The current conflict in Afghanistan dates back to the pro-communist coup in Kabul in 1978. The changes in the international order in the subsequent four decades can be traced in alliances which have helped sustain Afghanistan’s warring parties. For the first decade of the war, a combination of the Cold War and regional politics defined the external linkages. The Soviet Union and the Eastern Bloc supported the government in Kabul, while the US and allies plus the Muslim world supported the anti-government mujahidin. Iran and Pakistan combined hosted in the order of five million refugees and provided rear bases for their preferred mujahidin fighters.

The disengagement of the great powers after the collapse of the Soviet Union meant that in the second decade of the conflict, Afghan actors mainly depended on linkages with regional powers. Initially Pakistan maintained its support to what its intelligence service considered the most effective or reliable of the mujahidin parties, Gulbuddin Hekmatyar’s Hezb-i Islami. India rapidly developed its links with Jamiat Islami, the opposing faction in the old mujahidin alliance, which by now constituted the government of Burhanuddin Rabbani. After 1996, Pakistan found itself supporting a Taliban government in Kabul, which had been formed by a newly emerged faction that had displaced both sides of the old mujahidin alliance. Meanwhile, regional powers, including India, Iran and the Central Asian states channelled materiel to the Northern Alliance resistance to the Taliban, which was formed from the remnants of the Rabbani government plus the old northern militias.

In this light, the post-2001 period after the US and allies had removed the Taliban regime following 9/11 can be seen as a continuation of an internationalised intra-Afghan conflict, but with a major realignment of the patrons. For the first decade or so, Pakistan was widely considered the sole significant backer of the Taliban as they launched their insurgency against the new Kabul government and the NATO forces defending it – a role which Pakistan of course denied. More recently, the resumption of strategic competition between major global powers has strained the international consensus on peace in Afghanistan. The Afghan government has continued to depend upon the US and allies for military and economic support. Meanwhile, the Taliban have broadened their international linkages. They retain access to a ‘safe haven’ in Pakistan, operate a US-endorsed political commission in Doha, are widely considered to receive limited covert assistance from Iran and Russia and have cultivated wider international contacts. An internationalised conflict has added layers of complexity to the peacemaking challenge.

The presence of international troops in both the Soviet and US periods has clearly been one of the key issues around which the Afghan warring parties have mobilised international and domestic support. The Taliban have been at pains to draw the parallels between the two interventions, so as to appeal to the national spirit which inspired resistance to the Soviet presence. But plenty of the drivers of conflict operate independently of the existence of international troops in Afghanistan. In the case of an abrupt withdrawal of US troops, the most likely outcome would be a new reconfiguration of the conflict.

In comparing two of the most significant attempted general peace settlements of the Afghan war, at Peshawar in 1992 and Bonn 2001, it is possible to track efforts to address the challenge of belligerent inclusiveness. Table 3 summarises the Peshawar and Bonn processes.

Peshawar
In 1992, then Pakistani Prime Minister Nawaz Shareef presided over a meeting of representatives of the Pakistan-based Afghan mujahidin parties, which produced the Peshawar Agreement. This provided for a two-month interim government, to be headed by Sibghatullah Mojadidi, leader of one of the weaker mujahidin parties, followed by a four-month initial transitional government headed by Burhanuddin Rabbani, leader of one of the militarily stronger mujahidin parties. An ‘Islamic Council’ was supposed to give a mandate to another two-year transitional government. The agreement included a detailed carve up of cabinet and political appointments between the mujahidin parties. It made some attempt at balancing power, for example by allocating the prime ministership and the defence ministry to the two strongest mujahidin parties, which were rivals.

In the long run of the Afghan war, the Peshawar Accord should be seen as the political corollary to the Geneva Accords. It addressed the issue of government formation, which was parked when Geneva focused on the military issue of Soviet troop withdrawal. The Peshawar Accord was also the precursor to the Bonn Accord, with which it shares a core structure and rationale. In terms of inclusiveness, the key step from ‘Geneva’ to ‘Peshawar’ was that the main Afghan belligerents, the mujahidin parties, dominated the process. At the time, having acted as the figureheads for the military victory over the People’s Democratic Party of Afghanistan regime, the parties seemed to have the legitimacy and military clout necessary to agree the transition.

But only in the most limited sense could Peshawar be thought of as an inclusive process. Firstly, the parties themselves, as political entities, were deeply flawed. There had been a certain organic logic to the way that each of the parties had built itself up. Each was dominated by a leader and recruitment reflected the leader’s ideological appeal and the tribal alliances he could mobilise. Hekmatyar’s Hezb-i Islami had its powerbase among his Ghilzai Pashtuns, while the main appeal of Burhanuddin Rabbani’s Jamiat-e Islami was to fellow Tajiks. Pakistani support enabled the mujahidin to extend these networks across Afghan society. Because each of the parties claimed to be a national movement, they tried to organise in all provinces. However, strongholds also emerged where a single party or a commander affiliated to a single party had the upper hand. For example, Rabbani’s Jamiat became the dominant force in Tajik-majority areas of the north-east.

By 1992 many mujahidin field commanders were deeply suspicious of the Peshawar-based leadership and had formed an alternative alliance, the Commander’s Shura. In the latter stages there had been some cooperation between Pakistan and Iran to push their respective proxies to cooperate. The Pakistani authorities and Peshawar-based mujahidin leaders working on the Peshawar Accord reserved a modest share in power for the Shia mujahidin of Hizb-e Wahdat, on whose behalf Iran advocated. But they excluded the Commander’s Shura, which was considered potentially subversive of the Pakistan-backed mujahidin leadership.

The Peshawar Accord completely excluded the northern militias, which had emerged as a force over the preceding year. And the desire of all the sponsors of the Accord to project the changes under way in Afghanistan as a ‘victory of the Islamic revolution’ dictated that no formal role be given to the former regime – although in practice many of the incumbent deputy ministers were asked to serve in the interim and transitional administrations and senior military figures, such as Chief of Army Staff Asif Dilawar, sought the protection of mujahidin parties and continued to play a public role.

The Peshawar Accord provided for 45 political appointments to the interim administration and state bodies. All of the appointees were male as were the party delegations which agreed the Accord. There was an active Afghan civil society at the time, with a strong presence in Peshawar. UN envoys had built strong relations with this civil society and took some trouble to address societal inclusiveness, for example by ensuring participation of women in consultations. But the UN attempt, under Benon Sevan, at shaping the 1992 transition with a government of neutrals and technocrats failed decisively.

The UN ceded responsibility for convening this stage of the peace process to the Pakistan authorities who, consistent with their approach since the Soviet intervention, worked with the Afghan mujahidin parties. The mujahidin brought their own political culture, which included a significant degree of hostility to civil society and ideological commitment to restricting women’s role. Civil society was comprehensively excluded from the Peshawar process and became critical of the power grab it envisaged. And it was not just organised urban elites among civil society who felt excluded. The rise of the mujahidin parties had helped shift power in rural Afghan society away from tribal structures.

The power carve-up between mujahidin parties in the Peshawar Accord was the culmination of this shift. Furthermore, the Peshawar Accord precluded any move to broaden inclusion during the transitional process. It provided for an ‘Islamic Council’ of the parties’ nominees, to take top-level decisions guiding the transitional process. The lack of societal inclusiveness in the Peshawar process was a predictable consequence of the move from a UN-convened process to one involving the Pakistan authorities working with the mujahidin parties.

The failure of the Peshawar Accord to achieve a peaceful transition in 1992 is well known. A violent power struggle ensued in Kabul, among the parties to the Accord, the excluded northern militias and the marginally included Shia parties. This opened the way for the rise of the Taliban. The Peshawar Accord was the least internationalised of the major efforts at peacemaking during the Afghan conflict. The US and allies had largely disengaged with the end of the Cold War. UN envoy Benon Sevan headed up a diplomatic effort to reach a peaceful transition but did not achieve the level of international backing and local cooperation needed to implement his plan. That failure led to President Najibullah Ahmadzai living out his last three-and-a-half years in a UN compound in Kabul.

Pakistan was left as the sole sponsor of the Peshawar Accord. As a legacy of their support for the anti-Soviet struggle, the Pakistan authorities had well-established relations with all the mujahidin parties. Pakistan also initially provided food aid and cash assistance to the new mujahidin government. But, prudently, Pakistan did not attempt to deploy troops or provide overt security assistance in support of the accord. There was therefore no external mechanism to guarantee the implementation of the accord and few carrots or sticks to induce the Afghan parties to comply with their obligations and cooperate with the interim and transitional administrations.

Meanwhile, regional powers such as Iran, Turkey and India were all prepared to continue backing their favoured proxies, even as the accord fell apart. Measured against the criteria of both belligerent and societal inclusiveness, the Accord was clearly flawed and rapidly collapsed. The included parties calculated that they could grab more power than they had been allocated in Peshawar, and there was no external mechanism to coordinate regional engagement and deter them from this zero-sum behaviour.

Bonn Process
The post-Taliban Bonn Accord of December 2001 constituted, in essence, a reworking of the Peshawar Accord. Indeed, some of the key participants in the Bonn Process, not least President Karzai, had personally experienced the collapse of the 1992 transition. The 2001 process was both more inclusive and more internationalised than that of 1992. But, although the architecture of Bonn was greatly improved relative to Peshawar, it still replicated some of the flaws.

In common with the Peshawar Accord, the basic logic of the Bonn Accord was that Afghan factions reached agreement on the composition of a government in a transitional process, this time convened and witnessed by the UN. The conference that agreed the Accord was more inclusive than the meetings in Peshawar in 1992 because the UN, in addition to the main anti-Taliban armed faction, the Northern Alliance, invited delegations representing different unarmed coalitions with which the UN had maintained diplomatic contacts during the Taliban period. The non-belligerent delegations included the ‘Rome group’ of figures associated with the exiled former king, the ‘Peshawar group’ of figures associated with Pir Ahmad Gailani and moderate mujahidin based in Pakistan, and the ‘Cyprus group’ of mujahidin with some links to Iran. The non-belligerent delegations provided a vehicle for some participation of women, civil society and moderates, and even this limited degree of societal inclusiveness was in stark contrast to the 1992 process.

Critically, the Accord provided for an ambitious roadmap to broaden the base of the administration and to allow real popular participation. It further provided for a tribal assembly (Loya Jirga) to elect the head of the transitional administration and for constitution-making and elections. There was a strong international consensus in support of the Bonn process. In the first place, the US fully backed it, but also supported the format of UN-led talks. Backed by this consensus, UN diplomats were even able to draw on the regional powers to influence their clients to cooperate.

In strengthening the prospects for Bonn to be implemented, the key contrast relative to Peshawar 1992 was that it provided for an international presence and ongoing political, humanitarian and military assistance. The full-spectrum international support enabled a process that was more inclusive and ambitious than would have been possible for any single convener to deliver. Furthermore, although the regional powers such as Iran and Pakistan had a history of cooperation with key participants, these participants had their own agency and supporter base, and acted independently at the negotiating table.

One aspect of the Peshawar Accord that Bonn replicated, however, and which contributed to the next round of violent conflict, was the exclusion of the defeated power – in this instance the Taliban. The UN mission had responsibility for inviting participants and thus takes direct accountability for this exclusion. They were operating closely with US envoys Khalilzad and James Dobbins and, in the immediate aftermath of 9/11, overt Taliban participation would have been unpalatable in Washington. However, Taliban proxy participation, for example by inclusion of Taliban in the Rome Group delegation, would have been possible if organisers had prioritised this.

Arguably, the supporters of the defeated regime in 1992 had been far more successful at associating themselves informally with the post-1992 power structures than were the Taliban during the Bonn process. Military officers, political figures and officials invoked tribal and clan ties to receive the protection of the mujahidin parties and many of them were able to hold onto their positions and work with new allies. Thus, the Bonn Accord succeeded in installing a system of national government which has endured for nearly two decades and which put an end to a cycle of violent power struggles in the capital. But, despite significant provisions for progressive inclusiveness, through the Loya Jirga, constitution-making and elections, the Accord helped sow the seed of another round of violent conflict, as yet unresolved.

The four case studies of internationally backed peace initiatives illustrate differing levels of international engagement and evolving approaches to the challenge of inclusiveness in Afghan peacemaking. The key challenge is how to include belligerents so that agreement on ending violence sticks, and how to include society so that the settlement enjoys broad enough support and effects sufficient change to achieve a lasting peace.

The Geneva Accords rank globally among the most significant diplomatic initiatives of the Cold War. They were part of a fully international process, with a UN mediator and super-power and regional-power involvement. The principal outcome of the process was the removal of one international element of the conflict – the Soviet troop presence. But, ultimately, the Accords did not address the issue of how to achieve a peaceful transition and could not meaningfully have done so given that they were not sufficiently inclusive of the key Afghan belligerents.

The Peshawar Accord was an attempt to settle the political issues which could not be addressed in Geneva and, unlike Geneva, included the ascendant Afghan belligerents as parties. But the process had a single regional sponsor and no support, guarantee or enforcement mechanisms. This process too was inadequately inclusive of the belligerents as it excluded the regime and the northern militias, and it marginalised the mujahidin parties’ influential field commanders. It also had no societal inclusiveness. The parties made a show of complying with the Accord but then engaged in a destructive power struggle.

The Bonn Accord was supported by a strong international process, involving global and regional powers. The conference which produced the accord was already more inclusive of the Afghan parties than Peshawar. But even more significant was the way in which the Bonn Accord provided for a progressive and ambitious expansion of inclusiveness in subsequent decision-making. The period of political stability at the centre achieved through the Bonn Accord was unprecedented in modern Afghan history. This aspect of the outcome qualifies the Bonn process as a success, although the flaws in the political system it established and the resurgence of violence show that the Bonn Accord was insufficient to conclude the peace. Bonn included the seed of the next round of armed conflict because it lacked any provision to bring the defeated Taliban on board.

The launch of a Taliban political office in Qatar was a US-sponsored attempt to overcome the problem of the exclusion of a key belligerent from the Bonn process. The final outcome of that initiative is not yet known, as attempts to bring the Taliban movement into a meaningful peace process, via Qatar, continue. In the interim, installing the Taliban Political Commission in Qatar increased the Taliban’s international exposure, allowing them to project their movement’s position in various fora and to open up channels of communication with multiple international actors. But the movement has opted to pursue this political engagement alongside its war effort, without seriously committing to a peace process.

Previous successes and failures provide indications regarding possible strategies for enhancing inclusiveness. Sequencing the peacemaking agenda is key and all four case studies presented here include evidence on potential and pitfalls. The effective participation of warring parties (belligerent inclusiveness) is vital to obtaining agreement on controlling the violence. But the peace process can be sequenced so that subsequent negotiation and decision-making processes address governance, institutions and the social contract, all of which are essential parts of peacebuilding. Societal inclusiveness is important to the success of these efforts to settle the broad political issues. But it is also important that the peace process receives the sustained backing required to see it through the sequence of stages of winding down the conflict, from ceasefire to the renewed social contract.

The Afghan presidency has so far sought to claim a monopoly on decision-making around the peace process. But, Afghan democratic institutions more broadly can have a fundamental role in enhancing inclusiveness. For example, any meaningful process of peacebuilding through overhauling the social contract would require legislation. This implies a role for parliament. One of the radical contributions of the Bonn Process was that it provided for rapid transition to electoral democracy, with a national parliament, provincial councils and directly elected president. In a plural society such as Afghanistan, these elected institutions potentially have a key role in ensuring that all social groups are represented in decisions around peacemaking.

The five parliamentary and presidential elections since 2004 have provided occasions for societal mobilisation, in which all tribes and ethnic groups have built alliances and influenced outcomes, in which urban elites have been forced to engage with rural interest groups and in which women, helped by reserved parliamentary seats, have come to the fore in public affairs. The chaotic 2018 parliamentary election indicates that lack of integrity in the electoral institutions constitutes as formidable an obstacle to the functioning of Afghan democracy, as does the opposition of the Taliban. One of the most effective ways in which international donors may be able to ensure inclusiveness in a peace process is by helping to restore the integrity of electoral institutions and thus the legitimacy of elected representatives involved in peacemaking.

Experience also indicates many pitfalls in trying to balance different objectives for inclusiveness. Most major actors in Afghanistan have endorsed the notion that any peace process must be ‘Afghan-owned’, while at the same time assuming that there will be significant international involvement. The case studies suggest that international involvement can underpin inclusiveness. The contrast between Peshawar and Bonn indicates that full-spectrum international involvement, which in the Afghanistan case means the US, regional powers and the UN, can provide a vehicle to ensure both belligerent and societal inclusiveness. But, in assessing the degree of inclusiveness, it is important to factor in Afghan belligerents’ tradition of relying on international support as a substitute for domestic support.

The pitfalls also provide a reminder of the need for international peacemakers to adopt a do-no-harm approach, by avoiding legitimising flawed processes (such as stolen elections) and by remaining duly sceptical about opaque armed actors’ unproven claims regarding the extent of their popular base. Experience suggests that international peacemakers may have limited scope to champion successfully particular political schemes to promote inclusiveness or stability in Afghanistan. At times, a better alternative may be to promote consensus, rather than a particular scheme per se.