When the government talks, what should you believe?

Our ZDNet Government columnist takes a deep dive into the question of leaks, background briefings, and the believability of government statements.
Written by David Gewirtz, Senior Contributing Editor

When it comes to government transparency (or the lack thereof), it's been quite the season, hasn't it?

Week after week, we've had the slow dribble of reports from the Edward Snowden collection of purloined documents. Week after week, we've had clarifications and denials, official statements, and on- and off-the-record background briefings from government officials.

We've also seen some of the most ridiculous (and no, I'm not linking to them) conspiracy theory claims from the nutball fringe of the blogosphere and from mainstream media who should have known better.

Who can you trust and what is the truth?

I've given this question a great deal of thought, going all the way back to 2007 and my work exploring the missing White House emails reported by the Bush administration. I even wrote a book about it.

In that book, I pointed out one example of how popular media distorts official statements for editorial gain. In an episode of The Daily Show, Jon Stewart's team cut together statements from two White House briefings.

They took a statement from one briefing on one specific topic that occurred on a Monday and edited it so it played right before a statement (on a completely different topic) that had actually been made on the prior Friday. When the two statements played together, in their cut-and-out-of-order form, the White House sounded like it was saying something completely different from what actually was said.

What was disappointing about this set of altered-statement editing cuts was that both statements were disturbing on their own, and could have provided a base for editorializing on their own, without having been altered in order to make whatever point Stewart and his team were trying to make.

Ever since I discovered The Daily Show's alteration of the public record, I've established a set of guidelines for where and how I will accept statements, and what I will consider fact and what I will consider unsupportable hearsay.

I will only accept a government statement as official if I can get it straight from the source. For example, I would consider the contents of a White House briefing as official if I watched that briefing via the White House's own feed. I will accept Congressional testimony as official if I retrieved the transcript from an official government repository.

What is official?

Before I go further, let me explain what I mean as "official". I'm not saying I accept these statements as fact. Statements by public officials are not, as we all know, necessarily the truth. But an official statement by a public official does become part of the public record and can be attributed with confirmation to that speaker.

Going back to the misleading Jon Stewart cuts from 2007, if a historian or political scientist were to try to base an analysis of the events of that particular weekend on the statements from The Daily Show, that historian would get an externally-altered view of what the government said.

Take, at the very extreme, a case where a reporter is claiming that the government lied about something. Quoting a potentially altered record like The Daily Show cut wouldn't make the case for the reporter, and might, in fact, backfire.

But if that same reporter made the case for a government lie based on official Congressional testimony, that story would be potentially far more explosive -- and possibly alter the future of the nation.

That brings me to two classes of government information that sometimes winds up in the public record: leaks and "on-background" statements. Let's do leaks first.

The unprovable veracity of leaks

We're now all too familiar with WikiLeaks and Edward Snowden. Bradley Manning, now convicted, stole documents which WikiLeaks then published to the public. Edward Snowden stole documents from the NSA and provided them to the British Guardian newspaper.

These two sets of leaks have ignited a firestorm of press. But can we trust them as historical fact and can we consider them undeniable elements of the public record?

The answer to that is "no," even though the overwrought blogosphere and weak-willed mainstream media has had a feeding frenzy on these documents.

While it is likely that the NSA does much of what the documents indicate, we have no proof whatsoever that these documents are being provided to us in pristine, unaltered, preserved-chain-of-evidence state. They're probably accurate, but even though just about everyone is treating them as fact, they're all sourced from one individual (who happened to run first to China and then Russia), and there's just no externally verifiable evidence of truth.

On-background briefings

That brings me to "on-background" briefings. These are briefings that some reporters and analysts get from high-ranking government officials, but which can't be officially attributed to those officials.

I reported on one of these briefings just a few weeks ago. I had the opportunity to talk to a few, specific, high-ranking members of the U.S. Intelligence Community about one of the NSA issues and get some detailed information.

While I know who I spoke to, I made the promise that in return for the rather open and unrestricted Q&A session, I would not attribute the statements made to any specific individual or individuals.

The good part about this sort of briefing is it does help us understand what happened. The bad part about this sort of briefing is it can't enter the public record as fact. After all, I restated what a government official said, and I withheld the specific information about who said it.

While the historical record on this particular NSA issue can report that I reported it, it won't ever hold the weight of truth that a statement that could have been attributed to a particular public official would have had.

Another example of this is the Stuxnet story where the NY Times claimed US released Stuxnet with Israel and it accidentally escaped.

Because none of the reports on Stuxnet have actually been attributed to specific American officials, we can't be sure -- 100 percent sure as a matter of record -- that America was involved. Oh, sure. America was probably involved, but the lack of an official statement means there's no smoking gun.

There's no surety.

Some would argue that it doesn't matter. After all, since we can't believe what our politicians say anyway, what does it matter whether the news comes from a hearsay source or from a so-called official source?

It matters. It matters because the historical record is very much like a court proceeding. For us, as a society, to know what happened -- what really, actually happened -- we need reliable source data. A government statement itself isn't necessarily reliable, but as I wrote in my White House email book, if the government is willing to officially state they misplaced five million email messages, you can be sure that's at least the best case for the story.

In other words, while you can't fully trust a government statement, if an official government statement includes a mea culpa, you can reliably expect that the worst of what the statement released is at least the least of the actual badness.

That brings me back to the original question of this column: when the government talks, what should you believe?

The answer has two parts. Part 1 is this: if the government makes an official statement and you can confirm that the statement is, in fact, from the government and not altered by the media, that statement can become part of the historical record.

It's not necessarily the truth about whatever incident the statement is about, but it's truth in the sense that it can reliably be attributed to the government.

Part 2 is this: if a government statements admits culpability, you can reliably assume that the baseline of the culpability is what is covered in the statement. It might be a lot worse, but at least you have a baseline for the historical record.

So there you go. I will still accept on-background briefings because it provides me with insight I can often share with you that otherwise wouldn't be available. But I will also push for attributable public statements that we can use to build up a better-sourced and potentially more accurate historical record.

Editorial standards