Afghanistan Analysts Network – English

Regional Relations

“The US’s Greatest Strategic Failure”: Steve Coll on the CIA and the ISI

Ann Wilkens 11 min

“Directorate S” is Steve Coll’s second major study of the CIA’s role in recent Afghan wars. While “Ghost Wars” chronicled the years 1979-2001, “Directorate S” – referring to a subdivision of Pakistan’s inter-services intelligence directorate that covers Afghanistan – takes up the story in 2001 and follows it through to 2016. AAN Advisory Board member Ann Wilkens found Coll’s renderings of the lack of cohesion between the US and its Western allies, as well as between various US institutions, particularly compelling. Equally powerful was Coll’s startling account of the shifting and frequently contradictory views the US held of its Pakistani ally – and the slow unraveling of the bilateral relationship.

Cover of "Directorate S: The CIA and America’s secret wars in Afghanistan and Pakistan, 2001-2016" written by: Steve CollCover of "Directorate S: The CIA and America’s secret wars in Afghanistan and Pakistan, 2001-2016" written by: Steve Coll

Steve Coll has shed more light on the murky politics that govern the relations between the intelligence services of United States, Afghanistan and Pakistan than any other writer. His seminal work “Ghost Wars, The Secret History of the CIA, Afghanistan and Bin Laden, from the Soviet Invasion to September 10, 2001”, first published in 2004, chronicled the role of the CIA in the defeat of the Soviet army in Afghanistan during the emergence of the Taleban movement and in the pre-9/11 hunt for Osama bin Laden. Earlier this year, it was followed by “Directorate S, The C.I.A. and America’s Secret Wars in Afghanistan and Pakistan, 2001-2016”. Directorate S refers to a branch within the Inter-Services Intelligence (ISI), the most powerful of Pakistan’s intelligence agencies, that deals with Afghanistan. It provides a rich and detailed account of the prolonged failure of the international community to bring stability to Afghanistan, recording the minutiae of ‘who-said-what-to-whom-and-when’in an accessible narrative form. This review and dispatch will concentrate on Coll’s coverage of Pakistan’s intelligence service.

Western incoherence

The lack of cohesion among members of the international intervention in Afghanistan has been well-documented previously, and emerges once again in “Directorate S” as a major cause of Western failures in Afghanistan. The divide between the United States and ISAF partners is richly illustrated through the book’s focus on the CIA (which ran its covert war in parallel with, not inside, ISAF). Coll cites as one example the “Riedel review”, compiled in 2009 by former CIA officer Bruce Riedel to help define the Obama administration’s approach to Afghanistan, in which Riedel “found that the United States had only one truly ‘vital’interest in the region: to defeat Al Qaeda. […] America had other interests in the war, such as stability in South Asia and the reduction of heroin trafficking, but Al Qaeda trumped all others.” (p. 366) State-building in Afghanistan, an important goal for ISAF partners, is shown not to have been an American priority, at least not in the early stages of the intervention. When it does become more prominent, with the counter-insurgency (COIN) strategy launched by ISAF commander Stanley McChrystal in 2009, it is accompanied by a military surge emulating developments (then deemed successful) in Iraq two years earlier. Similarly, partner countries, with the exception of the United Kingdom, hardly figure in the many conversations rendered from decision-making circles in Washington. People talk about Afghanistan as if it were an American war, not the joint international effort hailed in official contexts.

Incoherence stands out as a hallmark of decision-making within the US as well. Already during the Bush administration an “incoherent command structure […] had grown up in Afghanistan as a result of ad hoc compromises with N.A.T.O. and within the American military.”In Kandahar and Helmand, the units deployed included “American ‘black’or covert Special Operations units, ‘white’or Green-Beret-style American Special Forces, British forces, Dutch forces, U.S. Marines and multinational Provincial Reconstruction Teams.”(p. 331-332) At the beginning of 2010, Coll writes, there were three different strategies for Afghanistan in Washington: “From ISAF headquarters, Stanley McChrystal commanded an intensifying ground war based on the clear-hold-build-transfer principles of counterinsurgency. […] From the Global Response Center in Langley, the CIA independently ran a drone war against al Qaeda and the Taliban holed up in Waziristan. […] Simultaneously, from the ground floor of the State Department, [the US’s special representative for Afghanistan] Richard Holbrooke and his aides […] pursued a third: trying to talk to [Taleban leader] Mullah Mohammad Omar’s lieutenants about peace. […] On paper, Obama’s National Security Council supported all three policies. But it would require feats of mental gymnastics to call these lines of action synchronized.”(p. 438)

Among these actors, Holbrooke –who passed away in 2010 –seems to be the only one to have focused on the wider, regional picture. Coll renders a private conversation Holbrooke had with a reporter in 2010: “There are three countries here – Pakistan, Afghanistan and India – with vastly different stages of political, social and economic development. They share a common strategic space. As has happened so many times in history, the weak state is the one that sucks in the others. That’s the history of Afghanistan and now the Great Game is being played with different players. The India-Pakistan relationship is an absolutely critical driver.” (p. 430-431) His boss, Foreign Secretary Hillary Clinton, is also wary of the possibility that the US might get bogged down in contradictions: “What was the ‘end-state vision’that the United States sought in Afghanistan? Clinton asked. That was perhaps why Karzai pressed so hard for Israel-like guarantees – perhaps he sensed correctly that the Obama administration did not know the answer. ‘Pakistan knows what end state they want,’Clinton said. ‘They have gotten more threatening to Afghanistan recently. They are letting loose the Haqqani network. But we don’t know our end-state vision because we don’t have one. We don’t have a Pakistan strategy or a reconciliation strategy. Just words and process.’” (p. 455-6)

In his concluding chapter, Coll comes back to the US-Pakistan-India relationship: “The rising, embittered skepticism toward Pakistan at the Pentagon, in Congress, and at the C.I.A. engendered by America’s experience of the Afghan war after 2001 helped to solidify ties between the United States and India; after 2001, the two countries judged increasingly that they shared a common enemy. Yet India proved to be cautious about working too closely or explicitly with Washington in Afghanistan or the region. The country’s noisy democratic politics contained a large strain of skepticism about American power. And India’s security establishment remained wary of taking risks in Afghanistan – say, by providing lethal military aid and troops to bolster Afghan forces against the Taliban – that might confirm Pakistan’s fears of encirclement and thereby provoke I.S.I. to retaliate by sponsoring more terrorism inside India.” He also touches on the Pakistan-China relationship: “The fallout from the Afghan war also persuaded Pakistan’s leaders, after 2011, to give up on any strategic partnership with Washington and to deepen ties to Beijing. This effectively opened Pakistani territory to Chinese companies and military planners, to construct transit corridors and bases that might improve China’s regional influence and links to the Middle East. Overall, the war left China with considerable latitude in Central Asia, without having made any expenditure of blood, treasure or reputation.” (p. 663)

Pakistans consistently ambiguous stance

In contrast, Pakistan’s policy stands out as consistent, ie, as being consistently ambiguous. Coll describes Pakistan’s support to the Taleban as “just enough to keep the war broiling, while avoiding aid so explicit that it might provoke the international community to impose sanctions on Pakistan or withdraw military sales.”(p. 679) Still, while consistent, the strategy was not cohesive. While Pakistan used a variety of channels to supply the Taleban, the theory of a “rogue I.S.I.” is refuted: “American intelligence reporting on individual, serving I.S.I. case officers, who managed contacts with the Quetta Shura or the Haqqanis /…/ showed that they were clearly in the Pakistan Army’s chain of command.” However, the picture is complex and confusing: “Overall, it was very difficult to reach a judgment that ‘Pakistan’did this or that or even that there was such a thing as ‘Pakistan’s policy’, when there were so many actors and when Directorate S was engaging diverse militant groups for different purposes at different times. In the tribal areas, I.S.I. sometimes made deals with violent radicals for defensive, tactical reasons – to forestall attacks on themselves or to get military supplies through to isolated bases. Other times the I.S.I made deals for strategic reasons – to encourage the groups to enlarge their influence inside Afghanistan or to attack Indian targets there. Still other times the army attacked these same groups in retaliation for attacks inside Pakistan.” (p. 289)

Throughout the book, Washington deals with the Pakistani army, not its government, as its natural counterpart. The civilian government structure hardly figures, much less parliament or civil society. After the replacement of the Musharraf regime by a civilian PPP-led government, the US ambassador in Islamabad warned her government: “’Let’s not fool ourselves that we have a democracy’to work with in Islamabad. The United States had to work with the Pakistan army.” (p. 403)

Ashfaq Parvez Kayani

The period covered throughout the book largely coincides with Ashfaq Parvez Kayani’s position at the helm of, first the ISI and then of the army, for a prolonged tenure (2004-2013). Kayani, thus, is the central Pakistani character in the drama surrounding Afghanistan. He comes across as sophisticated (more so than some of his American counterparts), low-key and circumspect. And consistent – he never comes close to giving up on the idea that Pakistan needs to exert influence in Afghanistan to counter the threat from India, ie the old concept of “strategic depth”. He is also better at keeping his cool when bullied by Americans than Hamed Karzai, the Afghan president who keeps irritating his US sponsors. With Kayani, there is no shouting, no show-downs, just quiet reservation and, yes, consistency in the face of a host of different American interlocutors.

One of them is CIA deputy director Steve Kappes, dispatched to Islamabad to challenge Kayani after the bombing of the Indian embassy in Kabul on 7 July 2008, which, according to “American, British and other allied intelligence services”, had been carried out by “a special Haqqani unit […] under I.S.I. orders to hit hostile targets in Afghanistan, including Indian ones.”During the meeting, Kayani “was reticent, professional, a listener, but his method was to never really say yes and never really say no.”(p. 308) Kayani’s main counterpart as army chief, however, was Mike Mullen, the most senior military officer on the American side. Between 2008 and 2013, Mullen visited Kayani in Pakistan 27 times, in addition to many meetings elsewhere and frequent telephone conversations. Mullen’s hypothesis about the ISI was a layered one: “At the very top of its hierarchy, I.S.I. was a black-and-white organization, fully subject to discipline and accountability […]. In the middle the organization started to go gray, fading into heavily compartmented operations that drew upon mid-level officers, civilians, contractors, and retirees. Then there were retired I.S.I. director-generals or senior brigadiers with their own followings among militants.” (p. 322) Other American analysts “started to grasp that the Taliban forces operated on a formal rotation system – training in Pakistan, field deployment, and then rest and recuperation back in Pakistan. Pakistan Army and Frontier Corps troops along the Pakistani border were firing on American border posts to provide covering fire for the Taliban to infiltrate into Afghanistan and return – the same tactics Pakistani forces employed for Kashmiri militants along the Line of Control.” (p. 329-330)

Providing a sign that the relationship between Kayani and Mullen went quite deep, Kayani discusses even his possible prolongation as army chief with his American counterpart: “When he met Mullen, Kayani returned to a delicate subject they had been reviewing privately for months. Should Kayani engineer and accept a three-year extension as chief of army staff and de facto head of state? Mullen wanted him to extend but talked to him gently about the pros and cons. In public, the Obama administration emphasized the importance of Pakistani democracy and civilian rule; in private, it negotiated for the continuation of favorable military control.” (p. 500).

Osama bin Laden

Ironically, while these rather intimate conversations were taking place, CIA analysts started investigating a certain compound in Abbottabad, suspected of housing Osama bin Laden and his family. On this subject, Coll writes: “Kayani had been I.S.I. director for less than a year when Bin Laden set up in Abbottabad. The Al Qaeda emir and his family enjoyed support from a sizable, complex network inside Pakistan – document manufacturers, fund-raisers, bankers, couriers, and guards. His youngest wife, Amal, gave birth to four children in Pakistani hospitals or clinics after 2002. Bin Laden limited his movements, rarely leaving his homes, but he did travel on Pakistani roads numerous times without getting caught, as did his sons and wives. Amal traveled at least once on an internal flight. In one case a man dressed as a policeman accompanied Bin Laden, according to one of the women who traveled with him. It is entirely plausible that I.S.I. ran a highly compartmented, cautious support operation involving a small number of case officers or contractors who could maintain deniability. Yet there remains no authoritative evidence – on-the-record testimony, letters, or documents – of knowing complicity by I.S.I. or the Pakistani state. […]

C.I.A. and other administration officials have said that they possess no evidence – no intercepts, no unreleased documents from Abbottabad – that Kayani or Pasha or any other I.S.I. officer knew where Bin Laden was hiding. Given the hostility toward Pakistan prevalent in the American national security bureaucracy by 2011, if the United States possessed such hard evidence, it almost certainly would have leaked.” (p. 548-549)

If Kayani had indeed been unaware of Osama bin Laden’s presence in Pakistan, the same may not have been true of Mullah Omar’s presence. When US Foreign Secretary John Kerry hosted Kayani and Karzai in Brussels in 2013 to discuss the possibility of peace negotiations with the Taleban, “Kayani insisted that he did not know where Mullah Mohammed Omar was. More than two years later, the Taliban would admit that on [that] very day […] Omar died of tuberculosis in a Karachi hospital. If Kayani knew of the Taliban emir’s dire condition, he kept it to himself while working on the statement in Omar’s name. None of the Americans had a clue. Kayani continued to represent to the Americans that he was carrying messages from Omar. Afghan intelligence did have a sense that Omar might be dead, but it could not prove it to the satisfaction of the Americans.” (p. 637-638) (1)

No advice, please

After his appointment as commander of US and ISAF forces in 2009, General McChrystal flew to Brussels to meet Kayani, who had been invited to talk at a NATO meeting there. Together with Mullen and General David Petraeus, later to become McChrystal’s successor, he met separately with Kayani to discuss the situation on the ground in Afghanistan. According to one of the meeting’s participants quoted in the book, McChrystal talked of the need “to hit the center of gravity.”Kayani disagreed: “’You don’t identify the center of gravity for the purpose of attacking it […]. You find ways to unbalance it without going straight at it.’He might have been describing I.S.I.’s twenty-year strategy against Kabul. ‘This will become a revolving door in the south – you’ll go in and out, the Taliban will go in and out’.” The Americans, however, ”were in no mood to take military advice from Kayani. Petraeus became aggravated. The last person he wanted to take advice from about the war in eastern Afghanistan was a general whose refusal to tear down the Taliban leadership in Quetta or to clean their militias out of North Waziristan was itself undermining N.A.T.O. strategy enormously. Pakistan’s sanctuaries were probably the biggest vulnerability in their military plan. […] Petraeus made his irritation plain and Kayani went outside to cool off with a smoke.” (p. 369)

A long history of schizophrenia

Overall, Coll describes the relationship between Washington and the Pakistani army as being one of “a long history of schizophrenia.”(p. 314) Apart from the dependence on Pakistan for transit traffic supplying the troops in Afghanistan, a major reason for the US’s continued wooing of Pakistani generals with aid and consultations was Pakistan’s nuclear arsenal: “The [Bush] administration had ‘regular’reports of Al Qaeda and other groups plotting to steal nuclear weapons. They did not want to do anything that would destabilize Pakistani command and control.” (p. 312). The Obama administration, in spite of mounting pressure to deal more harshly with Pakistani counterparts, by and large follows the same pattern. In his conclusion, Coll states: “America failed to achieve its aims in Afghanistan for many reasons: underinvestment in development and security immediately after the Taliban’s fall; the drains on resources and the provocations caused by the U.S.-led invasion of Iraq; corruption fed by N.A.T.O. contracting and C.I.A. deal making with strongmen; and military hubris at the highest levels of the Pentagon. Yet the failure to solve the riddle of the I.S.I. and to stop its covert interference in Afghanistan became, ultimately, the greatest strategic failure of the American war.” (p. 667)


The conclusions drawn in “Directorate S” are relevant. As the Afghan war lingers on with yet new decision-makers in Washington, a number of old truths illustrated in the book remain clear. While not always immediately apparent, they are also significant factors in the 25 July national election in Pakistan:

–   Geography will not change. Since the 1947 partition, Pakistan has defined its strategic interest as having access westwards, in Afghanistan, in the face of a threat from the East, ie, India. This position has survived periods of great turmoil without any substantial change. The likelihood that this will now change, with China emerging as Pakistan’s default supporter, seems remote;

–   Relations between Pakistan and India remain at the centre of the regional conflict, which cannot be solved unless the international community works on these relations too – from both sides;

–   On the Pakistani side, the strategy should not be continued one-dimensional support to its army, which has a vested interest in maintaining its central role. The democratic process has to be supported, strengthened and used for unlocking the stalemate. “Directorate S” needs a counterweight.



(1) For a Taleban version of the circumstances of Mullah Omar’s death, see this AAN dispatch.


ISI Pakistan US