HomeFeaturesDailyBriefingsRapidReconSpecial ReportsAbout Us

A Domestic Intelligence Agency?

This question was raised nearly two years ago in an On-line Debate at the Council for Foreign Relations. So it’s a returning question, and one that is certain to raise some controversy, both pro and con. The question is “Should the United States have a domestic intelligence agency?” And if the answer to that question is “yes,” what form should it take?

Ever since September 11th and its immediate aftermaths of the creation of the Department of Homeland Security, the passage of the Patriot Acts, and the on-going discussion over FISA, domestic intelligence, from some points of view, domestic spying and surveillance has been debated. At least according to mission, the roles of the CIA and the FBI are not supposed to superimpose or cross.

Undoubtedly, there will be arguments that the domestic security structure, now including the Department of Homeland Security, is flawed and not as streamlined as was originally hoped. It could be argued that the creation of the DHS has made it harder instead of simpler to implement security programs. It is also true that the domestic and international security and intelligence infrastructure still suffers from inefficiencies and cultural dissonance, creating difficult working arrangements among the legacy agencies.

A recently released report by the RAND Corporation discusses a framework to follow along one of two separate paths. The first would create a new agency using the intelligence components of the FBI, the DHS and the Intelligence Community. The second option would create an agency within an agency.

This all presents some very complicated policy questions and at the same time is concerning in its implications. Congress apparently asked the DHS to enlist an outside study to look at the feasibility of creating a separate counterterrorism intelligence agency, but to not make any recommendations. The report makes some very “interesting” statements:

There is no “right” balance among the many factors that intelligence affect and, even if there were, differences among individual citizens’ preferences would mean that right balance would differ from person to person. Any balance will also be unstable over time: The costs that society is willing to pay will be driven in part by how grave the threat is from terrorism. The threat will change, and, even at a single point in time, judgments about its magnitude can both differ greatly across citizens and be volatile in response to events across the body politic. Major events, such as the September 11 attacks, can cause seismic readjustments in public perceptions of threat and risk, leading to demands for major expansion in security and intelligence activities that can overwhelm other concerns—at least in the short term.

My own question is why the RAND study was contracted in the first place. Aside from the reality that the DHS has difficulties, what is the motivation behind looking for new intelligence solutions? Further, there seems to be a broadening of the intelligence gathering role of the F.B.I.


Better to be looking at this stuff now than after an attack or near-miss when we would be looking at it in a panic.

I do think that any domestic intel agency should NOT be limited to counterterrorism. You could classify some domestic gangs as terrorist organizations if you want to use the dictionary definition. And why would you be monitoring drug smuggling that aims to fund terrorism but [not] smuggling that aims to make money? Doesn't make much sense operationally.

Organizational theory has long offered conceptual tools with which to address problems of this type. Consider the concepts of 1) functional, 2) market-divisional and 3) matrix structure to organizations. Functional structure would exist in an intelligence agency where one unit does signals, another recruits spies, etc. Market division where a unit does Far East, another Africa, etc. And matrix structure would come up where the issue is whether to create a new unit to tackle an emergent problem, say, "domestic terrorism." The first two structure types are relatively permanent. The third - which would draw resources from the first two - is hoped to be temporary but in practice "nothing is more permanent than the temporary." Chains of command tend to be linear within functional and market-divisional structures. By contrast, in matrix structures each person or sub-unit is "steered" by several supervisors of differing expertise. In the textbooks such matters are addressed in detail. No, intelligence agencies do not appear to me be particularly extraordinary in the light of principles from organization theory.