The World Until Yesterday

positions heavily defended by machine guns; about 100,000 Japanese killed on or after August 6, 1945, by the American atomic bomb dropped on Hiroshima (Plate 37); and over 50,000,000 total deaths as a result of World War II. By these standards, the Dani fighting that I just summarized was a tiny war, if it is considered a war at all.

Yes, as measured by the absolute number of people killed, the Dani War was indeed tiny. But the nations involved in World War II were far more populous, and offered far more potential victims, than did the two alliances involved in the Wilihiman-Widaia war. Those alliances numbered perhaps 8,000 people in all, while the major participants in World War II had populations ranging from tens of millions to nearly a billion. The relative death toll of the Dani War—the number of Dani killed as a proportion of the total population involved—rivaled or eclipsed the casualty rates suffered by the U.S., European countries, Japan, or China in the world wars. For example, the 11 deaths suffered by the two Dani alliances on the Gutelu southern front alone, in the six months between April and September 1961, represented about 0.14% of the alliances’ population. That’s higher than the percentage death toll (0.10%) from the bloodiest battle on the Pacific front during World War II: the three-month struggle for Okinawa, employing bombers and kamikaze planes and artillery and flame-throwers, in which about 264,000 people (23,000 American soldiers, 91,000 Japanese soldiers, and 150,000 Okinawan civilians) died, out of a total American/Japanese/Okinawan population then of around 250,000,000. The 125 men, women, and children killed within an hour in the Dani massacre of June 4, 1966, represented about 5% of the targeted population (about 2,500), the southern confederations of the Gutelu Alliance. To match that percentage, the Hiroshima atomic bomb would have had to kill 4,000,000 rather than 100,000 Japanese, and the World Trade Center attack would have had to kill 15,000,000 rather than 2,996 Americans.

By world standards, the Dani War was tiny only because the Dani population at risk of being killed was tiny. By the standards of the local population involved, the Dani War was huge. In the next chapter we shall see that that conclusion also applies to traditional warfare in general.

CHAPTER 4 A Longer Chapter, About Many Wars

Definitions of war Sources of information Forms of traditional warfare Mortality rates Similarities and differences Ending warfare Effects of European contact Warlike animals, peaceful peoples Motives for traditional war Ultimate reasons Whom do people fight? Forgetting Pearl Harbor

Definitions of war

Traditional warfare, as illustrated by the Dani War described in the previous chapter, has been widespread but not universal among small-scale societies. It raises many questions that have been hotly debated. For example, how should war be defined, and do so-called tribal wars really constitute wars at all? How do the death tolls from warfare in small-scale societies compare to death tolls from state warfare? Does warfare increase or decrease when small-scale societies become contacted and influenced by Europeans and other more centralized societies? If fighting between groups of chimpanzees, lions, wolves, and other social animals furnishes precedents for human warfare, does that suggest a genetic basis of warfare? Among human societies, are there some especially peaceful ones? If so, why? And: what are the motives and causes of traditional warfare?

Let’s begin with the question of how to define warfare. Human violence assumes many forms, only some of which are normally taken to constitute war. Anyone will agree that a battle between large armies of trained professional soldiers in the service of rival state governments that have issued formal declarations of war does constitute war. Most of us would also agree that there are forms of human violence that don’t constitute war, such as individual homicides (the killing of one individual by another individual belonging to the same political unit), or family feuds within the same political unit (such as the feud between the Hatfield and McCoy families of the eastern United States beginning around 1880). Borderline cases include recurrent violence between rival groups within the same political unit, such as fighting between urban gangs (commonly referred to as “gang warfare”), between drug cartels, or between political factions whose fighting has not yet reached the stage of declared civil war (such as the fighting between armed militias of fascists and socialists in Italy and Germany leading up to Mussolini’s and Hitler’s assumptions of power). Where should we draw the line?

The answer to that question may depend on the purpose of one’s study. To future soldiers in training at a state-sponsored military college, it may be appropriate to exclude from a definition of warfare Chapter 3’s stories of violence between rival Dani alliances. However, for our purposes in this book, which is concerned with the whole spectrum of related phenomena observed from the smallest human bands of 20 people to the largest states of over a billion people, we must define warfare in a way that doesn’t define traditional warfare between small bands out of existence. As Steven LeBlanc has argued, “Definitions of war must not be dependent on group size or methods of fighting if they are to be useful in studying past warfare…. Many scholars define warfare in such a way that it refers to something that only complex societies employing metal tools can have [i.e., pitched battles and professional soldiers]. Anything else—say, a raid or two now and then—is not ‘real’ warfare, they believe, but is something more akin to game playing and not a subject of much concern. Such an approach or attitude, however, confuses the methods of war with the results of war…. Does conflict between independent political units lead to significant deaths and loss of territory, while resulting in some territory being rendered useless because it’s too dangerous to live in? Are people spending a great deal of time and energy defending themselves?…If fighting results in significant impacts on people, it is war regardless of how the fighting is conducted.” From that perspective, war should be defined sufficiently broadly so as to include the Dani fighting described in Chapter 3.

Consider one fairly typical definition of war, that from the Encyclopaedia Britannica’s 15th edition: “A state of usually open and declared armed hostile conflict between political units, such as states or nations or between rival political factions of the same state or nation. War is characterized by intentional violence on the part of large bodies of individuals who are expressly organized and trained to participate in such violence…. War is generally understood to embrace only armed conflicts on a fairly large scale, usually excluding conflicts in which fewer than 50,000 combatants are involved.” Like many other apparently common-sense definitions of war, this one is much too restrictive for our purposes, because it requires “large bodies of individuals who are expressly organized and trained,” and it thereby refuses to admit the possibility of war in small band societies. Its arbitrary requirement of at least 50,000 combatants is more than six times the entire population (men warriors, women, and children) involved in Chapter 3’s Dani War, and far larger than most of the small-scale societies discussed in this book.

Hence scholars studying small-scale societies have come up with various alternative broader definitions of war, similar to each other and usually requiring three elements. One element is violence carried out by groups of any size, but not by single individuals. (A killing carried out by one individual is considered a murder, not an act of war.) Another element is that the violence is between groups belonging to two different political units, not belonging to the same political unit. The remaining element is that the violence must be sanctioned by the whole political unit, even if only some members of the unit carry out the violence. Thus, the killings between the Hatfield and McCoy families didn’t constitute war, because both families belonged to the same political unit (the U.S.), and the U.S. as a whole did not approve of that family feud. These elements may be combined into a short definition of war that I shall use in this book, and that is similar to definitions formulated by other scholars of small-scale as well as state societies: “War is recurrent violence between groups belonging to rival political units, and sanctioned by the units.”