The problem is inherent to the CIA's approach to analyzing the data and intelligence it collects, argues Jeffrey Cooper, an analyst with the Center for the Study of Intelligence at the CIA.

Cooper's target is the "ingrained habits and practices of the intelligence community's analytic corps," rather than the structures and wiring diagrams the White House and Congress continue to adjust to improve intelligence.

There is little doubt intelligence analysis needs to be improved. Cooper cites not just the surprise terrorist attacks on Sept. 11 and the flawed National Intelligence Estimate that asserted Saddam Hussein had restarted his weapons of mass destruction programs, but failures from the last decade: North Korea's surprise launch of a three-stage missile, India and Pakistan's surprise tests of nuclear warheads; the emergence of globally networked Islamic extremists; and the development of a worldwide black market in nuclear, chemical and biological weapons.

Despite these massive failures, Director of National Intelligence John Negroponte said Monday there is no consideration being given to taking the analysis capability out of the CIA, and the White House remains committed to increasing the CIA's human intelligence collection and analytic capabilities by 50 percent.

"We still view the agency as the premier all-source analytic agency within the United States government," Negroponte said. "We want to build on the existing strengths of the CIA."

Cooper lays out a detailed diagnosis for the analysis malaise. It does not involve an organizational chart or new interagency panels, despite the fondness in Washington for blaming "stove-piped" information for the failure. The real pathology is in the "dysfunctional behaviors and practices within the individual agencies," Cooper states.

"Although previously cited reports on intelligence failure usually point to organizational stove-piping and technical shortcomings as the most important contributors to failures in collaboration , the sources of such failures are actually more widespread and complex and more frequently reflect shortcomings in work practices and processes, organization culture and social networks," Cooper writes in the December 2005 monograph, "Curing Analytic Pathologies: Pathways to Improved Intelligence Analysis."

The first step is analyzing the root causes of each failure. There can be many and they cascade throughout organizations, suggesting they are at least as much a fault of culture as they are individual human failures, according to Cooper.

The CIA was built up under the shadow of the Soviet Union -- a monolithic threat that manifested a vast amount of information of its threats and intentions in the deployment of missiles, massive training exercises and the weapons it invested in.

The threat the CIA faces now is quite different. Cooper asserts that it has not changed its methods of analysis from the Cold War, when its task was to penetrate and understand denied areas. But with the move away from superpowers and force-on-force conflict toward terrorism and asymmetric warfare, the challenge has changed to understanding and penetrating "denied minds."

At the top level, an organization like the CIA or DIA might find itself unable to say "no" to a customer to take on a specific tasking. It therefore shoulders more work with the same number of analysts, diluting the time and attention given to all products, and therefore degrading their quality.

Agencies and units are often unable to protect "slack" time -- down time that allows analysts to collaborate with others, "follow their nose" on leads and indulge their curiosity.

Cooper describes two kinds of scientific analysis that could be used as models by the intelligence community, deductive and inductive. Deductive starts with a theory -- for instance, that a major terror attack is being planned -- and looks for evidence to support or undermine that theory. The intelligence community is well versed in that method.

It is less comfortable with inductive analysis, which is far more time-consuming, creative, and open-ended -- and potentially far more valuable. It considers existing intelligence and attempts to tease out possible plots from those indications without a preconceived notion of what may happen. Cooper compares this to "connecting the dots" but says it is far more difficult -- more akin to putting together a jigsaw puzzle with mixed pieces and no picture of what it is supposed to look like when it is done. It's a method often used in scientific inquiry -- using creativity and intuition to create a new theory, but then applying the rigors of proof to the end product.

To get this kind of analysis, the CIA must recruit and develop analysts with expertise, independence and curiosity, Cooper said.

related report

Rumsfeld Supports Hayden, Denies Tension

by Pamela Hess
UPI Pentagon Correspondent Washington (UPI) May 10 - Defense Secretary Donald Rumsfeld energetically denied there is any turf battle over intelligence responsibility, and endorsed the nomination of an Air Force general to head the CIA.

"If you look at the debate and the articles in the newspaper and the comments that are being made, they are about theoretical conspiracies, they're about theoretical bureaucratic turf fights -- they're all off the mark," Rumsfeld said Tuesday at a Pentagon press conference. "There's no power play taking place in Washington. People can run around and find somebody who will tell them almost anything they want. But it's interesting how little facts ever get attached to any of these thumbsuckers that get printed in the press."

One of the "theories" Rumsfeld attempted to defuse is that Rumsfeld and Gen. Michael Hayden, because they have been at loggerheads on some issues, would not get along or be able to work together. Hayden is currently serving as the deputy director of national intelligence under DNI John Negroponte, was head of the National Security Agency. President George W. Bush nominated Hayden to replace Porter Goss on Monday.

Suggestions that Rumsfeld and Hayden do not get along can be traced back to 2004, when the Office of the director of national intelligence was being created. Hayden testified to Congress he thought control of the National Security Agency should be transferred from the Pentagon to the DNI. Rumsfeld disagreed.

"My view was, as someone who was appointed and nominated by the president, and serves at the pleasure of the president, that I ... thought that the Department of Defense ought to wait and see where the president comes down on that issue," Rumsfeld said.

"When the president decided not to move it over, I was in favor of that. Now, is that a big deal?" Rumsfeld said.

"In my view, (former NSA Director) Mike Hayden is a true professional, and he'll do an excellent job for the country," Rumsfeld said. "There is no tension. I've been working with Mike Hayden for five-plus years."

A second diametrically opposed theory is that Hayden, as a uniformed officer, would be unduly influenced by Rumsfeld.

Rumsfeld dismissed all suggestions that tensions exist between him and intelligence officials like Negroponte.

"We do not have issues between John Negroponte. We do not have issues between Porter Goss or George Tenet, nor will we have with General Hayden, assuming he's confirmed. We have good linkages down in the field. We have good linkages in Washington. I'm sure there are people in the middle who bulge out from time to time, but those things get worked out," Rumsfeld said.

The Pentagon has taken new steps into intelligence activities that encroach on areas traditionally governed by the CIA. Among them was the creation of the controversial Office of Special Plans prior to the Iraq war, which reviewed some of the same raw intelligence the CIA had on Iraq's weapons programs.

More recently, the Pentagon disclosed plans to access all of the CIA's databases for use in Joint Intelligence Operations Centers around the world, an effort to create an "all source" database that would allow analysts rapid access to all available intelligence. Lt. Gen. Jerry Boykin, the deputy under secretary of defense for intelligence, told reporters April 12 an early version of the JIOC had literally cut hours from the process of analyzing intelligence related to improvised explosive devices in Iraq.

Rumsfeld said these are not power grabs but improvements in intelligence gathering and operations agreed to by all parties.

"There isn't anything we're doing in the Department of Defense on intelligence that has not been worked out with the Department of State, that has not been worked out with the White House, that has not been worked out with the Central Intelligence Agency, that has not been worked out with the Director of National Intelligence," Rumsfeld said.

"It is a collegial and open process, and it is a comfortable process and it is an ongoing process, all of the articles, hundreds of articles, you're reading to the contrary notwithstanding," Rumsfeld said.

Related Links
-