Senate Intelligence Committee

The Senate Intelligence Committee appears as a congressional investigator of the U.S. intelligence community’s failure to anticipate major terrorist attacks between 1993 and 2001. The book credits the Committee with concluding that better processes could have improved the use of information already available, while questioning its proposed remedy: the difficulty was not decentralization alone, but decentralization without a way to aggregate information and judgment.

Investigation and Findings

The Committee’s conclusion, as the book presents it, was that improved processes would have produced a better result. It stressed the lack of information sharing among agencies: instead of assembling a coherent picture of threats to the United States, the agencies held separate, localized snapshots. The distinction matters because the argument does not claim that officials could necessarily have predicted the precise attacks on the World Trade Center and the Pentagon. It suggests instead that a reasonable, concrete estimate of the likelihood of such an attack might have been possible.

Senator Richard Shelby, the sharpest critic of the agencies’ performance, focused particularly on the FBI’s decentralized organizational structure. In his account, information holdings were fragmented into largely independent fiefdoms, and the intelligence community as a whole failed to put the right information in the right hands. Shelby argued for an agency standing above and independent from the disputatious bureaucracies, reviving an idea advanced decades earlier by “Wild Bill” Donovan: place one agency in charge of U.S. intelligence. The Committee’s emphasis on sharing and Shelby’s call for central authority thus answered a real organizational failure, but they also raised a further question about how integration should work.

Decentralization and Aggregation

The book distinguishes decentralization’s benefits from its risks. Specialized agencies can develop expertise, preserve independent perspectives, and interpret information differently; a clue treated as irrelevant noise by one group of analysts may reveal an important relationship to another. But decentralized systems can also leave valuable information stranded. Before September 11, the intelligence community had specialized bodies without an effective means of aggregating not just data but judgments. In the book’s formulation, there was decentralization without aggregation—and therefore no effective organization of collective knowledge.

Shelby’s proposed central agency could make information sharing easier and give one institution responsibility for the whole picture. Yet concentrating authority would also risk sacrificing the diversity, local knowledge, and independence that specialization provides. The book therefore accepts the force of Shelby’s criticism while rejecting centralization as the necessary answer: the central need was aggregation. Even relatively mechanical measures, such as linking agency databases, could have improved information flow while allowing agencies to retain autonomy. The book notes that, two years after September 11, the government still lacked a unified watch list drawing on data from across the intelligence community.

Institutional Significance

The Committee’s findings serve as a test of the difference between collecting information and making it collectively useful. A system can contain many informed specialists and still fail if there is no reliable path for observations and judgments to meet. The book’s alternative is not to erase specialization but to connect it—to preserve local knowledge while enabling the larger group to recognize patterns that no single agency could see alone.

That tension gives the Committee’s account significance beyond the immediate inquiry. Its call for information sharing identifies a central weakness, while the book’s response reframes the institutional remedy: intelligence requires coordination and aggregation, but those aims need not entail placing all interpretation in the hands of one central authority. The problem is how to join independent expertise without extinguishing the independence that makes the expertise valuable.