Concerning the second point, the constitution has already, in effect, been amended by the power-sharing agreement between President Ghani and Chief Executive Abdullah Abdullah. This is supposed to be ratified by a loya jirga. Moreover, this agreement creates a kind of prime ministerial position, which the pragmatic Taliban would clearly prefer since it allows the division of power between different forces represented in parliament. As to the supremacy of sharia law, this is recognized formally at least by a large number of Muslim states including some democracies; it should not be beyond the wit of man—or God—to devise a formula by which sovereignty comes from the divine but is exercised by the people.
Much more difficult are the issues of power—in the central government and, more importantly, in core Taliban strongholds. Chief among these issues is military force. The Taliban is not going to disarm. No one ever does in Afghanistan, unless they are defeated or dead. This has been emphatically true of the warlords and ethnic militias that Washington and the new Afghan state inherited in 2001. The Taliban would therefore have to be turned into an autonomous part of the Afghan security forces—as warlord forces have been in various parts of the country.
Territory subjected to these security forces would also see their elections fall under Taliban control. Leading supporters of the Afghan state in the core Taliban areas would be destroyed or forced to leave, possibly under the official banner of a new anti-heroin campaign, like that of 1999.
Central to the collapse of the latest peace process is the fact that the idea of accepting permanent armed forces and control of large parts of the country has been utterly unacceptable to the Afghan government and the great majority of its supporters. Such an arrangement, they said, would inevitably lead to a new war, with the government forces in a greatly weakened position. Surely, the Taliban would only pretend to accept such a compromise, would wait until U.S. forces had withdrawn and then launch a new attack to achieve total victory.
Analysts in Kabul told me that such a settlement could not be reached because it would radically change the terms of the patronage arrangements for the distribution of Western aid on which the present system is founded. A peace settlement would almost certainly lead to a reduction in existing levels of aid from the United States and Europe, thereby reducing the overall pool and increasing competition between groups. The introduction of the Taliban into a share of government would reduce the available patronage still further. In particular, it would greatly increase tensions over the issue of who represents the Pashtuns of Afghanistan, who enjoy a plurality of the population (most estimates put them between 40 and 45 percent, though almost all Pashtuns think they are an absolute majority).
Ghani has been heavily criticized by many Afghans and some Western observers for allotting too many jobs to Pashtuns (and in particular to Ghilzai Pashtuns, the loose tribal grouping from which he himself originates). He appears, however, to be merely following the logic of the power-sharing arrangement brokered by John Kerry in 2014, which involved a tacit assumption that the president must be Pashtun and that Abdullah (whose father was Pashtun and mother Tajik, but who was closely associated with the Tajik-dominated Northern Alliance) would be compensated with a quasi–prime ministerial position. It seems, therefore, to be President Ghani’s view that he will favor the Pashtuns with jobs, and Abdullah can look after the rest. Pashtun forces long accustomed to these benefits are not eager to share their patronage spoils with the Taliban.
GIVEN THE deadlock in direct talks with the Taliban, both Washington and Kabul had to fall back on the strategy of persuading Pakistan to force the Taliban leadership to make peace on Afghanistan’s terms. Since Pakistan provides safe havens (well, more or less safe, as Mansour’s death demonstrates) and a limited measure of support for Taliban operations, Pakistan obviously has the ability to bring pressure to bear. Moreover, Pakistani support for the Taliban is not as strong as in the past. The rise of the Pakistani Taliban insurgency changed Pakistani military attitudes to the Afghan Taliban considerably since the 1990s, when Islamabad backed the group unconditionally and was instrumental in its rise to power.
Pakistan is also influenced by China. Though China is generally sympathetic to Pakistan’s agenda in Afghanistan, Beijing most definitely does not want Taliban rule over the whole country, which could then lead to a wave of Islamist militancy in Central Asia and Chinese Xinjiang. Finally, Islamabad’s successful offensive against the last Pakistani Taliban strongholds in North Waziristan in 2014–15 seemed to make possible further action against the Afghan Taliban. In pursuit of help against the Taliban, President Ghani wooed Pakistan intensively and even visited Pakistani military HQ in Rawalpindi, a move bitterly criticized by many Afghans.
This effort by Kabul and Washington foundered both because Ghani did not offer any real incentives to Pakistan and because of a set of interlinked Pakistani beliefs and goals, a kind of strategic dogma in Islamabad. Pakistani generals and officials believe that India is seeking to undermine Pakistan through Afghanistan, above all through support for the ethnic separatist rebellion in Baluchistan; that the Afghan intelligence agency, the National Directorate of Security (NDS), is controlled by bitter enemies of Pakistan and is closely allied to India’s intelligence agency, the Research and Analysis Wing; that since the remaining leadership and cadres of the Pakistani Taliban were driven across the border into Afghanistan, they have become tools of the NDS against Islamabad (something that is also believed by many well-informed observers in Kabul, and was strongly implied by the arrest of a Pakistani Taliban leader, Latif Mehsud, in Afghanistan in March 2013 after he had been in contact with Afghan officials); that the Afghan state since 2001 has been dominated by Tajiks, and that its Pashtun leaders (presidents Hamid Karzai and Ashraf Ghani) have been impotent figureheads and American stooges; that the Afghan Taliban are the real representatives of the Pashtuns of Afghanistan and have strong support in the Pashtun territories; and that Kabul is hopelessly weak, the country is divided and dependent on Western support, and is bound to collapse.
Whether developments in Afghanistan lead to a peace settlement with the Taliban or anarchy and civil war, the Pakistani establishment is determined to ensure that the bulk of the Taliban will remain friendly to Pakistan. Islamabad believes that India, Iran and Russia will all acquire their own client groups in Afghanistan, that Pakistan will require the same and that the Taliban is the most viable contender. Central to Pakistani policy is an absolute determination not to fight another war against the Afghan Taliban on home soil, just when Islamabad has managed to end an extremely bloody war against the Pakistani Taliban, in which thousands of troops and police and tens of thousands of civilians lost their lives.
Any analysis of Islamabad’s strategy hinges on uncertainty over the patron-client dynamic between Pakistan and the Afghan Taliban. The problem here is that statements by the Pakistani and the Afghan secret services are almost completely worthless. Until recently, Pakistan even denied giving shelter to the Taliban. During a March 1 talk at the Council on Foreign Relations, the prime minister’s foreign-affairs adviser, Sartaj Aziz, justified harboring the Taliban leadership and their families. Only with the resulting leverage could Pakistan drag the Taliban to the negotiating table.
Western intelligence ought to be more reliable—but a close reading of WikiLeaks reveals that too much of its information comes originally from the NDS and from the interrogation of Taliban prisoners. I have little doubt that after a few days in NDS custody anyone would say whatever his/her captors wanted, as would prisoners in the hands of Pakistan’s Inter-Services Intelligence directorate (ISI).
The difficulty of pinning down anything on the subject of Pakistani support—as opposed to shelter—was made abundantly clear during my latest visit to Afghanistan in May and June of this year. Concerning the source of a massive truck bomb in Kabul on April 19, which killed sixty-four people, one very well-informed Western official with an international organization told me that the truck had contained weapons-grade explosives, part of a consignment of almost a ton that the ISI gave to the Taliban earlier this year. Another equally well-informed Western official in the same organization told me that the bomb had been made from a base of ordinary nitrate fertilizer, and that only the detonator was weapons grade—nitrate fertilizer is freely available in Afghanistan and Pakistan, and has been used in the past to make many bombs, including in Oklahoma City.
While safe havens in Pakistan have undoubtedly been of great importance to the Afghan Taliban, overall support has been much more limited and mostly limited to attacks on Indian targets conducted by the Haqqanis. If Western intelligence services are correct in assessing the Taliban’s annual budget between $300 and $600 million (around one-tenth of what the United States provides the Afghan security forces), then the Taliban can finance its own campaign from heroin and sympathizers in the Gulf without Pakistan’s help. So far, Taliban weaponry has also been limited and far beneath what the mujahideen in the 1980s enjoyed as a result of U.S., Pakistani and Saudi aid: rifles, machine guns, rocket-propelled grenades and IEDs, but nothing heavier than that. In other words, the notion that Pakistan could assert greater pressure on the Taliban is obvious. That they could simply shut down the Taliban campaign is greatly exaggerated.