Empires of the Silk Road: A History of Central Eurasia From the Bronze Age to the Present (85 page)

29.
Virtually every account of successful Chinese campaigns into Central Eurasia includes information on the booty acquired, but it is generally ignored by modern historians, who, regardless of where their sympathies may lie, generally list only the number of “plundering raids” by Central Eurasians against the Chinese, for example, the “Chronological Table of Plunder by the Hsiung-nu” of Hayashi (1984: 86–92), who also argues that the Hsiung-nu raids were the major source of agricultural manpower in the steppe empire. The source material presented in his article, and to some extent his own arguments, show that the Chinese in question were refugees from China who had escaped either on their own or in tandem with one of the Hsiung-nu raids in question. Cf. Di Cosmo (2002a: 202, 204).

30.
It is important to realize that before the advent of telecommunications a unitary language could
only
be maintained by continuous, direct intercommunication among its speakers. Some have proposed that the various Indo-European daughter languages were simply the autochthonous, primordial languages of the areas in which they are first attested. For example, Van de Mieroop (2004: 112–113) says:

Under the influence of an outdated nineteenth-century idea that there was an Indo-European homeland somewhere north of India, much attention has been devoted to finding out when and where the Indo-Europeans entered Anatolia and to finding evidence for an invasion. This search is futile, however. There is no reason to assume that speakers of Indo-European languages were not always present in Anatolia, nor can we say that they would have been a clearly identifiable group by the second millennium.

This claim makes no sense either linguistically or historically.

31.
The scenario presented in the text accepts the traditional view of Indo-Iranian, which is based on the understanding of Avestan as the most archaic form of Iranian. However, this view now seems to me to be incorrect. It is interesting to note that some early Indo-Europeanists did not consider Avestan to be an Iranian language. “The similarities between the two languages were so great that some thought that the Avestan language was merely a dialect of Sanskrit” (Mallory and Adams 2006: 6–7). If Avestan’s traditional linguistic status is indeed an error, it will not be possible to support the highly exceptional Indo-Iranian family, the early date of Zoroaster, the putatively shared early Indo-Iranian religious beliefs and practices, and much else. If current theory is wrong, all of the latter belong to Indic alone. However, because the phonological problem has just been discovered, and criticism of the traditional view will certainly be controversial until linguists investigate the problem in detail, I have refrained from modifying the text. See
appendix A
for further discussion.

32.
The current consensus is that the Proto-Indo-European homeland was somewhere between the southern Urals and central Volga in the north and the North Caucasus and Black Sea in the south. However, the distribution of *mori ‘lake, sea’ in the daughter languages (Mallory and Adams 1997: 503–504; 2006: 127) suggests that this word was acquired during the Proto-Indo-Europeans’ initial expansion, as posited here, meaning they would have earlier been nearer the Urals and middle Volga, which region is now considered by many to have been the homeland.

33.
In the literature it is generally known as the Bactria-Margiana Archaeological Complex, or BMAC. See Witzel (2003), who discusses the “body of loan words preserved independently from each other in the oldest Indian and Iranian texts that reflects the pre-Indo-Iranian language(s) spoken in the areas bordering N. Iran and N. Afghanistan, i.e. the Bactria-Margiana Archaeological Complex. These loans include words from agriculture, village and town life, flora and fauna, ritual and religion.” As noted above, the scenario presented here depends to a great extent on the received views of Indo-Iranian, and especially of the Avesta and the date of Zoroastrianism. In particular, the conclusions based on the once accepted idea of the Zoroastrian demonization of much that is thought to have been Indic now appear to be highly questionable. If the received view is incorrect, this scenario must be revised.

34.
The lack of domestic horse bones from earlier sites in the pre-Chinese area, and the exclusion of wild eastern Eurasian horses (Przewalski’s horse) from having contributed to the genetic makeup of domestic horses indicates that, as now accepted, domesticated horses were introduced to the western pre-Chinese area by the Indo-Europeans. As noted below, it is probable that the Proto-Tokharians brought the horses with them, though they seem to have kept them primarily for food. Second-wave Indo-Europeans would appear to be responsible for introducing chariot horses along with the chariot itself and a number of other cultural innovations.

35.
Records in ancient Near Eastern languages, which go back centuries earlier than these references, are totally silent on Indo-Europeans anywhere and do not contain any Indo-European words until this very point in time. Considering that the bodies of people who must have spoken the Proto-Tokharian dialect of Indo-European were buried in the eastern Tarim Basin region beginning around 2000
BC
, as noted above, and that the Proto-Anatolian dialect belonged, like Proto-Tokharian, to the Group A languages of Indo-European, the two groups would seem to have migrated at about the same time. They were the speakers of the first wave whose languages survived long enough to be recorded. It must be emphasized that membership in one or another of the three groups does not imply any genetic subgrouping. For example, Indic (Group B) and Iranian (Group C) are traditionally believed to belong to the same genetic subgroup, Indo-Iranian; however, see
appendix A
.

36.
The earliest name by which the Hittites are known is Nesili (written
Ne
š
ili),
by which they refer to themselves in Hittite, but this name is derived from the name of the Assyrian colonial town of Kanesh
(Kaneš
or
Kanes)
and means simply ‘man of (Ka)nesh’. Melchert rightly notes that “Hittite is an unmistakeably Indo-European language in all respects.” However, he also says, “Earlier claims about heavy non-Indo-European ‘substrate’ or ‘adstrate’ effects on Hittite … were grossly exaggerated” (Melchert 1995: 2152). The problem with this is that the Proto-Indo-European language itself is not attested, and
all
of the daughter languages (or branches) are significantly different not only from Proto-Indo-European but also from each other. The great differences among these languages can be explained best, if not solely, by assuming they were formed by individual creolizations of an original more or less unitary language. This explanation accords with what is known about language change in historical records from Antiquity to modern times. See Garrett (1999, 2006) and Beckwith (2006a), and
appendix A
.

37.
The Old Indic chariot warriors of Mitanni—the
maryannu
(written
ma-ri-ia-an-nu),
from Old Indic
márya
‘young warrior’ (plus the Hurrian plural
-nnu)—
and the Old Indic
marut
‘chariot warrior’ are both connected specifically with horses and chariots
(EIEC
277). The word for these warriors has a cognate in Old Persian
mar
ī
ka
(from Proto-Indo-Iranian *mariyaka) ‘member of a retinue’
(EIEC
630), that is, a band of warriors attached to a lord. “The OInd
márya
‘young man’ (cf. Av[estan]
mairy
? ‘villain, scoundrel’) is employed to describe the wildly aggressive war-band [the Maruts—
CIB
] assembled around the leadership of Indra or Rudra in the Vedas. Although the Indo-Iranian form is usually derived from an e-grade
*merio-
with cognates in other Indo-European stocks (e.g., Mayrhofer 1986–2000: 329–330), McCone suggests that the underlying form may well be an o-grade
(*morios)
with a precise cognate in OIr[ish]
muire
‘leader, chief’ “
(EIEC
31). The correspondence of these forms suggests that the ‘young warrior’ words—from the Proto-Indo-European zero-grade root *mr- and the o-grade root *mor of words for ‘to die, death, mortal, youth’, and so on
(EIEC
150; Pok. 735: *mer-, *moro-s; Wat. 42: *mer)—are related to the derived word *marko (with the highly productive suffix *-ko) ‘horse’
(EIEC
274 *márkos; Pok. 700 *marko-; Wat. 38 *marko-), the ancestor of English
mare,
attested only in Celtic and Germanic *marko ‘horse’, which thus originally meant ‘chariot warrior’s horse’.

38.
In the continuing absence of any clear archaeological dating of the appearance of the first Old Indic peoples in that region the question remains, were the Old Indic chariot warriors involved in the collapse of the Indus Valley civilization after all, as some believe the Rig Vedas say they were? Barbieri-Low (2000: 7) remarks, “Attendant on the collapse of Harappan civilization around 1500 B.C., was an influx of people from the north known as the Aryans. This Indo-European speaking group immortalized their ritual and culture in an epic known as the Rig Veda. In the Rig Veda, the Aryans use wheeled vehicles of several types, but the one they prize most is the horse-drawn chariot.” If this is not correct—and so far no consensus seems to exist on the date of or reason for the collapse of the Indus Valley civilization—what other Indian non-Indo-European urban civilization was located in northwestern India whose towns could have been overthrown by non-urban Old Indic chariot warriors in the middle of the second millennium
BC
, as the Vedas would seem to describe? The problem cannot be brushed aside, as it now generally is. On the controversy—much of it politically motivated—over this and many other problems involved with the Old Indic entrance into India, see Bryant (2001), Bryant and Patton (2005), and especially Hock (1999a, 1999b); in addition, recent general works on Indo-European (Mallory 1989; Gamkrelidze and Ivanov 1995; Mallory and Adams 1997, 2006) contain important material relevant to the controversy.

39.
The design in a seal ring from Mycenae (Drews 1988: 161) shows a hunting scene with an archer on the chariot. Most Mycenaean portrayals, which are a couple of centuries or more later than the Shaft Graves, show a warrior holding a spear rather than an archer. There is much other material on Mycenaean chariots. M. A. Littauer, J. H. Crouwel, and Peter Raulwing, in many publications over several decades (e.g., Littauer and Crouwel 2002; Raulwing 2000), have argued strenuously against the idea that the Mycenaeans were Indo-European-speaking invaders who brought the chariot with them and used it in warfare. The remarkably intrusive culture of the Shaft Graves is not, for them, evidence of any new people, but only of “the rise of vigorous local chieftains” who mysteriously developed for reasons that are “not fully understood” (Littauer and Crouwel 2002: 70). They also argue that the chariot was mainly used for “conspicuous display” and “as an adjunct of the greater military,” not as a key weapon in the armies of the day, the way it was certainly used in Anatolia and neighboring areas of the Near East, according to both textual and pictorial evidence. In addition, they argue that although the chariot “was not introduced to Mycenae by conquerors,” it could have been brought in later—they suggest “gifts” from foreign royalty in the ancient Near East, where the first wheeled vehicles were developed and, they argue, predecessors of the chariot are first attested. They also contend—in face of direct evidence, such as the above seal ring from Mycenae showing a chariot with driver and archer—that, “In Greece, there is no evidence for the association between the military chariot and the bow so well documented in the Near East and in Egypt. Instead, chariots here functioned as a means of transport for warriors who fought not from the vehicle but on the ground with close-range weapons” (Littauer and Crouwel 2002: 70–71). Hunting from a chariot was the same as using it for war. They also ignore the evidence of the “disc-shaped bridle cheek-pieces which are attested in Mycenae from c 1600
BC
and are found somewhat earlier in the steppe region”
(EIEC
245), which agree with other correlations between the Shaft Grave burials and the North Caucasus Steppe burials. Littauer, Crouwel, and Raulwing’s arguments do not make sense historically. They do not accord with the data that they and others present, according to which chariots are first attested archaeologically from the Volga-Ural area in about the twentieth century
BC
, and pictorially in the Near East at Kanesh (Littauer and Crouwel 2002: 45–46, figure 1), in the Kanesh Karum II site dated to ca. 1950–1850
BC
(EIEC
245). This is the same site where the first linguistically attested evidence of an Indo-European language—Hittite—has been found, and the Hittites are also the first people known to have used chariots in warfare, in the seventeenth century
BC
. The Mycenaean Greeks probably did not invent the chariot, nor did they ride in chariots from the steppe to Greece, but the area from which the Mycenaeans undoubtedly did come, or in any case through which they passed, namely the Caucasus region, had known chariots long enough for the Mycenaeans to acquire them if they did not already have them. Littauer and Crouwel’s idea that the distinctive Mycenaean culture arose spontaneously out of nothing is disproved by the archaeology. As Drews (1988: 176) points out, the Mycenaean Shaft Graves simply have no antecedents in Greece, and “to explain the shaft graves as the result of the growth of an indigenous ruling class is circular: the only evidence for the growth of such a class [is] the shaft graves.” See also Drews (2004) for corrections of some of his earlier arguments on the way the chariot was used in warfare. Most of the Near East–centered arguments about the origin, diffusion, and use of the horse and war chariot do not accord with the evidence and must be rejected.

40.
It is often stated that ‘China’ is an anachronism when referring to any polity of East Asia before the unification of the country by the Ch’in Dynasty (whence our name
China)
under Ch’in-shih huang-ti in 221
BC
. Other cultures that later were subsumed by the expanding Chinese people retained distinctive languages into the first millennium
BC
, at least, and it has been argued that it is not even certain if the Shang or Chou ruling strata spoke Chinese. However, though it is true that the
name
China is not earlier than the Ch’in kingdom—as is the idea of a unified country consisting of ethnolinguistically related parts that had earlier not been unified—it is not true that there was no earlier unitary state in the area of the Chinese homeland, the Yellow River region of the North China Plain. The ancestors of the Ch’in and Han empires were the Shang and Chou dynastic states centered in that same region. It is also incorrect to claim that the language of the Shang and Chou people was not Chinese. The Chou Dynasty, founded in 1046 or 1045
BC
, as well as the preceding Shang Dynasty it replaced, were both unitary states in which Chinese was the only written language. It is true that it is unknown whether the native
spoken
language of the Shang and Chou conquerors who created their respective states was different from the local language of the region they conquered, but Ockham’s razor tells us it must have been. What is significant linguistically is that in the Shang and Chou inscriptions essentially one language is recorded—though slightly different in some respects due perhaps to dialect and period changes—and that language is ancestral to the modern Chinese languages. Therefore, ‘China’ may be used without further quibbles to refer to the area that at any particular time in history was occupied by people who spoke a form of Chinese as their native language, from the Shang period to the contemporary era (bearing in mind the very small territory covered by the Shang realm). But the source of that language, which is clearly intrusive typologically in its “homeland,” remains at present unknown, though it was undoubtedly a result, at least in part, of the Indo-European intrusion into the area. That is, it is still uncertain whether Chinese is ultimately a minimally maintained Indo-European language or a local language influenced by Indo-European. This so far largely neglected problem deserves careful attention. For attempts to deal with it, see Beckwith (2002a, 2004b, 2006a).