The Columbia School of Journalism’s brutal takedown of Rolling Stone’s article describing a sexual assault on the University of Virginia’s campus is a must read for anyone who cares about the pitfalls of investigative journalism. As I read the 13, 000 word article, however, I kept thinking about of the all-too-human mistakes made by a different organization under different circumstances—that is, the CIA and its analysis of Saddam Hussein’s WMD efforts leading up to the 2003 invasion.
Luckily, we can mostly compare the two fiascos, since CIA mostly declassified its post-mortem of the WMD debacle a few years ago. And lo and behold, there are real parallels between Rolling Stone’s botched article and the mistakes made by the Agency’s analysts in the leadup to the war.
While the stakes were much different—CIA analysts had a much more complex task than Rolling Stone’s team—the comparisons are still instructive. Here are a few:
Lousy reporting and a lack of hard, corroborating information. Rolling Stone’s writer, Sabrina Rubin Erdely, neglected to follow multiple leads to corroborate the story of “Jackie”—the student whose terrifying assault formed the core of the RS article. At Jackie’s request, Ms. Erdely did not contact the alleged rapist, “Drew,” nor did she contact Jackie’s three friends who were with her after the incident supposedly took place.
This lack of corroborating information, along with mostly single-source reporting, was glossed over in the editing phase. As Columbia’s post-mortem noted, “[Managing Editor Will] Dana said he was not told of reporting holes like the failure to contact the three friends or the decision to use misleading attributions to obscure that fact.”
A paucity of hard intelligence was also true for efforts to figure out what exactly Baghdad was up to in its weapons development efforts. The U.S. relied heavily upon UN weapons inspectors following the first Iraq War for intelligence, as there seemed to be few, if any, human sources in Iraq itself. Furthermore, much of it was garbage – think the intelligence that came from “Curveball.”
Plus, there was an “absence of evidence” issue at work in Baghdad. As CIA’s post-mortem noted, “Shocked by the unexpected aggressiveness of early UN Special Commission (UNSCOM) inspections in 1991, Iraq secretly destroyed or dismantled most undeclared items and records that could have been used to validate the unilateral destruction, leaving Baghdad unable to provide convincing proof when it later tried to demonstrate compliance.”
The Iraqis destroyed evidence of their perfidy. Yet without hard proof of destruction, the very understandable assumption that Iraq was still in the WMD business was still in effect.
Preconceived notions and established mindsets coloring the reporting/analysis. The age-old analytic trap snared both Rolling Stone’s writers/editors and CIA’s analysts. The Columbia report notes confirmation bias—the tendency to highlight data that are agreeable to one’s argument and dismiss data which is not—“seems to have been a factor here.”
The report continues: “Erdely believed the university was obstructing justice. She felt she had been blocked. Like many other universities, UVA had a flawed record of managing sexual assault cases. Jackie’s experience seemed to confirm this larger pattern. Her story seemed well established on campus, repeated and accepted.” This conventional wisdom—that sexual assault on college campuses was occurring and was being covered up—was not challenged.
Over a decade beforehand, CIA (and the rest of the world) basically viewed Baghdad as being continually deceptive on a whole host of issues—and they were more or less right—despite the truth Iraq truly did not have WMD programs past the mid-1990s. As the report notes, “Ironically, even at key junctures when the [Hussein] regime attempted to partially or fully comply with UN resolutions, its suspicious behavior and destruction of authenticating documentation only reinforced the perception that Iraq was being deceptive.”
Ironically, Saddam Hussein’s son-in-law’s defection to the west in 1995 reinforced the assumption Baghdad it was still hiding something. In reality, the absence of evidence was considered evidence of…something. Or as CIA’s post-mortem states, “Faced with inconclusive or uncertain data, analysts made judgments with conviction that Iraq could successfully conceal damaging data.”
To be fair, no other intelligence service, even European or Arab ones that might have better understood Baghdad better, had a fundamentally different view on what Iraq was, or was not, up to. There were few credible alternative ideas about Saddam’s behavior out there, despite it not being the truth. Everyone believed the simplest answer. And it turned out to be dead wrong.
Editors who didn’t push hard enough or ask the right questions. The Columbia report was blistering in its critique of Rolling Stone’s editors. It noted, “Investigative reporters working on difficult, emotive or contentious stories often have blind spots. It is up to their editors to insist on more phone calls, more travel, more time, until the reporting is complete. [RS editor Sean] Woods did not do enough.” Furthermore, Managing Editor Will Dana “might have looked more deeply into the story drafts he read, spotted the reporting gaps and insisted that they be fixed. He did not. “It’s on me,” Dana said. ‘I’m responsible.’”
CIA’s 2006 post-mortem doesn’t touch upon the failures of its management chain, but all of CIA’s finished intelligence travels through multiple layers of bureaucracy and editing before being disseminated to the policymaking customer. These editors have theoretically more experience working on intelligence issues than the average analyst, so they should be cognizant of problems, hidden assumptions, biases, and the like. No particular paper could ever go out without the approval of multiple people, from the analyst to his/her team chief, the relevant Office Director, the PDB editorial staff and finally the Director himself. Many more people had their hands on every single document—and since the faulty analysis saw the light of day, each stop along the way bears some degree of responsibility. Bottom line: it was never a failure of an individual analyst.
More broadly, both the WMD debacle and the UVA story are case studies of what can go wrong when discussing high-profile issues that most people—thoughtful, smart, well-educated and well-intentioned individuals—had a strong opinion about. As Sean Woods noted, “It’s been an extraordinarily painful and humbling experience. I’ve learned that even the most trusted and experienced people – including, and maybe especially, myself – can make grave errors in judgment.” I’m sure there have been many people at CIA and throughout the government who felt the same way in the lead up to the 2003 Iraq invasion. It’s also interesting to note that at Rolling Stone and CIA, no one was demoted or fired for what happened on their watch.
Finally, the more important issue is in both institutions, this systemic failure did not cause a systemic change in the way each organization operates. Sure, CIA warns new analysts about the WMD debacle, and the Columbia School of Journalism is probably going to incorporate a whole “Rolling Stone” module into training greenhorn journalists, but the business of intelligence analysis or investigative reporting will probably not fundamentally change. Or, as Will Dana said, “It’s not like I think we need to overhaul our process, and I don’t think we need to necessarily institute a lot of new ways of doing things. We just have to do what we’ve always done and just make sure we don’t make this mistake again.”
Perhaps that’s should be an expected outcome. As Robert Jervis noted in his 2006 commentary on the Iraq WMD challenge for those who want to burn everything down and start anew:
Reforms are not likely to bring great improvement. Of course, any improvement is to be welcomed and may be worth great effort, but the very fact that intelligence failures have occurred in all countries and all eras indicates that while there may be better or worse systems in terms of accuracy (and we do not even know whether this is the case), errors are likely to be frequent even if we do better.
In ourselves, we place our trust.
photo: Karen Blaha: The lawn and rotunda. (Wikimedia Commons)