Introduction

The Afghan Local Police (ALP) has its origins in an international counterinsurgency (COIN) programme that sought to raise village-level defence forces from within communities that would help expel the Taliban from their areas. The programme was meant to further the COIN objective of defeating insurgents. However, it departed from ‘mainstream’ COIN in that it is based more on a light footprint of international forces who would tap into the purported power of ‘traditional’ systems of local decision-making and security provision to tip the balance against the insurgency in rural Afghanistan. In this, we suggest, the ALP can be understood as a COIN security ‘sub’-culture (Kaldor and Selchow 2015) that pursues COIN objectives by means of different practices intended to correct the shortcomings of the overall international ‘project’ (Suhrke 2011).

In emphasising the purported power of supposedly local, traditional or informal forms of order, the ALP parallels a broader shift in zeitgeist. This shift can be observed on two levels. On the policy level, in response to widespread loss of conviction in the power of externally engineered ‘state-building’ efforts to fix supposedly ‘failed’ states, the international development and security policy discourses increasingly see ‘bottom-up’ as a more cost-effective, efficient, appropriate and viable alternative to prevailing modalities of intervention. The ALP, for example, can be interpreted as a policy that has taken to heart exhortations for ‘alternative non-state-centric approaches to governance, the control of violence, peace-building, and development’ and to be based on an appreciation of ‘the strengths of the societies in question, acknowledging their resilience, encouraging indigenous creative responses to the problems, and strengthening their own capacities for endurance’ (Boege et al 2009a: 14). Meanwhile, this policy-level shift has been paralleled by, but needs to be distinguished from, an increasingly prominent and diverse scholarship that seeks to overcome the perceived intellectual failures of the failed states discourse by redefining statehood and suggesting ‘more empirically grounded or more conceptually innovative’ (Hagmann and Péclard 2010) research into the emergence and practice of local governance and authority in situations of violent conflict. While scholarship in this vein is opening the way to a much more sophisticated analysis of social phenomena and processes currently affecting the lives of billions of people across the world, the policy turn to the local contains highly problematic implications that have yet to be fully unpacked or sufficiently considered by either the policy or the academic communities.

The ALP programme has proven extremely problematic in practice. We draw attention to the wide diversity and changeability of ALP practices and distinguish between their contribution to local experiences of security and their contribution to international counterinsurgency objectives. The bewildering diversity and ambiguity of impacts (Noori 2015) is captured by the US military Special Operations Forces’ (SOF) expression, ‘if you’ve seen one village stability platform, you’ve seen one village stability platform’ (Hanlin 2011; Ives 2013).1 According to one assessment:

[r]oughly one third of ALP units are enhancing local security, undermining insurgent influence, and facilitating better governance (…). Another one third are not producing such outcomes and may, in certain respects, be engaged in collusion with the enemy or in abusive behavior that abets the enemy. The last third falls somewhere in between the first two groups’ (Joint Special Operations University, quoted in Smith 2015).

In addition to great spatial variation, some units appear to have been involved in serious abuses against the populations that they supposedly protect, whilst others morphed quickly in local perceptions from contributing to security to driving insecurity (Aikens 2014; HRW 2011:3; ICG 2015; Mashal 2011).

In this article we seek to explore why the ALP has proven so messy and to explain the gap between ALP objectives on paper and the way the programme has played out in practice. The ALP case indeed suggests that policies seeking to reflexively tap the purported power of the ‘local’ – depicted as more culturally appropriate, effective and legitimate – do not simply overcome the limitations of mainstream forms of intervention, but introduce new complexities of their own. Here the distinction becomes important between ‘bottom-up’ policy interventions and the more theoretically-driven scholarship that is often perceived to underpin and justify such policies. While acknowledging the difficulties of the ALP in practice, the more theoretically-driven work should not be dismissed too hastily. Far from justifying the ALP, the emerging literature should instead make policymakers cautious about the current turn to the local both in wider US SOF engagements around the world and in the broader international development and security policy communities. Moreover, it offers useful ways to interpret the contrast between the neat ALP blueprint and the practice.

We seek to highlight two strengths of the emerging theories. First, the heuristic of dynamics of contestation has advantages over dominant approaches to the political order of spaces where a monopoly of the legitimate use of force is not an established fact. A second, less well-observed, strength of such approaches, is their ability to encompass different forms of international intervention into the frame of study as another layer of actors engaged in dynamic contestation processes. The emerging theoretical work therefore provides a useful lens not only for understanding the dynamics of conflict-torn spaces, but also for understanding the supposedly unexpected consequences of policy interventions seeking to tap such dynamics.

To demonstrate these theoretical strengths we examine evidence from the District of Andar in Ghazni Province and apply the heuristic of dynamic contestation to explore why an apparently ‘messy’ ALP programme played out as it did. In analysing this context, part of our intention is to demonstrate the analytic purchase of more dynamic and emergent conceptions of authority in places where the formal state is at best one among many authorities. We also seek to show how this heuristic approach can incorporate international intervention within the analytic frame, including interventions like the ALP, that seek to reflexively harness insights about the strength of local, traditional or informal authority to achieve ostensibly more legitimate and effective interventions. At heart, we argue that the ALP reveals a tension between ascribing power and agency to local dynamics and assuming that local interests can be harnessed and aligned to external objectives. The messy, spatially varied and mercurial character of the programme in practice reflects that it added to, but did not overwrite, existing local dynamics of contestation between an array of actors. When the ALP was introduced into Andar’s complex local landscape, it became a resource which a wide array of local actors mobilised to influence according to their own perceived interests. The US forces sent to establish ALP units saw the programme in terms of the COIN objective of encouraging local actors to align behind a shared practice of resisting insurgents. The ‘security gap’ between COIN objectives, pursued through the ALP modality, and the ‘messiness’ of the ALP in practice, is therefore better understood not in terms of stated international counterinsurgency objectives but as the contingent outcome of multiple ‘power poles’ (Bierschenk and Olivier de Sardan 1997: 441, quoted in Hagmann and Péclard 2010: 542) vying to make the programme serve their own interests. Moreover, the reasons for the problems encountered can actually be well understood and even anticipated by taking seriously the more theoretically-driven literature from which the ALP and other ‘bottom-up’ policy interventions supposedly draw inspiration.

The article is arranged as follows. First, we present the ALP as depicted ‘on paper’ as a contemporary programme that, while rooted in COIN doctrine and objectives, emphasises the importance of ‘local’ tradition and custom. We then locate the programme within a wider turn in zeitgeist entailing growing interest in spontaneous and evolving forms of authority emerging in situations of violent conflict and attenuation of formal state authority. Thirdly, we explore the case of the ALP in Andar by contrasting official depictions of the programme with the way it has played out ‘in practice’. The case illustrates both the complexities and difficulties involved with policies of ‘going local’ and the value of emerging heuristic approaches that jettison state centric analytic lenses in favour of emphasis on dynamic contestation processes.

The ALP on Paper: Empowering the Local

The ALP is a US-sponsored, NATO-backed security programme that trains ‘local Afghans in rural areas to defend their communities against threats from insurgents and other illegally-armed groups’ (DOD 2011: 68). It was inaugurated in 2010 as part of the US-led ‘surge’ effort, at a time when tens of thousands of additional troops were deployed on top of the existing international military forces that were attempting to turn the tide against mounting insurgency. Given Afghanistan’s size and its varied and often daunting terrain, even the combined Afghan National Security Forces (ANSF) and international military presence were insufficient to extend security to the local level or to dominate the countryside – much of which remained contested space. International forces came to see villages as caught in the middle, threatened by international and national forces for cooperating with insurgents and menaced by the insurgents if they did not.

The ALP, as presented in official documents, was intended to help address these difficulties. Lacking sufficient forces to pervasively dominate the countryside at the district and sub-district level, the programme envisaged enlisting villagers themselves against the insurgency, ‘targeting’, according to NATO, ‘rural areas with limited to no ANSF presence… to enable conditions for improved security, governance and development’ (NATO 2012). By providing villagers with the means to provide ‘security’ themselves, the programme was intended to change the basic equation confronting villagers. In this, the programme’s basis in US COIN doctrine was apparent. COIN holds that insurgency cannot be defeated by military force alone but is achieved politically when the population is persuaded to consent ‘to the government’s legitimacy and stops actively and passively supporting the insurgency’ (US Army 2006: 1–14). COIN forces seek local opposition to insurgents, and ways to enable that opposition, while trying to generate a tipping point in which momentum swings decisively away from the insurgency (Jones and Muñoz 2010: 74). The ALP should be understood as designed to enable villagers wanting to ‘stand up for themselves’, and the programme grew from the realization that some villages across Afghanistan seemed to be spontaneously resisting insurgent groups (Catanzaro and Windmueller 2011). Signalling that the programme intends to respond to spontaneous bottom-up resentment rather than being imposed on villages, villagers must ‘have either demonstrated active resistance to the insurgency, or have recently asked for assistance to do so’ to be eligible (Stevens 2011: 65).

ALP advocates portray village-level self-defence forces as ‘traditional’ – and hence legitimate and efficient (Jones and Munoz 2010; Rector 2012). Indeed, in many parts of Afghanistan community level security arrangements existed for generations (Karokhail and Schmeidl 2006; Schmeidl and Karokhail 2009a). The traditional security mechanisms of the Pashtun provinces in southern and eastern Afghanistan, called ‘Arbakai’ in Pashto (‘Salwishti’ or ‘Shalgoon’ in FATA, and ‘Paltanai’ in Kandahar) (Tariq 2008: 3), for example, were volunteer initiatives for enforcing law and order and providing community security (Schmeidl and Karokhail 2009a; Tariq 2009). Arbakai have received particular attention from international forces, who often ‘use the term to depict well-meaning, disciplined, traditional community defence forces, obedient only to the tribal assembly (jirga)’ (Lefèvre 2010: 3).

Building on this tradition the ALP design, as presented in official public documents, entails Afghan and international forces working with village shuras (councils) to agree to the establishment of a local ALP unit, thereby ensuring local ownership, legitimacy and accountability. The shura then nominates adult men from within the community as recruits. Before enrolment, nominees were to be vetted both by the Afghan Government and international forces. If accepted, recruits were then supposed to be equipped and given three weeks of training with US SOF (DOD 2011: 68–70; DOD 2012). ALP units (officially consisting of around 30 patrolmen) were supposed to be raised solely through SOF units, embedded in villages as part of wider ‘Village Stability Operations’ who, augmented by a variety of other initiatives, were to pursue an ‘integrated approach to governance, security, and development’ (DOD 2011: 68; Felbab-Brown 2013; Robinson 2013; Saum-Manning 2012: 8; Huslander and Spivey 2012).

On paper at least, the ALP are not ‘police’, having no powers to arrest, and are only empowered to investigate crimes if specifically requested by the Afghan National Police (ANP). Rather, they are intended as a lightly armed, defensive, pro-government presence at the village level capable of deterring insurgents – a ‘night watch with AK-47s’ (Radin 2011), that can call upon ANSF and international forces for support. While the US provided funding, training, equipment and technical assistance, this was directed through the Afghan Ministry of Interior (MoI) in which the initiative is formally housed. By making ALP units formally accountable to the District Police Chief, complementing local decision-making with national and international vetting processes, and SOF training and mentorship, the ALP was presented as creating a fairly reliable, limited, pro-government presence with which the ANSF and international forces could work.

The programme became a critical component of the overall international military strategy during the surge period. General Petraeus, credited with the apparent successes of the ‘Sons of Iraq’ programme, saw the ALP as the ‘game changer’ in Afghanistan, and having made winning Karzai’s approval for the plan a ‘top initial goal’, invested considerable time lobbying the President (DeYoung and Chandrasekaran 2010; Rubin 2010). The ALP was, in Petraeus’ view, ‘arguably the most critical element in our effort to help Afghanistan develop the capacity to secure itself’ (Norris 2013); it was seen as an initiative capable of turning the tide against the insurgency, ending the drawn-out confrontation with the insurgency, and making possible the exit of international forces without risking the collapse of the Afghan government.

Following its inauguration in 2010 the ALP grew quickly. Initially slated as a temporary (2–5 years) force that would peak at 10,000 men, in mid-2011 it was announced that it would be tripled to 30,000. In 2013 it was announced that it would expand to 45,000 with funding guaranteed by the Pentagon until at least 2018 (Pessin 2010; Goodhand and Hakimi 2014: 14–15, quoting US Special Operations Command figures). The US envisaged that 30,000 ALP personnel in 154 districts by the end of December 2014 would be assigned to 1,320 checkpoints across 29 of the 34 provinces of Afghanistan (SIGAR 2014). By December 1, 2014, there were 27,837 ALP personnel (SIGAR 2015: 99). In part this expansion can be interpreted as a measure to fill a gap in capacity as international forces drew down but before ANSF reached full strength (Chayes 2011). Although the programme is small relative to total ANSF (with over 350,000 personnel as of January 2014) the distribution of the ALP across ‘key’ rural locations, where government and international presence has remained tenuous and challenged by the insurgency, may endow it with disproportionate significance. Additionally, if what some of its proponents have said about the power of the specific programme design proves correct, its relatively small size may belie its significance to future security in Afghanistan.

Following the security transition process and the arrival of the national unity government of President Ghani and ‘CEO’ Abdullah after the protracted 2014 presidential elections, the Afghan government ‘has not determined the final disposition of the ALP’ (SIGAR 2015: 99). As of January 2015 it was unclear whether the US Department of Defense (DOD) still envisaged funding the programme to 2018 (SIGAR 2015: 99), although the Ghani administration may be ‘seeking money to continue the program’ beyond that time (ICG 2015:i). Whether or not the programme is continued, it has trained and armed some 28,000 people in villages of tenuous government reach. The ALP (and broader VSO), moreover, may well prove to have significance beyond Afghanistan. Reflecting on the VSO/ALP concept, Robinson argues that ‘[u]ltimately, success really involves applying this model elsewhere’ and notes that ‘[t]his is the vision Admiral McRaven [the then-outgoing Commander of US Special Operations Command] is driving toward’ (Manea 2014). US Special Forces now contemplate their possible role in future operations around the world, in areas besides the pursuit of terrorists (Robinson 2013). US policymakers seem (for now) reluctant to engage in the prolonged and involved forms of intervention that evolved in Afghanistan. The possibility of distributed SOF teams raising indigenous forces that address US objectives, without large-scale deployments, and in ways perceived as less likely to degenerate into ‘militia’, appears to be under active consideration as a policy option.

The ALP in Context: A Changing Zeitgeist

While official depictions of the ALP programme clearly suggest its basis in counterinsurgency, other aspects of the programme suggest that it is better located at the confluence between COIN doctrine and a wider contemporary shift in zeitgeist regarding how best to conceptualise, and most effectively intervene in conflict-torn spaces. In describing this turn, we emphasise a distinction between two related efforts. The first involves theoretically-driven work motivated by the desire to overcome the perceived limitations of dominant approaches to understanding such settings. The second is more oriented to designing and advocating policy interventions that respond to the perceived problems associated with existing approaches. The conceptual distinction between these different projects is less clear-cut in practice since several authors, implicitly or explicitly, seek to address both questions. However, we highlight the distinction to draw attention to what is involved in moving between seeking to better understand social phenomena and attempting to intervene to achieve particular, typically externally derived, objectives.

We present this turn in more detail in three steps. First, we set out the neo-Weberian terms in which the dominant theoretical and policy discourses were framed, and how perceived shortcomings of mainstream policies created a policy appetite for alternative approaches. In its emphasis on tapping the purported power of traditional forms of authority, the ALP programme clearly embodies the wider policy turn in international development and security circles towards overcoming the limits of the dominant ‘top down’ state-building model by devising interventions that work ‘with the grain’ of existing local governance (Kelsall 2008). Second, although justifications of the ALP superficially appear to be close cousins of many of the arguments being developed by scholars more oriented to theoretical challenges of conflict-torn spaces, this impression is misleading. Taking seriously the more theoretically driven literature should make us cautious of policies seeking to reflexively harness local dynamics in pursuit of international objectives. Third, as we seek to demonstrate through our discussion of Andar District, the emerging theoretical work provides a useful lens not only for understanding dynamics of conflict-torn spaces, but also for understanding the supposedly unexpected consequences of policy interventions seeking to tap such dynamics.

Our point of departure is the observation that previously dominant approaches to understanding the political order of conflict-torn spaces in ‘neo-Weberian’ terms have recently been rivalled by increasing interest in alternative forms of analysis.2 Neo-Weberian understandings of statehood derive from the Weberian ideal-typical bureaucratic-rational state seen, at root, as a monopoliser of the legitimate means of coercion. Although Weber intended his ideal types as ‘pure’ logical categories for analytical purposes that had ‘no connection with value judgments’ (Weber 2011 [1904]: 98; emphasis in the original), his formulation has in recent decades become a normative-teleological benchmark dominating both the scholarly and policy discourses (Migdal and Schlichte 2005; Hagmann and Péclard 2010; Weigand 2015). The view that deviation from this condition is a dangerous pathology threatening the wider community of states has underpinned the tendency to categorise states with a low degree of monopolisation of force as ‘fragile’ or even ‘failed’ (Fukuyama 2004; Rotberg 2004). As the problems besetting such places came to be defined in terms of state failure, so the view emerged that the solution lay in concerted international efforts to establish supposedly Weberian ideal-typical state structures. This state-building agenda is widely seen as the lodestar that guided a range of international interventions since the end of the Cold War, including that in Afghanistan. Yet after a period in which international actors grew increasingly ambitious, attempting the wholesale external engineering and transformative ‘modernisation’ of post-conflict countries, the visceral difficulties encountered in Afghanistan reinforced a sense of deepening doubt about the feasibility of such international endeavours (Suhrke 2007).

The apparent ‘poor performance and high cost of statebuilding’ (Meagher 2012; Paris and Sisk 2009) provoked a search for explanations as well as a new receptiveness at the policy level to alternative or complementary approaches. A view emerged that the travails of state-building were rooted in the gulf between externally derived understandings and objectives and local realities and demands. State-building stood accused of being overly ‘top down’, both in the sense of being derived from international prescriptions and in being disproportionately focused on central organs of state that would project government across the territory. Intervenors in Afghanistan, for example, were accused of assuming that strengthening the centre would have a ‘cascading’ effect in which ‘the rest of the country would become successively subject to the Afghan state’ (Schetter 2013: 8). Such interventions were thought to lack traction because they either largely ignored the periphery, assumed a tabula rasa on which they could build afresh (Ucko 2013: 529) or, influenced by the ‘pathologizing’ failed states and ungoverned spaces discourses (Bell 2012; Hughes and Pupavac 2005), treated areas beyond state control as sites of breakdown or Hobbesian anarchy (Call 2010; Hagmann and Hoehne 2009; Hagmann and Péclard 2010). A sense emerged that an external vision of government could not simply be imposed upon a given country and that such efforts could and would be locally resisted or co-opted (Englebert and Tull 2008).

Meanwhile, more theoretically-oriented scholarship had been expressing dissatisfaction with the intellectual blind-spots created by prevailing neo-Weberian state centrism, which caricatured areas where effective control of the de jure state was limited or contested as sites of failure and chaos (Hill 2005: 148, cited in Hagmann and Péclard 2010: 541). Consequently, they were incapable of seeing them for what they were and inattentive to the insight that ‘absence of the state does not mean a void in its place’ (Bierschenk and Olivier de Sardan 1997: 441). Scholarship emerged decrying the ‘failures of the state failure debate’ (Hagmann and Hoehne 2009), seeking to reconceptualise and empirically document the ways social life continues to be ordered and reordered where state control is limited or contested (eg Hagmann and Péclard (ed) 2011; Menkhaus 2006/7; Migdal and Schlichte (ed) 2005; Raeymaekers et al 2008; Schetter (ed) 2013; Vlassenroot and Raeymaekers 2004; Doornbos 2010; Friedrichs 2010; Weigand 2015; Schlichte and Wilke 2000). In the process, researchers sought new ways to describe and conceptualise the phenomena they encountered, avoiding dependence on prevailing ‘essentialist, teleological and instrumentalist conceptions’ of the state leaving ‘little room for alternative models’ (Hagmann and Hoehne 2009) and documenting the roles played by a range of ‘non-state’ actors in governance processes. Migdal and Schlichte (2005), for example, presented a dynamic understanding of political order and statehood that redirects the analytical focus from the degree of monopolisation of force to interactions between the involved actors or authorities. Hagmann and Péclard (2010) use this definition to develop their heuristic framework of ‘negotiating statehood’, describing ‘processes of state (de-)construction’ (ibid.: 544) as a non-linear and at least partly undetermined product of ongoing, dynamic interactions between differently situated actors (ibid: 545; Doornbos 2010: 752). Their emphasis upon processes of ‘negotiation, contestation and bricolage’ echoes the language of a number of scholars similarly interested in understanding and documenting processes through which ‘governance’ is produced in contexts of apparently limited government (in the neo-Weberian sense) (eg Lund (ed 2007); Menkhaus 2006/7; Mielke et al 2011).

We emphasise two analytic benefits of drawing upon the kinds of heuristic approach being developed in this research tradition. First, the notion that social order ‘never ceases to exist but rather changes its institutional and normative contents’ (Wilde and Mielke 2013: 353) draws attention to the ways new governance arrangements, however provisional, emerge as people negotiate confusing post-conflict settings in which questions of political order remain at least partly unsettled. A second benefit, which has received less attention, however, is that such approaches are capable of bringing international intervenors into the frame of study as additional sets of ‘stakeholders’ vying to influence the emerging political order. The ‘failed states’ discourse framed such settings more in terms of threats posed to neighbouring states and the wider state system but tended to neglect or caricature local settings. Consequently, while the failed states discourse served as a justificatory basis for intervention it was less well-equipped either to anticipate how particular forms of international engagement would be received, or how they would ‘play’ into fluid and contested local dynamics.

Where much of this work focused on the theoretical and empirical exposition of social and political life under such conditions, many working in this vein have also considered the policy implications arising from their ideas. An influential body of work developed by Boege et al., for example, advocates ‘re-conceptualising fragile states as hybrid political orders’ (2008: 15) that combine ‘state institutions, customary institutions and new elements of citizenship and civil society in networks of governance which are not introduced from the outside, but embedded in the societal structures on the ground’ (ibid.: 17). They express the hope that their ideas could ‘contribute to a reorientation of external assistance’, so that ‘possibilities of externally influencing governance structures can be re-examined, shifting the focus from narrow models of state-building to understanding and engaging with hybrid institutions’ (ibid). Kaplan, similarly, argues that rather than ’trying to foist a Western style top-down state structure on Somalia’s deeply decentralized and fluid society, the international community needs to work with the country’s long-standing traditional institutions to build a government from the bottom up. Such an approach, he adds, ‘might prove to be not only Somalia’s salvation but also a blueprint for rescuing other similarly splintered states’ (Kaplan 2010: 81).

The parallels between the broad policy turn described above and the ALP programme are difficult to ignore.3 The programme was developed against a backdrop of mounting insurgency, deteriorating relations between the Afghan government and its international partners, and deepening doubt about the feasibility of the international ‘project’ in Afghanistan (Suhrke 2011). The ALP was explicitly developed from a critique that echoed wider trends in locating the shortcomings of the Afghanistan intervention in an unhealthy fixation on neo-Weberian state-building and neglect and ignorance of local politics. One ALP advocate argued that since 2001, international policy had been based on ‘a fatally flawed assumption: The recipe for stability is building a strong central government capable of establishing law and order in rural areas’ (Jones 2009a; Jones and Muñoz 2010: 6–7). That assumption, rooted in the reconstruction and state-building template of the 1990s and 2000s, was contrasted with a portrayal of Afghan history as a series of top-down, centralizing attempts to project a (neo-Weberian) monopoly of force across the territory, typically provoking the rural, traditional and conservative majority of the population into ‘social and political revolts’ (Jones 2009a). Exemplified mainly by the failed efforts of Amanullah Khan between 1919–1929 (Jones 2009a; Jones 2009b; Jones 2010; Jones 2012: 21), similar claims are frequently made about the pro-Soviet Khalqi regime’s ‘all-out assault on rural conservatism’ in 1978–79, which sparked rounds of violence in Afghanistan that have yet to be concluded (Giustozzi and Ibrahimi 2012: 1). This influential account of Afghan history argues not only that much of Afghanistan existed beyond effective state authority, but that the very effort to exert such control had driven repeated cycles of violence (Jones and Muñoz 2010: 84). In this context, the contemporary insurgency began to look like the latest rural, conservative, traditionalist backlash against externally-driven centralising and modernising projects. To these critics, the idea that central state authority could be built and then projected into the Afghan countryside revealed the collective failure of the international community, academics and western-educated Afghan government technocrats to ‘grasp the local nature of Afghan politics’ (Jones 2009a; Jones and Muñoz 2010: 6–7). From this perspective, the empirically oriented and conceptually innovative scholarly work being developed in parts of the academy suddenly seemed exactly what US military planners wanted in order to understand and effectively engage the decisive authority structures in Afghanistan.

Reflecting the rising policy interest in alternative modes of intervention that work through existing authorities at the local level, the ALP programme asserted that the ‘key’ to stability in rural Afghanistan lay in the approach taken by the Musahiban dynasty (spanning the rule of Zahir Shah, Nadir Shah and Daoud Khan from 1929–1978). The Musahibans were portrayed as having recognised the futility of seeking to forcibly project their state into rural areas. Instead they viewed statecraft as the work of engaging with customary, tribal and other forms of governance and security provision beyond the state while managing local security ‘under the auspices of legitimate tribal institutions’ (Jones 2009). While such an approach might appear at odds with the mental template of an ideal-typical Weberian state, this conception of statecraft as encompassing diverse repertoires for negotiating relations with a range of non-state actors aligns closely with historical accounts of state formation processes in early modern Europe. As Goodhand and Hakimi (2014: 8) point out, ‘states and imperial powers have frequently acted as brokers rather than monopolists, seeking to extend their control through franchising the means of coercion’.4 This account of the Musahiban ‘secret’ to state-building, moreover, would not seem out of place among contemporary portrayals of ‘hybrid political orders’ or ‘mediated states’.

The view that a strategy of harnessing local institutions was superior to top-down imposition was used to argue that existing efforts should, at least, be complemented by strategies for understanding and working through existing local political and security institutions. In this the ALP paralleled the suggestion, made in relation to Somalia, that the ‘best hope for state revival may lie in the explicit pursuit of a mediated state – in which a central government with limited power and capacity relies on a diverse range of local authorities to execute core functions of government and mediate relations between local communities and the state’ (Menkhaus 2007: 103). Politics in Afghanistan were ‘local’, but having fixated on central state institutions, the international community had largely neglected the countryside and the villages. While the international community was ‘looking in the wrong place’ – focusing on national government institutions (Jones 2009a) – the Taliban had implemented a skilful bottom-up strategy aimed at persuading, co-opting or coercing local leaders, particularly of majority communities that under Karzai found themselves ‘marginalized by ruling minority tribes’ (Jones 2010: 334).

Given the extent to which local politics in Afghanistan remained opaque to interveners, the discovery of the importance of local and rural dynamics by international military actors was welcomed (Ucko 2013). Three issues with the historical account used to frame the ALP programme stand out, however. First, as Giustozzi has observed, the ‘successful’ Musahiban state-building strategy of achieving progressive rural domination through intermediaries and co-optation was an ‘imperial’ model of governance that worked not on the basis of impartial administration ‘but on certain communities ruling over others or on a strategy of divide and rule among local leaders’ (Giustozzi 2009: 71; Karokhail and Schmeidl 2006). From this perspective, the strategies of ‘successful’ periods were implicated in producing a ‘precarious’, crisis-prone and inherently unstable system. The local security mechanisms attributed to the Musahiban period had not existed in seclusion from the state but had been sites of negotiation and contestation as the government sought to increase its grip upon the periphery. Local institutions were profoundly (and differentially) affected by such processes – local security institutions, for example, varied considerably in terms of relations between local mechanisms and national security forces (Tariq 2008). State agents seeking to influence or instrumentalise local institutions, meanwhile, required high levels of understanding and skill. This suggests that contemporary efforts to ‘go local’ may risk generating similarly precarious dynamics, especially given the limited local level understanding possessed by outside interveners.

Secondly, the literature indicates that in the past Arbakai and similar institutions were embedded in a wider cultural framework, having been raised and controlled through customary councils that enjoyed significant local legitimacy. However, more than thirty years of continuous violent conflict in Afghanistan has, in addition to destroying physical infrastructure and state institutions, also had profoundly destructive and transformative effects on village level social structures (Noelle-Karimi 2006; Noelle-Karimi 2013). These dynamics empowered new classes of actors and drove local responses to insecurity that contributed to producing new forms of order and authority (Giustozzi and Ullah 2006). This compounds the difficulty in identifying what is ‘traditional’ or equating what is ‘local’ with legitimacy (Schmeidl 2009).

Thirdly, it is not obvious that insurgency in Afghanistan is best understood as a rural rebellion against a state-building effort in the tradition of earlier delusional visions of modernising grandeur. Portraying the insurgency as a rural mass rejection of state-building or modernisation efforts obscures that the day-to-day experience of many rural Afghans was not best described as an encounter with the grand, internationally-backed Afghan government-led, state-building and modernisation mission that was deemed to have failed or to have fuelled insurgency (Schmeidl and Karokhail 2009b). Particularly in the south and southeast, post-2001 rural experience was not primarily of an ambitious neo-Weberian state-building experiment, but of a heavily militarised international presence prosecuting a war on terror and empowering a government characterised by a combination of corruption, predation and ineptitude (Fishstein and Wilder 2012; Rangelov and Theros 2012).

The ALP in Practice: Impressions from Ghazni Province

In this section, we consider the evidence from Andar District of Ghazni Province, supplementing the available literature with interviews with local people. In order to investigate how the ALP programme works in a specific context, we conducted three rounds (spring 2013, 2014 and 2015) of phone interviews with a small number of selected community elders in the district. The case underscores the gap between portrayals of the programme ‘on paper’ and the way the programme has played out in practice. It also reinforces the sense, emerging from the body of existing literature, that the ALP is highly ambiguous in terms of its contribution to security/insecurity at the local level and in terms of a confusing patchwork of wildly divergent outcomes not only from village to village, but with ALP units morphing rapidly from apparently locally supported security provider to predatory actor.

In exploring the Andar case, we highlight some of the problems implied by international attempts to modify or invent local governance and security institutions by invoking tradition, history and culture to advance their objectives. Such approaches, we suggest, do not simply overcome the problems associated with ‘top down’ forms of intervention, but introduces tensions of its own. However, we also seek to demonstrate the analytic value of dynamic approaches to political order in conflict-torn settings by using such an heuristic to interpret what initially appear to be messy and unpredictable outcomes. Such approaches, we suggest, are capable of encompassing both local setting and forms of international intervention – even interventions seeking to reflexively harness the ‘local’ or ‘traditional’. At heart, we argue, the ALP reveals the tension in ascribing power and agency to local dynamics while continuing to assume that local interests can be harnessed and aligned to external objectives.

Several provincial and district-level studies explore the ALP in some depth (eg on Baghlan (HRW 2011; Goodhand and Hakimi 2014), Helmand (Stevens 2011), Herat (HRW 2011), Kandahar (ICG 2015), Kunduz (HRW 2011; Goodhand and Hakimi 2014; ICG 2015), and Wardak (Goodhand and Hakimi 2014)). These studies both underline the complexity of local circumstances into which particular ALP programmes have been inserted and create a chaotic impression of ‘highly uneven’ outcomes (Felbab-Brown 2013: 138). In some times and places the programme has appeared to deliver intended outcomes both in furthering anti-Taliban counterinsurgency objectives as well as in generating local perceptions that the ALP was making people safer. However, most observers provide evidence that in a significant number of cases – despite the claim that the programme had learned the lessons of its predecessors, and the emphasis on local accountability mechanisms in the programme design – ALP units have committed serious abuses against the population they supposedly protect, thereby deepening, rather than alleviating, insecurity (Aikins 2014; HRW 2011: 3; ICG 2015; Mashal 2011; Sarfraz and Norland 2012; Yoshikawa and Pennington 2011; Yousafzai and Moreau 2013). An unpublished US SOF study reportedly found ‘every fifth ALP is involved in the drug trade, extorting illegal taxes, land grabbing, murder, rape, running secret detention facilities and violent internal power struggles’ (Ruttig 2013: 5). As with similar past efforts, there has been concern that this latest internationally-sponsored armed actor may one day ‘go rogue’ (Goodhand and Hakimi 2014: 45), turning to banditry, being co-opted into militias or joining the insurgency and violently opposing the state (Borger 2012). There is some suggestion that this may be already underway. As the international presence in Afghanistan has receded, so has oversight of ALP units by their US SOF mentors, leaving some local residents ‘shivering with fear’ not at the Taliban but at the local ALP military entrepreneurs (Goldstein 2015; Stancati 2014). The reports of abuses that have dogged the ALP have also fed a nagging sense of unease that rather than the ALP helping villagers to resist the Taliban, villagers may actually be reaching out to the insurgency for protection from the ALP. Such a trend would indicate that besides often creating insecurity for local communities, the ALP can actually fuel the insurgency it was set up to defeat.

Competing Authorities in the Andar Uprising

The establishment of the ALP in Andar District of Ghazni Province is particularly interesting because of an apparently spontaneous popular uprising against the Taliban that broke out in spring 2012, generating widespread interest in Afghan and international media (Peter 2012; Moreau 2012). US Special Operations Command, perhaps reading events through the lens of COIN doctrine as evidence of the population breaking from the insurgency, became ‘seized’ with the uprising, seeing it as a ‘springboard’ for introducing the ALP (Robinson 2013: 203; Lubold 2012; Trofimov 2013). Subsequent efforts to replicate and spread the ‘Andar model’ underline the importance of this particular case in wider contemporary US military thinking on Afghanistan.

The dominant narrative about events in Andar holds that, having steadily strengthened their position (Chivers 2011; Harpviken 2012; Reuter and Younus 2009), the Taliban introduced a number of unpopular measures in Ghazni, culminating in the closing of schools in response to government efforts to ban motorcycles, which they relied upon (Foschini 2012; Habib 2012a). In this reading, the ban on schooling caused such strong local resentment that villagers spontaneously rose up and began to forcibly expel the Taliban from their villages. This account is largely followed by Robinson (2013), who reflects the US SOF perspective, and receives at least qualified support from contemporary media (Farmer 2012; Peter 2012). In an interview, a former mujahidin member and current community elder in the district offered a similar interpretation of the origins of the uprising. He pointed out that international and Afghan forces had regularly conducted military operations designed to drive out the Taliban, but that after military operations finished and forces withdrew, the insurgents immediately returned and resumed harassing and even executing local people. According to this interviewee, ‘The Uprising’ (‘Paatsoon’) was a locally developed initiative that reflected resentment of Taliban behaviour. However, the interviewee also stressed the failure of the government to retain control over the area following anti-Taliban clearing operations. The way our interlocutor framed his response in terms of government shortcomings and how he saw the uprising as an attempt to expel the Taliban and provide local security directly, offers an important clue. Many involved in the uprising were rejecting the Taliban without welcoming the government, and did not see their movement as a bridgehead for government control, as it was interpreted by outsiders (Foschini 2012; Moreau 2012). This middle position, neither Taliban nor government, however, would prove difficult to maintain.

While compelling, the emerging international narrative of a purely ‘popular’ uprising concealed how people in Andar, and even different actors involved in the uprising, were using widely differing narratives to explain the situation. Moreover, each of these narratives, including that emerging in international coverage, was politically charged; the language of ‘uprising’ can be viewed as a kind of ‘symbolic repertoire’ through which different actors seek to ‘defend and to challenge…power relations’ (Hagmann and Péclard 2010: 547). Foschini and Habib (a pseudonym) provide evidence that the uprising was not simply driven by popular resentment, but was marked from the outset by competition between local power-brokers with roots in the anti-Soviet jihad, each seeking to mold the uprising to serve their own objectives (Foschini 2012; Habib 2012a, 2012b, 2012c, 2013a, 2013b). Most accounts acknowledge the role of figures affiliated with Hekmatyar’s Hizb-i Islami, a mujahedin-era faction that remains influential in post-Taliban Afghanistan. After the Taliban resurgence in Andar from around 2003 (Reuter and Younus 2009), descendants of mujahedin era ‘Hizbis’ were incorporated into the Taliban movement but held significantly different views from their comrades on questions such as education and development work (Habib 2012a). Robinson, perhaps reflecting the US SOF narrative, maintains that there was a spontaneous uprising but suggests that when its leader was killed, a Hizbi ‘political figure’ then ‘insinuated himself to assert leadership’ (2013: 204). Foschini (2012) similarly believes that the uprising was driven by ‘a broader and deeper malaise’ conceding only that ‘[t]he Hezb-e Islami connection has certainly contributed to strengthen and militarise the revolt’. On the basis of detailed interviews with local informants, however, Habib relates alternative local narratives portraying ‘the “uprising” as a revolt by the “Hezb Taleban” group against their Taleban brothers’ (Habib 2012a) – a view that is supported by the way the revolt apparently began in a series of historically Hizbi villages south of the district centre. An alternative account, which also interprets the uprising as a Taliban rift, plays down the Hizbi doctrinal schism theory, instead emphasising local dissatisfaction at growing Pakistani influence in the local Taliban movement (Economist 2012; Felbab-Brown 2013: 149; Rubin and Rosenberg 2012).

An additional narrative, corroborated by a range of sources, emphasizes the role played by Assadullah Khaled, a Karzai-affiliated figure who would soon head the Afghan National Directorate of Security (NDS). Depicted by Habib as ‘probably the single most powerful figure in Ghazni calling the shots from behind the scenes’ (Habib 2012a), Khaled is alleged to have reached out to three rival local power-brokers with mujahedin-era roots, seeking to influence the uprising and realign it as an anti-Taliban and pro-government movement. In return for mobilizing pro-government arbakai, ‘each would be able to hire his men for it, thereby gaining power in the area again’ (Habib 2012a). One of those apparently approached by Khaled grew suspicious of his Hizbi rivals and felt ‘sidelined’ and then ‘betrayed’ when his arbakai commanders were killed (ibid). However, the Hizbi leaders who appear to have split with the Taliban and initiated the uprising were themselves soon marginalized. If it had been ‘started by a staunch, anti-government, anti-ISAF Hezbi group, fighting for ideological reasons’ within months the uprising ‘turned into what looked and acted like an arbakai – an anti-Taleban militia which fights on the government’s behalf and is supplied and supported by ANSF and coalition forces’ (Habib 2012a). Khaled may have sought to capitalise on internal division within the Taliban, and then to assume control of the movement, progressively marginalizing rivals through alliances with more amenable local commanders.

The rivalry between local power-brokers who had emerged during the war years was proving to be as significant to local politics as were the ‘traditional’, ‘tribal’ sources of authority invoked in the ALP concept. Even if the uprising had begun in spontaneous and popular terms, within months local struggles to instrumentalise it were resulting in fragmentation, with some villages apparently flying the flag of Hizb-i Islami, and others the government of Afghanistan. This had important consequences. In addition to generating an escalation of local violence as a result of the Taliban backlash – within five months more than one hundred people had been killed (Habib 2012b) – there were reports of clashes between different arbakais and growing wariness as they began to harass local people suspected of Taliban sympathies. In the early months, people in Andar referred to the anti-Taliban rebels as arbakai. As described above, international forces and ALP architects had attached positive connotations to this term, but within a year people in Andar were using the term ‘unanimously and perjoratively’ to ‘denote a government or foreign-backed local force which fights, not for the protection of the local community, but because they are paid by outsiders and therefore act as irresponsible mercenaries’ (Habib 2013a).

The Uprising exposed the difficulty faced by villagers wishing to reject the Taliban without aligning with the government. Once the Taliban determined to use force against the rebels, the latter had little choice but to accept or reach out to power-brokers capable of keeping their movement alive. Intra-uprising divisions, both between ‘Hizbis’ and those affiliated to Khaled and within the latter group made the symbolic appeal of the uprising vulnerable to charges that it had been co-opted. As noted by Foschini (2012), while support of government-linked powerbrokers might enable the uprising to survive the Taliban backlash, this might not mean ‘better security and increased possibilities for local kids to attend school – the originally stated objective of the rising’. Moreover, the more the uprising was perceived to be moving into a pro-government orbit, the more likely it would be to elicit concerted Taliban opposition. All in all, as one observer told The New York Times, the uprising had become ‘a bit of a mess’ (Rubin and Rosenberg 2012).

Enter the ALP: A ‘Helping Hand’ – but for whom?

In late September 2012, barely six months into the uprising and in the midst of these complex local dynamics, US SOF arrived in Andar and soon began formal training for the ALP. The uprising they encountered did not simply reflect unified opposition to the Taliban, but a politically fragmented environment in which competition between local power-brokers seems to have been more decisive than the ‘traditions’ the programme was meant to tap. The Hizbi faction was militarily significant but hostile to the US, which may explain why international commentary emphasised the ‘popular’ character of the uprising, and why US SOF, wishing to marginalise Hizbi influence, distinguished between the ‘legitimate’ ALP that they trained, and the original defenders (Robinson 2013: 205). Adding to the complexities of a ‘popular’ movement already steered to some extent by Hizbi leaders as well as Khaled’s behind-the-scenes activity, the US was now ‘essentially trying to set up a competing local defense force that was not under HIG [ie, Hizb-e Islami] influence’ (ibid). Where Hizbis were critical of the Taliban and of the government, US SOF sought to draw the uprising into their orbit as a pro-government counter-insurgency force. On paper the ALP merely offered a helping hand to villagers tired of insurgency and wanting to ‘stand up for themselves’ (Catanzaro and Windmueller 2011). The way the ALP appears to have been established in Andar, however, exposes that the objectives of local uprisers, themselves not unified, were not synonymous with international objectives. A programme that claimed to simply reinforce spontaneous local resistance in practice sought to harness the uprising ‘repertoire’ to fulfil overarching counterinsurgency objectives to the extent that it built a parallel force that alienated existing uprisers. Those uprisers who did not view their rejection of the Taliban as an endorsement of the government were nevertheless forced, however reluctantly, to rely on outside help by military necessity.

A further perception was that the ALP was not set up in the way envisaged on paper. An interviewee told us that rather than being appointed through local consultation in the way formally envisaged, he believed recruits were applying directly to district and provincial police officers, with positions filled by ‘young, unemployed people who have a background in petty crime’, and motivated by a salary. This perception echoes a range of other sources, who judged the ALP by its actions rather than its branding and were no less suspicious of this latest armed group. For example, Habib (2013a) reports a local teacher’s view that ‘[t]heir treatment of the people did not change. They are the same arbaki guys. They only changed their façade’. The community elder we interviewed was disappointed with the government for not supporting the uprising and instead setting up the ALP as an alternative ‘local’ force, which did not provide more security but was an additional source of insecurity, saying ‘we demanded the government to support the Uprising instead of sending drug-addicts and thugs pretending to police our communities’. From the interviewee’s point of view, the way the ALP was set up reflected that ‘the government and provincial police authorities are more concerned with demonstrating their presence’ but were ‘not concerned about the demands and needs of the communities’. The case of Andar suggests that even where villagers showed a desire to ‘take a stand’, international/state efforts to harness such movements were sometimes locally perceived not as a helping hand, but as an attempt to align these efforts with international/state security objectives with which they did not identify.

However, Andar also illustrates how national and local actors were not passive in the face of this ‘bottom-up’ international counterinsurgency intervention, but sought to harness the ALP to agendas other than (and often at odds with) the international leitmotifs of opposing and defeating the insurgency and extending state authority. While the ‘formal’ process of progressively incorporating the uprising into the ALP programme appeared quite successful, this did not simply replace pre-existing contestation over the uprising among Afghan stakeholders. Local ALP recruits continued to assert their independence, one suggesting to journalists that they had agreed to join up because of ‘the shortage of weapons and supplies to continue the fight’ but were ‘not like the other ALP units’ (Habib 2013a). By accepting external support, the uprisers risked their claim to local legitimacy by allowing the Taliban to portray them as being ‘in the foreigners’ embrace’ (Trofimov 2013).

While the ALP on paper envisaged reinforcing an uprising such as that in Andar, in practice the ALP was building a new force that soon sought to expand into additional villages. Violence continued to escalate throughout 2013. By November the local conflict in Andar had claimed more than 300 lives ‘exceeding all the dead of the conflict between summer 2003 and summer 2012’ as well as becoming qualitatively worse, with opponents even denying burial to fallen adversaries (Habib 2013b). Youths engaged to fight on both sides ‘do not feel bound by any outside authority or rule book and their way of fighting is entrenching hurt and anger’ (ibid). Following the wider post-2014 transition process, the support provided by local US SOF has been removed. Goldstein (2015) reports that the ALP in Andar has subsequently become more vulnerable and also less controlled, and reports accusations that ALP units in Andar engage in kidnappings, beatings, extortion and extrajudicial killing ‘partly to feed themselves and partly because there is no one to stop them’.

The Andar uprising was not as it was portrayed internationally – the heuristic of dynamics draws attention to disparate understandings, internal divisions, and how the uprising ‘narrative’ was itself contested by different actors, including international actors, vying to define it in their own preferred terms. The Andar ALP was not inserted into the ‘tribal’ world evoked by the imagery of village elders and shura decision-making, but a landscape in which the class of military entrepreneurs that emerged during decades of conflict were significant. A range of influential figures actively sought to manipulate the very attempts at negotiation and co-option employed by international actors to advance their objectives at the local level. Around the country, wartime commanders were finding ways to have their militias enrolled in the ALP to obtain arms, salary and backing from US forces (Felbab-Brown 2012; see also Shirzay 2012). As a result, the ALP became a means through which a range of actors might obtain powerful external support and resources while claiming the legitimacy of tradition, or as was the case in Andar, of a popular movement. For all that the ALP invoked the power of the local on paper, in practice the attempt to harness local dynamics to international COIN objectives proved problematic. The ALP did not simply align with and empower local anti-Taliban mobilisation, nor did it reconfigure local dynamics behind international counterinsurgency objectives. Instead, as international actors sought to advance their objectives through the programme, so too did local actors seek to harness the programme to their own agendas.

Rather than evaluating the ambiguity of the Andar ALP in terms of the degree to which it appeared to advance or undermine stated counterinsurgency objectives, we suggest its practices are better thought of as a contingent outcome of contestation dynamics between multiple actors. Rather than aligning local actors with international COIN objectives, the ALP in practice expresses the disparate security interests of these actors playing out under a nominally singular banner. Thus, the ALP’s behaviour in Andar is better understood as a ‘merged’ practice, its contribution to security or insecurity driven by ongoing contests between actors pursuing quite different security ‘objectives’. In this sense the ALP ‘modality’ of intervention, precisely because it sought to work through the local, produced distinctive, apparently messy, security outcomes.

Conclusion

On paper the ALP expresses COIN doctrine, but in ‘going local’ it also shadows a wider shift in zeitgeist. The programme might first appear as a logical policy conclusion from some of the emerging analytic research on conflict-torn spaces, purportedly overcoming the unsustainability of heavy footprint counterinsurgency by recognising and harnessing the latent power of Afghan culture and traditions to achieve ‘smarter’ forms of intervention. These attributes were combined to create a programme that aligned international counterinsurgency objectives with an overwhelming local desire for security, even during the drawdown and withdrawal of international forces from combat roles. In light of the grave problems associated with the programme, the ALP may then appear as a cautionary tale warning policymakers of the less palatable implications of ‘bottom-up’ interventions. Here, however, we suggest that a more careful examination of the move from analysis to intervention is needed. While highlighting that such policies do not simply overcome the problems of ‘top-down’ approaches, but also introduce tensions and ambiguities of their own, we emphasise the analytic strength of the underlying heuristic approaches supposedly inspiring bottom-up policies. Moreover, we have suggested such approaches enable useful insights into the dynamics of local settings and external interventions, and help to anticipate why the ALP, for all its emphasis on locally appropriate forms of intervention, has still proven so messy in practice.

Taken at face value, the literature rethinking the political order of conflict-torn spaces closely parallels particular claims underpinning the programme. In particular, interpreting Afghan history as a tale about the inappropriateness of seeking to project state authority into a rural periphery governed by ‘traditional’ authority structures is used to depict existing international policy as misguided and justify the turn to ‘go local’. This argument echoes broader critiques of state fragility discourse and neo-Weberian state-building policy that constitute a point of departure for the literature rethinking the political order of conflict-torn spaces. Arbakai and similar pre-existing security institutions are then presented as the key to past success in rural security provision: the government having worked with the grain of existing non-state security institutions rather than provoking conflict by seeking to override such institutions. This reasoning, again, closely parallels debates in the literature rethinking the political order of conflict-torn spaces regarding the need to better understand and engage with existing local institutions.

The widely observed difficulties with the programme at first suggest that the policy turn to ‘bottom-up’ may not simply correct mainstream approaches but also introduce new complications. Here we insisted on distinguishing policy advocacy from more analytically-driven work since the latter, far from simply justifying the bottom-up policy turn, actually helps to anticipate some of the difficulties. To demonstrate this analytic strength, we adopted a more dynamic understanding of political order as an heuristic analytical lens. In the absence of a monopoly of force, the political order of Afghanistan can be thought of as an arena of competition and negotiation of various authorities with different security objectives, degrees of influence and relationships that vary spatially and change temporally. Depending on the specific dynamics of each locality, the competition over the ownership of the ALP resulted in a context-specific merged practice. Hence, even though the ALP on paper is described as being embedded in the counterinsurgency ‘security culture’, its implementation in the context of the outlined dynamics make it look very differently in practice. The changing zeitgeist regarding the political order of conflict-torn spaces and the evolving literature on dynamic statehood is analytically valuable but should not be mistaken for a new policy blueprint guaranteeing legitimacy and efficiency in attempts to transform political order.

In the case of Andar, the ALP’s contribution to international counterinsurgency objectives and to community level security was highly ambiguous. We have illustrated that this ‘messiness’ in practice can be better understood by using the literature rethinking the political order of conflict-torn spaces as an analytical framework – and not as a new basis for designing international interventions such as the ALP. Just as ‘neo-Weberian’ approaches are widely perceived to have substituted a normative-teleological idea of the state for Weber’s ideal-typical analytic tool, the ALP programme can be interpreted as demonstrating the dangers in moving from concepts such as hybridity for analytic purposes to employing them as the grounds for new modalities of intervention. Using this heuristic it becomes clear that although the ALP claims to take seriously the power of local agency, this claim is undermined by the assumption that that power can be harnessed and aligned with international objectives. The way the programme plays out reflects the outcome – to some extent unintended by any single actor – of processes of contestation between multiple actors – local, national and international – all vying for ‘ownership’.

This is not the end of the story. Despite having been inspired by US COIN efforts and still being sustained by US funding, the security transition process has meant that ALP units receive less direct mentorship than during the surge period. Consequently, the ALP is both less supported and less controlled and has, to some extent, taken on a momentum of its own as it becomes enmeshed in and reinforces local conflict dynamics. The ALP never became the animal that its creators depicted ‘on paper’, but its practices continue to be at least as much propelled by local dynamics as they have been successful in harnessing them. Meanwhile, the ALP appears to have a community of supporters who see it as a model with applicability to a range of conflict-torn spaces. Here we reiterate that emerging analytic work on the importance of dynamics in such spaces, often seen as the basis for ‘going local’, actually provides a valuable lens for anticipating some of the supposedly unintended consequences likely arising from the jump to ALP-like policies in other contexts.

Competing Interests

The authors declare that they have no competing interests.

Acknowledgments

Our sincere thanks go to Mary Kaldor, Sabine Selchow, Anouk Rigterink, Ali Ali, Sally Stares, David Brenner, Theowen Gilmour, Louis-Alexandre Berg and two anonymous reviewers for their invaluable comments and suggestions. We also want to express thanks to the participants at the workshop on the ‘Unintended Consequences of Statebuilding Interventions’ at Queen Mary University of London in March 2014 and at the International Studies Association’s (ISA) 56th Annual Convention in New Orleans, Louisiana in February 2015. Florian Weigand gratefully acknowledges the support of the Economic and Social Research Council (ESRC) (grant number ES/J500070/1). We are also grateful for the support of the Security in Transition Research Programme, funded by the European Research Council (ERC) (grant reference 269441) at LSE’s Civil Society and Human Security Research Unit, and for the opportunity to contribute to the special issue on the security gap.