Nomadic affinity with oracle bone divination

« previous post | next post »

Anyone who has studied the history of writing in China is aware that the earliest manifestation of the Sinitic script dates to around the 13th century BC, under the Shang Dynasty (ca. 1600- BC).  It is referred to as jiǎgǔwén 甲骨文 ("oracle bone writing") and was used primarily (almost exclusively) for the purpose of divination.  The most ideal bones for this purpose were ox scapulae, since they were broad and flat, and had other suitable properties, which I shall describe below.

The bones used for divination were prepared by cleaning and then having indentations drilled into their surface, but not all the way through.  A hot poker was applied to the declivities, causing cracks to radiate from the heated focal point.  This cracking was called bǔ卜, a pictograph of the lines that form in a heat-stressed bone.

Middle Sinitic: /puk̚/

(BaxterSagart): /*pˤok/
(Zhengzhang): /*poːɡ/


Cantonese (Jyutping): buk1

Hakka (Sixian, PFS): puk

Min Nan (POJ): pok / poh


The etymology of the word itself is likely onomatopoeic for the sound of the popping, bursting, cracking of the bone.  Later, bǔ卜 came to be used as a general term for divination.

Since this type of divination was a combination of scapulimancy and pyromancy, we may refer to it as pyromantic scapulimancy or pyro-scapulimancy.

What sort of people are likely to have developed this practice and utilized it for divining the outcome of impending events?  Nomads with herded animals whose flesh they roasted on campfires.

Incidentally, the Shang and their Zhou successors found a good substitute for ox scapulae when and where the latter were not readily available in the form of turtle plastra, large numbers of which remain in the archeological records and are written about in such texts as the Zhuang Zi 莊子 (Wandering on the Way [late 4th-early 3rd c. BC]):

Master Chuang was fishing in the P'u River. The king of Ch'u dispatched two high-ranking officials to go before him with this message: "I wish to encumber you with the administration of my realm."

Without turning around, Master Chuang just kept holding on to his fishing rod and said, "I have heard that in Ch'u* there is a sacred tortoise that has already been dead for three thousand years. The king stores it in his ancestral temple inside of a hamper wrapped with cloth. Do you think this tortoise would rather be dead and have its bones preserved as objects of veneration, or be alive and dragging its tail through the mud?"

"It would rather be alive and dragging its tail through the mud," said the two officials.

"Begone!" said Master Chuang. "I'd rather be dragging my tail in the mud."

[Victor H. Mair, tr., Wandering on the Way: Early Taoist Tales and Parables of Chuang Tzu (New York:  Bantam, 1994), p. 164.]

[*Note that Ch'u is a southern kingdom.]

As for the ratio of cattle scapulae to turtle plastra, Keightley (Sources of Shang History) did a rough calculation and found that turtle plastra are slightly more numerous than cattle scapulae, but he did not take into consideration regional and temporal differences. His findings are based on Anyang data, but it is noteworthy that, according to Ken-ichi Takashima, in Zhouyuan, Shaanxi all the oracle bones found there are turtle plastra. The same is true in Daxinzhuang, Shandong, but in Zhengzhou, Henan they are mixed — with some scapulae and some plastra.  Certain authors think that they were brought in from Anyang, but according to Takashima's analysis of graphs (Ken-ichi Takashima, "Literacy to the South and the East of Anyang in Shang China:  Zhengzhou and Daxinzhuang", in Li Feng and David Prager Branner, eds., Writing & Literacy in Early China:  Studies from the Columbia Early China Seminar [Seattle:  University of Washington Press, 2011], pp. 141–172) they were local. The problem is that in Zhengzhou some are “practice” inscriptions made on a sheep (?) rib. Another problem is that they discovered these bones when they were bulldozing to repair the Yellow River bank. They were brought to Chen Mengjia (1911-1966) in Beijing for examination, and due to the poor management (guǎnlǐ 管理) they–only a few pieces—got lost. But in the late 1990s archeologists found a few more pieces in the Zhengzhou castle remains. Takashima says that he personally examined them (now kept at the Henan Provincial Institute of Archeology) with his own eye.

It would appear from the available evidence that scapulae are earlier than plastra in the pre-Anyang period.  Going back to the Neolithic, ca. 3500 BC, osteomancy seems to have started in the northwest, and then gradually spread southeasterly to the rest of the East Asian Heartland (EAH). In the earliest periods, there was no plastromancy, but it began to appear roughly around 2000 BC.

The observations in the above paragraph are based on my own investigations and on information supplied to me by colleagues who are knowledgeable in oracle bone studies.  Further research in the detailed archeological record is necessary to confirm these working hypotheses, but overall they complement our impression that Anyang constitutes the main vector of Shang interaction with the steppe peoples to the north and northwest, and that it was along this trajectory that pyro-scapulimancy came to the predecessors of the Shang, and from the Shang it spread to other parts of the EAH, developing into pyro-plastromancy as it went through time and space.  This would have been a natural consequence of the lack of adequate numbers of herded animals, especially large ones such as cattle, that is characteristic of steppe nomadism.

Following the work of Roderick Campbell and colleagues, it is evident that at Anyang itself there were plentiful cattle (as well as pig and deer) to supply the bone-working industry located there:  Roderick B. Campbell, Zhipeng Li, Yuling He, and Yuan Jing, "Consumption, exchange and production at the Great Settlement Shang:  bone-working at Tiesanlu, Anyang", Antiquity, vol. 85, issue 330 (December 2011), 1279-1297.

Takashima also observes that osteomancy was practiced by hunters in the north(east) (for the locations, see Rowan Flad’s paper cited in the "Recommended readings").  Algonquian hunters in North America also practiced pyro-osteomancy until not so long ago.

To give a taste of what pyromantic scapulimancy must have been like on the steppe for millennia, here is what William of Rubruck (Willem van Ruysbroeck, Guillaume de Rubrouck, or Willielmus de Rubruquis; ca. 1215-ca. 1295), a Belgian Franciscan friar, witnessed in 1253 during his own private mission to convert the Mongols to Christianity:

    After he arrived at Mongke's camp, William observed a servant carrying charred shoulder-blade bones of sheep out of the khan's tent.  He later learned that the Khan "does nothing in the world without first consulting these bones; consequently he does not allow anyone to enter his dwelling until he has consulted them."  The khan observed the traditional religious practices of the Mongols, including each day sending bones to be burned.  "When the bones, therefore, have been burned until they are black they are brought back to the Khan and he thereupon examines them to see if with the heat of the fire they have split lengthwise in a straight line.  If they have, the way is clear for him to act; if, however, the bones have cracked horizontally or round bits have shot out, then he does not do it."

[From Valerie Hansen, The Silk Road:  A New History with Documents (New York, Oxford:  Oxford University Press, 2017), p. 401.]

These observances are at the imperial level of the Mongols, but one can well imagine how smaller groups or even private individuals on the steppes might themselves more informally (or perhaps we should say, less officially) wish to consult the cracks in the scorched bones left over from their meat meals cooked upon an open fire.

The Shang rulers, on the other hand, made oracle bone divination into something still more grand and ritualistic than the Mongols.  In this respect, we may say that the Shang perfected a more rustic, rough and ready steppe practice, just as the medieval denizens of the EAH domesticated and industrialized tea usage when they turned a medicinal, fermentable caffeine bearing plant leaf much favored by "southern barbarins" (Nán Mán 南蠻) into an aestheticized, steeped beverage.  The Shang did not just throw away the oracle bones after consulting them, but kept them in archives, which is why we have over a hundred thousand oracle bones from the main site where they were discovered, Anyang.

By way of conclusion, let us connect the Shang royal divinations (13th c. BC) with the Mongol imperial prognostications (13th c. AD) through space and time with Tibetan charred bone augury.  The latter I was stunned to see with my own eyes as a humble exhibit in the Ürümchi Regional Museum.  Totally unexpected, I beheld a medieval scorched sheep scapula with Tibetan writing on it.


Selected readings

  • "Paleographers, riches await you!" (10/28/16)
  • "The concept of word in Sinitic" (10/3/18)
  • "Oracle" (3/21/14)
  • Rowan K. Flad, "Divination and Power: A Multiregional View of the Development of Oracle Bone Divination in Early China", Current Anthropology, 49.3 (June 2008), 403-437.
  • David N. Keightley, Sources of Shang History: The Oracle-Bone Inscriptions of Bronze Age China  (Berkeley, Los Angeles:  University of California Press, 1978).
  • Ken-ichi Takashima,  A Little Primer of Chinese Oracle-Bone Inscriptions with Some Exercises; 1., Aufl. (Harrassowitz Verlag [April 1, 2015]), 188 pages; 2nd rev. ed. (Harrassowitz; 2nd Revised ed. edition (February 20, 2019), 202 pages.
    • This Little Primer introduces students of classical Chinese to the earliest extant body of Chinese texts dating from about the 13th to the 10th centuries BC. These texts are known as Oracle-Bone Inscriptions and relate to any matter that was deemed sufficiently important to require consultation with the ancestors and deities of the Shang aristocracy. Indispensable to the study of the history of Chinese religion, politics, agriculture, the calendar system, hunting, warfare, medicine, sacrificial and ritual practices, and other matters of life in China's first historical dynasty, these more than 130,000 pieces of inscribed turtle plastrons and bovine scapulas, though mostly fragmented, comprise more text in terms of number of characters than the combined transmitted traditional pre-Qin classical Chinese texts. The material is presented in three forms: normalized transcriptions of the texts into modern standard Chinese script, translations into English, and ink-squeezes or rubbings of the original texts. There is also a detailed linguistic and philological explanation of the text, plus an annotation, and commentary on the cultural and historical background of the material. No special background in analyzing grammar and syntax will be required to understand most, if not all, of the materials presented in this Little Primer.
  • Peter Hessler, Oracle Bones: A Journey Between China's Past and Present (alternative subtitle: A Journey Through Time in China) (New York:  HarperCollins, 2006) — a central figure in this book is Chen Mengjia, who was the most renowned scholar in oracle bone studies in China during the 50s and 60s until he committed suicide under political pressure; mentioned above in the o.p.
  • Victor H. Mair and Erling Hoh, The True History of Tea (London: Thames and Hudson, 2009).

[Thanks to Ken-ichi Takashima, Edward Shaughnessy, Jonathan Skaff, and William Boltz]


  1. David Marjanović said,

    November 25, 2020 @ 11:42 am

    It is referred to as jiǎgǔwén 甲骨文 ("oracle bone writing") and was used primarily (almost exclusively) for the purpose of divination.

    Well, if it was also used for any other purpose, it was probably written on less durable materials than bone, so we wouldn't know, right?

  2. Cervantes said,

    November 25, 2020 @ 1:16 pm

    I have a vague memory that this is connected with the origins of the hexagrams of I Ching. Is that correct, or in any event possible?

  3. Chris Button said,

    November 25, 2020 @ 1:45 pm

    @ David Marjanović

    The name in Chinese is based solely on the materials it was found inscribed on.

    甲 means shell/plastron, 骨 means bone.

    The English rendition "oracle bone" doesn't quite capture that.

  4. Victor Mair said,

    November 25, 2020 @ 1:53 pm

    From Axel Schuessler:

    Divination by shoulder blades was still practiced by Mongols in the Tsaidam-Golmo[d] area around 1906, see Albert Tafel “Meine Tibetreise” (Deutsche Verlagsgesellschaft Stuttgart 1914), p. 284 (also p. 287):

    Tafel’s Golmo guide was unarmed, carried not even a short sword. “Instead, every night he tossed a few shoulder blades from sheep into the camp fire in order to find out the safest trail for traveling [the next day], and figured out for himself the future from the cracks that were caused by the heat. Thus he learned which trail we had to choose in order not to be attacked [by robbers]. Every night [at the end of that day’s trip] he was pleased with himself because his calculation had turned out to be correct, we had not encountered robbers….” (p. 284).

    Another Mongol guide followed a similar practice a few days later: “At every rest stop [the Mongols] prophesied with the help of sheeps’ scapulae, nine dice, and rosaries what would happen in the next few hours….” (p. 287).

  5. Christopher Nugent said,

    November 25, 2020 @ 2:24 pm

    @ David Marjanović

    This is matter of some controversy. Two good places to start are:

    Bagley, “Anyang Writing and the Origin of the Chinese Writing System,” in The First Writing: Script Invention as History and Process, 191–249.

    Smith, “The Evidence for Scribal Training at Anyang,” in Writing and Literacy in Early China: Studies from the Columbia Early China Seminar, 173–205.

  6. Dan Martin said,

    November 25, 2020 @ 2:43 pm

    Mike Walter wrote a few papers on the practice, not only in Tibet, but in other neighboring countries. I know that where Indians saw Jambu Island (our world continent) in the shape of a chariot, Tibetans saw it in the shape of a sog-kha ('shoulderbone'). The Yarlung Valley, considered the imperial heartland, and site of the royal tombs, really does resemble the shape of a shoulderbone, and Tibetan sources speak of Yarlung Sogkha, thereforre. So it's not just about divination, but also larger geo conceptions. I'm sure this one you can download for free online: Michael Walter, ‘Scapula Cosmography and Divination in Tibet,’ Kailash, vol. 18, nos. 3-4 (1996), pp. 107-114. The other I know about is his ‘Areal Religious Phenomena in Tibet & Central Eurasia,’ contained in: Historical Themes and Current Change in Central and Inner Asia, ed., Michael Gervers & Wayne Schlepp (Toronto Studies in Central & Inner Asia, no. 3) Joint Centre for Asia Pacific Studies (Toronto 1998), pp. 122-133. It may be a little more difficult to locate.

  7. Chris Button said,

    November 25, 2020 @ 10:12 pm

    @ David Marjanović and Christopher Nugent

    Slightly tangential, but Takashima & Yue's "Evidence of Possible Dialect Mixture in Oracle Bone Inscriptions" (2000) provides interesting evidence that the language represented in the oracle-bones was a living language rather than an artificial divinatory language.

  8. Mary Kuhner said,

    November 25, 2020 @ 10:14 pm

    My spouse and I were finishing up the tail end of a roleplaying game session over dinner at a Tibetian restaurant in the US when a staff member approached us and diffidently asked me if I was using the dice for divination. I explained that I knew how to read cards but not dice, and she thanked me and went back to work.

  9. Jonathan Silk said,

    November 27, 2020 @ 4:39 am

    I'm curious; ignorant of the field, I had never encountered the name Ken-ichi Takashima. But I immediately wondered at the unusual romanization. When I checked, I find that this is how the author treats his name but I wonder if you or readers know of other cases. We would, following "the rules," rather write Ken'ichi, although as far as I know, many Japanese just use Kenichi. My own dear teacher wrote his name Gadjin Nagao, although strictly it should have been Gajin, although at least for me it is somewhere between very hard and impossible to not insert that -d- sound. Thinking about it, I suppose I also pronounce *tro-djan for Trojan. (NB: I know nothing much of phonology, obviously!)

  10. Lasius said,

    November 27, 2020 @ 4:41 am

    Since domestic cattle only entered East Asia in the late third millenium BC, did they have to use scapulae of wild aurochsen if they already practiced this form of divination by 3500 BC?

  11. Philip Taylor said,

    November 27, 2020 @ 9:43 am

    Jonathan — "I also pronounce *tro-djan for Trojan" — "dj" as in /dj/ or as in "adjudicate" ? I.e., /ˈtrəʊ djən/ or /ˈtrəʊ dʒən/ ?

  12. Wolfgang Behr said,

    November 30, 2020 @ 5:36 am

    Joachim Gentz has a research project running at the University of Edinburgh, which tries to reconstruct the bone cracking process in some detail: "Analysis and reconstruction of the production of controlled divinatory cracks in Shang Dynasty oracle bones from the 13th and 12th century BCE held in the National Museum of Scotland" (see, go to "Research" for an abstract). Joachim is cooperating not only with the museum but also with a local butcher and a buffalo farm in North Scotland for some hands-on "experimental archaeology". There was a BBC documentary for the "History of magic" exhibition at the BL in 2018, where you could see him cracking a scapula, I believe.

RSS feed for comments on this post