WikiLeaks and the Culture of Classification
By Scott Stewart
On Friday, Oct. 22, the organization known as WikiLeaks published a cache of 391,832 classified documents on its website. The documents are mostly field reports filed by U.S. military forces in Iraq from January 2004 to December 2009 (the months of May 2004 and March 2009 are missing). The bulk of the documents (379,565, or about 97 percent) were classified at the secret level, with 204 classified at the lower confidential level. The remaining 12,062 documents were either unclassified or bore no classification.
This large batch of documents is believed to have been released by Pfc. Bradley Manning, who was arrested in May 2010 by the U.S. Army Criminal Investigations Command and charged with transferring thousands of classified documents onto his personal computer and then transmitting them to an unauthorized person. Manning is also alleged to have been the source of the classified information released by WikiLeaks pertaining to the war in Afghanistan in July 2010.
WikiLeaks released the Iraq war documents, as it did the Afghanistan war documents, to a number of news outlets for analysis several weeks in advance of their formal public release. These news organizations included The New York Times, Der Spiegel, The Guardian and Al Jazeera, each of which released special reports to coincide with the formal release of the documents Oct. 22.
Due to its investigation of Manning, the U.S. government also had a pretty good idea of what the material was before it was released and had formed a special task force to review it for sensitive and potentially damaging information prior to the release. The Pentagon has denounced the release of the information, which it considers a crime, has demanded the return of its stolen property and has warned that the documents place Iraqis at risk of retaliation and also place the lives of U.S. troops at risk from terrorist groups that are mining the documents for tidbits of operational information they can use in planning their attacks.
When one takes a careful look at the classified documents released by WikiLeaks, it becomes quickly apparent that they contain very few true secrets. Indeed, the main points being emphasized by Al Jazeera and the other media outlets after all the intense research they conducted before the public release of the documents seem to highlight a number of issues that had been well-known and well-chronicled for years. For example, the press has widely reported that the Iraqi government was torturing its own people; many civilians were killed during the six years the documents covered; sectarian death squads were operating inside Iraq; and the Iranian government was funding Shiite militias. None of this is news. But, when one steps back from the documents themselves and looks at the larger picture, there are some interesting issues that have been raised by the release of these documents and the reaction to their release.
The documents released in this WikiLeaks cache were taken from the U.S. government's Secret Internet Protocol Router Network (SIPRNet), a network used to distribute classified but not particularly sensitive information. SIPRNet is authorized only for the transmission of information classified at the secret level and below. It cannot be used for information classified top secret or more closely guarded intelligence that is classified at the secret level. The regulations by which information is classified by the U.S. government are outlined in Executive Order 13526. Under that order, secret is the second-highest level of classification and applies to information that, if released, would be reasonably expected to cause serious damage to U.S. national security.
Due to the nature of SIPRNet, most of the information that was downloaded from it and sent to WikiLeaks consisted of raw field reports from U.S. troops in Iraq. These reports discussed things units encountered, such as IED attacks, ambushes, the bodies of murdered civilians, friendly-fire incidents, traffic accidents, etc. For the most part, the reports contained raw information and not vetted, processed intelligence. The documents also did not contain information that was the result of intelligence-collection operations, and therefore did not reveal sensitive intelligence sources and methods. Although the WikiLeaks material is often compared to the 1971 release of the Pentagon Papers, there really is very little similarity. The Pentagon Papers consisted of a top secret-level study completed for the U.S. secretary of defense and not raw, low-level battlefield reports.
To provide a sense of the material involved in the WikiLeaks release, we will examine two typical reports. The first, classified at the secret level, is from an American military police (MP) company reporting that Iraqi police on Oct. 28, 2006, found the body of a person whose name was redacted in a village who had been executed. In the other report, also classified at the secret level, we see that on Jan. 1, 2004, Iraqi police called an American MP unit in Baghdad to report that an improvised explosive device (IED) had detonated and that there was another suspicious object found at the scene. The MP unit responded, confirmed the presence of the suspicious object and then called an explosive ordnance disposal unit, which came to the site and destroyed the second IED. Now, while it may have been justified to classify such reports at the secret level at the time they were written to protect information pertaining to military operations, clearly, the release of these two reports in October 2010 has not caused any serious damage to U.S. national security.
Another factor to consider when reading raw information from the field is that, while they offer a degree of granular detail that cannot be found in higher-level intelligence analysis, they can often be misleading or otherwise erroneous. As anyone who has ever interviewed a witness can tell you, in a stressful situation people often miss or misinterpret important factual details. That's just how most people are wired. This situation can be compounded when a witness is placed in a completely alien culture. This is not to say that all these reports are flawed, but just to note that raw information must often be double-checked and vetted before it can be used to create a reliable estimate of the situation on the battlefield. Clearly, the readers of these reports released by WikiLeaks now do not have the ability to conduct that type of follow-up.
Few True Secrets
By saying there are very few true secrets in the cache of documents released by WikiLeaks, we mean things that would cause serious damage to national security. And no, we are not about to point out the things that we believe could be truly damaging. However, it is important to understand up front that something that causes embarrassment and discomfort to a particular administration or agency does not necessarily damage national security.
As to the charges that the documents are being mined by militant groups for information that can be used in attacks against U.S. troops deployed overseas, this is undoubtedly true. It would be foolish for the Taliban, the Islamic State of Iraq (ISI) and other militant groups not to read the documents and attempt to benefit from them. However, there are very few things noted in these reports pertaining to the tactics, techniques and procedures (TTP) used by U.S. forces that could not be learned by simply observing combat operations — and the Taliban and ISI have been carefully studying U.S. TTP every hour of every day for many years now. These documents are far less valuable than years of careful, direct observation and regular first-hand interaction.
Frankly, combatants who have been intensely watching U.S. and coalition forces and engaging them in combat for the better part of a decade are not very likely to learn much from dated American after-action reports. The insurgents and sectarian groups in Iraq own the human terrain; they know who U.S. troops are meeting with, when they meet them and where. There is very little that this level of reporting is going to reveal to them that they could not already have learned via observation. Remember, these reports do not deal with highly classified human-intelligence or technical-intelligence operations.
This is not to say that the alleged actions of Manning are somehow justified. From the statements released by the government in connection with the case, Manning knew the information he was downloading was classified and needed to be protected. He also appeared to know that his actions were illegal and could get him in trouble. He deserves to face the legal consequences of his actions.
This is also not a justification for the actions of WikiLeaks and the media outlets that are exploiting and profiting from the release of this information. What we are saying is that the hype surrounding the release is just that. There were a lot of classified documents released, but very few of them contained information that would truly shed new light on the actions of U.S. troops in Iraq or their allies or damage U.S. national security. While the amount of information released in this case was huge, it was far less damaging than the information released by convicted spies such as Robert Hanssen and Aldrich Ames — information that crippled sensitive intelligence operations and resulted in the execution or imprisonment of extremely valuable human intelligence sources.
Culture of Classification
Perhaps one of the most interesting facets of the WikiLeaks case is that it highlights the culture of classification that is so pervasive inside the U.S. government. Only 204 of the 391,832 documents were classified at the confidential level, while 379,565 of them were classified at the secret level. This demonstrates the propensity of the U.S. government culture to classify documents at the highest possible classification rather than at the lowest level really required to protect that information. In this culture, higher is better.
Furthermore, while much of this material may have been somewhat sensitive at the time it was reported, most of that sensitivity has been lost over time, and many of the documents, like the two reports referenced above, no longer need to be classified. Executive Order 13526 provides the ability for classifying agencies to set dates for materials to be declassified. Indeed, according to the executive order, a date for declassification is supposed to be set every time a document is classified. But, in practice, such declassification provisions are rarely used and most people just expect the documents to remain classified for the entire authorized period, which is 10 years in most cases and 25 years when dealing with sensitive topics such as intelligence sources and methods or nuclear weapons. In the culture of classification, longer is also seen as better.
This culture tends to create so much classified material that stays classified for so long that it becomes very difficult for government employees and security managers to determine what is really sensitive and what truly needs to be protected. There is certainly a lot of very sensitive information that needs to be carefully guarded, but not everything is a secret. This culture also tends to reinforce the belief among government employees that knowledge is power and that one can become powerful by having access to information and denying that access to others. And this belief can often contribute to the bureaucratic jealously that results in the failure to share intelligence — a practice that was criticized so heavily in the 9/11 Commission Report.
It has been very interesting to watch the reaction to the WikiLeaks case by those who are a part of the culture of classification. Some U.S. government agencies, such as the FBI, have bridled under the post-9/11 mandates to share their information more widely and have been trying to scale back the practice. As anyone who has dealt with the FBI can attest, they tend to be a semi-permeable membrane when it comes to the flow of information. For the bureau, intelligence flows only one way — in. The FBI is certainly not alone. There are many organizations that are very hesitant to share information with other government agencies, even when those agencies have a legitimate need to know. The WikiLeaks cases have provided such people a justification to continue to stovepipe information.
In addition to the glaring personnel security issues regarding Manning's access to classified information systems, these cases are in large part the result of a classified information system overloaded with vast quantities of information that simply does not need to be protected at the secret level. And, ironically, overloading the system in such a way actually weakens the information-protection process by making it difficult to determine which information truly needs to be protected. Instead of seeking to weed out the overclassified material and concentrate on protecting the truly sensitive information, the culture of classification reacts by using the WikiLeaks cases as justification for continuing to classify information at the highest possible levels and for sharing the intelligence it generates with fewer people. The ultimate irony is that the WikiLeaks cases will help strengthen and perpetuate the broken system that helped lead to the disclosures in the first place.