A few weeks after Rumsfeld resigned, I remarked to a friend over e-mail that I had some respect for him because of his “known unknowns” speech. He was laughed at for saying something very important: that there are things we don’t know, and we ought to not only try to find out what we don’t know, but that we must also recognize that there are some things we know so little about that we don’t even know that we don’t know them.

Similarly several months ago, when Secretary of State Rice said that we’ve learned from the thousands of mistakes we’ve made in Iraq, the mainstream media chose to not learn from mistakes, and instead use headlines with words such as “admits” and “concedes”. That only encourages future officials not to talk about what they’ve learned.

Most recently, the leaked Stephen Hadley memo about Iraqi Prime Minister Nouri al-Maliki is being cast as Bush doubting al-Maliki. In the past, the Bush administration has been criticized for not having a diversity of opinions; but when it clearly does so as in this case, that’s not evidence of multiple points of view, it’s proof that the Bush administration itself has already come to this conclusion.

If we’re lucky, this is just Bush Derangement Syndrome: it doesn’t matter what comes out of this administration, it must be spun to criticize the president. Hopefully, our next administration will be free to continue learning from mistakes and considering multiple possibilities without fear that the mainstream media will spin it into “we were wrong” and “we don’t trust our allies”. Unfortunately, this sort of thing tends to last: once the press tastes blood, they don’t care who they get it from.

There are some questions about how and why this was leaked, but this is a beautiful example of why administrations (and companies) often don’t encourage diverse opinions: when leaked, they’re usually spun in a way to reflect badly on the organization.

Possibly the most controversial over the long term have been plans for war against our allies. Considered “mostly academic”, they’ve still been characterized as the product of a military with “too much time on its hands”. But such plans are extremely useful precisely because they are in response to extremely unlikely threats. They highlight areas that we don’t know we don’t know.

We might have known in the seventies, for example, that Canada would never invade us. What was so far off of our radar that we didn’t even know we didn’t know it, however, was that our long border with Canada might provide a means for terrorists to bring explosives into our country.

An academic plan for war with Canada might have recognized that one major problem was the huge border and the ease of crossing it. A large truck crossing the border at any point could carry some massive explosives. Even if we never realize, when considering this possibility, that the large truck doesn’t have to be driven by a Canadian trooper, we still have a plan for dealing with mobile explosives crossing the border by truck. A plan for dealing with this could be useful if terrorists choose to use Canada (or Mexico) as staging grounds for launching attacks against the United States. It gives us a head start, and it highlights weaknesses in our current “real” plans even if we don’t realize it at the time.

It is extremely important that we consider unlikely scenarios. When we only plan for what we know about, we can easily not just not plan for what we don’t know about, we can encourage what we don’t know about. Our “plans” for avoiding loss of life during an airliner hijacking, for example, may have reduced lives lost to hijackings. But they encouraged hijacking airliners for other purposes. A culture supportive of rather than dismissive of wild “academic” defense plans might have spotlighted that flaw in our reasoning.