I mentioned to Robin Hanson that his near/far dichotomy is reflected in Randall Collins’ “Violence: A Microsociological Theory”, which he had recently discussed. Here are relevant bits from chapter 2.

Soldiers who have been in combat and had direct contact with the enemy tend to depict him as courageous; it is enemies on more distant combat zones who are not respected; and soldiers in rear areas, and even more so civilians at home, who express a low regard for the enemy (Stouffer et al. 1949: 158-65). [p. 41]

We would expect that higher-ranking officers would be least likely to perceive the problem [of low rates of competent firing]. In all organizations, those at the highest ranks are least likely to have accurrate information on what is going on at the lowest level of practical action; in addition, the higher the rank, the more the person identifies with the formal frontstage ideals of the organization and is likely to talk in official rhetoric. Frontline combat soldiers of the lower ranks would have a different viewpoint. Yet another bias comes from the contrast between detailed observation of what is happening in each micro-situation, and summary accounts of an ideal-typical version of performance; the latter would tend to be more idealized toward a favorable image, and we would expect that this bias would grow with as the actual memories of combat experience become more distant. [pgs 47-8].

A higher firing ratio (and more generally across history, a higher frequency of using one’s weapons) is the result of a number of conditions. (Many of these are summarized in Grossman 1995).

  1. Group-operated weapons: teams of soldiers manning machine-guns, bazookas or rocket launchers, mortars, and other weapons where part of the team feeds ammunition or provides ancillary services.
  2. Greater distance from the enemy: artillery has a high firing rate; snipers firing at long distance with telescopes have a high rate; infantry firing in close combat has a low rate; and use of weapons in hand-to-hand combat is very low. [pgs 55-6].

A few individuals have been highly accurate combat shooters. Chief among these are snipers, but these have been a very small proportion of troops, and individuals firing with great accuracy are not how most casualties in battle are inflicted. In the modern era, casualties were caused primarily by artillery fired at long distance. In the musket era of parade-ground formations, cannon operating closer to the battle line generally accounted for more than 50 percent of the casualties; the most successful generals, including Gustavus Adolphus in the seventeenth century and Napoleon at the turn of the nineteenth, emphasized small mobile field-pieces spread throughout combat units that could be fired at close range, especially with grapeshot that gave a machine-gun-like effect (Grossman 1995: 11; 154). In World War I, artillery caused almost 60 percent of British casualties, bullets about 40 percent; in World War II, artillery and aerial bombs caused about 75 percent, bullets less than 10 percent. In Korea, shells and mortars caused about 60 percent of American casualties, while small arms caused 3 percent of deaths and 27 percent of wounds (Holmes 1985: 210).  [...] How are we to interpret this pattern? As Grossman documents, artillery achieves a higher level of firing than riflemen and frontline troops generally; the sheer distance from the enemy, and especially being shielded from personally seeing the men one is trying to kill increases the level of performance. Additionally, artillery are group-operated weapons; local small-group solidarity and emotional entrainment keeps men on the job [...] The tension/fear of combat is almost completely debilitating at close range, and in the individual use of small arms; at longer range, tension/fear is overcome, but something like it remains in a degree. [pgs 58-59]

Evidence for this argument [about the pleasure of battle] needs to distinguish the kinds of situations in which men (and in some instances, women) express feelings of joy about combat. One type is pre-battle elation. Bourke (1999: 274) quotes a World War I British chaplain who describes his own, and his troops’ “strange and fearsome delight at being at last up ‘really’ up against it”. This is a case of feelings prior to these men’s first battle, still in the phase of rhetoric. Ulysses S. Grant (1885/1990: 178) similarly describes troops in November 1861 in his first Civil War command, who were so eager for battle that he felt he could not maintain discipline if he did not find an engagement for them to fight.

Related to this type is the bloodthirsty rhetoric expressed at a distance from the front. Given the size of the logistics and support train for twentieth- and twenty-first century armies, a considerable proportion of troops are in no position actually to fire at the enemy, and in relatively little danger of receiving fire; but they often carry weapons and have been trained to use them, and so they have some plausible self-description as fighting soldiers19. Soldiers in rear areas express more hatred of the enemy, and more ferocious attitudes toward them, than frontline troops (Stoufffer et al. 1949: 158-65). Whereas combat soldiers are more likely to treat prisoners well–once the danger point has been passed when they are actually being captured, often sharing food and water with them–rear area troops tend to treat prisoners more callously, or even brutally (Holmes 1985: 368-78, 382). Continuing the progression, civilians at home are more likely to express violent rhetorical hatred against the enemy and blood-thirsty joy in killing them (Bourke 1999: 144-53). Given the relatively high proportion of women on the civilian homefront, there is reason to doubt that gender per se, rather than situational differences, accounts for differences in ferociousness.

The farther from the front, the more rhetorical ferociousness is expressed, and rhetorical enthusiasm for the whole fighting enterprise. This fits the general pattern of all fights: surrounded by bluster and gesture up until the actual fight situation, when the emotion shifts drastically and tension/fear takes over (documented in Holmes 1985: 75-8, citing many observers). The proportion of empty rhetoric expands with each step toward the rear; war is successively more idealized, the enemy successively more dehumanized, attitudes toward killing successively more callous, and the whole affair more like the cheering of sports fans. [pgs 66-7]

[T]he circumstances that cause the most fear not necessarily those that are objectively the most dangerous. Artillery shells and mortars, as we have seen, cause by far the most casualties–and the soldiers themselves generally know that (Holmes 1985: 209-10)–but the greatest difficulty in combat performance is in confronting small-arms fire at the forward edge of the combat zone. Some surveys show relatively high fear of being killed by bayonet and knife, events rare to the point of fantasy but indicating the quality of soldiers’ imagery about what they feel is in front of them. Nor do persons in highly dangerous situations all show the signs of incapacitating fear that affects frontline troops (Grossman 1995: 55-64): navy personnel are subject to the same dangers as army soldiers of being blown apart by enemy shells–the largest source of ground combat casualties–in addition to prospects of drowning, but data on long-term breakdown from combat stress–which is one measure of combat fear–shows much lower rates of breakdown for sailors in combat zones. Similarly for civilians under bombardment, including long-term blitzes such as the German attack on England, or the Allied bombings of German cities; casualties included being burned alive or sustaining extreme bodily mutilations resulting from burns. Nevertheless, civilian psychiatric casualties were low in these areas compared to army troops.

Several refined comparisons are revealing of the precise source of tension/fear. Prisoners of war under fire or aerial bombardment had no increase in psychiatric casualties; whereas their guards apparently underwent increased tension, since their psychiatric rates went up (Grossman 1995: 57-58; Gabriel 1986, 1997). That is to say, the guards were still in a combat mode, perhaps because they were confronted with enemies under their own eyes, at the same time they were struggling to maintain control over them; whereas the stance of the POWs was merely to endure. Grossman (1995: 6-61) also points out that reconnaissance patrols behind enemy lines, although extremely dangerous, do not lead to psychiatric casualties. The reason, Grossman argues, is that such patrols are attempting to gather information by stealth and, above all, to avoid attacking the enemy. And combat officers, although in most wars they have had physical casualty rates considerably higher than their men, have had a much lower rate of psychiatric casualties (Grossman 1995: 64). Here we see that the source of strain is neither fear of death and injury, nor aversion to killing in principle, since officers are in charge of directing their men toward killing, and indeed pressing them to overcome their fear and incompetence. What is different, and what seems to buffer them from tension/fear, is that they personally do not have to do the killing. The same goes for non-firers, who often perform other useful tasks on the battlefield, such as helping load ammunition into the firers’ guns (Grossman 1995: 15). This shows that they are often willing to expose themselves to as much danger as the firers. It isn’t even that they are opposed to killing; they just can’t bring themselves to do it themselves.

Medics in ground combat are subject to the same sorts of dangers as infantry troops; yet their rate of combat fatigue is much lower (Grossman 1995: 62-64, 455). [pgs 74-6]

There’s lots more I’d like to post elaborating on Collins’ explanation, but what I’ve put up is plenty. I actually received Collins book in the mail two days ago, James Scott’s “Seeing Like a State” today, and picked up The 10,000 Year Explosion from the library yesterday, which I’ll be focusing on in the near future. I actually had a post about the Quark and the Jaguar in my head since the weekend, but it will probably be a while until I get to that.