President Obama: Ask THESE Questions

By Marc Ambinder

As President Obama meets with his intelligence cabinet today, here's hoping that he asks a couple of pointed questions:

-- What does the National Counterterrorism Center actually do?  Why is it staffed with people who are very junior?

-- Is there too much intelligence? Whose job is it to act on hunches and run them through the system?

-- Who is responsible for rapidly upgrading the intelligence community's IT infrastructure?

-- How can the White House pressure Congress to do a better job of oversight?

-- If an ambassador, or a policy maker, senses something suspicious, why shouldn't he (or she) be held accountable for failing to flag said item as suspicious?

-- Why am I meeting with so many different people to figure out what went wrong? Does that give me an indication of the problem itself?

For the sake of the future of U.S. intelligence, let's ban the dot connecting metaphor. The Christmas Day terrorist attempt was fundamentally not about a failure to connect the dots. The dots were already connected. An elaborate portrait was drawn, with chiaroscuro shading and an antique frame. All that was needed is a policy maker to act on the information -- someone whose job it is to look at the pictures and decide how to change policy accordingly.  This could have been the administrator of the FAA. It could have been the chief of station in Lagos. It could have been Mike Leiter, the director of the National Counterterrorism Center. It could have been a senior FBI agent working counterterrorism. It could have been a senior SIGINT analyst at the National Security Agency. It could have been a regional security officer in the Diplomatic Security Service. It could have been Janet Napolitano, or someone working for the TSA's intelligence division. All of them had plenty of information. No one felt compelled to shake the tree.

"In some ways, this case is worse than 9/11," said Amy Zegart, the highly regarded UCLA researcher of national security. When an FBI agent suspected the presence of two Al Qaeda terrorists inside the United States, she made the decision to call the FBI's Bin Laden unit in New York and impress upon its head that the information in her possession was important. "In this case, no one seems to have even picked up the phone."

Before 9/11, Zegart says, agencies refused to share information even when they were sitting in the same room. After 9/11, information sharing has become routine -- too routine -- paper shuffling. What good is a State Department cable if its originator does not believe it is important? Why have continually updated terrorist watch lists that don't trigger anything in particular? When the National Security Agency, which routinely intercepts about 200,000,000 communications per day, actually finds a diamond in the rough -- chatter about a Nigerian who is going to attempt an attack on an airplane -- what happens when everyone knows about it but no one feels compelled to act upon it?

One person who advises the administration on intelligence issues compared the process of producing intelligence nowadays to Gulliver, constrained by 1,001 nodes for dot collection. Fusion, for the sake of fusion, provides policy makers with no advantage.

The National Counterterrorism Center is responsible for all-source analysis. As envisioned by the 9/11 commission and then by Congress, its highly trained, experienced analysts would fuse data from all 16 intelligence agencies and produce "products" that policy makers would then use to make decisions. Critics worried that the NCTC's establishment would displace the decision makers themselves; they would rely on others to connect dots and tell them what to do. Centralizing intelligence analysis when fighting a broadly distributed and evolving enemy did not make sense; separating the function of high-level counterterrorism analysis from intelligence collectors -- the folks at the CIA's Counterterrorism Center -- was like a blueprint for a game of telephone.

There have been worthy initiatives from the ODNI -- science-based analytical guidelines, new joint-duty requirements for managers -- heck, former DNI Mike McConnell even managed to eliminate one very large and not very useful technical collection program.

But the DNI is not yet the enabler, the facilitator, of intelligence collection and production that Congress intended. And you can blame Congress that. They've lost focus on intelligence reform; the professional staffs on the intelligence committees are talented, but many of their bosses don't have the time, or interest, to learn about intelligence. And committee term-limits in the House make it difficult for members to specialize.

Congress bears responsibility for making sure that the information spigot is filtered; for making sure that the departments and agencies it creates are working properly; for making sure that the DNI has the power and budget authority to back up the statutory obligations; for being forward-looking, and not just reactive. When Congress holds its hearings on the Christmas Day plot, here's hoping that it asks questions about its own oversight functions.

The intelligence community's information infrastructure remains out of date and duplicative. A bevy of talented chief technology officers and CIOs are trying to figure out what to keep and what to ditch, but it is an "absolute disaster" when compared to what private companies routinely build, said AEI's Norm Ornstein, who has been worried about the issue for years. Again, the problem isn't information sharing. It's that human beings can only do so much. Why it is that the TIDE database isn't regularly, instantly, cross-checked with the Border Patrol's APIS database, or the State Department's Visa and Passport Watch List -- why the computers can't talk to each other and flag potential problems instantaneously -- cannot be blamed on a "need to know" culture. Google can do this; FedEx DOES this. Why can't the IC? Blame leadership for failing to prioritize information technology.

Inevitably, because he is the director, Adm. Dennis Blair is receiving the lion's share of the blame. Blair is not popular at the White House, and he has picked several battles with the CIA that turned out to be battles he could not win. (First rule of institutional politics: don't pick battles you can't win!) Really, though, any DNI would be in Blair's position. He cannot de-layer the intelligence bureaucracy -- he hasn't the power to do so. He can't force the CIA to better integrate its operations with the national intelligence strategy.  Hiring "better" people won't work; there are great people throughout the intelligence agency who do incredible work. The problem, as one former CIA case officer explained to me, is this:

"Journalism is a lot like spying since it involves gathering information and talking to people. Imagine if you had to relay every word you published at The Atlantic through six or more people, each of whom edited what you had written. Some of them like to sit on things for a while, others like to wait overnight in order to sleep on it. Some are out for the afternoon and can't get to it until the next day. None of these people actually produce anything themselves, they just edit your work. Your work would eventually be published, but it would take a long time, and in some cases the final product would no longer resemble what you had originally written."

One wagers that, if there is reform to be had, Congress will insist that it consist of a new layer of bureaucracy. A Senior Overseer of Intelligence Prioritization, reporting directly to the Deputy Director of National Intelligence for Foreign and Domestic Intercepts and Threat Conditions. Why not solve this whole business with a few new business cards? It's easier than getting people to accept responsibility.

"It does not appear from public comments that the administration is forward leading on intelligence improvements," Zegart said. "They're in a reactive posture."

This article available online at:

http://www.theatlantic.com/politics/archive/2010/01/president-obama-ask-these-questions/32980/