campaign desk

How To Keep Secrets

Obama tries to get classification right
September 2, 2009

Sign up for The Media Today, CJR’s daily newsletter.

It’s no secret that many think America’s classification system is no longer up to the task of protecting the nation or its secrets.

In late 2008, a Homeland Security advisory panel chaired by William Webster—the only person to ever run both the FBI and the CIA—described the classification system as “broken,” and recommended that the department’s incoming secretary find a way to make Americans safer by expanding information access. The 9-11 Commission held that the “long-term costs of overclassification” were “substantial” to both safety and America’s pocketbook. Those in government also see problems.

“When you’re dealing with these things everyday, I very often ask: Why is this classified?” then-Secretary of Defense Donald Rumsfeld said while being questioned at a 2004 hearing by then Senator Hillary Clinton. “Overclassification is, I agree, something that very properly ought to be addressed in a serious way.”

The biggest question facing stakeholders in the classification system—journalists, historians, and government and military officials—is what the Obama administration will do to reform a rickety framework, reduce costs and backlogs, foster broader access to historical records, and promote sharing of important intelligence information between analysts working in many government entities.

Many of these problems are the legacy of the paper-based classification system born in March of 1940, when, with Europe at war, President Roosevelt signed Executive Order 8381, imposing penalties for the unauthorized distribution of specially marked documents.

Since then, ten other executive orders have attempted to plumb the twisted lines intersecting security and accountability, reform and bureaucratic intransigence. In late May, President Barack Obama entered the fray by signing a presidential memo tasking General Jim Jones, his National Security Advisor, with laying the groundwork for a new executive order.

Sign up for CJR’s daily email

While Obama’s memo called for recommendations by August 25, the group remains at work on a narrow range of issues. Representatives from the government agencies with the greatest stake in the classification system—the military, the intelligence community, nuclear secret keepers, and the National Archives among them—have been meeting on and off since Obama signed the order, sometimes weekly, working through subject areas that the order is likely to address.

“The outcome is uncertain, but at least some of the participants are taking the task seriously and want to do more than a token effort,” says Steven Aftergood, a longtime secrecy watcher at the Federation of American Scientists. “There is remarkable consensus—not just at the National Security Council but in the national security agencies themselves—that the classification system is broken. It is not serving the interests of the agencies in performing their missions efficiently. They are not concerned about freedom of information or things like government accountability, but they are concerned about having a logical, predictable, rational information security policy. And there’s a sense that this is the last chance that we have for a decade—maybe for a generation—to get the classification system right.”

But even if there is consensus about the system’s ills, it looks unlikely that the review process will likely produce any drastic changes.

“These E.O.s are typically not revolutionary. Rather, they are evolutionary,” says Jay Bosanko, head of the Information Security Oversight Office, the government’s internal classification monitoring body. “Typically they evolve and the concepts are modified. New concepts are introduced that are revolutionary, but the orders themselves tend to be a little slow to change.”

The biggest potential change suggested in Obama’s memo, one almost certain to end up in an eventual executive order, would fulfill his campaign promise to create a National Declassification Center in order to accelerate the declassification of historical records. The declassification center is not a new idea; it was described and endorsed in a 1997 report produced by Senator Daniel Patrick Moynihan’s blue ribbon commission on secrecy.

The general concept is simple. In order to declassify a document under the current system, every agency that has an “equity” in the document—meaning they produced material within—no matter how old the document or relevant information, must take a look and approve its release.

But this process is uncoordinated—there’s no permanent central location where the agencies’ declassification representatives can congregate to review documents together. Instead, documents are passed around between agencies, meaning that each agency can’t easily determine what has already been flagged or redacted by a colleague agency. While there’s some prioritization of the classification queue by relative age or topic, there’s no guarantee that any of these agencies—many of which have a poor reputation for understaffing and underfunding declassification efforts–will promptly examine a document once it’s been passed on to them for review.

A National Declassification Center, at its most basic, would provide a central location where agencies that deal extensively with classification could detail personnel and concentrate and coordinate their efforts. As the name makes clear, declassification would be the center’s central mission, rather than a poorly funded afterthought coming in behind the very real priorities of the agencies currently charged with doing classification in-house.

“I think that the National Declassification Center potentially has tremendous value, and it would be a major transformation in the way that declassification is done in this country,” says Meredith Fuchs, general counsel of the National Security Archive, a research library housed at George Washington University that regularly deals with classified documents. “Right now declassification is really mismanaged because it’s so dispersed among the agencies, and there’s a lot of redundancy, and there’s not a lot of best practices because its being done in so many places.”

But it is still far from clear what the center’s exact set-up and powers will be.

Martin Faga, who from 1989 to 1993 led the National Reconnaissance Office, then a top-secret agency responsible for the country’s spy satellites, is currently acting chair of the Public Interest Declassification Board, a federal advisory panel which issued a report in 2007 endorsing a National Declassification Center. He hopes that the Center will have some role in establishing rules of thumb for classification and declassification guides, the agency-level documents consulted by officials as they decide how a document should be classified at its creation, or a whether it’s safe to declassify a given document later on. Many guides have gone years without update, and a hard look at some of these could render new categories of documents ineligible for classification or eligible for release.

Steven Aftergood of the Federation of American Scientists hopes that the center will have an even stronger hand, and be given the power to formally review and approve or reject the guides. “If I could wish one thing into the executive order, that would be my wish,” says Aftergood.

Behind that hope is an argument that Aftergood recently made in an article appearing in the Yale Law & Policy Review: that successful classification policy adjustments promote accountability by taking decisions on classification out of the sole hands of the agency that originally made them.

Aftergood cites the Interagency Security Classification Appeals Panel, which was created by a Clinton-era executive order. ISCAP consists of representatives from the major classifying agencies, as the NDC would be. When a citizen feels that a document is unjustly classified, ISCAP is, short of judicial review, the final court of appeal. And while the agency hoping to keep the document has a chance to make their case, the panel of their colleagues can vote to overrule, in full or in part, the agency’s original decision. In fact, according to Aftergood’s data, ISCAP has done so in 64 percent of cases put before it.

It’s far from clear if the expected declassification center will have any power, like ISCAP does, to overrule a single agency’s classification decisions—and, if so, how that power will be exercised. Faga, of the Public Interest Declassification Board, favors placing the ability in the hands of the Center’s chair. An ISCAP-like voting process is also a possibility.

But without some authority to declassify documents over the objection of other agencies, Aftergood says, the NDC “is simply a way of streamlining the status quo, and making existing declassification procedures run more smoothly. So instead of waiting three or five years for an unsatisfactory response or denial, now, with a National Declassification Center you might be able to get your denial in a year and a half. Is that progress? I don’t think it is.”

While a National Declassification Center draws support from many in the access community, others question the assumptions upon which the idea itself is based.

“It’s a highly inefficient model to believe that every agency needs to have a say in the declassification of specific records, even if they have an equity in it,” says Bill Leonard, who headed the Information Security Oversight Office from 2002 to 2007. “The irony of it is that I was a big support of a National Declassification Center. I was really pushing for it. But quite frankly, that was in the Bush administration. And I was hoping in an Obama administration we’d actually move past that mindset where we’d actually need a National Declassification Center.“

An alternative, favored by Leonard and popular in the access community, would be to create a so-called “drop dead” date. Nearly all documents that were, say, twenty-five, thirty, or forty years old would be automatically declassified. Aftergood suggests that different agencies might have different drop dates to reflect the kinds of classified information they deal in. But no one expects the order to endorse a drop dead date.

Obama’s memo, after floating the idea of a National Declassification Center, goes on to mention restoring what the president describes as “the presumption against classification.” In 2003, George W. Bush removed instructions issued in Clinton’s executive order eight years before saying that if a classifier had “significant doubt” about whether a document should be classified, it should remain unclassified.

It’s not clear that this instruction made a difference. “I don’t think we saw any change whatsoever,” says Leonard. “Those are words without meaning.”

Others basically agree.

“I think certainly the presumption should be reversed, but I don’t think that’s going to make any difference,” says Fuchs. “It’s important rhetoric but in order to actually reduce classification, its necessary for there to be incentive and oversight built into the actual classification system, that will discourage unnecessary classification and will impose consequences such as retraining or improved training when unnecessary classification is identified.”

Still, slipping something akin to Clinton’s language back into the executive order has wide support, in part because it might be a way to signal the revisions’ intent.

“It speaks to tone, and tone is very important,” says Bosanko.

There’s another Bush era change to the executive order that went unmentioned in Obama’s May memo. In 1999, the Director of Central Intelligence argued that his—and therefore his agency’s—classification decisions were exempted from review by the ISCAP process. While the Justice Department’s Office of Legal Counsel flatly disagreed with his legal argument, four years later the CIA was able to obtain a carve out from Bush allowing the intelligence chief to unilaterally overrule ISCAP. Since then, there have been two reported cases where the Director of National Intelligence has overruled a decision requiring release.

The potential that the intelligence community’s special powers at ISCAP could be left intact troubles Bill Leonard, who in his capacity as ISOO director served as ISCAP’s executive secretary at the time the change was made.

“It will be interesting to see if that’s rolled back. If even that’s not rolled back,” says Leonard, “then we’ll know that we truly are just tinkering around the edges.”

There’s good reason that so many questions about the classification order remain unanswered: it is being discussed behind closed doors. The drafting process has received virtually no media coverage, perhaps because the principals are none too eager to talk about recommendations that they are privately producing for the president. (The National Security Council, which is leading the discussion, did not respond to multiple requests from CJR for comment.)

“They are meeting weekly as an interagency group, and meeting more frequently within the agencies to sort out their own wants and needs,” says Steven Aftergood. “They are not meeting weekly with people outside the government, who have different expectations for this process. And it won’t be surprising if the final product reflects that. I mean, you and I aren’t around the table.”

General Jones’s staff did ask the Public Interest Declassification Board to run a public input process, which included a forum at the National Archives, attended by government officials and a wide range of outside access advocates and classification stakeholders. The PIDB also worked with the White House’s Office of Science and Technology Policy to set up an online public forum allowing anyone to offer outside suggestions, no matter how far they were from Washington.

Many people contributed—historians, former government employees, and experts including Aftergood and Leonard. The PIDB summarized their recommendations and sent them along to General Jones. The accompanying letter offered a meeting between the PIDB and the interagency drafting group, and suggested that some form of the recommendations be released before they were finalized to allow further public comment. There has been no response on either point.

Even so, it’s indisputable that the online forum offered an unprecedented method for the access community and everyday citizens to offer suggestions and concerns about classification policy.

“We think it worked out quite well,” says PIDB acting chair Martin Faga, before acknowledging that it was “hard to say” how seriously the public input would be taken.

It would have been easy for a casual commenter, one unfamiliar with the structure of the review, to get the impression that they were offering comment directly to the decision makers, rather than to a board that would summarize and pass along their contributions.

“It was at one end of a very long funnel,” said Aftergood. “Their input may be briefly read and considered, but it won’t be given the same weight as the agencies that have to comply with the executive order.”

Bosanko, in his dual role as a member of the interagency group and the PIDB’s secretary, was a connection between the review process and the public comment generated on the project’s blog. While he declined to comment onwhat the interagency group would be recommending to the president, he says that ideas and proposals raised on the blog forum were regularly cited, weighed, and discussed by the decision makers.

The panel’s recommendations, whatever they may be, can’t begin to impact the classification system until they become a signed executive order. After that, there will be a months- to years-long period of ramping up, as the ISOO writes administrative guidance refining the implementation. There will be a major effort to retrain the nation’s classification officers. And if, as is expected, the order produces a new National Declassification Center, there will be all the effort required to create an entirely new government entity, requiring cooperation and staff from agencies throughout the executive branch.

“Just like the classification system itself, where there’s an inherent tension about what is open and what is closed, you have that same tension with policy and what is implementable,” says Bosanko, before listing some of the challenges to reforming the classification system: “The level of change you can foster in a bureaucracy in a given period of time, the amount of culture you have to over come, the budgetary limits you have on an agency’s ability to change, the cost to train a work force, when you factor in all of those different things, its very difficult to come up with something that all the agencies are going to embrace, and be willing to accept, and which the public accepts… it’ll be interesting.”

Clint Hendler is the managing editor of Mother Jones, and a former deputy editor of CJR.