Post-ISIL reconciliation in Iraq and the local anatomy of national grievances: the case of Yathrib

ABSTRACT The incentive for international agencies to promote ‘local’ peacebuilding is commonly premised on the hope that micro-level interventions can nourish grassroots activism and participatory citizenship. Tracing reconciliation processes across Iraq following the defeat of ISIL provides a window through which to view the transactional relationships between ‘national’ and ‘local’ forms of politics, conflict and peace in post-Ba’athist Iraq. This paper focuses on the example of Yathrib, Salah al-Din province, where over ninety per cent of residents were displaced in 2014, and an estimated eighty-five per cent subsequently returned following peace negotiations. The paper uses qualitative interview findings to demonstrate on the one hand that seemingly ‘local’ tribal solutions are built into national-level ‘peace strategies’, while on the other, state capture and power politics is infused into the management of apparently parochial disputes. While these observations are not an indictment of international efforts to intervene at the subnational level, they serve to thoroughly ‘de-romanticise’ the local.


Introduction
The 'local turn' to peacebuilding has undergone several theoretical and policy-oriented shifts since rising to prominence nearly thirty years ago. 1 In a 2015 critical review of the trend, Paffenholz identifies two 'waves' of scholarship advocating increased emphasis on the 'local'. 2 Both emphasise that peace cannot be simply imposed on conflict-wrought states by external actors, but differ in other respects. The first, following Lederech's work after the Cold War, broadly conceived the 'local' as actors at the middle level of society (such as NGOs and 'community' leaders) who could influence both national and grassroots levels, and were therefore deserving of capacity-building assistance from international development agencies. 3 The second, represented by the work of Mac Ginty and Richmond, emerged as a critical reaction to the failure of wholesale liberal statebuilding in Afghanistan and Iraq in the early 2000s. 4 These scholars problematised the international application of liberal peace projects, and instead promoted 'a post-liberal order founded on emancipatory local agency'. 5 In this latter context, the 'local' was often simply treated as pertaining to the domestic as opposed to the international, whether at the national, provincial, or district level. For her part, Paffenholz criticises the second wave's tendency to understand international and domestic peacebuilding initiatives as binary opposites (i.e. liberal versus local), with little regard for the fact that international interventions are not monolithically 'liberal', and that 'local' actors do not uniformly share an agenda for promoting their citizens' welfare. She also questions the assumption that local actors will resist liberal international interventions, citing examples of domestic actors at the mid-level of society resisting predatory elites which have captured the state at the centre. 6 Although the failure of the international statebuilding project in Iraq post-2003 shaped the second wave of local peacebuilding scholarship, surprisingly little academic attention has subsequently been paid to efforts promoting local peacebuilding in Iraq. In the contemporary Iraqi context, 'local' is commonly used interchangeably with 'subnational' to apply to processes at provincial, municipal, and district levels. What matters for prospective peacebuilders is not so much the unit of analysis, as the possibility that peacebuilding at whatever subnational level could be driven by different logics to those at the centre, and thereby provide intervening agencies with outlets to promote participatory citizenship and grassroots agency. Such hopes generated initial optimism for efforts to decentralise Iraq's provinces as a means of lessening conflict. 7 In practice, decentralisation has focused on administrative measures, with few political or revenue-raising powers granted to local governments. Lack of formally ceded autonomy impacts the ability of these 'middle' level government officials to shape how peace takes effect in their constituencies, but it has not rendered them inactive. As will be seen, provincial and subprovincial political actors operate within 'local' transactional political fields, which often overlap with, as opposed to merely being subsidiary to, politics at the centre, but do not present a fundamentally different logic of governance. With respect to local peacebuilding, externally backed development actors must be able to at least navigate underlying domestic logics if they are to help implement settlements.
The story of government in post-Ba'athist Iraq, at both the centre and peripheries, is one of progressive state capture. While technically based on competitive elections, Iraq's political settlement is a product of 'de-Baathifcation' policies endorsed by the Coalition Provisional Authority in 2003, and an allocative system, or 'muhasasa', whereby ministries, state-owned enterprises, political posts and public sector jobs are divided amongst dominant political parties and their followers. 8 While patronage politics was widespread under Saddam, the Coalition embedded a consociational division of power, consolidating patronage networks along sectarian lines, and establishing the Shi'a (roughly 65% of the population) as dominant within them. 9 Over time, while hostilities related to sect have persisted, allegiances based on sect have fractured, and multiple political factions, most of which enjoy pockets of subnational influence, have competed for a 'slice of the pie'.
The exclusion from federal government felt within Sunni-Arab majority provinces is perceived as a key factor in the rise of Islamic State of Iraq and the Levant (ISIL) from 2014, even if more 'parochial' pre-existing and largely non-sectarian disputes contributed to some supporting ISIL. In the course of ISIL's occupation and the subsequent 'liberation' campaign, around 6 million Iraqis were displaced from across central and northwestern Iraq. Many faced enormous challenges in returning, including the seizure of their property by 'liberating forces' (a combination of Iraqi government security forces, Kurdish Peshmerga, and volunteer forces organised around tribal and ethno-sectarian identities), death threats from the remaining population over suspected ISIL sympathies, or exorbitant demands for blood money payment to families of ISIL victims. In this context, various religious and tribal figures, municipal, provincial and national politicians, federal government representatives, formal and informal security actors, and external agencies have intervened to support dozens of lengthy reconciliation processes between hostile families to enable the displaced to return.
This paper looks at one such reconciliation process, pertaining to Yathrib in Salah al-Din province, a mixed Sunni-Shi'a community in the Sunni-Arab heartland, where over 80,000 residents were displaced between June and November 2014. Using findings from a series of qualitative interviews with key stakeholders in the negotiations as well as protagonists and bystanders from different sides of the dispute, the paper seeks to understand how 'peace' was manipulated by actors and issues that intersected multiple layers of governance, and how this necessarily framed the parameters of intervention by externally funded peacebuilding agencies. While the dynamics of reconciliation processes elsewhere in Iraq, including Qaim in Anbar, Tikrit in Salah al-Din, and Telafar in Ninewa may produce equally valuable insights about peacebuilding in divided communities, here, we focus on Yathrib because the findings we accumulated clearly demonstrate how the multi-scalar tribal and transactional politics which permeates post-Ba'athist Iraq affected peacebuilding in the district. 10 The article's findings are based on semi-structured qualitative interviews by the authors with 19 individuals based in Iraq, including 11 residents of Yathrib and the surrounding region in Salah al-Din, former employees of Sanad (an Iraqi NGO specialising in conflict resolution established in 2013 by the United States Institute for Peace -USIP) and Sanad's Network of Iraqi Facilitators (NIF), members of the United Nations Development Programme (UNDP), and a former advisor to the Iraqi government on reconciliation. The interviews were conducted between September 2019 and February 2020. While some interviews were conducted in person by one of the authors, 9 In 2018, Transparency International ranked Iraq 169 out of 180 countries in terms of corruption. 10 Overall, more international attention has been given to local reconciliation processes in the ethnically diverse Ninewa province, possibly in part because of donor interest in minority rights, and in part because Ninewa is relatively more accessible to donor agencies operating from the Iraqi Kurdish Region than other parts of the country. For an analysis of a parallel reconciliation process that occurred in Ninewa, see Sarah Sanbar, Barriers to post-ISIS reconciliation in Iraq: Case study of Tel Afar, Ninewa (Sciences Po, Kuwait Programme, 2021). https://www.sciencespo.fr/kuwait-program/wpcontent/uploads/2021/02/Sarah-Sanbar-Post-ISIS-Reconciliation-in-Iraq.pdf.
due to the prevailing security situation in the area and in Iraq more generally at the time of research, over half of the interviews were conducted remotely. We also had recourse to a collection of secondary Arabic language news articles and television bulletins retrieved via YouTube. The paper is divided into five sections. The first situates 'reconciliation' at national and subnational levels as a form of transitional justice in post-conflict societies in general, and in post-2003 Iraq in particular. The following two sections address events in Yathrib: part two focuses on the period between June 2014, when ISIL took control of the district, and December 2014, when Iraqi ground forces and international air forces defeated militants in Yathrib. Part three outlines the respective roles played by local, national and international actors in the messy process of brokering reconciliation between 2014 and 2019, demonstrating the extent to which Sanad's efforts were bound up with those of more selfseeking political actors. Part four questions assumptions about the inclusiveness of 'local' peacebuilding by highlighting the ambiguous divisions between different layers of governance, and the frequently exclusionary nature of tribal and transactional politics combined which influenced the shape of peace. The concluding part reflects on how Yathrib informs the debate on national versus subnational forms of peacebuilding, and the role played by external donor agencies. Ultimately, the authors reserve judgment on the inherent value of international actors supporting specific local peacebuilding initiatives but highlight considerations worth factoring into calculations about the potential benefits of intervention.

Post-conflict reconciliation and its particularities in post-ISIL Iraq
The need to restore conditions for sustainable intercommunal peace, if not harmony, through 'reconciliation', has been formally recognised in post-conflict states including South Africa, Sierra Leonne, Argentina, Haiti and Iraq as a key component of transitional justice. The UN defines transitional justice as 'the full range of processes and mechanisms associated with a society's attempt to come to terms with a legacy of large-scale past abuses, in order to ensure accountability, serve justice and achieve reconciliation'. 11 As such, it has included a combination of judicial and non-judicial proceedings including truth-telling commissions, reparations, prosecution, and institutional reform.
Clearly, the circumstances under which conflicts end influence the characteristics of reconciliation. Where a clear victor emerges, there is often a tendency to impose retributive forms of justice (punishing offenders), whereas negotiated peace provides greater potential, and indeed pressure, for restorative justice (restoring relationships between offenders and victims). Inevitably, transitional justice initiatives can be highly asymmetric. Writing about the recent Syrian civil war, Sosnowski argues that the regime has used local truces termed reconciliation agreements as 'tools of subjugation and control' akin to 'strangle' contracts since they leave the weaker party with little choice about the outcome. 12 By branding these arrangements 'reconciliation', she argues, the regime seeks to derive international legitimacy.
Since severe internal conflict almost invariably entails mass displacement, post-conflict peacebuilding and reconciliation is often concerned at least partially with resettlement. 13 The degree to which refugees and/or internally displaced individuals are given a voice in post-conflict negotiations has varied, generally in accordance with the relationship between the displaced and ruling elite, but is overall low. 14 Conflicts which have become national in scale are often bound up with 'localised' grievances, which, to address the particular needs of the constituencies in question, arguably require reconciliation processes which are distinct from national-level political settlements. 15 Support from Western development agencies to reconciliation processes has been predominantly accorded at the national level, or via the delivery of training in conflict-resolution skills to community actors. 16 In post-ISIL Iraq, however, Sanad actually intervened into specific local dispute-resolution processes, providing important case-study material for those interested in how externallysupported local peacebuilding plays out in practice.
National reconciliation has been a recurrent theme on Iraq's political agenda since 2003. Premised initially on the need to grant justice to the thousands of victims of Saddam's regime, the meaning of 'reconciliation' has subsequently evolved, reflecting the need to make amends for wrongs committed during sustained bouts of sectarian warfare in mixed Sunni-Shi'a communities in Iraq and the emergence of Sunni-Arab militant insurgencies. Many conflicts premised on sectarian or ethno-sectarian grievances have 'tribal' undertones, and reconciliation processes commonly follow tribal customary models. In Arab tribal societies, 'reconciliation' (musalaha) is a key component of customary law that regulates relations between disputing parties. When blood is shed as a result of conflict between two or more individuals, families, or tribes, the family of the offending party must offer to pay compensation or diya (a Quranic term for blood money) to the family of the victim, as well as satisfying other terms to ensure that the victim's family or tribe will not seek revenge. While civil laws (and in some cases shari'a) have formally replaced tribal laws across the Middle East, customary laws pertaining to dispute resolution are still widely observed. 17 This is particularly the case in less urbanised and conflict-prone settings, where the civil legal system has no teeth. In Iraq, tribal dispute resolution processes have frequently filled the vacuum left by the state's inability to enforce civil law and order and were commonly pursued in the years of heightened sectarian violence between 2005 and 2008, and after ISIL's defeat. 18 During ISIL's campaign across central and northern Iraq between 2014 and 2017, ISIL militants killed tens of thousands of Iraqi security personnel as well as civilians, targeting Shi'a and other minorities, but also Sunni-Arabs whom they identified as apostates. ISIL achieved easy victories over the Iraqi army and police in many areas. Security forces were unable to defeat the militants until mid-2017, assisted by international air strikes and the Popular Mobilisation Forces (PMF). The majority of the PMF hail from Shi'a-majority provinces, and a number are backed by Iran.
The Iraqi government's position towards ISIL after its defeat was unequivocal. Members were declared terrorists and those captured sentenced to death. ISIL family members were also treated harshly: many were forcibly detained in camps, and only freed if they publicly denounced relatives and cooperated with the authorities in identifying other members. The question of how to deal with suspected ISIL sympathisers, however, has been more complex. During the ISIL campaign and counter-campaign, an estimated six million Iraqis (15 per cent of the entire population) were displaced. 19 They included Shi'a Arabs, Kurds, Christians, Shabak, Yezidis and Turkman; but the majority were Sunni-Arabs. Some fled to escape ISIL; others, who for various reasons remained under ISIL rule, fled during the 'liberation', fearing damage to their possessions and/or sectarian reprisals from Iraqi security forces and PMF.
The security landscape in 'liberated' Iraq has changed dramatically. In many areas, the PMF and their political allies retain de facto control over local security. The PMF were formally incorporated as an independent unit into the state security forces by the Iraqi parliament in November 2016. 20 Nevertheless, a number of units, particularly those loyal to Iran, act largely autonomously, and have ignored government directives to withdraw from urban areas. The newly acquired political clout of these organisations (the Fatah alliance, which incorporates leaders of the key 'pro-Iran' components of the PMF won over 14% of the vote in the 2018 parliamentary elections) has enabled them to play instrumental intermediary roles in brokering agreements concerning the conditions under which displaced families can return. The following sections consider how this process played out in Yathrib.

ISIL's campaign and the counter-campaign on Yathrib
Yathrib is a small district in Balad municipality in the south of Salah al-Din, 100km from Baghdad. The district covers approximately 4.5 square kilometres and according to the central directorate for statistics in Salah al-Din, 2019, has a population of around 83,151. The land is highly fertile with an abundant water supply from the Dijla River to the north and east. To the west is Balad town, to the south is Dujail district. Yathrib is close to Balad air base, as well as the Imam Al-Assiyd Muhammad bin Ali Al-Hadi Shi'a shrine, which was stormed in July 2016 by ISIL militants.
Most residents are cultivators who farm dates, citrus orchards and grapes. An intensely tribal character has prevailed in the area, and strong allegiances are paid to rural traditions. The tribesmen hail from the Bani Tamim Tamim comprise the largest tribes in the area: prior to ISIL's campaign, around 40,000 Bani Tamim lived in 'Albu-Hashma settlement in Yathrib; 7,000 (known as the Albufadus) in Hatemiya; and roughly 5,000 in Albu Hassan settlement. Around 12,000 Zubayd Albu Sultan lived in al-Mazari'a settlement; 19,000 al-'Izza in Albu Jayli settlement; 10,000 Bani Assad in Ahbab village; 9,000 Al-'Aniyyin in the Hadera village; 7,000 Bani Saad in 'Al-Saoud' settlement; and 5,000 Karaghul in a settlement south of Yathrib. 21 While Yathrib (and indeed Salah al-Din more generally) is predominantly Sunni-Arab, the Bani Sa'ad, Albufadus Hatemiya, and some Albu Hassan are Shi'a, as are several of the tribes in neighbouring Balad town. 22 Since 2003, Salah Al-Din has been repeatedly identified by the Iraqi government as a source of Sunni Islamist militancy. It is situated within the so-called 'Sunni Triangle' northwest of Baghdad. Tikrit forms the northern tip of the triangle, Ramadi and Baghdad form the southwestern and southeastern points. After the 2003 US-led occupation of Iraq, much of the Sunni militant opposition to coalition forces and the new Shi'a-dominated government emanated from this area. Balad municipality itself was a source of recruits for Al-Qa'ida, the 'Army of the Men of the Naqshbandiya Order' and Ansar al-Sunna. 23 On 11 June 2014, the local police in Yathrib surrendered unexpectedly to a small convoy of ISIL militants entering from the north and east of the province. A few hours later, the army units protecting the perimeter withdrew, mostly to Balad airbase. On the same day, ISIL militants seized control over all Yathrib's settlements barring the predominantly Shi'a Bani Sa'ad, Albufadus Hatemiya and Albu-Hassan settlements, where tribesmen resisted. According to informed local sources interviewed by the authors, the number of fatalities was in the hundreds, whilst thousands were injured.
In Yathrib's centre and surrounding settlements where ISIL militants took control, they forced residents, civil servants and members of the security forces to declare allegiance to their self-proclaimed Caliph, Abu Bakr al-Baghdadi, and to disavow the Iraqi state. They then introduced a range of restrictions on residents, including shutting hairdressers (which they proclaimed immoral), forbidding smoking and internet use, seizing and destroying satellite dishes, demanding regular mosque attendance, imposing conservative dress codes, and prohibiting women from going out except when accompanied by male family members. Villagers reported that those who disobeyed received punishments, which in their most extreme form included stoning to death. Young men were obliged to join the militants and to execute those who refused in public squares. 24 Some tribesmen, including members of the Albu-Jayli, Al-Mazari'a and al-Aaniyyin became ISIL leaders. Many Albu Hashma, alienated by what they viewed as the anti-Sunni-Arab policies of the Maliki government, also swore allegiance to ISIL. 25 Many residents fled, heading to other parts of Salah al-Din, Diyala, Kirkuk or Suleymaniya. Some left Iraq entirely and sought refuge in Turkey. Others subsequently attempted to flee, telling ISIL members manning checkpoints that they were going to hospital or to visit relatives. They promised to return within a set period, or else their 21 Interview with former Yathrib resident, 21 January 2020. 22 Demographic details were compiled via interviews with current and former residents of Yathrib and Balad. 23  houses would be seized. But ISIL subsequently forbade residents from leaving altogether, executed those who tried to flee in the town centre, and blew up the homes of alleged 'apostates'. 26 The tribes resisting ISIL appealed to Iraqi forces and PMFs to assist in the campaign to liberate the district. In late 2014, the Iraqi air force began extensive strikes on Yathrib, targeting ISIL assets and forcing the vast majority of the population to flee. In late December, Iraqi security forces announced a campaign to liberate Yathrib, involving 17 Division of the Iraqi Army, Federal Police, and PMFs -specifically Asa'ib Ahl al-Haq (AAH), Kata'ib Hizbullah, Badr, the Imam Abbas Combat Division, and the Hussayniya Threshold Brigades (which are all associated with Iranian backing). The Iraqi army, air force and coalition forces provided air support. ISIL militants used around 13 thousand residents as human shields, but were nonetheless defeated and the Iraqi flag was raised above the district building on 27/12/2014. The security file was handed over to Brigade 41 pertaining to AAH. 27 Land and property in Yathrib were extensively damaged by ISIL and the 'liberation' campaign. An official source in the area indicated that over 70% of the houses were blown up and most agricultural plains were damaged by fires and disruptions to water supplies. Thousands of cars were abandoned in fields by fleeing families, and many destroyed in subsequent fighting. Following ISIL's defeat, mass graves containing bodies of residents executed by militants were discovered in the orchards. 28

Peace-brokers and the long road to reconciliation
Mass displacement across Iraq during ISIL's campaign presented the federal and provincial governments with staggering humanitarian responsibilities. The authorities, while intent on dispensing justice to ISIL members, were also keen for the displaced to return home as soon as possible. In Yathrib, however, where over 90% of the pre-2014 population had fled, Shi'a tribal shaikhs, as well as Sunni-Arab tribal leaders in Dujail and Balad whose tribesmen had been killed resisting ISIL, refused to allow thousands of families to return home, claiming that they or their relatives had collaborated with ISIL. Some families did attempt to return and were warned that they would be subject to tribal revenge attacks if they did not leave.
Over the following five years, efforts by USIP, Sanad, the NIF and the UNDP to return the displaced to their homes were intermingled with other initiatives sponsored by a combination of municipal, provincial and national figures. Amongst the high-profile figures who attempted to broker or impose solutions were Salim al-Jabouri, former Speaker of the Council of Representatives; Qassim Daraji, Baghdad member of parliament for al-Sadiqun, the political wing of AAH; Ra'id al-Jabouri, and subsequently his cousin Ahmed al-Jabouri, the governors of Salah al-Din along with other members of the provincial council; Abdullatif al-Humaym, Chairman of the Sunni Endowments Fund; and Abu Ali al-Saadi, the central and northern regional military commander of AAH. Other security commanders also chipped in, as did tribal leaders in the district who had previously worked with the prime minister's office on the reconciliation file since the days of Al-Qae'da. 29 The overall reconciliation strategy was to broker agreements at the level of the tribe or sub-tribe between resident and displaced tribesmen (known as the 'resisting' and 'returning' tribes). In practice, these tribes were not unified: indeed, the Bani Tamim were split between resisting and returning tribes. Initial attempts to achieve all-encompassing agreements faltered, not least because some of the would-be 'brokers' were not viewed as neutral by all parties. In the summer of 2015, some tribesmen agreed on the behind-the-scenes intervention of Sanad and its Network of Iraqi Facilitators (NIF). Sanad had achieved notable success in easing tensions between tribes after the Camp Speicher massacre in June 2014, when ISIL militants murdered an estimated 1700 army recruits. 30 Sanad's approach involved analysing the underlying causes of conflict, identifying enablers and spoilers, and engaging various factions (including displaced tribesmen in Suleymaniya and elsewhere) to identify their claims and demands before bringing the parties together. 31 A Sanad facilitator emphasised that in all of their work, they sought to bring in as many perspectives as possible, including those of women, noting that women often highlighted practical aspects necessary for coexistince (for instance, relating to ration cards), which men might not.
Sanad's initial efforts were cut short when donor funding ran out after six months, at which point several of the more influential PMF in the province took the reins, using their own intermediaries to exhort both resisting and returning tribes to compromise on the terms of return, but without conclusive results. When Sanad's activities resumed with new funding some months later, they were able to establish several preliminary agreements-in-principle with disputing parties, including the agreement that families of ISIL victims, rather than receiving 'blood money' from the returning tribes, might instead be paid by Iraqi government's official compensation fund.
These tentative agreements provided the framework for a large conference in Baghdad in May 2016, hosted by the Chairman of the Sunni Endowments Council, the provincial governor and provincial council chairman. The conference was attended by representatives of various (though not all) parties to the dispute, who signed a 'Community Peace Document', 32 committing themselves, and, by extension, their tribes, to a range of clauses. The onus of responsibility naturally fell on the 'returning' tribes, who agreed to publicly denounce tribesmen who had joined or cooperated with ISIL; to never allow ISIL members, their fathers, sons or brothers to return; to 'reveal the fate of tribesmen whom Daesh [ISIL] had kidnapped', and likewise to reveal where the bodies of martyrs had been buried and to treat anyone concealing this information as a participant in the crime; to refrain from initiating malicious lawsuits related to the bombing of houses and theft of property; and to comply with the authorities' security checks enabling their return. The resisting tribes agreed not to close the main road in the area or the roads leading to it; to stop operating illegal checkpoints and preventing tribesemen who had complied with legal procedures from returning. All sides agreed to form a tribal council in Yathrib to preserve 'good neighbourliness' and resolve tribal disputes. Salah al-Din provincial government agreed to ask the Supreme Judicial Council to expedite relevant lawsuits and to establish a popular mobilisation battalion composed of residents, who were to preserve security alongside the army and local police. 33 Additional financial, administrative and security conditions were added during the conference, including the requirement that victims' families be paid the equivalent of blood money in form of 500,000 dinars on family ration cards. The matter of compensation was not, however, settled, and as a result, the agreement was not fully implemented. 34 Both the Salah al-Din governor and Sunni Endowments chairman offered to contribute large sums as compensation. However, several resisting tribes insisted that returning tribes must pay blood money (even where the latter denied cooperating with ISIL), which they were not always able or willing to do.
Some families were permitted to return, after paying diya, or offering assurances that the governorate would pay compensation on their behalf. The returnees underwent arduous security checks by local authorities in tents outside Yathrib before being finally admitted, and those whose houses had been destroyed had to live in tents provided by the UN and provincial government. 35 But for the majority, negotiations stumbled on over the subsequent months and years. Individual agreements were gradually concluded between different tribal subgroups concerning payment of compensation, with various actors playing more-or-less prominent roles in mediation.
From early 2017, the UNDP, under the rubric of its Integrated Reconciliation Programme, assigned Sanad with forming a 30-member local peace committee for Dujail and Balad comprising tribal leaders, activists, youth representatives and local government officials including the district mayor. Some committee members were already involved in mediating between tribes and dealt primarily with promoting reconciliation between clans -not only from Yathrib but also from other nearby tribes affected by ISIL. Sanad trained committee members in conflict resolution and negotiating. In February 2018, Sanad and the UNDP hosted a press conference in which around 40 community leaders around Balad and Dujail announced their support for principles of peaceful co-existence and stabilisation in Yathrib, which included supporting security agencies and the rule of law, supporting victims and preventing retaliation and collective punishments. 36 By early 2019, an estimated 45,000 of the displaced had returned. In March, in another tribal conference hosted by the governor (who by then was Ammar al-Jabar al-Jabouri), the Albufadus finally agreed to allow a further 30,000 of the Albu Hashma to return, pending payment of diya. 37  authors that thanks to the committee's efforts, around 80% of the displaced had eventually been able to return. 38

Assessing inclusiveness amidst tribal and transactional politics
The return of the majority of Yathrib's residents was an achievement: elsewhere in Iraq, including parts of Ninawa, displacement figures remain far higher. But at what cost did it occur? While Sanad and the UNDP sought to promote the type of inclusiveness and participatory citizenship that international actors commonly associate with 'the local', a closer examination of the dynamics of the peace process indicates that, to the extent that it was a 'local' process, it was not particularly equitable, and was strongly influenced by the logic of tribal and transactional politics which permeate Iraq as a whole, and which are far from inclusive.
Yathrib's reconciliation process was 'local' in that it pertained to a particular subnational administrative territory, but the actors involved and issues at stake intersected different spheres of district/municipal/provincial/national and even international government and governance. Our interviewees in Balad municipality indicated that the terms of reconciliation were affected by the way that pre-existing tensions within the district, particularly within the Bani Tamim interacted with national dynamics. Until ISIL's takeover, the Albu Hashma had dominated local political affairs, including positions on the local council. With the rise of the Shi'a-dominated federal government post-2003, some Shi'a members of the much smaller Albu Hassan sought greater influence over local affairs, and after ISIL's invasion, sought PMF assistance to liberate Yathrib. After ISIL's defeat, AAH in particular gained a decisive role in local politics. The former council chairman (an Albu Hashma) was expelled and accused of supporting terrorism, and the governor's office in Yathrib appointed one of the Albu Hassan as the (first Shi'a) chair of the council, although the Albu Hashma retained the directorship of the district council. 39 At the same time, interviewees in Balad indicated that part of the reason why some tribes were so vehemently opposed to the return of others was because in their absence, they had profited from the use of their lands.
The widespread pursuit of tribal dispute resolution in ISIL's aftermath is unsurprising given that many ISIL members were recruited through patrilineal kinship networks. But contrary to the idea that 'tribal' solutions are fundamentally parochial or 'local', they have been incorporated into national peace strategies. This is not merely a product of the post-2003 settlement. During the early years of Saddam's rule, the Baath regime suppressed the prerogative of tribal leaders to autonomously settle disputes along customary lines. 40 However, over time Saddam reactivated aspects of tribalism as a strategy of governance, sometimes giving precedence to tribal law above the civil legal system. 41 Following regime change, the 2005 Iraqi Constitution enshrined 'the advancement of the Iraqi clans and tribes' and 'their noble human values in a way that contributes to the development of society'. But it also noted that: 'The State shall prohibit tribal traditions that are in contradiction with human rights'. 42 The Iraqi government has further upheld tribal identity by creating tribal affairs 38 Interview with UNDP regional coordinator 8 January 2020. 39 Interviews with Balad and former Yathrib residents, November and December 2019. committees to operate in the Prime Minister's Office, the Ministry of Interior and in parliament, and the Ministry of Justice has established formal communications with tribal shaikhs across the country. The National Office for Reconciliation, created under former Iraqi PM Nouri al-Maliki's rule, frequently intervened in tribal affairs. 43 In Yathrib, district, provincial, national and occasionally international actors played a part in brokering tribal reconciliations, showing varying levels of commitment to ensuring that Iraqi civil law, and individuals' rights, were preserved. Certain customary practices of collective punishment (such as jalwa, whereby a wrongdoer's relatives up to the 4th or 5th degree of patrilineal descent are exiled from their hometown), are clearly difficult to reconcile with principles of participatory citizenship. In 2016, Salah al-Din provincial council adapted jalwa to civil law, passing a resolution stating that 'anyone proven to have been complicit or affiliated with ISIL has no right to return to the governorate', and ordering the expulsion of immediate relatives of ISIL-members from Salah al-Din for 10 years to life, noting that they were only allowed to return if deemed 'safe." 44 The decree also established a committee to seize ISIL-affiliates' property and suspend their ration cards, and those of their families. Families that killed ISIL-affiliated relatives or handed them to the authorities were exempted.
Sanad sought to soften some of the more exclusionary aspects common to tribal customary practices, for instance by including women in peace negotiations behind the scenes, but a member of the NIF acknowledged that in reality women had no place at the table in tribal musalahat. 45 The UNDP-sponsored local peace committee for Balad and Dujail also appointed a few women, but according to a male member, they were unable to play a particularly active role in the reconciliation file. Similarly, Sanad encouraged resisting tribes to accept government compensation funds in place of diya from returning tribes. This initiative was only partially successful; however, intermediaries at several levels exploited the collection of diya for political ends, and the distribution of compensation funds was subject to multiple claims of corruption.
Members and aspiring members of Salah al-Din council clearly used the reconciliation file to garner political support. Immediately after ISIL's defeat, the governor of Salah al-Din, Ahmed al-Jabouri (Abu Mazen) left the post to his cousin to become Minister of Provincial Affairs in Haider al-Abadi's government. The latter position was abolished in 2016 and, to garner support from Salah al-Din tribes and the Prime Minister's Office to be reinstated as governor, Abu Mazen assumed a prominent role in mediation between tribes. On reinstatement, he spent large amounts of provincial funding on compensating victims of ISIL. Abu Mazen, who had already been investigated for corruption, was subsequently removed as governor in July 2017 and sentenced to prison for misusing authority and federal funds and appropriating lands for personal use. Undeterred, on his release, he successfully campaigned in the 2018 parliamentary elections, winning a seat after his electoral bloc won the highest number of votes in Salah al-Din, and his staying power in the province is undoubtedly partly attributable to his accommodation of Iran-sponsored PMFs in Salah al-Din. 46 Meanwhile, AAH, which had been vying for popular support with Badr and Kata'ib Jund al-Imam Abbas, used its pivotal (if hardly 'neutral') role in the 'mediation' process to consolidate territorial control over the district. 47 Conversations with actors on different sides of the dispute suggest that even those who disliked and/or feared AAH believed that, as the most consistent force on the ground, it was best-placed to enforce agreements, largely thanks to the authority exercised by Abu Ali al-Saa'di, who forged effective working relationships with several tribal shaikhs, and, for instance, identified a Zubayd tribal shaikh in Dujail as AAH's point of contact for 'tribal affairs'. 48 AAH collected funds from families who wished to return, redistributing them to victims' families, and pressing the latter to accept the returnees. Families could not return without obtaining AAH's approval. This practice appears to have paid dividends, as AAH's political candidate, who stood in the 2018 elections on the PMF list (Fatah -which came second overall in the province), won the Balad seat based on electoral support from over 5,000 Sunni returnees to Yathrib.
Since consolidating its territorial control over Yathrib, AAH has reportedly retained an effective stranglehold over social and political life within the area in return for preserving stability and fighting resurgent ISIL cells (of which there have been a growing number since mid-2019). Several interviewees indicated that reconstruction contractors, humanitarian agencies, and other commercial operatives require approval from AAH to operate in the town, and some residents indicated that the group made money from taking a cut of royalties and profits.
The actual monetary cost of return remains obscure. Interlocuters in Salah al-Din told us that in 2016 the Sunni Endowments chairman offered to pay 2 billion dinars (approximately 1.8 million USD) as compensation to ISIL victims' families; that the governor had used government funding allocated to Salah al-Din for reconstruction in areas damaged by Daesh to pay 300 million dinars (over 250,000 USD) as part of a deal in which displaced local clans agreed to pay 100 million dinars (around 84,000 USD) blood money; and that AAH collected 500,000 dinars (around 420 USD) from each returning family, which it paid to victims' families. Another interlocuter indicated that AAH had taken a cut from the sums it collected. And news reports on the March 2019 reconciliation agreement indicated that only a third of the 700 million dinars (500,000 USD) in blood money 'owed' to the families of victims had been raised.

Conclusion: revisiting the international in local peacebuilding
The complexities of achieving reconciliation in Yathrib raise questions about the role that external agencies can or should play in subnational peace processes in Iraq, and, by extension, in other post-conflict states.
In the first instance, it bears remembering that relationships between international actors and subnational 'local' actors are heavily influenced by relationships between those same international actors and the federal government. The practice of offering capacitybuilding, training and funding for community leaders and organisations at the subnational 47 Interview with NIF adviser, 5 February 2020. 46 In July 2019, the US government sanctioned Abu Mazen for corruption, claiming that Abu Mazen 'has been known to protect his personal interests by accommodating Iran-backed proxies that operate outside of state control.' ('Treasury Sanctions Persons Associated with Serious Human Rights Abuse and Corrupt Actors in Iraq', US Treasury, July 2019. https://home.treasury.gov/news/press-releases/sm735 (accessed May 20, 2020). 48 Interview with a member of the local peace committee for Balad and Dujail, 22 January 2020.
level to promote civic participation in post-conflict and 'fragile' states is now well recognised. Nonetheless, as a rule, external actors only support subnational political capabilities that challenge the authority of federal government if their relationship with the centre has conclusively broken down. The UN 2015 Peacebuilding Review concedes that the UN is not in the best position to lead the process for sustaining peace but can only assist and facilitate national stakeholders in doing so. 49 In Iraq's case, the US and the 'international community' more broadly (understood as Western liberal democracies, their allies and associated multilateral agencies) created the foundations of the federal Iraqi government after regime change, and as a result these actors have been deeply vested in its success. Notwithstanding increasing discomfort over the Iraqi government's trajectory, there is little prospect of Western international actors forsaking national political processes. So, even where external actors have intervened into subnational peace processes, they have done so with the support of the National Office for Reconciliation and its successor office. Hence, there is little opportunity to bypass the centre in order to reach and empower 'ordinary citizens', except to the extent that key national partners can cut through bureaucracy. Second, contrary to representations of 'localism' as inclusive, emancipatory, and romantic, 50 subnational politics is often equally if not more exclusionary and transactional than national politics. Where local governance is dominated by transactional politics, local peace processes are similarly likely to be so. That does not mean there is no will amongst 'local' citizens to produce a more equitable settlement, it simply means that overriding self-interested actors is extremely difficult. This is something that internationally funded projects designed to promote civic participation have to contend with often they are ill-equipped to compete with other 'local' actors who rely on force and on funding sources which are unconstrained by accountability mechanisms. 51 The micropolitical economy of peacebuilding is a key consideration. Several interviewees from Balad district contrasted the consistent presence of AAH in Yathrib with the more patchy role played by Sanad, which was forced to cease activities 6 months after it started when funding ran out, only to re-engage 18 months later. As a result, interlocutors questioned how they could consider Sanad or the UNDP as reliable local partners.
Third, in the Iraqi context at least, Sanad and the UNDP were faced with multiple political and security actors already involved in seeking solutions to displacement. In what is often a crowded subnational political marketplace, the process of identifying relevant stakeholders and community representatives is often complex and politicised. In fact, conversations with individuals from Balad suggest that many people who had not been directly engaged in the negotiations were unaware of the role that Sanad or the UNDP played in the reconciliation file. Their work was complicated by difficulties in establishing a common agenda not only amongst the disputing tribes but also amongst the various 'mediating' actors, or a solid line of communication with the Iraqi government's Reconstruction Fund administrators.
Fourth, the non-inclusive and often unequitable dimensions of subnational peacebuilding processes in post-conflict settings such as Yathrib present ethical dilemmas to international actors regarding intervention. On the one hand, Sanad and the UNDP had to decide how to interact with the military factions including the Iran-backed PMFs on the ground. A member of the NIF acknowledged that Sanad was obliged to accept the reality that the PMF were the primary guarantors of security in the area, and that their buy-in was essential to achieve any kind of settlement, so Sanad took advantage of the competition between them, and in particular AAH, Badr and Kata'ib Jund al-Imam Abbas) to secure reconciliation settlements and thereby accrue political capital to complement its own efforts to facilitate peace-talks. Similarly, when the UNDP supported the establishment of local peace committees, in keeping with UN operational guidelines, no PMFs or military personnel were enlisted, but since it was impossible to exclude the PMFs from mediation, Sanad coordinated indirectly with them via the National Office of Reconciliation and the National Security Advisory. 52 On the other hand, USIP and Sanad decided to engage with tribal musalahat, something which most external development agencies have avoided, on the basis that they undermine civil law. Conversely, USIP has maintained that in many cases, obtaining buy-in from tribal leaders in negotiations has been simply the only way of restoring stability. Instead of trying to ignore tribal figures in the implementation of peace strategies, the objective should be to encourage them to support formal process of law and to forego the implementation of exclusionary elements of tribal law. 53 Sanad and the USIP sought to establish bridges between national and local peace processes because, as the USIP programme manager argued, there need to be national level solutions that can consolidate local gains. 54 Overall, our findings suggest that in order to assert themselves as relevant and effective partners, externally funded peacebuilding actors need to be able to recognise and manipulate the existing logic or 'rules of the game' that inform how business -including the business of peacebuilding -is done. This is particularly the case with respect to subnational peace processes, where they find themselves in the thick of multiple overlapping political fields. In the Iraqi case, the logic is tribal and transactional, and in Yathrib at least, Sanad did accept these parameters, even when seeking to modify them. Naturally, the kind of reconciliation agreements that Sanad helped to secure are not the end of the story: after returning home, residents face the much longer process of rebuilding their place in the community. The UNDP's local peace committees were designed with this longer-term goal in mind, but their sustainability is questionable, and one committee member suggested to us that the committee's success was more attributable to the fact that it contained members already active in brokering reconciliations, than to the fact that they had been organised into a committee. 55 None of the caveats discussed necessarily negate the value of external actors promoting peace at the subnational level. The efforts by USIP, Sanad and the UNDP to facilitate dialogue between disputing parties in Yathrib over the course of several years do appear to have been integral to achieving a number of agreements for return -even if few Iraqis are aware of them. In their absence, the process would likely have been more exclusionary 52 Interview with member of the NIF, 7 January 2020. 53 Interview with USIP programme manager, 8 December 2019. 54 Osama Gharizi & Haidar al-Ibrahimi, 'Baghdad must seize the chance to work with Iraq's tribes', USIP, January 17, 2018.
https://warontherocks.com/2018/01/baghdad-must-seize-chance-work-iraqs-tribes/ (accessed February 22, 2020), and authors' interview with USIP programme manager, 8 December 2019. 55 see note 48 above. and corrupt. These limited gains may, indeed, be considered a positive outcome in view of the overriding Iraqi political status quo. But in the final assessment, there is nothing romantic about local peacebuilding in Iraq.

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

Funding
This article is an output of the Conflict Research Programme, led by the London School of Economics and Political Science, and funded by UK aid from the UK government; however the views expressed do not necessarily reflect the UK government's official policies.

Notes on contributors
Jessica Watkins is a research fellow at the Institute for Middle East Studies at GIGA. From 2017-2021 she worked as a post-doctoral researcher at the Middle East Centre, London School of Economics, on the Conflict Research Programme, focusing on regional and domestic drivers of conflict in Iraq. Her PhD from the War Studies Department, King's College London, was on civil policing and dispute resolution in Jordan.
Mustafa Hasan is a freelance journalist an independent researcher in the field of human rights, political sociology and Islamic political movements based in Baghdad. He worked previously as the deputy editor-in-chief at the Institute for War and Peace Reporting in Iraq.

Ethics approval
This research involved human participants and was approved by the LSE Research Ethics Board in December 2019. Reference number: 1041.