War Before Civilization

by Lawrence H Keeley

  21. Vayda 1976: 23; Heider 1970: 78, 119; HNAI vol. 8, 1978: 506, 510, 513, 674, 687; HNAI vol. 6, 1981: 286–87, 494; Slobodin 1960: 83; Chagnon 1968: 141; Herdt 1987: 54–55.

  22. Heider 1970: 105; Herdt 1987: 54; HNAI vol. 6, 1981: 287; Cannon 1992: 509–10.

  23. (Middle Missouri) Zimmerman 1980; Willey 1990; Bamforth 1994; (Southwest) Haas 1990: 187, and personal communication.

  24. Milner et al. 1991: 595. Milner and his colleagues also provide a long list of references to evidence of warfare deaths among prehistoric Native Americans in the eastern prairies and woodlands. A popular account of the Crow Creek site is given in Zimmerman and Whitten (1980); a detailed analysis of the bones appears in Willey (1990).

  25. Wahl and Konig 1987; Courtin 1984: 448.

  Chapter 5

  1. Regarding the identity of the Skraelings, archaeology indicates that they were Indians ancestral to the historic Beothuk, since the Dorset Eskimo had disappeared from Newfoundland and adjacent parts of Labrador several centuries before the Norse appeared (HNAI vol. 15, 1978: 69; Fitzhugh 1985: 25–29). Given that their technology and diet were the same as the historic Beothuk, the population density (no more than one person per 25 square miles) and settlement pattern (small-band encampments scattered along the coast) of the proto-Beothuk Skraelings were probably comparable. That being the case, there would have been no more than 100 potential Skraeling warriors within a 200-mile radius of the Viking settlements. Since the Viking colony consisted of 250 men and women (Morison 1971: 54), the Vikings could never have been outnumbered or attacked by a “multitude” of Skraelings as the sagas claim.

  2. Morison 1971: 32–62; see also Roesdahl 1991: 274–75; Fitzhugh 1985: 28.

  3. Parker 1988: 120.

  4. For examples, see Allred et al. 1960; Utley and Washburn 1977; Utley 1984; HNAI vol. 4, 1988: 128–63; HNAI vol. 10, 1983: 496.

  5. For example, Utley 1984; HNAI vol. IS, 1978: 625.

  6. The Narragansett fort successfully stormed by Massachusetts militia in 1675 was incomplete.

  7. One would think that the repeated thrashing of better-disciplined European and fanatically disciplined Asian armies by American “rabble” (and the nasty treatment dealt the French by highly “irregular” Mexican Juaristas) would eventually disabuse these Colonel Blimps of their condescension. But it shows no signs of abating; see, for example, the works of Max Hastings and Dan van der Vat.

  8. Morris 1965; Edgerton 1988 (this work, written by an anthropologist, is especially recommended). The Zulu polity was a true state, and Zulu regiments were disciplined units that fought in massed formations and thus were easier for a civilized army to defeat than hit-and-run skimishers like the Apaches.

  9. Porch 1986: 120–23, 140, 165–72, 227–30.

  10. Edgerton 1988: 119–213.

  11. Bodley 1990:46.

  12. Utley and Washburn 1977: 53, 134, 210; HNAI vol. 4, 1988: 130, 159, 162, 164, 170–72; HNAI vol. 15, 1978: 99–100; Utley 1984: 95–96; Porch 1986: 209–10; Parker 1988: 119–20, 207 n. 49; Eid 1985.

  13. Eid 1985: 139; HNAI vol. 15, 1978: 99–100. Malone (1991) also makes this point in greater detail, marred only by his unsupported belief that the precontact warfare of the New England tribes did not inflict many casualties.

  14. Malone 1991: 6. Apologies to Patrick Malone for appropriating the title of his fine book The Skulking Way of War to head this chapter.

  15. Utley 1984: 95–96.

  16. This is abstracted from accounts in Utley 1984; Utley and Washburn 1977; and HNAI vol. 4, 1988: 168, 174–76.

  17. Dudley 1975: 90–91, 98, 157–70; Connolly 1989: 165–67; Dobson 1989: 205–12.

  18. McGovern 1985: 311–14; HNAI vol. 5, 1984: 551–55; Fitzhugh 1985: 27–31. Some academics dismiss these Inuit and Norse accounts because they contain certain fantastic or stylized elements. This dismissal is analogous to doubting that Magellan’s expedition sailed around the world, simply because Pigafetta’s account of the voyage mentions fabulous animals and minor miracles. In any case, scholarly scepticism has not discouraged the Greenlanders (a term now used only for those of Inuit descent) from mischievously celebrating this ancestral genocide of Scandinavians in every conceivable form of art.

  19. Mercer 1980: 157–58, 184–93; Crosby 1986: 79–89.

  20. For example, HSAI vol. 3, 1948: 218, 467, 505, 509, 563, 618, 729.

  21. (Pawnee) Weber 1992: 171; (Seminoles and Red Cloud) Utley and Washburn 1977: 128, 131–35, 211–15.

  22. Crosby 1972, 1986.

  23. Aztec casualties from Thomas (1993: 528–29) do not include the 170,000 estimated to have died from starvation and disease during the siege of Tenochtitlàn. Disease mortality is calculated from various estimates of the decline in central Mexico’s population cited by Thomas (1993: 609–13); Crosby (1972: 53); and Fagan (1984: 286).

  24. Crosby 1986: 133–47 and 1972: 35.

  25. Williams 1970: 198–99.

  26. Malone 1991.

  27. Schiefenhövel 1993: 327; P. Weissner, personal communication.

  28. One popular folly in the antebellum American South was that the United States could “annex” Mexico proper, specifically in order to expand the slave economy. Similar southern ambitions were entertained regarding Cuba and Central America. Any such southern-supported “filibuster” would undoubtedly have been a bloody failure, as was the only actual attempt in Nicaragua from 1855 to 1857 (McPherson 1988: 47–116).

  29. Updating Laqueur’s (1984: 441–42) list, my count is as follows. Guerrilla wins: Mexico (1911–1919), Ireland (1919–1922), Arabia (Ibn Saud), Nicaragua (Sandino 1927–1933), Yugoslavia (World War II), Albania (World War II), Palestine (Israelis 1944–1948), Indochina (1945–1954), Indonesia, Algeria, Cyprus (independence), Cuba (Castro), Vietnam, Laos, Cambodia, Angola, Mozambique, Guinea-Bissau, Rhodesia (Zimbabwe), Aden-Yemen, Namibia, Afghanistan, Nicaragua (Sandinista).

  Guerrilla victories correlated with conventional military victories: Arabia (World War I), China (1927–1945), USSR (World War II), Greece (World War II), France (World War II), Italy (World War II).

  Guerrilla losses: Boers (1899–1902), Philippines (1899–1902), Soviet Union (Basmatchi 1919–1930), Morocco (1921–1927), Brazil (Prestes), Palestine (Arab 1936–1939), Poland (1944), Iraqi Kurds (1945–1975), Philippines (Huk 1946–1956), Greece (communists 1947–1949), Malaya (communist), Kenya (Mau-Mau), Venezuela (1962—1965), Peru (1962–1965), Oman (1962–1976), Guatemala (1964–1967), Bolivia (Guevara).

  Ties: German East Africa (1914–1918, surrendered only at general armistice), southern Sudan (1955–1972), Syria (1925–1936), and Yemen Civil War (1962–1970) (in the three latter cases, the guerrillas were granted considerable concessions, as were the defeated Boer commandos, the Filipino Huks, and the Kenyan Mau-Mau).

  Although a number of guerrilla wars are currently unresolved, by my count, guerrilla wins outnumber losses by almost two to one. Also note that many of these “losses” amd “ties” have engendered several new and continuing guerrilla wars, such as in the Philippines (Huk) and in the Sudan.

  It is difficult to classify urban terrorists as guerrillas, since their actions never intend or accomplish the slightest harm to the military strength or basic economy of their enemies, they never clear or control any territory, and they have an unbroken losing record for at least the last two centuries. Since they rely on the mass media for their theatrical effects, they seem better classified as dramatic performers (Laqueur 1984: xi).

  It is fair to note that some military analysts find little to appreciate in guerrilla techniques (for example, Laqueur 1984 and Van Creveld 1989). They note the dependence of modern guerrillas on external logistic support and suitable geography. But they also denigrate guerrilla victories by attributing them primarily to the political or “moral” failings of their conventional opponents rather than the military effectiveness of the guerrillas. Thus they argue that the Dutch in Indonesia, the British and the Portuguese in countless wars of decol
onization, the French in Indochina and Algeria, and the Americans in Vietnam lost because of a lack of will at home, a liberal squeamishness about human rights and the brutalities required to win, or an inability to commit their full military resources to the struggle. These arguments are merely special pleadings. In all wars, the defeated side loses its will to continue, either because it has suffered intolerable human and economic losses or because it judges continuing warfare would be more costly than making a disadvantageous peace. Guerrilla wars have been won and lost for the same reasons as conventional wars. The totalitarian, unsqueamish Soviet Union was neither hindered by a free press nor domestic political dissent in its war in Afghanistan; yet it was defeated. The division of military resources also affected the victors in some notable conventional wars: the United States in World War II and Korea (the Pacific/Asia versus Europe/NATO), Israel in 1967 and 1973 (Syria versus Egypt), and Britain in the Falklands War (Northern Ireland/NATO versus Argentina).

  30. Weigley (1991) details the futility of 200-year quest by European armies for “decisive” battles.

  31. For example, McNeill 1982; Parker 1988.

  32. One defense analyst (Friedman 1991: 251) attributes all of Iraq’s military deficiencies to general social and economic factors and concludes: “It is difficult to escape the conclusion that military superiority is a function of social organization.”

  Chapter 6

  1. HNAI vol. 5, 1984: 177, 218, 333, 477; HNAI vol. 6, 1981: 408, 455; HNAI vol. 7, 1990: 215, 336, 465, 495; HNAI vol. 8, 1978: 199, 380, 393, 488, 547; HNAI vol. 10, 1983: 329; HNAI vol. 15, 1978: 262, 278, 316, 386, 676; HSAI vol. 1, 1946: 195, 314, 391, 498; HSAI vol. 3, 1948: 88, 112–26, 188, 278, 291, 528, 647, 701, 756; Matthews 1877: 61; Slobodin 1960: 83; Stewart 1965: 381–82; Ray 1963: 143; Meggitt 1977: 89–91; Handy 1923: 133–34; Evans-Pritchard 1940: 128; Brown 1922: 85.

  2. Meggitt 1977: 102–103.

  3. Edgerton 1988: 130. Actually, other Nguni tribes and the Mtetwa (Zulu) a hundred years before the Zulu War sometimes took male prisoners who were later ransomed for cattle; the custom of killing prisoners was an innovation of King Shaka (Otterbein 1967: 352).

  4. Oliver 1974: 395–98; HNAI vol. 8, 1978: 380, 393, 547; HNAI vol. 10, 1983: 329; HNAI vol. 15, 1978: 220, 316, 386, 628, 676; Hudson 1976: 253–57; HSAI vol. 3, 1948: 119–26, 291, 339.

  5. HNAI vol. 15, 1978:316, 386.

  6. Vayda 1960: 70; Handy 1923: 138; Carneiro 1990: 199; Balee 1984: 246–47; Whitehead 1990: 155; Steward and Faron 1959: 209, 236, 244, 305, 323–27, 331, 335; HNAI vol. 8, 1978: 330, 380; Morren 1984: 175, 179, 193; HNAI vol. 15, 1978: 386, 676; HSAI vol. 3, 1948: 291, 339, 701.

  7. For example, Glasse 1968: 93; Pospisil 1958: 93.

  8. Of the 230 or so tribal groups (worldwide but the majority in the Americas and Oceania) for which I have notes on this issue, I have found only 8 that sometimes spared adult male captives for any reason: the Shawnee and Fox of the midwestern United States, the Mojo and Baure of central South America, the Macushi Carib of Guiana, the Nandi and Meru of East Africa, and the Nguni of southeastern Africa (HNAI vol. 15, 1978: 628, 642; HSAI vol. 3, 1948: 418, 852; Huntingford 1953: 77–78; Fadiman 1982: 46; Otterbein 1967: 352). In all these cases, the ethnographic accounts indicate that such acts of mercy were at least unusual, if not exceptional.

  9. Evans-Pritchard 1940: 128–29; Kelly 1985: 55–57.

  10. HNAI vol. 5, 1984: 177; HNAI vol. 6, 1981: 455; HNAI vol. 7, 1990: 215, 336, 465; HNAI vol. 8, 1978: 219, 547. HNAI vol. 15, 1978: 676; Ray 1963: 143; HSAI vol. 3, 1948: 262, 710, 786; Steward and Faron 1959: 188.

  11. For example, Hogbin 1964: 59; Meggitt 1977; Pospisil 1963: 59; Robbins 1982: 188.

  12. Porch 1986: 80.

  13. Jorgensen 1980: 514.

  14. Fadiman 1982: 46; Meggitt 1962:38; HNAI vol. 7, 1990: 465; HNAI vol. 8, 1978: 380, 547; HNAI vol. 10, 1983: 329; HNAI vol. 15, 1978: 157.

  15. HSAI vol. 4, 1948: 549; Steward and Faron 1959: 322; Rouse 1986: 188. Whitehead (1990) dismisses both the Island Carib traditions and the complementary reconstructions of linguists as being the result of early confusion between the Caribs of the mainland and the linguistically complex islanders. He also attributes this confusion to exclusive reliance on Spanish sources. Since the only sources on the early contact period islanders are Spanish, it is difficult to locate an alternative. He claims that the Island Carib “pidgin” was just a “trade language” shared with their friends the mainlanders. The idea that Island Carib men would use a mere trade lingo when speaking with their wives and among themselves seems far more improbable than the traditional explanation.

  16. For example, Early and Peters 1990: 20, 67–70; see also Kelly 1985 on the demographic effects of the capture of Dinka women by the Nuer.

  17. HNAI vol. 11, 1986: 382; HNAI vol. 10, 1983: 476; HNAI vol. 8, 1978: 199, 329, 440, 547, 694–700; HNAI vol. 6, 1981: 408, 494; HNAI vol. 5, 1984: 218, 333, 477; HSAI vol. 1, 1946: 498; HSAI vol. 3, 1948: 721; Steward and Faron 1959: 358; Slobodin 1960: 83; Oliver 1974: 398; Handy 1923: 133–34; Vayda 1960: 92; Brown 1922: 85.

  18. Zegwaard 1968: 443; Morren 1984: 176. Among an Asmat population of about 5,000 people, eighty-three such raids were recorded in just one year. If each raid killed only one person (a very conservative estimate), the annual war death rate would be 1.66 percent, or higher than any group listed in Figure 6.1.

  19. Pakenham 1991: 609–15; Edgerton 1988: 210–12.

  20. Utley and Washburn 1977: 42, 206–7, 217–18; HNAI vol. 15, 1978: 106–7; HNAI vol. 7,1990: 586, 592; HNAI vol. 8,1978:107–108,178, 195, 205, 249, 362–63. In the United States, public condemnation of such slaughters came primarily from representatives of the federal government: Indian agents and officers of the regular army. Similarly infamous genocides of native Tasmanians and South African “Bushmen” were committed by colonial militias, commandos, and allied native tribes.

  21. Some readers may be unconvinced by percentage comparisons between populations of hundreds or thousands of people and populations of millions and tens of millions—that is, they are more impressed by absolute numbers than ratios. However, consistent with such views, such skeptical readers must also disdain any calculations of death rates per patient or passenger-mile and therefore always chose to undergo critical surgery at small, rural, Third World clinics and fly on small airlines. At such medical facilities and on such airlines, the total number of passenger or patient deaths are always far fewer than those occurring on major airlines or at large university and urban hospitals. These innumerate readers should also prefer residence on one of the United States’s small Indian reservations to life in any of its metropolitan areas since the annual absolute number of deaths from homicide, drug abuse, alcoholism, cancer, heart disease, and automobile accidents will always be far fewer on the reservations than in major cities and their suburbs.

  22. Lindeman 1987: 115.

  23. Keegan 1976: 313.

  24. Meggitt 1977: 112.

  25. See Appendix, Table 6.2, for references for Figure 6.2. Several of these prehistoric war death percentages are comparable (and actually higher than) that of the Jivaro, whose high figure is claimed to be the result of the introduction of firearms.

  26. Cybulski, n.d. in press.

  27. Keegan 1989: 592.

  28. Recent figures indicate that nearly 25 million Soviet citizens died during World War II, of whom less than once-third were military (Weinberg 1994: 894). Almost all these military casualties (about 8 million) would have been male and, given that almost 25 percent of all Soviet males were in the military, a conservative estimate of female casualties is 60 to 65 percent of the civilian deaths. By this crude estimate, the Germans killed almost 15 million Soviet males and over 10 million civilian females, giving a female: male death ratio of less than 1:2. A more complete estimate must include the millions of non-Soviets of both sexes killed by Nazi death squads and camps. In modern history, Nazi Germany is unique in both the scale and the indiscriminateness of its homicides.<
br />
  29. See Appendix, Table 6.2.

  30. (New Guinea) Robbins 1982: 212–13; the Auyana cited this as a high, but not extraordinary, casualty rate; (Papuans) Pospisil 1963: 45, 57; (Blackfoot) Livingstone 1968: 9.

  31. The Soviet Union was 25 percent deficient in males (sex ratio 133:100) at the end of World War II, and West Germany’s population still showed a 22 percent deficiency in males (sex ratio 128:100) in 1960. Poland’s losses, amounting to 20 percent of its prewar population, were proportionally the most severe of any nation during World War II (Keegan 1989: 591–92). In its 1865 to 1871 war against Brazil, Argentina, and Uruguay, Paraguay lost 65 percent of its population and 80 percent of its adult males (McPherson 1988: 856).

  32. Vayda 1976: 23; Meggitt 1977: 174; HNAI vol. 6, 1981: 329, 454–56; HNAI vol. 9, 1979: 392; HNAI vol. 11, 1986: 370, 381; Ferguson 1984b: 277, 281, 285; Bean 1972: 130–31; Chagnon 1968: 129. See Kelly 1985: 19–20 for a possible African case.

  33. Rhodes 1986: 779 (citing Gil Eliot).

  34. This scenario is not ecologically impossible. Many horticultural tribes in New Guinea and Africa sustained population densities ranging from 100 to over 200 persons per square mile (see Brown 1978: 106; Murdock and Wilson 1972: 257–58, 266–72). In 1988, the continental population densities of Europe and Asia had still not exceeded 180 persons per square mile and those of North America, South America and Africa were approximately 60 persons per square mile. The Huron tribe of Canada (lacking plows and domestic animals) supported a precontact population density of 60 per square mile (HNAI vol. 15, 1978: 369). The 1990 Times Atlas (Plate 5) shows this same area today having a population density between 50 and 100 persons per square mile.

  35. (Mae Enga) Meggitt 1977: 101; (Mohave) Stewart 1965: 377–79.