Strafor.com: An insiders' view of the inside

George Friedman writes an insightful piece titled simply, The Intelligence Problem. Not that the good Doctor does not fail to take advantage of a little free PR in his own newsletter, but his analysis of the national Intelligence Community, from the perspective of a private IC analyst, stands on its own.

Stratfor: Geopolitical Intelligence Report - May 9, 2006

The Intelligence Problem

By George Friedman

Porter Goss has been fired as director of the CIA and is to be replaced by Gen. Michael Hayden -- who is now deputy to Director of National Intelligence John Negroponte and formerly was director of the National Security Agency (NSA). Viewed from beyond the Beltway -- and we are far outside the Beltway -- it appears that the Bush administration is reshuffling the usual intelligence insiders, and to a great extent, that is exactly what is happening. But there is more: White House Chief of Staff Joshua Bolten, having decided such matters as who the new press secretary should be, has turned to what is a very real problem for President George W. Bush: a vicious battle between the White House and the CIA.

The fight is simply about who bears the blame for Iraq. The White House and the Defense Department have consistently blamed the CIA for faulty intelligence on Iraqi weapons of mass destruction and over the failure to predict and understand the insurgency in Iraq. The CIA has responded by leaking studies showing that its intelligence indeed was correct but was ignored by Bush and Defense Secretary Donald Rumsfeld. The truth, as usual, lies somewhere in the middle. There certainly were studies inside the CIA that were accurate on the subject -- but given the thousands of people working for the agency, someone had to be right. The question is not whether someone got it right, but what was transmitted to the White House in then-Director George Tenet's briefings. At this point, it really does not matter. There was a massive screwup, with plenty of blame to go around.

Still, it is probably not good for the White House and the CIA to be in a vicious fight while a war is still going on. The firing of Goss, who was a political appointee brought in to bring the agency to heel, is clearly a concession to the CIA, where he and his aides were hated (that is not too strong a word.) Hayden at least is an old hand in the intelligence community, albeit it at the NSA and not the CIA. Whether this is an attempt to placate the agency in order to dam up its leaks to the press, or whether Bush is bringing in the big guns to crush agency resistance, is unclear. This could be a move by Rumsfeld to take CIA turf. But in many ways, these questions are simply what we call "Washington gas" -- meaning something that is of infinite fascination within Washington, D.C., but of no interest elsewhere and of little lasting significance anywhere.

The issue is not who heads the CIA or what its bureaucratic structure might be. The issue is, as it has been for decades, what it is that the CIA and the rest of the intelligence community are supposed to do and how they are supposed to do it. On the surface, the answer to that is clear: The job of the intelligence community, taken as a whole, is to warn the president of major threats or changes in the international system. At least that appears to be the mission, but the problem with that definition is that the intelligence community (or IC) has never been good at dealing with major surprises, threats and issues. Presidents have always accepted major failures on the part of the IC.

Consider. The IC failed to predict the North Korean invasion of South Korea. It failed to predict Chinese intervention there. It failed to predict the Israeli-British-French invasion of Suez in 1956. It failed to recognize that Castro was a communist until well after he took power. It failed to predict the Berlin Wall. It failed to predict or know that the Soviets had placed missiles in Cuba (a discovery that came with U-2 overflights by the Air Force). It failed to recognize the Sino-Soviet split until quite late. It failed to predict the tenacity of the North Vietnamese in the face of bombing, and their resilience in South Vietnam. The IC was very late in recognizing the fall of the Shah of Iran. It was taken by surprise by the disintegration of communism in Eastern Europe and the Soviet Union. It failed to predict the intentions of al Qaeda. And it failed in Iraq.

Historically, the American IC has been superb when faced with clearly defined missions. It had the ability to penetrate foreign governments, to eavesdrop on highly secure conversations, to know the intentions of a particular foreign minister at a particular meeting. Given a clear mission, the IC performed admirably. Where it consistently failed was in the amorphous mission of telling the president what he did not know about something that was about to change everything. When the IC was told to do something specific, it did it well. When it was asked to tell the president what he needed to know -- a broad and vague brief -- it consistently fell down.

This is why the argument going on between the CIA and the White House/Defense Department misses the point. Bush well might have ignored or twisted intelligence on Iraq's WMD. But the failure over Iraq is not the exception, it is the rule. The CIA tends to get the big things wrong, while nailing the lesser things time and again. This is a persistent and not easily broken pattern, for which there are some fundamental causes.

The first is that the IC sees its task as keeping its customers -- the president and senior members of his administration -- happy. They have day-to-day requirements, such as being briefed for a meeting with a foreign leader. The bread-and-butter work of the IC is the briefing book, which tells a secretary of state what buttons to push at a ministerial meeting. Ninety-nine percent of the taskings that come to the IC concern these things. And the IC could get 99 percent of the task right; they know that this minister is on the take, or that that minister is in a terrible fight with a rival, or that some leader is dying. They do that over and over again -- that is their focus. They are rarely rewarded for the risky business of forecasting, and if they fail to forecast the invasion of South Korea, they can still point to the myriad useful things at which they did succeed.

When members of the IC say that no one sees the vital work they do, they are right. And they are encouraged to do this work by their customers. If they miss the fall of the Soviet Union, it is the bread-and-butter work that keeps them going. If the nuts and bolts of intelligence compete with the vital need of a government to be ready for the unexpected, the nuts and bolts must win every time. The reason is simple: The unexpected rarely happens, but meetings of the G-8 happen every year. The system is built for the routine. It is hard to build a system for the unexpected.

A second problem is size. The American IC is much too big. It has way too many resources. It is awash in information that is not converted into intelligence that is delivered to its customers. Huge organizations will lose information in the shuffle. The bigger they are, the more they lose. Little Stratfor struggles to make sure that intelligence flowing from the field is matched to the right analyst and that analysts working on the same problem talk to each other, and it is tough. Doing it with tens of thousands of sources and intelligence officers, thousands of analysts and hundreds of briefers is a failure waiting to happen. All of the databases dreamt of by all of the information technology people in the IC cannot make up for total overload.

It can be argued that there is no alternative. The United States has global interests and thus must have global and massive resources. But the fact is that global interests are not well-served by a system that is too large to function efficiently. Whatever the need is, the reality is that managing the vast apparatus of the IC is overwhelmingly difficult, to the point of failure. Moreover, the management piece is so daunting that finding space to look for the unexpected -- and transmit that finding efficiently to the customer -- has been consistently impossible. The intelligence services of smaller countries sometimes do much better at the big things than massive intelligence services. The KGB was an example of intelligence paralysis due, among other things, to size.

A third issue is the cult of sourcing. There is a belief that a man on the ground is the most valuable asset there is. But that depends on where he is on the ground and who he is. A man on the ground can see hundreds of feet in any direction, assuming that there are no buildings in the way. It always amuses us to hear that so-and-so spent three years in some country -- implying expertise. We always wonder whether an Iranian spending three years in Washington, D.C., would be regarded as an expert around whom analysis could be built. Moreover, these three-year wonders frequently start doing freelance analysis, overriding analysts who have been studying a country for decades -- after all, they are "on the ground." But a blond American on the ground in the Philippines is fairly obvious, especially when he starts buying drinks for everyone, and the value of his "intelligence" is therefore suspect. Sourcing is vital; so are the questions of who, where and for how long.

The most significant weakness of the cult of sourcing is that the most important events -- like the Chinese intervention in Korea -- might be unreported, or -- like the fall of the shah -- might not be known to anyone. These things happened, but there was an intelligence collection failure in the first case; the second failure stemmed not from a collection problem, but from a purely analytic one. In any case, the lack of a source does not mean an event is not happening; it just means there is no source. There is no question but that sources are the foundation of intelligence -- but the heart of intelligence is the ability to infer when there is no source.

Another problem is the IC's obsession with security, compartmentalization and counterintelligence. The Soviet Union's prime mission was to penetrate the U.S. IC. Huge inefficiencies were, therefore, appropriately incurred in order to prevent penetration. The compartmentalization of sensitive information increases security, but it pyramids inefficiency. Al Qaeda is not engaged in penetrating the IC. It is dangerous in a different way than the Soviets were. Security and counterintelligence remain vital, but shifting the balance to take current realities into account also is vital. Intelligence work involves calculated risk. The current system not only keeps smart and interesting people out of jobs, but more important, it keeps them from access to the information they need to make the smart inferences that are so vital. That would seem to be too high a price to pay in the current threat environment. Information on China can be compartmentalized; information on the Muslim world could be treated differently.

The IC wants consistent messaging. They want to produce one product that speaks with a single coherent voice. The problem is that the world is much messier than that. Giving a president the benefit of the official CIA position on a matter is useful, but not as useful as allowing him to see the disputes, discomfort and doubts stemming from the different schools of thought. Those disagreements are sometimes treated as embarrassing by the IC -- but honest, public self-criticism builds confidence. Stratfor -- and we are not comparing our tiny outfit to the IC, with its massive responsibilities -- publishes an annual report card with our forecasts, specifying where we succeeded and failed. We may as well; our readers and clients know anyway.

This may not be what the president wants, of course, and Negroponte and Hayden will want to give him what he wants. But the head of an intelligence agency is like a doctor: He must give the patient what he needs and try to make it look like what the patient wants. In the end, it doesn’t matter what you do, as Porter Goss has just found out. Negroponte and Hayden will probably lose their jobs anyway -- through resigning or being sacked, or through Bush's second term ending. Even if they are lucky, their jobs won't last much more than two years. There is no percentage in hedging, when you think of it that way.

Perhaps the single greatest weakness of the IC is its can-do attitude. It cannot do everything that it is being asked to do -- and by trying, it cannot do the most important things that need to be done. It has had, as its mission, covering the world and predicting major events for the president. It has failed to do so on major issues since its founding, finding solace in substantial success on lesser issues. But it is possible that the bandwidth of the IC, already sucked up by massive management burdens, is completely burned up by the lesser issues. It may be that the briefing book to the president for his next meeting with the president of Paraguay or Botswana will be thinner, or he might just have to wing it. The republic will survive that. The focus must be on the things that count.

Rethinking why there is an intelligence community and how it does its job is the prerequisite for Hayden and Negroponte to be successful. We do not believe for a minute that they will do so. They don't have enough time in office, they have too many meetings to attend, they have too many divergent views to reconcile into a single coherent report. Above all, the CIA has to be prepared to battle the real enemy, which is the rest of the intelligence community -- from the Defense Intelligence Agency to the FBI. And, of course, the odd staffer at the White House.

This report may be distributed or republished with attribution to Strategic Forecasting, Inc. at www.stratfor.com.