Here's an interesting (if frustratingly inconclusive) article from a guy named Nassim Taleb, a mathematician/philospher. It's about what Taleb calls "Black Swans", large-scale historical anomalies which were basically impossible to predict (such as 9/11).

He notes that the current 9/11 commission is flawed in three specific ways:

• Excessive and naïve specificity -- By focusing on the details of the past event, we may be diverting attention from the question of how to prevent future tragedies, which are still abstract in our mind. To defend ourselves against black swans, general knowledge is a crucial first step.
  
• Hindsight distortion -- Before 9/11, the risk of terrorism was not as obvious as it seems today to a reasonable person in government (which is part of the reason 9/11 occurred). Therefore the government might have used its resources to protect against other risks — with invisible but perhaps effective results.
  
• Flawed system of rewards -- We can set up rewards for activity that reduces the risk of certain measurable events, like cancer rates. But it is more difficult to reward the prevention (or even reduction) of a chain of bad events.
  

All this is very logical-sounding, and Taleb concludes that we must be forward-looking, not backward-looking (which I'm generally an advocate of), but honestly I'm not sure what modes of thought or courses of action Taleb would recommend.

Let's look at his flaws one by one:

If we didn't look at specifics, we might overlook important patterns in the future. Box cutters, flying lessons, one-way airline tickets...these are all specific points of reference, and they now send up red flags. According to Taleb, should we not pay attention to specific methods and tools that have led to black swans in the past? If we ignore them in favor of general trends, aren't we putting ourselves at risk from a repeat event? Many people think "Oh, they'll never use planes again...they've already done that." But in terms of actual policy and action, does that mean we shouldn't take measures to prevent exactly the same thing from repeating?

On hindsight risk, he's saying that a reasonable person would not have advocated allocating resources to fighting terrorism with the vigor we are now because it was not seen as a very great risk. I've got news for him...some people still don't see it that way. I still hear the lame-brained argument that since smoking or fatty foods or diabetes or disuse of seatbelts kill more people every year than terrorism, the money spent fighting terrorism would be better spent in those areas.

And in terms of reward, he's right that it's difficult to reward a person or organization for something that doesn't happen. But how to change that? There are, I hope, dozens (perhaps hundreds) of different specialists in our intelligence agencies, each devoted to a particular specialization (port security, cyberterrorism, bioweapons, nukes/dirty bombs, kidnapping, hijacking, and so on). And then there are all the scenarios related to terrorism that haven't been contemplated. How do we distinguish between a threat that was avoided through vigilance and systemic success from one that was really never a viable threat to begin with?

In short, what kind of policy is Taleb proposing? How, in particular, do we reward people for "thinking the impossible", as he suggests?

I basically think there's always going to be this tension between what's reasonable to expect, and how to prepare for it, and the unexpected, which in many ways is either impossible to prepare for, or which many people will not be compelled to prepare for without reasonable cause.

We'd make fun of any American who would build a bio-tent in their backyard and buy gas masks for their entire family. But if there were multiple, coordinated attacks in many major cities that killed hundreds of thousands of people with poison gas, people would be stocking the hell up on those things, and there'd be a commission asking why we didn't see it coming and why the government didn't provide every American with gasmasks before the event.

There are people now who mock the precautions we are taking, even after 9/11. Do we really expect the government to take all necessary means to prevent every possible type of attack? Of course not. The means and possible results are far too varied.

The best course of action, it seems to be, is to do your best to create conditions in which black swans are less likely to arise. Ideally, power internationally would be diffused, and even the notion of strong Presidents or Prime Ministers would be antiquated (one day, perhaps, we'll get there). But so far the best form of governance we've come up with is one which entails checks and balances, most directly in the form of elective office.

Of course there are those who disagree with our actions in Iraq, but it seems to me the idea of regional reform in the Middle East was the most proactive and forward-looking policy we could have adopted. Because there are too many potential black swans, and we would never be able to anticipate or prevent them all...but the corrupt, autocratic conditions that give rise to them, we might.