Comments & Responses

Comments & Responses

Mini Teaser: 

by Author(s): Pat RobertsJohn J. MearsheimerGeoffrey Peter Hugh Loane


Intelligence Reform

Not long ago, during one of the many terrorism alerts we have endured in Washington, it was made clear that even on matters related to the terrorist threat to the U.S. homeland, our intelligence agencies still refuse to share information. Two important intelligence agencies had different assessments concerning the seriousness of a particular terrorist threat. These differing assessments came about not because of different tradecraft or aggressive alternative analysis, but because one of the agencies did not have access to all the relevant information concerning that threat.

Unfortunately, this situation is not unique. The examples sound like this: Two intelligence agencies, both working on terrorism, will not share intelligence information with each other, even though both agencies are made up of patriotic Americans with the same top secret clearances. This failure to share is glossed over with arguments about security and "need to know", but it mostly comes down to petty bureaucratic politics.

Over the years, the intelligence community has evolved into a system of "have" and "have not" agencies. The agencies that collect the intelligence "have" the intelligence, and the agencies that perform mostly analysis "have not." The intelligence collection agencies enjoy significant clout in our government through their control over the flow of information.

Key terrorism analysts in our intelligence agencies must be given access to every single piece of relevant intelligence data concerning threats to the homeland. When analysts have uneven access to information, policymakers can't tell whether two analysts disagree because one of them has done a better job of sorting through the information, or if they disagree because one just doesn't have access to an important piece of information.

To make matters worse, analysts in the "have not" agencies often don't know what they don't know. What this means is that these analysts often don't even know the right questions to ask in seeking out information because they don't know that the information exists. The intelligence community is not a "level playing field" when it comes to information access.

Another illustration of this sad state of affairs is the National Counterterrorism Center (NCTC). If you visit the intelligence watch center, and look under one of the analysts' desks, you will find an amazing collection of ten or more computers, each with a different connection back to one of our intelligence agencies. In 2005 this bailing-wire solution is the only way we can bring together our vast holdings of intelligence data.

Why is it that this strange arrangement exists? Why don't our intelligence agencies work off of one system in which a properly cleared intelligence analyst can instantly search the data holdings of the entire intelligence community from a single computer?

Simply put, we have this arrangement because knowledge is power. Absent some outside pressure, bureaucracies usually only share valuable information when there is something in it for them. Accordingly, every time the 15 different intelligence agencies have met over the last decade or more to discuss intelligence-sharing, they work for 15 different bosses, and no one person is in charge to force them to make the compromises necessary to link our intelligence networks.

Those with top secret clearances at one intelligence agency can't imagine how those with top secret clearances at another intelligence agency could possibly be trusted to protect their precious intelligence data, so the networks and databases never get connected.

We count on the analysts at the NCTC to detect terrorist threats to the homeland. It is preposterous that those analysts have to toggle between ten to 15 different computers to find what they need. They should have instant access to every piece of data the intelligence community collects through a single search on a single computer. We need to change the way business is done.

And as bizarre and inefficient as that arrangement at the NCTC is, I'm afraid that's not the worst of it. At least the NCTC's terrorism analysts have access to the relevant computer networks.

If you are a terrorism analyst who works outside the NCTC, you don't get access to all of those different computer networks. And for other targets that are no less important, like North Korea, China or the proliferation of nuclear weapons, none of the analysts, no matter where they work, has access to all the intelligence data we collect on those targets.

If we are to move toward an intelligence community where all analysts, no matter what they work on or where they work, have the full benefit of every relevant piece of data the intelligence community collects, we must reject the concept of "information-sharing" in favor of what I call "information access."

I believe that information sharing is a limited idea that falsely implies that the data collector is also the data owner. The concept of information sharing relies on collectors to push information to those analysts who they deem need it. We need new thinking on this issue. While we must continue to protect intelligence sources and methods, cleared analysts with a need to know should be able to use a single computer to pull information from all intelligence databases, without waiting for any one agency to deem them worthy.

Essay Types: Essay