1. ‘Nearer Jushi’ 車師前 refers to the kingdom or state centred in the Turfan oasis or, sometimes, to the tribe which controlled it. There can be no question that Nearer Jushi refers here to the Turfan Oasis. See for example: CICA, p. 183, n. 618; also note 1.5 above. For the etymology of the name Turfan see Bailey (1985), pp. 99-100, which is ummed up in his sentence: “The name turpana- is then from *druva-pāna- ‘having safe protection’, a name suitable for a walled place.”

   “One other oasis town is currently under excavation. At Yarghul (Jiaohe), 10 km (16 miles) [sic – this should read 10 miles (16 km)] west of Turpan, archaeologists have been excavating remains of the old Jushi capital, a long (1,700 m (5,580 ft)) but narrow (200 m (656 ft)) town between two rivers. From the Han period they uncovered vast collective shaft tombs (one was nearly 10 m (33 ft) deep). The bodies had apparently already been removed from these tombs but accompanying them were other pits containing form one to four horse sacrifices, with tens of horses for each of the larger burials.” Mallory and Mair (2000), pp. 165 and 167.

“Some 300 km (186 miles) to the west of Qumul [Hami] lie [mummy] sites in the vicinity of the Turpan oasis that have been assigned to the Ayding Lake (Aidinghu) culture. The lake itself occupies the lowest point in the Turpan region (at 156 m (512 ft) below sea level it is the lowest spot on earth after the Dead Sea). According to accounts of the historical period, this was later the territory of the Gushi, a people who ‘lived in tents, followed the grasses and waters, and had considerable knowledge of agriculture. They owned cattle, horses, camels, sheep and goats. They were proficient with bows and arrows.’ They were also noted for harassing travellers moving northwards along the Silk Road from Krorän, and the territories of the Gushi and the kingdom of Krorän were linked in the account of Zhang Qian, presumably because both were under the control of the Xiongnu. In the years around 60 BC, Gushi fell to the Chinese and was subsequently known as Jushi (a different transcription of the same name).” Mallory and Mair (2000), pp. 143-144.

“History records that in 108 BC Turpan was inhabited by farmers and traders of Indo-European stock who spoke a language belonging to the Tokharian group, an extinct Indo-Persian language [actually more closely related to Celtic languages]. Whoever occupied the oasis commanded the northern trade route and the rich caravans that passed through annually. During the Han Dynasty (206 BC-AD 220) control over the route see-sawed between Xiongnu and Han. Until the fifth century, the capital of this kingdom was Jiaohe.” Bonavia (1988), p. 131.

“Turpan is principally an agricultural oasis, famed for its grape products – seedless white raisins (which are exported internationally) and wines (mostly sweet). It is some 80 metres (260 feet) below sea level, and nearby Aiding Lake, at 154 metres (505 feet) below sea level, is the lowest continental point in the world.” Ibid. p. 137.

“The toponym Turfan is also a variation of Tuharan. Along the routes of Eurasia there are many other place names recorded in various Chinese forms that are actually variations of Tuharan.” Liu (2001), p. 268.

The Turfan oasis contains a number of ancient sites including Kharakhoja or Khocho where the Maoji xiaowei or ‘Commandant responsible for the military colonies’ at Cheshi [Ch’e-shih] or Gaozhang [Kao-ch’ang] during the Han dynasty – see note 1.5 above.

“By native tongues it is called Apsūs (Ephesus), also the city of Dakianus (after the Roman Emperor, Decius [249-251 CE], the persecutor of the Christians) and Idikuchahrī. The old Chinese name was Kao-chang. The name of Dakianus finds its explanation in the circumstance that in the immediate neighbourhood, in the valley of Tuyoq, there is a sacred shrine of the Seven Sleepers, which, even at the present day, has the reputation of great sanctity, and is visited by Mohammedan pilgrims from such distant lands as Arabia and India. But the Mohammedans were not the first to bring the legend there – Islam did not reach these districts before the thirteenth or fourteenth century – for it dates from Buddhist times, as I shall show later on.
          The old town is an enormous square, covering about a square kilometre or 256 acres…. The massive old wall in many places is still in good preservation. It is almost twenty-two yards high and made out of stamped mud in the fashion common even at the present time from Persia to China. Numerous towers – there are still seventy of them existing – strengthen this wall, which diminishes in solidity towards the top, but which in the lower part is so massive that the builders could have arranged whole suites of rooms within its bulk, especially near the gates.
          The masonry of the gates is destroyed, but there appears to have been a fortified gate in the middle of each of the four walls enclosing the town, and apparently there was a fifth gate in the north-west corner of the wall.
          The buildings are too much destroyed to allow the course of the streets to be plainly traced, but two wide streets, one running from north to south and the other from east to west, seem to have crossed each other in the centre of the town by that Ruin K which we recognized as the shrine of the Manichæan kings of the Uighur Turks…. The ground plan of the town, therefore, doubtless follows the pattern of the Roman castrum.
          The buildings of the city are, without exception, temples, monasteries, tombs – in short, nothing but religious buildings…, for we did not find a single exception to this class. The architecture in every case is either Iranian (with dome-shaped roof,…) or Indian (stūpa,…). No Chinese buildings are to be found in the Turfan oasis or in any other of the old settlements visited by us.
          It was a city of temples, and a necropolis whose strong fortifications in time of war provided a refuge for inhabitants living in simple mud-houses outside the gates. These mud-houses, probably of the same kind as dwellings to-day, have disappeared without leaving a single trace.” von Le Coq (1928), pp. 56-57.

2. Jiaohe 交河 [Chiao-ho] = Yarkhoto, 20 li (8.3 km) west of Turfan. Stein (1928), p. 563 says: “That Chiao-ho, the ancient capital of Turfan, literally ‘[the town] between the [two] rivers,’ is identical with the ruined site of Yar-khoto, ‘the town between the Yars,’ is subject to no doubt.” This false etymology has obviously influenced later writers – see below.
          Pavel Lurje of the Department of Ancient Near East, St. Petersburg Branch of the Institute of Oriental Studies, Russian Academy of Sciences, in an email of 18 May 2000, wrote to correct my draft notes and said that the name Yarkhoto is, “a compound of Monghol khoto ‘town,’ and Turkic Yar = ‘gorge,’ ‘valley’.” This interpretation I later found confirmed by von Le Coq (1928), pp. 53-54:

   “Coming down from the mountain we reached a remarkable valley called, from its great steep loess cliff (yār), the Yār ravine. Here two small streams unite in swampy land, and, on their shores, rise high cliffs in whose steep sided there are cave temples, but on the horizontal surface at the top we find the ruins of a fortified town. This town, where the arrangement of the streets cans still be easily recognized in certain parts, was for a time, the capital of the Uighur realm. To-day it is often called by the half-Turkish, half-Mongolian name of Yar-khoto (the cliff town), whilst its Chinese name is Kiao-ho ….”

   “Chiao-ho..., “connected rivers”, located 20 li or ca. 13 km. [sic – the correct conversion is 8.3 km] East of Kuang-an. Hsü Sung states that the ruins of this town lie near Yarkol, ca. 12 km. NW of Turfan and so does Chavannes (1907), p. 155, note 1. Huang Wen-pi (1954), pp. 3 and 8, identifies it with the ancient town of Yarkhoto or Idiqutshari.” CICA, p. 183, n. 619.

“The ruined city of Jiaohe is ten kilometres (six miles) west of Turpan, perched atop a narrow terrace ravined on all sides by the confluence of two rivers. (The city’s Chinese and Uygur names both mean ‘confluence of rivers.’) [See, however, Pavel Lurje’s comments above]. The cliffs rise 20 to 30 metres (yards) above the riverbeds, forming a natural defence that served instead of city walls.
            The city was established early in the Western Han Dynasty (206 BC–AD 24) and proved an effective fortress when troops and peasants took refuge there from the raiding bands of Xiongnu horsemen. It reached its cultural peak under the Uygurs in the ninth century but was gradually abandoned after the Yuan Dynasty (1271- 1368).” Bonavia (1988), pp. 142-143.