Well, the incompetents in the Government have done it again, releasing a Pentagon report in redacted form — only we can read much (if not all) of what was scrubbed out. You’d think that maybe they’d learned something in light of the report on the shooting death of Nicola Calipari, where redacted sections were “uncovered” and posted in the Italian media and various blogs around the world. But you’d be wrong.
On Tuesday, the DoD’s Inspector General submitted to the Senate Armed Services Committee a report on the refueling-tanker aircraft leaseback contracting fiasco that led to jail time for several key players, including Boeing’s former CFO and the Pentagon’s No. 2 procurement official (who’d later obtained a lucrative position at Boeing).
The report, as released to the public on Tuesday in conjunction with the hearing, contains two different types of redactions. One type, however — containing material removed at the request of the White House Counsel staff — can actually be read, after a fashion. Follow the jump to see how.
The redactions (or masking) in the report, “Management Accountability Review of the Boeing KC-767A Tanker Program” (warning: 270-page PDF), are of two types: text that’s been blacked out, presumably by the Pentagon, and text that’s screened off by gray boxes with stars. [Update as of 10:45am 6-8-05: At some point overnight, the DoD Office of the Inspector General pulled down the report, presumably in order to replace it with a version where we can’t read any of the redactions. I had downloaded a copy of the entire report but would need a place to host the PDF file (4.4MB).] Examples of the former can be found at pages 19, 27, 31, 124, 127-131, etc., of the PDF, while starred-gray redactions can be seen at pages 28, 29, 32, 122, etc.
As noted in the Washington Post, in a story posted prior to the report’s formal release,
What the story fails to mention, however, is that text within the gray redactions can be determined, albeit indirectly. As noted by TPMCafe user Cy Guy (see response to this comment I’d posted there), the PDF is fully searchable — including the supposedly redacted text. Here’s how to read the “hidden” text:
Using your standard Adobe Acrobat Reader, scroll down in the document to page 231 of the PDF (page 219 of the report). This page contains both black and gray redactions. Activate the Search function and enter the term “White House”. This should result in 45 hits total. There should be a line in the Results box for this page that reads, “a White House meeting on the tanker“. This matches up exactly to the second line of the visible paragraph near the top of the page, and the short starred-gray portion will be highlighted (click in any white space on the report page if the highlighting appears in black to see this).
Okay, so we can see that they took out this reference to the words “White House” here. “Not such a big deal,” you say? “What I’d like to see is the text of the e-mail just below, which they redacted in its entirety.” Read on.
Now click on the next line from the Results box, which reads, “a White House official] and [an“. See how it highlights within the big gray box? Now click on the next result (“White House] this AM as scheduled.“). Using this as a new base, say, run a New Search on the phrase “AM as scheduled”. You should get one result (this same page), only now there are another few words to fill in (“this AM as scheduled. I think the meeting“). You could now search for “think the meeting” and get the next few words. And so on, and so on . . .
All that’s needed to build out each of these starred-gray sections is one word within it, and from there, the entire redaction can be reconstructed, piece by piece, using words or phrases that can be inferred from the visible text. Moving forward through each gray redaction seems to be much simpler than working backwards, since only one word shows up in the Results box before the searched-for item. Perhaps someone more technically adept than I could figure out a way to run these searches automatically.
Now, while the Calipari report could apparently be downloaded into a text file allowing for the unmasking of redacted text, the same doesn’t appear to be true here. But the fact for this report remains that with a bit of time and effort, anyone can crack open and expose a whole lot of text that the White House Counsel’s office doesn’t want us to see. Anyone want to take on this project?
What’s truly astounding about this redaction screwup is that (a) the Administration had just been through a similar situation barely a month earlier, and (b) anyone with half a brain would know that this could have been prevented merely by generating a hard copy of the redacted document and then rescanning it — Adobe Acrobat would thus have no way of “reading” and indexing the text in the blocked-out areas. I’m betting that whoever was responsible for this gets sacked rather promptly.
Only a few months into Bush’s second term, and the pervasive incompetence is really starting to seep through. We can all hope that this is just the beginning of a recurring pattern of inattention on their part. [UPDATE: But seeing how the report got pulled down for cleansing, it’s also important that we be quick in downloading these files before they’re made to disappear.]