On Killing: The Psychological Cost of Learning to Kill in War and Society

by Dave Grossman

The results were consistently the same: only 15 to 20 percent of the American riflemen in combat during World War II would fire at the enemy.

That missing factor is the simple and demonstrable fact that there is within most men an intense resistance to killing their fellow man. A resistance so strong that, in many circumstances, soldiers on the battlefield will die before they can overcome it.

Submission is a surprisingly common response, usually taking the form of fawning and exposing some vulnerable portion of the anatomy to the victor, in the instinctive knowledge that the opponent will not kill or further harm one of its own kind once it has surrendered.

When a man is frightened, he literally stops thinking with his forebrain (that is, with the mind of a human being) and begins to think with the midbrain (that is, with the portion of his brain that is essentially indistinguishable from that of an animal), and in the mind of an animal it is the one who makes the loudest noise or puffs himself up the largest who will win. Posturing can be seen in the plumed helmets of the ancient Greeks and Romans, which allowed the bearer to appear taller and therefore fiercer to his foe, while the brilliantly shined armor made him seem broader and brighter. Such plumage saw its height in modern history during the Napoleonic era, when soldiers wore bright uniforms and high, uncomfortable shako hats, which served no purpose other than to make the wearer look and feel like a taller, more dangerous creature.

Gunpowder’s superior noise, its superior posturing ability, made it ascendant on the battlefield. The longbow would still have been used in the Napoleonic Wars if the raw mathematics of killing effectiveness was all that mattered, since both the longbow’s firing rate and its accuracy were much greater than that of a smoothbore musket. But a frightened man, thinking with his midbrain and going “ploink, ploink, ploink” with a bow, doesn’t stand a chance against an equally frightened man going “BANG! BANG!” with a musket.

Griffith estimates that the average musket fire from a Napoleonic or Civil War regiment (usually numbering between two hundred and one thousand men) firing at an exposed enemy regiment at an average range of thirty yards, would usually result in hitting only one or two men per minute! Such firefights “dragged on until exhaustion set in or nightfall put an end to hostilities. Casualties mounted because the contest went on so long, not because the fire was particularly deadly.”

(Cannon fire, like machine-gun fire in World War II, is an entirely different matter, sometimes accounting for more than 50 percent of the casualties on the black-powder battlefield, and artillery fire has consistently accounted for the majority of combat casualties in the twentieth century. This is largely due to the group processes at work in a cannon, machine-gun, or other crew-served-weapons firing. This subject is addressed in detail later in this book in the section entitled “An Anatomy of Killing.”)

This trend can be seen in the roots of the Western way of war. Sam Keen notes that Professor Arthur Nock at Harvard was fond of saying that wars between the Greek city-states “were only slightly more dangerous than American football.” And Ardant du Picq points out that in all his years of conquest, Alexander the Great lost only seven hundred men to the sword. His enemy lost many, many more, but almost all of this occurred after the battle (which appears to have been an almost bloodless pushing match), when the enemy soldiers had turned their backs and begun to run. Carl von Clausewitz makes the same point when he notes that the vast majority of combat losses historically occurred in the pursuit after one side or the other had won the battle. (Why this occurs is a subject that will be looked at in detail in the section “Killing and Physical Distance.”)

Marshall makes it clear that in most cases the firers were aware of the large body of nonfirers around them. The inaction of these passive individuals did not seem to have a

demoralizing effect on actual firers. To the contrary, the presence of the nonfirers seemed to enable the firers to keep going.[3]

A 1986 study by the British Defense Operational Analysis Establishment’s field studies division used historical studies of more than one hundred nineteenth-and twentieth-century battles and test trials using pulsed laser weapons to determine the killing effectiveness of these historical units. The analysis was designed (among other things) to determine if Marshall’s nonfirer figures were correct in other, earlier wars. A comparison of historical combat performances with the performance of their test subjects (who were not killing with their

weapons and were not in any physical danger from the “enemy”) determined that the killing potential in these circumstances was much greater than the actual historical casualty rates. The researchers’ conclusions openly supported Marshall’s findings, pointing to “unwillingness to take part [in combat] as the main factor” that kept the actual historical killing rates significantly below the laser trial levels.

The concept of drill had its roots in the harsh lessons of military success on battlefields dating back to the Greek phalanx. Such drill was perfected by the Romans. Then, as firing drill, it was turned into a science by Frederick the Great and then mass-produced by Napoleon.

An average engagement would take place at thirty yards. But instead of mowing down hundreds of enemy soldiers in the first minute, regiments killed only one or two men per minute. And instead of the enemy formations disintegrating in a hail of lead, they stood and exchanged fire for hours on end.

The process of some men electing to load and provide support for those who are willing to shoot at the enemy appears to have been the norm rather than the exception. Those who did fire, and were the beneficiary of all of this support, can be seen in countless reports collected by Griffith, in which individual Civil War soldiers fired one hundred, two hundred, or even an incredible four hundred rounds of ammunition in battle. This in a period when the standard

issue of ammunition was only forty rounds, with a weapon that became so fouled as to be useless without cleaning after firing about forty shots. The extra ammunition and muskets must have been supplied and loaded by the firers’ less aggressive comrades.

Yet despite the obvious options of firing over the enemy’s head (posturing), or simply dropping out of the advance (a type of flight), and the widely accepted option of loading and supporting those who were willing to fire (a limited kind of fighting), evidence exists that during black-powder battles thousands of soldiers elected to passively submit to both the enemy and their leaders through fake or mock firing. The best indicator of this tendency

toward mock firing can be found in the salvage of multiply-loaded weapons after Civil War battles.

Now we can begin to fully understand the reasons underlying Paddy Griffith’s discovery of an average regimental hit rate of one or two men per minute in firefights of the black-powder era. And we see that these figures strongly support Marshall’s findings. With the rifled muskets of that era, the potential hit rate was at least as high as that achieved by the Prussians with smoothbore muskets when they got 60 percent hits at seventy-five yards. But the reality was a minute fraction of this.

It took over eight hours, not eight minutes, to inflict those horrendous casualties on Grant’s soldiers. And as in most wars from the time of Napoleon on down to today, it was not the infantry but the artillery that inflicted most of these casualties. Only when artillery (with its close supervision and mutual surveillance processes among the crew) is brought into play can any significant change in this killing rate be observed. (The greater distance that artillery usually is from its targets, as we will see, also increases its effectiveness.)

That the average man will not kill even at the risk of all he holds dear has been largely ignored by those who attempt to understand the psychological and sociological pressures of the battlefield. Looking another human being in the eye, making an independent decision to kill him, and watching as he dies due to your action combine to form one of the most basic, important, primal, and potentially traumatic occurrences of war.

The point is that (1) an intense, traumatic, guilt-laden situation will inevitably result in a web of forgetfulness, deception, and lies; (2) such situations that continue for thousands of years become institutions based on a tangled web of individual and cultural forgetfulness, deception, and lies tightly woven over thousands of years; and (3) for the most part there have been two such institutions

about which the male ego has always justified selective memory, self-deception, and lying. These two institutions are sex and combat…love and war.

Even after Marshall’s World War II revelations, the subject of nonfirers is an uncomfortable one for today’s military. Writing in Army magazine—the U.S. Army’s foremost military journal—Colonel Mater states that his experiences as an infantry company commander in World War II strongly supported Marshall’s findings and noted several World War I anecdotes that suggest that the problem of nonfirers was just as serious in that war.

According to studies by Marshall, these changes resulted in a firing rate of 55 percent

in Korea and, according to a study by R. W. Glenn, a 90 to 95 percent firing rate was attained in Vietnam. Some modern soldiers use the disparity between the firing rates of World War II and Vietnam to claim that Marshall had to be wrong, for the average military leader has great difficulty in believing that any significant body of his soldiers will not do their job in combat. But these doubters don’t give sufficient credit to the revolutionary corrective measures and training methods introduced since World War II.

As a psychologist I believe that Marin is quite correct when he observes, “Nowhere in the [psychiatric and psychological] literature is one allowed to glimpse what is actually occurring: the real horror of the war and its effect on those who fought it.”

One of Freud’s most valuable insights involves the existence of the life instinct (Eros) and the death instinct (Thanatos). Freud believed that within each individual there is a constant struggle between the superego (the conscience) and the id (that dark, lurking mass of destructive and animal urges residing within each of us), and that the struggle is mediated by the ego (the self). One wit once referred to this situation as “a struggle in a locked, dark basement; between a homicidal sex-crazed monkey and a puritanical old maid; being mediated by a timid accountant.”

There can be no doubt that this resistance to killing one’s fellow man is there and that it exists as a result of a powerful combination of instinctive, rational, environmental, hereditary, cultural, and social factors. It is there, it is strong, and it gives us cause to believe that there just may be hope for mankind after all.

Swank and Marchand’s much-cited World War II study determined that after sixty days of continuous combat, 98 percent of all surviving soldiers will have become psychiatric casualties of one kind or another. Swank and Marchand also found a common trait among the 2 percent who are able to endure sustained combat: a predisposition toward “aggressive psychopathic personalities.”

The British in World War I believed that their soldiers were good for several hundred days before inevitably becoming a psychiatric casualty. But this was made possible only by the British policy of rotating men out of combat for four days of rest after approximately twelve days of combat, as opposed to America’s World War II policy of leaving soldiers in combat for up to eighty days at a stretch.

Some psychiatric casualties have always been associated with war, but it was only in the twentieth century that our physical and logistical capability to sustain combat outstripped our psychological capacity to endure it.

Until the post-Vietnam era, when hundreds of thousands of PTSD cases appeared, this was the only treatment believed necessary to permit the soldier to return to a normal life. But the problem is that the military does not want to simply return the psychiatric casualty to normal life, it wants to return him to combat! And he is understandably reluctant to go.

But if soldiers begin to realize that insane soldiers are being evacuated, the number of psychiatric casualties will increase dramatically. An obvious solution to this problem is to rotate troops out of battle for periodic rest and recuperation—this is standard policy in most Western armies—but this is not always possible in combat. Proximity—or forward treatment—and expectancy are the principles developed to overcome the paradox of evacuation syndrome. These concepts, which have proved themselves

quite effective since World War I, involve (1) treatment of the psychiatric casualty as far forward on the battlefield as possible—that is, in the closest possible proximity to the battlefield, often still inside enemy artillery range—and (2) constant communication to the casualty by leadership and medical personnel of their expectancy that he will be rejoining his comrades in the front line as soon as possible. These two factors permit the psychiatric casualty to get treatment and much-needed rest, while not giving a message to still-healthy comrades that psychiatric casualty is a ticket off the battlefield.

Let us use a model as a framework for the understanding and study of psychiatric casualty causation, a metaphorical model representing and integrating the factors of fear, exhaustion, guilt and horror, hate, fortitude, and killing.

Both the Berkun and Shalit studies indicate that fear of death and injury is not the primary cause of psychiatric casualties on the battlefield. Indeed, Shalit found that even in the face of a society and culture that tell the soldiers that selfish fear of death and injury should be their primary concern, it is instead the fear of not being able to meet the terrible obligations of combat that weighs most heavily on the minds of combat soldiers.

Fear, combined with exhaustion, hate, horror, and the irreconcilable task of balancing these with the need to kill, eventually drives the soldier so deeply into a mire of guilt and horror that he tips over the brink into that region that we call insanity. Indeed, fear may be one of the least important of these factors.

Nonkillers are frequently exposed to the same brutal conditions as killers, conditions that cause fear, but they do not become psychiatric casualties. In most circumstances in which nonkillers are faced with the threat of death and injury in war, the instances of psychiatric casualties are notably absent.

The carnage and destruction, and the fear of death and injury, caused by the months of continuous blitz in England during World War II were as bad as anything faced by any frontline soldier. Relatives and friends were mutilated and killed, but in a strange sort of way, that

was not the worst of it. These civilians suffered one indignity that most soldiers need never face. In 1942, Lord Cherwell wrote: “Investigation seems to show that having one’s house demolished is most damaging to morale. People seem to mind it more than having their friends or even relatives killed.”

And yet, incredibly, the incidence of psychiatric casualties among these individuals was very similar to that of peacetime. There were no incidents of mass psychiatric casualties. The Rand Corporation study of the psychological impact of air raids, published in 1949, found

that there was only a very slight increase in the “more or less long-term” psychological disorders as compared with peacetime rates. And those that did appear seemed to “occur primarily among already predisposed persons.” Indeed, bombing seemed to have served primarily to harden the hearts and empower the killing ability of the nations that endured it.

The prisoners were unarmed, impotent, and strangely at peace with their lot in life. They had no personal capacity or responsibility to kill, and they had no reason to believe that the incoming artillery or bombs were a personal matter. The guards, on the other hand, took the matter as a personal affront. They still had a capacity and a responsibility to fight, and they were faced with the irrefutable evidence that someone was intent on killing them and that they had a responsibility to do likewise. The psychiatric casualties among the guards—as among most other soldiers in the same circumstances—represented an accepted method of escape from the unbearable responsibility inherent in their roles as soldiers.

But in the twentieth and twenty-first centuries, psychiatric casualties during naval warfare have been nearly nonexistent. The great military physician Lord Moran noted the remarkable absence of psychological illness among the men he ministered to aboard ships in World War II. Discussing his experience in two ships, he said, “One was sunk after surviving more than two hundred raids and the whole of the first Libyan campaign. The other was in four major actions in addition to many raids at sea and in harbour, and twice sustained actual damage.” Yet the incidence of psychiatric casualties was almost nonexistent. “There were more than five hundred men in the two ships and of these only two came to me about their nerves.”

Instead of killing people up close and personal, modern navies kill ships and airplanes. Of course there are people in these ships and airplanes, but psychological and mechanical distances protect the modern sailor. World War I and World War II ships often fired their weapons at enemy ships that could not be seen with the naked eye, and the aircraft they fired at were seldom more than specks in the sky. Intellectually these naval warriors understood that they were killing humans just like themselves and that someone wanted to kill them, but emotionally they could deny it. A similar phenomenon has occurred in aerial combat. As previously noted, World War I and World War II pilots, in relatively slow-moving aircraft, could see enemy pilots, and thus large numbers of them failed to fight aggressively. Modern pilots, fighting an enemy seen only on a radar screen, have no such problems.

And even though line-officer casualties in most wars are consistently much higher than those of their men (in World War I, 27 percent of the British officers who served on the western front were killed, compared with 12 percent of the men), their psychiatric casualties are usually significantly lower (in World War I, the probability of a British officer becoming a psychiatric casualty was half that of the men).

It would appear that, at least in the realm of psychiatric casualty causation, fear does not reign supreme on the battlefield. The effect of fear should never be underestimated, but it is clearly not the only, or even the major, factor responsible for psychiatric casualties on the battlefield.

The magnitude of the exhaustion and the horror suffered by combat veterans and victims of strategic bombing is generally comparable. The stress factors that soldiers experienced and bombing victims did not were the two-edged responsibility of (1) being expected to kill (the irreconcilable balancing of to kill and not to kill) and (2) the stress of looking their potential killers in the face (the Wind of Hate).

“At least,” they say, “you had the guts to try.” And the graduates