Skip to content

National Security Archive Director Tom Blanton Testifies Today on “Arbitrary and Capricious Classification System”

December 7, 2016


This morning the Archive’s Executive Director, Tom Blanton, is testifying before the House Oversight and Government Reform Committee for its hearing on “Examining the Costs of Over-classification on Transparency and Security.” Below is a copy of Blanton’s “must read” testimony.

Testimony of Thomas Blanton, Director, National Security Archive, George Washington University  

Before the Committee on Oversight and Government Reform,

U.S. House of Representatives 

Hearing: Examining the Costs of Over-classification on Transparency and Security

Wednesday, December 7, 2016

Rayburn House Office Building, Room 2154, 9 a.m.

Chairman Chaffetz, Ranking Member Cummings, distinguished members of the Committee, it is an honor to be invited to testify today before this Committee, which provided so much leadership for the passage this year of the Freedom of Information Act amendments, signed into law by President Obama on June 30.

Already your far-sighted reforms have driven real change in the bureaucracy.  For example, the CIA finally had to release their internal draft history of the Bay of Pigs disaster, revealing – horrors! – that the Agency suffered a nasty internal power struggle afterwards – hardly a national security secret, just bureaucratic “dirty linen,” as the suppressed history remarked.  The 25-year sunset you imposed on bureaucratic drafts, on agency deliberative process, the 5th exemption to the FOIA, really works.  Thank you!


That success, the 25-year rule, and your whole legislative approach to reforming FOIA, needs to be applied here today, to the classification system.  It’s time for Congress to step up to its Constitutional Article I responsibilities and write a law to govern an out-of-control, dysfunctional, counter-productive classification system.  Until now, you’ve pretty much left it to the Article II folks, who claimed as much power as they could get away with in the name of the Commander-in-Chief.

A law to govern classification – that was the number one recommendation of the Moynihan Commission 20 years ago.  They asked me to testify back then, and I’m sorry to report today that most of what they recommended never happened.  It is worth looking back in order to look forward.

That Moynihan Commission was quite an effort.  The formal title was “The Commission on Protecting and Reducing Government Secrecy” – some people saw that as contradictory, protecting and reducing, but I believe this is just the common sense notion that the only way you can truly protect the real secrets is by releasing the non-secrets.  The Commission report quoted Justice Potter Stewart in the Pentagon Papers case, saying “when everything is classified, then nothing is classified.”  My own metaphor at the time was:  We have low fences around vast prairies of classified information, when what we need is high fences around small graveyards of what could really hurt us.

I’m here today to tell you, we’re still stringing two strands of barbed wire around the prairies.  To compound the problem, we’re deploying our armored cars to go round up unclassified emails like Hillary’s, instead of focusing on the real hazards, like the millions of hacked security clearance files at the Office of Personnel Management.  I’ll come back to that point, about priorities, about whether you can believe anything a securocrat tells you about what’s classified (you can tell from the outset that I’m a skeptic).  But let’s start with the lessons from the Moynihan Commission.

The Moynihan Commission came up with unanimous bipartisan recommendations – and not just the usual suspects – not only Pat Moynihan, Democrat of New York, but also Jesse Helms, Republican of North Carolina, John Podesta of Washington D.C., and Larry Combest of Lubbock, Texas, then the Republican chair of the House Intelligence Committee.  Among others.

The Moynihan Commission reported a range of findings about how much over-classification there was.  From insiders administering the classification system, they received testimony that the problem was only a 5-to-10-per-cent overage.  But the final report treated with far more credence the estimate from President Reagan’s top National Security Council staffer, Rodney B. McDaniel, that only 10 per cent of classification was for “legitimate protection of secrets.”  (Commission report, p. 36).  That is, 90% over-classification.

My own experience at the National Security Archive, with more than 50,000 Freedom of Information Act requests over 30 years, and millions of pages of declassified documents, tells me that McDaniel is especially on target with his 90% when it comes to historical records.  For more current information, you can’t get a more informed and independent view than from the head of the 9/11 Commission, Republican Governor Tom Kean.  Kean was looking at all the intelligence on the 9/11 attacks, all the signals intercepts and CIA assessments of Al Qaeda, and commented publicly that too much secrecy had been part of the problem that left our country vulnerable: “Three quarters of what I read that was classified should not have been.” (Cox News Service, July 21, 2004)  So 75% over-classification on very current national security information.


The Moynihan Commission was greatly impressed with the costs of secrecy – not so much the dollar costs, however substantial, but the detriment to open research that would keep the United States ahead of the rest of the world technologically.  A central theme of the Moynihan report concerns the ways classification retards scientific and technical progress by compartmenting information and stifling the scientific method.  One 1970 study organized by scientists and cited by the Commission even suggested that “more might be gained than lost” if the U.S. unilaterally adopted “a policy of complete openness in all areas of information,” but given existing realities, recommended a 5-year sunset on scientific and technical classification.  In effect, the Moynihan Commission attributed the American national security advantage to our society’s open flow of information, rather than the potential development of thicker vaults to rival the Kremlin’s.

That finding is still true today.  No less an authority than former Los Alamos Laboratory director Siegfried Hecker describes in his latest book, Doomed to Cooperate (p. 402), how excessive secrecy and compartmentalization actually produces a “negative impact on nuclear weapons stewardship.”  Hecker criticizes government “overreaction” to allegations of security breaches, ramping up security at the expense of the research environment in ways that have “undermined the effectiveness of the labs.”

After extensive hearings, the Moynihan Commission concluded that our secrecy system was broken and needed a statute to fix it.  That law would mandate changes to the thought process around making the initial classification decision.  The classifiers should have to consider the public interest in release, the cost-benefit ratio, the actual vulnerability of the information, the long-term cost of keeping the secret, and not just whether it might damage national security, but all the other factors including the benefit from disclosure.

Key to the new statute would be a new concept of a “life cycle” for secrets.  Restrain the decision on the front end so you have fewer to start with.  Continuously push the unneeded secrets out of the system so they don’t stack up and gum up the information flows you need in any efficient decision-making process.  Minimize the amount you keep for the long haul, by using sunsets like the 25-year-rule and automated processes for release.

I have to say, this was especially prescient.  The World Wide Web was only a couple of years old at the time of the Moynihan Commission report (1997).  Google was still a year away from even launching.  But already electronic systems were proliferating documents at a rate the old carbon-copy secretaries would never have imagined.  What we now know is no matter how far you reduce the number of “original classification authorities” and no matter how far you bring down the number of “original classification” decisions, the capacity of computer systems to produce infinite copies means that the classified universe is expanding faster than the Big Bang.  That means the costs keep going up – $16 billion plus in the last fiscal year – and even more of a problem, the declassifiers will never catch up.  That’s why we need a statute that puts some automatic sunsets into the mix:  no more page-by-page reviews, if a document is in a certain category, it’s public after 10 years or 25 years.

The bureaucracy will object.  They’ll say every document has to be reviewed in case there’s a Social Security number in there, or a phone number, or other data protected by the Privacy Act.  But this is a formula for perpetual backlogs, a system that chokes on its old secrets, and of course, full employment for retirees doing the reviews.  Instead, we need to apply computing power to search and sort and protect privacy – standard formats like SSNs and phone numbers should be the easiest for automation to deal with.

The Moynihan Commission also recommended creating a central office to run classification policy.  They found all kinds of confusion and bureaucratic tussling between the Security Policy Board and the Information Security Oversight Office.  Frankly, as an outsider, I didn’t see this issue nearly as interesting as the bureaucrats found it.  But what we ended up with, as a combination of the Commission’s attention to this problem, and the 1995 Executive Order that set up an appeals process for mandatory declassification review requests, was an interagency panel that has been a rousing success – not least as a result of its staffing from the ISOO.

This is the Interagency Security Classification Appeals Panel – or Icecap as we call it.  The Panel has overruled the agencies in favor of requesters more than 70% of the time – yet another hard data point about over-classification.  Turns out that simply moving the decision about declassification out of the hands of the original agency makes a huge difference, even when the originators still have a say.  Yet, as useful as the Panel’s decisions have been, we’ve seen little evidence that the bureaucracy has learned anything from them.  We have to keep going back to the Panel, and the backlog there keeps growing, with some cases dating back a whole decade.

In my Moynihan testimony 20 years ago, I highlighted the huge successes in declassification that Congressional statutes had accomplished in creating the Nazi War Crimes board and the Kennedy Assassination Records Board – those two reviews combined to compel the release of tens of millions of pages of historically valuable records that would have otherwise remained secret indefinitely.  Without these statutes, we would never have seen the CIA’s file on Adolf Eichmann, or on Eichmann’s deputy whom the CIA recruited after the war and installed as a well-paid vice president at Proctor&Gamble in Cincinnati.  These records were technically still classified, but the Congress made a finding in law that the public and historical interest outweighed the bureaucratic factors.  We need such a finding across the board on classification, in statute, with an oversight board that can order openness and re-balance the secrecy teeter-totter.

But instead of a government-wide Declassification Board that could break the logjam on whole series of historical files, we got the limited ISCAP handling only mandatory review appeals (and only a few hundred of them, usually for individual documents), authorized by Executive Order rather than statute.

Congress did legislate an advisory function in this arena, the Public Interest Declassification Board.  The Pidib (as we call it) has become a helpful and responsive sounding board, producing useful recommendations, and even weighing in on some priority declassifications; but it is not the kind of drag-the-quarry-back-to-the-cave operation we need, or that the JFK and Nazi war crimes boards provided.

The statute you write needs to combine the ISCAP and the PIDB, by adding outside blue-ribbon nominees to an inter-agency panel of insiders, and giving the new body the power to overrule agencies and order the release of batches of documents.  The new body should turn its decisions into binding guidance on the agencies.  Such guidance is desperately needed.

The Moynihan Commission recommended that the CIA Director produce a new directive that would only withhold sources-and-methods information if there was a demonstrable harm from release, not just any and every method.  Such a directive has never happened, and there’s hardly a CIA Director born who would ever give away power so cavalierly.  So Congress has to do it, put this recommendation into the statute, there has to be demonstrable harm from the release of the source/method or else it can’t be withheld.

Instead of a rational cost-benefit approach, however, the last 20 years have only demonstrated the CIA’s burka approach to redaction.  Look at the President’s Daily Briefs that the CIA produced for Presidents Kennedy and Johnson and finally declassified (partly) last year.  We had gone to court to get the Briefs released for their historical value, but the CIA opposed us on the grounds that the very document itself was an intelligence method.  After the courts finished laughing at this, they allowed the CIA to withhold the two Briefs we had asked for (from Lyndon Johnson’s presidency), but denied the CIA’s claim for a “per se” exemption for all the Briefs.


With some pushing from the CIA’s own historical advisory group, the CIA finally started releasing the Briefs last year, even though many looked like Swiss cheese from the redactions.  One white blotch seemed familiar – the claim was “sources and methods” – but we already had a copy we had used in the lawsuit, found at the LBJ Library before the CIA began its absolutist claim.  That censored paragraph?  Our other copy showed the redacted source was our United Nations mission quoting foreign officials in New York.

What’s the secret?  My guess is that the CIA doesn’t want us to know that sometimes somebody in the State Department can actually come up with useful information.

Another major Moynihan Commission recommendation focused on centralizing declassification in a National Declassification Center.  This took about 10 years to see the light of day (that’s one measure of bureaucratic resistance).

The NDC does exist, and cranks out the low-hanging fruit from the classified trees, but it has little power over the agencies, and continues to pursue a hugely wasteful approach where one classified word can keep a document denied from release, and send it into the pile that has to be re-reviewed down the road.  That pile has taken on Jack-and-the-Beanstalk proportions.  Here the Moynihan Commission apparently bowed to the wishes of CIA director John Deutch and said the NDC “would not supersede agency control” over declassification decisions.  A decade of real experience shows that if NDC keeps avoiding any superseding, the backlog of historical classified records will overwhelm the system, especially with the tsunami of electronic records already in the pipeline.

We need to draw a line at least on the historical records – after 25 years, agencies have to turn over to the NDC the authority to declassify, and if the agencies want to keep a hand in, they have to put in real funding and real detailees into the NDC process.  Even so, the NDC should make the decision, not the cold dead hands of the bureaucracy.

Which step ensures improper withholdings are not occurring. (Hint: none.)

Which step ensures improper withholdings are not occurring. (Hint: none.)

This is especially true at the Presidential Libraries, where the process to open records is excruciatingly slow.  My organization obtained Mikhail Gorbachev’s transcript of his Malta summit with President George H.W. Bush two decades before the Bush Library was able to declassify the American version.  Now, 25 years after the end of the Soviet Union, we’re finally able to publish all the summit conversations between Gorbachev, Reagan, and Bush – and the American side, not the Russian side, was responsible for almost all the delay.  At the Presidential Libraries, a researcher has to file a Freedom of Information request just to get a group of files “accessioned,” which can take years, and then come back to the library, go through the boxes full of withdrawal sheets listing still-classified documents, and file individual Mandatory Declassification Review requests for those, which takes more years.  The National Declassification Center should be a geyser of Presidential records, centralizing the review, saving time and money.  A new statute on classification could make it so.

As for the other Moynihan Commission recommendations on areas like standardizing security clearance procedures, I can’t speak to those.  Not my expertise.  I’ve never had a clearance, and I don’t want one.  I remember the late 18-term Congressman George Brown, who saw the commercial potential in spy satellites (we take it for granted today in our traffic apps and Google maps), but the securocrats prohibited him even from talking about the possibility, so he resigned from the House intelligence committee so he wouldn’t be bound and gagged.

Well, that’s what a single securocrat can do today.  Bind and gag by claiming classification.  Other officials with equal or more seniority and expertise may well disagree, but all it takes is one securocrat and the whole system grinds to a halt.

Let me show you our latest example.  At the National Security Archive, we’re hardly even surprised any more.  We’ve been publishing these compilations called “Dubious Secrets” on our Web site for more than a decade now – side-by-side examples of different versions of the very same document, one section blacked-out here, but left in full over there.  Sometimes the documents have almost mirror image redactions, so when you slide them together you get the whole text.

In other words, I’m about to show you some documents that senior government authorities with the power to say so insist are classified.  Yet at the same time, these very same documents have been declassified by senior government authorities also with the power to say so.  All of which is to say, don’t believe them until you see for yourself.  Always ask, where’s the damage?

Here’s a document still classified, you can see all the black blotches, this was a decision just a month ago, in November.  The Joint Staff at the Pentagon deemed this document very sensitive, even though it dated all the way back to 1986, 30 years ago, and it was about the Soviet Union, a country that no longer exists.  But the document looked familiar.  Our expert on the topic, Dr. Bill Burr, thought he’d seen that headline and title before, and he poked around in our files.  Sure enough, back in 2010, the Headquarters staff at the Pentagon had declassified this document from a copy in another file.  In full.

So now we can read from six years ago the text that the Joint Staff thinks, right now today, is really sensitive, classified, worth spending taxpayers’ dollars on protecting, can’t be looked at by you or us in any public setting.  And what’s in there?  Just the Joint Chiefs’ comments on a draft presidential directive for our mid-1980s strategy against the Soviet Union.  No weapons systems design.  No intelligence assets.  It’s a waste of the Committee’s time even to read this out loud.

That made us go back to the cover letter on this document, this November.  In there, the Pentagon tells us that neither OSD (Office of the Secretary of Defense) nor NSC (National Security Council) had any objection to declassification in full, but a single securocrat, Mr. Mark Patrick of the Joint Staff Information Management Division, decided to exercise his Sharpie, or his computerized black-out system. What a travesty of national security.

Two versions of the same page of White House e-mail, addressed to then deputy national security adviser Colin Powell. The reviewer classified the deletions each time at the Secret level, meaning he believed their release would cause "serious damage" to U.S. national security.

Two versions of the same page of White House e-mail, addressed to then deputy national security adviser Colin Powell. The reviewer classified the deletions each time at the Secret level, meaning he believed their release would cause “serious damage” to U.S. national security.

It gets worse.  At least with the Joint Staff example, it’s one office against another.  But consider this piece of White House e-mail, originally sent to Colin Powell because he had missed the meeting.  The declassification review, going through several thousand White House e-mails, looked at the version from Powell’s user area first, and blacked out chunks from the middle section.  A little over a week later, the review dealt with the sender’s copy, as written by the meeting note taker.  This time, the top and bottom sections were whacked.  Classified SECRET, meaning there would be serious damage to our national security if released.  Except they were already released and there was no damage.

We found out the punch line once both versions arrived and we put them together.  The same person did the review both times – a highly experienced reviewer with TS/SCI clearances.  He told me later there must’ve been something in the Washington Post the first time around that made Egypt and military aid seem sensitive, and the second time around he had forgotten the document and the only news stories were on the Iran-contra arms deals, so he blacked that out.

Fast forward from this piece of Colin Powell e-mail from the 1980s to February of this year, when somebody in the inspector-general line of work grabbed two of Colin Powell’s e-mails from the account he had on the State Department unclassified system (his main account was with and deemed them classified.  When reporters called Powell for comment, the former 4-star general, Presidential national security adviser, chairman of the Joint Chiefs, and secretary of state described the messages as “fairly minor” notes from ambassadors, and remarked, “I do not see what makes them classified.”  Later, Powell told NBC News (February 4, 2016), “I wish they would release them, so that a normal, air-breathing mammal would look at them and say, ‘What’s the issue?’”

That’s what we’ll ask when the purportedly classified Hillary e-mails ever see the light of day.  When we actually get to read the declassified versions of those 110 or so messages in 52 chains, my bet is we’ll find that those 8 chains supposedly containing TOP SECRET information started with newspaper stories, like the one in the New York Times about drone strikes in Waziristan in 2011 while the then-chair of the Senate Foreign Relations Committee was visiting Pakistan – and neither he nor the ambassador Cameron Munter apparently were informed ahead of time.  So millions of Americans can read the newspaper story and talk about it over the breakfast table or around the office water cooler, but not the Secretary of State.  The CIA has effectively extended its capriciously high classification level covering the drone program, which was anything but secret even in 2011, to constrain the most basic diplomatic discussion of what’s in the newspaper that day.

Already, one of the Hillary e-mails now classified at the SECRET level has emerged with enough metadata (the date and the To/From/Subject lines) to check with the author, Dennis Ross, a veteran of three decades at State and the National Security Council in highest-level negotiations and highest-level security clearances.  Ross had emailed the Secretary of State in September 2012 with unclassified thoughts about the back-channel talks between the Israelis and the Palestinians.  Ross told the New York Times (February 13, 2016) that nothing about the discussion should be classified, “It shows the arbitrariness of what is now being classified.”

That’s the problem:  an arbitrary and capricious classification system that lacks internal and external credibility and contains too many secrets.  This system shields government misconduct, obstructs Congressional and public oversight, retards scientific progress, and cedes enormous power to its enforcers, the securocrats.  It’s time to write a law that reduces government secrecy.  Thank you for your attention and I look forward to your questions.


Archive Director Tom Blanton during his first appearance on The Colbert Report in 2010.

Archive Director Tom Blanton during his first appearance on The Colbert Report in 2010.

Tom Blanton is the director since 1992 of the independent non-governmental National Security Archive at George Washington University (, which won the George Polk Award in 2000 for “piercing self-serving veils of government secrecy.”  He was the Archive’s first Research Director starting in 1986, and he has also served as the Henry M. Jackson Lecturer at Whitman College, the Lazerow Lecturer at Long Island University, and the Henderson Lecturer at the University of North Carolina at Chapel Hill.  Educated at Harvard University, he is series editor for the Archive’s online and book publications of more than a million pages of declassified U.S. government documents obtained through the Archive’s more than 50,000 Freedom of Information Act requests.  His articles have appeared in Diplomatic HistoryForeign PolicyWilson Quarterly, and the Cold War International History Project Bulletin, among other journals, as well as The New York Times, Washington Post, Boston Globe, Wall Street Journal, Los Angeles Times, and other newspapers.  His new book is The Last Superpower Summits: Gorbachev, Reagan, and Bush: Conversations that Ended the Cold War (New York/Budapest: CEU Press, 2016, with Svetlana Savranskaya).  Other books include “Masterpieces of History”: The Peaceful End of the Cold War in Europe 1989 (New York/Budapest: CEU Press, 2010, with Svetlana Savranskaya and Vladislav Zubok), which won the Link-Kuehl Prize for documentary editing from the Society for Historians of American Foreign Relations; The Chronology (New York: Warner Books, 1987) on the Iran-contra scandal; and White House E-Mail (New York: New Press, 1995) on the landmark lawsuit that saved over 220 million records from Reagan through Obama.  His honors include the Emmy Award (2005) for individual achievement in news and documentary research, the Jean Mayer Global Citizenship Award (2011) from Tufts University, Harvard University’s Newcomen Prize in History (1979), and the American Library Association’s James Madison Award Citation (1995) for “defending the public’s right to know.”  The National Security Archive receives no government funding or contracts, and instead relies on publication royalties and donations from individuals and foundations for its $3 million annual budget.


Comments are closed.

%d bloggers like this: