The twilight war

Ten years after 9/11, what have we learned? An exclusive extract from the new afterword to the 9/11 Commission report
August 24, 2011
In tribute: since 2003, the anniversary of 9/11 has been marked by twin shafts of light, projected upwards from near Ground Zero

The 2004 report by the 9/11 Commission, or the Final Report of the National Commission on Terrorist Attacks Upon the United States, was a landmark in government analysis of a catastrophe. The report by the bipartisan commission also became a bestseller; unusually for a government report, it was praised for its literary quality, and was a finalist in the National Book Awards for non-fiction.

Philip Zelikow, the executive director of the commission, has written a new afterword to the report for the tenth anniversary of the attacks. An exclusive edited extract from the new section, which he has called “The Twilight War,” runs below. At the start, he explains the commission’s unusual approach. “Above all, we were most interested in understanding what happened and why. As citizens we are conditioned to judging public officials as having been good or bad—with every bit of information mined to reinforce such a judgement. But a good historian’s job is different: to understand the whats and the whys. If there was contradictory or inconclusive evidence, we laid that out too and then we left it to the readers to be the jurors.”

In the rare cases when a historical event, especially a traumatic event, stirs emotions on a massive scale, touching many millions of people, it enters popular culture. Great numbers of people soon form beliefs about what happened and why.

It is difficult to overstate the significance of these narrative beliefs. The December 1941 Japanese surprise attack on Hawaii, centred on the American naval base at Pearl Harbor, was such a thunderbolt. It shaped many images and beliefs, including about the necessity of war, the vulnerability of America, and the character of the Japanese enemy.

People will try to make sense of events in ways that fit their prior understanding of how the world works. The duty of government is to organise the best possible contribution of factual understanding to the inevitable river of information and argument, so that this river might leave behind a healthier silt of beliefs. A thorough self-assessment is also necessary if governments want to understand and mend any of their own failings.

Of course it is not easy to conduct such a self-examination. Not least, after a major shock, relevant officials are utterly preoccupied with urgent demands in the present. President George W Bush and Congress (and the government of New York City) were slow to muster this self-examination. An accurate account only emerged years later, due to the commission’s work. The historical work of the commission about what happened before and on 9/11 does not yet need any significant amendment.

So far, several studies of new evidence have confirmed the commission’s conclusion that the flirtations between Iraq and al Qaeda during the mid-1990s did not end up in a collaborative operational relationship linked to the 9/11 attack or any others. On Iran, the commission did find some interesting but inconclusive circumstantial evidence, and asked the US government to examine further Iran’s pre-9/11 relationship with al Qaeda. As far as I know, this has not happened.

In 2004, the commission wrote that “countering terrorism has become, beyond any doubt, the top national security priority for the United States.” It may no longer be the top, but it is still a top priority. Measured in constant dollars, spending on national defence in the last ten years has gone up about 67 per cent.

When the commission first issued its report, the United States was catching its breath. Now, in the aftermath of Osama bin Laden’s death, the threat of violent extremism has become a matter of routine. Rou-tine, but not ordinary.

For those seeking historical analogies, the most suggestive one is the rise of violent anarchist groups in the late 19th and early 20th centuries. Centered in Spain, France, Italy, Germany and Russia, the movement spread into the US and around the world, mainly through expatriates and immigrants from those homelands. Inspired by a few charismatic leader figures, exalting the “propaganda of the deed,” anarchists carried out sensational attacks, including the assassination of a US president (William McKinley) and several other leaders. Looking back, historians—like contemporaries at the time—see this movement as a symptom of displaced and alienated people in societies undergoing wrenching change, reacting to traditions of harsh, repressive rule. In no case did anarchists seem likely to establish governments of their own. The movement was instead an outlet of fury, ominous as a precursor to movements that would more effectively exploit such anger and such dreams.

What the commission called the “second enemy,” the scattered fanatical adherents drawing inspiration from a shared creed, is no longer “gathering.” It has crystallised. Cells of actual and would-be Islamist terrorists have drawn the attention of intelligence and law enforcement agencies on every inhabited continent. The most deadly attack on US soil since 9/11 was actually carried out by a solo fanatic, an army psychologist inspired by Islamist propaganda to murder 13 people at the base where he was stationed.

The danger of global Islamist terrorism is greatly reduced from what it was on 9/11. The core al Qaeda organisation in Pakistan is probably down to no more than a few hundred reliable operatives. Richard Reid, the so-called shoe bomber, was overwhelmed by airline employees and passengers as he tried to ignite explosives in his boots on an airliner preparing to land just before Christmas in 2001. Eight years later, on Christmas in 2009, trying somehow to attack the United States, the best al Qaeda’s affiliate in Yemen could do was to repeat the Reid operation, this time with an operative who was overwhelmed by airline passengers and employees as he tried unsuccessfully to ignite explosives in his underwear.

But the risk of an attack is not zero—one could still happen at any time. There have been other recent attempts—including a 2009 plan to bomb New York City subways (akin to the Madrid railway bombings of 2004 and the London bombings of 2005) and a 2010 attempt to explode a car bomb in Times Square in New York. Any attack will be publicised sensationally.

Any president who downplays the danger, trying to put it into a more normal proportion, invites humiliation if there is an attack. He doesn’t dare boast about success. If there is no attack, the reassurance invites an unwanted dulling of concern. The paradox is that efforts to rightsize a reduced risk seem… too risky.

Yet the most serious threats are posed by a relatively tiny number of people, fewer in number and less well organised than the production crew of any one of Hollywood’s larger films. A handful of deluded zealots derive most of their power not from their strength or the power of their ideals. They get their power from us—from our society and our culture. To quote one of our former commissioners, Richard Ben-Veniste, perhaps Americans might best “combat terrorists by taking away the terror.”

The paradox is genuine. The death of bin Laden is, however, one of those potentially catalytic moments that open minds to a fresh narrative. Contemporary societies will remain vulnerable to the abilities of even a few people to do terribly disruptive things. That feature of our age is not unique to the danger posed by Islamist fanatics. A principal function of 21st-century government will be to manage a healthy adjustment to the kinds of risks that are endemic to this generation.

There will be failures. The supreme measure of a mature, professional institution—or government—is how it handles failure: its capacity for honest self-examination and thoughtful accountability.


The commission’s report put great emphasis on eliminating terrorist sanctuaries. One reason the 9/11 attacks succeeded was because the perpetrators not only had a sanctuary in Afghanistan and Pakistan but could safely stage and train inside the US. The US is now a much more difficult place in which to deploy and prepare unnoticed.

But the focus on sanctuaries draws the power of the US and its allies into the chaotic areas of the world. As the report put it: “In the 20th century, strategists focused on the world’s great industrial heartlands. In the 21st, the focus is in the opposite direction, toward remote regions and failing states.”

Unfortunately, even as the commission was issuing its report, one of those failing states—US-occupied Iraq—had become another major base for al Qaeda. During the war’s peak years, between 2003 and 2008, it was a principal recruiting vehicle and killing field for al Qaeda, which launched attacks against Americans and other foreigners, although most victims were Arabs and fellow Muslims. Al Qaeda in Iraq is currently a broken force, repudiated even by most Sunni Arabs, although significant remnants still persist.

While the war lasted it was a source of consternation for many counterterrorism experts who believed that, at best, it diverted attention and resources away from what they considered the main theatre of conflict, in Pakistan and Afghanistan. That attention and resources were diverted is undeniable. Whether the conflict in Pakistan and Afghanistan would have been amenable to more attention and resources is doubtful, especially since the US’s interventions there have been beset by more fundamental problems of strategy and policy design.

In Pakistan, the current government’s double-game does not seem to be sustainable for the long haul. For the sake of continued effective operations against al Qaeda and open supply lines to Afghanistan, the US has tolerated Pakistan’s duplicity. Rather than recoiling from the whole country in disgust and giving up on it, the US government bet significant sums on a programme of assistance first to Pervez Musharraf’s government, and then to Asif Ali Zardari’s. Successive administrations and Congresses have hoped that, somehow, such assistance might raise the odds of a Pakistan that would be part of the solution to violent Islamist extremism rather than an accelerant to it.

It is more and more difficult to be optimistic about this bet. The problem is not that most Pakistanis prefer extreme Islamist ideologues, as those political parties have never done well in Pakistani elections. But the extremist revolutionaries are well positioned to aggravate and exploit the longstanding diseases—like weak public order, poor education, fragile civic institutions in a country of patron-client governance, sectarian conflict, and extreme inequality between rich oligarchs and desperately poor masses. These conditions seem to be bringing Pakistan closer to state failure. Politicians speaking out for tolerance are murdered; other government officials are afraid even to attend the funerals. The obvious double-dealing in matters large and small, coloured by a sense of deep corruption and selfishness among the Pakistani ruling elite, is more corrosive than ever.

In 2010 the US made a renewed effort to base its aid programme on a more durable compact of trust rather than a merely transactional basis, without evident success. Although the aid programme may not be especially efficient or effective, it is well intentioned and does some good. Nonetheless, anti-Americanism is the popular mode among Pakistanis, especially since America routinely carries out missile strikes in their country, an intrusion that Pakistan’s leaders will barely acknowledge, much less defend. That said, the strikes are not the principal cause of the anti-Americanism. Like Pakistan’s obsession with the threat from India, the general anti-Americanism is a symptom of profound internal discontent. These frustrations find an outlet in the contemplation of lurid phantom enemies paired, in the case of America, with a historical narrative in which the US has always been a faithless friend.

If Pakistan has any friends its government regards as faithful, China and Saudi Arabia probably head the list. The United States should compare notes with both of those governments about Pakistan’s prospects and what can be done.

Once a country enters a process of terminal decline, its governing institutions can die. Pakistan’s condition may not be terminal. Yet it is close enough that other states, including the US, should prepare to contain and mitigate the spillover dangers.

In Afghanistan, the commission’s recommendation for a “redoubled effort to secure the country” happened, although it went into top gear only four years later. Unfortunately, the effort to “disarm militias and curtail the age of warlord rule’” never really happened. Plenty of money was allocated from 2003 on, in relation to the small Afghan economy. Most went to foreign agencies, much of it misallocated or spent on foreign contractors. Meanwhile the institutions of the Afghan government were starved for resources and found many of their best people hired away by the foreigners. Between 2004 and 2006 the government of President Hamid Karzai allied itself with key warlords in an attempt to create a powerful central ruler linked through patronage networks to these power brokers. This design was backed by the US.

The result was a backlash in which the Afghan Taliban fed off popular fury toward Karzai and his corrupt allies. By 2008–09 the insurgency was gaining strength and spreading into non-Pashtun regions of the country.

The commitment of more US troops has made a difference, pacifying some of the worst districts. As of summer 2011, however, no viable strategy of Afghan governance has emerged that would seem likely to keep the lid on if the US releases its grip.

The Taliban is not a terribly formidable organisation. Its adherents are proudly responsible for most of the killings of innocent Afghan civilians. It is not well funded or well armed, at least in comparison to the forces organised by the Afghan government and bankrolled with American money.

Yet “Afghanistan” is the veneer covering a collection of local rulers and local fights, where the Taliban is usually more effective than the equally illiterate but more reluctant police and soldiers.

The Americans can be effective militarily. They also wield influence as a source of aid and enrichment. But they are a heavy, noisy presence. Culturally, they are utterly alien.

Over the years, the Americans have worn out their welcome in much of the country. From the soldiers wearily putting on their helmets for another patrol, to the contractors putting up a new dining hall at some T-walled installation, to the accountants poring over invoices in a double-wide trailer, all the toil and trauma, both shocking and dreary, is animated by the memory of 9/11. The United States finds itself running a protectorate in a country that is as far away and difficult to supply as any on the planet. It does this out of fear that, if it lets up, the conditions that produced 9/11 will occur again.

But history does not play reruns. Bin Laden is dead. Even if American forces do relax their grip, a return to pre- 9/11 conditions is not probable. The anti-Taliban portions of the country are much stronger than they were in the 1990s. They would have access to international support. They would probably win many local fights. In any case, the US will keep a careful watch on developments in Afghanistan, find receptive local partners, and retain bases from which its forces can operate.

Besides Pakistan and Afghanistan, al Qaeda has operated in Yemen since its beginning. All three—Pakistan, Afghanistan, and Yemen—illustrate a more general strategic issue. In its effort to get at al Qaeda and its allies, the US and its allies will forge working relationships with local authorities. But the US should not assume that massive interventions of ground forces can directly pacify and police such areas of the world. Aside from issues of cost and feasibility, the apparent intrusion of US forces in Arab and Muslim communities is an enormous recruiting point for al Qaeda and other Islamist groups.

If these premises are right, US policy has unavoidable limits and a critical risk: that the local partnership will break down, with no good substitute. Flooding the zone with US troops cannot be the default alternative.

The US must have a strategy that copes with this risk. That might mean investing in a capability to target enemy combatants without nearby bases, to support foreign policing and justice, to control- international travel and shipping, and to spot illicit financial transfers. It might also mean working to build a global political and economic system that rewards those local governments that choose to hunt terror groups at war with the US.


The next ten years will probably be very different from the last ten. The US cannot police all the wilderness areas of the world, dispersing its might and resources into every swamp and desert. Needed now is a more fundamental reconception of policy that can take the offensive against the most dangerous groups, rely principally on local allies, and have regional strategies to shape conditions and contain dangers where local partners falter. The US and its friends must have an overall policy that hedges against bad outcomes in South Asia, especially in Pakistan, that their best efforts may not be able to prevent.

In the Islamic world, anti-Americanism, or suspicion about America, remains high. But US forces are leaving Iraq. A new US president has reintroduced his country to the world, with Barack Obama’s 2009 address in Cairo. The US role in the Muslim narrative is moving away from centre stage.

Al Qaeda has now engineered so many indiscriminate killings of Muslims, especially in Iraq, that its already marginal standing has eroded even more. Arabs may not be pro-American. But they are not pro-al Qaeda either.

Arab attention is riveted more and more on the future of their own societies. From north Africa to the Persian Gulf, revolution has convulsed the Arab world. These revolutions were not sponsored or led by Islamists. The general push was for an end to tyrannical, corrupt dictatorships, with a mainly secular agenda promoted in its place: more law and justice, more choice, more jobs, more human dignity.

The revolutions may not turn out happily. But they are a healthier response, opening a richer range of possible out-comes than recourse to cells of alienated young people seeking martyrdom.

Outsiders can frame choices without trying to steer them. They can also influence choices, at the margin, as new Arab leaders develop a message of opportunity to go with the promise of more justice and dignity. But outsiders should not put themselves at the centre of these choices. They should avoid being cast as saviours or scapegoats.

I judged that how the US handled the detention and treatment of Muslim captives—a set of issues evoked by words like “Guantánamo” or “torture” or “Abu Ghraib”—was a problem doing more damage to US interests in the world than any other.

Five years of experience with the new rules for interrogations has not revealed any notable hindrances in the conduct of operations, and the new procedures have made it easier to conduct operations with some of America’s most important friends.

No one knows how the current period of turmoil will turn out. The situation in Pakistan and Afghanistan is bad. The Arab revolutions of 2011 are in midstream. In Iran, a popular revolt in 2009 was crushed, but its embers are still red-hot.

Extreme political movements will try to exploit the unrest. Israel remains unpopular everywhere in the Muslim world; the US association with Israel will not go away. But the principal focus of attention is back on choices to be made closer to home. The streams of Arab and Muslim political life are flowing down some new channels. More productive issues are on the agenda than those of the 1990s, when inchoate resentments displaced hatreds onto the “far enemy” in America because the problems at home seemed so impervious to change.

The problem with intelligence

Intelligence has a mystique, so the public and policymakers often expect too much from it, and not enough. Conditioned to expect predictions, they should expect insight. Conditioned to expect startling finds, they should expect the patient construction of incomplete mosaics.

The core 9/11 problem was not the mere failure to provide tactical warning of the attack. In addition to certain specific operational slips in case management, the real problem was the intelligence establishment’s failure to apply its own best practices for how to provide warning of attack. This was a craft laboriously honed for nearly 60 years, and then not systematically applied to the enemy that everyone knew was most likely to attack the United States. The core indictment of the commission that examined the intelligence community’s analysis of Iraqi WMD capabilities was similar: the failure to apply the best practices that were already developed and understood within those very institutions.