The Dangerous Science of Intelligence Analysis

WikiLeaks' release of 92,000 military intelligence documents, now being parsed by media outlets across the globe, has sparked a groundbreaking experiment in open-source intelligence analysis. For the first time, journalists and amateurs can access and draw conclusions from the intelligence typically restricted to professional government analysts. But we in the public may be about to learn the same lesson that the Bush administration learned the hard way in the months and years following September 11: Analyzing intelligence is a science. It has specialized practices and methods. Done wrong, it can be dangerous.

As Jane Mayer documents in The Dark Side, one of President Bush's first and most serious mistakes after 9/11 was cutting out the intelligence analysts. Mayer writes:

In the days immediately after the attack, he and Cheney demanded to see all available raw intelligence reports concerning additional possible threats to America on a daily basis. Cheney had long been a skeptic about the CIA's skills, and was particularly insistent on reviewing the data himself. "The mistake," [high-ranking National Security Council official Roger] Cressey concluded later, "was not to have proper analysis of the intelligence before giving it to the President. There was no filter. Most of it was garbage. None of it had been corroborated or screened. But it went directly to the President and his advisers, who are not intelligence experts. That's when mistakes got made."

In reviewing raw intelligence without the standard rigors of the analysis process, the Bush White House was led to policies of historic significance. The two most famous were their conclusions, based on highly fragmentary intelligence, that Saddam Hussein was actively seeking weapons of mass destruction and that Iraq had ties to al-Qaeda. The mistakes were so serious that eventually the Bush administration not only reinstated the role of analysts but actually increased their prominence in the intelligence process. The long-standing "wall" between the CIA's analysis and operations wings, the latter of which is responsible for gathering intelligence, has been partially dismantled so that analysts can better oversee and vet the gathering process.

Intelligence generally comes in several forms, and during CIA Director George Tenet's tenure from 1997 to 2004, one of the most common sources of intel on Afghanistan was HUMINT, or human intelligence, which is derived from conversations and interrogations. But all intelligence, especially HUMINT, can be notoriously unreliable and fragmentary. Much of it comes from speculation, second-hand rumors, or outright lies. Often the source has an agenda; sometimes he or she simply wants to get paid.

One of the most important jobs of an intelligence analyst is determining the veracity of intelligence, most of which he or she is likely to discard. Former CIA analyst and presidential adviser Bruce Riedel tells Foreign Policy's Charles Homans:

It's like trying to put together a jigsaw puzzle. You maybe have 200 pieces of the puzzle. The first thing you don't know is, is this a 500-piece or 1,000-piece puzzle? And then with the 200 pieces you have, maybe half of them don't belong to this puzzle at all. They're in the wrong box. And then every hour or so, someone comes along and dumps 10 more pieces on your desk -- and nine of them aren't even part of it.

This week, as tens of thousands of pieces of intelligence are made publicly available on the Internet, the media may risk falling into the same trap as did the Bush White House of treating junk intelligence as gold. We are already seeing various reports, based on conjecture in the leaked intelligence documents, that Osama bin Laden is linked to Iran, that he bought remote-controlled rockets from North Korea, and that he died in June 2007 in Peshawar.

Clearly, not all of these reports can be true, and most likely none of them are. Will the media seize on the allegations of Iranian support for al-Qaeda as actionable fact or as the ambiguous intelligence fragment that it is? As Joshua Foust has explained, there is little evidence for and significant evidence against a substantive connection between al-Qaeda and the Iranian regime. Additionally, were Iran actually involved in fueling the Afghan conflict, we would almost certainly see the proliferation of explosively formed penetrators (EFPs), an extremely dangerous form of roadside bomb that was deployed in Iraq, allegedly with Iran's assistance (although this is disputed). A study put out by the Institute for the Study of War, a think tank that has not been shy about pointing out potential Iranian involvement in Iraq, only found:

A multi-ethnic, stable Afghanistan serves Iran's economic and security interests. While Tehran does not want to see a Taliban comeback in Kabul, it is wary of the presence of so many US troops on its frontiers and Washington's increasingly aggressive posture in recent years. Thus, some analysts argue that Iran favors the maintenance of a low level insurgency as long as US troops remain in Afghanistan. Such a low intensity conflict would tie down the US military and alleviate US and international pressure on Tehran over its nuclear program and other controversies.

Unlike the president of the United States, the media does not have the power to launch a war in the Middle East. But it does have the power to influence public opinion, which can lead to changes in policy. The Bush administration learned the risks and the lessons of evaluating raw intelligence outside the rigors of professional analysis. With 92,000 pieces of raw intel now accessible to anyone with an Internet connection, every curious journalist, editor, and reader will have to develop tremendous skepticism. It may be time to modify the adage of media in the Drudge era: We are all analysts now.