in , , , , , ,

Three Lessons for the Intra-Afghan Talks, Courtesy of Iraqi Power-Sharing

Written by Colm Trant

This essay argues that Iraq’s destabilising power-sharing experience can be avoided in consociational Afghan power-sharing if Western actors apply a model defined by three interrelated lessons from Iraq: the need for accommodation, comprehensiveness, and commitment in power-sharing. ‘Accommodation’ instructs Western actors to identify key political interests within a state and ensure power-sharing includes these interests. ‘Comprehensiveness’ instructs Western actors to facilitate power-sharing that spans multiple dimensions—not wholly investing in one—which better equips power-sharing to accommodate diverse interests. ‘Commitment’ instructs Western actors to uphold the stability of this power-sharing by policing participants’ compliance. Each lesson comprising the model identifies an instructive failure of Western actors in Iraq that remains avoidable in Afghanistan: Western actors failed to accommodate key Iraqi interests, like Sunni Arabs and pro-unitary Shia Arabs; this informed their failure to ensure comprehensiveness, allowing autonomy alone to dominate and destabilise Iraqi power-sharing; and they committed foremost to military victory, not to policing compliance with details of Iraqi power-sharing. This essay contributes to existing debates around Iraqi power-sharing by demonstrating how three lessons from Western failures in Iraq can be applied to develop stable Afghan consociationalism.

Structurally, the essay begins by briefly clarifying its consociational approach and corresponding academic debate, then proceeds in two sections: First, it assesses Iraq’s power-sharing instability to derive the model’s three interrelated lessons for Afghanistan; second, these lessons are then applied to Afghanistan and ongoing Doha negotiations, finding that Western actors can produce stable consociationalism in Afghanistan by learning from these failures in Iraq’s power-sharing experience.

Outlining the essay’s theoretical approach clarifies the model’s definition of power-sharing dimensions and introduces central academic debates concerning Iraq’s power-sharing experience. Consociationalism typifies four dimensions of power-sharing: Autonomy, proportionality, grand coalitions, and mutual vetoes. That is, ‘autonomy’ of territories or cultures, ‘proportionality’ in representation and state resources, ‘grand coalitions’ including significant communal segments within government, and ‘mutual vetoes’ allowing these segments to obstruct sensitive policies.[1] These aim to regulate conflict in plural societies, like Iraq and Afghanistan, by recognising and managing deep divisions. Yet, the instability characterising Iraq’s power-sharing experience provokes debate between consociationalists and ‘integrationists’ concerning whether this instability is caused by the absence or presence respectively of consociational dimensions. Integrationists here argue consociationalism is ineffective for conflict regulation, preferring voluntary minimum-winning coalitions, majoritarian systems, and autonomy determined by administrative practicalities, not cultural divisions—all aiming towards constructing civic nationalism rather than empowering ascriptive identities.[2]

While both sides recognise Iraq’s instability, debate seeks to answer whether consociationalism was the cause: For example, Younis argues Iraq’s consociationalism was comprehensive—providing power-sharing over multiple dimensions—and that its instability demonstrates how consociationalism is ineffective for conflict regulation, which would favour an integrationist model for Afghanistan.[3] Contrarily, Ltaif argues Iraq’s instability emerged from incomplete consociationalism—it was insufficiently comprehensive and did not accommodate key interests like Sunni Arabs.[4] Yet other academics and practitioners pin Iraq’s instability neither to its accommodation nor comprehensiveness, but from failures to police compliance with power-sharing, with examples from Kirkuk and the Erbil Agreement—both raising Western actors’ commitment to power-sharing.[5] However, none of these arguments alone offer a complete model guiding Afghanistan’s uncertain power-sharing future; rather, each broaches interrelated failures of Western-led conflict regulation in Iraq which must now be synthesised to demonstrate how similar instability can be avoided in Afghan consociational power-sharing. The first of these lessons concerns accommodation: Consociational power-sharing requires accommodating key interests, but Iraq’s unstable power-sharing did not.

Iraq’s lesson on accommodation in power-sharing agreements begins with Western planning failures for a post-invasion political settlement: the US failed to prearrange either an inclusive political settlement or the institutions in which an inclusive settlement might be negotiated. On 16 December 2003, Secretary of Defense Donald Rumsfeld wrote to Vice President Cheney, responding to this very criticism: “With opponents saying we had no ‘plan’, it is important that we keep referring to our ‘plan’.”[6] Yet Rumsfeld’s ‘plan’ in December 2003 merely asserted ownership of the US’s ad hoc scramble to assemble a post-invasion Iraq; he did not address the planning that occurred in Washington since 2001, because these plans were blundered.[7] Rumsfeld participated in planning for the Iraq War since September 2001, and by early 2002 he was lobbying the Bush administration for a minimalist US occupation of Iraq reflecting ostensibly successful Afghan regime change.[8] Rumsfeld lost to a Department of State (DoS) alternative involving US-led organisation of a prearranged Iraqi provisional civilian government, institutionalising democracy post-haste.[9] The transition’s immediacy was paramount: No plans were made for accommodating diverse Iraqi interests in a political settlement, having presumed their interests would be accommodated by the immediately instated Iraqi provisional government, not US agencies or military officers.[10] However, through 2002 and early 2003, attention drifted to combat planning while peacebuilding plans languished, and when Saddam’s regime collapsed in April 2003, these DoS plans were not ready: no US agency was equipped to support a transition until the Coalition Provisional Authority’s (CPA) establishment in May 2003, which then spent five further months scrambling to organise Iraq’s provisional government. Not only were details for an accommodative political settlement absent from US planning, but their planning to develop institutions conducive to accommodation was improperly executed, so that the institutions which might have communicated with Iraqi interests to develop nationally acceptable, inclusive power-sharing did not exist at the crucial moment when Saddam’s regime collapsed.

This vulnerable intermission of governance was then joined by a second US failure of accommodation, further condemning Iraqi power-sharing to instability: The Bush administration’s failure to define limits for de-Baathification and the subsequent underrepresentation of antagonised Sunni Arabs in power-sharing negotiations. Bush resolved to purge Baathist leadership, yet never prescribed a limit.[11] The CPA chose to purge deeply, dismissing 30,000 bureaucrats comprising four layers of administration, destroying Baghdad’s governance over Iraq.[12] Subsequently, Iraq’s Baathist military was disbanded, leaving 400,000 predominantly Sunni-Arab soldiers unemployed and disempowered.[13] This vacuum of governance and security was answered by insurgencies, reflecting how Western-led conflict regulation made accommodative power-sharing increasingly unlikely: Many Sunni Arabs, antagonised by de-Baathification, perceived the new state as exclusionary and illegitimate, expressed violently through insurgency and then politically through boycotting January 2005’s elections. Sunni Arab turnout reached lows of 2% in Sunni-dominated Anbar Province and below 20% in other majority-Sunni Arab provinces.[14]

Provoked by a failure to accommodate Iraq’s Sunni Arabs, this boycott is consistently identified historiographically as a watershed moment for Iraq’s unstable power-sharing experience because the elections determined the balance of voting members for the 2005 constitutional drafting.[15] On Western advisors’ recommendations, the election used one nation-wide constituency and party-list, which was exclusionary because the absence of constituencies meant low Sunni-Arab turnout caused overrepresentation among high-turnout segments—Kurds and Shia Arabs.[16] Consequently, broadly pro-unitary Sunni Arabs were insufficiently represented against Kurdish and Shia Arab pro-federalism interests. Rather than ensuring Iraqi power-sharing was negotiated by an inclusive set of elites accurately representing each significant segment, as consociationalism advises, the US was instead preoccupied with insurgency and rushed the CPA, and then the US embassy, towards whatever settlement the January elections might produce, disinterested in whether power-sharing accommodated Iraq’s pro-unitary interests.[17]

The error of integrationist critiques here is to misidentify this failure of accommodation as functional consociationalism, as Younis does when she argues de-Baathification was a direct product of consociationalist primordialism—that  consociationalism requires Sunni Arabs’ exclusion because their grievances are immutable and cannot be reconciled.[18] Yet the CPA, responsible for de-Baathification, was not consociationalist—expressing integrationist views on federalism—and its successors at the US embassy were not primordialists, instead striving belatedly and unsuccessfully to mend Sunni representation after the boycott.[19] Likewise, consociationalist theorists emphasise inclusivity and—as of the 1990s—are not primordialists.[20] Rather, it was Western actors’ failure of accommodation, not healthy consociationalism, that would spawn Iraq’s unstable power-sharing constitution: Western actors failed to include significant pro-unitary interests in the power-sharing negotiations, contrary to consociationalism, ultimately producing excessively pro-federalism, non-comprehensive power-sharing responsible for Iraq’s unstable experience.

This pro-federalism, non-comprehensive power-sharing was finalised in Iraq’s 2005 Constitution, which developed the consociational dimension of autonomy—its federalism—at the expense of comprehensive consociationalism, with destabilising consequences. Kurdish parties KDP and PUK together pushed for maximalist federalism—their preferred alternative to then-unrealistic independence.[21] Shia Arab parties, despite running on a united list, were more divided on constitutional questions: SCIRI favoured federalism, hoping to rule an autonomous Shia region, while Dawa and the Sadrists favoured unitary rule from Baghdad, acknowledging Kurdish autonomy but opposing further federalism.[22] However, Kurdish and pro-federal Shia interests proved too powerful given their overrepresentation from January’s elections—pro-unitary Shia Arabs alone could not curb the pressure for extensive autonomy.[23] The resulting 2005 Constitution established a Kurdish federal region (KRG), ceding all governance competencies except nine policy areas exclusive to Baghdad, mostly inconsequential besides foreign policy.[24] Meeting SCIRI’s ambition, a pathway was provided to form new federal regions, although Shia Arab voters in subsequent elections instead favoured unitary, pro-centralisation politics, befitting their overall majority.[25] Western actors failed to perceive the long-term risks of permitting a power-sharing agreement that was not comprehensive but instead invested power-sharing almost entirely in one dimension, autonomy, favouring Kurds and failing to accommodate pro-unitary interests. This non-comprehensiveness was only problematic due to the preceding failure of accommodation: the US was enabling a one-dimensional power-sharing agreement unduly skewed towards overrepresented Kurdish and SCIRI interests. Without external pressure to broaden power-sharing, only the fraction of pro-unitary Shia Arabs remained to demand comprehensiveness, but their gains were insufficient to achieve stability through the inclusion of diverse interests over multiple consociational dimensions.

These remaining dimensions—proportionality, mutual vetoes, and grand coalition—clarify how Western actors’ failure to ensure comprehensiveness contributed to Iraq’s unstable power-sharing experience. Proportionality within the constitution was only partly substantiated as a side-payment for autonomy: The PUK, KDP, and SCIRI offered concessions on oil ownership and revenues—forms of proportional economic power-sharing—but Kurdish negotiators were eager to exchange some oil for autonomy and federal subsidies.[26] Likewise, the only vetoes serviced Kurdish autonomy: vetoes between the federal layers, where the KRG held a veto within its territory over almost all federal legislation.[27] This is a form of self-rule that is definitionally not consociational because it is not mutual.[28] Nor was there any grand coalition in the constitution to placate unaccommodated Sunni Arabs freshly dethroned from historical political dominance; there was only a temporary two-thirds legislative majority for selecting transitional presidents, who appointed the Prime Minister-Designate responsible for government formation, but this lapsed after the transition to a simple-majority run-off.[29] This non-comprehensiveness in the 2005 Constitution worsened Iraq’s unstable experience by producing a political system designed to service a pro-federal clique accommodating Kurdish and some Shia Arab interests at the expense of pro-unitary interests in a strong Baghdad government.

Dixon, a critic of consociationalism, was correct for different purposes when he noted that these provisions do not treat all four dimensions of consociational power-sharing, and elsewhere the same observation of incomplete consociationalism has induced the labels “consociationalism ‘light’” and “informal consociationalism”.[30] The instability of Iraqi power-sharing therefore cannot be attributed to the robust presence of consociational power-sharing; rather, instability originated with Western actors’ failure to accommodate Iraqi interests, which then precluded the comprehensiveness of power-sharing in these one-sided negotiations. Unsurprisingly, pro-unitary ambitions survived the constitution’s ratification: Sunni Arabs participated in subsequent elections and pro-unitary Shia Arab parties surged in popularity against SCIRI, suddenly leaving Iraq with accurate representation of those unaccommodated interests opposed to extensive autonomy, pitting this more-representative government against a non-accommodative, non-comprehensive power-sharing constitution. The consequence was, and continues to be, a constant destabilising tension between Baghdad, Erbil, and their national constitution.

Yet, this destabilising relationship between Baghdad and Erbil is not wholly explained by the lack of accommodation and comprehensiveness alone: it further requires incorporating the model’s third lesson—how the lack of commitment from Western actors to police compliance with power-sharing ensured no settlement was ever reached for these destabilising disputes. The US was committed to military victory against insurgents, but this commitment did not extend to policing power-sharing provisions in an Iraqi state they wanted to present internationally as independent and competent—metrics for successful liberal peacebuilding.[31] For Washington, Iraqi power-sharing disputes were for sovereign Iraq to resolve, the destabilising consequences of which are well-illustrated by Kirkuk and the Erbil Agreement. First, Kirkuk is a disputed territory of Iraq, historically Kurdish but settled with Arabs during Saddam’s regime alongside ethnic cleansing against Kurds, who presently form a slim majority.[32] As a sensitive, symbolic, and oil-rich issue, a three-stage process was agreed in transitional legislation and the 2005 Constitution to reach a power-sharing settlement for Kirkuk before 2008.[33] This process involved milestones like its referendum and deadline that Western actors could have used to police compliance through easily observed implementation.[34] However, the US did not commit to guaranteeing the process, instead focusing on counter-insurgency, interested in Kirkuk only where it risked upsetting timely US withdrawal.[35] Without external policing, Baghdad and Erbil’s approach to Kirkuk soured, resembling an interstate dispute rather than domestic politics. Baghdad, expecting a referendum to return a majority voting to join the KRG—and thus retreat under Kurdish autonomy—refused to comply with the agreed procedures.[36] Erbil, eager to reclaim historical territory, responded coercively, with Peshmerga occupying Kirkuk in 2014, to which Baghdad replied militarily in 2017, regaining control of Kirkuk.[37] Whether a power-sharing settlement would have involved Kirkuk joining the KRG or attaining special autonomy from either the KRG or Baghdad, such cooperative solutions required external policing that never materialised because Western actors were uncommitted to the details of Iraqi power-sharing. Although motivated both by liberal peacebuilding aversions to interference in Iraq’s domestic politics and their focus on security issues, Western actors’ lack of commitment ultimately destabilised Iraq’s power-sharing experience by entrenching hostilities between Baghdad and Erbil, who could not alone resolve this dispute cooperatively.

Similarly, within Baghdad, Western actors’ lack of commitment first enabled a destabilising crisis of government formation and subsequently failed to guarantee a fragile grand coalition agreement—the 2010 Erbil Agreement—aiming to stabilise Iraqi power-sharing, which required external support. Iraq’s prime-ministership was an atypically weak premiership due to the 2005 Constitution’s extensive autonomy.[38] Entering office in 2006, and dissatisfied with the premiership’s weakness, Nouri al-Maliki developed patronage networks within Iraqi intelligence and security forces, circumventing chains of command.[39] Western actors, concerned with Iraqi military effectiveness, fuelled this by helping al-Maliki centralise civilian control of Iraq’s military within one advisory position subject to his appointment.[40] Empowered atop this patronage network, al-Maliki signalled he would not accept defeat after 2010’s indecisive elections, with parties struggling to form a voluntary minimum-winning coalition. For integrationists like Younis, this is evidence that consociationalism is the problem because it recommends parliamentary systems which exacerbate competition for the premiership in coalition formation.[41] However, this misses the importance of comprehensiveness, because the consociational answer to this struggle was to join proportional electoral systems with the further dimension of grand coalitions—formulaically allocating positions rather than haggling around minimum-winning coalitions—which Iraqis sought to achieve through 2010’s ‘Erbil Agreement’.

Here al-Maliki retained the premiership in exchange for promising to formulaically distribute executive portfolios and committee chairs between parliamentary blocs: a form of consociational grand coalition that would improve the comprehensiveness of Iraqi power-sharing and curb destabilising crises of government formation.[42] Yet Western actors again failed to commit, here to policing the Erbil Agreement—an issue of Iraqi domestic politics. Toby Dodge presaged the agreement’s failure for similar reasons: that it lacked enforceability because Iraqis could not police their own executive and its corrupt military.[43] Both Kirkuk and the Erbil Agreement demonstrate the long-term destabilisation of preceding Western failures to produce accommodative, comprehensive power-sharing, worsened by this absence of commitment even to the flawed power-sharing that was produced. At each step, Western actors’ failures in accommodation, comprehensiveness, and commitment caused the Iraqi experience of power-sharing to become characterised by instability. And it is precisely these three lessons which therefore provide the model for Western-led consociational power-sharing in Afghanistan.

To avoid replicating Iraq’s instability, Western actors must first apply its lesson on accommodation: If Doha’s intra-Afghan negotiations fail to accommodate the breadth of Afghan interests, they risk a destabilising first step analogous to alienating and antagonising Iraq’s Sunni Arabs. Specifically, two internally diverse interests risk going unaccommodated in Doha’s negotiations: First, elites in Kabul who are not loyal to Afghanistan’s current government, and second, local elites constituting the decentralised Taliban insurgency. A risk of destabilising exclusion is already visible in Kabul amidst Doha’s negotiations: On 21 October 2020, a Pashtun Islamist elite in Kabul, the leader of Hezb-e Islami, publicly denounced Doha’s negotiations, explicitly citing their exclusion of many Afghan leaders, himself included.[44] It is unsurprising that Kabul’s elites are threatened by a power-sharing future—the distribution of offices and valuable patronage networks will be diluted by the Taliban’s inclusion. Islamist groups like Hezb-e Islami will face the most electoral and clientelist competition from Taliban inclusion in Afghan politics, and their leadership is signalling a warning familiar to post-invasion Iraq: a failure to accommodate their interests may push them to delegitimise or spoil a power-sharing settlement.[45]

Procedurally, it is unrealistic that Hezb-e Islami would be seated in Doha at this point, yet US mediators nonetheless must tackle these potential spoilers when they arise, working to ensure their interests are accommodated in Doha’s final agreement despite their absence. There is currently no indication of US efforts to accommodate these dissenting Kabul elites; US diplomacy treats Kabul’s Doha delegation as representing Kabul’s interests inclusively, but it truly represents a mixture of Ashraf Ghani and Abdullah Abdullah’s preferred appointees.[46] While no population is being near-wholly excluded like Sunni Arabs in Iraq’s 2005 negotiations, the exclusion remains important because it antagonises political elites associated with militias and histories of insurgency: Hezb-e Islami and their Pashtun Islamist base could easily destabilise Afghan power-sharing if they are not accommodated.[47] Neither Doha delegation will voluntarily accommodate Hezb-e Islami at their own expense, just as nobody expected Kurds or Shi’ites to charitably accommodate Sunni interests in Iraq; in both cases, the responsibility falls on external actors to ensure this foundational stability.

Equally, just as Doha’s Kabul delegation cannot be expected to fully accommodate Kabul’s elite interests, so too must Western actors not mistake Doha’s Taliban delegation as wholly representative of Taliban interests across Afghanistan: accommodating the Taliban must include the so-called ‘Taliban Caravan’, which describes the Taliban insurgency’s diversely motivated and highly decentralised leadership.[48] Taliban at Doha lack strong authoritarian control over this insurgency and cannot guarantee that all Taliban commanders will abide by Doha’s agreement if these commanders find further conflict preferable.[49] Three key interests in the ‘Taliban Caravan’ include: Pakistani-sponsored Taliban, Taliban Islamists, and ‘local’ Taliban.[50] The first, Pakistani-sponsored Taliban, can be regulated indirectly by accommodating Pakistan, whose security interest in preventing stronger Indian-Afghan relations is addressed by guaranteeing Taliban presence in government through grand coalitions and proportionality as elaborated below.[51] Conveniently, for Hezb-e Islami’s interests, grand coalition and proportionality also offer accommodation: proportional electoral systems will enable competitivity despite Taliban presence, and grand coalitions will preserve their presence in government despite diluting the Islamist vote.[52]

The second ‘Taliban Caravan’ interest, hard-line Islamists, requires delicate handling of the ‘Emirate question’ on reconciling ideas of an Islamic Republic and Islamic Emirate, which can be managed consociationally through mutual vetoes. The third, ‘local’ Taliban, describes communal powerbrokers who participate in the movement because it provides basic security and justice in otherwise self-reliant communities—these interests are accommodated through autonomy for local, traditional political institutions. Accommodation in Afghan power-sharing is therefore dependent upon its comprehensiveness: To avoid the instability of Iraq’s autonomy-centric power-sharing, Afghan power-sharing must treat each consociational dimension above rather than privileging one segment’s preferred dimension, like Kurdish autonomy. There is a risk of emulating Iraq’s experience by again failing to accommodate interests and motivating spoilers. This instability will have emerged from non-comprehensive power-sharing. The essential next question, then, is how to design comprehensive consociational power-sharing in Afghanistan given the failures of this process in Iraq.

Presently, Afghanistan’s constitution precludes a consociational grand coalition through its integrationist winner-takes-all presidentialism, which has already destabilised Afghan politics without Taliban participation.[53] Despite Afghans being highly fragmented across ethnic and sub-ethnic groups, none of whom form a majority like Iraq’s Shia Arabs, Western actors in the early 2000s hoped an indivisible presidency would produce strong centralisation and civic nationalism to overcome Afghanistan’s historically weak central government.[54] Instead, divisions between these segments, like the large Tajik minority and Pashtun plurality, were deepened by exclusionary contest over the presidency—Ashraf Ghani and Abdullah Abdullah twice brought the Afghan state to crisis by disputing the prized presidential election results. Their disputes were managed by US and UN mediation producing ad hoc power-sharing which persists in Abdullah’s role heading the intra-Afghan talks.[55] With majoritarian presidentialism already struggling to survive Afghanistan’s deep divisions, introducing the Taliban to compete for this exclusive office promises a return to violence and highlights the futility of applying integrationist, majoritarian models for Afghanistan—these models require a basis for big-tent, centrist ‘Afghan nationalism’ which does not exist.[56]

Instead, semi-presidentialism like Iraq’s would ease post-conflict reconciliation by portioning the presidency into a prime ministership and a strengthened assembly speaker, producing two newly empowered offices to accommodate further interests within leadership roles. The utility of this grand coalition component was so apparent in Iraq that it emerged informally, an ‘Iraqi National Pact’ of a Kurdish president, Shia premier, and Sunni speaker; yet the informal, unenforceable nature of Iraq’s pact is a weakness that needlessly risks instability in Afghanistan whenever one segment appears prepared to renege.[57] Whatever Doha’s teams might agree for an inclusive distribution of these offices, Afghanistan’s grand coalition must guarantee the Taliban at least one such leadership role to accommodate both the Quetta Taliban and their Pakistani allies. Even this first step towards consociationalism demonstrates why stability requires further comprehensiveness: How could Afghan semi-presidentialism with grand coalitions function without proportionality?

Specifically, proportionality overcomes two obstacles for Afghan grand coalitions: First, ensuring that accommodating the Taliban does not itself alienate Kabul’s worried Islamists like Hezb-e Islami; second, preventing threats of majoritarian domination by one segmental interest, which would ruin any intra-Afghan agreement but would require a level of disproportionality only conceivable in majoritarian systems. Afghanistan’s legislative elections use the single non-transferable vote (SNTV), a majoritarian system risking high disproportionality and wastage in constituencies where multiple parties appeal to the same voter bases.[58] The Taliban and Hezb-e Islami draw similarly on Pashtun, Islamist constituents, and SNTV risks subsuming Hezb-e Islami beneath Taliban electoral victories.[59] Proportional electoral systems prevent this by aiding the competitivity of niche parties,  awarding surplus ranked votes and formulaically allocating seats. By negotiating a proportional system at Doha, the Taliban and Pakistan will be assured that Pashtun and Islamist voters will be accurately represented in government, while Hezb-e Islami will be assured that Taliban competition does not eclipse them as it could under majoritarianism. Further, proportionality within Afghanistan’s multiple balance of power guarantees that no segment will exercise a majoritarian share in government even between parties—there is no majority like Iraq’s Shia Arabs which could threaten a majoritarian relapse of the kind Horowitz terms the ‘degradation problem’, where a majority segment rallies and legislates an end to power-sharing.[60]

Grand coalitions and proportionality in Afghanistan will preclude any risk of power-sharing degradation given its ethnic fragmentation. And just as proportionality facilitates grand coalitions, Iraq’s experience illustrates how proportionality without grand coalitions is dysfunctional in post-conflict settings, again advising comprehensive consociationalism: Iraq’s parties were fragmented by legislated proportionality yet needed to voluntarily form minimum-winning coalitions amidst raw civil-war divisions.[61] The Erbil Agreement sought to join proportionality with a grand coalition for this very reason during 2010’s prolonged coalition negotiations. To evade Iraq’s experience, Doha’s agreement must not only overhaul Afghanistan’s governmental system with a formal grand coalition accommodating interests between Kabul, Quetta, and Islamabad, but moreover must reinforce this dimension with proportionality, together warding against reversion to armed conflict from failures to consensually form minimum-winning coalitions.

Yet even with a grand coalition and proportionality to accommodate Pakistan, Quetta Taliban, and Kabul Islamists, further comprehensiveness through mutual vetoes is required for the predictable flashpoints within this grand coalition that further includes the many ‘independents’ like Ghani and parties like Jamiat-e Islami and Hezb-e Wahdat. Here, Iraq’s lesson is that vetoes must be mutual if they are to promote stability, not unidirectional and destabilising as from Erbil to Baghdad. Although no single party will gain a majority given proportionality, broader ideological interests threaten to win destabilising majorities: Specifically, the ‘Emirate question’ concerns whether Afghanistan should preserve civil liberties and democratic institutions gained since 2002 or introduce Islamist reforms. For Afghan power-sharing stability, each side must be empowered to protect these sensitive political needs. Taliban, for example, cannot be enabled to rally broader Islamist support beyond their own mostly Pashtun base to undermine women’s rights in the 2004 Constitution, which they have signalled.[62] Inversely, pro-Republic Afghans cannot be allowed to freely trample Islamist needs like religious justice or education, yet an Afghan Uzbek military commander recently signalled an eagerness to answer growing Islamism with violent suppression.[63]

The pragmatic solution is security through mutual vetoes—preventing destabilising, non-cooperative policies on sensitive issues by requiring consensus. Given Afghanistan’s multiple balance of power, it could suffice to allow one-third of the assembly to pass a motion designating a bill as a key decision requiring a two-thirds majority, similar to a mutual veto mechanism in Northern Irish consociationalism.[64] This threshold approximates recent polling where 25% of Afghans supported transitioning to an Islamic Emirate, but the threshold is negotiable and simply aims to stabilise Kabul’s government through mutual political security.[65] Iraq’s handling of vetoes was destabilising precisely because it was not mutual: Baghdad was effectively powerless to govern the KRG due to the constitution’s provisions on regional legislation’s precedence, and this one-sidedness informed the refusal of Baghdad to comply with provisions treating Kirkuk. Consequently, Western actors must swallow their liberal preferences and push Doha’s negotiations to include mutual vetoes: if vetoes are absent or one-sided as they were in Iraq, then even combined proportionality and grand coalitions risk unravelling over sensitive policy needs.

While central government interests are accommodated through these three consociational dimensions, it would be a mistake to assume this is sufficiently comprehensive for Afghanistan: Autonomy remains necessary to accommodate local actors in Afghanistan’s conflict—the ‘local Taliban’ of the decentralised insurgency, powerbrokers, and warlords, who all threaten stability if neglected. While Iraq’s treatment of autonomy emerged from overrepresenting Kurdish and Shia Arab interests during negotiations, Afghanistan suffers an opposite problem—strong local interests exist for autonomy across Afghanistan’s rural communities, yet little appetite exists to accommodate these interests in Doha’s negotiations, which are primarily concerned with power in Kabul.[66] Autonomy in Afghanistan would not be federal, but rather involves formalising the informal political institutions already governing Afghanistan’s rural communities since at least the 1970s, surviving decades of civil war and earning strong legitimacy.[67] A great failure of Western-led conflict regulation from 2001 was expecting ‘trickle-down centralism’ to spread nationwide from a Kabul government that was highly centralised on paper; this project failed, just as it failed when Abdur Rahman first tried it in the late-nineteenth century, and it currently risks undermining the legitimacy of a power-sharing Kabul.[68] The Taliban’s rapid spread in the 1990s, and again in the 2000s, was facilitated by these local actors who primarily sought security and justice from a superior government, whether Kabul or insurgents, demonstrating these interests’ strength.[69] For everyday governance, these communities are self-reliant and distrust Kabul’s bureaucrats: Localities most often form autonomous extra-legal councils staffed variously by elders, maliks, mullahs, or khans depending on the culture, but in every case handling dispute resolution and local politics through traditional institutions perceived as legitimate.[70]

Amidst civil war, these communities are increasingly militarised, whether joining the Taliban or mobilising against it. Consequently, many now include militia commanders whose cooperation in ‘Disarmament, Demobilisation and Reintegration’ processes is essential for a successful Doha ceasefire.[71] Their interests must be accommodated with guarantees to preserve these institutions in which their social and political capital is invested. This requires Kabul to formalise limitations on its power over rural communities, which is a concession requiring Western pressure without voices at Doha representing these local interests.[72] There is little directly analogous between Iraqi and Afghan autonomy given their immensely different contexts, yet the generalised lesson of Iraq’s experience remains instructive: comprehensiveness in power-sharing is crucial for accommodating the diverse interests within a polity and thereby avoiding an experience characterised by instability.

Although Afghan power-sharing’s stability relies upon this laborious accommodation of interests through comprehensiveness, Iraq’s experience demonstrates how merely agreeing to these terms is insufficient without corresponding commitment from external actors policing compliance. The US must therefore push at Doha for a role as an external guarantor of Afghan power-sharing’s implementation, and this commitment must coexist with its February 2020 withdrawal agreement, upon which Taliban participation depends.[73] Military withdrawal risks being interpreted as non-commitment, yet this is a moment where US commitment must only change tracks from military to political.[74] The failure of US commitment in Iraq before, during, and after the 2005 Constitution was to minimise Western commitment to Iraqi power-sharing while instead focusing on invasion and counter-insurgency. Once agreed, Afghan power-sharing must be implemented and upheld, which requires constitutional amendments, legislative reforms, and elections—fragile processes risking reversion to war.[75] Without external commitment, Iraq’s experience showcased elites refusing to implement agreed power-sharing, like Kirkuk and the Erbil Agreement. In Afghanistan, the risk of conflict over comparable failures has recent precedents in the 2014 and 2019 elections. In these elections, the necessity of Western commitment was demonstrated by US and UN election monitoring, discovering substantial voter fraud and offering essential mediation.[76] This crisis was managed by external commitment and from Iraq’s experience of non-committed external actors to imagine that subtracting external policing while adding Taliban participation would produce stability. The commitment of external actors to policing Doha’s agreements will determine the stability of any comprehensive, accommodative, consociational power-sharing in Afghanistan by managing these inevitable moments of crisis and dispute.

In conclusion, Western actors must address three interrelated components of power-sharing in Afghanistan to avoid an unstable experience: Power-sharing must include diverse interests which may not presently have a voice at Doha yet credibly risk destabilising an unaccommodating intra-Afghan agreement; it must achieve this accommodation through comprehensiveness, broaching all consociational dimensions of power-sharing; and it must obligate commitment from Western actors, primarily the US, to police compliance with this agreement, without which the risk of disputes and reneging which threaten to spoil peace. Each of these lessons is inseparable and together forms an instructive model derived from the failures of conflict regulation in Iraq, where consociational power-sharing was not joined by accommodation, comprehensiveness, or commitment, and so produced an Iraqi experience characterised by instability. By applying these lessons fully, an intra-Afghan agreement would finally be on a meandering road to stable, consociational power-sharing.

Notes

[1] Arend Lijphart, Democracy in Plural Societies: A Comparative Exploration (New Haven: Yale University Press, 1977), 25–43.

[2] John McGarry, Brendan O’Leary, and Richard Simeon, ‘Integration or Accommodation? The Enduring Debate in Conflict Regulation’, in Constitutional Design for Divided Societies: Integration or Accommodation?, ed. Sujit Choudhry (Oxford: Oxford University Press, 2008), 45–51; John McGarry and Brendan O’Leary, ‘Iraq’s Constitution of 2005: Liberal Consociation as Political Prescription’, International Journal of Constitutional Law 5, no. 4 (2007): 670–76.

[3] Nussaibah Younis, ‘Set Up to Fail: Consociational Political Structures in Post-War Iraq, 2003–2010’, Contemporary Arab Affairs 4, no. 1 (2011): 2.

[4] Eduardo Abu Ltaif, ‘The Limitations of the Consociational Arrangements in Iraq’, Ethnopolitics Papers 38, no. 1 (2015): 5–12.

[5] Toby Dodge, ‘Iraq’s Road Back to Dictatorship’, Survival 54, no. 3 (2012): 150–57; Raad Alkadiri, ‘Federalism and Iraq’s Constitutional Stalemate’, Research Paper, Middle East and North Africa Program (Chatham House, 2020), 2, https://www.chathamhouse.org/sites/default/files/2020-11/2020-11-27-iraq-federalism-alkadiri.pdf.pdf.

[6] Donald Rumsfeld, ‘Referring to the “Plan”’, 16 December 2003, https://www.esd.whs.mil/Portals/54/Documents/FOID/Special_Collection/Rumsfeld/DocumentsReleased_to_Rumsfeld_Under_FOIA.pdf?ver=2017-05-05-104646-373.

[7] Toby Dodge, ‘Intervention and Dreams of Exogenous Statebuilding: The Application of Liberal Peacebuilding in Afghanistan and Iraq’, Review of International Studies 39, no. 5 (2013): 1201; Special Inspector General for Iraq Reconstruction, Hard Lessons: The Iraqi Reconstruction Experience (Washington, D.C.: US Government Printing Office, 2009), 7–28.

[8] Special Inspector General for Iraq Reconstruction, Hard Lessons, 7–8.

[9] Special Inspector General for Iraq Reconstruction, 8.

[10] Phebe Marr, The Modern History of Iraq (Boulder: Westview Press, 2012), 260–65; Special Inspector General for Iraq Reconstruction, Hard Lessons, 40–41.

[11] Special Inspector General for Iraq Reconstruction, Hard Lessons, 73–74.

[12] Jonathan Monten, ‘Intervention and State-Building: Comparative Lessons from Japan, Iraq, and Afghanistan’, ANNALS of the American Academy of Political and Social Science 656, no. 1 (2014): 182; Marr, The Modern History of Iraq, 267–68.

[13] Dodge, ‘Liberal Peacebuilding in Afghanistan and Iraq’, 1206.

[14] Renad Mansour and Faleh Jabar, ‘Inter- and Intra-Ethnic Relations and Power Sharing in Post-Conflict Iraq’, European Yearbook of Minority Issues Online 11, no. 1 (2014): 192.

[15] Marr, The Modern History of Iraq, 287–92; Gareth Stansfield, ‘Intervention in Iraq: From Regime Change to de Facto Partition’, in Routledge Handbook of Security Studies: Second Edition, ed. Myriam Dunn Cavelty and Thierry Balzacq (New York: Routledge, 2017), 281; Charles Tripp, A History of Iraq (Cambridge: Cambridge University Press, 2007), 295–96; Younis, ‘Set Up to Fail’, 12–13.

[16] Special Inspector General for Iraq Reconstruction, Hard Lessons, 204; Mansour and Jabar, ‘Power Sharing in Post-Conflict Iraq’, 191–93.

[17] Paul Bremer, ‘Strategy for Iraq’s Political Transition’, 10 November 2003, https://www.esd.whs.mil/Portals/54/Documents/FOID/Reading%20Room/CPA_ORHA/06-F-2495_doc_1.pdf; Special Inspector General for Iraq Reconstruction, Hard Lessons, 205–6.

[18] Younis, ‘Set Up to Fail’, 4 and 9–10.

[19] Coalition Provisional Authority, ‘CPA 0767: Administrator’s Weekly Governance’, 12 March 2004, § 3, https://www.esd.whs.mil/Portals/54/Documents/FOID/Reading%20Room/CPA_ORHA/08-F-1338_CPA_0767_Administratior_weekly_Governance_March-12-2004.pdf; Special Inspector General for Iraq Reconstruction, Hard Lessons, 229–31.

[20] Arend Lijphart, ‘Power-Sharing, Ethnic Agnosticism, and Political Pragmatism’, Transformation 21, no. 1 (1993): 94–96.

[21] Alkadiri, ‘Federalism and Iraq’s Constitutional Stalemate’, 4; Marr, The Modern History of Iraq, 269–71.

[22] Gareth Stansfield, ‘Accepting Realities in Iraq’, Briefing Paper, Middle East Program (Chatham House, 2007), 6, https://reliefweb.int/sites/reliefweb.int/files/resources/97ACCEBF0B94A5FC492572DF000F82CA-Full_Report.pdf; Alkadiri, ‘Federalism and Iraq’s Constitutional Stalemate’, 4–5; Marr, The Modern History of Iraq, 293.

[23] Ashley Deeks and Matthew D. Burton, ‘Iraq’s Constitution: A Drafting History’, Cornell International Law Journal 40, no. 1 (2007): 63–65.

[24] Constitute Project, ‘Iraq’s Constitution of 2005’, 2005, §§110 and 114-15, https://www.constituteproject.org/constitution/Iraq_2005.pdf.

[25] Alkadiri, ‘Federalism and Iraq’s Constitutional Stalemate’, 7–8; Dodge, ‘Iraq’s Road Back to Dictatorship’, 154.

[26] Lijphart, Democracy in Plural Societies, 38–39; Deeks and Burton, ‘Iraq’s Constitution’, 63–65 and 72–74; Mansour and Jabar, ‘Power Sharing in Post-Conflict Iraq’, 199.

[27] Constitute Project, ‘Iraq’s Constitution of 2005’, §§ 110, 114-15.

[28] Lijphart, Democracy in Plural Societies, 37; Matthijs Bogaards, ‘Iraq’s Constitution of 2005: The Case Against Consociationalism “Light”’, Ethnopolitics, 2019, 6–8.

[29] Constitute Project, ‘Iraq’s Constitution of 2005’, §§ 70 and 76.

[30] Paul Dixon, ‘Is Consociational Theory the Answer to Global Conflict? From the Netherlands to Northern Ireland and Iraq’, Political Studies Review 9, no. 3 (2011): 316–17; Bogaards, ‘Iraq’s Constitution of 2005’, 1–2 and 8; Toby Dodge, ‘Iraq’s Informal Consociationalism and Its Problems’, Studies in Ethnicity and Nationalism 20, no. 2 (2020): 145–48.

[31] Dodge, ‘Liberal Peacebuilding in Afghanistan and Iraq’, 1206–12.

[32] Liam Anderson, ‘Power Sharing in Kirkuk: The Need for Compromise’, in Power Sharing in Deeply Divided Places, ed. Joanne McEvoy and Brendan O’Leary (Philadelphia: University of Pennsylvania Press, 2013), 364.

[33] Constitute Project, ‘Iraq’s Constitution of 2005’, § 140; Coalition Provisional Authority, ‘Law of Administration for the State of Iraq’, 8 March 2004, § 58, https://web.archive.org/web/20090423064920/http://www.cpa-iraq.org/government/TAL.html.

[34] Anderson ‘Power Sharing in Kirkuk’, 367.

[35] Stefan Wolff, ‘Governing (in) Kirkuk: Resolving the Status of a Disputed Territory in Post-American Iraq’, International Affairs 86, no. 6 (2010): 1372; Anderson, ‘Power Sharing in Kirkuk’, 367.

[36] Wolff, ‘Governing (in) Kirkuk’, 1369–71.

[37] Alkadiri, ‘Federalism and Iraq’s Constitutional Stalemate’, 10.

[38] Dodge, ‘Iraq’s Road Back to Dictatorship’, 150–51; Mansour and Jabar, ‘Power Sharing in Post-Conflict Iraq’, 193–94.

[39] Marr, The Modern History of Iraq, 320; Dodge, ‘Iraq’s Road Back to Dictatorship’, 151–53.

[40] Dodge, ‘Iraq’s Road Back to Dictatorship’, 151–52.

[41] Younis, ‘Set Up to Fail’, 4–6.

[42] UNAMI, trans., ‘Erbil Agreement’, 7 November 2010, https://peacemaker.un.org/sites/peacemaker.un.org/files/IQ_101107_IraqErbilAgreement%20%28English%29.pdf; Ltaif, ‘The Limitations of the Consociational Arrangements in Iraq’, 7 and 13–14.

[43] Dodge, ‘Iraq’s Road Back to Dictatorship’, 157.

[44] BBC Monitoring, ‘Afghan Politician Decries Doha Peace Talks’, 21 October 2020, https://monitoring.bbc.co.uk/product/c20242hz.

[45] Joanna Nathan, ‘Reading the Taliban’, in Decoding the New Taliban: Insights from the Afghan Field, ed. Antonio Giustozzi (London: Hurst & Co., 2009), 32; Anatol Lieven, ‘The War in Afghanistan: Its Background and Future Prospects: Analysis’, Conflict, Security & Development 9, no. 3 (2009): 337–38.

[46] BBC Monitoring, ‘Afghan Reconciliation Council Chief Criticises Appointments’, 1 September 2020, https://monitoring.bbc.co.uk/product/c201zyqy; US Department of Defense Inspector General, Operation Freedom’s Sentinel: Lead Inspector General Report to the United States Congress (Arlington: US Department of Defense, 2020), 10–11.

[47] Special Inspector General for Afghanistan Reconstruction, Reintegration of Ex-Combatants: Lessons from the U.S. Experience in Afghanistan (Arlington: SIGAR, 2019), 53–57.

[48] Martine van Bijlert, ‘Unruly Commanders and Violent Power Struggles: Taliban Networks in Uruzgan’, in Decoding the New Taliban: Insights from the Afghan Field, ed. Antonio Giustozzi (London: Hurst & Co., 2009), 160.

[49] Alex Strick van Linschoten and Felix Kuehn, An Enemy We Created: The Myth of Taliban-Al Qaeda Merger in Afghanistan (London: Hurst & Co., 2012), 246–47; van Bijlert, ‘Unruly Commanders and Violent Power Struggles’, 162–66.

[50] van Bijlert, ‘Unruly Commanders and Violent Power Struggles’, 160–61.

[51] Lieven, ‘The War in Afghanistan’, 343–44; van Linschoten and Kuehn, An Enemy We Created, 243–44.

[52] Special Inspector General for Afghanistan Reconstruction, Reintegration of Ex-Combatants, 55.

[53] Constitute Project, ‘Afghanistan’s Constitution of 2004’, 2004, §§ 60-4, 71-3, and 90-5, https://www.constituteproject.org/constitution/Afghanistan_2004.pdf.

[54] Lieven, ‘The War in Afghanistan’, 339; Thomas Barfield, Afghanistan: A Cultural and Political History (Princeton: Princeton University Press, 2010), 277–78; Jennifer Brick Murtazashvili, ‘Informal Federalism: Self-Governance and Power Sharing in Afghanistan’, Publius: The Journal of Federalism 44, no. 2 (2014): 326–27.

[55] Colin Cookman, ‘Elections as a Disputed Power-Sharing Mechanism in Afghanistan’, SAIS Review of International Affairs 38, no. 1 (2018): 82–84.

[56] McGarry, O’Leary, and Simeon, ‘Integration or Accommodation?’, 70–71; Barfield, Afghanistan: A Cultural and Political History, 19–21; Lieven, ‘The War in Afghanistan’, 338–40.

[57] Ltaif, ‘The Limitations of the Consociational Arrangements in Iraq’, 6–7; Mansour and Jabar, ‘Power Sharing in Post-Conflict Iraq’, 198–99.

[58] Cookman, ‘Elections as a Disputed Power-Sharing Mechanism in Afghanistan’, 84–86.

[59] Lieven, ‘The War in Afghanistan’, 338; Barfield, Afghanistan: A Cultural and Political History, 6–7.

[60] Lijphart, Democracy in Plural Societies, 55–57; Donald L. Horowitz, ‘Ethnic Power Sharing: Three Big Problems’, Journal of Democracy 25, no. 2 (2014): 10–11.

[61] Marr, The Modern History of Iraq, 298–300.

[62] BBC Monitoring, ‘Afghan Peace Talks in Doha’, 11 December 2020, https://monitoring.bbc.co.uk/product/c2028ka8; Constitute Project, ‘Afghanistan’s Constitution of 2004’, §§ 83-4 and 149.

[63] BBC Monitoring, ‘Military Official Seeks Permission to “Suppress” Taliban’, 19 November 2020, https://monitoring.bbc.co.uk/product/c20242hz.

[64] Northern Ireland Office, ‘The Belfast Agreement’, 10 April 1998, Strand One, § 5(d), https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/136652/agreement.pdf.

[65] BBC Monitoring, ‘Survey Result Backs Republic System in Afghanistan’, 21 September 2020, https://monitoring.bbc.co.uk/product/c2021m3r.

[66] Alex Thier, ‘What Are the Prospects for Power-Sharing in the Afghan Peace Process?’, United States Institute of Peace, 16 September 2019, https://www.usip.org/publications/2019/09/what-are-prospects-power-sharing-afghan-peace-process; Ulrich Pilster, ‘Afghanistan: Peace through Power-Sharing?’, The Washington Quarterly 43, no. 1 (2020): 137.

[67] Murtazashvili, ‘Informal Federalism’, 324–25.

[68] Christine Noelle-Karimi, ‘Jirga, Shura and Community Development Councils’, in Local Politics in Afghanistan: A Century of Intervention in the Social Order, ed. Conrad Schetter (Oxford: Oxford University Press, 2014), 42–43; Barfield, Afghanistan: A Cultural and Political History, 161–63; Murtazashvili, ‘Informal Federalism’, 327.

[69] Nazif Shahrani, ‘Centre-Periphery Relations in Afghanistan’, in Local Politics in Afghanistan, ed. Schetter, 31–34; Antonio Giustozzi, ‘Local Politics and the Taliban’, in Local Politics in Afghanistan, ed. Schetter, 79–83 and 86–87; Thomas Barfield, ‘Continuities and Changes in Local Politics in Northern Afghanistan’, in Local Politics in Afghanistan, ed. Schetter, 140.

[70] Giustozzi, ‘Local Politics and the Taliban’, 78–79.

[71] Barfield, ‘Local Politics in Northern Afghanistan’, 138–40; Special Inspector General for Afghanistan Reconstruction, Reintegration of Ex-Combatants, 62–64, 84–85, and 94–95; US Department of Defense Inspector General, Operation Freedom’s Sentinel: Lead Inspector General Report to the United States Congress, 13.

[72] Thier, ‘What Are the Prospects for Power-Sharing in the Afghan Peace Process?’; Pilster, ‘Afghanistan: Peace through Power-Sharing?’, 136.

[73] Ryan C. van Wie, ‘A Peacekeeping Mission in Afghanistan: Pipedream or Path to Stability?’, Journal of Indo-Pacific Affairs 3, no. 4 (2020): 182–84; US Department of State, ‘Agreement for Bringing Peace to Afghanistan’, 29 February 2020, 2, https://www.state.gov/wp-content/uploads/2020/02/Agreement-For-Bringing-Peace-to-Afghanistan-02.29.20.pdf.

[74] van Wie, ‘A Peacekeeping Mission in Afghanistan: Pipedream or Path to Stability?’, 197–98.

[75] Pilster, ‘Afghanistan: Peace through Power-Sharing?’, 136–38; Cookman, ‘Elections as a Disputed Power-Sharing Mechanism in Afghanistan’, 83–84.

[76] Cookman, ‘Elections as a Disputed Power-Sharing Mechanism in Afghanistan’, 83–84.

References

Alkadiri, Raad. ‘Federalism and Iraq’s Constitutional Stalemate’. Research Paper. Middle East and North Africa Program. Chatham House, 2020. https://www.chathamhouse.org/sites/default/files/2020-11/2020-11-27-iraq-federalism-alkadiri.pdf.pdf.

Anderson, Liam. ‘Power Sharing in Kirkuk: The Need for Compromise’. In Power Sharing in Deeply Divided Places, edited by Joanne McEvoy and Brendan O’Leary, 364–85. Philadelphia: University of Pennsylvania Press, 2013.

Barfield, Thomas. Afghanistan: A Cultural and Political History. Princeton: Princeton University Press, 2010.

BBC Monitoring. ‘Afghan Peace Talks in Doha’, 11 December 2020. https://monitoring.bbc.co.uk/product/c2028ka8.

———. ‘Afghan Politician Decries Doha Peace Talks’, 21 October 2020. https://monitoring.bbc.co.uk/product/c20242hz.

———. ‘Afghan Reconciliation Council Chief Criticises Appointments’, 1 September 2020. https://monitoring.bbc.co.uk/product/c201zyqy.

———. ‘Military Official Seeks Permission to “Suppress” Taliban’, 19 November 2020. https://monitoring.bbc.co.uk/product/c20242hz.

———. ‘Survey Result Backs Republic System in Afghanistan’, 21 September 2020. https://monitoring.bbc.co.uk/product/c2021m3r.

Bogaards, Matthijs. ‘Iraq’s Constitution of 2005: The Case Against Consociationalism “Light”’. Ethnopolitics, 2019, 1–17.

Bremer, Paul. ‘Strategy for Iraq’s Political Transition’, 10 November 2003. https://www.esd.whs.mil/Portals/54/Documents/FOID/Reading%20Room/CPA_ORHA/06-F-2495_doc_1.pdf.

Coalition Provisional Authority. ‘CPA 0767: Administrator’s Weekly Governance’, 12 March 2004. https://www.esd.whs.mil/Portals/54/Documents/FOID/Reading%20Room/CPA_ORHA/08-F-1338_CPA_0767_Administratior_weekly_Governance_March-12-2004.pdf.

———. ‘Law of Administration for the State of Iraq’, 8 March 2004. https://web.archive.org/web/20090423064920/http://www.cpa-iraq.org/government/TAL.html.

Constitute Project. ‘Afghanistan’s Constitution of 2004’, 2004. https://www.constituteproject.org/constitution/Afghanistan_2004.pdf.

———. ‘Iraq’s Constitution of 2005’, 2005. https://www.constituteproject.org/constitution/Iraq_2005.pdf.

Cookman, Colin. ‘Elections as a Disputed Power-Sharing Mechanism in Afghanistan’. SAIS Review of International Affairs 38, no. 1 (2018): 81–91.

Deeks, Ashley, and Matthew D. Burton. ‘Iraq’s Constitution: A Drafting History’. Cornell International Law Journal 40, no. 1 (2007): 1–87.

Dixon, Paul. ‘Is Consociational Theory the Answer to Global Conflict? From the Netherlands to Northern Ireland and Iraq’. Political Studies Review 9, no. 3 (2011): 309–22.

Dodge, Toby. ‘Intervention and Dreams of Exogenous Statebuilding: The Application of Liberal Peacebuilding in Afghanistan and Iraq’. Review of International Studies 39, no. 5 (2013): 1189–1212.

———. ‘Iraq’s Informal Consociationalism and Its Problems’. Studies in Ethnicity and Nationalism 20, no. 2 (2020): 145–52.

———. ‘Iraq’s Road Back to Dictatorship’. Survival 54, no. 3 (2012): 147–68.

Giustozzi, Antonio, ed. Decoding the New Taliban: Insights from the Afghan Field. London: Hurst & Co., 2009.

Horowitz, Donald L. ‘Ethnic Power Sharing: Three Big Problems’. Journal of Democracy 25, no. 2 (2014): 5–20.

Lieven, Anatol. ‘The War in Afghanistan: Its Background and Future Prospects: Analysis’. Conflict, Security & Development 9, no. 3 (2009): 333–59.

Lijphart, Arend. Democracy in Plural Societies: A Comparative Exploration. New Haven: Yale University Press, 1977.

———. ‘Power-Sharing, Ethnic Agnosticism, and Political Pragmatism’. Transformation 21, no. 1 (1993): 94–99.

Linschoten, Alex Strick van, and Felix Kuehn. An Enemy We Created: The Myth of Taliban-Al Qaeda Merger in Afghanistan. London: Hurst & Co., 2012.

Ltaif, Eduardo Abu. ‘The Limitations of the Consociational Arrangements in Iraq’. Ethnopolitics Papers 38, no. 1 (2015): 1–22.

Mansour, Renad, and Faleh Jabar. ‘Inter- and Intra-Ethnic Relations and Power Sharing in Post-Conflict Iraq’. European Yearbook of Minority Issues Online 11, no. 1 (2014): 187–209.

Marr, Phebe. The Modern History of Iraq. Boulder: Westview Press, 2012.

McGarry, John, and Brendan O’Leary. ‘Iraq’s Constitution of 2005: Liberal Consociation as Political Prescription’. International Journal of Constitutional Law 5, no. 4 (2007): 670–98.

McGarry, John, Brendan O’Leary, and Richard Simeon. ‘Integration or Accommodation? The Enduring Debate in Conflict Regulation’. In Constitutional Design for Divided Societies: Integration or Accommodation?, edited by Sujit Choudhry. Oxford: Oxford University Press, 2008.

Monten, Jonathan. ‘Intervention and State-Building: Comparative Lessons from Japan, Iraq, and Afghanistan’. ANNALS of the American Academy of Political and Social Science 656, no. 1 (2014): 173–91.

Murtazashvili, Jennifer Brick. ‘Informal Federalism: Self-Governance and Power Sharing in Afghanistan’. Publius: The Journal of Federalism 44, no. 2 (2014): 324–43.

Northern Ireland Office. ‘The Belfast Agreement’, 10 April 1998. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/136652/agreement.pdf.

Pilster, Ulrich. ‘Afghanistan: Peace through Power-Sharing?’ The Washington Quarterly 43, no. 1 (2020): 121–41.

Rumsfeld, Donald. ‘Referring to the “Plan”’, 16 December 2003. https://www.esd.whs.mil/Portals/54/Documents/FOID/Special_Collection/Rumsfeld/DocumentsReleased_to_Rumsfeld_Under_FOIA.pdf?ver=2017-05-05-104646-373.

Schetter, Conrad, ed. Local Politics in Afghanistan: A Century of Intervention in the Social Order. Oxford: Oxford University Press, 2014.

Special Inspector General for Afghanistan Reconstruction. Reintegration of Ex-Combatants: Lessons from the U.S. Experience in Afghanistan. Arlington: SIGAR, 2019.

Special Inspector General for Iraq Reconstruction. Hard Lessons: The Iraqi Reconstruction Experience. Washington, D.C.: US Government Printing Office, 2009.

Stansfield, Gareth. ‘Accepting Realities in Iraq’. Briefing Paper. Middle East Program. Chatham House, 2007. https://reliefweb.int/sites/reliefweb.int/files/resources/97ACCEBF0B94A5FC492572DF000F82CA-Full_Report.pdf.

———. ‘Intervention in Iraq: From Regime Change to de Facto Partition’. In Routledge Handbook of Security Studies: Second Edition, edited by Myriam Dunn Cavelty and Thierry Balzacq, 280–90. New York: Routledge, 2017.

Thier, Alex. ‘What Are the Prospects for Power-Sharing in the Afghan Peace Process?’ United States Institute of Peace, 16 September 2019. https://www.usip.org/publications/2019/09/what-are-prospects-power-sharing-afghan-peace-process.

Tripp, Charles. A History of Iraq. Cambridge: Cambridge University Press, 2007.

UNAMI, trans. ‘Erbil Agreement’, 7 November 2010. https://peacemaker.un.org/sites/peacemaker.un.org/files/IQ_101107_IraqErbilAgreement%20%28English%29.pdf.

US Department of Defense Inspector General. Operation Freedom’s Sentinel: Lead Inspector General Report to the United States Congress. Arlington: US Department of Defense, 2020.

US Department of State. ‘Agreement for Bringing Peace to Afghanistan’, 29 February 2020. https://www.state.gov/wp-content/uploads/2020/02/Agreement-For-Bringing-Peace-to-Afghanistan-02.29.20.pdf.

Wie, Ryan C. van. ‘A Peacekeeping Mission in Afghanistan: Pipedream or Path to Stability?’ Journal of Indo-Pacific Affairs 3, no. 4 (2020): 182–203.

Wolff, Stefan. ‘Governing (in) Kirkuk: Resolving the Status of a Disputed Territory in Post-American Iraq’. International Affairs 86, no. 6 (2010): 1361–79.

Younis, Nussaibah. ‘Set Up to Fail: Consociational Political Structures in Post-War Iraq, 2003–2010’. Contemporary Arab Affairs 4, no. 1 (2011): 1–18.


Written at: King’s College London
Written for: Professor Michael Kerr
Date written: December 2020

Further Reading on E-International Relations







This article by Colm Trant, originally published on E-International Relations is licensed under Attribution-NonCommercial 4.0 International(CC BY-NC 4.0).

What do you think?

122 Votes
Upvote Downvote
Avatar Of Global Governance Veteran

Written by Global Governance

Content AuthorYears Of Membership

Comments

Leave a Reply

GIPHY App Key not set. Please check settings

Loading…

0
File 20210318 23 Dnatl1

When our evolutionary ancestors first crawled onto land, their brains only half-filled their skulls

File 20210320 15 Fszc79

‘Sex addiction’ isn’t a justification for killing, or really an addiction – it reflects a person’s own moral misgivings about sex