guardian, 4 October 2019) implied some semi-official representation capacity which pushed them across the sovereignty line to act like state actors.

In the case of migrant detention centres, humanitarian support practices bore similar sovereignty line-drawing effects: ‘Support for detention centres may be seen as giving semi-state authority to criminal groups’ (interview 22). The swing here was between totally unofficial (non-state) detention centres and those nominally operating under the authority of the DCIM, though still de facto controlled by VNSA, whose number constantly varied (interview 17; Eaton 2018, 11; Human Rights Watch 2019, 15–16). Altogether, the combination of problematic material empowerment, political legitimation and sovereignty line-drawing effects on Libyan migration governance actors confronted the international community with the policy dilemma between humanitarian presence and withdrawal from migrant detention centres. The positions adopted in this respect ranged along a continuum from disengagement (UN agencies and NGOs) to engagement without recognition (EU) to overt bilateral engagement (Italy). Several UN agencies agreed on the discontinuation of humanitarian assistance in these facilities in 2018, according to them, in defiance of pressures from both the Libyan authorities and international donors, and at the expense of much of the latter’s funding: ‘This choice is negatively impacting on [our] funding, as international donors are currently prioritising aid to migrant detention centres. This is where most of the money is going now’ (interview 17). International NGO staff similarly highlighted the price paid for the winding down of their activity in this area in terms of EU funding, especially from the EUTF. In both cases, such a costly decision was justified by invoking conflict sensitivity as well as the ‘do no harm’ principle of humanitarianism (interviews 9, 17).

By contrast, occupying an in-between position, EU officials emphasised that their official policy was to lobby with the Libyan GNA for the closure of the migrant detention centres and that, accordingly, there was no EU direct funding for or involvement in any of these facilities. At the same time, the EU institutions considered that ‘a total withdrawal would be worse from a humanitarian perspective’ and preferred to maintain a ‘minimal support’ by funding UN agencies and NGOs that provided relief there (interviews 20, 21, 13, 9). Such indirect support practices arguably amounted to engagement without recognition. Finally, Italy stood out as the only foreign state to have its own bilateral programme, amounting to 14 million euros and channelled through NGOs, ‘to improve the conditions directly in the migrant centres – in the legal ones, of course’. This was defended as a lesser evil: ‘The international community has only two [options]: either we complain that the migrants are mistreated and close our eyes and go back home, […] or we act and we try to improve as much as possible the conditions of the migrants’. This line of argument also stressed the importance of constant dialogue with and discreet pressure on the Libyan authorities, as opposed to public shaming. Italy’s negotiations with the GNA, for instance, were claimed to have played an essential role in the establishment of the UNHCR’s Gathering and Departure Facility (GDF) in Tripoli as a (partial) alternative to migrant detention centres (interview 19).

Conclusion

This article has demonstrated the potential of deploying insights from the literature on international recognition, and social-theoretical approaches in particular, to study international interactions with intrastate conflict actors and contexts. Arguing that social interaction and recognition are indissolubly bound to each other as well as process-oriented, non-dualistic and performative in nature, we have shown that any sort of international engagement with domestic actors bears some recognitional effects on such agents and the wider structures they partake in. Thus understood, the influence of international recognition-through-interaction goes beyond two dichotomies that have so far dominated the specific research on international recognition in intrastate armed conflicts. The first of these is the governments vs. rebels binary, which may be overcome by expanding our analytical lenses from the perspective of violent conflict to that of ALS, so as to shed light on engagement with governance rather than (just) belligerent actors. The case studies of security/armed actors and migration governance actors in post-2011 Libya reveal the extent to which international recognition-through-interaction has configured agency and by extension structures in this conflict and ALS, beyond the protracted contest about the international recognition of the Libyan government.

The second, related limitation that we have aimed to redress here is the binary view of (mis)recognition as a source of either (violent) conflict or conflict transformation which has prevailed in the application of recognition theory to IR and conflict and peace studies. We have instead theorised and empirically substantiated the extent of a broader range of effects on domestic (conflict) actors. In order of importance, (re)drawing lines appears to be the most distinctive and consequential power of international recognition-through-interaction in Libya, especially when it comes to tracing the sovereignty boundaries – highly artificial in this and other ALS – between state security forces and VNSA considered as ‘militias’, or official and unofficial migrant detention centres. Material empowerment effects were tangible at all times, yet their subjection to sovereignty line-drawing significantly varied with the fluctuations of the conflict context and the governance area. Resource provision to security/armed actors was attached to SSR objectives and required a certain level of formal international recognition in non-war phases, but it tended to revert to informal if not covert practices during wartime. Among migration governance actors, material reinforcement was straightforward and formalised in the interaction with the Libyan Coast Guard, yet it equally took place in spite of European recognition avoidance, as a side effect of humanitarian aid provision, in the case of detention centres.

International recognition-through-interaction also influenced the identity formation and change of Libyan domestic security/armed actors in terms of their larger-scale self-identification and positioning within the broader game of the conflict. It similarly contributed to framing the nature and activity of migration governance actors. In both cases, however, identity transformation effects were only partial and shallow: deep-seated or revamped local attachments and pre-existing attitudes remained stronger. Finally, the impact on the Libyan actors’ political legitimacy was the most mixed, as international engagement interacted with their local legitimacy in complex and conflicting ways, with empirical (output) legitimacy sometimes increasing at the expense of normative (principle-based) legitimacy.

Acknowledgements

This article would not have been possible without funding from the British Academy and Leverhulme Trust, or without the time and generosity of our interviewees. It has similarly benefitted from spot-on and constructive feedback from the Journal of Intervention and Statebuilding's two anonymous reviewers and editorial team. Our thanks also go to colleagues from the University of Exeter's Centre of Advanced International Studies (CAIS) and Miguel Hernando de Larramendi for their comments on an earlier related draft resulting from the same research project.

Disclosure statement

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

Correction Statement

This article was originally published with errors, which have now been corrected in the online version. Please see Correction http://dx.doi.org/10.1080/17502977.2023.2288425

Additional information

Funding

This work was supported by a British Academy/Leverhulme Small Research Grant [Grant Number SRG18R1\181252].

Notes on contributors

Irene Fernández-Molina

Irene Fernández-Molina is a senior lecturer in International Relations at the University of Exeter, United Kingdom. Her research deals with international relations and foreign policies in the Global South, IR theory (international socialisation, recognition, practices), North Africa as well as EU foreign policy and Euro-Mediterranean relations.

Alfonso Casani

Alfonso Casani is a lecturer at the Department of Political Science and Public Administration of the Complutense University of Madrid, Spain. His work focuses on social movements and opposition dynamics as well as political Islam and the interplay between religion and politics, with a special focus on the Maghreb region.

Notes

1 From the outbreak of hostilities to the conclusion of ceasefire agreements.