A New Executive Order on Secrecy Policy

A Bush Administration executive order on national security classification policy leaves much to be desired from a public policy point of view and will do nothing to curb the Administration's excessive secrecy, though it is not quite as egregious as critics had anticipated.

	“The order does not touch the roots of dysfunction in the classification system, which allow agencies to make extravagantly false classification claims.”

In recent decades, whenever the Presidency shifted from one party to another, the new President has issued an executive order on secrecy policy to serve as the foundation of the classification system. Typically, and at least rhetorically, the orders issued by Democratic presidents have emphasized disclosure, while those of Republican presidents have stressed secrecy.

President Clinton's 1995 executive order 12958 dramatically accelerated declassification, inaugurating a process which has now yielded close to a billion pages of historically valuable declassified documents.

The Bush Administration's initiative to craft a new executive order, which began in August 2001, has therefore been a source of anxiety for those who feared that the Administration's predilection for official secrecy would lead to dramatic changes in classification policy.

Bush Administration Secrecy

Secrecy has become a distinguishing trait of the Bush Administration that is acknowledged even by its ideological partners. "An iron veil is descending over the executive branch," complained Rep. Dan Burton (R-IN) in 2001 after the Bush Administration rebuffed some of his inquiries regarding Justice Department oversight.

The barriers to information access are increasingly numerous and diverse. Census data, information about the role of industry in the Vice President's Energy Task Force, and budget estimates for the war on Iraq are just a few of the topics on which battles for access have lately been fought.

Of course, in many cases, there are two sides to the story, as in disputes over disclosures of inventories of toxic materials, for example. Opponents of disclosure argue that indiscriminate release of information could expose security vulnerabilities. (Environmentalists counter that disclosure is a prerequisite to correcting the vulnerability.)

But in other cases, the secrecy is mindless, arbitrary and unwarranted. For example, although the Central Intelligence Agency declassified the intelligence budget totals in 1997 and 1998 (under pressure of litigation), the Agency today says that the same information from 1947 and 1948 must remain classified and would damage national security if disclosed! This extreme case illustrates the bad faith that pervades much of the secrecy system today. (An FAS lawsuit opposing the CIA's claim is pending under the Freedom of Information Act.)

The Bush Administration's new executive order on national security classification, signed in late March, does not eliminate all constraints on official secrecy, as some had feared, but neither does it move beyond the parameters of the Cold War secrecy system into a truly twenty first century information policy.

The new Bush order generally affirms the single most important achievement of the 1995 Clinton executive order, which was its aggressive declassification regime for historically valuable documents that are 25 years old or older. The order includes a provision for automatic declassification of 25 year old documents, of which there are many millions, but will defer its effective date from April 2003 until December 31, 2006.

The new executive order also preserves the Interagency Security Classification Appeals Panel (ISCAP), which has proven to be an exceptionally powerful tool for correcting classification abuses by subjecting them to the scrutiny of an interagency review panel. The new order will somewhat blunt the ISCAP's effectiveness, however, by permitting the Director of Central Intelligence to reject Panel rulings unless he is overridden by the President.

The order directs that "the unauthorized disclosure of foreign government information is presumed to cause damage to the national security" and such information would therefore be presumptively classified, which has not previously been the case.

It includes the new category of "infrastructure" vulnerabilities as potentially classifiable information, and it would ease the reclassification of previously declassified information.

"It could be a lot worse," several officials spontaneously agreed. "Keep in mind that this is the Bush Administration we're talking about, it's post 9/11, and we're about to go to war," said one agency official. "It could be a lot worse."

It could also be a lot better.

The order does not touch the roots of dysfunction in the classification system, which allow agencies to make extravagantly false classification claims. Strengthening and expanding the ISCAP review process, rather than curtailing it, might have been one way to improve the correction of classification errors and abuses.

More fundamentally, the order is a vestige of a Cold War information policy that is now obsolete and increasingly counterproductive. It could have been implemented without any problem thirty years ago, but is simply oblivious to the implications of the information revolution of the past decade.

Not only do the order's authors fail to acknowledge the qualitative distinction between old fashioned paper records and digital data, they do not know what Senator Richard Shelby has lately pointed out: namely, that imposing a strict "need to know" standard of information control can diminish information's utility, and must inevitably exclude many of those who could productively exploit it to the detriment of national security.

Steven Aftergood directs the FAS Project on Government Secrecy. A version of this article was published as an op-ed in The Forward on March 28, 2003.