In one incident that illustrates the reach of this order, federal agents removed documents from the personal papers of the late Senator Scoop Jackson, housed at the University of Washington. Far more troubling was the revelation in 2006 that more than twenty-five thousand documents had been pulled from the stacks at the National Archives and Records Administration, which are open to the public. Matthew Aid, a part-time historian who first noted the disappeared documents, catalogued a bizarre list of what had gone missing: cables documenting a widely known intelligence failure from the Korean War; a 1948 message chastising the State Department for not predicting riots in Bogotá; talking points on how to handle questions about Japanese peace offers before the end of World War II; etcetera. Archive officials conducted an audit and determined that a third of the documents that had been pulled were not eligible to be reclassified, even under the new Bush standards.
Alongside the official classification system exists a murky system sometimes called “pseudo-secrecy.” More formally known as “sensitive but unclassified” or “controlled unclassified” information, it functions with little regulation, monitoring, or clear force of law. Attempts to fully measure the use of this category are frustrated by the fact that there is no single definition for what qualifies as sensitive-but-unclassified (SBU) information.
Concerns about sensitive-but-unclassified information date back at least to 1972, when a House committee held hearings deploring the ways that similar labels were being used to keep information from coming to light under the Freedom of Information Act, which carries an exemption for properly classified documents. Explicit presidential support for such pseudo-secret labeling dates to a 1977 presidential directive on telecommunications technology by Jimmy Carter, and it has been used in every subsequent administration.
In 2007, a Defense Department official charged with improving interagency information sharing estimated that there were 107 different labels in the category—from “Official Use Only” to “Sensitive Internal Use”—and none is monitored by the Oversight Office, which means that there are no official numbers to describe the trend. But openness advocates and some journalists have suggested that the Bush administration has significantly expanded both the number of such labels and the volume of documents being labeled. For example, in April 2002, officials at the Department of the Interior instructed their employees that “all unclassified DOI systems” should be “considered SBU.” The 2002 law authorizing the creation of the Department of Homeland Security specifically said that all scientific research produced by the department, wherever possible, should be unclassified, but President Bush used a signing statement, which spells out how the executive branch will interpret and implement a law, to make clear that he intended to mark much of that information as sensitive. In 2006, when the National Security Archive conducted a study, via FOIA, of the use of sensitive-but-unclassified labels at thirty-seven major government agencies, it found that roughly two-thirds of the SBU programs were operating without any statutory justification. Only one program had an automatic procedure for removing the designation, as is required by law with classified information. According to Barton Gellman, a reporter with The Washington Post, Vice President Cheney’s office routinely took to stamping papers with “Treated As: Top Secret/SCI,” a designation some classification experts believe Cheney invented. “At least with classification, you have ISOO overseeing what gets classified and what doesn’t get classified, keeping track of how much is classified,” says Rebecca Carr, who covered access and First Amendment issues for Cox Newspapers from January 2005 to May 2008. “The pseudo-classification category is like the Wild West. There’s nobody watching the store.”