Tuesday, April 6, 2010

Human Decisions and War

Back in July 1997, when the situation in Iraq was really bad, a US Army helicopter killed two journalists and about a dozen civilians. Yesterday, gun camera footage of the incident, along with the radio communication between the pilots and other soldiers, was leaked.

If you have not heard about this yet, you probably will. News shows and websites will probably show an edited version with added political commentary. Here is the full, clean, version. The shooting is in the first few minutes:

This incident brings up a lot of important points of human decision making.

Start with the decisions of the soldiers. People have been throwing around words like 'murder' and 'war crime'. This is wrong. There was certainly no intention to kill civilians. You can tell that the pilots honestly believed that the people were armed. The first part of the video clearly shows someone pointing something that looks like a rocket-propelled grenade at the helicopter.

But the thing being pointed at the helicopter was a camera. The people were not armed. The pilots made a mistake. The human mind, no matter how well-trained, is prone to all sorts of errors. People see things that are not there. They derive incorrect inferences from situational cues. In this case, an American patrol had come under attack in the area, which primed the pilots to expect, and see, armed resistance.

All of these cognitive errors are made much more likely in stressful situations. These pilots probably knew other pilots who had been killed by RPG's. They may have seen the mangled bodies of their friends. They had doubtless been shot at many times themselves. So when someone pointed a big black thing at them, they immediately assumed they were under attack, with tragic consequences.

This kind of thing is inevitable. It will happen, no matter how well you train the soldiers. You can reduce the odds of these mistakes, as our Army has done, but they will never go away.

Now, consider the decision-making process of political leaders who make strategic choices.

Making the decision to fight a counterinsurgency means that you are creating hundreds of thousands of situations where multiple innocent people could be killed by a single error in judgment. Even if our troops were 99.99% correct in their decision making process (an unrealistically high goal that will never be attained), there would be several incidents like this one where things went badly.

Anyone who does not take this fact into account will make bad strategic decisions. People have a very bad habit of making choices based on ideals and goals and hopes, rather than a hard-headed consideration of the facts on the ground. They know that we have a very well-trained military full of people who want to protect out country and would never intend to commit any war crime. But they fail to understand that good intentions do not necessarily generate good results. Chaos will happen.

Fighting a war in the modern world means that you will, almost certainly, create situations where innocent civilians are killed by your troops and this fact gets broadcast to a world with a lot of people who are looking for any excuse to make you look bad. If your warfighting strategy cannot survive multiple such incidents, then it is doomed to fail.

Soldiers, who know what combat is like, understand the chaos of war and its consequences. Most civilian leaders, who live comfortable, stress-free lives where things usually go as planned, do not. Good strategic decisions can only come from better communication, and an understanding of probability.

Whenever making any deployment decision, the politicians need to be told something like this: "This plan will cause at least three incidents where our troops kill a large number of innocent civilians, and this fact is released to the world. There is no way of reducing that number. Are you prepared to deal with the consequences of that?"

But nobody wants to hear that. They want to pretend that bad things will not happen. And when bad things do happen, they blame someone at the bottom of the hierarchy. This is wrong. Statistically speaking, these civilians were killed the instant a politician signed an order to go to war. Any moral sanction should be assigned to the people who pushed for war.

Now consider the decisions and thought processes of people who react to this and similar incidents.

People respond emotionally to images, not statistics. This video could be what ends up defining the Iraq conflict in people's minds; the modern version of the picture of the Vietnamese girl running from a burning village. They will not trust the official report from the Army. They will look at the video and make up their own minds.

They will fail to consider the state of mind of the soldiers. When I look at this video in my office, it is clear to me that the group of civilians is unarmed. I am not stressed, my life is not at risk, and I have the benefit of hindsight. It is all too easy to conclude that I would never open fire in that situation, and then to conclude that the pilots were wrong. But when I think of the mistakes I have made in stressful situations, I have a lot more sympathy for the pilots.

People will also not understand that this is perhaps the worst of hundreds of thousands of incidents where our soldiers had to make life-or-death decisions. It is the exception, not the rule. This video says nothing about the day-to-day existence of the soldiers in a combat zone, of the hundreds of times they put their own lives at risk to save civilians. But people will only remember the video.

The only way for the Army to counteract this effect is by releasing lots of other real-life unedited videos, showing terrorists committing crimes like using civilians as human shields. I know they must have lots of footage like that, but I have not noticed any effort to release and publicize it.

The critics of the Army are also making the mistake of assuming that intention equals results. They see a bad thing happening and assume that it was intended to happen. They are holding our troops to an unreasonably high standard (perfection) and calling them criminals when this standard is not met.

Yes, this video is dramatic evidence of a major blunder. Yes, it shows us that we need to work harder to prevent such things in the future. But it does not mean that anyone in the Army committed a crime.

One final note. It may seem that these things could be reduced by more use of robot drones in combat. If the pilots were not personally at risk, then they might be less likely to launch attacks because they feel threatened. But the evidence suggests the opposite. From what I have seen, the CIA drone controllers are far more likely to kill innocent civilians than the Army pilots who are in the middle of a fire zone.

No comments: