in search of a language unrecognised (part 2)

***This post is a “Part 2”.  If you’ve clicked directly to this entry, please check out Part 1 here.

Around 2005 I began to take an active interest in Tibetan again.  In the back pages of my Tibetan book I reread my notes from Rilong, and still wondered about this language.  My best conclusion was that I didn’t know what else it would likely be if not a variety of Kham Tibetan.  By chance, I found a Tibetan tutor that year who happened to come from the Kham region of western Sichuan.  He had left the area as a teenager and escaped to India where he lived in a Tibetan community before emigrating to Vancouver, Canada, where I was living at the time.  After I learned of his story, I brought my notes to our next lesson feeling quite certain he was going to settle my doubts once and for all. However, to my disappointment, he said it didn’t sound like any kind of Kham Tibetan that he’d ever heard and had no suggestions of what language it could be.

I returned to the internet, as one does in times of confusion, and was happy to find that there was significantly more information available this time.  The combination of an uncensored internet and the passage of 4 or 5 years yielded a good amount of results on the topic of the languages of western Sichuan.  Clicking around on Ethnologue, I came across a language listed as Jiarong, spoken in north central Sichuan, including Xiaojin County, where Si Guniang Mountain is.  The Xiaojin dialect was known as Situ.  Could this be it?  The entry said Jiarong was phonologically and lexically similar to Tibetan, with complex consonant clusters, though grammatically more similar to Qiang.  It certainly seemed to fit the bill.  The listed names for the language included Chiarong, dGyarung, Gyarong, Gyarung, Jarong, Jyarung, Keru, Rgyarong; Jiarong seemed to possibly be a Chinese pinyin rendering of a more “Tibetan-ish” sounding name.  With a little more poking around I came across the name rGyalrong.  Every other western Sichuan language that I found didn’t quite match geographically or had already been ruled out. I had a good feeling about this rGyalrong.  I was getting closer to an answer!

Now all I had to do was corroborate these findings with a language sample to compare the words and phrases I collected in Rilong with words and phrases in verified rGyalrong.  And here I met another road block.  I didn’t keep any record of it, but if memory serves me right, I was only able to find a paragraph long transcribed piece of a folk tale in an unspecified dialect of rGyalrong.  The text didn’t really help to confirm anything, but I did observe definite phonological similarities between the sample text and my collected data.  With hesitant joy, I tentatively concluded that the language I was after was probably rGyalrong…maybe.  Not exactly the definitive victory I was hoping for.

At any rate, another 9 or 10 years passed by, and I decided it was high time to start writing about all the language-related musings going on in my head.  The rGyalrong mystery was one of the first ideas that occurred to me because a.) everyone loves a good mystery and b.) I thought if I published the story, I might get some actual answers back.  So about three weeks ago, I started searching around online yet again for rGyalrong related information and found that in the past ten years rGyalrong studies and research has made leaps and bounds.  Before I could even begin to write my story, I was getting answers left and right.

The first search yielded exactly what I was looking for, a site actually called the rGyalrongic Languages Database.  It seems that during the years I was waiting for more information to become available, two linguists were working to collect, document, and record lexical items from rGyalrong dialects.  I couldn’t believe my eyes.  In their database I was able to specify the dialect of Rilong Town and find 200 sentences transcribed, all with audio.  There was also a lexicon of hundreds of individual words!  It was an incredibly thorough project.  I scrambled for my notes and began to compare.

Screen Shot 2015-11-07 at 08.36.58 p.m.

A few discrepancies popped out, but at first glance, the two number sets looked extremely similar.  My data tends to favour unvoiced consonants, while the database uses voiced consonants, but that isn’t terribly surprising.  As a native speaker of English I am bound to have the tendency to hear unvoiced, unaspirated consonants as voiced.  My list apparently also has six and seven mixed up, but that could be the result of confusion collecting data on my part or the part of my informant.  Finally, number nine is a bit of a mystery.  I couldn’t find anything in the database resembling /sa’sɔm/, so I assume I wrote down the wrong item and/or it means “I don’t know”, “I forget”, or “I’m bored – can we please stop doing this?”.  At any rate, things were looking good.  It seemed that after all this time, rGyalrong was indeed the language!  But just to double check…

Screen Shot 2015-11-07 at 08.36.37 p.m.

Further confirmation!   My informant’s version of hello/你好 is an obvious calque from Chinese, which is apparently used by at least one speaker but is very probably not a native expression.  The database gave a different phrase, but as a calque the word-to-word translation checks out.  “Where are you going?” was also confirmed, though with the same voiced/unvoiced error.  And finally, the second word for “good” also checks out.  Incidentally, /kʊˈsna/ was the word for good listed under Rilong, whereas almost all the other towns in Xiaojin County used /dɛˈla/.  I’m not sure of the background of my friend in Rilong, but he seems to use both.  In addition to what is listed above, I was able to corroborate at least 5 or 6 other scribblings of mine, and at long last I could put a name to this language I’d though so much about:  rGyalrong!

After a moment of contentment, the questions came flooding.  So who were the rGyalrong anyway?  And why did they call themselves Tibetans?  And why had nobody ever mentioned them to me while I was in Sichuan?  As luck would have it, an ethnography of the rGyalrong people was published just last year.  The book is titled  rGyalrong, Conservation and Change, written by David Burnett, who lived in Sichuan and studied the cultures of the western half of the province right around the time I first encountered the term rGyalrong online.  This wealth of information was incredible after such a long wait.


The rGyalrong are one of the many ethnic and linguistic groups of western Sichuan.  Due to the extreme nature of the mountains and valleys where they live, they have been relatively isolated throughout much of their history, resulting in a wide variety of divergent dialects.  There is documentation of the rGyalrong as early as the Tang dynasty (AD 608-917) describing a legendary matriarchal society, ruled by a queen, protected by fortress-like watchtowers along a river of gold.  Ancient towers still stand, and the Dajin River (大金川, Big Gold River) flows through rGyalrong territory, historically rich in gold.  The term rGyalrong is actually an abbreviation of rGyal Mo Tsha Ba Rong (རྒྱལ་མོ་ཚ་བ་རོང), which means “The Queen’s River Valley” in Tibetan.  But why are they called by a Tibetan name? And why didn’t the people in Rilong tell me they spoke rGyalrong or even identify themselves as such?

The term rGyalrong is actually not a self-designation but a term used by outsiders (rendered in Chinese as Jiā Róng, 嘉绒).  The rGyalrong refer to themselves as Keru (/kəru/ or /kɯrɯ/ depending on the dialect), but to understand why they also identify as Tibetans, we have to look at the history of ethnic minorities in the People’s Republic of China.  Before the establishment of the PRC the rGyalrong were ruled by local chieftains, Tu Si (土司).  They had religious and cultural ties to Tibet via Gelug-pa Buddhism and had contact with the Han Chinese to the east, but they enjoyed relative independence from their larger neighbours.  With the establishment of the PRC and the occupation of Tibet, however, the rGyalrong were swallowed up and lost all autonomy.  In the 1950’s when the government set about to identifying and categorizing all of the minority nations in the newly expanded China, the rGyalrong, being small in numbers, were lumped in with the Tibetans.  To the present day, the rGyalrong are officially categorized as Zāng Zú (藏族), or Tibetan, and because of this denomination and the PRC’s approach to “promoting” ethnic minority cultures, many rGyalrong look towards Tibet (or rather an official, stereotyped version of Tibet) as a frame of reference for their own cultural identity.

So that basically answers most of my questions.  My host on Si Guniang Mountain was speaking rGyalrong, but identified both himself and his language as Tibetan, as per official policy.  After almost 15 years, the mystery of rGyalrong is solved!

I am grateful to have taken this journey.  Along the way we’ve encountered Standard Tibetan, Chinese, Qiang, Amdo Tibetan, Kham Tibetan, and finally rGyalrong.  I hope that rGyalrong continues to gain increased recognition and is able to resist the pressures of more dominant languages in the region.  It seems to me a very dangerous situation when a language is denied its own name and one group’s cultural identity is appended onto that of another group.  The resulting invisibility does not bode well for the future of the language.  I hope that by sharing my story, I can add a tiny bit to the ever growing store of information available about the Keru, or rGyalrong, people and encourage others to further document, support, and defend this language tucked away in the mountains of western Sichuan.


Well, there was one question that was left unanswered, in case you missed it.  From all of the new sources of information I have found, it sounds as if rGyalrong has never been put into writing, yet my informant wrote his name for me as I showed in the last post.  I haven’t been able to figure out what this script is, but I do have a vague guess.  In the bookstore in Hong Yuan, I found a beautiful book of different styles of Tibetan calligraphy, some of them quite different from the standard print.  This book probably deserves an entire post of its own, but for now I’ll share one page.


I can’t quite match any of the hand-written letters with the letters above, but it does suggest the possibility that there could be a style of handwriting for the Tibetan alphabet with a similar form that could be used to write words or names in rGyalrong.  After all, rGyalrong practitioners of Buddhism will still read scriptures in Tibetan and will at least be familiar with the Tibetan script.  This is merely conjecture though.  If anyone has any ideas about the writing sample above, or even about the usage of different styles of Tibetan calligraphy, please share!

in search of a language unrecognised (part 1)

As a native speaker of a major world language (and a learner of many others), I have come to realise that I take for granted the precision with which we define and delineate individual languages.  English is English, Korean is Korean.  Sometimes Dutch and Flemish cause some confusion, but even then we can slap the label Tussentaal on the grey area between the two.  And there are obviously many more examples of linguistic grey areas where opinions will differ on what to call a given community’s speech, but for the most part we’re not used to ambiguity when it comes to identifying languages.  Below I’d like to share the story of a language I encountered about 15 years ago and how I never really knew what it was until now.

In 2001 I moved to Chengdu, the capital city of Sichuan province in southwestern China.  The western half of the province is quite linguistically diverse, often referred to as the “ethnic corridor”.  In the mountains and grasslands of western Sichuan dialects of Tibetan and many varieties of Qiangic languages abound, many of them very isolated and mutually unintelligible with their neighbouring languages.  For most Han Chinese in Chengdu, though, the western regions of the province are simply “Tibetan”.

At my first chance, I took advice from a friend in Chengdu to visit a famous mountain called Si Guniang Shan (རི་བོ་སྐུ་བླ་འི།, 四姑娘山, Four Maidens Mountain) in Xiaojin County (小金县), Aba Tibetan and Qiang Autonomous Prefecture (རྔ་བ་བོད་རིགས་དང་ཆང་རིགས་རང་སྐྱོང་ཁུལ , 坝藏族羌族自治州).  

The scenery of Xiaojin County was incredible.  Having driven narrow, winding roads, teetering above the churning, cobalt Minjiang River, we eventually began to ascend into the clouds.   As the air became thinner and thinner, we finally arrived in Rilong Village (日隆镇).  I was excited by the change in scenery, new faces and mostly by the promise of a new language.  It was my first time in a Tibetan area, and understanding nothing about the broad variety of Tibetan dialects, I naïvely brought along my Learning Practical Tibetan text book, a perfectly adequate book for learning and practicing Standard Lhasa Tibetan. 


I arranged to stay in a family-owned guesthouse and quickly found myself sitting on the terrace chatting with one of the owners in Mandarin.  I asked him about the language he spoke with his family, and he confirmed that it was , or Tibetan.  I told him I had learned some Tibetan and tried a couple phrases, but nothing seemed to register.  I pulled out my Tibetan book and showed him some of the phrases written in Tibetan script.  “Oh, that’s not the Tibetan we speak here.  Ours is different,” he informed me.  I jumped to the conclusion that the local language was a divergent dialect of Tibetan and set about to asking him how to say different things in his dialect.

Here are a few of the things I jotted down on the spot in (questionable) IPA inside the back cover of my Tibetan book.

hello / 你好 – nɔ ɡəˈsna

where are you going? / 你去哪里? – nɔ ɡəzdəˈtʃʲɛ

good / 好 – dɛˈla

And the numbers 1-10:

Screen Shot 2015-10-27 at 07.13.31 p.m.

Comparing this set of numbers with the Standard Tibetan numbers, some loose parallels can be found – some vowels match and some consonants match at least in place of articulation , but it is clearly a different set of numbers.  The most obvious cognate is the number three – taking into account the written but not pronounced prefix g-, Standard Tibetan “three” is almost identical to my informer’s “three”.  

As for the sentences above, they bare no resemblance to their Tibetan counterparts.  Taking // to be “you” (the phrase for hello could possibly be a Chinese calque:  你好 “you good”), it seems completely different from ཁྱེད་རང་ /kʼʲɛɾaŋ/, you in Standard Tibetan.  None of this really gives any definitive information, but it is clear that the “Tibetan” in Rilong is something other than Tibetan as spoken in Tibet proper.

And finally I asked my new friend to write his name for me in Tibetan.  He said his Tibetan name was pronounced  /ɡʷɔɹ ʝʲa/.  To my surprise he wrote:


I could not identify this script at all, and it definitely didn’t resemble any Tibetan script I had ever seen up until that point.  This language was becoming curiouser and curiouser.

On the bus back to Chengdu, I couldn’t stop wondering about what this “different Tibetan” was exactly.  Back in the city I went online to see what bits of information I could piece together.  The government tends to block a lot of sites having to do with ethnic minorities in China, so I could only find basic facts at the time.  What I was able to find out was that much of Aba Prefecture, where I had been, was part of Amdo, one of three traditional regions of Tibet.  I also learned of Kham, another historical region of Tibet that encompasses much of western Sichuan Province.  Both of these regions were said to have their own Tibetan dialects.  In my searches I also came across the Qiang, an ethnic group living in mountainous areas west of Chengdu, who spoke a Tibeto-Burman language.  Could my friend in Rilong have been speaking Amdo or Kham Tibetan?  Could it have been Qiang, but just referred to as “Tibetan” due to cultural affinity or custom?  I began to wish I had been more systematic about collecting data on Si Guniang Shan.  I had had no idea I would have a linguistic mystery on my hands.


My next trip into western Sichuan allowed me to visit Qiang territory, specifically Taoping Village, which is also northwest of Chengdu, in Wenchuan County (汶川县).  In Taoping I found people living a largely traditional lifestyle.  Their stone, fortress-like houses topped off with tall watchtowers reflect a tumultuous past when the threat of the Chinese, Tibetans, or even neighbouring Qiang villages was ever looming.

In Taoping, the people were speaking a language with some phonological similarities to the language I heard in Rilong:  complex consonant clusters and a similar array of vowels sounds.  However, when I asked what language they spoke, people gave me a definitive answer:  Qiang (羌语).  When I asked them if their language was Tibetan or if they spoke Tibetan, the answer was always no.  The Qiang I met had no ambiguity about their language or identity.  Pulling out my notes from Rilong to ask about a few phrases, it became clear that Qiang was not the language I was looking for.

A Qiang conversation in Taoping Village

My third trip to northwestern Sichuan was back to Aba Prefecture, but this time a bit further from Chengdu, past the mountains and into the grasslands of Amdo.  This time I visited a place called Hongyuan (红原, རྐ་ཁོག་), a kind of one-yak town inhabited by Tibetans, Han Chinese, and some Hui Muslims, and frequented by Tibetan nomads from the surrounding areas.  Hongyuan was a stark contrast to the more mountainous regions I had visited.  From the edge of town all you could see in any direction was grassland extending to the horizon.  The Tibetans in Hongyuan emulated the image of Tibetans we often see in the west:  nomadic herders, long fleece-lined coats, hands busy with prayer wheels and prayer beads during idle moments.  I wondered if the language here would also be more recognisable.

As luck would have it, Hongyuan had a bookstore, where I was able to find some books for teaching children to read and write Tibetan, as well as an English textbook for Amdo Tibetan speakers.  I was happy to see that Amdo Tibetan was not all that different from Standard Tibetan.  Some differences in vocabulary, pronouns and verb endings were apparent, but the speech in Hongyuan was definitely Tibetan.  The biggest difference I noticed was that pronunciation in Amdo was much more conservative and much closer to the written word than in Standard Tibetan.

Tibetan syllables tend to be written with consonant prefixes and/or suffixes around the core of the syllable.  In Standard Tibetan these prefixes and suffixes are not usually pronounced per se, but instead they make their presence known by influencing the tone or vowel sound of the syllable.  In Amdo Tibetan, however, most of these prefixes and suffixes seemed to be pronounced, yielding quite complex consonant clusters.  Below are the numbers, as pronounced by a monk I met in Hongyuan.  Observing the spelling, you’ll see that he basically pronounces the numbers as they are written, unlike Standard Tibetan speakers.

Screen Shot 2015-10-27 at 08.30.06 p.m.

(On a side note, the consonant clusters are so prevalent in Amdo speech that I noticed some people inserting them into English words when they asked me to teach them how to say things in my language.  One monk bid me farewell by saying “zGood-rByez”.)

I left Hongyuan with a clear impression of Amdo Tibetan (at least as spoken in Hongyuan) and was loaded up with books to study and notes to review.  I felt no closer, however, to figuring out anything about the language I first encountered in Rilong months earlier.  If Amdo Tibetan wasn’t so drastically different from Standard Tibetan, then it seemed unlikely that Kham Tibetan would be any more different either.  I wasn’t sure Rilong fell within Kham territory anyway.  The Qiang theory had been disproven as well.  Some piece of information was missing, and nobody seemed to have any answers.  Soon after Hongyuan I left China, and for the next several years my language mystery would remain unsolved until new resources made themselves available.