Ghost Wars: The Secret History of the CIA, Afghanistan, and bin Laden, from the Soviet Invasion to September 10, 2001

• Introduction
• Remarks
• Questions and Answers

Introduction

JOANNE MYERS: Good morning. I’m Joanne Myers, Director of Merrill House Programs, and on behalf of the Carnegie Council I’d like to thank you for joining us as we welcome Steve Coll to speak about Ghost Wars: The Secret History of the CIA, Afghanistan, and bin Laden, from the Soviet Invasion to September 10, 2001.

In the wake of what is currently perceived as the CIA’s historic failure on September 11, much of the postmortem on the attack has focused on the failures of communication and information-sharing among the CIA, the FBI, and the National Security Agency, and their inability to connect the so-called dots.

While it now appears that there certainly was a paucity of dot-connecting prior to 9/11, this begs the more important question as to whether there were realistically enough dots to connect.

The literature on this and related topics flows on in an unabated fashion. However, with the publication of Ghost Wars, at last we have a comprehensive and penetrating account of the inside story of the U.S. involvement in Afghanistan as to what actually happened, or more precisely, what did not happen.

This book details the CIA’s role in this unfamiliar and distant land, a relationship with foreign intelligence groups, and the rise of bin Laden and his al Qaeda network throughout the decade leading up to September 10, 2001.

Mr. Coll gathers his information from approximately 200 interviews with American, Afghan, Pakistani, and Saudi participants, as well as some pertinent documents and some secondary sources.

Additionally, Ghost Wars reveals publicly for the first time material that was contained in the sections of the Congressional Joint Inquiry Final Report on the September 11 attacks that was redacted from government classifiers.

Our speaker is a Pulitzer Prize-winning journalist. He worked out of New Delhi from 1989 to 1992 as The Washington Post South Asia’s bureau chief. While there he traveled frequently to Pakistan and Afghanistan, covering the war and also U.S. policy, including CIA covert action during that period.

After September 11, he correctly surmised that there was a story to tell about the Afghan antecedents of the attacks, a story that he, himself, had lived through, at least for a while.

Please join me in giving a very warm welcome to our guest this morning, Steve Coll. Thank you for joining us.

Remarks

STEVE COLL: Thank you so much for that introduction, which was so coherent that it has stolen the first paragraph of my remarks.

I plan to walk through the history that I’ve written and tease out some highlights and themes to show you how these threads are woven together.

The book aspires to be a narrative history of the antecedents of the September 11 attacks as they were located in Afghanistan, with a particular emphasis on the CIA, Pakistani intelligence (ISI), Saudi intelligence, and other hidden networks, particularly al Qaeda and its affiliates, and how they interacted over 20 years in the run-up to September 11.

Part of my premise is that in the long history of surprise attacks, one of the distinguishing features of September 11 is the relatively important role of clandestine networks in the preceding events. This was not a surprise attack ordered from the map room of a formal standing army; this was an attack that not only came from a stateless group but was preceded by covert action aimed at disrupting that group immediately before the attacks.

I’ll divide the history into three parts.

• The first part begins with the Soviet invasion in 1979 and ends with the withdrawal of Soviet troops: the period of the anti-Soviet jihad; the intensive collaboration between the CIA, ISI, and Saudi intelligence; the most successful covert action certainly in the modern history of the CIA and one of the most successful in the history of intelligence.

  Structures that were built up and patterns established in those years become crucial to an understanding of what follows.

• The second chapter begins with the withdrawal of Soviet troops and runs to early 1998. From an American perspective it’s a narrative of withdrawal and retreat.

  Initially there is an intense debate about whether the U.S. had continuing interests in Afghanistan and in the Sunni Islamic fountainhead that it was creating, radicalism or not.

  And then ultimately the U.S. withdraws, the Taliban rise, and a new and somewhat distorted form of American relations with Pakistani intelligence settles in.

• And finally, the third part is the period from the spring of 1998 until September 10 when the CIA, under Presidential order, returns to the region in an attempt to disrupt al Qaeda, capture or kill bin Laden, even before the Africa embassy bombings in August 1998.

I’ll spend the least time on the 1980s but I will mention two things during that period.

1) A very broad pattern was established in which the CIA subcontracted the anti-Soviet jihad to ISI. Pakistani intelligence is a division of the Pakistani army and not organized as a civilian intelligence service. ISI is generally commanded by a two-star general, and its cadres are drawn from the officer corps of the Pakistan army.

They are organized in clandestine regional bureaus. The Afghan Bureau became the instrument of the anti-Soviet jihad. These were often Pashtuns, who had language and local identity and were seconded to the Bureau for long periods of time.

Why did the CIA turn over its political program in the jihad to ISI? Partly the Agency was scarred by its experience in Vietnam, and there was a sense of no more "hearts and minds" for us. We’ll let the Pakistanis figure out who the winners and losers are politically. If they have a complicated regional agenda that is even more Islamist than we would like, so be it. We will focus on the main adversary, the Soviet Union. We won’t try to tell the Pakistanis how to run politics in the region.”

That established a pattern in which the United States and the Saudis together turned over enormous sums of money to ISI and said, “You pick the winners.” ISI chose Hekmatyar as their primary winner, and Hekmatyar, in turn, created a nexus in which al Qaeda thrived by the end of the 1980s.

2) The other piece of this period that I report on in more depth than we’ve had available before is the relationship between Saudi intelligence and bin Laden.

I find no evidence that the CIA had direct contact with bin Laden, but they were allies with Saudi intelligence during the 1980s. The formal alliance with Saudi intelligence was a check-writing operation in which the Congress would appropriate covert funds each year and then somebody would fly to Riyadh where Prince Turki al-Faisal would write matching-funds checks, which would go into the formal accounts of the CIA administered out of Washington, Switzerland, and elsewhere.

But the CIA and Saudi intelligence each had separate compartmentalized unilateral operations. The CIA’s were modest in the beginning; they became more substantial in the late 1980s. The Saudi unilateral operations involved the support charities, networks and Islamist clients along the Afghan-Pakistani frontier. These were carried on in substantial part by Ahmed Bahdib, the chief of staff to Prince Turki al-Faisal, the longtime chief of Saudi intelligence.

I interviewed Bahdib in Jeddah on the record, and he also gave an interview to Lebanese television which he provided me. He tells a story of quite an intimate relationship with bin Laden in the mid-1980s. He had taught bin Laden at secondary school in Jeddah, and their families came from the same part of southwestern Saudi Arabia.

Bahdib was Turkey’s operations guy on the frontier, a rough-and-ready character, brought a lot of cash around, worked with the charities. He paints a fragmentary sketch that suggests that bin Laden was an operational partner of Saudi intelligence during the mid-1980s. He probably didn’t require a stipend; he was very wealthy. But he does seem to have operated under contracts, bringing in construction equipment, bulldozers, other equipment to the frontier and building specific projects that the Saudis wanted for clients such as Sayyaf and others whom they particularly favored—depots, medical clinics, roads.

Here bin Laden would have collaborated intimately with the Afghan Bureau of ISI, and so the whole structure that we became familiar with ten years later—ISI, al Qaeda, and Saudi proselytizing money and official money and murky money—was constructed in the mid-1980s, and Saudi intelligence was right there on top of it in a fairly operational way.

In the second chapter of this history, after the Soviets withdrew in February 1989, I was in the region as a reporter. This was a time of intensifying although somewhat obscure internal debate and struggle over whether the United States had any further business in the region, and if so, what?

After the Soviets left, they remained engaged in support of President Najibullah, a Pashtun Afghan communist former Secret Police chief, who had started to position himself as more of a nationalist and moderate unifier and who was a Pashtun. The UN became involved in trying to construct a political settlement in the aftermath of the Soviet withdrawal.

The American Government was divided internally over whether to participate in this project. The CIA argued that it was a fool’s errand and that we had no interest to justify the risks involved.

The State Department, primarily represented by Peter Thompson, who is Second Special Envoy to the Afghan resistance, argued forcefully that the U.S. had an interest in constructing a peaceful transition, investing in stable politics in Afghanistan and in its reconstruction. He was supported by the British and Europeans to some extent and also by the UN.

The narrative of the American debate and argument over whether in effect to support the Pakistanis who sought a military solution or whether to support the United Nations is an important part of where we end up, because eventually all falls apart and politics and guerrilla war are left to the Islamists.

The United Nations and the British in particular were advocating a government that resembles almost exactly the government that we got after September 11, which was a federation of the guerrilla leader Ahmed Shah Massoud’s northern organization, royalist Pashtuns, such as the Karzai family, and exiled intellectuals.

The alternative was the Muslim brotherhood-influenced Islamist networks that were supported by Pakistani intelligence, led by Hekmatyar, who had a coup-making strategy in alliance with other extremists. Nobody quite won. Massoud blocked Hekmatyar from taking full power, and thus the civil war began.

As the war became a stalemate, Afghanistan itself collapsed, became the site of one of the world’s most severe and almost universally neglected humanitarian crises, and gave rise to the Taliban.

An important part of the second chapter of this history is the relationship between Benazir Bhutto’s notionally democratic civilian cabinet and the Taliban, between its birth in 1994 and her second ouster by the army in 1996. After a long interview with Benazir about her experience as well as many other documents to rely on, we can see that Benazir boldly, repeatedly and effectively lied to the U.S. at every level for two years.

In the first months of 1995, under pressure and reluctantly, she granted ISI a blank check to build the Taliban into a national force in Afghanistan because she had vowed not to mess up her relations with the army this time. It didn’t come at a great price to her domestic agenda, so she gave in.

Then she came to the U.S. and told every interlocutor she met—Senators, Assistant Secretaries of State, the United States Congress in a speech in a joint session—that she had no idea what was going on with the Taliban, where they were coming from, and absolutely denied that Pakistan had anything to do with them.

The Americans meanwhile were withdrawn from the region. The Agency had no real budget or program. We had no station in the country. The embassy had been closed in the late 1980s for security reasons. Afghanistan wasn’t even on the operating directive of the Islamabad station in the mid-1990s. There was no formal part of the Directorate of Operations that was responsible for Afghanistan except back in Langley, a long way from the subject.

There were a few other programs floating around the region, like Stinger recovery, that occasionally put the Islamabad station in touch with Afghanistan, but we were not reporting on the mystery of the Talibans’ rise, nor on the question of Pakistani support. We weren’t reporting on their agenda, whether they were as extreme as they appeared, their connections, their sources of funding.

One of my favorite stories is a history of the Islamabad station. Its chief in these years was an interesting character named Gary Schroen, an Afghan specialist and Farsi speaker. He came out in 1996 and was interested in building back a grip on Afghanistan out of Islamabad. But his only budget was for Stinger recovery. With about 600 Stingers missing by the mid-1990s, there was a program of tens of millions of dollars a year in covert appropriations to recover as many as possible worldwide.

After the Taliban took Kabul and established themselves as the aspiring national force, Schroen flew in to Kandahar, where they were headquartered, and Mullah Omar wouldn’t meet with him because he doesn’t meet with any non-Muslims, but he was able to meet with [Taliban Foreign Minister] Muttawakil and a couple of other senior Taliban officials. He said, “We’re here because we understand that you’ve got 53 Stingers in the basement. We’d like to buy them back, and the going rate is about $100,000 a pop. That’s pretty good money for the Taliban.”

The Taliban reacted to this offer by saying, “No, we’re going to keep them because we’ll need them for our existential war with Iran which we expect is coming up shortly, and we’re going to use them to neutralize the Iranian air force, so we’re keeping the batteries fresh.” That was the first and perhaps the most substantial direct contact between the CIA and the Taliban.

Bin Laden came back to Afghanistan in the spring of 1996 from his previously forced exile in Sudan. He initially didn’t connect with the Taliban but went to eastern Afghanistan, back to the world that he knew in the late 1980s, this world shaped by ISI and Hekmatyar and international Islamists around Peshawar and Jalalabad. But that world was rapidly being overtaken by the Taliban, and so bin Laden started contributing funds to their cause, and moved south to Kandahar where he was established by 1997.

In the late 1980s, the CIA had established a Counterterrorist Center, organized by region and by target. Over the years, they’ve had cells for almost every terrorist target that has menaced the U.S. or its allies. They established a unit to track bin Laden in early 1996 at the time that they closed the embassy in Khartoum, which had been the primary platform for this objective previously.

The unit didn’t have an operations mandate in 1997. It was a collection and analytical unit, with operations of espionage but not disruption.

They were looking for a way to watch bin Laden, as he kept issuing increasingly menacing statements in press interviews or by fax. He was part of the anti-royal Saudi opposition and had many means of communicating his agenda, particularly after he was forced out of Sudan. He blamed the Americans for this, and he first started to explicitly declare the U.S. as al Qaeda’s main enemy when he arrived in Afghanistan.

All of this attracted the attention of the CIA, and gradually they start to develop ideas to operate against him, and to acquire Presidential authority to do so.

The first operations against bin Laden were rooted in another cell of the Counterterrorist Center which was dedicated to tracking down a Pakistani migrant named Mir Aimal Kansi, who had shot up the entrance to the CIA in 1993, killing two people. He was a fugitive in the same borderlands, and the CIA unit was dedicated to returning him to the U.S. to face trial.

The Islamabad station recruited a group of paid Afghan agents who had worked with the agency in the 1980s to find Kansi, and they moved around southern Afghanistan trying to get a fix on him. Kansi eventually was captured by a walk-in informer. After he was bundled off to the U.S., where he was executed in Virginia, the bin Laden unit took over the paid agents that had been funded and equipped with vehicles and tracking devices, GPS equipment and weapons and other gear.

They began to develop plans to capture bin Laden and render him to justice in the United States. Their first set of plans were drawn up before the African embassy bombings in August 1998. They had a couple of different versions of a plan that was part of the scheme to capture Kansi and involved kidnapping bin Laden in southern Afghanistan and holding him in a cave for 30 days.

These agents, under the direction of the station, had provisioned a cave with enough food and water to last for 30 days. The idea was to pluck him off of a road or pull him out of a farm where he was living outside of the Kandahar airport, and pull him into this cave.

He wasn’t yet under indictment in the United States, inconveniently, but a grand jury investigation was under way, and the station was told that they would likely have an indictment if they needed one. There was some discussion about whether it would make sense if they did get hold of him to contact the governments of Egypt and Saudi Arabia and ask them if they would be willing to take him for trial because there was substantial evidence that bin Laden had been involved in an assassination attempt on Mubarak as well funding for the Islamic group and other militant Egyptian groups, and the Saudis notionally had 1,000 charges against him but a reluctance to level any.

This was the beginning of an intensifying three-year plan in which the CIA repeatedly sought allies and agents to either capture or kill bin Laden, and their own mandate became a subject of intense debate inside the National Security Cabinet. How much risk was President Clinton prepared to take? To what extent was the President prepared to authorize lethal operations? To what extent did he insist upon operations designed under a law enforcement framework to render bin Laden for trial?

In the end, the President never chose between those two paths, but pursued both at the same time. He did demonstrate his willingness to kill bin Laden outright two weeks after the Africa embassy bombings when he ordered cruise missile strikes against a camp where he believed bin Laden was located.

But in the aftermath, as he signed various findings and memoranda authorizing CIA covert action, he tended to authorize programs constructed with capture in mind, which meant relatively complicated planning, extraction—how can you hold him safely? How do you get a plane in, how do you get him out? How do you transfer him to the FBI? How do you maintain a chain of evidence along the way?—all of which complicated the planning.

A group of paid agents continued to work for the station. They were tasked with following bin Laden around Afghanistan in the hope that they could get a fix on his location that was confident enough to warrant either cruise missile strikes or a raid to capture him.

In early 1999, they followed bin Laden to a southern hunting camp where he was falcon hunting with a group of visiting Arab princes. The agents put up GPS devices, the Agency put up satellites and took pictures of the camp. It was a typical Persian Gulf camp, complete with generators, refrigerators and big tents, and there was even a C-130 on an airstrip nearby which the photographs allowed them to trace back to the United Arab Emirates.

This episode went on for about ten days or two weeks. Inside the National Security Cabinet, the question was, “Do we have enough information to shoot? And even if we do, can we risk killing unknown possible members of one of the United Arab Emirates’ royal families?” In the end Clinton chose not to shoot.

Many of the officers who were involved in running this operation felt very frustrated by the decision, because they had thought that this was about as good a look at him as they could get, but on the other hand, it’s easy to understand why you wouldn’t, in the middle of the Senate impeachment trial, risk destroying the royal family of the United Arab Emirates without any guarantee that you would get bin Laden; and even if you did get bin Laden while destroying half the royal family of the U.A.E., it’s not clear whether that would have been judged a success.

Another similar story that signals some of the tension and debate and unresolved framework occurred a year later after the millennium scare, which you remember was a frightening series of near-misses, at which point the Counterterrorist Center started to concentrate on its alliance with Ahmed Shah Massoud, in the hope that he would be the instrument that finally successfully disrupted bin Laden.

But again, the National Security Cabinet and the legal group that writes the very sensitive findings and instructions for these operations wasn’t really sure what it wanted to say to Massoud. In the debate about whether to take sides in the Afghan war, or to join Massoud in his war against the Taliban or to limit the relationship with Massoud to intelligence collection, the latter view prevailed.

The CIA teams that went into the Pashtun Valley to start to work with Massoud to track bin Laden were instructed that they could not fundamentally alter the Afghan battlefield by the assistance they might render Massoud. This is a fairly abstract standard if you’re a captain or major-level operator, and you’re not sure whether the satellite map or the intelligence equipment you’re giving qualifies as fundamentally altering the battlefield. There was much debate about this within the CIA, and between Massoud and the operators who visited him.

Shortly after the millennium, the CIA Counterterrorist Center picked up a report that bin Laden was visiting a camp in eastern Afghanistan called Derunta, near Jalalabad, which is relatively close to Massoud’s northern lines.

They transmitted this via the communications with Massoud’s people, and Massoud sent a team out with Soviet rockets to shell the camp and try to kill bin Laden while he was making a speech. They went by mule over footpaths to get to an overlook that Massoud had previously established above this camp.

After the team had gone, Massoud’s people wrote a message back and said, “Okay, we got your report. Our team’s on the way. They're going to shell the camp within 36 hours.”

The lawyers at Langley went nuts, and they said, “You are not authorized to give intelligence to support lethal operations; your instructions are clear: It’s about collection, it’s about identifying his movements, it’s not about murdering him with rockets from a hillside.”

So there was panic. The Counterterrorist Center was instructed to pull the team back. They called Massoud and asked, “Can you tell them to turn around and go back?” And he said, “These guys are on mules, they don’t have radios, they’re up in the hills. They’re gone; there’s nothing we can do.” So everyone sat around drumming their fingers waiting to see what would happen.

The team did shell the camp but, so far as anyone knew, they didn’t hit him or anybody else, and the whole incident passed. This sort of incident frustrated the people who worked on this day to day, because they weren’t sure what they were permitted to do.

Massoud’s struggles against bin Laden became increasingly interlocked with the Americans, and bin Laden understood that these two enemies were collaborating and that striking each of them simultaneously was by far the best tactic. Because if the 9/11 plot did not succeed, he knew that he would be at war in Afghanistan, and that Massoud was by far the most formidable tactician that he faced.

We don’t have proof that bin Laden set these two plots in motion with an idea that they would hit within two days of one another, but there’s no doubt in a somewhat broader contextual sense that he saw these two targets as interconnected. The irony is that Americans didn’t understand that their fate was intertwined with Massoud’s in such a way.

Questions and Answers

QUESTION: Do we assess the September 11 attack as an intelligence failure or a policy failure?

Why you are so cautious in referring to the connection between bin Laden and the Saudis? Even Prince Turki was very open about it. In 1992 there was a very direct cooperation for practical reasons.

Does your research provide a way to understand the underlying element of the last 20 years, that is the connection between Saudi Arabia and Pakistan?

On the one hand, you have a country which is rich in money and energy, and, on the other, a country which is rich in population and know-how. Was this underlying relationship of two complementary countries a relationship that we have not understood?

STEVE COLL: You’re right, Prince Turki has acknowledged meeting bin Laden five times, but he describes the relationship as one of two notables in Saudi society encountering each other at the Islamabad embassy.

I’m trying to describe a much more month-to-month, project-to-project—this budget, this contract—relationship.

Let me go back to your first question about whether September 11 was an intelligence failure or a policy failure. It has been frustrating to watch the commission investigations unfold in such a state of intellectual isolation. They are only focusing on the tactical warning issue, without putting it in the broader context of policy. Little bits and pieces of the broader context of policy are coming out, but the entire question has been reduced to whether there was adequate information to give tactical warning of the specific 9/11 plot, and if not, isn’t that a terrible intelligence failure?

Very robust strategic warning was being given. Everybody knew that al Qaeda was in motion with spectacular attacks. The President was briefed on that all summer. But they didn’t have intelligence adequate to predict the when and where.

Penetrating a terrorist organization like al Qaeda is a difficult task. There are occasional instances in history where intelligence services have been able to achieve something analogous. The U.S. did not have an agent inside the core al Qaeda leadership. They understood that something huge was on the way, but they couldn’t identify it.

People involved will say, “Look, we went in and sweated every day over this. We knew better than anybody how awful this was going to be.” But it’s impossible on the tactical side to stop every plot carried out by a secretive terrorist organization. Even the British in Northern Ireland—an operation in a much more favorable environment, sharing culture, geographical space, language, intimacy—couldn’t stop every attack.

There were specific avoidable errors about the 9/11 plot, particularly the failure to watch-list the two known al Qaeda sympathizers who were surveyed at a meeting in Malaysia early in January 2000.These were errors in the context of a million tasks to do that, had they been carried out, might have disrupted the plotters with the imagination and skills that delivered the particular power of September 11.

This would not have stopped bin Laden. It would not have taken al Qaeda off the field. We might still have been tolerating the Taliban and tolerating an Afghan sanctuary, but the United States might not have been attacked with such power.

QUESTION: The pessimistic view today for Afghanistan would be descent into feudalism, growing drug production, and early stages of the return of the Taliban. Do you see that as much too pessimistic? Are there more hopeful signs?

STEVE COLL: I don’t live there anymore, and so I’m reluctant to try to be authoritative about it. But I do think it’s too pessimistic. The north, the west and the Shi’ite-dominated center of Afghanistan are currently more stable and hopeful than they have been in a while. We’re talking about the east and the south, the Taliban heartland.

There is a substantial amount of instability and uncertainty that threatens the entire project. Afghans have not come to believe that the United States will finish what it began; and the U.S. and its partners have not yet delivered on their pledges, in part because doing so will take time, and only time will demonstrate both pledges.

The Taliban is attempting to reconstitute itself. It has two strategic vulnerabilities:

• Mullah Omar is still alive, and he is a cult leader, and if he is eventually taken off the radio one way or another, the Taliban as we have known it for ten years will be substantially changed.
• The sanctuary that the Taliban enjoys across the border in Pakistan, and particularly around Quetta and the border areas is very important. Pakistan has been playing a double game since September 11, feeding the Americans, foreigners, Chechens, Arabs, and others that they are seeking as part of the al Qaeda network, but tolerating and to some extent nurturing Taliban figures as a reserve card in the Pashtun political game to come.

As soon as the Quetta sanctuary is withdrawn by the Pakistani State to the best of its ability, the Taliban will begin to wither.

The south will be unstable for a long time, but the question is will the Taliban threaten the entire project in Afghanistan, or will they be isolated?

QUESTION: What was bin Laden’s game plan in executing 9/11? You said that he understood that there would be retaliation. What was he planning, and how would it unfold to his way of thinking?

STEVE COLL: What he had in mind was to shock the United States, shake its grip and at the same time change its relationship with Saudi Arabia. And he succeeded to the extent that the United States no longer has military bases and the same footprint in Saudi Arabia as it did before.

The other part of his program, however, was based on the fallacy, that he and his allies could win the war in Afghanistan, that they would bog down and defeat the United States the way they had defeated the Soviet Union.

But what bin Laden didn’t understand, in part because he had such a distant relationship from them and in part because he didn’t understand Afghanistan or Afghans any better that the Americans or the Soviets did, was just how weak the Taliban’s grip on the country really was. The Taliban were isolated, they were unpopular, they had no roots, and so they collapsed without much of a push.

QUESTION: The congressional committee dealing with terrorism in the United States had done a report back in the 1990s which showed the extent of both the domestic and the international threat of these organizations, including al Qaeda.

But the problem was that American law and executive directives from the Presidential office said that you cannot pick these people up. The FBI couldn’t even compile dossiers on them.

So until 9/11 struck and things changed with both the Patriot Act and by executive order, you couldn’t do much towards rounding up all of these suspects. The State Department has also come under criticism for having allowed many of the people on watch lists to enter the country anyway.

There are some problems in terms of our civil liberties in dealing with these matters: How do you stop them before a crime is committed when the law says you can't do anything until a crime is committed? Now that they’ve changed it to conspiracies, how will this affect what we’re doing in the future?

STEVE COLL: This is an area that I know substantially less about than the authorities governing operations. There’s no doubt that the record that I narrate shows multiple instances where the actual and perceived legal constraints on cooperation between the CIA and the FBI and between the FBI and everybody else in the world hampered the response that anybody applying common sense with the benefit of hindsight would wish for.

Faced with the specific challenge of bin Laden between 1998 and 2001, we failed to recognize the Taliban as an enemy and lacked the determination to deprive bin Laden of the extraordinary sanctuary he was operating from in in Afghanistan. From his point of view, it had certain limitations: lack of electricity, lack of water, lack of airline flights. But otherwise, it was a wonderful place to base a global campaign of the kind that he was attempting to mount, and he was able to recruit and attract with impunity thousands of adherents.

The United States wanted to narrow the problem of bin Laden to the smallest possible degree. They didn’t want to confront the Taliban as a military opponent or to confront the Taliban’s networks of support in Pakistan and in the Gulf. They didn’t want to challenge Saudi Arabia’s double game. They didn’t want to do all of the complex things that unfolded after September 11. They wanted to thread a needle, pluck bin Laden off the field or strike him with an immaculate cruise missile if they could.

QUESTION: Having watched the CIA operations in this part of the world, can you give us an informed opinion about whether the Agency has recovered sufficiently from the serious blow of dismantling in the 1970s? Has it recovered the infrastructure that was knocked out? How does it compare to sister agencies like Mossad, MI6, KGB, in terms not of the toys, but of penetration, language efficiency, understanding the local politics on the ground, and maintaining the secrecy of operations?

STEVE COLL: This is an enormous project that is still in its relatively early chapters. The CIA during the 1990s was flat in notional terms, declining in relative terms, and since September 11 has been surging in funding, personnel, recruitments, training. There are very large classes of case officers and other operators who are now only two or three years into their training and careers.

Their effectiveness and the effectiveness of the programs that have trained, developed, and deployed them is likely to be better measured in five or six or seven years. Right now they’re suffering from growing pains with many young people and retired people. People at the height of their powers in mid-life are relatively thin.

As to their capacity, they still operate best against states and are weakest against stateless adversaries. The American experience during the Cold War was in the main a confrontation between two very large states and also proxy contests primarily involving state targets.

Against China, North Korea, Russia, other governments, they are pretty good in part because of their huge technology advantage over everybody else. But as al Qaeda has demonstrated, against a stateless actor you have to array yourself in an entirely different way.

QUESTION: You mentioned that after the Soviet withdrawal, there was a debate on the extent of American engagement, and particularly the UN’s efforts to cobble together a political settlement.

What happened in the policymaking apparatus that deprived the UN of a strong, muscular, diplomatic political support that the U.S. alone could have provided, particularly vis-à-vis Pakistan? Clearly that could have spared us the situation that we have today after an expensive war and after 9/11.

Could you give more detail about this extraordinary policy failure?

STEVE COLL: During the first Bush Administration, 1989 to 1992, there was an intensive debate about whether to seek a political settlement that might leave Soviet proxy Najibulla in power or at least include him in the negotiations.

The CIA was determined to support the Afghan guerilla leaders, clients of Pakistani intelligence, who insisted that only a military solution was acceptable. They were not prepared to endorse as policy anything that would leave Najibulla in the game.

The State Department had a more nuanced view: They wanted to maintain the military program but also to participate in negotiations, to set political stability and sideline the extremists, as the policy document put it, as an explicit goal of both covert action and diplomacy.

But James Baker, who was then Secretary of State, was not prepared to argue the case at the cabinet level, and the Agency prevailed in its view. As a result, the subcontracting to Pakistan prevailed, and ISI started promoting Hekmatyar’s coup-making strategy, his attempt to rain rockets down on Kabul, and to combine with the communist extremist defense minister Tanai to take power, and other such plots.

More broadly the explanation lies elsewhere, which is to remember what was going on in the councils of Washington in those years. The Berlin Wall fell in November 1989. The hard-line coup against Gorbachev was staged in the summer of 1990. The Soviet Union dissolved in December 1991. Reunification of Germany, the end of history—everything is afoot, and Afghanistan slid steadily in those years down to the point where it never even surfaced on anybody’s lists. It was as much neglect as calculation.

QUESTION: Bush, Powell, CIA, and DOD believe that there’s a relationship between al Qaeda and Hussein, whereas most Democrats believe otherwise. Is or was there a relationship?

STEVE COLL: I see no evidence in the record of any substantial relationship. There is evidence of periodic contacts which could give us a few strands to work with.

When bin Laden was in Khartoum in the mid-1990s, Sudan was an internationale for Islamist terrorist-types, even Shi'as and Sunnis. The Iraqis showed up occasionally in Khartoum, and evidence dating back to the mid-1990s from defectors and others would indicate that they perhaps talked about training, or issues like chemical weapons on the sidelines of conferences.

I have not seen any convincing evidence that Saddam Hussein was in operational direct contact with al Qaeda after bin Laden was in Afghanistan.