The Earliest Evidence of Bonpo Rituals?

A record of a ritual to a local deity, found in Miran (IOL Tib J 255)

The four-sided, pointed stick pictured above was found in the desert fortress of Miran by Aurel Stein in 1907. Along with the most of Stein’s acquisitions, it was then sent to London, where it was placed in the India Office Library, to be ignored by almost everyone except the librarian FW Thomas, who attempted to read the Tibetan writing on all of its four sides, and published his translation in his Tibetan Literary Texts and Documents in the 1950s.

The stick is all that remains of a ritual performance, which is recorded in Tibetan writing on each of its four sides. The writing tells us that this was a ritual for a local deity (yul lha) carried out by a team of ritual specialists including a bon po. Like the other documents from Miran, it dates from the time when the fort was an outpost of the Tibetan empire, which began to fall apart in the middle of the 9th century. This stick probably dates from a few years (perhaps a few decades) before that collapse.

So what we have seems to be a record of the actual performance of a ritual dating back to the time of the Tibetan empire. I think this must be by far the earliest reliable documentary evidence of the actual ritual activities of people identifying themselves as bon po.

Why is this interesting? There has been a debate going on in Tibetological circles for some time about the early non-Buddhist Tibetan religion, which was probably not known as Bon but was practised by ritualists known as bon po. The relationship between this early complex of ritual practices and the religion known as Bonpo (now accepted as one of the schools of Tibetan Buddhism) is complicated. Modern scholarship has cast doubt on the accounts of the Bonpo tradition about its own history, transmitted in texts which generally date from after the 10th century. Those attempting to understand the nature of the early non-Buddhist Tibetan religion have often turned to the Dunhuang manuscripts as an alternative source of evidence (I wrote more about this a while ago in this post).

*  *  *

A fresco from one of the stupas near the Miran fort, predating the Tibetan occupation by several centuries

There are quite a few manuscripts from Dunhuang about non-Buddhist ritual practices like funerals, divination and healing. I could write much more about them, but to show why the wooden dockets from Miran are so interesting, I’ll just say why the Dunhuang sources are somewhat unsatisfactory. First, as the Dunhuang cave seems to have been sealed in the early 11th century, these manuscripts may not date from much earlier than the transmitted texts of the Bonpo tradition, weakening claims by scholars that they are the more authentic sources. Second, the Dunhuang manuscripts are literary sources (though probably derived from oral traditions), mostly narratives or paradigms which would have presumably have supported ritual practice, but are not evidence for what people were actually doing.

On the other hand, the dockets from Miran can be dated, with some confidence, to the 9th century, and probably to Tibet’s imperial era. As records of actual ritual events, they let us know that this was not a merely a literary tradition, but a living practice. And unlike the literary texts, they are firmly local, telling us who the officiants of the ritual were, why the ritual was carried out, and the local deities to whom the ritual was addressed. The offer us the chance to see the activities of the bon po (as well as other ritual officiants like gshen), “on the ground.”

The remains of the Tibetan fort at Miran (Tib. Nob cung)

So, what kinds of rituals were being performed for the Tibetan military officials of the Miran fort? Unsurprisingly, there are quite a few records of funerals (see for example IOL Tib N 330). It is difficult to work out exactly what happened in the course of these rituals (despite Thomas’s valiant attempts at translation). It looks to me like the main aim of the funeral was to guide the “mental principle” (thugs) of the deceased to the right level (gral). One of the practices accompanying this seems to be a libation offering: most of the funeral records specify a precise number of spoonfuls (yams) of a sacred beverage (skyems) to be offered.  Reference to a “beverage offering” (skyems gsol) in the Old Tibetan Annals suggests that some form of this practice goes back to the 7th century or earlier.

But it is only in another kind of ritual, the supplication of local deities, that we find the four-sided pointed sticks like the one at the top of this post. I don’t know the reason for the stick’s being carved into this shape, and any ideas would be welcomed (could it represent an arrow, for example?). The ritual supplications are directed to a variety of deities, including the local deities (yul lha), and minor spirits like sman and g.yang. In these rituals, the main officiant is called lha bon po, that term lha presumably indicating his special role towards deities. The other officiant is the gshen, and it is interesting to see that it was the norm, rather than the exception, for these two types of ritualist to work together.

*  *  *

There’s much more to be said about these ritual dockets, but I’ll conclude with a thought about the people who produced them. Clearly they were the soldiers and officials manning the outposts of the Tibetan empire in Central Asia. In two cases, we have the names of the people who either officiated or commissioned the ritual, and they both have the high official rank of blon. So it seems that well after the official adoption of Buddhism as the Tibetan state religion, the practice of non-Buddhist rituals was common (perhaps even standard) among the Tibetan ruling class. In a sense, this shouldn’t surprise us. Perhaps more surprising is that one of the dockets (IOL Tib N 279) mentions the presence of 21 Buddhist monks (dge ‘dun) at a funeral ritual.  It is difficult to say from this source whether these monks were carrying out the role normally performed by the bon po or were just in attendance at a (non-Buddhist) funeral for a deceased member of their sangha. Either scenario is intriguing.

*  *  *


I have written an article on these and other early sources on non-Buddhist Tibetan ritual practice, provisionally titled “The Naming of Tibetan Religion: Bon and Chos in the Imperial Period,”which  will come out at some point in the near-ish future, and I’ll post a notification when it does.

FW Thomas’s translations can be found in the section “Government and Social Conditions” of his Tibetan Literary Texts and Documents concerning Chinese Turkestan, Part II: Documents, Royal Asiatic Society, London, 1951.

For some interesting discussion of the term bon in the early period, and the dates of the Dunhuang sources, see Henk Blezer, “Ston pa gshen rab: Six Marriages and Many More Funerals.” Revue d’Études Tibétaines 15 (2008): 421–479. PDF available right here.

The reference to skyems gsol in the Old Tibetan Annals is in the year 682-3. See the translation at p.94 of Brandon Dotson’s The Old Tibetan Annals: An Annotated Translation of Tibet’s First History. Vienna: Verlag der Österreichischen Akademie der Wissenschaften, 2009. The Tibetan text of the Annals and many early ritual texts from Dunhuang are freely available over at OTDO.

I also recommend having a look at Vincent Bellezza’s translation of a narrative on the “golden libation” (gser skems) recently found in the Gathang stupa.

Finally, for all other matters bibliographic see Dan Martin’s extensive online Bon bibliography.

*  *  *

Afterthought on the date of the Miran documents

While we know that Dunhuang was swept away from the Tibetans in the year 848, the exact date of the fall of Miran is unknown. In The Tibetan Empire in Central Asia (p.172), Christopher Beckwith states that Miran remained in Tibetan hands into the 850s, but then “passed out of the historian’s ken”. It seems likely to me that this passing out of history was accompanied by the swift decline of the fort as a functional part of the Tibetan empire. Without the imperial support network that kept these outposts going (of which we know quite a lot from many of the other wooden documents from Miran), it is not likely that they could have continued to function for very long. Their Tibetan inhabitants would then have returned to Tibet proper, or to the nearest cities with large Tibetophone populations, like Liangzhou. In their language and palaeography, the ritual dockets belong among the military documents that form the bulk of the Miran manuscripts, and thus I think should be considered a part of the culture of imperial Tibet, even if their exact terminus ad quem is not known.

14 thoughts on “The Earliest Evidence of Bonpo Rituals?

  1. they look like the giant ‘toothpicks’, used in empowerment divinations.
    are the names indicating sponsorship of a big ritual with a feast? sonam sponsored this?/ in honour of dorje?

  2. In India the dice for games and divination were four-sided, not six-sided. That’s why there are four world ages (yugas), they say.

  3. Dear D,

    Yes, and I think I’ve seen a Central Asian version of those four-sided dice as well. And we do have Dunhuang documents on divination which line up the results of the roll of two dice, each one as rows of dots (not stacked on top of each other as in our six-sided dice). There’s a very similar method found in a Turkic manuscript as well. Anyway, despite the similarity in shape, I don’t see that the ritual sticks having a similar role to dice; the text just continues from one side to the next, in a continuous flow. The previous comment on the modern use of wooden sticks inscribed with the names of donors or recipients of the ritual might be closer to what we have here, as there are a couple of sticks with personal names on them.


  4. Is there a search-category that will return only woodslips at the IDP database, or one that will return (for example) only inscribed wood that is four-sided? I don’t find anything like this in the “advanced search” . The category “manuscript” seems way too broad, given that so much of what is in the database is covered by that term (not distinguishing the types of material that are inscribed upon, and not distinguishing the various means of inscription like ink, chisel, brush, fingernails…).

  5. If you click on “Advanced Search” and then go to “Form” you can select “woodslip” (as well as “wood bracket”, “wood shavings” etc.) you can also select Tibetan from the Languages/Scripts search box. Unfortunately there isn’t a separate category for the four-sided sticks. By the way, an excellent introduction to the woodslips is Tsuguhito Takeuchi’s article “The Tibetan Military System and Its Activities from Khotan to Lop Nor” in The Silk Road: Trade, Travel, War and Faith (now out of print I think but it’s on Google Books, and the whole of Takeuchi’s article is there).


  6. Or just search “IOL Tib N” and browse the slips that way.

    Some of the pointy slips, like IOL Tib N 1443, are notched.

    One thinks of the cutting of the record into two parts (khram gnyis bgyis in PT 1042), one for the deceased and one for the living. (Actually, one for the skyibs lug and one for the chief assistant in the funeral, or something similar.) Perhaps the record kept by the living was invested with power or had a talismanic effect. It is interesting to go back to Rona-Tas’ article on the tally stick in Tibetan wrathful iconography in light of these sticks from Miran.

  7. Thanks for the interesting post! Just a comment: the two scenarios you suggest at the end seem to be based on the implicit premise that the funeral mentioned on IOL Tib N 279 is a “bon po” one, and thus a “non-Buddhist” one. I have not tried to locate and decipher the text, and am not well versed in these early historical materials, but could it not be also that the local perceptions that inform that text were not characterized by this kind of exclusivist view of religious affiliation?

  8. Dear Nicolas,

    Thanks for your comment. It’s hard to get the nuances right in a brief blog post! A likely scenario is that Tibetan rituals with pre-Buddhist roots — performed by people inhabiting the ritual roles of bon po, sku gshen and the like — continued alongside Buddhism during and after the Tibetan imperial period. The tensions of this situation can be seen in polemical texts like the one I mentioned in this post, and in the “Buddhicized” funeral ritual in Pelliot tibétain 239. One interesting aspect of the wooden dockets is that they do suggest whatever adherence to Buddhism the soldiers felt or were expected to show, it was not exclusive of participating in these sorts of rituals.


  9. Sam, this is such valuable data. I am particularly interested in the types of deities listed in these 9th century sources. You mention jul lha, sman and g.yang. In Khu tsha zla ‘od’s 13th century Phur pa gter ma, he includes a remarkably long register of non-Buddhist deities. I am interested to know which ones were already attested in 9th century sources. Rob

  10. Rob,

    There are not very many other types of deities mentioned in the wooden slips. As well as yul lha we have the compound yul lha yul bdag which seems to refer to a single deity (though it is not clear). We have sman and g.yang, and we also have rtse lha and rtse sman.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s