Trust, But Please Don't Verify


The government's Information Security Oversight Office has come out with its annual report, which means that we can now compare the classification and declassification numbers of Bush's first four years in office. If I was handier with computers, I'd present this in graphical form, but I'm guessing you can still detect the trend lines:

Number of documents classified by the Executive Branch, annually:
2001: 8.7 million
2002: 11.3 million
2003: 14.2 million
2004: 15.7 million

Number of documents declassified by the Executive Branch, annually:
2001: 100 million
2002: 44 million
2003: 43 million
2004: 28 million

At those rates the classification/declassification numbers should converge by, what, 2006? I guess we'll just have to trust 'em!

My take on Bush/Cheney/Rumsfeld secrecy, and its historical roots in the Gerald Ford presidency, here.

UPDATE: One graph, coming up! (Thanks to commenter rich.)

UPDATE 2: See comments for Clinton's poorly-trended second-term numbers as well.

NEXT: Drudge Under On Fire

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. But Kerry would have been worse!


    Damn, I’ve been reading too much Slashdot….

  2. Shouldn’t it be “… if I WERE handier with computers …”?

    You, Henry, try to find a more sophisticated way to defend Bush. Switching the discussion to (speculative) behavior of Kerry is silly. Stick to the topic, i.e. to Bush!

  3. I think the secrecy of the Bush/Cheney administration has been over-the-top and decidedly harmful, but I’m not sure you can lay these particular numbers at their feet without context (or with only the context of Bush/Cheney’s admitted desire to not weaken the presidency).

    The number of declassified documents has been in freefall since 1997, when it hits its apogee of more than 200 million pages. During the last year of the Clinton administration, that number was down to 75 million. The executive order signed to release this stuff went into effect about 10 years ago, and most of the agencies involved are now running low on “low-hanging fruit” that can be quickly declassified, and they’re left with a load of documents that reference information from multiple agencies. You can imagine how much of a red tape pain in the ass it could be to declassify those.

    I’d focus on the specific cases of secrecy the administration has perpetrated, rather than the actual numbers of documents (which can be misleading).

  4. In fact, here’s Clinton’s second term if someone wants to make a graph.

    Number of documents classified by the Executive Branch, annually:
    1997: 6.5 million
    1998: 7.3 million
    1999: 8.0 million
    2000: 11.1 million

    Number of documents declassified by the Executive Branch, annually:
    1997: 204 million
    1998: 193 million
    1999: 127 million
    2000: 75 million

  5. Lend another hand, rich?

  6. Thanks for the update, Matt. The converging numbers are of some concern, but somewhat predictable due to the finite backlog of documents and the ever-expanding bureaucracy? of the federal government. It’s actually kinda amazing they bothered to go back and declassify so much old junk– props to Clinton for the executive order (and lesser props to Bush for not canceling it). Now, number of classified documents versus number of total documents generated by the feds, per year — that would be interesting, especially broken down per agency.

    I think the “number of total documents generated by the feds” alone would be enough to inspire a frown in libertarians.

  7. Check your mail Matt. I sent you Clinton’s broken out like Bush’s as well as two views for comparison.

  8. Wouldn’t it be better to just have presumption that documents X years old should automatically be declassified? There are no legitimate national security reasons for some WWII-era documents to remain classified but they are–all that remains is the possibility of embarrassment. Tough shit in a free society, I say. All of this bullshit is on the taxpayer dime to begin with.

    Forty, maybe thrity, years should be more than enough–automatic declassification, unless the Executive can show a real, present national security concern (or danger to lives). And, of course, many documents less than 30 or 40 years old should be let out of jail, too.

  9. Henry,

    That’s pretty much exactly what the executive order says. Check section 3.4. Isn’t it a rare and magnificent thing when government is one step ahead of you?

    It’s 25 years, with loopholes of course.

  10. Henry-

    In regard to your 40 year horizon, I would certainly agree that troop movements from WWII should be declassified. But anything that might reveal the identity of a spy or informant should probably remain secret.

    For instance, the last thing we need is for the name of a Japanese mole in WWII to be declassified, and then he gets a visit in his Tokyo nursing home from some angry guy whose father died in a battle where that info proved crucial.

    I know, unlikely, but you can never be sure, and anybody who risked his life to deliver information to us deserves to have his privacy maintained.

    And declassifying documents 40 years old could endanger the lives of informants in Cuba, China, North Korea, or other places where the regimes are more than 40 years old.

  11. Phocion:

    In theory, yes–but in practice it’s a joke. No doubt many interesting and illuminating documents have been released, but the WW2 example that came to mind arose from the still(!)-classified docs on Pearl Harbor. Robert Stinnett (regardless of whatever you think of his theories) has done yeoman’s work in getting some of this stuff declassified, but what possible legitimate purpose is there in keeping this shit under wraps? I’m just suppose to presume the government is acting in good faith, and not covering long-dead asses?

    Thoreau: Yes, your N. Korea example would squarely fit into protecting lives–no argument. Your Japanese argument is a bit more farfetched–if somebody in Japan might attack the decendents of such a spy, does that forever lock up these papers? Also, it seems you could solve much of the problem through very minor redactions, not wholesale classificiation.

  12. Henry-

    I’d be fine about releasing documents with only minor redactions to protect past informants.

  13. While W’s numbers look worse, looking at Clinton’s suggests to me that as an administration ages it becomes more secretive. I wonder if there’s data to see if there is an historical trend.

  14. According to something I once read at The Straight Dope, the only still classified documents dating back before the 30s all deal with a formula for some kind of invisible ink.

    I may be slightly off, but I’ll post the link if I can find it.

  15. This guy has a good graf as well, showing the crazy & ahistorical spike in declassifications that has been plummetting for five years now…. I’ve been too busy to add new ones to this post; maybe tomorrow.

    Of the two number-sets, it’s the *new* classification that worries me much more.

  16. Wow, during a war, more documents are classified and fewer are unclassified! Whodathunkit?

    Seriously Matt, this is about #329 on the list of things to worry about these days.

  17. thoreau and Henry, Re: WWII documents.

    I imagine that rather than trying to protect troop movements and individual spies from that era, they most likely are more concerned about ways and means and exactly how US intelligence goes about recruiting agents. They’re probably not interested so much in covering up 60+ year-old info as much as they’d like to keep people in the dark about how they actually go about obtaining info.

  18. One thing I’d want to check on these numbers — what are the TOTAL number of classified documents held?

    Clinton’s trend in declassifying documents could be increasing secrecy — or simply that the White House had worked through the backlog of “stuff that should be declassified but isn’t”.

    Hard numbers don’t mean squat — we should really be looking at percentages, and trying to guess how much was low-hanging fruit.

Please to post comments

Comments are closed.