July 28, 2004

A Director of National Intelligence?

Should we create the position of a national intelligence director, in other words, separate the position of Director, Central Intelligence from the position of the head of the Central Intelligence Agency?


The National Security Act of 1947 created both the Central Intelligence Agency and the position of Director, Central Intelligence (DCI). The DCI, the senior intelligence official in the United States government, also serves as the director of the Central Intelligence Agency (DCI). This arrangement created the environment that has manifested itself in the intelligence failures of the last decade. No one person can effectively serve as the senior intelligence advisor to the President, manage the intelligence community and direct the diverse operations of the CIA.

Although on paper there is a separation between the DCI and the CIA, in practice the line blurs to the point of extinction and CIA the agency takes on the trappings of the DCI staff. CIA has come to regard itself as the action arm of the DCI – a phenomenon referred to as “mission creep.” The blame rests with the DCI as well as the bureaucrats at CIA who often attempt to wear the more senior DCI mantle.

For example, each major combatant military command has a DCI Representative, invariably an officer from the CIA. That officer is accredited to the Commander of the military organization. The command has an intelligence director – the DCI Representative should be accredited to the command’s Director of Intelligence (J-2).

I can attest from personal experience that the DCI Representative often refuses to deal with the J-2. A case in point: during operations Desert Shield and Desert Storm, the DCI staff in Washington forwarded the National Intelligence Daily, the “NID,” to the DCI Representative. The NID is the consensus of the entire intelligence community, cleared by CIA, NSA and DIA, and the military services if applicable. The DCI Representative would allow the Commander to read the NID, but not share that same intelligence report with the command J-2. Then the Commander would attend the morning intelligence briefing prepared by the command intelligence staff (without benefit of the DC-area analysis contained in the NID). Invariably, the Commander would blindside the command director of intelligence with things like, “Well, Washington doesn’t agree with that assessment, General….” This is inexcusable.

On the human intelligence (HUMINT) operational side, the DCI is supposed to control the national source registry, but in reality, CIA controls it. The national source registry is the mechanism by which potential intelligence assets are registered, vetted and exploited. Since the CIA controls the registry, they can (and do) determine which organization controls which assets. While CIA sees all source registrations, other agencies do not. Specifically, when the Defense Department spots and attempts to develop a potentially valuable asset, CIA has the final decision on the disposition of the asset. Conversely, when a CIA station spots a potential asset, DOD never even knows of the asset. There are documented cases, and again I speak from personal experience, when CIA surprisingly declares “prior interest” in someone you have attempted to register, or after registration, arbitrarily assumes control of the asset – in the name of the DCI. Again, inexcusable.

The DCI is charged to approve all clandestine and covert operations in the community. In reality, again, this is done at CIA. So no matter what the Defense Department organizations want or need to do, CIA has to approve it. This is way too much disapproval power for an organization that has no responsibility to the Secretary of Defense, combatant commands, or deployed American forces. Many times, DOD has had reasonable proposals that CIA claimed would interfere with ongoing CIA operations. CIA is not obligated to coordinate, explain or even acknowledge their operations with DOD, although DOD is obligated to coordinate all intelligence proposals with CIA (not the DCI, the CIA). Hopefully, with a real national director, the entire community would all be on the same playing field. Operations, assets, coordination between agencies would be transparent, and likely more effective.

National HUMINT Collection Plans are developed by the DCI. Having participated in the development of these plans, it is easy to see the dominant role of the CIA. The community has a meeting to determine collection strategies for HUMINT collectors. Although many agencies provide input, the draft that is circulated for review is always that authored by CIA. Rather than have your input considered on its face, you have to disprove the CIA position to get yours in. CIA, as one agency, should have its input considered equally with the DOD agencies.

On the analysis side, CIA assumes a dominant role, again taking on the mantle of the DCI. The DCI organization responsible for the National Intelligence Estimates (NIE) is the National Intelligence Council, the NIC. With very few exceptions, these are CIA officers, working in CIA offices, using CIA databases and CIA staffs. For all practical purposes, they are nothing more than a branch of the CIA. However, the NIE is a “national” intelligence estimate, not a “central” intelligence estimate. Opinions other than those of CIA area rarely considered.

That brings us to a major problem – information sharing inside the intelligence community. Too often, CIA compartments information so that other appropriately cleared analysts never see the information. As a result, only CIA analysts have access to some of the best information available, although they may not be the most qualified on the subject, or have the greatest interest in the subject. For example, the Air Force-run National Air Intelligence Center analysis of Iraqi unmanned aerial vehicles was hindered because CIA did not share all available information. There was no reason the Air Force analysts, all possessing the required security clearances, were not provided the information. Not only is this inexcusable on the surface, but when dealing with military capabilities, this potentially puts American forces more at risk.

Creation of national centers is a good idea, but they should not replicate the existent DCI centers. The DCI centers are nothing more than thinly disguised CIA-lead analytical or coordination centers, fraught with the same compartmentation problems in the community. The new centers need to be true community-wide centers, with operational tasking authority over all intelligence community assets and responsibility to the national intelligence director, not the CIA.