The Horse, the Wheel, and Language: How Bronze-Age Riders From the Eurasian Steppes Shaped the Modern World

by David W. Anthony

  It is unlikely that such a large bundle of common innovations, vocabulary, and poetic forms arose independently in two branches. Therefore, Pre-Greek and Pre-Indo-Iranian almost certainly were neighboring late Indo-European dialects, spoken near enough to each other so that words related to warfare and ritual, names of gods and goddesses, and poetic forms were shared. Greek did not adopt the ruki rule or the sat∂m shift, so we can define two strata here: the older links Pre-Greek and Pre-Indo-Iranian, and the later separates Proto-Greek from Proto-Indo-Iranian.

  The Birth Order of the Daughters and the Death of the Mother

  The ruki rule, the centum/sat∂m split, and sixty-three possible variations on seventeen other morphological and phonological traits were analyzed mathematically to generate thousands of possible branching diagrams by Don Ringe, Wendy Tarnow, and colleagues at the University of Pennsylvania.19 The cladistic method they used was borrowed from evolutionary biology but was adapted to compare linguistic innovations rather than genetic ones. A program selected the trees that emerged most often from among all possible evolutionary trees. The evolutionary trees identified by this method agreed well with branching diagrams proposed on more traditional grounds. The oldest branch to split away was, without any doubt, Pre-Anatolian (figure 3.2). Pre-Tocharian probably separated next, although it also showed some later traits. The next branching event separated Pre-Celtic and Pre-Italic from the still evolving core. Germanic has some archaic traits that suggest an initial separation at about the same time as Pre-Celtic and Pre-Italic, but then later it was strongly affected by borrowing from Celtic, Baltic, and Slavic, so the precise time it split away is uncertain. Pre-Greek separated after Italic and Celtic, followed by Indo-Iranian. The innovations of Indo-Iranian were shared (perhaps later) with several language groups in southeastern Europe (Pre-Armenian, Pre-Albanian, partly in Pre-Phrygian) and in the forests of northeastern Europe (Pre-Baltic and Pre-Slavic). Common Indo-Iranian, we must remember, is dated at the latest to about 1700 BCE. The Ringe-Tarnow branching diagram puts the separations of Anatolian, Tocharian, Italic, Celtic, German, and Greek before this. Anatolian probably had split away before 3500 BCE, Italic and Celtic before 2500 BCE, Greek after 2500 BCE, and Proto-Indo-Iranian by 2000 BCE. Those are not meant to be exact dates, but they are in the right sequence, are linked to dated inscriptions in three places (Greek, Anatolian, and Old Indic), and make sense.

  Figure 3.2 The best branching diagram according to the Ringe–Warnow–Taylor (2002) cladistic method, with the minimal separation dates suggested in this chapter. Germanic shows a mixture of archaic and derived traits that make its place uncertain; it could have branched off at about the same time as the root of Italic and Celtic, although here it is shown branching later because it also shared many traits with Pre-Baltic and Pre-Slavic.

  By 2500 BCE the language that has been reconstructed as Proto-Indo-European had evolved into something else or, more accurately, into a variety of things,—late dialects such as Pre-Greek and Pre-Indo-Iranian that continued to diverge in different ways in different places. The Indo-European languages that evolved after 2500 BCE did not develop from Proto-Indo-European but from a set of intermediate Indo-European languages that preserved and passed along aspects of the mother tongue. By 2500 BCE Proto-Indo-European was a dead language.

  CHAPTER FOUR

  Language and Time 2

  Wool, Wheels, and Proto-Indo-European

  If Proto-Indo-European was dead as a spoken language by 2500 BCE, when was it born? Is there a date after which Proto-Indo-European must have been spoken? This question can be answered with surprising precision. Two sets of vocabulary terms identify the date after which Proto-Indo-European must have been spoken: words related to woven wool textiles, and to wheels and wagons. Neither woven wool textiles nor wheeled vehicles existed before about 4000 BCE. It is possible that neither existed before about 3500 BCE. Yet Proto-Indo-European speakers spoke regularly about wheeled vehicles and some sort of wool textile. This vocabulary suggests that Proto-Indo-European was spoken after 4000–3500 BCE. As the Proto-Indo-European vocabulary for wheeled vehicles has already been described in chapter 2, let us begin here with the Proto-Indo-European terms for wool.

  THE WOOL VOCABULARY

  Woven woolen textiles are made from long wool fibers of a type that did not grow on wild sheep. Sheep with long wooly coats are genetic mutants bred just for that trait. If Proto-Indo-European contained words referring unequivocally to woven woolen textiles, then those words had to have entered Proto-Indo-European after the date when wool sheep were developed. But if we are to use the wool vocabulary as a dating tool, we need to know both the exact meaning of the reconstructed roots and the date when wool sheep first appeared. Both issues are problematic.

  Proto-Indo-European contained roots that meant “sheep,” “ewe,” “ram,” and “lamb”—a developed vocabulary that undoubtedly indicates familiarity with domesticated sheep. It also had a term that in most daughter cognates meant “wool”. The root *HwlHn- is based on cognates in almost all branches from Welsh to Indic and including Hittite, so it goes back to the archaic Proto-Indo-European era before the Anatolian branch split away. The stem is unusually long, however, suggesting to Bill Darden of the University of Chicago that it was either borrowed or derived by the addition of the -n- suffix from a shorter, older root. He suggested that the shorter root, and the earliest form, was *Hwel- or *Hwol– (transcribed as *Hw(e/o)l). Its cognates in Baltic, Slavic, Greek, Germanic, and Armenian meant “felt,” “roll,” “beat,” and “press.” “Felt” seems to be the meaning that unites them, since the verbs describe operations in the manufacture of felt. Felt is made by beating or pressing wool fibers until they are pounded into a loose mat. The mat is then rolled up and pressed tightly, unrolled and wetted, then rolled and pressed again, all this repeated until the mat is tight. Wool fibers are curly, and they interlock during this pressing process. The resulting felt textile is quite warm. The winter tents of Eurasian nomads and the winter boots of Russian farmers (made to fit over regular shoes) were traditionally made from felt. If Darden is right, the most ancient Pre-Proto-Indo-European wool root, *Hw(e/o)l-), was connected with felt. The derivative stem *HwlHn-, the root retained in both Anatolian and classic Proto-Indo-European, meant “wool” or something made of wool, but we cannot be certain that it referred to a woven wool textile. It could have referred to the short, natural wool that grew on wild sheep or to some kind of felt textile made of short wool.1

  Sheep (Ovis orientalis) were domesticated in the period from about 8000 to 7500 BCE in eastern Anatolia and western Iran as a captive source of meat, which is all they were used for during the first four thousand years of sheepherding. They were covered not with wool but with long, coarse hair called kemp. Wool grew on these sheep as an insulating undercoat of very short curly fibers that, in the words of textile specialist Elizabeth Barber, were “structurally unspinnable.” This “wild” short wool was molted at the end of the winter. In fact, the annual shedding of short wild wool might have created the first crude (and smelly) felts, when sheep slept on their own damp sheddings. The next step would have been to intentionally pluck the wool when it loosened, just before it was shed. But woven wool textiles required wool thread.

  Wool thread could only be made from unnaturally long wool fibers, as the fibers had to be long enough to cling to each other when pulled apart. A spinner of wool would pull a clump of fibers from a mass of long-fiber wool and twist them into a thread by handfeeding the strand onto a twirling weighted stick, or hand spindle (the spinning wheel was a much later invention). The spindle was suspended in the air and kept twirling with a motion of the wrist. The spindle weights are called spindle whorls, and they are just about the only evidence that survives of ancient thread making, although it is difficult to distinguish spindle whorls used for making woolen thread from those used for making flaxen thread, apparently the oldest kind of thread made by humans. Linen made from flax was the oldest woven text
ile. Woolen thread was invented only after spinners of flax and other plant fibers began to obtain the longer animal fibers that grew on mutant wool sheep. When did this genetic alteration happen? The conventional wisdom is that wool sheep appeared about 4000–3500 BCE.2

  In southern Mesopotamia and western Iran, where the first city-based civilizations appeared, woven wool textiles were an important part of the earliest urban economies. Wool absorbed dye much better than linen did, so woolen textiles were much more colorful, and the color could be woven in with differently colored threads rather than stamped on the textile surface (apparently the oldest kind of textile decoration). But almost all the evidence for wool production appears in the Late Uruk period or later, after about 3350 BCE.3 Because wool itself is rarely preserved, the evidence comes from animal bones. When sheep are raised for their wool, the butchering pattern should show three features: (1) sheep or goats (which differ only in a few bones) or both should make up the majority of the herded animals; (2) sheep, the wool producers, should greatly outnumber goats, the best milk producers; and (3) the sheep should have been butchered at an advanced age, after years of wool production. Susan Pollock’s review of the faunal data from eight Uruk-period sites in southern Mesopotamia, northern Mesopotamia, and western Iran showed that the shift to a wool-sheep butchering pattern occurred in this heartland of cities no earlier than the Late Uruk period, after 3350 BCE (figure 4.1). Early and Middle Uruk sheep (4000–3350 BCE) did not show a wool-butchering pattern. This Mesopotamian/western Iranian date for wool sheep was confirmed at Arslantepe on the upper Euphrates in eastern Anatolia. Here, herds were dominated by cattle and goats before 3350 BCE (phase VII), but in the next phase (VIa) Late Uruk pottery appeared, and sheep suddenly rose to first place, with more than half of them living to maturity.4

  The animal-bone evidence from the Near East suggests that wool sheep appeared after about 3400 BCE. Because sheep were not native to Europe, domesticated Near Eastern sheep were imported to Europe by the first farmers who migrated to Europe from Anatolia about 6500 BCE. But the mutation for longer wool might have appeared as an adaptation to cold winters after domesticated sheep were introduced to northern climates, so it would not be surprising if the earliest long-wool sheep were bred in Europe. At Khvalynsk, a cemetery dated about 4600–4200 BCE on the middle Volga in Russia, sheep were the principal animal sacrificed in the graves, and most of them were mature, as if being kept alive for wool or milk. But animals chosen for sacrifice might have been kept alive for a ritual reason. At Svobodnoe, a farming settlement in the North Caucasus piedmont in what is now southern Russia, dated between about 4300 and 3700 BCE, sheep were the dominant domesticated animal, and sheep outnumbered goats by 5 to 1. This is a classic wool-sheep harvesting pattern. But at other settlements of the same age in the North Caucasus this pattern is not repeated. A new large breed of sheep appeared in eastern Hungary at Kétegyháza in the Cernavoda III–Boleraz period, dated 3600–3200 BCE, which Sandor Bökönyi suggested was introduced from Anatolia and Mesopotamia; at Bronocice in southern Poland, in levels dated to the same period, sheep greatly outnumbered goats by 20 to 1. But beyond these tantalizing cases there was no broad or widespread shift to sheep keeping or to a wool-butchering pattern in Europe until after about 3300–3100 BCE, about the same time it occurred in the Near East.5

  Figure 4.1 Locations of early sites with some evidence for wool sheep. The drawing is from a microscopic image of the oldest known woven wool textile published by N. Shishlina:(1) Uruk; (2) Hacinebi; (3) Arslantepe; (4) Novosvobodnaya; (5) Bronocice; (6) Kétegyháza; (7) Khvalynsk. After Shishlina 1999.

  No actual woven woolen textiles are firmly dated before about 3000 BCE, but they were very widespread by 2800 BCE. A woven woolen textile fragment that might predate 3000 BCE was found in a grave in the North Caucasus Mountains, probably a grave of the Novosvobodnaya culture (although there is some uncertainty about the provenience). The wool fibers were dyed dark brown and beige, and then a red dye was painted on the finished fabric. The Novosvobodnaya culture is dated between 3400 and 3100 BCE, but this fabric has not been directly dated. At Shar-i Sokhta, a Bronze Age semi-urban trading center in east-central Iran, woven woolens were the only kinds of textiles recovered in levels dated 2800–2500 BCE. A woven wool fragment was found at Clairvaux-les-lacs Station III in France, dated 2900 BCE, so wool sheep and woven wool textiles were known from France to central Iran by 2900–2500 BCE.6

  The preponderance of the evidence suggests that woven wool textiles appeared in Europe, as in the Near East, after about 3300 BCE, although wool sheep may have appeared earlier than this, about 4000 BCE, in the North Caucasus Mountains and perhaps even in the steppes. But if the root *HwlHn- referred to the short undercoat wool of “natural” sheep, it could have existed before 4000 BCE. This uncertainty in meaning weakens the reliability of the wool vocabulary for dating Proto-Indo-European. The wheeled vehicle vocabulary is different. It refers to very definite objects (wheels, axles), and the earliest wheeled vehicles are very well dated. Unlike wool textiles, wagons required an elaborate set of metal tools (chisels, axes) that preserve well, the images of wagons are easier to categorize, and the wagons themselves preserve more easily than textiles.

  THE WHEEL VOCABULARY

  Proto-Indo-European contained a set of words referring to wheeled vehicles—wagons or carts or both. We can say with great confidence that wheeled vehicles were not invented until after 4000 BCE; the surviving evidence suggests a date closer to 3500 BCE. Before 4000 BCE there were no wheels or wagons to talk about.

  Proto-Indo-European contained at least five terms related to wheels and wagons, as noted in chapter 2: two words for wheel (perhaps for different kinds of wheels), one for axle, one for thill (the pole to which the animals were yoked), and a verb meaning “to go or convey in a vehicle.” Cognates for these terms occur in all the major branches of Indo-European, from Celtic in the west to Vedic Sanskrit and Tocharian in the east, and from Baltic in the north to Greek in the south (figure 4.2). Most of the terms have a kind of vowel structure called an o-stem that identifies a late stage in the development of Proto-Indo-European; axle was an older n-stem derived from a word that meant “shoulder.” The o-stems are important, since they appeared only during the later end of the Proto-Indo-European period. Almost all the terms are derived from Proto-Indo-European roots, so the vocabulary for wagons and wheels was not imported from the outside but was created within the Proto-Indo-European speech community.7

  Figure 4.2 The geographic distribution of the Indo-European wheel-wagon vocabulary.

  The only branch that might not contain a convincing wheeled-vehicle vocabulary is Anatolian, as Bill Darden observed. Two possible Proto-Indo-European wheeled-vehicle roots are preserved in Anatolian. One (hurki- ‘wheel’) is thought to be descended from a Proto-Indo-European root, because the same root might have yielded Tocharian A wärkänt and Tocharian B yerkwanto, both meaning “wheel.” Tocharian is an extinct Indo-European branch consisting of two (perhaps three) known languages, called A and B (and perhaps C), recorded in documents written in about 500–700 CE by Buddhist monks in the desert caravan cities of the Tarim Basin in northwestern China. But Tocharian specialist Don Ringe sees serious difficulties in deriving either Tocharian term from the same root that yielded Anatolian hurki-, suggesting that the Tocharian and Anatolian terms were unrelated and therefore do not require a Proto-Indo-European root.8 The other Anatolian vehicle term (hišša- ‘thill’ or ‘harness-pole’) has a good Indo-European source, *ei-/ *oi- or perhaps *h2ih3s-, but its original meaning might have referred to plow shafts rather than wagon shafts. So we cannot be certain that archaic Proto-Indo-European, as partially preserved in Anatolian, had a wheeled-vehicle vocabulary. But the rest of Proto-Indo-European did.

  WHEN WAS THE WHEEL INVENTED?

  How do we know that wheeled vehicles did not exist before 4000 BCE? First, a wheeled vehicle required not just wheels but also an axle to hold the vehicle. The wheel, axle, a
nd vehicle together made a complicated combination of load-bearing moving parts. The earliest wagons were planed and chiseled entirely from wood, and the moving parts had to fit precisely. In a wagon with a fixed axle and revolving wheels (apparently the earliest type), the axle arms (the ends of the axle that passed through the center of the wheel) had to fit snugly, but not too snugly, in the hole through the nave, or hub. If the fit was too loose, the wheels would wobble as they turned. If it was too tight, there would be excessive drag on the revolving wheel.