8. Don Ringe communicated the argument against hurki- to me in a letter in 1997. Bill Darden discussed the Anatolian terms in Darden 2001.
9. I am indebted to Mary Littauer for alerting me to draft experiments carried out in 1838–40 with wagons and carts on different road surfaces, where it was determined that the draft of a wagon was 1.6 times greater than that of a cart of the same weight. See Ryder 1987.
10. For the earliest wheeled vehicles, see Bakker et al. 1999; and Piggott 1983. For European wheels, see Häusler 1992; and Hayen 1989. For Mesopotamia, see Littauer and Crouwel 1979; and Oates 2001. The most comprehensive anlysis of the steppe vehicle burials, still unpublished, is by Izbitser 1993, a thesis for the Institute of the History of Material Culture in St. Petersburg. Izbitser is working on an English-language update from her post in the New York Metropolitan Museum. Other key steppe accounts are in Mel’nik and Serdiukova 1988, and the section on wagons in Gei 2000:175–192.
11. Sherratt’s essays were compiled and amended in Sherratt 1997. He continued to suggest that horseback riding in the steppes was inspired by Near Eastern donkey riding; see 1997:217. An early critical response to the SPR is Chapman 1983.
12. For Neolithic sleds in Russia, see Burov 1997. Most of them were joined with mortice-and-tenon joints, and equipped with bent-wood curved runners. These are the same carpentry skills needed to make wheels and wooden-slat tires.
13. The version of the Renfrew hypothesis I use here was published as Renfrew 2001. For assenting views among archaeologists, see Zvelebil and Zvelebil 1988; Zvelebil 1995; and Robb 1991, 1993. Robert Drews (2001) began in a different place but ended up supporting Renfrew.
14. For the north Syrian origin of the Anatolian Neolithic population, see Bar-Yosef 2002; for the likely Afro-Asiatic linguistic affiliation of these first farmers, see Militarev 2002.
15. See Gray and Atkinson 2003, reviewed by Balter 2003. The linguist L. Trask criticized Gray and Atkinson’s methods, and Gray responded on his homepage, updated March 2004, at http://www.psych.auckland.ac.nz/psych/research/Evolution/GrayRes.htm.
16. Buck 1949:664, with Indo-European terms for turn, turn around, wind, and roll. Gray’s argument for a natural independent development of the term wheel from to turn (wheel = the turner) is further complicated by the fact that there are two reconstructed Proto-Indo-European terms for wheel, and the other one was based on the Proto-Indo-European verb *reth- ‘run’ (wheel = the runner), a different semantic development.
17. Renfrew 2001:40–45; 2000. Renfrew’s hypothesis of a very long-lived Proto-Indo-European phase, surviving for many millennia, is supported by some linguists. For a view that Proto-Indo-European was spoken from the Mesolithic through the end of the Corded Ware period, or about 6000–2200 BCE, see Kitson 1997, esp. 198–202.
18. Childe 1957:394.
19. Mallory 1989:145–146; and Anthony 1991a. For Africa, see Nettles 1996.
CHAPTER 5. LANGUAGE AND PLACE
1. For homeland theories, see Mallory 1989, chap. 6. For political uses of the past in the Soviet Union, see Shnirelman 1995, 1999; Chernykh 1995; and Kohl and Tsetskhladze 1995. For the belief in an Aryan-European “race,” see Kühl 1994; and Poliakov 1974.
2. The Pontic-Caspian steppe homeland hypothesis was defended in English most clearly by Gimbutas 1970, 1977, 1991; and Mallory 1989, updated in Mallory and Mair 2000. Although I agree with Gimbutas’s homeland solution, I disagree with her chronology, her suggested causes for the expansion, and her concept of Kurgan-culture migrations, as I explained in detail in Anthony 1986.
3. See Dixon 1997:43–45. Similarly for Zimmer 1990:312–313, “reconstructions are pure abstracts incapable of being located or dated … no philological interpretation of the reconstructed items is possible.”
4. The tree model does not exclude or deny some areal convergence. All languages contain elements based on both branching structures and convergence with neighbors. On areal borrowing, see Nichols 1992.
5. See Thomason and Kaufman 1992; Nichols 1992; and Dixon 1997. All support the derivation of the Indo-European languages from Proto-Indo-European. Dixon (1997:31), although a critic of the criteria used to create some family tree models, stated: “The genetic relatedness of the Indo-European languages, in a family tree model, has of course been eminently proved.” A good brief review of various approaches to convergence can be found in Hock and Joseph 1996:388–445.
6. Gradual convergence between neighboring languages can result in several different kinds of similarities, depending on the social circumstances. The range of possibilities includes trade jargons, crude combinations of words from neighboring languages barely sufficient to communicate for purposes of trade or barter; pidgins, which evolve from trade jargons or from a multitude of partially known languages in a colonial encounter where a colonial target language supplies much of the content of the pidgin; and creoles, which can evolve from pidgins or can arise abruptly in multiethnic forced labor communities where again a colonial target language supplies much of the content. Unlike pidgins, creoles contain the essential grammatical structures of a natural language, but in a reduced and simple form. They can, of course, be as expressive in song, poetry, and metaphor as any natural language, so the fact that they are grammatically simple is not a value statement. All these ways of speaking pass through a bottleneck of great grammatical simplification. Indo-European grammar is not at all like a creole grammar. See Bickerton 1988; and Thomason and Kaufman 1988.
7. Pulgram, in 1959, suggested that the comparative method, applied to the modern Romance words for coffee, would produce a false Latin root for coffee in Classical Latin. But Pulgram’s claim was rebutted by Hall (1960, 1976). Pulgram’s argument was cited in Renfrew (1987:84–86) but corrected in Diakonov (1988: n. 2).
8. For Pre-Indo-European substrate terms in Balto-Slavic, see Andersen 2003. For Greek and pre-Greek place-names, see Hester 1957; Hainsworth 1972; and Renfrew 1998. In northern Europe, at least three different extinct non-Indo-European languages have been identified: (1) the “language of Old European hydronomy,” preserved principally in non-Indo-European river names; (2) the “language of bird names,” preserved in the names of several kinds of birds, including the blackbird, lark, and heron, and also in other terms borrowed into early Germanic, Celtic, and Latin, including the terms for ore and lightning; and (3) the “language of geminates,” which survives only in a few odd sounds quite atypical for Indo-European, borrowed principally into Germanic but also into a few Celtic words, including doubled final consonants and the wordinitial [kn-], as in knob. See Schrijver 2001; Venneman 1994; Huld 1990; Polomé 1990; and Krahe 1954.
≠ 9. For beech and salmon as terms that limited Proto-Indo-European to northern Europe, see Thieme 1958. Friedrich 1970 showed that the beech root referred variously to beech, oak, and elder trees in several branches, and that in any case the common beech grew in the Caucasus Mountains, making it useless as a diagnostic northern European tree word. Diebold 1985 summarized the evidence against salmon as a limiting geographic term. For the honeybee argument, see the excellent study by Carpelan and Parpola 2001. See also the articles on salmon and beech in Mallory and Adams 1997.
10. This interpretation of Proto-Indo-European *peku is that of Benveniste 1973:40–51.
11. This reconstruction of Proto-Indo-European society is based on Benveniste 1973, numerous entries in Mallory and Adams 1997, and Gamkrelidze and Ivanov 1995.
12. For Proto-Uralic linkages with Proto-Indo-European, see Carpelan, Parpola, and Koskikallio 2001, particularly the articles by Koivulehto and Kallio. See also Janhunen 2000; Sinor 1988; and Ringe 1997.
13. For a Yeniseian homeland, see Napol’skikh 1997.
14. Koivulehto 2001.
15. Janhunen (2000) has somewhat different forms for some of the pronouns. Nichols pointed out in a note to me that the -m and -n shared inflections are not very telling; only a whole paradigm of shared inflections is diagnostic. Also, nasal consonants occur in high fre
quencies and apparently are prone to occur in grammatical endings, and so it is the pronouns that are really important here.
16. Nichols 1997a.
17. For the glotallic theory, see Gamkrelidze and Ivanov 1973; see also Hopper 1973. For their current views, see Gamkrelidze and Ivanov 1995.
18. For discussions of the glottalic theory, see Diakonov 1985; Salmons 1993; and Szemerényi 1989.
19. For critical discussions of the Semitic-Proto-Indo-European and Kartvelian-Semitic-Proto-Indo-European loan words, see Diakonov 1985:122–140; and Nichols 1997a appendix. On the chronology of the Proto-Kartvelian dispersal or breakup, see Harris 1991.
CHAPTER 6. THE ARCHAEOLOGY OF LANGUAGE
1. My definitions are adapted from Prescott 1987. A different set of definitions was suggested by Parker 2006. He suggested boundary as the general term (what I am calling borders) and border as a specific term for a political or military boundary (more or less what I am calling a boundary). Parker tried to base his definitions partly on vernacular understandings of how these words are normally used, a noble goal; but I disagree that there is any consistency of usage in the vernacular, and prefer to use established definitions. In their review of the borderland literature, Donnan and Wilson (1999:45–46) followed Prescott in using border as the general or unspecialized term. The classic work to which I owe a great deal of my thinking is Barth 1969. For archaeological treatments of ethnic borders, see Shennan 1989, and Stark 1998.
2. For the growth of Medieval European regional identities, see Russell 1972; and Bartlett 1993. For the anthropological deconstruction of tribes and bounded cultures, see Fried 1975; and Wolf 1982, 1984. See also Hill 1992; and Moore 2001. For good archaeological uses of this border-deconstructing approach to ethnicity see Wells 2001; Florin 2001; MacEachern 2000; and James 1999.
3. See Hobsbawm 1990; Giddens 1985; and Gellner 1973. Giddens (1985:120) famously referred to the nation-state as a “bordered power-container.” For a different interpretation of ancient tribes and borders, see Smith 1998. He is accused of being a “primordialist”; see his defense in chapter 7. Also see Armstrong 1982.
4. For projectile points and language families in South Africa, see Weissner 1983. For a good review of material culture and ethnicity, see Jones 1997, esp. chap. 6.
5. For New Guineau, see Terrell 2001; see also Terrell, Hunt, and Godsen 1997. For the original argument that biology, culture, and language were separate and independent, see the introduction to Boaz 1911. For California, see Jordan and Shennan 2003. For the other examples, see Silver and Miller 1997:79–98.
6. Persistent frontiers were the subject of a flurry of studies in the 1970s; see Spicer 1971 and a volume dedicated to Spicer by Castile and Kushner 1981. The focus in these papers was the maintenance of stigmatized minority identities. In archaeology, the long-term persistence of prehistoric “culture areas” was discussed long ago in Ehrich 1961. The subject was revisited by Kuna 1991; and Neustupny 1991. My first paper on the subject was Anthony 2001.
7. For the persistence of the Hudson-Valley Iroquoian/Algonkian frontier, see Chilton 1998. For the Linear Pottery frontier, see Zvelebil 2002. For the Jastorf/Halstatt frontier, see Wells 1999.
8. Emberling (1997) used the term redundant rather than robust for material-culture borders that were marked in multiple categories of material culture, and he recognized that this redundancy suggested that these borders were particularly important socially.
9. For Wales, see Mytum 1994; and John 1972. For the genetic border at the Welsh/English frontier, see Weale et al. 2002. For the border near Basle, see Gallusser 1991. On Breton culture, see Jackson 1994; and Segalen 1991. For the German/Romansh frontier in Italy, see Cole and Wolf 1974.
10. For the Ucayali quotation, see DeBoer 1990:102. For language and genetic correlations, see Jones 2003.
11. For the Iroquois, see Wolf 1982:167; 1984:394; and, in contrast, see Tuck 1978; Snow 1994; and Richter 1992. Moore (2001:43) also used intermarriages between Amerindian tribes as an index of general cultural and linguistic mixing: “These [marriage] data show a continual movement of people, and hence their genes, language, and culture, from society to society” (emphasis mine).
12. For the borders of functional zones, see Labov 1994. For functional zones, see Chambers and Trudgill 1998; and Britain 2002.
13. See Cole and Wolf 1974:81–282; see also Barth 1969. Cole and Wolf wrote a perceptive analysis of a persistent frontier in Italy, and then in 1982 Wolf published his best-known book, which suggested that tribal borders outside Europe were much more porous and changeable. In making this argument he seems, in my view, to have made some statements contradicted by his own earlier field work.
14. For the billiard-ball analogy, see Wolf 1982:6, 14. On migration processes generally, see Anthony 1990, 1997. Archaeologists of the American Southwest have pushed migration theory further than those of any other region. For a sampling see Spielmann 1998. For migration theory in Iroquoian archaeology, see Sutton 1996.
15. For the four Colonial cultural provinces, see Fischer 1989; Glassie 1965; and Zelinsky 1973. Although anthropology veered away from cultural geography in the 1980s and 1990s, historians and folklorists continued to study it. See Upton and Vlach 1986; and Noble 1992. For a review of the historians’ interest in cultural geography in North America, see Nash 1984.
16. Clark 1994.
17. Kopytoff 1987.
18. For the Nuer, see Kelley 1985. For the effect of changes in bride-price currencies on basic subsistence economies, see Cronk 1989.
19. On dialect leveling among colonists, see Siegel 1985; Trudgill 1986; and Britain 2004. The degree of leveling depends on a number of social, economic, and linguistic factors; see Mufwene 2001. For Spanish leveling in the Americas, see Penny 2000. On the history of American English dialects, see Fischer 1989.
20. For charter groups, see Porter 1965; and Breen 1984. On German immigrants in Ohio, see Wilhelm 1992. On Puritan charter groups in new England, see Fischer 1989:57–68. On the Maya, see Fox 1987, although now there are criticisms of Fox’s migration-based history; on apex families, see Alvarez 1987; and on the Pueblo, see Schlegel 1992.
21. On leveling and simplification in material culture among colonists, see Noble 1992; and Upton and Vlach 1986. Burmeister (2000) noted that the external form of residential architecture tends to conform to broad norms, whereas ethnicity is expressed in internal details of decoration and ornament.
22. The Boasian approach to borders is reviewed in Bashkow 2004.
23. On the provinces of France, see Chambers and Trudgill 1998:109–123; on the Maasai, see Spear and Waller 1993; on Burma, see Leach 1968, 1960; and for a different interpretation of Burma, see Lehman 1989.
24. On language and ecology, see Hill 1996; and Nettles 1996. Hill’s paper was published later in Terrell 2001:257–282. Also see Milroy 1992.
25. The concept of ecologically determined “spread zones” for languages came from Nichols 1992. Similar ideas about arid zones and language expansion can be found in Silver and Miller 1997:79–83. Renfrew (2002) applied the term spread zone to any region of rapid language spread, particularly any expansion of pioneer farmers, regardless of ecology. Campbell (2002), however, warned against mixing these definitions.
26. For China, see DiCosmo 2002; and Lattimore 1940.
27. For Acholi origins, see Atkinson 1989, 1994.
28. A similar model for the growth of Bronze Age chiefdoms, described long before Atkinson’s case study was published, was by Gilman 1981.
29. For the Pathan-Baluch shift, see Mallory 1992; Barth 1972; and Noelle 1997.
CHAPTER 7. HOW TO RECONSTRUCT A DEAD CULTURE
1. For the history of Christian J. Thomsen’s Three-Age System, see Bibby 1956.
2. I generally follow the Eneolithic and Bronze Age chronology of Victor Trifonov at the Institute of the History of Material Culture in St. Petersburg; see Trifonov 2001.
3. For the impact of radiocarbon dating on our under
standing of European prehistory, see Renfrew 1973.
4. The old carbon problem in freshwater fish is explained in Cook et al. 2002; and in Bonsall et al. 2004. I used their method to create the correction scale that appears in the appendix.
5. A good historical review of radiocarbon dating in Russian archaeology is in Zaitseva, Timofeev, and Sementsov 1999.
6. For a good example of cultural identity shifting in response to changing historical situations, see Haley and Wilcoxon 2005. For Eric Wolf’s and Anthony Smith’s comments on situational politics alone being insufficient to explain emotional ties to a cultural identity see Cole and Wolf 1974:281–282; and Smith 1998, chap. 7.
7. For technological style and cultural borders, see Stark 1998.
CHAPTER 8. FIRST FARMERS AND HERDERS
1. The three sky gods named here almost certainly can be ascribed to Proto-Indo-European. Dyeus Pater, or Sky/Heaven Father, is the most certain. The Thunder/War god was named differently in different dialects but in each branch was associated with the thunderbolt, the hammer or club, and war. The Divine Twins likewise were named differently in the different branches—the Näsatyas in Indic, Kastor and Polydeukes in Greek, and the Dieva Deli in Baltic. They were associated with good luck, and often were represented as twin horses, the offspring of a divine mare. For Trita, see Watkins 1995; and Lincoln 1981:103–124. More recently, see Lincoln 1991, chap. 1. For the twins, see Puhvel 1975; and Mallory and Adams 1997:161–165.
2. For the tripartition of Indo-European society, see Dumezil 1958; and Littleton 1982. There is a good review in Mallory 1989:128–142. For an impressive example of the interweaving of three’s and two’s in Indo-European poetry, see Calvert Watkin’s analysis of a traditional Latin poem preserved by Cato in 160 BCE, the “Lustration of the Fields.” The structure is tripartite, expressed in a series of doubles. See Watkins 1995:202–204.
The Horse, the Wheel, and Language: How Bronze-Age Riders From the Eurasian Steppes Shaped the Modern World Page 51
