First appeared in print in The Rhinoceros Times, Greensboro, NC
Intelligence "Failures" and the Blame Game
The joint congressional committed investigating 9/11 has come up with a remarkable conclusion.
We need another layer of bureaucracy in the intelligence community.
We've heard all kinds of talk about how 9/11 could have been prevented if only the CIA and FBI and NSA and the Immigration Bureau and the Defense Intelligence Agency and the State Department and my cousin Bob had all shared the information they had.
And these congresspersons, who have never lost their childlike faith in the ability of bureaucracy to solve problems caused by bureaucracy (called "fighting fire with fire"), think that this can be solved by putting one superbureaucrat in charge of all the intelligence-gathering bureaucracies.
True, there was key information that wasn't shared, and should have been.
But the most telling information was missed, not because the agencies don't talk to each other, but because the primary role of each layer of bureaucracy in each intelligence-gathering organization is to filter out information and keep it from rising to the top.
Think about it: Say you have a thousand field agents. All of them are gathering information. Some of it will be valuable, some of it will not.
So they all report their information to superiors, who decide which of that information to pass on up the chain, and which to ignore.
Each level removes more information, because there is always far more data than any one human mind can grasp.
Furthermore, the way data is removed from the stream is highly selective. Each bureaucrat passes on the information that seems important to him.
This obviously means that he is not passing on the information that seems unimportant to him. Trivial information ... like the number of Arabic-speaking foreigners learning how to fly, but not land, large commercial jet aircraft.
We had that information and didn't use it, not because the agencies didn't talk with each other, but because the bureaucrats didn't recognize its importance and pass it up the chain.
Nobody meant to be stupid. But the pressure is always to remove data from the stream. You can lose your job in the intelligence bureaucracies by passing along meaningless, useless information as if it were important.
But information that is important often doesn't look important until it is combined with other information that is, but doesn't look, important.
Putting a new superboss in charge of intelligence gathering is a guarantee that less, not more, important information will get to the President and others who have to make key decisions about national defense.
Right now, the President -- and, perhaps more importantly, his top advisers on defense and international matters -- see reports from competing agencies that use different methods and sources to gain information. Which means that while each agency filters out what that agency thinks is unimportant, no agency can filter out what the other agencies think is important.
So the odds of surprising but important information sneaking through the system are actually higher than they would be if we put yet another layer of bureaucracy between the intelligence gatherers and those who make decisions based on what they gather.
What we need is not a single "intelligence czar," but rather to strip down and de-layer the existing intelligence bureaucracies. Allow more sharing of information among agents at low levels, without the filtration imposed by supervisors.
A lot of time will be wasted on making ridiculous, meaningless connections.
But I'll take a time-wasting system that can find unexpected connections over an "efficient" system that filters out everything that isn't expected.
As for Alabama Senator Richard Shelby, who singled out CIA director George Tenet for personal blame, along with NSA director Hayden and former FBI director Freeh, I think it would be wise for congresspersons to refrain from blame-throwing.
A lot of the obstacles that kept us from finding out about 9/11 and other intelligence "failures" in advance were put in place by Congress. Always with the best of intentions, of course ... just as each bureaucrat who removes data from the intelligence stream thinks he's helping focus our attention on the things that really matter.
For that matter, I'm tired of hearing newspeople nattering on about "intelligence failures."
We start out, in every dangerous situation, not knowing a thing about what our enemies are doing -- or even, in many cases, who they even are.
Any information that our intelligence agencies gather moves us to slightly higher ground and gives us slightly better perspective.
But the information we get will never be complete. Indeed, our decision-makers will never even have complete information about what we are doing. There is no such thing, outside the mind of God, as "complete" information.
In other words, intelligence agencies always "fail" because they can never find out everything. Also, much of what they find out will be false -- lies from the enemy, incorrect data, human error.
There are always gaps, and the gaps are always filled with guesswork.
Students of military history know that those gaps and guesses can be crippling. The Civil War would not have lasted as long as it did if northern commanders had not consistently overestimated the size of Confederate armies.
How did this happen? The Pinkerton detective agency gathered information; so did reconnaissance patrols. At each level, the estimates of enemy numbers got inflated. Why? Because the worst error they could commit was to underestimate the enemy and send an army into battle outnumbered without knowing it.
For if you underestimate the enemy's capabilities, you risk losing your army entirely -- total defeat.
While if you overestimate, the worst that can happen will be hesitancy and inaction. You never win a war that way ... but you're less likely to be the one blamed if you lose it.
Intelligence agencies' worst failures come from their own fear of being wrong. And there are failures -- wrong conclusions that lead to wrong decisions.
But you can never regard lack of information as a "failure." Lack of information is the natural state; each piece of information that turns out to be useful and accurate is a victory. But since we can't know what we don't know -- and don't know that we ought to know -- it is absurd to call lapses in information "failures."
What we get from intelligence services is the best information they can get, along with their best advice about which information is useful and which is not to be trusted. They will be wrong most of the time about most of their information. That is and always will be true, and it is not failure.
And if we create a climate in which every successful attack by our enemies is blamed on our own intelligence-gatherers, then we will find our intelligence agencies filled with very timid, careful, by-the-book, inside-the-box bureaucrats whose primary purpose in life is to avoid doing anything they might get blamed for.
It's like kickers in football. They get called in most noticeably when the game is going badly. When the team failed to get a touchdown but needs to get at least a field goal. When the team failed to get a first down but needs the punt to place the opponent as far from the goal as possible.
There is no kicker so good that he can always succeed in every circumstance. And when they fail, it often has dire consequences for the outcome of the game.
The coach doesn't fire the kicker every time he misses. Nor does he -- or the team, if they're smart -- blame the kicker for losing the game.
After all, the only time a game ever comes down to a single kick is when the rest of the team failed to prevent the opponent from scoring and failed to score enough touchdowns themselves.
The "blame" for 9/11 falls on Al-Qaeda. They are the ones who deliberately did what they did.
Beyond that immediate blame, we can find thousands of contributing factors, from deep flaws in the politics and culture of many Muslim nations to mistakes made and problems overlooked by our own government, leadership, and, yes, citizenry. There's plenty of "blame" to throw around, and not a thing to be gained by it.
It does help to find correctable flaws in our system and try to correct them so that we can get more and better information about what our enemies intend.
But we will never create a system that always prevents every attack by our enemies.
Our leaders will always have to make decisions based on imperfect information. Usually the consequences are trivial, or at least invisible to most of us. But sometimes the consequences will be terrible and obvious.
It's worth remembering that there is no great leader, politically or militarily, who has not made very bad decisions from time to time.
The great leaders are the ones who, despite the lack of sufficient information, nevertheless decided, acted, and when things went wrong made new decisions to deal with the consequences.
Here's what both General Grant and General Lee understood, that most of the failed commanders in the Civil War did not: The key to victory is not figuring out what the enemy is going to do, but to make the enemy worry constantly about what you are going to do.
The key to defeating terrorism is not to try to figure out what they're going to do in advance and prevent every attack. That job simply cannot be done. The key is to deprive them of every support they currently have -- recruitment, supply, shelter -- so that no matter what they might want to do, they won't have the means to carry it out.
Copyright © 2002 by Orson Scott Card.
Reproduction in whole or in part without permission is prohibited.