The firestorm surrounding the story that U.N. Ambassador Susan Rice told on Sunday morning talk shows five days after the attack on the American diplomatic facility in Benghazi has drawn wildly different reactions around Washington. Bad data? Innocent mistake? Political deception? From the president's staunch defense of Rice to John McCain's repeated attacks, different people see different things. But, as a former CIA analyst, I've taken a separate lesson from the episode: The office of the Director of National Intelligence (DNI), created to facilitate interagency analysis and operations, has become a serious bureaucratic obstacle.
For a long time, the CIA ruled the intelligence cycle of collection, aggregation, analysis, and dissemination. But in 2004, the 9/11 Commission recommended the United States unify the intelligence community. Thus, the DNI was born. Today, according to its website, the DNI "serves as the head of the Intelligence Community, overseeing and directing the implementation of the National Intelligence Program and acting as the principal advisor to the President, the National Security Council, and the Homeland Security Council for intelligence matters related to national security." But, however noble and sensible the intent, the DNI has done very little to remedy the coordination issues -- and Benghazi is a perfect example.
On September 16, Rice went on Meet the Press and stated, "Our current assessment is that what happened in Benghazi was, in fact, initially a spontaneous reaction to what had just transpired hours before in Cairo, almost a copycat of the demonstrations against our facility in Cairo, which were prompted, of course, by the [Innocence of Muslims] video." Rice has since publicly stated that there was an error in the intelligence she was given -- there was not a protest at the facility in Benghazi. But the initial set of talking points from the CIA is said to have indicated that the attackers had links to al Qaeda. So why did she not mention that? A U.S. intelligence official has said, "The information about individuals linked to al Qaeda was derived from classified sources, and could not be corroborated at the unclassified level; the links were tenuous and therefore it made sense to be cautious before naming perpetrators." So was Rice just being prudent? It's possible, though her performance was clumsy at best.
I have no access to classified information from the CIA or DNI on Benghazi, but here's what might have happened. On the eve of the attack, the analysts at the CIA are reading reports from regional embassies, watching news reports, and poring over cables from assets as the ridiculous anti-Islam YouTube video sparks protests across the region. As the attack in Benghazi ensues, they scramble to assess the situation and draft products for the various consumers -- the most sensitive pieces go to select people at the White House, Pentagon, Office of the DNI, and National Security Council. If time allows, an alternative product is scrubbed of the most sensitive information for release across various parts of the U.S. government. In other words, the CIA may (or may not) have disseminated two different sets of talking points: one highly classified to protect sources and methods, and a second for broader dissemination.
Although the CIA's role has changed in recent years, informing and warning the senior ranks of the U.S. government on tactical issues is supposed to be its purview, because over the course of decades, the agency has established processes to create, coordinate, and disseminate the right intelligence to the right audience. This is not to say that the CIA has always been correct in its assessments -- the very nature of analysis, particularly on breaking events, all but guarantees some degree of error -- but it has an established way of producing analytic products.
Alas, one of the responsibilities of the DNI is to "ensure the most accurate analysis of intelligence is derived from all sources to support national security needs" (my emphasis), and the intelligence community, for better or worse, is not a public affairs dynamo: it is made up of 16 agencies. It's likely that the DNI passed the document to other intel agencies, which all added views from their own analysts, and created another set of talking points to be disseminated. These talking points could easily have emphasized the role of the video in Benghazi -- not all analysts from differing agencies have access to the same information. In the case of the DNI, collective intelligence analysis can be just as flawed as independent analysis: a consensus view can reveal differences of opinion within the intelligence community, but it also can result in the loss of nuance or, worse, become a convenient excuse for adopting a safe position.
So a policymaker could end up receiving many different sets of talking points, not all of which necessarily say the same thing. See how confusing this could be to someone in Rice's position? An amalgamation of analysis might be more tempered or more diluted than the analyses that contributed to it.
As a former CIA analyst and news consumer, I think we are looking at three distinct problems:
First, interagency coordination on breaking events is very difficult, if not impossible. The DNI, with its acquisition of the National Intelligence Council (NIC), has a remarkable ability to coerce intra-agency collaboration on cross-cutting strategic issues. The NIC serves as a bridge between the intelligence and policy communities, employing senior subject matter experts from academia, government, and the private sector. Rightly so, the National Intelligence Estimates are the definitive position of the intelligence community. The extensive, careful interagency coordination that goes into producing an NIE takes months and cannot be replicated at a tactical level in a fast-moving situation like Benghazi.
Second, intelligence analysis of specific events rarely lends itself to a good sound bite suitable for the 24-hour news cycle. If, for instance, Amb. Rice had repeated the caveat, "The information is still being refined, we don't know exactly who the perpetrators are or why they attacked, but here is what we have so far" over and over, it probably would have been accurate. But it also would have been wildly unsatisfactory (or frustrating) to both political leaders and the American public. There are always varying degrees of certainty when it comes to intelligence analysis as new information arrives and is weighed against historical information.
Third, the CIA's cultural inclination to stay out of the media fray is part of its genetic code and it cannot -- will not -- be the public interface. The CIA's primary mission is to collect, evaluate, and disseminate foreign intelligence to assist the president and senior U.S. policymakers in making national security decisions. Unlike the Israel Defense Forces, which has both a Twitter and a Pinterest account -- although the NIC just started a Twitter account (@ODNI_NIC), which is a promising step in the right direction -- the CIA is likely to remain frustratingly silent about its work, both good and bad. Intelligence will always be filtered through policymakers. I experienced this firsthand in the run-up to the Iraq war, watching various policymakers on Meet the Press answer questions based on a piece I had written earlier in the week. I cheered every time Tim Russert asked a pointed question, forcing them to be more specific and accurate -- because the CIA will not go on record to correct a statement by a policymaker on a talk show.
In light of this, what should the DNI's role be in the intelligence community, if not disseminating a coordinated intelligence product? The CEO of a company is typically the one planning strategy, interfacing with board members, stockholders, and consumers. A CEO doesn't typically write the chief financial officer's year-end summary or the marketing director's strategy -- instead, he views both products from 25,000 feet to ensure the company is on steady footing. The DNI should have a similar role: rather than replicating work, it should focus on reviewing the source material from the various agencies and collaborating to ensure all of the information has been reviewed. In the case of the Benghazi talking points, the intelligence community all had a role in editing the talking points once passed from the CIA. Other points of view make sense, but in the immediate aftermath of something like Benghazi, the arrival of new (and possibly conflicting information) is likely to confuse, not improve, the product. It is best to leave the dissemination, in the immediate aftermath, in the hands of the agency that owns the source of the information and is in the business of disseminating intel products -- in this case the CIA.
The political churn around Benghazi masks the real issue: how best to strike a balance between informing U.S. policymakers in a timely manner and continuing to foster coordination (if not cooperation) between the members of the intelligence community. The DNI has a purpose, but it does not serve taxpayers to have another bureaucracy replicating work being done by those it oversees. Instead, it would be worthwhile -- especially for those hard-working, well-qualified DNI employees -- to reassess the office's mission.