Friday, September 07, 2007

On Killing

The Psychological Cost of Learning to Kill in War and Society, by Lt. Col Dave Grossman

The subtitle tells the point of the book. Why do so few soldiers kill, and what happens to them when they do?

The first clause may seem counterintuitive: quite a few soldiers were killed during the Civil War, WWI and WWII, what is the author referring to?

He brings up the Napoleonic wars as evidence as well, and tries to reference the wars among the Greek cities (though he's on very shaky ground there). If you use the accuracy of the Prussian soldiers (measured at firing ranges), count the number of men in a square who can fire at any one time, and see how close the enemy soldiers were standing; you estimate a casualty rate several orders of magnitude higher than what actually appears to have been the case. From this he concludes that most of the soldiers were “posturing” (firing in the air to frighten) because they couldn't bring themselves to kill. In this case there's another explanation: the smoke helped conceal, and, as Devil Anse Hatfield pointed out, it isn't all that easy to hit somebody who doesn't stand still.

His Civil War inferences are more solid: the same problem of far more shots than casualties; but in addition there's the tidbit that when scavenging the battlefield afterwards for discarded muskets, almost half were not just loaded but double-loaded. Your buddies beside you can't quite tell if you shoot, but loading was done standing up, and everybody could see if you were reloading your weapon. A few muskets had been reloaded almost to the muzzle! Misfires could not have been that common—it wasn't raining.

World Wars I and II were conducted with a lot more attention to effectiveness, and after-the-war interviews disclosed that in WW II only 15-20 percent of men in combat fired their weapons.

So how come so many died? Napoleon thought the answer was artillery, and it was certainly lethal during the Civil War as well. In WW I the answer was the machine gun—which is a remote weapon manned by a crew. It would be useful to cross-check this with casualty reports from army hospitals, but this the author does not do. Were most injuries/deaths from infantry rifle bullets or machine gun bullets or artillery damage? The data is probably available somewhere.

Aha. Found something. An essay on medical care in WW I. “In the American Expeditionary Force (AEF), 20,420 men were treated for wounds inflicted by rifle balls, a combined 51,226 were treated for wounds by either shrapnel or shell and an additional 74,883 were treated for wounds by an unspecified gunshot missile.[14] The Surgeon General's post-war report concluded that, "Taken as a whole, the percentage of wounds from exploding missiles probably varied, from 50 to 80 percent being the highest when battle conditions were most stabilized, as in trench warfare."[15] This differs significantly from previous wars, such as the American Civil War, in which the Minie ball caused 94 percent of all battle wounds and artillery shell and canister approximately six percent. “

”The most prevalent battle injuries admitted into Medical Corps hospitals involved the lower extremities; 78,032 men were admitted with these injuries, with 5,722 (7.33 percent) eventually succumbing to their wounds. A list of the causes of these injuries, with the concurrent fatality rates listed in parenthesis, shows the emergence of artillery: rifle ball, 10,582 (3.69%), shell, 9,371 (9.10%), shrapnel, 18,182 (5.64%), and not specified, 39,897 (8.65%).[34] “

This does not deal with fatalities, which of course could result in a major bias in the data, unless rifle shots are relatively randomly aimed. (Gas killed about 3%, and can be ignored here.)

The Army took this to heart, and in Korea the rate was 75%, and in Vietnam 90%. Now whether this meant actually shooting at the enemy is an open question, as the author emphasizes. Current training is far more realistic, and American soldiers are significantly more effective than before.

Which, of course, means that they are more effective and accurate at killing. (And not killing when the they're not supposed to. I've used that shoot/no-shoot police training system—you have a fraction of a second to tell danger from innocence. They get good at it. Our group wasn't.)

At a distance, where there's no personal interaction, it is quite easy to kill. When surrounded by comrades running a war engine (as in artillery or machine guns) the social pressure is high and the guilt is divided (and the distance is usually large as well). Closer up, it is more traumatic for most men, and a common feature of the stories he quotes are of soldiers going off somewhere to throw up for a while.

The result, except for the 2-3% who are not empathetic, is a strong sense of guilt for those who had to kill an enemy close up, especially if they watched him die or found some humanizing details (family photo, etc) about him. Sometimes they could suppress this feeling of guilt, and sometimes not. The stress of guilt, lack of sleep/food/shelter, constant danger, and sense of hatred by enemies generally makes psychological casualties of soldiers after a few months in constant combat, which is why we do lots of rotations.

So what do you do with them when it is time to come home? He says many cultures demanded a purification ceremony first, but does not pursue this very interesting possibility. Instead he mentions the salutary effects of: spending a few weeks coming home with the buddies you had spent your time with, so that you could talk through things with people who understood; and being celebrated (parades, etc) by the population at home. Vietnam had neither. Soldiers were moved around quickly, did not apparently spend substantial time together in the field, much less on the way home; and the welcome they received is notorious (and apparently every bit as bad as was claimed).

OK, let's think about that “purification” angle again. The idea seems promising: a soldier feels guilty and needs a transition from “killing-ok” to civilian life. The first problem is that we as a nation do not have a single culture anymore, and any given religious ceremony of requesting/granting forgiveness, while it might be very beneficial, won't be understood/accepted by most of the troops. You would need many such types of ceremony, and the large chunk of effective atheists would be left out.

It is possible to train your population and soldiers to regard their opponents as subhuman, and this makes it much easier for them to kill, and to kill without guilt. Of course it means they're more apt to commit atrocities as well, which often work against you in the long run. In the short run, terrorism works quite well—much better than bombing from a distance.

It is also possible to desensitize your soldiers, so that they are not bothered so much by shattered bodies. We do this to some degree in training; other armies have been much more deliberate about it. The child soldiers in Africa are required to kill someone (preferably a family member) as part of their initiation. The SS were required to kill the pet dog they had trained.

And, as the author points out, our movies and video games are admirably designed to desensitize us (and especially impressionable youth) to violence and gore. This is intuitively obvious to any observer, but is also backed up with considerable research, but apparently we just do not care as a society that we immerse ourselves in extremely brutal violence. What you are immersed in affects you; advertisers know this quite well.

One major difference between the Friday the 13th desensitization program and that used by the army is that the latter includes discipline: when not to shoot. It will not surprise you to know that violent crime rates keep rising even though overall crime rates have fallen recently.

I suppose I should get back to that 2-3% of non-empathetic folks who kind of like killing. It turns out that they can come in several flavors. Some are the psychopaths who make ordinary life hell for everybody else: wolves. Others are disciplined sorts--“lawful good” for the gaming crowd--who like downing baddies: like sheepdogs. They're useful and good to have around.

Since various subsections of the book are meant to stand alone, there is a lot of repetition in the book. His anecdotes are compelling. (He describes himself sitting in a swamp eating live frogs as though this were the most natural thing in the world.) The book is not for the squeamish. If you can't read it, at least think over some of the problems he mentions.

UPDATE: See a later post about some problems with the numbers Grossman relied on. I don't think it invalidates his work.

No comments: