William Donovan's campaign for a national intelligence agency

William Donovan's campaign for a national intelligence agency sought to place U.S. intelligence under an agency standing above the existing bureaucracies. The book revisits that proposal after the intelligence failures preceding September 11, using it to frame a central organizational question: how can scattered knowledge be made collectively useful without eliminating the independence and specialization that produce it?

Origins and Proposal

The campaign is introduced through an account of Donovan’s argument for a national intelligence agency in the years after the Second World War. The cited history of the campaign places it within public debate over American intelligence from 1944 to 1953; a collection of documents on the postwar evolution of U.S. intelligence provides further context for that debate. The proposal was not simply to add another participant to the intelligence system. It was to establish an agency with authority and independence sufficient to stand above the disputatious bureaucracies.

The book returns to Donovan’s idea when discussing the failure to recognize the collective significance of intelligence available before September 11. The Congressional inquiry emphasized that agencies had produced localized snapshots rather than a coherent account of the threats facing the United States. Senator Richard Shelby, a sharp critic of the agencies’ work, argued that the Federal Bureau of Investigation’s decentralized structure had left information holdings fragmented into largely independent fiefdoms. In that setting, Donovan’s proposed agency represented a way to put intelligence above the competing bureaucracies and ensure that information reached the people who needed it.

The Case for Central Coordination

The proposal’s appeal rests on a real weakness of decentralization: information found in one part of a system may never reach another. Specialized agencies can develop expertise and close knowledge of their own areas, but without a means to share and combine what they know, important connections can remain invisible. In the book’s account, the intelligence community before September 11 lacked not only information sharing but a way to aggregate judgments across agencies.

That distinction matters. A central agency could make it easier for information to be shared and could address the absence of an organization-wide picture. But simply putting one agency in charge does not ensure that the best interpretation will emerge. Shelby’s proposed remedy assumed that a person or small group at the center could sift through the information and decide what mattered. The book argues that even a highly capable small group cannot be presumed wiser than the larger community whose knowledge it would oversee.

The Trade-Off with Decentralized Knowledge

Decentralization has an opposing strength: it preserves independence, specialization, and knowledge rooted in particular jobs and experiences. People close to a problem may be best placed to understand it, and different analytical perspectives can reveal significance that another group would dismiss as noise. Removing those benefits in pursuit of a single central authority could make the intelligence system more unified while narrowing the knowledge available to it.

The book therefore distinguishes centralization from aggregation. The problem, in its analysis, was not decentralization as such, but decentralization without an effective mechanism for bringing information and judgments together. Linking agencies’ computer databases could have improved information flow while allowing agencies to retain their autonomy. Yet two years after September 11, the government still lacked a single unified watch list drawing on information from across the intelligence community. The campaign for a national agency remains a consequential reference point, but the proposed institutional answer is not treated as sufficient by itself.

Significance in the Book’s Argument

Donovan’s campaign anchors a broader argument about how groups should organize knowledge. A market, corporation, or intelligence agency needs both specialized local understanding and a means of making that understanding useful across the whole system. The book’s preferred principle is not to replace dispersed judgment with a central interpreter, but to connect dispersed sources so that the collective significance of their information can be recognized.

The campaign thus illuminates a recurring organizational tension: central authority can address fragmentation, but it can also displace diversity and local expertise. In the account of the intelligence community, information needed to be shared; the harder question was how to aggregate it without making the center the sole source of judgment. Donovan’s proposal supplies the historical alternative against which that distinction is drawn.