Learning to Kill

John A. Nevin

According to Lt. Col. Dave Grossman, whose book On Killing (Little Brown, 1995) prompted this essay, the Army developed new training procedures that mimic combat. A recruit crouches in a foxhole, rifle at the ready, and shoots at man-shaped targets clothed in olive drab that pop up randomly in the “battlefield.” Some of them move; some have concealed jugs of red paint. When a target is shot, it falls over and the red paint flows. This “quick-kill” exercise is repeated again and again, under a variety of different conditions, with rewards in the form of praise, public recognition, badges, and weekend passes for recruits who do especially well. And nearly all of them do well.

Their training carries over to actual combat. Grossman quotes a Vietnam veteran: “Two shots. Bam-bam. Just like we had been trained in 'quick kill.' When I killed, I did it just like that. Just like I'd been trained. Without thinking.” This veteran, and thousands of others, have to live with the tragic fact that they violated the most fundamental human commandment mindlessly, as a result of their training.

We can understand their training in terms of principles developed by the science of behavior, but this does not indict behavior science as a servant of military violence. Scientific principles are just there, like gravity, to be applied deliberately or inadvertently. You can't really object to gravity, but you can learn how it works – and if you object to organized killing, as I do, it would be worthwhile to understand the principles that underlie it.

The basic unit of behavior analysis involves three terms: the situation, the response, and its consequences. During training, the immediate situation is being in a foxhole with targets popping up, and the feel of a rifle in hand; the response is aiming the rifle and shooting the targets; and the consequences include the immediate “kills” and the subsequent rewards. But there is another crucial aspect of the environment: The entire military training situation, which instills obedience to orders and authorizes controlled, directed violence. Repeated training in situations that mimic combat and with consequences that reward accurate shooting virtually guarantees that soldiers will shoot to kill when authorized by their commanders and confronted with human targets, even when they are terrified by the noise and confusion and constant danger of real combat.

The underlying process is known as behavioral momentum, according to which previously learned patterns of action tend to continue in the face of severe challenges because of a history of frequent reward in similar situations. Usually, the momentum of learned behavior is valuable for the individual and society, as when an emergency medical technician uses lifesaving skills learned during training to save a life under the stress of a real emergency. But it can also have tragic consequences, as in the massacre at My Lai, Vietnam where training in the use of weapons and obedience to orders led soldiers to shoot old men, women, and children. Whether valuable or not, the behavior in question becomes obligatory: If training included frequent rewards for repeated responding in a given situation, a similar situation will produce the trained response.

The dependence of responding on situational similarity is important. For example, very few veterans of infantry training go around shooting people on the streets after returning to civilian life. The reason is that a trainee's learning was specific to the military situation, where strict obedience to orders is instilled and unauthorized action is severely punished. Firing at a target – or, in combat, killing another person – is specifically authorized by a commander's orders. In other words, military training produces controlled killing.

Are the recent episodes of killing by school children related to military training techniques? Lots of children are taught to fire real guns during target practice or hunting trips, with adult approval for hitting bullseyes or animals; and in some video games a child holds a simulated weapon and “fires” at human-shaped targets on a screen, where bloody chunks fly off after hits. The situation, confronting a target while holding a weapon, the response of aiming and firing, and the consequences of feedback and approval are similar to those arranged in military training. There is at least one important difference, however: Unless parents act like army officers and allow children to shoot only under strict orders with severe penalties for unauthorized firing, even in target practice and video games, the specific authorization to shoot and kill that controls and directs military violence is missing.

Many people have agonized over how children in their own families and communities could become killers, and many of their questions remain to be answered. One question, suggested by the argument above, is whether a child had been allowed to shoot and was rewarded for doing so in the absence of strict control. I don't know the answer, but this can be said for certain: The killings wouldn't have happened if the killer hadn’t had access to a gun. To prevent killing, don't own a gun. If you must own a gun, keep it securely locked away. And if you teach a child to shoot, make sure that each shot is fired only when an adult gives the order.

John A. Nevin is Professor Emeritus at the University of New Hampshire in Durham. He is best known for his experimental research on behavioral momentum. Discover Magazine has produced a television documentary entitled “Born to Kill” that deals in part with the topic of this essay. It includes commentary by Lt. Col. Grossman and describes current military training practices in relation to basic principles of behavior.

Return Home