China Qinghai Tibet Translation

Some Early Chinese Descriptions of the A Mye rMa Chen Range




Above upper: The holy peak of the Great Grandfather of the Yellow River, A Mye rMa Chen.
Above middle: Tibetan nomads shifting pastures on the plains of rMa Stod County, south of the mountain.

Above lower: An encampment on the plains by the Yellow River in rMa Stod.

Early last month, I ordered a cheap reprint of the 1695 Wanli era “Gazetteer of Xining Commandery” 西寧衛志. The second chapter, titled “Mountains and Rivers” 山川, begins as follows:

崑崙山 在衛治西北故臨羌縣境. “漢書: 地理志” 注云: “崑崙在臨羌, 西北有王母石室, 仙海, 鹽池. 西有弱水, 崑崙山祠. 唐長慶中, 劉元鼎使吐蕃云: 三山中高四下, 曰紫山, 古所謂崑崙. 魯曰悶摩黎.” 元潘昂霄 “黃河志” 云: “吐蕃朵甘思東北鄙有大雪山, 即崑崙. 自山腹至頂皆雪, 炎夏不消, 遠年成冰.” 洪武間, 西平侯沐英, 征西將軍鄭愈追羌俱至此山. […]
Mount Kunlun: This mountain is located within the borders of Bordering-on-the-Qiang County, which is to the north-west of the commandery seat. The Geography section of the Han History says of it, ‘Mount Kunlun is in Bordering-on-the-Qiang. To the north-west there is the stone house of the Queen Mother [of the West], the Sea of the Immortals, and the Salt Lake. To the west there is the Weak Water and the Temple of Mount Kunlun. During the Changqing reign of the Tang Dynasty (AD 821-824), an envoy to the Tibetans, Liu Yuanding, wrote: ‘There are three mountains of which the middle is the tallest, and it descends steeply on all four sides. This is called the Purple Mountain, and it is this which was in ancient days called Mount Kunlun. The barbarians call it Men Mo Li. During the Yuan dynasty (1271-1368) Pan Angxiao wrote in ‘Gazetteer of the Yellow River’: ‘In the north-eastern borders of Tibetan mDo Khams, there is a great snow mountain. This is Mount Kunlun. From the waist of the mountain to the peak it is entirely covered in snow. Even in the heat of summer it does not melt, and over many years it has become ice.’ During the Hongwu reign (1368-1398) the Duke of Xiping, Mu Ying, and the general of the western armies Zheng Yu both pursued the barbarians and reached this mountain.”

Mount Kunlun is a mythical paradise or axis mundi of Chinese legend. It is the pillar of heaven: shamans live on this mountain, various Daoist visionaries journey to it, the Yellow River and other rivers as well flow out from it. The most detailed classical sources for this place are the “Narrative of Mu, Son of Heaven” 穆天子傳  and the Classic of Mountains and Seas 山海經. (I’ve translated all of the sections about the Kunlun from the latter source in a previous post.) I was thus intrigued here to discover that in the 16th century, the inhabitants of Xining City where I live in modern Qinghai province considered the Kunlun to be an actual mountain, located within the jurisdiction of their commandery. The place was distant and the accounts contradictory. Nonetheless, various historical people who were not flighty Daoist poets had reached this mountain and left descriptions of the place and the journey, including transcribed toponyms from the local non-Chinese languages.  

The obvious candidate for such a mountain is the most famous peak in north-eastern Tibet, A Mye rMa Chen (Amne Machen). At 6282 meters, this mountain absolutely dwarfs anything in hundreds of kilometers. The god who lives in this mountain, rMa Chen sBom Ra, is worshiped throughout mDo-Khams as the great genius loci (gZhi bDag) of the entire upper Yellow River regions. (The name A Mye rMa Chen means “Great Yellow-River Grandfather”). In a very real sense, this mountain defines the Yellow River in its upper reaches. The Yellow River rises from pools on the plateaus some 350 kilometers to the west, passes through the two great lakes of sKya Ring and sNgo Ring, and continues east along the southern flank of A Mye rMa Chen. When it reaches the plains of mDzod dGe, the river loops north and then west, flowing parallel to its earlier route along the northern slope of A Mye rMa Chen. In this way the mountain defines the first thousand kilometers of the Yellow River’s flow. Although the mountain was located deep in the country of the feared mGo Log tribes, the deity was worshiped all over Amdo. By 1695, surely knowledge of this had made its way to Xining, the nearest Chinese city. If we should search for an actual holy Kunlun Mountain which is the pole of heaven and from which the Yellow River flows, and which also was located somewhat south of Xining and had a name that was knowable in the Tibetan language, surely this is our candidate.

Since my good friend Anna Sehlanova is writing a PhD thesis at Oxford about this mountain and its god, I decided to see if I could pull up any references in early Chinese sources that seemed to be clearly describing this mountain. This was done by following up the references in the above article and also by scanning through this book “Study on the Kunlun and the Origins of the Yellow River”, 崑崙河源考, a Qing-dynasty kaogu 考古學 work that compiles various legendary and true accounts. Through this method I was able to find four accounts that seemed like they referred to A Mye rMa Chen, all translated in full below. These accounts stretch in time from AD 635 to 1382. Of them the first account seems most sketchy to me (it could refer to A rMye rMa Chen, it could not). The second seems like it probably does and the third and fourth accounts certainly do describe this mountain. The fourth account even provides a poem about the place and a brief description of rMa Chen sBom Ra‘s worship by Tibetans.

The essential bit of evidence demonstrating that all of these accounts do refer to A Mye rMa Chen is a correct hydrological description of the region. All of these accounts state, in one way or another, that the Yellow River rises in a series of pools and then flows into lakes. (The accounts usually refer to these as the “Lakes where Stars Rest” 星宿海, I’m not sure where this name originates.) From there after several hundred kilometers (the number differs) the river passes under the foot of a great snowy mountain, and this is usually given a name in the barbarian language and is stated to be the Kunlun of Chinese legend. This hydrology, given separately in each account, is essentially correct – and the “Kunlun Mountain” in question is mt. A Mye rMa Chen.




Above upper: The plain of the Yellow River in rMa sTod County.
Above middle: The view from a pass in rTsi Kong Thang County, on the way to A Mye rMa Chen, and a ceiling in a monastery near Xia Dawu.
Above lower: A nomad encampment west of the mountain in rMa Stod County.

The earliest account is the most sketchy, but it seems worth translating if only for interest.

A bit of explanation is needed for this and further accounts. In at some point between BC 138 and BC 125, the Han emissary Zhang Qian reached Lop Nor in modern Xinjiang. The account of this provided in the Sima Qian’s “Historical Records” 史記 would confuse Chinese geographers literally for the next two thousand years:

“West of Khotan, the water all flows to the west, and from there enters the Western Sea. East of Khotan, the water flows east, and flows into the Salt Marsh (eg. Lop Nor). From the Salt Marsh the water sinks down and goes underground. South of it [Khotan?] is the origin of the Yellow River. There is much jade here. The Yellow River flows into the Middle Kingdom.

史記, 列傳, 大宛列傳, 6
Historical Records, The Exemplary Account of Da Wan, 6

From this Zhang Qian deduced that the water in Lop Nor lake (which has no outlet to the sea) in fact flowed underground and burst out into China as the Yellow River. The mountains south of Khotan were, therefore, the Kunlun:

“The envoy of Han [eg. Zhang Qian] exhausted the source of the Yellow River. It flows from Khotan, and in these mountains there is much jade, which is mined and brought forth. The Son of Heaven consulted with old maps and books, and called these mountains from which the Yellow River emerges “The Kunlun”.

史記, 列傳, 大宛列傳, 26
Historical Records, The Exemplary Account of Da Wan, 26

Interestingly, Sima Qian himself didn’t totally buy this story about Kunlun. In his “appraisal” at the end of the biography of Zhang Qian, Sima Qian quotes the now-lost “Basic Record of Yu” for a description of the Kunlun:

“The Grand Historian Says: According to ‘The Basic Record of Yu’, ‘the Yellow River flows out of the Kunlun. The Kunlun is three thousand five hundred li tall. This is where the sun and moon hide from each other in order to shine and glow by turns. On the top there are the Sweet Springs and the Jade Lake’. In recent times the envoy Zhang Qian reached behind Parthia, and exhausted the source of the Yellow River. Did he actually see what the ‘Basic Record’ calls the Kunlun?

Therefore I say that that for the mountains and rivers of all the nine parts of the world, the record in the “Book of Documents” [eg., ‘The Tribute of Yu’] is near to describing them correctly. As for the fantastic creatures described in ‘The Basic Record of Yu’ and the ‘Classic of Mountains and Seas’, I won’t even speak of them.”

史記, 列傳, 大宛列傳, 37
Historical Records, The Exemplary Account of Da Wan, 37

Despite this well-placed skepticism of the emperor’s pronouncement that the Kunlun had been located, Sima Qian was doubting only that the mountains Zhang Qian had seen were indeed the mythical axis mundi. He did not dispute that source of the Yellow River had been found. Thus the idea that Zhang Qian had discovered the source of the river in Xinjiang remained a widely accepted in China until the 19th century. The statement “The envoy of Han exhausted the source of the Yellow River” 漢使窮河源 even became a set phrase, and we shall see it repeated in the later accounts. Indeed, this is why the range that borders the Tarim Basin to the south even today bears the official name “Kunlun”. Later on multiple people would reach the actual source and write accurate accounts of it in Chinese. Nevertheless, the powerful orthodoxy of these classical accounts kept everyone arguing back and forth for the next two millenia.

As a corollary to the above, a second mountain was identified as the point from which the Yellow River spilled out of its underground channel and into the known lands of China. This was called the Mountain of Piled Rocks 積石山. The basic source for this is a brief mention in the “Tribute of Yu” section of the Book of Documents, which was one of the Four Books and Five Classics 四書五經 of Confucian orthodoxy.

[Yu the Great] channeled the Yellow River from the Mountain of Piled Rocks to the Dragon Gate, and thence south to the northern side of Mount Hua.

尚書, 夏書, 禹貢, 17
The Book of Documents, The Book of Xia, the Tribute of Yu, 17

In Yu’s great re-ordering of the world during the legendary Deluge, the Mountain of Piled Rocks was the highest point along the Yellow River to which he reached, and therefore the western boundary of known China. For all we know this Mountain of Piled Rocks was probably A Mye rMa Chen too. Later (at least by the fourteenth century) on it came to refer to a particular hill on the river-side west of Lanzhou, which bears this name today. In any case in the following account it seems that the “Mountain of Piled Rocks” may refer to A Mye rMa Chen.

The below comes from the section of the New Tang History 新唐書 on the lands of the Tuyuhun. The Tuyuhun were a nomadic state in that occupied the northern part of the Tibetan Plateau from the third century AD until the first half of the seventh, at which point they were defeated and their lands seized by the expanding Tibetan Empire. The section below describes the victorious campaigns of the celebrated Tang general Li Jing against the Tuyuhun Khagan Murong Fuyun. This took place in AD 635, or just before the advance of the Tibetan armies under Srong bTsan sGam Po into Qinghai:

[The Tuyuhun Khan] Murong Fuyun had gone to the plain of Tulun, and from there intended to escape to Khotan. [The general] Ci Wanjun led riders in pursuit for about a hundred li, and defeated him. But the troops lacked water, to the extent that they had to cut their horses and drink the blood. [The generals] Hou Junji and Li Daozong rode across the empty wasteland for over two thousand li. [In this place] there was frost at the height of summer, and little grass or water. The soldiers ate ice and the horses had snow as fodder. After a month, they passed the River [sic] Where Stars Rest and arrived at the upper part of Bo Lake. They gazed upon the Mountain of Piled Rocks, and viewed the source of the Yellow River. [The general] Chi Shi Si Li led riders to strike at the baggage train of the barbarians. The two armies met at Da Fei river and the valley of Po Luo Zhen.

新唐書, 列傳146上, 西域上, 32
New Tang History, Exemplary Accounts 146 A, “The Western Regions A”, 32

The Old Tang History 舊唐書 biography of Hou Junji 侯君集 gives a slightly different account: 

…[In this place] there was frost at the height of summer, and on the mountains there were great masses of snow. In continuous battles they fought past the River Where Stars Rest, and reached the Bo Lake. Each time they met the barbarians, they achieved great victories and spoils. In the north they could see a Mountain of Piled Jade [sic]. They observed that the source of the Yellow River flowed out of it.

舊唐書, 卷七十三, 二
Old Tang History, Volume 73, 2

The battles around this mountain are described in somewhat more detail in the Exemplary Accounts biography of Xue Wanjun 薛萬均:

At this point Xue Wanjun was promoted to the vice-commander of the Qiemo Route army, and followed Li Jing on his campaign against the Tuyuhun. The army passed by Qinghai Lake. Xue Wanjun and [his younger brother] Xue Wanche each took a hundred soldiers and rode ahead of the main army. They came across the barbarians. Xue Wanjun rode out alone against them, and none dared face him. He returned and told all of the generals, “It will be easy to deal with these bandits.” He returned and pressed the attack, killing over a thousand. He was as brave as three armies. The barbarians fled to the Mountain of Piled Rocks. A great wind lifted the banners, and Xue Wanjun said, “The barbarians are coming!” and he roused the soldiers. Thereupon the barbarians arrived. Xue Wanjun rode straight ahead and slew their general. The rest of them fled away. He pursued them to the wastes of Tulun and then returned, meeting Li Jing at Qinghai Lake.

新唐書, 列傳 19, 侯張薛 17
New Tang History, Exemplary Accounts 19, “Hou, Zhang, and Xue”, 17

In the wake of these campaigns, the entire region of modern Amdo was put under direct Chinese control. The area was organized into thirty two counties (ch. zhou 州), with the regional command at Song Zhou 松州, modern Songpan. The section the Western Regions in the New Tang History comments on the Tang dominion in this area as follows:

Under this system, east from the head of the Yellow River and the Mountain of Piled Rocks, all the country belonged to the Middle Kingdom.

新唐書, 列傳146上, 西域上, 5
New Tang History, Exemplary Accounts 146 A, “The Western Regions A”, 5

As to the identity of this Mountain of Piled Rocks, a few things can be pointed out. The first is that the description of Li Jing and his generals’ campaigns puts them in roughly the right area to stumble upon A Mye rMa Chen – and all the accounts have this mountain at or near the Yellow River source. Second, the Tang soldiers are described as seeing both the source of the Yellow River and a “River Where Stars Rest”, which presumably corresponds to the lake of the same name in later descriptions. 

Third, the “Bo Lake” 柏海 is an interesting question. Earlier in the same New Tang History passage quoted above the general Li Daozong says that “The Bo Lake is near to the source of the Yellow River. No one has ever reached it before.” 柏海近河源,古未有至者。Six years later in AD 641, Li Daozong would visit this lake a second time, this time escorting the famous princess Wencheng 文成公主 to meet her betrothed, the Tibetan emperor Srong bTsan sGam PoThe Old Tang History says, “Srong bTsan sGam Po led his tribes and soldiers past the Bo Lake, and came himself to meet [Princess Wencheng and Li Daozong] at the source of the Yellow River.” 弄贊率其部兵次柏海,親迎於河源。[舊唐書, 卷二百七, 五 / Old Tang History, Volume 207, 5.] The supposition that Li Daozong’s “Bo Lake[s]” is the present day sNgo Ring and sKya Ring thus makes some sense – these lakes are near the source of the Yellow River and are a natural point at which to meet an envoy, located roughly half-way between the last Chinese cities on the Huangshui River and the Tibetan capital at Lhasa.

Finally, this Mountain of Piled Rocks, whether or not it was A Mye rMa Chen, was considered important enough to Li Jing’s campaigns that a replica of it was actually built on his wife’s grave in inner China. 

When Li Jing’s wife died, the emperor ordered that her tomb be built according to the example of [the Han Dynasty generals] Wei Qing and Huo Qubing. It was constructed in the shape of the Steel Mountain and the Mountain of Piled Rocks, in order to display Li Jing’s victories.

新唐書, 列傳 18, 二李勣 11
New Tang History, Exemplary Accounts 18, “The Two Lis, Jing and Ji”, 11

So it may be that Li Daozong and Hou Junji and their troops riding across the highlands in pursuit of the Khaghan of the Tuyuhun in AD 635 were the first Chinese people in history to lay eyes on the great mountain of A Mye rMa Chen. Or it may have been a totally different mountain and the Bo Lake a totally different lake. Hard to prove with certainty, but it seems possible, and interesting to imagine…

IMG_1301 (1)

two golok images

Above upper and lower: Two views of the A Mye rMa Chen massif from Great Horse Pass (rTa mChog Nya Kha) on the south-western side of the sKor Ra route. 
Above middle: A herdsman in rTsi Kong Thang county

The second account seems much more certain than the first. It takes place about two hundred years after the first in AD 822, during the waning years of the Tang dynasty. The account and particularly the famous stele inscription of the treaty described in the account are well-known in Tibetological literature, and both have been translated in full before (see Bushell, 519 for the below account). Nevertheless just for fun I’ve done it again here, with commentary. 

The first account we have read above, of Li Jing, his gallivanting generals, and their triumphant adventures across the highlands of Qinghai, is full of the strength and confidence of the early Tang 初唐. Two hundred years later, the second account is redolent of the decline and conquest of that once-great dynasty. A bit of background is needed to make sense of the narrative in this account, which is intended not simply to describe a diplomatic mission but to also evoke a particular historical memory.

After Li Jing’s original conquests in the 630s, the area of modern Amdo would remain contested between the Tang and the Tibetans for over a hundred years. The Tang built military garrisons and tuntian forts where they could, and the mobile Tibetans would attack these settlements or raid their fields and flocks when opportunities presented themselves. This long struggle culminated for the Chinese at least in the siege of Rock-Castle Fort 石堡城. The traditional location for this fort has it atop the Mountains of the Sun and Moon (ch. 日月山, tb. Nyi Zla Ri) in Huangyuan County, just over the pass from the eastern edge of Qinghai Lake. (There’s an interesting minority opinion that has it on the south side of the Tao River in modern Jo Ne County 卓尼縣, but I’ll leave that argument for now.) 

hacheng and riyue shan small

Above: The walls of the Qing-dynasty Black Fort 哈城 in Huangyuan County 湟源縣, with the Mountains of the Sun and Moon and the putative location of the Rock-Castle Fort in the center background. According to the generally accepted geography, the “Red Ridge” 赤嶺 which was the Tang-Tibetan boundary established in 733 was at the top of the pass suggested to the left of this picture.

The Tibetans occupied the Rock-Castle Fort in 741 and the Tang could not ignore it – the fort commanded one of the main passes leading onto the northern plateau and directly threatened the Chinese farmland of the Huangshui Valley. The general Huangfu Weiming 皇甫惟明 laid siege to the place in the following year but was unable to take it. The place remained a dagger aimed at the heart of Tang for eight more years until in 749 the famous general Geshu Han 哥舒翰 was sent to reduce it, which he did, with catastrophic losses. The “Comprehensive Mirror of Resources for Governance” 資治通鑑 gives an account of the battle: 

“The emperor ordered the commandant of Right-of-the-Long-River Commandery Geshu Han to take the troops of the Right-of-the-Long-River and West-of-the-Yellow-River Commanderies, as well as the troops of Abusi of the Turks, and add to these with troops from the Northern Regions and East-of-the-Yellow-River Commanderies, in total sixty three thousand, and take the Rock Castle Fort of the Tibetans. This fort was defended by sheer cliffs on three sides, and there was only one path up to it. Although there were only a few hundred Tibetans within the fort, they had got much rations. They had also piled up wooden rams and boulders [for rolling down on attackers]. The Tang soldiers attacked the fort from both the front and the rear, but were not able to subdue it. Geshu Han made an assault [continually] for several days but could not take it. He summoned his lieutenants Gao Xiuyan and Zhang Shouyu and wished to execute them [for their failure]. The two of them begged for three days of life in which to achieve victory. They were able to take the fort within the limit. They captured the Tibetan Xi Nuo Luo of the Steel Knife [Fort], as well as four hundred others. The dead among the Tang armies were numbered in the tens of thousands.”

資治通鑑 216, 24
The Comprehensive Mirror of Resources for Governance 216, 24

Incidentally, this is probably what the famous poet Du Fu was talking about with his reference to “the head of Qinghai Lake” in his famous “Song of Soldiers and Carts” 兵車行. In the poem he stands on the bridge at Xianyang 咸陽橋 north of the capital, watching soldiers march west towards the front.

君不見,青海頭,”For sir have you not seen – at the head of Qinghai Lake
古來白骨無人收。Those ancient white bones which now no man collects.
新鬼煩冤舊鬼哭,The new ghosts lament their sorrows, the old ghosts cry –
天陰雨濕聲啾啾。Beneath the darkened heavens and in the wet of rain, their voices rise up.

In any case, when the An Lushan rebellion broke out seven years later, Geshu Han and all his armies holding the garrisons on the Tibetan border were recalled to defend the capital. In a series of events immensely famous in Chinese history, Geshu Han was posted to the Tong Pass 潼關 east of Chang’an, forced by court politics to make an attack on the rebels at Luoyang 洛陽, defeated, and captured, and eventually executed. The rebel armies marched over the pass and took and sacked what was perhaps the greatest city in the world. The Tang dynasty was crippled and for a thousand years no Chinese army marched again into the  highlands of Amdo.

Sixty years later, the account given here is in some sense a bitter journey through that history. In 821, the envoy Liu Yuanding was sent to Tibet to negotiate the fixing of the border between the two nations. This border stood nowhere near the original Tang garrisons along the Yellow River high up in Amdo, but instead ran along the line of the Qinling 秦嶺 range, a dangerous and humiliating few hundred kilometers from the Tang capital. The account describes Liu Yuanding crossing this range and traveling up the valley of the Wei , Yellow 黃, and Huangshui 湟水 rivers through Tibetan-held territory, meeting hordes of Chinese people eager to greet a Tang envoy after sixty years of foreign rule. Everywhere he sees the remnants of China’s past greatness – the destroyed walls of Lanzhou, the toppled boundary stones at the Red Ridge, the abandoned fortifications built by Geshu Han all those years ago. When he reaches the Stone-Castle Fort he beholds it with a sad fascination – it is the site of China’s last, bloody victory over the Tibetans before the whole empire was shattered, the scene of Geshu Han’s last pyrrhic victory before the catastrophe at the Tong Pass, the grave of tens of thousands of Chinese soldiers, the end of Chinese civilization before the high wilderness of Tibet, and he probably the first Chinese official in sixty years to lay eyes on the fabled walls of this place.

Beyond the Red Ridge and the Stone-Castle Fort, Liu Guanding has entered into an alien land. Strange tombs of dot the hillsides, and he journeys on an ancient and crumbling road through a weirdly fertile waste. Finally (and the length of this journey is elided), he reaches the bTsan Po’s summer court on the upper pastures of the sKyid Chu river, and here is all barbaric splendor. A huge ring of pikes is set in the grassland, guarded by soldiers and shamans, and surrounding a massive tent where the Tibetan emperor sits in state, flanked by war chieftains and Buddhist monks. But even the very opulence of this foreign court reminds him of his own dynasty’s fall – the entertainers are all Chinese, no doubt carried off by the Tibetans, and as the envoys feast they play sad old Chinese court songs. 

Finally, on the return journey, Liu Yuanding inquired to his Tibetan escorts as to the source of the Yellow River. Perhaps this took place as they were fording it along the upper reaches in rMa sTod, because he mentions the conditions of the crossing here. The Tibetans told him, and perhaps he saw in the distance, that the river arose in a place called “flooding and welling” 洪濟, and that it flowed down from here a distance of some many li, and that there stood a great mountain which the local people called Men Mo Li and the Chinese had once known as Kunlun.

The 822 stele recording the terms of this treaty still sits outside the Jo Khang temple in Lhasa. Liu Yuanding’s name is recorded on it in Chinese 劉元鼎 and in Tibetan, as Le’u ‘Gwan Deng. I’ve translated the full account from the New Tang History section on Tibet below, with some help from the modern-Chinese version found here. I took the Tibetan names in this version from Li Fang-Kuei and Pan Yihong’s articles, see the bibliography for details.

“In the first year of the Changqing Reign (AD 821), when they heard that a princess had been given in a marriage alliance to the Uyghurs, [the Tibetans] attacked Clear-Border Fort. They were driven away by Li Wenyue. Then the envoy Qi Li Tuo Si arrived at the court and asked for a treaty to be made. The emperor allowed it. Cui Zhi, Du Yuanying, and Wang Bo assisted in the process. They recommended that the treaty be reported to the Imperial Temple. The Official of Rites told them, ‘The emperors Suzong and Daizong both tried to make treaties with the Tibetans, and they did not report it to the Temple. The Dezong Emperor made a treaty during the Jianzhong reign, and in order to strengthen the bonds of the agreement, began the custom of reporting it to the Temple. After the treaty meeting at Pingliang, it was not again reported. The practice has been ceased.’ Therefor the officials dropped the matter.

[Note: All of the names here are preceded by elaborate bureaucratic titles, which I’m not going to try to translate, so this is approximate.] They appointed the official Liu Yuanding as the envoy to the treaty meeting. His subordinate was Liu Shilao. The prime minister as well as the high officials Han Gao and Niu Sengru, Li Jiang of the Ministry of Personnel, Xiao Fu of the Ministry of the Armies, Yang Yuling of the Ministry of Households, Wei Shou of the Ministry of Rites, as well as the other high officials Zhao Zongru, Pei Wu, Liu Gongchuo, Guo Cong, as well as the Tibetan envoy Na Luo all had a treaty meeting in the western suburbs of the capital.

For a message to the treaty meeting, the bTsan Po sent the following: ‘Our two nations should not raid each other. If there are any who should flee to the other side, they should be given grain and clothes and returned.’ The emperor assented to this. All of the the officials signed their name onto the treaty document. Just at the time this treaty was taking place, the Tibetans sent riders and attacked Lu zhou. The commandant of Ling zhou Li Jincheng met them in battle at the Great Stone Mountain, and defeated them. After this the barbarians sent another envoy, Zhao Guozhang. They also sent gifts of confidence [?] to the Prime Minister.

The next year, envoys were received asking to settle the border. Liu Yuanding met with the Tibetan official Na Luo in his nation, and demanded that all of the barbarians affix their names to the treaty documents. Yuanding passed through Chengji, Wuchuan and arrived at the Guangwu range near the Yellow River. Previously, before the walls of this city had been destroyed, the area around Lanzhou was all paddy fields, with peaches, pairs, and lush green willows, and the inhabitants were all people of the Tang dynasty. Now when they saw the banners and canopies of an envoy, they crowded either side of the road to look on. When Yuanding arrived at the city of Longzhi, a thousand old men came weeping to pay their respects, asking if the Son of Heaven was well. One of them said to Yuanding, “It was not long ago that the Chinese armies left here – even today our sons and grandsons have not forgotten the garb of the Tang dynasty. Does the court remember us here? When will our armies return?” When he finished speaking they all began to wail. Yuanding pressed this man more closely, and found that he was originally from Feng County [in the interior].

From there they passed by the Stone Castle Fort. This is built on a cliff wall which is extremely sheer, and the path leading to it is twisting. The barbarians call this the Steel Knife Fort. Going right from there about ten li, the rocks are all red, and the barbarians call this the Red Ridge. The boundary stones that An Wangyi and Zhang Shougui once set here [in 733] had all been upended, and only those stones put their by the barbarians still stood. The Red Ridge is more than three thousand li from Chang’an. In previous days this was located within the commandery of Longyou. This is called the Menhuanlu River, from which it is about a hundred li south to the Lha Sa [ch. Luo suo] river. This is a tributary of the gTsang [ch. Zang] River.

South west of the Yellow River, the land is flat as a whetstone. The plain is a wilderness but very fertile. In the valleys along the rivers there is much tamarisk, and on the hills there are cedars. On the slopes there are a great many grave tumuli, and besides each one is built a little room. These are all painted red, and upon them are drawn white tigers. These are all the graves of barbarian nobleman who committed great deeds in war. In life they wore tiger skins as clothes, and in death these drawings advertise their bravery. Those who died along with them are buried to the sides. From there they crossed the Xi Jie Luo Range. The road here is made of drilled stones, such that carts could travel upon it. This was the road that Princess Jincheng once traveled to Tibet on. From there they arrived at the Valley of Elk [ch. Mi Gu, perhaps Tb. Nyen Kar or Mer Khe?], and stayed in that place.

The northern tributaries of the gTsang River are the summer encampment of the bTsan Po. At the four corners of the camp were spears, and every ten paces were planted long white pikes. In the middle were set great banners so as to create three gates, which were separated at about a hundred paces. Soldiers guarded these gates, as well as shamans with bird hats and tiger belts, holding drums. Anyone who entered this enclosure was searched. In the center of the encampment there was a tall pavilion, surrounded by treasured balustrades. The bTsan Po sat here in a great tent, which was ornamented with gold in the manner of dragons, tigers, and leopards. His hair was tied in a knot which looked like the morning clouds, and he carried a sword enameled with gold. [The monk named] dPal Chen Po [ch. Bo Che Bu] stood  on the bTsan Po’s right, and all of the ministers arrayed themselves beneath the pavilion.

When the Tang envoys first arrived here, the official named Xi Da Re came to the treaty meeting. A great entertainment was put on to the right of the court, with food and wine, as it would have been done in China. For music they sang “The Song of the King of Qin Breaking Formations”, as well as “Liang County”, “The Barbarians on Wei River”, “Important Records” and various other songs. All of the hundreds of performers were people from the Middle Kingdom. The plinth for the meeting was ten paces wide, and six feet high. The Tang envoys and about ten of the barbarian officials sat across from each other, while about a hundred chieftains sat beneath. Above them was set a great couch. dPal Chen Po ascended this, and announced the treaty, while another man standing at his side translated. The envoys smeared the blood of the sacrificial animals on their lips [to formalize the treaty], while dPal Chen Po did not. At the conclusion of the meeting, they once again swore their allegiance to the image of the Buddha. They drank water with tumeric, and all the envoys wished each other well, then descended from the meeting plinth.

On Liu Yuanding’s return journey, he was a guest of the barbarian general Zhang lTa bZan (ch. Shang Ta Zang) on the Great Xia River. He gathered more than a hundred of the eastern officials [of the Tibetan empire] and posted the treaty text upon a high pavilion, so that all would know it. In this way each would be restrained by the treaty to protect his own land and not invade the other side. He dated it to the seventh year of the Tibetan sKyid rTag reign.

Zhang lTa bZan said to Liu Yuanding, “The Uyghur are a small nation. Once I undertook an expedition against them. I reached to within three days of their city and was about to take it. Then I received word that our bTsan Po had died, and so I returned. And yet they are not truly my enemies. Why are the Tang so afraid of them, that they favor them so?”

Liu Yuanding replied to him, “The Uyghurs have merit, and they respect the terms of treaties. Never have they caused chaos or advanced soldiers to seize a single inch of our territory. Therefore we favor them.” By this Zhang lTa bZan was silenced.

Liu Yuanding forded the Huangshui River, and reached the Valley of the Dragon Springs. From there to the north-west one can see the Barbarian-Killing River. The old walls built by Geshu Han were scattered all over this place. The Huangshui River exits from the Meng Valley, and when it arrives at the Dragon Springs it merges with the Yellow River. As for the upper course of the Yellow River, it rises at the Flooding-and-Welling Ridge and then flows to the south-west for two thousand li. The water here is narrow. In the spring it can be forded, while in the summer and autumn it must be crossed on boats. About three hundred li south of it are three mountains, of which the middle is highest, and they descend steeply on all four sides. This is called the Purple Mountain. It is located in the land of Yang Tong. What was in ancient days called the Kunlun is this. The barbarians call it Men Mo Li Mountain. To the east it is a five thousand li distance to Chang’an. The Yellow River has its source in the midst of this, and from there it flows slowly down, gathering together many streams. It’s color here is red, and the distance of its flow is very great. Where it joins other waters it becomes muddy. For this region for generations the lands of the western barbarians have been called the lands of the Yellow and Huangshui Rivers. From the source of the Yellow River, it’s about five hundred li to the north-east to the “sand tail” called Mo He Yan. The sands are about fifty li wide, north of Sha County. To the south-west where they enter the land of the Tuyuhun, the sands narrow, and for this reason it’s called the “sand tail”. It can be conjectured that this is located south of the place called South-of-Sword-Gate. That which Liu Yuanding experienced and saw, can be roughly summarized as above.

新唐書, 列傳141下, 吐蕃下, 25-27
New Tang History, Exemplary Accounts 141 B, “The Tibetans B”, 25-27

Despite some garbling as to the direction of flows and which river is a tributary of which, the hydrology here is essentially correct. The Yellow River wells out of the ground beneath high ridges and flows down for at least a thousand Chinese li (Liu Yuanding has it at two thousand, going south-west) before reaching a great snow mountain of multiple peaks. This is the Chinese Kunlun of old and the local people of Yangtong call it (Middle Chinese) Mòn Mwâ Liei 悶摩黎. This name may (or may not) be the earliest attestation of sBom Ra, and seems to have some continuity with the name given in the next account three hundred years later, (Modern Mandarin) Yi Er Ma Bu Mo La or Middle Chinese Jek Ńɨ́ [Ma] Pǝw Mâk Lât 亦耳麻不莫剌. The brief description also carries with it the interesting suggestion that this name sBom Ra (which has no clear meaning in Tibetan) might have originated in the non-Tibetan language of Yangtong 羊同, the land in which this mountain is supposed to be located. Yangtong is a very obscure kingdom which receives a one-paragraph description in the Tang Comprehensive Encyclopedia 通典, which tells us little useful about this place and people other than that it was indeed located in modern Qinghai province. 





Above upper: A valley on the road to A Mye rMa Chen in rTsi Kong Thang county.
Above middle and lower: Three scenes along the Yellow River, I think near the area that Du Shi called “The Nine Fords” (Modern Khalkha. 
Yesön Garam), now in rMa sTod County. In the middle two A Mye rMa Chen (Du Shi’s “Mountain of Heaven”, Turkic. Tenghri Tagh) is visible on the northern horizon.

The third account comes from the Geography 地理 section of the Yuan History 元史. The text more or less tells its own story. As a brief background, over the course of the early 13th century, the Mongol Empire conquered most of Asia, including by the 1240s the Tibetan Plateau. In 1271 Qubilai Qaγan declared the Yuan Dynasty in Northern China, and by 1279 the conquest of all of China was complete. The next year, presumably in response to having become the most powerful man in the history of the world, Qubilai Qaγan decided to settle once and for all the whole issue of where the Yellow River came from. To this end he chose a Mongol officer named Du Shi and sent him off to Amdo to figure this out. 

The description of Du Shi’s journey is preserved in a peculiar dual-format, which is explained in the text itself below. The description is fairly garbled and the toponyms are hard now to trace, but nevertheless the basic description seems clear. With google maps and some familiarity with the region, one can easily follow the river from the source of the river west of rMa sTod down past sNgo Ring and sKya Ring, along the southern flank of A Mye rMa Chen, down through the gorges in the south of mGo Log, and out to the great riverine plains of mDzod dGe.

A few clues in the text allow us to reconstruct in more detail the Mongol explorer’s precise itinerary through this country. The narrative states clearly that he began his voyage from He County 河州 (modern Linxia 臨夏) and proceeded over the mountains and west from there. The description of the Yellow River between modern roughly rMa Chu County and Gui De 貴德 is accurate but extremely sparse, suggesting that the Du Shi did not actually follow the river in this section. Finally, there’s a confused reference apparently to a lake and a river which are well north of the Yellow River’s route, Alag Lake and the Ulaan River in modern Dulan County.

From this we can speculate that the account given below (which runs from the source down-stream) is the opposite of the direction in which Du Shi actually made his journey. He set out from He County, reached the river at the garrison and postal relay station somewhere on the plains of modern mDzod dGe, and followed the river up from there along the south side of the A Mye rMa Chen range to the source. Having achieved his objective, he took a faster route home, riding north via the above-mentioned Alag Lake and Ulaan River back to the city of Gui De. From there he used river transport to return all the way back along the Ordos Loop to the Central Plains and made it back to Khanbaliq by winter. Later on, the Chinese compilers of his journey confused Alag Lake with rGya Ring and sNgo Ring and thoroughly muddled the various Mongol and Tibetan hydronyms of the mDzod dGe plains, leaving the garbled account below.

I’ve made this speculative map of Du Shi’s journey in Amdo with annotations here. The blue represents the route of the Yellow River, the black is my guess at Du Shi’s route.

A few other interesting things can be pointed out. Perhaps most curious is the great diversity of languages found in the toponomy – Du Shi gives place names in both in Mongol and in Tibetan, with a few more apparently in Chinese, as well as at least one (that of A Mye rMa Chen!) in some form of Turkic. Another worthwhile point to notice here is that sometime between the 8th century and the 13th, the “Mountain of Piled Rocks” seems to have migrated from it’s high-Tang position at A Mye rMa Chen to its current location as a hill (and county) on the Yellow River west of Lanzhou. Finally we can note the mention of the “Envoy of Han” 漢使” and the “exhausting” 窮 of the Yellow River source, a reference to which we will return again later. 


In ancient times, nobody knew the source of the Yellow River. “The Tribute of Yu” traces the source of the Yellow River, but it stops at the mountain called Piled Rocks. In the Han Dynasty, The Envoy of Han Zhang Qian was granted a seal and sent out on a mission to reach the Western Regions. [He wrote that] passing the Jade Gate, he saw two rivers meeting in confluence. [One of these rivers] took its source in the Onion Range, passed by Khotan, and flowed into the Salt Lake. From there it went underground for a thousand li, until it arrived at Piled Rocks Mountains and there re-emerged. During the Tang, Liu Yuanding was sent to Tibet. He made inquiries as to the source of the Yellow River, and arrived at the answer at Men Mo Li Mountain. Therefore for all of the years and months of history, to plumb the depths of this was a great difficulty, and what knowledge had been attained was not more than the above. Of all those over the generations who have discussed the source of the Yellow River, only those two truly sought out the root of the matter. The other explanations are strange and circuitous, and none of them are true. The reason for this is that during the ages of the Han and Tang, the outer barbarians had not all been fully conquered, and the roads were not all fully opened. Therefore such a journey was winding and difficult, and nobody was able to easily reach these places and exhaust their extremities.

The Yuan Dynasty possessed all beneath heaven. From the edge of the sea and all within and without, wherever the footsteps of men reached, they set up postal relays, and wherever these messengers went it was [as convenient] as if they were travelling in the very center of the country. In the seventeenth year of the Zhiyuan Era (1280), Du Shi was made an Official with Power of War and Amnesty, granted a Golden Tiger Seal, and sent to find the source of the Yellow River. Du Shi accepted the order, and that year arrived in He County. Sixty li east of the county seat, there is a place called Pacifying-the-River Station. Another sixty li south-west of the station there is a mountain called Horse-Killing-Pass, where the forest is thick and the barriers are high. From there lifting his feet he went gradually higher, until after a day he had reached the top. Journey west the land became even higher. After four months travel Du Shi reached the source. That winter he returned to make his report, and described the locations of the things he had heard [?]. Later on, the Hanlin scholar Pan Angxiao got a report of Du Shi’s journey from Du Shi’s younger brother Kuokuo, and wrote it up as “Gazetteer of the Yellow River”. Zhu Siben of Linchuan also got from Balijisi’s [an obscure Mongol official] family an illustrated volume in Brahmi characters [sic, probably means Tibetan] which came from the archive of the imperial tutor, and translated it into Chinese. When compared to the gazetteer by Pan Angxiao, both have areas which are more or less detailed. Therefore today we have taken the two books and set down their narratives here, and where they differ, we have annotated them below.

According to these accounts, the source of the Yellow River is in the western extremity of Tibetan mDo Khams.  In this place there are a hundred springs welling forth, pouring out in all directions. Du Shi was not able to examine them all closely, since the area is about seventy or eighty li square, all looked down on by a great mountain. The springs sparkle and shine like stars, and therefore they are called huodun nao’er. “Huodun” means “The Resting Place of Stars”. [Modern Khalkha. Oddyn Nuur “Lake of Stars”] Balijisi’s version reads, “The Yellow River has it’s origin in the south-west of the central continent, three thousand li directly west from the Southerner [ch. Man] Department of Sichuan’s Horse Lake, a thousand five hundred li north-west of Yunnan’s Lijiang district, and two thousand li south-west of Sa Si Jia [tb. Sa sKya?] in the imperial military region. The water comes welling out of the ground here as if it were a well. There are about a hundred such wells. They flow to the north-east for about a hundred li, and then combine to form a great lake, which is called Huodun Nao’er.”

The various streams here flow together, and at nearly five or seven li from this place, they converge into two huge pools, which are called Ala Nao’er. [There is today a lake Alag Nuur “The Speckled Lake” located about sixty kilometers across the ranges north of sKya Ring and sNgo Ring. There seems to be some confusion here with this lake and the Ulaan River (Du Shi gives Hu Lan; it means “red”) which flows down from it.] Going west and then east, [the two lakes] are joined and are subsumed into each other. Travelling for one day, gradually it turns laterally to the east and becomes one river. This is called the Red Treasure [ch. Chi Bao] River. After another two or three days, another river comes from the south-west, which is called the Yi Li Chu. This joins with the Red-Treasure River. After another three or four days travel, another river comes from the south, which is called Hu Lan. After this comes another river, which is called the Ye Li Shu, which flows into the Red Treasure River. By this point the flow has gotten large, and it begins to be known as the Yellow River. At this point the color is very clear, and it can be forded by a person.

Balijisi’s version reads, “The source of the Hu Lan River is in the mountains to the south. In this place there are great mountains and lofty ranges, which stretch laterally for a thousand li. This river flows for about five hundred li, until it merges with the Ye Li Chu river. The source of the Ye Li Chu river is also in the southern mountains. This river flows north-west for five hundred li, until it finally joins the Yellow River.”

After another one or two days’ travel, the river branches into eight or nine legs. The name of this place is Ye Xun Guan Lun, which can be translated as “Nine Crossings” [Modern Khalkha Yesön Garam]. This place is about five or seven li wide, and can be forded by horses. Traveling another four or five days, the water becomes muddy. The local people inflate leather sacks, and ride these to cross it. In other cases, they bind together wood and shields in the form of a boat, tie it together with hair and leather, then cross on it. These boats can only fit two men at a time. From here two mountains press the river on either side to form a gorge, at a distance of sometimes one li, sometimes two or only half a li. The depth of this gorge can only be guessed.

In the north-east of mDo Khams there is a great snow mountain, which is called Yi Er Ma Bu Mo La [tb. A Mye sBom Ra?]. This mountain is the highest of all, and its name can be translated as Teng Qi Li Ta [Presumably Turkic Tenghri Tagh, “The Mountain of Heaven”]. This is Mount Kunlun. From the waist of the mountain to the peak it is all snow, and it does not melt in winter or summer. The local people say, that after many years it becomes ice, and this can be seen in the sixth month [sic]. From the where the river splits into eight or nine legs, it flows to Mount Kunlun, traveling for about twenty days.

Balijisi’s version says, “From the place with the muddy water, the river flows north-east for two hundred li, until it merges with the Huai Li Huo Tu river. The Huai Li Huo Tu river has its origins in the mountains to the south. It flows directly north and then turns west, flowing for about eight hundred li until it meets the Yellow River. From here the Yellow River once again flows to the north-east for another hundred li, until it passes the region called Guo Lang Ma Ha. From there it flows directly north a hundred li, then turns and goes to the north-west for two hundred li. It turns once again and flows directly north for a hundred li, then turns once again and flows east, passing beneath Mount Kunlun. The Tibetans call this Yi Er Ma Bu Mo La. This mountain is extraordinarily tall. The mountain range stretches across the land for about five hundred li, and the Yellow River follows its foot, flowing east. It passes the two places called Sa Si Jia Kuo Ji and Kuo Ti.

The Yellow River flows about half a day’s ride south of Mount Kunlun, and from there west for about five days, until it reaches the regions called Kuo Ji and Kuo Ti. These two places are connected to each other. After another two days, it reaches the place called Ha La Bie Li Chi Er. This is an important place from which roads lead in all directions, and there are many brigands here. An official and soldiers have a garrison here. Close to it, about two days ride to the north, the Yellow River passes by.

Balijisi’s version reads, “the Yellow River passes Kuo Ti, and merges with the Yi Xi Ba Si Jin River. The Yi Xi Ba Si Jin River arises north of the Steel Leopard Range, and flows directly north for a total of five hundred li, until it merges with the Yellow River.”

West of the Mount Kunlun, people are extremely few. Most of them live south of this mountain. The mountains here are not particularly tall, and the river is therefore scattered and winding. Of wild animals there are yaks, wild horses, wolves, leopards, and ibex. East of the Kunlun, the land gradually drops down, and there are cliffs and gorges and barriers, and places so narrow that a fox could jump [from one side to the other]. Travelling for about five or six days, another river comes from the south-west, which is called Na Lin Ha La, which can be translated as “The Narrow Yellow River”. [Presumably Modern Khalkha Nariin Shar, “Narrow Yellow”. It seems to me also that this river has been conflated with the Qi Er/Li Ma Chu (tb. dKar rMa Chu, “White Yellow River”?) below, and that both of them probably refer to the river which is marked on modern maps as the White River 白河.]

Balijisi’s version reads, “The Ha La River arises from north of the White Dog Range. The water flows to the north-west for five hundred li, until it merges with the Yellow River.”

After another two days, a river comes from the south, which is called the Qi Er Ma Chu. The two rivers [sic] merge into the Yellow River.

Balijisi’s version reads:  “From the place where the Ha La River and the Yellow River merge, the river flows due north for two hundred li, passing the A Yi Bo Postal Station, then turns to the north-west. From here it flows along the northern flank of the Kunlun for two hundred li, where it merges with the Qi Li [sic] Ma Chu. The Qi Li Ma Chu arises to the north-west of Wei and Mao Counties, on the north side of the Min Mountains. The water flows north, passing through the territory of ancient Dang County, where it flows due north and then turns to the west for about four hundred li, turning then to the north west, and after another five hundred li, it merges with the Yellow River.”

The Yellow River flows north, and then turns west, passing north of the Kunlun. Then it flows to the north-east, and after about half a month’s travel, it arrives at Gui De County. This place is called Bi Chi Li. From early times it was made a county and had a government by officials. This county is within the prefecture which governs the Tibetans and other areas, which has its seat at He County. Another four or five days from there, one reaches the county called Piled Rocks. This is the “Piled Rocks” from the “Tribute of Yu”. After another five days, one reaches the Peaceful-Village Border Gate in He County. Another day’s journey down, there is the place called the Casting-Nets Pit. From here the Yellow River goes north-east for one day, until the Tao River merges with it from the south.

Balijisi’s version reads, “From the place where the Qi Li Ma Chu river merges with the Yellow River, the river goes north-west, and merges with the Peng Za river. The Peng Za river has its source north-west of Peng Za mountain. The water flows due west for seven hundred or so li, passing the place called Zha Sai Ta Shi, until it meets the Yellow River. Then the river turns to the north-west and flows for three hundred li, then turns to the north-east, where it passes Xi Ning County, Gui De County, Horse Ridge, in total eight hundred li, until it merges with the Miao river. The Miao River has its source from near the military hostel at Qing Tang valley. It flows due east for five hundred li, passing Er Ba Postal Station until it merges with the Yellow River. From there the river flows to the north-east, passing Earth-Bridge Postal Station and the Receiving-the-Qiang Fort in old Piled Rocks County. From the edge of Gou Mi Postal Station in Guo County the capital [sic] is in total about five hundred li. The river passes He County and merges with the Wild [ch. Ye Pang] River. The Wild River has its source north of Westward-Leaning Mountain. The water flows north-east for about a hundred li, where it merges with the Yellow River. From there the river flows north-east for about a hundred li, passing Treading-On-White Fort and Golden River Postal Station to where it merges with the Huangshui River and the Great Flow [ch. Hao Men] River. The Huangshui River has its source beneath the Qilian Mountains. It flows due east for about a thousand li, and the Great Flow River pours into it. The Great Flow River has its origin in Shan Dan County, south of the county seat beneath Shan Dan Mountain. The water flows south-east for seven hundred li, and then merges into the Huangshui River. Then both of these merge with the Yellow River. The Yellow River flows again about a hundred li to the north-east, where it merges with the Tao River. The Tao River has its source north of the Sheep-Scattered [ch. Yang Sa] Range. It flows to the north-east, passing Along-the-Tao Governorate, in total about eight hundred li. Finally it merges with the Yellow River.”

After another day, the Yellow River arrives at Lan County [eg. Lanzhou]. It passes the North Bu Ford and arrives at Singing Sands County. From here it passes Ying Ji Li and flows east, passing south of Ning Xia Governorate. From there it flows east, arriving at Eastern-Victory County, which belongs to the Da Tong Route. From its source to the lands of the Han, this river merges and strings together all of the narrow creeks and streams from the north and the south. And yet nobody knew or had recorded its extremity. All of the mountains were covered in grass and stones, and once you arrived at Piled Rocks, they were entirely covered in forest which was extremely thick. It is said that the Yellow River has nine bends. This area recorded has two bends, [one of which is?] from Qi Er Ma Chu to Bi Chi Li at Gui De County.

思本曰:「[…] 大概河源東北流,所歷皆西番地,至蘭州凡四千五百餘里,始入中國。又東北流,過達達地,凡二千五百餘里,始入河東境內。又南流至河中,凡一千八百餘里。通計九千餘里。」
Balijisi’s version reads, “[Here follows a longish paragraph describing the known route of the Yellow River from Lanzhou to the point where it exits onto the North China plain. This isn’t terribly interesting so I’ll leave it out.] It can roughly be said that as the Yellow River flows north-east from its source, all of the land that it passes belongs to the Tibetans [ch. Xi Fan]. From the source to Lan County [Lanzhou] it’s about four thousand five hundred li, and it is at this point that the Yellow River enters the Middle Kingdom. From there it flows again to the north-east, passing the lands of the Tatars, at a length of two thousand five hundred li. From there it first enters the prefecture of East-of-the-River. There it flows south into the prefecture of Central-River, in total a thousand eight hundred li. In total it can be computed at nine thousand li.

元史, 志15, 地理6, 500-504
Yuan History, Gazetteer 15, Geography 6, 500-504


IMG_1335 (1)


Above upper: Another view of the A Mye rMa Chen, at sun-down at Great Horse Pass.
Above middle and lower: nomad children in a school at rMa Chu Zhang (ch. Huang He Xiang 黃河鄉) and an encampment in the snow on the plains of the Yellow River south-west of the mountain in rMa sTod.

The final reference to A Mye rMa Chen comes from a text called Compilations on the Western Regions 西域集, which was written by a Chan monk named Zong Le 宗泐 sometime in the 1380s. This text is apparently lost but it’s briefly quoted in the “Study on the Kunlun and the Origins of  the Yellow River”, 崑崙河源考. In 1369 the Yuan Dynasty fell and the Ming Dynasty was declared in its place. Ten years later in 1379 the monk Zong Le had the back luck to annoy the Hongwu Emperor Zhu Yuanzhang 朱元璋. The famously irascible emperor told him to go across the Himalayas to India and “find more Buddhist scriptures”. Zong Le was 61 years old at the time. Judging from his account and the poetry attached to it he seems to have taken the whole thing with a sense of humor.

He traveled to He County and from there made his way across the Tibetan Plateau, crossing the Yellow River on the frozen ice, to reach mNga’ Ris. From there he crossed the Himalaya to Nepal and visited the Buddhist holy sites of northern India and then returned the way he came, arriving back in China five years after he left in 1382.

Zong Le’s writing is both vivid and erudite, and he is refreshingly willing to question his own Sino-centric knowledge. He is also interested in Tibetan religion and seems to have understood some of their language. The below is as quoted in “Study on the Kunlun and the Origins of  the Yellow River”, 崑崙河源考:

“The origin of the Yellow River is at Mao Bi Er Che Bo Er Mountain. The Tibetans call the Yellow River Mao Chu [tb. rMa Chu]. The Yak River is called the Bi Li Chu [tb. ‘Bri Chu, “The Female-Yak River” eg., the upper Yangzi]. Chi Ba [Mountain] divides the regions [of flow]. The water which arises from the south-west of this mountain flows into the Yak River. The water which comes out on the north-east side is the source of the Yellow River. When at the western Mountain of Returning Stars I tasted this water, the Tibetans all joked with each other, ‘The Chinese is drinking the Chinese water!’.”

“From the source of the Yellow River it is about seven or eight hundred li east to reach Mount Kunlun. For about three hundred li the river is fordable. Then it goes under the Kunlun and merges with the Kunlun rivers. The traditional account of the Middle Kingdom is that the Yellow River has its source at the Kunlun, but this is not true. The Kunlun is called Mao Bu La Ke. This mountain is the greatest and tallest. In all the four seasons it is covered with snow, and a god lives there. The Tibetan books have descriptions of how within the realm [of this god] he is to be worshiped. There are nine mountains, and this is one of them. These are also recorded [sic].”

The same source, “Study on the Kunlun and the Origins of  the Yellow River”, 崑崙河源考, also contains the following wry poem by Zong Le. The poem is about the the mythical Kunlun, the real mountain A Mye rMa Chen, the god rMa Chen sBom Ra, and that “Envoy of Han” who claimed to have “exhausted the source of the Yellow River” all those fifteen hundred years before. I’m sure there are other interesting Chinese accounts of A Mye rMa Chen, but right now this poem is the last thing I feel like translating.


積雪覆崇岡 The massed snows covered high in mists
冬夏常一色 Whether winter or summer, the color never changes.
群峰譲獨雄 The ranks of peaks give way to a lone hero –
神君所棲宅 The abode of a gentleman god.
傳聞嶰谷篁 Legend has it that in the bamboo thickets of Mount Xie
造律諧金石 They make flutes with sound sweet as gold.
草木尙不生 But here no grasses have ever grown
竹産疑非的 And I doubt you could get any bamboo.
漢使窮河源 “The Messenger of Han exhausted the source of the Yellow River”
要領殊未得 And yet perhaps in truth we still haven’t yet reached the point –
遂令西戎子 And it’s been making these western barbarians
千古笑中國 Laugh at our Middle Kingdom for the last thousand years.
老客此經過 This old traveler passing here
望之長嘆息 Gazes upon it all and heaves a long sigh
立馬北風寒 With our horses stopped the north wind is cold
回首孤雲白 and when I look back, the lonely clouds are white.

IMG_1322 (1)


Below Upper: Looking east towards the A Mye rMa Chen massif on the plain of rTa Bo Zhol Ma.
Below Lower: The glacier at Little Cliff Settlement Pass (tb. ‘Brag sDe’u Nya Kha) on the A Mye rMa Chen circumambulation road. 

[Note: all the pictures in this post come from a walking trip from sKu ‘Bum Monastery to rMa sTod County via A Mye rMa Chen and its circumabulation route in spring 2009]



Generally speaking, please excuse my atrocious bibliography habits.

The following works I got from the Chinese Text Project: (Sturgeon, Donald. Chinese Text Project. Web. 18 May 2016. <>.) This is bad bibliographic practice, but, the precise editions I used are “wiki” versions found there. In rough chronological order:

  • 尚書 “The Book of Documents”
  • 史記 “Historical Records”
  • 舊唐書 “The Old Tang History”
  • 新唐書 “The New Tang History”
  • 通典 “Comprehensive Encyclopedia”
  • 資治通鑑 “Comprehensive Mirror of Resources for Governance”
  • 元史 “Yuan History”
  • 崑崙河源考 “Study on the Kunlun and the Origins of  the Yellow River”

Secondary works are as follows:

  • Bolor Dictionary. Bolorsoft LLC. Web. 18 May 2016.
  • Bushell, Stephen W. The Early History of Tibet from Chinese Sources. London: Trübner, 1880. Print.
  • Dotson, Brandon, and Guntram Hazod. The Old Tibetan Annals: An Annotated Translation of Tibet’s First History. Wien: Verlag Der Osterreichischen Akademie Der Wissenschaften, 2009. Print.
  • Dorofeeva-Lichtmann, Vera. “Where Is the Yellow River Source? A Controversial Question in Early Chinese Historiography”. Oriens Extremus 45 (2005): 68–90. Web.
  • Enoki, Kazuo. “Tsung-Le’s Mission to the Western Regions, 1378-1382.” Oriens Extremus 19 (1979): 47-53. Print.
  • 何, 孝荣. “元末明初名僧宗泐事迹考.” 中國民族宗教網. 中國民族报社, 22 Dec. 2014. Web. 18 May 2016.
  • Li, Fang-Kuei. “The Inscription of the Sino-Tibetan Treaty of 821-822.” T’oung Pao44.1 (1956): 1-99. Web.
  • 劉, 敏寬. 西寧衛志 (明). Ed. 繼光 王. 西寧: 青海人民出版社, 1993. Print.
  • Pan, Yihong. “The Sino-Tibetan Treaties in the Tang Dynasty.” T’oung Pao 78.1 (1992): 116-61. Web.
  • “新唐書, 列傳一百四十一譯文.” 詩詞吧. 古詩文網. Web. 18 May 2016.

EDIT: One last source; the title in English is “Machen Pomra in Chinese-Language Historical Sources”. I didn’t actually use the below when writing this, but afterwards someone sent it to me and it covers basically the same territory as what I wrote, with a few more Qing Dynasty sources (I didn’t bother with these) and interesting stabs at getting the original Tibetan names. The author of this paper (a scholar at Lanzhou University of Nationalities named Liu Tiecheng) and I came to most of the same conclusions. So, if you want an actual well-cited academic paper about this topic, see the below, it’s very good! 

  • 劉, 鐵程. “漢文史料中的”瑪卿邦熱”(19th-20th Century).” 中國民族學 10 (2013): 99-104. Print.


Art China Qinghai Tibet Translation

The Long Valley of Trefoils, or, Some of the Outer Regions of ‘Ol Mo Lung Ring

bon rgya dgom pa


'ol mo lung ring stretched FULL

Above upper: The monastery of Hundred Bon (tb. Bon brGya dGon Chen) in a the mountains above Rebgong.
Above middle: Inside the big new pagoda in Bon brGya
Above lower: A map of the Bon holy realm ‘Ol Mo Lung Ring, which hangs in the main prayer hall.

Last November, my friend Jasper Henderson and I made a trip up to Rebgong to attend a ‘Cham dance at Hundred-Bon Monastery (bon brgya dgon chen). While we were there, we took the opportunity to photograph the details of a giant map of the Bon holy land ‘Ol Mo Lung Ring which hangs in the main prayer-hall there (see picture above). This was apposite since at the time I was being bothered by some Ukrainian kid who showed up from this left-fascist group Sut’ Vremeni and in good Nazi-esotericist style seemed convinced that “Dark Bonpo” was the key to their movement’s world domination and that Sergei Kurginyan and the other luminaries at the top where in on this too. Against this background of general Russo-Tibetan weirdness it seemed useful to go and see some Dark Bonpo for ourselves. And since the whole Sut’ Vremeni Bon interest was apparently influenced by the zany late-60s theories of Whatshisface Kuznetsov and Lev Gumilyov about Bon geography being an elaborately disguised map of Sasanian Persia, it seemed worth it to photograph the big ‘Ol Mo Lung Ring diagram too.

(It’s worth stating at the outset here that I’m a skeptic about the whole “Bon texts contain information brought from Central Asia” theory. Among other things the whole Bon holy-land called sTag gZig equals “Tajik” idea, which is quoted as fact in a whole number of sources, seems extremely unproven to me. These Tibetological authors seem blithely content to ignore the complicated history of that ethonym in Central Asia, as well as the fact that sTag gZig already has a perfectly good mythical-toponym meaning in Tibetan [“Tigers and Leopards”]. The rest of the not-obviously-invented geography about this place seems to refer fairly explicitly to Mount Kailash and the regions surrounding it. I’m not saying that some aspects of Bon didn’t come from Central Asia but this seems very unproven to me at present.)

Anyways photographing this map was easier said than done – the map hangs quite high up in the dimly lit hall. The monks brought us two tall stools, and we balanced precariously there. Jasper shone a flashlight on the painting and I, trying to keep my hands steady, was able to photograph some of the lower bits. Then, because this blog is generally about sacred geography, world mountains, and visual lists and itineraries in Asian art, I’ve tried to translate some of the captions here.

Katie Buffetrille published an article in 2009 (“Khyung Mo Monastery [A’mDo] and it’s ‘Map’ of ‘Ol Mo Lung Ring”) which briefly describes a nearly identical map which hung another nearby Bon monastery, Khyung Mo dGon Pa in Tri Kha / Guide 貴德. Later on I went to this monastery but wasn’t able to see the map there. In any case her pictures and descriptions demonstrate that the two maps were almost identical. Buffetrille’s map in Khyung Mo monastery was produced in the ’70s or ’80s by a local painter who copied it from an original by one sBra Ser Pandita, who was active in the late eighteenth and early nineteenth centuries. At least one other copy of the map exists, located in dGa ‘Mal dGon Pa in Shar Khog (Buffetrille, 323). If there is such a detailed dedication on the Hundred-Bon map I didn’t see it, but it’s unarguably another late 20th century copy of this, no doubt made from one of the neighboring maps.

The dedication on the map (located just beneath the palace of Sham Po in the lower-center) reads as follows:

mchan yig bris pa dge slong bsod nams g.yung drung dang / lha bzo ‘od gsal lha mgon rnam sras sku mched gsum / rgyu ‘byor yon bdag dad ldan so nag dbang phyug gis  / dmangs sgor stong phrag gnyis bcas phul ba’i bsod nams kyi / dge ba’i mthu la brtan nas zhing ‘dir myur du skye bar shog / sarba mngaga lam /

The notes were written by dGe sLong bSod Nams g.Yung Drung / The artists [lit: “god-makers”] were the three brothers, ‘Od gSal, Lha mGon, and rNam Sras / The donor of the wealth [for the painting], the faithful one, Black-Tooth dBang Phyug / made an offering of two thousand yuan / may the power of the meritorious virtue quickly arise at this place / Sarva Mangalam!

As both Katia Buffetrille and Dan Martin point out, there seems to have been a slow process of elaboration of this sacred geography over the centuries. The earliest eleventh or twelfth century Bon geographical descriptions give a fairly sparse list of countries which seem to be derived partly from early Buddhist sources – of the non-Buddhist locations which appear in these lists, one of them is sTag gZig ‘Ol Mo Lung Ring (Martin, 260). Later on these geographies became increasingly elaborate and, to my eye, increasingly fantastical, until at least by the 19th century gigantic maps such as the one above were being produced, containing hundreds of individual place-names. The one below is reproduced in both Snellgrove (plate XXII) and Martin (p. 270) – Dan Martin speculates that it was created based on 12th century geographical texts and some other source, possibly the 14th century gZi brJid (Martin 275). He also points out that it seems to be mainly a gazetteer of places associated with the life of the mythical Bon founder, sTon Pa gShen Rab.

from snellgrove

I don’t think I’m up to a full comparison of all these sources at the moment, which would necessitate a full transcription of the entire map and a long paper in itself. In any case the map in Hundred-Bon and the one supplied by Martin and Snellgrove seems to have one major difference, which is that in their version there are in total six rings of earth and water, while the Hundred-Bon and Khyung Mo versions only have five.

There’s also the below Thang Ka. This is reproduced in Dan Martin (Olmo Lungring, a Holy Place Here and Beyond) and also on the wikipedia page for sTag gZig ‘Ol Mo Lung Ring, which is where I downloaded the image. According to the citation in Martin’s article, this is a 19th century Tibetan production which is currently in the possession of the Rubin Museum in New York, and that’s all the information I’ve been able to get on it.


Speculatively, it could also be pointed out regarding the apparently rather recent production of all of these Bon maps, that there are two fairly obvious and nearby Buddhist inspirations for the large-scale depiction of ‘Ol Mo Lung Ring. The first is the big depictions of Shambhala commonly found on the flanking walls of the entrances of large Buddhist monasteries in at least in Amdo and probably elsewhere too. These usually show the circular realm of Shambhala (tb. Sham BHa La) surrounded by mountains and the armies of Raudra Chakrin (tb. Drag Po ‘Khor Lo Can, “The Fierce One of the Wheels”) marching out to defeat the forces of evil at the end of this age. If Buddhists were going to depict their Central Asian holy realm in their monasteries, the Bonpo obviously felt the need to match them.



Above: The holy realm of Shambhala, arranged like an eight-spoked wheel. 


Above: Raudra Chakrin slaying the king of the Mleccha.

The second inspiration for the Bon ‘Ol Mo Lung Ring drawing is probably the great cosmological diagrams which can less commonly be found flanking the same entrances as the Shambhala drawings. These usually show the great axis-mundi Mount Meru, with the concentric rings of mountains, seas, and continents spreading out at its foot. This concentric cosmology is suggested in the Bon map by the nested squares of lands and seas, which increase in sanctity as one travels inward and culminate in the white mountain of Kailash (tb. Gangs Ti Se) at the center.


Above: An cosmological diagram showing Mt. Meru with the heavens extending from its tip, surrounded by rings of mountains and nine continents. Taken from sGo dMar dGon Pa in Rebgong.


Above: Another more contemporary-style image of the Buddhist cosmos from La Mo bDe Chen in gCan Tsha.

Depictions of Mount Meru go back to even before the birth of the Buddha, but the detailed diagrams of Raudra Chakrin’s ride out of Shambhala seem to have been a fairly recent topic in Tibetan monastic art. Speculatively, this may suggest something as to the impetus for the Bonpos to start producing pictures of their own holy land.

Returning to our original Bon map of ‘Ol Mo Lung Ring, I’ve tried to translate some of the place names attached to the pictures. Going in a widdershins Bon kora, I’ve started out from the lower-left corner and swung right from there to the central palace in the lower part of the picture. The upper parts, while maybe more interesting, were too high up to photograph easily and so I haven’t been able to reproduce or translate them in detail. East faces down in this map, hence the great palace of Sham Po on the eastern river Na Ra Dza Ra there. Thus west is at the top, north to the right, and south to the left. Excuse my bad Tibetan translation skills please, and correct all mistakes you find!


The Lower Left Corner:

lower left - FULL

lower left charnel ground

dur khrod // mya ngam thang nag sdobs chen khyab pa bsgral ba’i dur khrod // dpal mgon // dur khyi //  dur bya // ma mo yum // dbal bon // sa ‘dul mchod rten // tur me //

Charnel Ground // The Liberating Charnel Ground [called] Black Plain of Suffering, Full of Powerful Ones. // The Glorious Protector // Charnel Dog // Charnel Birds // Female Ma Mo // A Bonpo of Fire // An Earth Subduing Stupa // Charnel Fire /

lower left - bya ri gtsug ldan

g.yung drung gtsug gshen rgyal ba // bya ri gtsug ldan ‘bar ba naks tshal gyi dgon pa /

The Eternal Highest gShen King // The Forest Monastery of the Bird Mountain of Shining Peacocks


lower left - rtag gzigs bya ri

lower left - rtag gzigs bya ri - detail

rtag gzigs bya ri gtsug ldan gyi grub gnas /

Perfected Place of the Bird Mountain of Peacocks in rTag gZigs /

lower left - rang byung mchod rten

rang byung mchod rden dkar po // phyag [ ] mchod pa’i zhing mchog /

A Self-Arising White Stupa // The Pure Realm of Hand [ ] Offerings /

lower left - gser gling

lower left - gser gling - detail 01

gser gling gser rgyud ces pa rgyal po’i khab / gnod sbyin pho mos sgo khyi byas pa’i tshul /

The House of the King who Speaks the Golden Tantra of the Golden Realm / The Place Where Male and Female Yakshas Act as Door-Dogs /

lower left - gser gling - detail 02

gnod sbyin pho /

A Male Yaksha /

lower left - brag seng ge rgyab bsnol

lower left - brag seng ge rgyab bsnol - detail 01

lower left - brag seng ge rgyab bsnol - detail 02

brag seng ge rgyab bsnol /

The Lions of the Cliffs, Opposite to Each Other /

lower left - phya rje rgyal po

phya rje rgyal po’i rgyal sa / [ ] ling ma’i yul // btsun mo /

The Kingdom of the King [called] the Lord of Fate / The Land of the [ ] Ling Ma. // A Pure Woman /

lower left - dga' ldan lha yi yul

lower left - dga' ldan lha yi yul - detail

lower left - dga' ldan lha yi yul - detail 02

dga’ ldan lha yi gling / ston pas lha sa bon ‘khor bskor ba’i gnas yin /

The Realm of the Joyful God / This is the Abode where the Buddhas of the God’s Place Bon Turn the Wheels /

lower left - gsas khang pad ma

gsas khang pad ma ba bkra gling / lha mo su brgya’i pho brang / btsun mo phya za ‘gu li[ ] ma’i sgrub pa sbyang ba’i […]

The Brilliant Realm of the Lotus Temple / The Palace of a Hundred Goddesses / […] of the Practice and Attainment of the Pure Woman of Fate [‘gu li{ } ma?]


The Lower Left Center:

lower center left - FULL

lower center left - bdud ma hrang khagrub pa thob pa stong dang rtsa brgyad bzhugs / shin tu gnyan pa’i gangs yin sgom pa dka’ // bdud ma trang [?] kha chen btul ba’i gnas //

The Dwelling of the One Thousand and Twenty Eight Achieving and Attaining Ones / The Snow is Extremely Fierce, and [there is?] Very Austere Meditation // The Abode of the Taming of the bDud Demon, the Great Trang Kha.

lower center left - charnel ground

sos med khrag mtsho dge stig lam bsgral ba’i dur khrod // sa ‘dul mchod rden // dbal bon khro gtum gdang bkra // dur bya // dur khrod // dur me // bstan pa’i zhing// mkha’ ‘gro // / bdud rgyal stong dur shing // dpal mgon keng chen rag gcod /

The Charnel Ground of Liberation from the Road of Good and Evil, [Called] The Lake of Blood from which there is No Resurrection// Earth Subduing Stupa // A Wrathful and [gdang] Splendid Bon of Fire // Charnel Bird // Charnel Ground // Charnel Fire // The Plane of the Teachings // A Dakini // The Charnel Tree of the Thousand bDud Demon Kings // The Blood-Cutting [?] Glorious Lord, the Great Keng /

lower center left - chu bon ra dza ra la

chu bo na ra dza ra la gru gzings kyis bsgrod pa’i tsul /

The River Na Ra Dza Ra La, the Place Where One Must Travel by Boat /

lower center left - dben gnas shel brag

lower center left - dben gnas shel brag - detail 01

lower center left - dben gnas shel brag - detail 02

dben gnas shel brag gnam skas can /

The Crystal-Cliff Hermitage, which has a Sky Ladder /

lower center left - gsas rje lhos kyi zhing khams

gsas rje hos kyi zhing khams / hos dang ba yi rang gi rgyal sa /

The Realm of the Holy Lord of the Hos / The Royal Land of the Pure Hos /

lower center left - od kyi lha ri

rin chen khrus kyi rdzing bu / dpag bsam shing skyid tsul // ‘od kyi lha ri spo mthon / khams chen po brgyad dang dge bsnyen theg pa gsungs pa’i gnas // a mo li ka’i  rta rgyang grag gcig gis [ ] ba yong /

The Pool where Jewels are Washed / The Happy Realm of the Wish-Granting Tree // The Lofty Summit of the Mountain of the God of Light / The Eight Great Realms and the Realm of the Words of the Vehicle of Those who Approach Virtue / The Horse of A Mo Li Ka which Arrives [ ] over a Distance of One rGyang Grag [~two miles] /

lower center left - phyags gyung trung ba

lower center left - phyags gyung trung ba - detail 01

shar phyogs [?] gyung trung ba gling chen /

The Great Swastika Realm of the Eastern Regions /

lower center left - ri bo sha kya seng ge

ri bo shakya seng ge rtsa ba rgya mtsho gling drug ‘khyil ba’i dbus / mchod rtan bkra shis mang gi gtsos cir yang sprul sku’i mchod rtan brgya dang brgyad / rgyal bu dge ba’i ‘khor lo can la bstan pa rjes su bzhag /

In the Midst of the Six Ocean Realms of the Shakya Lion Mountain Peak / There are Many Auspicious Stupas and Everywhere There are One Hundred and Eight Stupas of Reincarnated Lamas / [?] These Were Placed After the Teachings [by the?] Virtuous Prince who Holds the Wheels /

lower center left - sman la nyams len

sman la nyams len mdzad pa’i dgon pa /

The Monastery of Acting Upon Experience in Medicine /


The Lower Central Palace:

lower center palace - FULL

shar phyogs gsas mkhar sham po lha rtse gnas / 
The Realm of the Holy Peak of Sham Po, in the Fortress of the gSas in the Eastern Regions.

lower center palace - mtha' yi rgyal phran

lower center palace - mtha' yi rgyal phran - detail 01

lower center palace - mtha' yi rgyal phran - detail 02

lower center palace - mtha' yi rgyal phran - detail 03

mtha’ yi rgyal phran nams kyis [ ] rigs kyi rgyal sa nas che rtags [?] yig tshang blangs nas phyir pebs pa’i tsul /

The Place where All the Greatest Feudal States Take The Insignia and Documents of their Greatness from the Kingly Realm of [ ], and then Return /

lower center palace - gyung drung dgu brtsegs

lower center palace - gyung drung dgu brtsegs - detail 01

lower center palace - gyung drung dgu brtsegs - detail 02

g.yung drung dgu brtsegs mthong ba’i shar phyogs na rgyal rigs la skad rigs mi gcig pa sum cu yod / grong khyer ‘bum tso brgyad yod / 

In the Eastern Regions there are Thirty Different Languages of Each Royal Line of the Nine Swastika of the Thousand Tiers / There are Ten Thousand and Eight Cities /

lower central palace - rin chen khrus gyi rdzing bu

rin chen khrus kyi rdzing bu /

The Pool of Washing Jewels /

lower central palace - rin chen khrus gyi rdzing bu - detail 02

lower central palace - rin chen khrus gyi rdzing bu - detail 01

chu bo chen po na ra dza ra la zam pa btsugs pa’i tshul /

The Place Where a Bridge Has Been Erected on the Great River Na Ra Dza Ra /

lower center palace - kun bzang rgyal ba

kun bzang rgyal ba rgya mtsho’i sku /

The Oceanic Body of the All-Good King /


The Dividing Rings:

The central realm of ‘Ol Mo Lung Ring is separated off from the outer realms by five rings of mountains and five rings of water. From the innermost to the outermost, these are the First Ring, the Mountains of Gold (ra ba gcig pa gser gyi ri), the Second Ring, Mountains of Silver (ra ba gnyis pa dngul ri), the Third Ring, the Mountains of Conch Shells (ra ba gsum pa dung gi ri), a fourth ring of red mountains which is not labelled in my pictures at least, and the Fifth Ring, the Mountains of Pearl (ra ba lnga pa mu tig gi ri). Each of these is separated by a lake (mtsho) of the same substance. This regions is marked as “Belonging to the King of the Nagas, Takshaka” (klu rgyal ‘jog pos bdag byed). A few pictures:

ringed rivers 01

ringed rivers 02

ringed rivers 03


The Inner Realm:

center sanctum FULL straightened

The inner realm of ‘Ol Mo Lung Ring is surrounded by a ring of great palaces and kingly seats. After another ring of water, it becomes a land of stupas, monasteries, and siddhas, rising up in nine tiers to the great world-mountain at the center. A sKor Ra path leading around the mountain can be seen at the base the first tier. Generally speaking though this was all too high up in the dim rafters of the prayer-hall for me to photograph in detail, so I’ve only got a general description.

The mountain is labeled:

g.yung drung dgu brtsegs ri’i rtse mo / srid pa’i gangs ri dkar po / shel dkar gyi mchod rten brtsegs [?] pa’i tshul /

The Female Peak of the Nine Swastika Tiered Mountain / The White Snow Mountain of the Universe / The Land of the Layered Stupa of White Crystal /

From the top it seems that there is a route leading up to a higher paradise, but this was entirely too high to see clearly or photograph.



The ‘Cham


initial procession panorama

As for the ‘Cham itself, it was pretty cool. I’ve attached a few pictures. The order of dances and processions was as follows:

  1. A parade of all of the members of the ‘Cham and monks from the monastery: carrying tall banners and following white chalk marks left on the courtyard floor.
  2. Zhwa Nag brGyad: “Eight Black Hats”
  3. mChod Pai Lha Mo bZhi“Four Offering Goddesses”
  4. Keng Rus bZhi“Four Skeletons”. These came out to bring a heart on a chopping block to the big torma in the center of the ‘Cham. Later on they walked around among the spectators to collect money.
  5. dMu bDud Tang Srid Pa rGyal Mo gNyis“The dMu bDud Demon and the Queen of Existence”
  6. sTag Ri Rong: “The Tiger of Mountains and Valleys” This was by far the most popular god in the ‘cham. The whole monastery went crazy when stag ri rong came out, screaming out the name and going into byin ‘bebs trances.
  7. rMa rGyal Pom Ra: “The King of the Yellow River, Pom Ra”. (Another name for A Mye rMa Chen)
  8. A bSeSo far as I know, this name doesn’t have a meaning.
  9. dMag dPon gNyis“Two War Chiefs”
  10. Shel Khrab Can“The Crystal-Armored One”
  11. Bya Tang Seng Ge gNyis“The Bird and the Lion”
  12. Once again the keng rus “skeletons”
  13. All of the dancers came down together, along with many of the monks: They continued to dance around the courtyard for almost an hour, as many different things took place – the people crowded together to pass under the torma, fireworks were set off, etc., until all the gods and monks were slowly sent back up into the monastery one by one.

019 the mass dance, a bse

gshin rje yab and dmu bdud 02

022 rma chen pom ra tang stag ri rong tang dmu bdudl

Works Cited:

  • Buffetrille, Katia. “Khyung Mo Monastery (A’mDo) and Its ‘Map’ of ‘Ol Mo Lung Ring.” East and West 59.1-4 (2009): 313-26. Web.
  • Martin, Dan.
    • “Olmolungring: A Holy Place Here and Beyond.” Bon, The Magic Word: The Indigenous Religion of Tibet. Ed. Samten G. Karmay and Jeff Watt. New York, NY: Rubin Museum of Art, 2007. 99-123. Print.
    • “‘Ol-Mo-Lung-Ring, the Original Holy Place.” Sacred Spaces and Powerful Places in Tibetan Culture: A Collection of Essays. By Toni Huber. Dharamsala, H.P.: Library of Tibetan Works and Archives, 1999. 258-97. Print.
  • Snellgrove, David L. The Nine Ways of Bon. London: Oxford U, 1967. Print.
China Gansu Inner Mongolia Photos Qinghai

A Shout-Out for a Friend


Above: Our camel, Subutai the Magnificent, resting on the grasslands in the Qilian Mountains

This is super after-the-fact, but I thought I’d put up a link here to my friend Adam Rouhana’s photo site, Among other things, this has got pictures from a walking trip we made from Lanzhou-ish to Dunhuang in the summer of 2014, partially in company of an extremely truculent camel that we bought in Alashan. The trip went Jingtai 景泰 – dPa Ris / Tianzhu 天祝 – Wuwei 武威 – Minle 民樂 – Zhangye 張掖 – Alashan Right Banner 阿拉善右旗 – Sunan 肅南 – Qilian 祁連 – Yumen 玉門 – Dunhuang 敦煌. I think Adam’s pictures at their best really get to the austere weirdness of rural west China, whether it’s camel-wrangling in the Gobi desert or the surreal lives of the petite-bourgeoisie in boom-town county-level Communist Gansu. Not many people genuinely attempt to see that world in all it’s weird glory, and in that sense I think these photos are actually something quite rare.

There’s also awesome photos from Morocco, the US, Palestine, and other places. So, you should check it out. Here are some of my favorite China pictures. All photos copyright Adam Rouhana, etc.:







Click here for more!

Art China Fortresses Hebei Shanxi Yu, Guangling, and Yangyuan

ILLUSTRATING GROOTAERS, or, the Principle Gods of Rural Xuan-Da and their Iconographies, Part 1

[Note that this post has been broken up into four parts; click to jump to part 2, part 3, and part 4.]



zhangjiakou old pictures

Images from the Xuanhua region in the early 20th century, taken from this blog.

Willem Grootaers (1911-1999) was a reverend father in the Congregatio Immaculati Cordis Mariae, a Belgian Catholic missionary organization that operated in East Asia and especially in northern China and Mongolia. Grootaers was posted to Datong in northern Shanxi and spent over ten years there, surviving through the civil war and the Japanese invasion of the ’30s, until he was finally expelled with the rest of the CICM missionaries at the Communist victory in 1951. He was a deeply intellectual man, whose abiding scientific fascination seems to have been the spread and flow of abstract traits across geographic space – beyond the mapping of village cults discussed below, he produced some of the first serious studies of Chinese and especially Japanese dialectology.

In the mid and late-1940s, with the Chinese civil war raging around him, Grootaers set out to produce a series of studies illustrating “the geographical method applied to folklore”. What this meant essentially was that he walked to every single village in three different counties of war-torn northern China and wrote down all of the temples that he could find in each one, noting any inscriptions or other interesting information he could acquire. The surveys were published in several volumes: (1945) Les Temples Villageois de la Region au Sud de Tat’ong (Chansi Nord): Leurs Inscriptions et Leur Histoire”, (1949) “Temples and the History of Wan-Ch’uan 萬全 (South Chahar): The Geographical Method Applied to Folklore”, (1951) “Rural Temples Around Hsuan-Hua (South Chahar), Their Iconography and History”. He also published a survey of the temples of Xuanhua City, “The Sanctuaries of a North-China City: A Complete Survey of the Cultic Buildings in the City of Hsuan-Hua (Chahar)” and a brilliant study of the movement of cultural traits between Chinese villages, “The Hagiography of the Chinese God Chen-Wu: The Transmission of Rural Traditions in Chahar”. These studies represent the most complete survey we have of the religious and cultural monuments of any chunk of the Chinese countryside, anywhere in China, before the Cultural Revolution wiped it all away.

grootaers young in xuanhua

Grootaers’ survey team in 1948 Xuanhua: Left to right, William A. Grootaers 賀登崧, Li Shiyu 李世瑜, a student named Delinger, and Wang Fushi 王輔世. Image from “Sanctuaries”, p.177.

grootaers old in tokyo

Left to right Wang Fushi 王輔世, Willem Grootaers 賀登崧, and Li Shiyu 李世瑜, reunited in Tokyo 1994. Image from “Sanctuaries”, p.235. Grootaers would pass away in his home in Japan five years later.

The villages that Grootaers saw have been changed totally in the last fifty years, and nearly none of the temples that he recorded now exist. That said, there is one place where the physical and religious culture he described can still be seen. Apparently due to a policy difference in building new villages starting from the 1970s, the three counties of Yu 蔚縣, Guangling 廣靈縣, and Yangyuan 陽原縣 have preserved in a somewhat intact state the original pre-Communist form of Xuan-Da villages. I spent nearly a year in 2014 surveying villages in this area, is adjacent to and more or less culturally coterminous with the areas that Grootaers worked in. Below is a map of our respective survey areas, in which my area is marked in ORANGE and Grootaers’ area is marked in BLUE.

Xuan Da Research Areas Map (small letters)

The temples in my research area were heavily impacted by the Cultural Revolution, and have been further decimated by neglect and looting since. I’m not aware of a single religious statue which survives from before the Chinese takeover, but in some cases frescoes have made it through wholly or partially intact. From these I’ve attempted to gather together here examples which can illustrate what Grootaers was talking about. The value of these is that they exist in archaeological situ – that is, with some minor regional variation, they can be securely tied to the cultural assemblage which Grootaers describes.

It’s also my opinion that Grootaers was wrong or at least confused about a bunch of things, and that later scholars haven’t understood him particularly well. To that end I’d like to offer a brief explanatory guide to both the works of Willem Grootaers and to the Xuan-Da villages and temples he was writing about, in the hopes that future scholars using him as a source will google this and find it useful. Those who aren’t interested can just skip down to the pictures.





Making Sense of Willem Grootaers:

1) The Villages and their History


inside the southern barbican of warm-springs fort

The villages surveyed by Willem Grootaers belonged to a specific cultural assemblage which was common to the whole area north of the Taihang Shan 太行山 and Yan Shan 燕山 in northern Shanxi and Hebei provinces. This area is known as Xuan-Da 宣大, referring to the cities of Xuanhua 宣化 and Datong 大同. The rural cultural assemblage was defined first by the existence of fortress walls around each village, and second by a specific package of temples which were built into and associated with these walls. (Unfortunately, Grootaers doesn’t really differentiate which settlements he visited were walled, although it’s clear from his descriptions that almost all of them were.) For all their emphasis on mapping regional variation, Grootaers’ studies clearly demonstrate that the form of these villages and the main religious figures worshiped within them were broadly common to villages across all of the areas he surveyed.

Although particular elements of this culture have existed all across China from antiquity, the cultural assemblage that Grootaers described essentially came into existence in the late fifteenth and the first half of the sixteenth century. This can be demonstrated simply enough by graphing the dates given in Grootaers’ surveys. Although Grootaers doesn’t provide a full range of all collected dates, for each different cult he gives a series of “earliest dates” found in temple epigraphy. These represent the earliest point from which he was able to establish the existence of a particular cult. The below graph represents all of these “rural” dates from the countryside of Datong, Wanquan, and Xuanhua. This omits dates from sites which Grootaers specifically notes were not villages, as well as dates from the small collection of stone dharani pillars found in Datong county. (If you want a detailed methodology, I suppose you can just email me.)

grootaers dates graph

The graph tells a fairly clear story: The religious landscape that Grootaers was describing came into existence basically out of nowhere in the second half of the fifteenth century and the first half of the sixteenth. The reason for this explosion of dates wasn’t clear to Grootaers; my research in Yu County and its surrounds can provide the answer. To make a long story short, it turns out that the villages themselves came into existence in the same period, or at least were destroyed and rebuilt en masse at this time. The below graph represents dates given on village fort gatehouses from my research area, which as we have seen in the map above lay adjacent to Grootaers areas in both the east and the west. These dates represent the year in which that particular walled village was constructed.

fortress dates from yu county

The reader will see that the two graphs match each other almost exactly. Historical records allow us to elaborate. After the Tumu Incident in 1449, the Ming dynasty gradually lost control of its northern borders and suffered repeated and destructive raids from the Mongols. Government policy during this period was to build walls – the present Great Wall system across northern China largely dates from this period. In the rural areas along the frontier, policy encouraged population consolidation and fortification. Even without this government encouragement, yearly Mongol raids made it imperative for villages to take defensive action. The result was the massive reshaping of the rural landscape over the course of roughly eighty years, and the large scale (re-)creation of villages and temples. The change was so complete that, as these graphs indicate, almost nothing remained from before the start of the fortification period.

Further, the religious culture Grootaers described physically could not have existed before the fifteenth century. These temples were built as structural and geomantic elements of the fortress, such that they could not have pre-existed it. Many of the most important cults were associated with the cardinal axis route of the fort (The Perfected Warrior 真武, the Jade Emperor 玉皇 and Avalokiteshvara 觀音), the fortress gate (Wenchang 文昌 and the Kui-Star 魁星), or crossroads within or without the fort (The God of the Five Ways 五道神). Even in cases where the temple habitually sat at some distance from the fort, the precise direction and distance from the walls could be strictly dictated by custom (The Dragon Kings 龍王). If the fortresses did not exist before 1475, then this culture of temple building could not have either, and the concurrence of the two graphs indicates that the fortresses and the temples came into existence hand in hand. Further, the epigraphy of these village temples strongly supports the idea that villagers were actively thinking about the geomantic relationship between temple and fort during the fifteenth and sixteenth centuries. Out of the chaos of the Mongol wars in the mid-Ming, a new religious system was created in the villages of Xuan-Da. (The one important exception to this story is the existence of large Buddhist monasteries, which were often fortified in their own right and seem to have been spared the brunt of the Mongol attacks. These are the only type of structure in rural Xuan-Da which was sometimes able to persist through this period.)

So to summarize here, Grootaers’ temples belong to a very specific village culture which existed in the region along the Sino-Mongol frontier, characterized by fortified settlements and a specific set of gods worshiped in temples that were structurally connected to the fortresses. This culture came into being as result of Mongol raiding and population consolidation schemes in the fifteenth and sixteenth centuries. I would caution the reader on this basis to treat with extreme caution Grootaers’ attempts to push his geographic analysis back beyond the mid-fifteenth century. It’s also worthwhile to note here that this was a culture that tended to think spatially – these temples were intentionally set in particular relationships to the village walls and streets, often in ways that suggest interpretation. We will return to this point when we talk about the composition of the temple frescoes.

2) The Temples and the Gods


a ruined temple to the three officials outside of a fortress gate

When looking at Grootaers’ numbers it’s crucial to understand that he was not counting temples, he was counting instances of worship of a particular god. Grootaers refers to this as a “cultic unit”. The need for this term arises from the fact that in some cases a god might be worshiped in its own temple, which is known by that name to all of the villagers, and in other cases a god might be found worshiped in a lateral shrine or image in the temple officially devoted to another deity.

The “cultic unit” is actually a very useful and necessary concept, but organizing a survey on this basis also elides some extremely important differentiations. According to this method of counting, all temples are created equal. Massive Buddhist monasteries with recorded histories stretching into the Song or even Tang periods are unceremoniously lumped in with tiny one-room village shrines, simply because the deity worshiped therein is called “Buddha” 佛. Particular deities jump or dive in the ratings because of this. The God of the Five Ways 五道神, a minor deity worshiped at small crossroad shrines, is the second most important god when measured by number of “cultic units”. Meanwhile the Perfected Warrior 真武, who is worshiped on immense towers which loom over the Xuan-Da landscape even today, drops down to the sixth most important god according to this metric.

Nor does this type of tallying explain the spatial or compositional relationships between gods. To give an example, the Jade Emperor  玉皇 takes a place on a tall tower attached to the northern wall of the fort, identical to that of the Perfected Warrior. In rare cases other male gods (for instance the Three Officials 三官) can occupy this tower as well. These towers are nearly always connected with a temple to Avalokiteshvara 觀音 or the Grandmother 奶奶 at the southern end of an axial street which runs through the fort. From this we can deduce that there is a sort of meta-cult: that of a paternal, martial god who watches over the fort from a high northerly tower, axially connected to a maternal, merciful god located at the other end of the village. This particular “meta-cult” or composition of temples, when counted either by total numbers of “cultic units”, by the resources involved in building all of these structures, or by its centrality in organizing the physical space of the village, is undoubtedly the most important expression of Xuan-Da village religion. Nevertheless the reader would have to read Grootaers’ descriptions extremely closely to realize its existence.

All this is not to say that the tallying of “cultic units” isn’t useful and important, but to stress that it cannot be used as the sole or even the main metric of a given cult’s actual importance or function in Xuan-Da society. (Nor, to his credit, did Grootaers ever take it as such.)

With that said, here’s the list of Grootaers’ cults. These are compiled from the combined rural surveys of Wanquan and Xuanhua; at the time of the Datong survey Grootaers had not yet developed the idea of a “cultic unit” and thus does not give such numbers. I’ve attached a brief and extremely unscientific note as to what it seems to me that the cult was “about”; the reader should take these with a grain of salt.

  1. 龍王 The Dragon Kings: 202 units [including the Black Dragon Kings 黑龍王,  the White Dragon Kings 白龍王, the Eight Dragon Kings 八龍王, and the Dragon Kings of Wells and Springs 井泉龍王] – Granting rain, regulation of the weather, successful harvests. 
  2. 五道神 The God of the Five Ways: 197 units – Guarding travelers, guiding the souls of the dead to the underworld, the registry of events in the lives of people. 
  3. 觀音 Avalokiteshvara: 155 units – Maternal compassion, Buddhist miracle working and succor to those in need. 
  4. 馬王 The Horse King: 113 units – Care of horses, mules, and other livestock.
  5. 關公 Lord Guan: 107 units – Association with the Romance of the Three Kingdoms and related folk traditions, embodiment of values such as martial strength, loyalty, etc.
  6. 真武 The Perfected Warrior: 104 units [including the Southern Perfected Warrior 南真武] – Father figure who watches over the fort from a high tower, guardian of the northern frontier against Mongol attack, center of a complex series of hagiographic tales.
  7. 河神 The River God: 72 units – Presumably regulation of irrigation and floods, although I’m not certain. Possibly synonymous with the cult of Erlang 二郎, a popular god in southern China who was known for great irrigation works.
  8. 三官 The Three Officials: 70 units – Bureaucratic Daoist deities who governed Heaven, Earth, and Water.
  9. 佛 The Buddha: 59 units – In some ways the center of two practices. In some cases the Buddha is simply another village “pusa” 菩薩 to whom incense is lit, and in other cases he is worshiped with the whole monastic and philosophical apparatus of the Buddhist religion.
  10. 胡神 The Barbarian God: 59 units – Unclear, but connected in some way with the bringing of rain and to the cult of the Dragon Kings.
  11. 文昌 Wenchang: 32 units – Ostensibly the god of Literature, although in practice this cult seems to be tied to that of the Kui-Star and be a god specifically associated with village gates. 
  12. 奶奶 The Grandmother: 27 units – The granting of children, mothership. This cult has been con-fused with that of Avalokiteshvara and also with the wives of the Jade Emperor, the Daoist deities referred to as the Empresses 娘娘.
  13. 玉皇 The Jade Emperor: 27 units – The Jade Emperor is ostensibly the head of all of the gods in the bureaucracy of heaven. In practice his cult is similar to that of the Perfected Warrior; a martial, paternal figure who sits on a tower overlooking the village.
  14. 魁星 The Kui Star: 20 units – Ostensibly astrology, in practice a sort of demon tied to the worship of Wenchang and set atop fortress gates.
  15. 地藏 Ksitigarbha: 20 units – The judgement of souls after death and the successful passage of the dead through the underworld, associated with the worship of the Ten Yamas 十閻王.
  16. 土地神 The God of the Soil: 19 units – A small god of specific localities.
  17. 大仙 The Great Saint: 17 units – According to Grootaers, the worship of fox spirits. 
  18. 火神 The Fire God: 16 units – Regulation of fire.
  19. 財神 The God of Wealth: 14 units – Accumulation of wealth. 
  20. 普明佛 The General-Enlightenment Buddha: 14 units – A recent (in Grootaers’ time) cult which involved the millenarian belief that a particular man named Li Bin 李賓 was the incarnation of Maitreya. 
  21. 玄壇 The Dark Altar: 13 units – According to Grootaers, protection against hail storms. 
  22. 蟲王 The King of Vermin: 13 units – Protection against vermin.
  23. 靈官 The Spiritual Official: 13 units – Protection against malicious geomantic influences.
  24. 彌勒佛 Maitreya: 11 units – The end of the world, and also being fat and happy, I’m not sure which variant.
  25. 山神 The Mountain Gods: 11 units – Mountains, and also the return of livestock lost there. 
  26. 三郎 The Three Youths / Sanlang: 9 units – Unclear. According to Grootaers, bringing rain, the cult of Yellow-Sheep Mountain 黃羊山, and a connection to secret societies. 
  27. 三皇 The Three Emperors: 7 units – According to Grootaers, this actually represents a confusion of two separate cults, one to the Holy Farmer 神農, Fu Xi 伏羲, and the Yellow Emperor 黃帝, and a second to the Emperor of Heaven 天皇, the Emperor of Earth 地皇, and the Emperor of Men 人皇
  28. 城隍神 The God of Walls and Moats: 7 units – A god of administratively designated cities, who holds a position in the heavenly bureaucracy analogous to that of the country magistrate on earth. Also involved in the judgement of souls after death. 
  29. 灶王 The King of the Hearth: 5 units – A hearth spirit which reports the doings of the household to the Jade Emperor, and therefore must be placated. 
  30. 達摩 Bodhidharma: 5 units – The original patriarch of Zen Buddhism in China, and the center of many legends.
  31. 三教 The Three Teachings: 4 units – Ecumenical respect for the Buddha, Confucius, and Laozi. 
  32. 韋馱 Weituo: 4 units – Guards the Dharma and Buddhist temples.
  33. 瘟神 The God of Disease: 3 units – Protects against disease. 
  34. 五穀 The Five Grains: 3 units – The god of cereals. 
  35. 三清 The Three Purities: 3 units – A trinity of philosophical Daoist deities. 
  36. 三星 The Three Stars: 2 units – The three stars are 福祿壽 “Happiness, Advancement, and Longevity”.
  37. 青苗 Green Shoots: 2 units – Apparently a harvest god.
  38. 眼光 Eye-Glow: 2 units – A Daoist goddess, according to Grootaers, worshiped in rural areas as a goddess of eyesight. 
  39. 黍神 The God of Millet: 1 unit – A harvest god.
  40. 倉官 The Granary Official: 1 unit – The god of granaries, and thus of plentiful food. 
  41. 牛王 The Ox King: 1 unit – Uncertain, possibly has some kind of tantric function due to his appearance and his association with the rite of the Great Feast of Water and Land. It’s not clear to me that this is even the correct name for the god Grootaers describes.
  42. 喜神 The God of Happiness: 1 unit – No idea, other than the name.
  43. 酒神 The God of Wine: 1 unit – No idea, other than the name.
  44. 老子 Laozi: 1 unit – The legendary founder of the Daoist religion. 
  45. 風神 The God of Wind: 1 unit – No idea, other than the name. 
  46. 北嶽 The Northern Peak: 1 unit – The worship of Mount Heng, located nearby in Shanxi province. 
  47. 譚公 Lord Tan: 1 unit – Lord Tan was a local hero who fought the Mongols in the early Ming. 

The reader will appreciate from this list that while a great number of gods might be worshiped, there was a small set which were obligatory and present in nearly every village. (Further, it could be pointed out that some gods were associated with the fort, while other gods were obligatory for the village, whether it consisted of one large fort or several smaller ones in close conurbation.) In any case, broadly speaking the sine qua non of village temples might be listed as the Dragon Kings, the Perfected Warrior, Avalokiteshvara, Lord Guan, and then a small shrine to the God of the Five Ways. Add to this an elective selection of a few more deities depending on means and inclination, and the village’s religious equipment is complete.

3) The Frescoes and their Compositions:


dragon king composition example

As noted above, rural Xuan-Da was a society which often expressed meaning through the composition of space. This was the case with the placement of temples around the village as it was with the interior accoutrement of the shrine rooms. Each temple was generally a square room with an altar on the main wall or stretching around on three walls. The walls would ideally be covered in frescoes, although some villages might have been too poor to afford this. In some cases statues would be set atop the altar with the frescoes as background, in other cases the room was devoid of statues and the images on the walls were the main object of devotion.

No religious statuary has survived the Cultural Revolution in my area, although many examples have been re-built more or less in traditional style. I also don’t have anything in particular to add to Grootaers’ comments on this subject, and so I won’t deal with the statues here. Rather, I will attempt here to define a general typology of Xuan-Da village temple frescoes. Of course there will be many exceptional cases. This typology will also need to be divided into those images found on the rear (ie. central 正) wall of the temple, upon which there is essentially only one composition, and those images found on the two lateral 側 walls of the temple, in which case we may define up to five different compositions.

We will begin with the rear wall of the temple, which faces the viewer directly as he enters:

The Front Court:



All of the rear temple walls contain essentially the same scene: the main god holds court facing the viewer. To his left and right stand civilian 文 and martial 武 attendants, palace ladies, generals, fan-and banner-bearers, and other supernatural flunkies as appropriate. The analogy to an imperial court is made clear by the throne upon which the main god sits and the hu 笏 (note-taking boards) held by the attending figures. In this case the human worshiper takes the spatial role of the supplicant approaching a king or magistrate in a palace or yamen hall. The statues which would have been commonly placed on the central altar and potentially the two lateral altars as well would only have accentuated the simile. 

We may point to one exceptional fresco on a lateral wall of a Temple to the Perfected Warrior 真武廟 which literally depicts this scene. In this case a supplicant has entered into the Palace of the Northern Polestar 北極宮, which is depicted as half-palace, half-fortress, and thrown himself upon the steps of the dais upon which the Perfected Warrior sits. Surrounding them in the court are the same collection of Primordial Generals 元帥 and palace ladies who surround the god in his depictions on the rear temple walls. (It should be noted here that the below fresco is in some sense an exception that proves a rule – it does not fall into any of our set categories below.)



A clear visual simile is being made of the rear wall of the temple as court, the god as emperor, magistrate or judge, and the worshiper as supplicant. We may also point out that the Perfected Warrior’s domain is depicted as being part fortress, and thus synonymous with the village itself. 

With respect to the compositions of the lateral walls flanking the rear one, we may define five different compositions. Of these the final two are somewhat conditional. These are numbered in rough order of commonality:

1) The Procession, or, The Pursuit of the Evil Ones: 


Grootaers uses the latter name, I prefer the former; anyway the two interpretations are not mutually exclusive. In this case the gods are seen to ride out along right-hand wall of the temple and ride back along the left-hand wall, usually accompanied by their many attendants, generals, warriors, mounts, etc. Grootaers states that peasants in his areas identified this to him as 拿八怪 “catching the eight monsters”, and indeed in some cases these monsters are visible being led back in chains. 

I would, however, point out that the object of the expedition is not always the capture of demons. In the case of the Dragon Kings it can be clearly seen that the gods are involved in dispensing water on the world below, and no demons are to be seen. Another important aspect of these drawings which Grootaers does not note (although it can be seen in his photographs) are the small images of common people going about their business beneath the main scene of the gods riding around in the sky. In many cases these scenes culminate with a religious procession arriving at a recursive drawing of the very temple in which the frescoes are housed. 

To me it seems that these drawings might be better understood as depicting a homology between the gods’ motions across the heavens and the motions of their devotees on earth. (The fact that the gods move clockwise around the space of the shrine room inevitably suggests the Tibetan practice of sKor Ra or circumambulation.) I myself have attended temple festivals in Yu County in which the entire village, led by a Daoist priest, processes with banners and instruments through the space of the fort, visiting each temple in turn. This scene is immediately familiar from the frescoes in the temples. The below images are dated to 1709, and depict a religious procession arriving at a Dragon King temple, which is accurately depicted down to the ancient pine tree 松樹 outside. The last scene was witnessed by the author in the village of Northern Gate 北門子村, during a temple fair in the summer of 2014:

temple process 001

temple process 002

temple process 003

temple process 004

For these reasons I’ve chosen to refer to this composition as “The Procession”, over Grootaers term “The Pursuit of the Evil Ones”. The procession in question can refer both to the procession of the gods across the temple walls and to the earthly processions which reflect this celestial motion. This seems like a useful and interesting analytical standpoint from which to understand these images. In any case this is probably the most common composition in Xuan-Da: it is found almost universally in temple to the Dragon Kings 龍王 and less frequently in those of the God of the Five Ways 五道神, the Three Officials 三官, the Grandmother 奶奶, and others. 

2) The Panel Series:

laoye miao lefthand wall full

The Panel Series is as described, a long sequence of “comic book” style images which either tell a story or visually portray a list. Grootaers refers to this as “illustrated biographies”; I would point out that while hagiographic content does predominate with these drawings, it is far from the only topic that can be portrayed this way, and some of these series are not even narratives. In some cases the content of the drawings is identified by small cartouches; in other case the viewer is assumed to know the story already or be able to understand it from the image. Grootaers produced an excellent study of the transmission of this type of paneled hagiography in “The Hagiography of the Chinese God Chen-Wu: The Transmission of Rural Traditions in Chahar”.

I have seen Panel Series in Yu County depicting the following topics: the hagiography of the Perfected Warrior; events in the life of Guan Yu, drawn from the Romance of the Three Kingdoms; the hagiography of Sakyamuni Buddha, drawn from various sources; scenes of Avalokiteshvara working miracles, in all cases that I’ve seen drawn from the Universal Gateway 普門 section of the Lotus Sutra; scenes from the Fifty Three Stations of Sudhana 善財童子五十三參 of the Entering the Dharma Realm 入佛界品 section of the Flower Garland Sutra; scenes of the “Hundred Trades” 百工, depicting various professions. These are the common topics in Yu County at least, and they seem to match well enough what Grootaers describes. I’ve also seen isolated examples of a few others: “The Old Gentleman’s Eighty One Transformations” 老君八十一化圖, a Daoist hagiography of Laozi; one heavily damaged set apparently to do with Wenchang 文昌 the god of literature. Grootaers mentions a few other examples of Panel Series; no doubt before the Cultural Revolution many more topics existed.

3) The Martial Array:


The Martial Array is in some ways simply a continuation of the images from the rear wall “Court” onto the two lateral walls. In all cases that I’ve seen this involves the retinue of some martial god, namely the Perfected Warrior 真武 or the Jade Emperor 玉皇. In this case the “Primordial Generals” 元帥 line both lateral walls, standing in a row facing the viewer, all striking martial poses. The effect is to accentuate the metaphor of the court on the rear wall and cause it to flank the worshiper on either side, with the added suggestion of a military drill. 

It’s also worth noting that even in temples which do not have this type of frescoes, the effect would frequently be replicated by the statues set on altar-tops along all three walls. This was the case in many temples dedicated to the Perfected Warrior, in which the Panel Series hagiographies simply provided background to a Martial Array of statues. So far as I know no intact pre-Revolution examples of this survive in Xuan-Da now, but Grootaers describes the scene, and I have seen analogous ones rebuilt since. 

4) The Judgement of Souls:

dizang full wall 01

This is some sense not truly a composition but more of a topic, which can be portrayed in a variety of ways. Nevertheless the format of some of these images matches none of the other tropes and so it must be included here. In this case the presiding god sits at a table surrounded by lackeys and yamen runners and judges the cases of souls brought before him. Underneath or on the flanking walls, we see scenes of the torture of souls. Grootaers notes that this scene is commonly found in temples dedicated to Ksitigarbha 地藏, and more occasionally in temples dedicated to the God of Walls and Moats 城隍神, the God of the Five Ways 五道神, the Three Officials 三官, and others.

5) The Feast of Water and Land:

taken from 故城寺壁畫, 16-17

[image from “Gucheng Monastery Frescoes” 故城寺壁畫]

Like the previous example, this both a composition and a topic. The argument could be made that the composition is some kind of very great elaboration and combination of the Panel Series and Procession themes. Images of the Feast of Water and Land are also found almost solely in large Buddhist monastic halls, in which the space is large enough that the statues can be set in the center of the room and all four walls be given to frescoes. In this sense it’s not truly a composition native to village shrines. Nevertheless it was an extremely popular theme in north Chinese Buddhist art (I’m aware of four examples in Yu County alone, and there were undoubtedly more before the Cultural Revolution), and the scene would certainly have been familiar to all villagers. The theme dates back to at least the Yuan Dynasty in Xuan-Da, and thus pre-dates the creation of most of the villages that concern us here. 

The image is that of a ritual titled, “The Great Feast of All on Water and Land” 水陸大齋. In this ritual a priest enumerates the names of the entire canon of Buddhist and Daoist gods and ghosts, in order that they descend to receive an offering and in the process be converted by the reading of scriptures. To this end, we see all of the luminaries of the heavens, earth and seas processing clockwise around the interior of the temple, often bearing long banners that identify each group. The cast is markedly textual, in that the members of the procession seem to have been drawn from some canonical list and many of the common deities worshiped in village shrines are not to be found. I have discussed this trope at length in previous posts

A few last comments can be made about the analytical use of this categorization. Gods can be depicted as doing something, (the Procession, the Judgement of Souls); as being associated with a particular story or list (the Panel Series); or as being positioned in a particular way (the Court, and especially its extension, the Martial Array). Generally speaking, the form of the depiction follows the function of the cult. The Dragon Kings do something; each year they either do or do not ride out to dispense rain, and so they are universally depicted via the Procession. Lord Guan is important because of a story, and the values associated with that story; he is universally depicted via the Panel Series narrating this story. The Perfected Warrior is important because of his position, the high tower from which he projects his axial, paternal, martial power; therefore he commonly has a Martial Array depicted either in frescoes or in statues.

And so, having got all this background down on paper, let us turn to the frescoes themselves.

Continued in Part Two, below:

Art China Fortresses Hebei Translation Yu, Guangling, and Yangyuan

ILLUSTRATING GROOTAERS, or, the Principal Gods of Rural Xuan-Da and their Iconographies, Part 2

Continued from Part 1:

1) THE DRAGON KINGS  龍王 – 201 Units:

dragon king temple


Temples to the Dragon Kings are most numerous in all of Xuan-Da, and therefore I have the most and the best preserved examples from this type of temple. In Yu County, these temples universally sit outside the village walls and at some distance. The direction of the temple with respect to the fort was governed by geomantic custom, to the extent that Grootaers was in some areas able to map the boundaries of particular cultural regions as defined by the angle to the fort at which the Dragon King temple sat. The Dragon Kings were aquatic deities, and hence in semi-arid Xuan-Da, their cult was mainly involved with the regulation of rain, and with rituals to that effect.

The images below come from a temple originally built in 1566, and then renovated again in 1709. The frescoes inside date from this later reconstruction, in which the name of the artist is given as Cui Wenxin 崔文新.

“(a) The back wall: It has the image of a female divinity, properly called Shui-mu 水母, “Mother of the Water,” by whose sides stand, first, some male Lung-wang, “Dragon kings,” secondly, the gods of Thunder and Lightning, having human shapes but with the beaked face of a bird of prey; finally, behind the main image is a numerous retinue of heavenly spirits, among them the spirits of Hours, Days, Months, and Years.

Whenever the temple is big enough, the western and eastern thirds of the central wall have frescoes of quite distinct gods, namely, of those who in still larger villages would have a separate sanctuary.”

Wan-Ch’uan, p.228

I’m not certain which the lateral gods are in this spread; they don’t seem to be the ones Grootaers suggests are most common.



Below are the “Four Officials of the Amounts” 四值功曹, who regulate the lengths of hours, days, months, and years, and thus presumably the timely passage of the seasons with their rains and harvests.



Who this colorful fellow is I’m not sure.


“(b) The lateral walls of the Lung-wang temple: The theme of both lateral walls of the Lung-wang temples is “The Pursuit of the Evil Ones”, as described in chapter 2, Wu-tao temples. Only temples too poor for elaborate frescoes have nothing on their walls; 56 Lung-wang temples have those frescoes, more than half of the total number (see map 1). A few peculiarities of this kind of fresco as represented in this type of temple ought to be noted: the “Mother of Waters” does not herself participate in the expedition but is shown seated under the archway of her palace Shui-ching kung 水晶宮, the “Crystal Palace”, surrounded by her female attendants; she either greets the leaving expedition (eastern wall) or welcomes it back (western wall). The members of the expedition are very numerous; they ride on dragons when going out (eastern wall), and on horseback when returning (western wall). The upper parts of the fresco show the heavenly spirits putting in action their instruments: the Thunder spirit rolling his thunder machine, the Rainbow spirit pouring a rainbow out of his urn, and so on. In the foreground of the Crystal Palace, two small figures are standing apart from the others, wishing godspeed to the expedition or making the ceremonial salute with both hands to welcome it back: they are the Wu-tao god and the Earth god (T’u-ti); at their sides stand the tiger and the wolf, which we find in the Wu-tao temple as companions of the two gods. In the upper right-hand corner of the western wall fresco a huge hand is thrust out of the clouds, to which a small figure on horseback presents a yellow scroll. This is the Yu piao 雨表 or “Report to Heaven on the Rain” (see photograph 11).”

Wan-chuan, p.229-230

The eastern lateral wall of the temple, in which the procession rides out from the Crystal Palace. Note the small figures of farmers and traders going about their business beneath the clouds.


A great flood bearing down from the upper left corner of the scene, bearing a red sun. Daoist depictions of the creation of the world tend to bear the image of red suns rising from water, although I’m not sure what the significance is.

sun wave

The Crystal Palace, with the Mother of Waters seeing off the procession.


On the way out, the riders are human and the mounts are dragons.






The central figure of each side of the composition is a palanquin being carried by draconic bearers. The occupant of the palanquin is always invisible behind the drapes. Grootaers states (Hsuan-hua, p.35) that in some of his areas the Mother of Waters remained within the palace, and in other areas she was visible sitting in the palanquin. In this case, it seems that the painter chose a compromise by depicting her in the palace and then hiding the palanquin’s precise occupant.


A thunder-drummer.


Meteorological scalies pour down rain.


Another thunder-cymbalist.


The western lateral wall of the temple, in which the procession returns to the Crystal Palace. On this side of the composition, the dragons have changed into horses.




This is the fellow who leads the palanquin.


The thunder-drummer from the western wall, having packed up his drums after the storm and carrying them back to the Crystal Palace.


The image of the Rain Report being presented to heaven in the upper right corner of the composition is the counterpart to the red sun being borne upon a flood on the opposite wall. This hint of a higher deity elicits some excitement from Grootaers:

“In the upper right hand corner of the western wall, a huge hand is thrust out of the clouds, to which a small figure on horseback presents a yellow scroll. This is the yupiao 雨表 Report to Heaven on the Rain. (fig. 15, top center, fig. 44, top left.) […]

There is a final personage whose importance must be stressed. If we ask ourselves to whom the “Rain Report” is destined, we see it must be a god higher placed than the Dragon Kings themselves. In the general description above, we found how a huge hand comes out of the clouds to receive the report. In fact more often than not, a moon-gate is painted in the upper corner of the western wall; this gate is called nan-t’ien-men 南天門 which could mean as well “Gate of the Southern Heaven” as “Southern Gate of Heaven”; the latter is more probable. We found a couple of times this name inscribed on the gate, as f.i. at DV 95. In some cases, all in the Liu-Ho plain (Dv 87, 95, 98, 122b, 125b, 125d), the hand of the unknown god is thrust through the half open gate, instead of through the clouds. But in most cases, a young being stands in front of the gate (fig. 44, top left), clothed in a checkered dress, with white and black squares; this dress is called ba-gua-i 八卦衣 “the dress of the eight hexagrams.” This youth standing in the southern gate of heaven sometimes sends a rainbow from his hands. As a corresponding feature on the opposite wall, in these cases where the Mother of Water joins the expedition, no Palace of Crystal is shown. Its place is taken by a youthful person (the same as in the Nan-t’ien-men?) from whose hand pour forth the clouds on which the expedition rides out. This is clearly shown in fig. 16. […]

The supreme god who lives behind this door is shown twice on the frescoes of the Lung-wang temples, at Dv 122b and Dv 125d; his head only peeps through the nan-t’ien-men. He wears a round cap with a semispheric ornaments jutting out above the ears; he has a small round beard in the center of the chin and a drooping moustache. No such god was noted elsewhere in our survey.”

– Rural Temples, p. 35-36

I don’t have any pictures of the South Gate of Heaven or the deity of heaven. In Yu County, these figures are represented by the giant hand reaching down from a flood to receive the report. Here they are:



Below are the God of the Soil 土地神 and the God of the Mountains 山神. Grootaers describes these as welcoming the expedition back at the Crystal Palace. In this temple this is not quite so; in fact they stand in the lower, “human” part of the image, behind a recursive image of the temple itself, welcoming the human participants of a religious procession which has arrived with music and offerings. I’ve not put up pictures of that procession here, since Grootaers doesn’t mention it and it’s interesting enough to deserve a full post of its own.


Here is the full set of photos from this temple:

Here’s a second set, from a different temple in a similar style, undated.



2) THE GOD OF THE FIVE WAYS 五道神 – 197 Units:



The relative importance of this cult can be overstated by the numbers of temples devoted to it. Unlike the other cults, all of the temples devoted to the God of the Five Ways are extremely small, being simple shrines set at crossroads, often no more than waist height. Despite its popularity, the precise purpose of this cult seems a little obscure. Grootaers points out two uses of the temples which might explain the iconography:

“In everyday life, the Wu-tao temple is used for the announcements which in a modern state are made at the registrar’s office: births, marriages, and especially deaths are called out loudly by the head of the family in front of the Wu-tao temple.

Another role still may be ascribed to this cult, i.e., that of the protector of the roads and of the people walking on them. This is implied by the name itself of the main god, Wu tao, “Five Roads”; this is even more clearly expressed by the vertical inscriptions referring to the person of the god; we have a great number of such inscriptions; here follows one typical example:

The body covered in golden armour, he wanders on all the roads;
The hand holding the precious sword, he patrols all the directions.”

Wanch’uan, p.238

These observations on the function of this cult are corroborated by Sidney Gamble in Ding County of southern Hebei. In this county, deaths in the family were reported with incense and paper money at the Temple of the Five Ways. Importantly, in the absence of such a temple nearby, the same announcement could just be made at the nearest crossroads. (Gamble, p.387)

With respect to interiors of the temples, it’s noted that the God of the Five Ways (wu-dao) is not always the central god worshiped in the Temples of the Five Ways (wu-dao miao). That said, there is more or less a set iconography:

“(a) The images of the northern wall: The main images (see photographs 4 and 5) are: In the center, the Wu-tao shen, “God of the Five (viz. all) Roads,” a benign, white-faced and black-bearded personage (most often having a three-pronged beard), wearing a military cap and a red coat under an armour, and holding a sword; in the east, the Shan-shen  山神, “God of the Mountain,” a fierce, blue- or black-bearded person, wearing a military cap and a blue-black coat under an armour, holding a sword or a whip; in the west, the T’u-ti shen 土地神, “God of the Earth,” a smiling, white-faced and white-bearded person, dressed like a scholar, with a yellow robe, holding a fan or simply having his hands in his sleeves.”

Wanch’uan, p.234

In total honesty, the inhabitants of the village in question identified the below temple as a being to the Three Officials 三官, not to the Five Ways 五道. However, it’s fairly clear from the iconography that this is a mis-attribution. It’s also one of only two good surviving example that I’m aware of, so it’s worth putting up:



“Finally, in front of the main images, two animals, a tiger and a wolf, are attached with iron chains to a stone pillar or a rock; the tiger is normally found in the east, and the wolf in the west. Their role will become clear when we describe the lateral frescoes.”

Wanch’uan, p.235

Here are the tiger and wolf, with a little statue of what I think is The Great Scholar of the Face 面然大士 or the King of Demons 鬼王.


“(b) The frescoes on the lateral walls: […] A narrow strip of territory in the west [of Wanquan County] has one theme that we may call “The Judgement of Souls”, whereas the whole region has another theme, found uniformly everywhere, viz., “The Pursuit of the Evil Ones”.

“The Judgement of Souls” represents Ch’eng-huang 城隍, the god of the city walls (on the eastern wall), and Yen-wang 閻王, the god of Hell (on the western wall), both sitting in judgement on the souls brought to them. The first step in the judgment is the torture of the souls before Ch’eng-huang – whether as a punishment or for extracting the truth is not clear. Subsequently they pass before Yen-wang, where the souls of the just are led across a bridge over Hell’s fire.

“The Pursuit of the Evil Ones” is a very important theme in the local iconography and with some variations is found in several types of temples (see Introduction, 4.). In the Wu-tao temples, this type of fresco pictures the three main gods of the temple (Wu-tao, Shan-shen, and T’u-ti) on horseback. On the eastern wall, we see them start on a punitive expedition with a great display of wrath; they are led by the Devil and the Judge (the two auxiliary figures in front of the main wall) and accompanied by the Tiger and the Wolf. They pursue evil spirits 捉妖, or, as local people say sometimes, na pa kuai 拿八怪, “catch the eight monsters”. These monsters – there are seldom eight of them on the walls – represent the powers of evil who lead people to sin and hell; one at least has always the head of a chicken, another that of a hare (both symbols of homosexuality), sometimes serpents are amongst them.

The western wall of the temple pictures the returning expedition, the three main gods cantering back in an orderly row, bearing a satisfied expression. The Tiger and Wolf carry between their teeth some of the vanquished foes, the Devil and the Judge lead a chained procession of captives (see photographs 4 and 5).”

Wanch’uan, p.236

The left wall, in which the procession rides out.


The right wall, in which it returns.


Some details of the riders and their attendants:






The capture of the monsters:



I’ve only seen one other surviving set, which has much the same basic arrangement. The front wall shows the God of the Five Ways flanked by his attendants:

The two side walls show the same procession scene.

There’s also this mammary-headed individual, giving us a thumbs up:




A gatehouse and a small temple

guanyin miao front flags

wushisancan panorama

In both Grootaers’ areas and in Yu County, temples to Avalokiteshvara are commonly found at the southern end of the fort, facing in (north), either outside the main gate or on top of it. The cult is in some cases confused with that of the Grandmother 奶奶, another female deity associated with mercy and the bearing of children. Grootaers says of the iconography:

“The Kuan-yin temple has almost everywhere the same interior decoration. On a central throne, made of a lotus flower, sits Kuan-yin, in the feminine shape found often in Buddhist iconography. The image has here always a bonnet made of cloth and a great mantle hanging from the shoulders. The lotus throne sometimes rests on the back of a red horse. It is only in a few cases that the Kuan-yin image is a fresco painted on the central wall.

In about fifteen Kuan-yin temples, irregularly scattered over the area, Kuan-yin has her two associates, Manjushri 文殊 and Samantabhadra 普雨 [sic], whose names are unknown to the people; those images are also sitting on lotus thrones, one resting on top of an elephant (western image), the other on top of a green lion (eastern image). This type of triple image was never found in painting. […]

The ceiling of the Kuan-yin temple disappears sometimes under a canopy of clouds, from which a multitude of heavenly spirits look down; among them and nearest to the main images, we find often the dove-like bird with a rosary in its bill that is a common companion of Kuan-yin.”

Wan-chuan, p.242-3

guanyin miao front

With respect to the lateral walls:

“On both sides of the temple, the 18 Arhat or Lo-han 羅漢 are disposed in two rows, either as images standing on the floor or as painted frescoes. Behind or above the Lo-hans a row of panels on the wall represents miracles wrought by Kuan-yin in favor of the people who invoke her: a drowning man is shown rescued out of the water by the hand of Kuan-yin thrust from the clouds; robbers attacking a traveler are driven off by her, and so on; the number of miracles represented is ordinarily 12 or 8, divided on the two walls.”

Wanch’uan, p.243

I have one partial example of the Eighteen Arhats; since this was found within a large multi-hall monastery 寺 I have put it in that section beneath, since Grootaers mentioned it was also occasionally found in that context in his areas. In both of the two cases where I have paneled images of miracle working, the source of the text is the Universal Gateway 普門品 section of the Lotus Sutra. I’ve already translated the best-surviving and most artistically attractive example in a previous post. The other intact example is translated below. The panels come from the heavily damaged temple building pictured below, in which the space was originally divided into two separate shrines, one to Lord Guan and the other to Avalokiteshvara. The text seems taken more or less at random from the sutra, so I’m not sure of the order.

guanyin miao panorama w guangong

guanyin panel 02

Or if three thousand great-thousand realms of Yakshas and Rakshasas wish to come and beset a person, if he calls out the name of Hears-the-Sounds, these evil demons will not even be able to turn their evil eyes upon that person, let alone do him harm.

guanyin panel 01

[Cartouche illegible; I’m guessing from the picture it’s probably this, or another passage like it:]

Or else if there is a person, whether he is innocent or guilty, if he has been put in chains or in the cangue, if he calls out the name of Hears-the-Sounds, then his restrains will be broken, and he will attain release.

guanyin panel 03

Or if you are floating upon a great water, cry out this name, and you will arrive at a dry place.

guanyin panel 04

[Cartouche illegible, I’m assuming from the picture it’s the following]

If you rely upon the name of Hears-the-Sounds, if you are thrown into a great fire, this fire will not be able to burn you.

guanyin panel 05

雲雷鼓掣電  降雹澍大雨  念彼觀音力  應時得消散
Amids clouds and the rumble of thunder and striking of lighting, as hail falls and a great rain pours down – think only of the power of Hears-The-Sounds, and in response right at that moment the clouds will all vanish and clear away.

guanyin panel 06

或在須彌峰  為人所推墮  念彼觀音力  如日虛空住
Or if upon the peak of Sumeru, you are pushed off by someone and fall – think only of the power of Hears-The-Sounds, and you will travel as the sun across the emptiness. 

guanyin panel 07

蚖蛇及蝮蝎  氣毒煙火然  念彼觀音力  尋聲自迴去
If you meet with vipers and scorpions, with evil vapors, smoke and fire – think only of the power of Hears-The-Sounds, and immediately hearing the sound they will flee back from where they came.

guanyin panel 08

或被惡人逐  墮落金剛山  念彼觀音力  不能損一毛
Or if you are thrust by someone evil, and fall from the Vajra Mountain – think only of the power of Hears-The-Sounds, and you cannot be diminished even by a single hair. 

guanyin panel 09

或遇惡羅剎  毒龍諸鬼等  念彼觀音力  時悉不敢害
Or if you meet an evil yaksha, or a poisonous dragon or any other demon – think only of the power of Hears-The-Sounds, and at all times they will not dare harm you. 

guanyin panel 10

若惡獸圍繞  利牙爪可怖  念彼觀音力  疾走無邊方
If you are encircled by evil beasts, with sharp teeth and claws so fearsome – think only of the power of Hears-The-Sounds, and you will be able to rapidly flee without any obstructions.

guanyin panel 11

[Cartouche illegible, I’m guessing the text might be this section, but maybe not:]

應以童男、童女身得度者,即現童男、童女身而為說法,應以天、龍、夜叉、乾闥婆、阿修羅、迦樓羅、緊那羅、摩 [目+侯] 羅伽、人非人等身得度者,即皆現之而為說法
For those who might be saved by a boy-child or a girl child, she will appear in the body of a boy-child or a girl-child and speak the Law. For those who might be saved by a heavenly being, a dragon, a yaksha, a gandharva, an asura, a garuda, a kinnara, a mahoraga, whether in human or in non-human form, she will appear in that shape and speak the Law.

guanyin panel 12

For those who might be saved by a spirit bearing a golden vajra, she will appear as a spirit bearing a golden vajra and speak the Law.

guanyin panel 13

And at that time, Hears-the-Sounds pitied the four assemblies as well as the beings of heaven, the dragons, people and non people, and received the necklace [from Limitless-Intentions]. He divided it into two, offering one part to the Buddha Sakyamuni, and the other to the Buddha-stupa of manifold treasures. 

guanyin panel 14

我為汝略說  聞名及見身  心念不空過  能滅諸有苦
假使興害意  推落大火坑  念彼觀音力  火坑變成池
I will explain it to you: if you hear her name or see her form,
if you think of her without idleness, then she can extinguish all suffering.
If a person has intent to harm you, and pushes you into a pit of fire,
Think only of the power of Hears-the-Sounds, and the pit of fire will become as a pool of water.

Grootaers says that in his areas, the main fresco subjects within temples to Avalokiteshvara were the figures of the Arhats and this miracle-working text of the Universal Gateway. As mentioned above, the Arhats are not in evidence in the surviving temples of Yu County (although they probably existed before the Cultural Revolution as statues along the altar-top). Further, a third iconographic topic, not mentioned by Grootaers, is common: the Fifty Three Stations of Sudhana 善財童子五十三參 from the “Entering the Buddha Realm” 入佛界品 section of the Gandavyuha Sutra. I’ve seen two or three examples of this, although unfortunately I neglected to completely photograph any of them. In any case the images are fairly repetitive: Sudhana, a small boy dressed in red, worships and studies at the feet of a series of Buddhas and other figures before eventually attaining enlightenment.

wushisancan detail 01

wushisancan detail 02

wushisancan detail 03

wushisancan detail 04




4) THE HORSE KING 馬王 – 113 Units:

mawang miao

qun miao

According to Grootaers, the Horse King (Ma Wang) seems to have been mainly associated with the worship of the Dragon Kings, being venerated as one of the lateral images in that shrines. I’ve not been able to securely identify him in that context, and in my research area only one temple specifically dedicated to the Horse King survives in even partially intact state. Nevertheless, the iconography is quite interesting.

“All representations of the god are unmistakable: he has three faces, one looking in front, the other two, to left and right, all identical, of a deep brown colour and with a fierce expression; his body is covered in armour; he has six arms, two in front and two on each side, brandishing a sword, a bell, a magic seal, or making a magic sign with his fingers. Among his attendants the principal ones are two generals, whose appearance varies a great deal in the different temples; the one at the right often leads a horse. On the whole, there seems to be a great liberty in the decor surrounding the main figure, and the local peasants are unable to give the names of most figures. One will get an idea through the description of the set-up as found in one place, Cz 254: Behind the usual image of the god, a boy and a girl similar to the chin-t’ung 金童 and yu-nv 玉女 found in burial ceremonies; at his left, a devil with a benign white face, at his right an old tribunal official with a black beard; in front, two generals, the one at the left holding a stone, the other at the right leading a horse. […]

Wanch’uan, p.256

I have one example of this frontal iconography, from the same late-Qing shrine as the second God of the Five Ways iconography above.

The two flanking walls show the six-armed Horse King riding out in procession with his generals. Grootaers describes this briefly:

The lateral walls of the few Ma-wang temples sometimes have frescoes. In three of them (Cz 315a, 353c, 278a) the theme is the Pursuit of the Evil Spirits (see ch. 2), with of course Ma-wang as the leader of the heavenly procession. In one instance, a small-sized temple at Dv 163, the frescoes represent two men, one on each wall, leading one horse by the bridle; the man on the western wall carries a headdress in the shape of a lion head; on the eastern wall, the man has an elephant in his headdress. We cannot identify these personages, but similar headdresses were found for the attendants of the God of the Fire (ch.13). At Dv 124, Ma-wang has the God of the Fire and Lao-tzu (ch. 24).

Hsuan-hua, p.57

These are from a different shrine. The right wall:


And the left wall:




The rather Indian or Central Asian appearance of this god (multi-armed, mounted, bearing banners and weapons) prompts Grootaers to speculate on a relationship to the Indic deity Hayagriva, “The Horse-Headed”.


In the first shrine above, there’s also this interesting horse with tendrils of fire coming out of it:

The remainder of the images from the first shrine:



5) LORD GUAN 關公 – 107 Units:

guan gong miao outside 03

axial guan gong miao

guan gong miao outside

Lord Guan is the late Han-dynasty general, Guan Yu 關羽 or Guan Di 關帝 “Emperor Guan”, hero of the “Romance of the Three Kingdoms” 三國演義. He’s also called “The Grandfather” 老爺, presumably in complement to the “Grandmother” 奶奶, another common village god. Grootaers says of the iconography:

“We found only three temples where the central image of Kuan-ti was painted on the wall. All the others have statues. These now may be of two different types: the military type, showing Kuan-ti as an army general, with full armour, either standing (seven cases) or on horseback (five cases); another type is that of Kuan-ti as a literate, sitting on a throne (29 cases). There is the possibility of a combination of the two types, one type of statue standing in front of the other; Dv 97 for instance has Kuan-ti as a literate, in front of which he is represented on horseback, and more to the front yet another smaller statue as a literate. The Kuan-ti on horseback has a special name in the spoken language, Le-ma Kuan-ti 勒馬關帝 ‘Guan-ti restraining a horse.'” […]

Hsuan-hua, p.64

Below is the only temple I’m aware of where back-wall frescoes of Lord Guan survive. Grootaers identifies the standard list of generals flanking the main image as Zhou Cang 周倉, Liu Hua 劉化, Wang Fu 王甫, Guan Ping 關平, and in one case Zhao Lei 趙累.

laoye miao front

guan gong miao interior panorama

The latter image comes from a different temple than the one examined below; although it’s heavily damaged the reader may get an idea of what the layout might have looked like in Grootaers’ time.

With respect to the flanking walls:

“Another important part of the set-up of the Kuan-ti temples are the frescoes of the lateral walls. There are a few themes less often found, as the horses of Kuan-ti (see above) or the two genii of wealth Tz’u-t’ung 梓潼 and Ts’ai-shen 財神 (ch.11) at Cz 279a. But most of the Kuan-ti temples have their lateral walls covered with a great number of panels depicting incidents from Kuan-ti’s life, without a doubt according to the same novel San-kuo-chih yen-i […]. We found these panels in 26 temples; in a few cases there were as many as 72 (Dv 87) or 48 (Dv 176a) of them. But in no case did we find titles along each panel as is the case for the Chen-wu temples. We had therefore no possibility of taking down a detailed description of each panel.”

Hsuan-hua, p.66

In Yu County, it was in fact quite common for these paneled stories to have titles. The right wall:

Frescoes narrating the Romance of the Three Kingdoms (1)

The left wall:

laoye miao lefthand wall full

I’ve translated the upper three rows of the left wall. The order is in bottom-to-top boustrophedon:

laoye 3.6

大戰周公瑾 – A Great Battle with Zhou Gongjin

laoye 3.5

鎮守荊州府 – Taking Control of the Government of Ji zhou

laoye 3.4

單刀驚魯肅 – Startling Lu Su with a Single Blade

laoye 3.3

胡班投降 – Hu Ban Makes his Surrender

laoye 3.2

興帥伐曹操 – Raising an Army to Campaign Against Cao Cao

laoye 3.1

用計取襄陽 – Taking Xiangyang by Stratagem 

laoye 2.1

立斬夏侯存 – Immediately Beheading Xia Houcun

laoye 2.2

進兵攻樊城 – The Army Advances to Attack the City of Fan

laoye 2.3

大戰龐令名 – A Great Battle with Pang Lingming

laoye 2.4

決水淹七軍 – Damming the Waters to Drown Seven Armies

laoye 2.5

于禁乞性命 – Yu Jin Pleads for his Life

laoye 2.6

箭射成何將 – Shooting General Cheng He with an Arrow

laoye 1.6

周倉擒龐德 – Zhou Cang Seizes Pang De

laoye 1.5

怒斬龐令名 – Beheading Pang Lingming in Anger

laoye 1.4

用水淹樊城 – Using Water to Submerge the City of Fan

laoye 1.3

大戰徐公明 – A Great Battle with Xu Gongming

laoye 1.2

活捉呂子明 – Capturing Lu Ziming Alive

laoye 1.1

 玉泉山顯聖 – A Saint Manifests Upon Jade-Springs Mountain

Here’s the full set from this temple:

Another late-Qing set from a different temple, now partially collapsed. (There’s a stele text saying when these were made, but I can’t be bothered to go hunting for it right now.)

Continued in Part Three, below: