September 11, 2001

September 11, 2001, becomes the decisive test of the book’s argument that a group can possess important information and still fail to make collective sense of it. The attacks exposed the weakness of an intelligence system divided among specialized agencies whose findings did not reliably converge. The book resists an easy verdict: hindsight can make scattered clues look unmistakable, yet a system that cannot connect them may lose a chance to assess a threat in time.

The fragmented intelligence system

After the Second World War, the Central Intelligence Agency initially occupied a leading place in American intelligence. Over time, however, the system became more fragmented: agencies with overlapping responsibilities operated with substantial autonomy, the director of the CIA exercised little practical supervision over them, and much intelligence funding came from the Department of Defense. The FBI, responsible for domestic law enforcement, operated almost entirely outside the intelligence community even though information about foreign terrorists inside the United States could matter to the CIA.

This arrangement had not anticipated the 1993 bombing of the World Trade Center, the 1998 bombings of American embassies in Kenya, or the attack on the USS Cole in Yemen. Until September 11, the structural weaknesses could still be overlooked. After the attacks, failures of intelligence gathering appeared undeniable: a congressional inquiry concluded that agencies had failed to recognize the individual and collective significance of information relevant to the plot and had missed opportunities to disrupt it.

Hindsight and the meaning of warning signs

The book qualifies the force of that retrospective judgment. A congressional inquiry conducted after an attack is a form of Monday-morning quarterbacking: once the outcome is known, information that seemed inconclusive beforehand can appear to have pointed directly toward it. The intelligence agencies handled enormous quantities of material, and relevance is easier to perceive when the event that gives it meaning has already occurred.

The account invokes Roberta Wohlstetter’s study of the warning failures before Pearl Harbor. Signals of an impending Japanese attack existed, but they were mixed into what Wohlstetter called “the buzzing and blooming confusion.” If it was difficult to identify the right signs ahead of an attack involving hundreds of planes and ships and thousands of people, anticipating a terrorist operation involving nineteen men was harder still. Strategic surprise, in this view, cannot be eliminated simply by insisting that someone should have recognized the clues.

Yet the prior record sharpened a different question. Between 1993 and 2001, the intelligence community had failed to anticipate four major terrorist attacks. Predicting the precise attacks on the World Trade Center and the Pentagon may have been impossible; estimating that such an attack was a concrete possibility may not have been. The issue was not only whether each clue could have been interpreted correctly in isolation, but whether a different way of organizing intelligence could have made the combined threat more legible.

Sharing information without erasing specialization

Congress emphasized the lack of information sharing. Agencies had produced localized snapshots rather than a coherent picture of threats to the United States. Senator Richard Shelby singled out the FBI’s decentralized structure, which left information fragmented among largely independent fiefdoms. His proposed remedy was to place one agency above the disputatious bureaucracies and restore the centralized intelligence structure that William Donovan had advocated decades earlier.

The book treats that diagnosis seriously but rejects centralization as a complete answer. Specialization can produce finer-grained analysis and greater expertise; independent perspectives can also reveal relationships another group misses. The missing capacity was not simply a central office with authority over every agency. It was a means of aggregating information and judgments across them. Decentralization without aggregation left agencies autonomous but did not make them collectively intelligent.

This distinction reflects a broader tension in the book’s account of decentralization. People close to a problem often possess valuable local and tacit knowledge that cannot easily be conveyed up a hierarchy. Specialization and independence can increase the information available to a group, but they do not guarantee that discoveries in one part of the system will reach the rest. The challenge is to preserve local knowledge while making it useful to the whole. In the intelligence community, the author’s proposed direction is therefore not to replace every specialist with a central judge, but to create better ways for specialized agencies to share what they know and combine their judgments.

The lesson carried into other organizations

The intelligence case leads into the book’s analysis of group decision-making. Effective collective judgment depends not just on gathering evidence but on making dissent and less widely held information available to decision-makers. The later account of the Mission Management Team during the Columbia mission gives that problem a concrete organizational form: engineers’ doubts and requests for images did not receive a serious hearing, and uncertainty failed to prompt adequate inquiry. The contrast is not that one episode is identical to the other, but that both make visible the distance between possessing information and building a shared, revisable assessment from it.

The same structural question appears in the discussion of corporations. A rigid hierarchy can obstruct the flow of information: a new product idea in one company had to traverse 223 bureaucratic paths before acceptance, while distance between executives and workers made it difficult for leaders to know whether their picture of the company matched reality. The comparison clarifies the book’s central point about September 11: information-processing problems are not solved by choosing between complete central control and isolated autonomy. A functioning group needs both room for specialized, local judgment and channels that let knowledge travel, meet competing interpretations, and inform collective decisions.