This page briefly discusses the different textual traditions within traditional Tibetan grammar, and the classic texts in each one. Beri Gyalse and Aachen Tenzin Thupten have posted many Tibetan-language lectures on these topics on YouTube.
A good Tibetan-language compendium of traditional grammatical literature is The Collected Tibetan Grammars བོད་ཀྱི་བརྡ་སྤྲོད་ཕྱོགས་བསྒྲིགས་ in 25 volumes, edited by སི་ཁྲོན་བོད་ཡིག་དཔེ་རྙིང་བསྡུ་སྒྲིག་ཁང་. I have indexed the table of contents from all 25 volumes in the form of a downloadable Tibetan-language spreadsheet:
This spreadsheet is in .ods format, which is the libre counterpart of a .csv file. It was created in LibreOffice, which is a libre counterpart to Microsoft Excel. You can also search the table of contents below:
[table id=1 /]Sections:
1. The Sumtak
The mythical figure of Tönmi Sambhoṭa is credited not only as the creator of the Tibetan alphabet, but also as the author of the two foundational texts of Tibetan grammar: The Thirty (སུམ་ཅུ་པ་ Sumchupa) and The Application of Signs (རྟགས་འཇུག་ Takjuk), together known as the Sumtak (སུམ་རྟགས་). The Thirty discusses Tibetan letters and the functions of Tibetan particles, whereas The Application of Signs discusses Tibetan spelling rules and their effects on the meaning of a word.
The Sumtak are the standard texts used for teaching Written Tibetan grammar in Himalayan communities. They have been the subject of many commentaries over the centuries, and even today they frequently serve as the root text for Tibetan language textbooks. The Thirty is frequently the very first text studied in Himalayan Buddhist monastic curricula.
The most popular commentaries today on the Sumtak are both by Yangchän Truppä Dorje:
- The Well-Explained King of Trees (ལེགས་བཤད་ལྗོན་དབང་ Lekshä Jönwang), which comments on The Thirty;
- The Mirror that Clarifies Difficult Points (དཀའ་གནད་གསལ་བའི་མེ་ལོང་ Kanä Säwä Melong), which comments on The Application of Signs.
The Well-Explained King of Trees has been translated on Lotsawa House, but The Mirror that Clarifies Difficult Points is not yet available in translation for free online. These two commentaries are regarded almost as highly as the Sumtak themselves, and frequently serve as the basis for subcommentaries and textbooks in their own right.
Since the Sumtak focus heavily on letters and spelling, it is perhaps no surprise that they are sometimes studied alongside texts on orthography (དག་ཡིག་ tak-yik). in which case they are sometimes collectively referred to as the Sumtak Taksum (སུམ་རྟགས་དག་གསུམ་).
An classic English-language book on Sumtak literature is Roy Andrew Miller’s Studies in the Grammatical Tradition in Tibet.
2. The Three Groups
A second prominent tradition in Tibetan grammar focuses on the idea of the Three Groups (ཚོགས་གསུམ་ tsok-sum). The Three Groups refer to three aspects of a language’s grammar, namely:
- letters (Skt. vyañjana, Tib. ཡི་གེ་ yi-ke)
- nouns (Skt. nāman, Tib. མིང་ ming)
- phrases (Skt. pada, Tib. ཚིག་ tsik)
These three topics are discussed in order, each one building on the last. The classic Tibetan text on the Three Groups is The Door of Speech (སྨྲ་སྒོ་ Mago) by the Indian mahāpaṇḍita Smṛtijñānakīrti. The idea of the Three Groups is of Indian origin, and it is mentioned in many Buddhist sūtras.
Just like the Sumtak, the Three Groups provide a replicable and customizable framework for analyzing Tibetan grammar. They continue to be used as the basis for modern-day textbooks, such as Thubten Jinpa’s བརྡ་སྤྲོད་སྨྲ་སྒོའི་ལྡེ་མིག་ (Institute of Tibetan Classics, 2010).
A good English-language discussion of the Three Groups can be found in Pieter C. Verhagen’s A History of Sanskrit Grammatical Literature in Tibet, vol. 2, p. 240-253.
Read more on the Three Groups →
3. Sanskrit Grammatical Literature
Of all the Sanskrit grammatical literature translated into Tibetan, the three most prominent texts are the three grammars:
- Candravyākaraṇa by Candragomin
- Kalāpavyākaraṇa (a.k.a. Kātantra) by Śarvavarman
- Sārasvatavyākaraṇa by Anubhūtisvarūpācārya
To my knowledge, as of December 2022, none of these texts have yet been translated into English.
Various other Sanskrit grammatical texts were translated into Tibetan and studied, such as the Varṇasūtra (ཡི་གེའི་མདོ་ Yikei Do) and its commentary. Also worth mentioning is the Mantrasya Paṭhopāya (སྔགས་ཀྱི་བཀླག་ཐབས་ Ngakki Laktap, “how to recite mantras”) genre, which consists of short guides that teach how to pronounce Sanskrit mantras, along with their commentaries. These guides were often composed by Tibetan lotsāwas, and include interesting discussion of contemporary Sanskrit pronunciation norms. One famous text in this genre is Narthang Lotsāwa’s How to Recite Mantras (སྔགས་ཀྱི་བཀླག་ཐབས་ Ngakki Laktap).
A good English-language resource on the Tibetan reception of Sanskrit grammatical literature is Pieter C. Verhagen’s two-volume work, A History of Sanskrit Grammatical Literature in Tibet.
Read more on Sanskrit grammar →
4. Other Tibetan grammar texts
An Aid for Children
The grammar text བྱིས་པ་ལ་ཕན་པ། (“An Aid for Children”) was composed by Sakya Paṇḍita in 1205-6. It is a commentary on Sönam Tsemo’s བྱིས་པ་བདེ་བླག་ཏུ་འཇུག་པ། (“An Easy Entrance for Children”), which discusses Sanskrit and Tibetan pronunciation. This text discusses late 12th century Tibetan pronunciation in quite a lot of detail. Below I have translated excerpts from this text that give examples of Tibetan pronunciation. The Tibetan text is available here; the translated excerpts run from the bottom of p. 167 to the bottom of p. 170.
Translation
“First, those who do not know the place of articulation will, for that reason, experience many wrong pronunciations. For example, pronouncing lda ba for ཟླ་བ་, sgyub pa for བསྒྲུབ་པ་, tra shis for བཀྲ་ཤིས་, and so on. Therefore, I will explain in order to make them know how to speak correctly.
[…]
People from dBus mix up syllables with a ra superscript and standalone letters, or ya-sta and ra-sta letters. For example, they make errors like sta for རྟ་ or tag for སྟག་; or like ‘gyo for འགྲོ་ or phya ba for ཕྲ་བ་, and so on.
People from gTsang mistake syllables with a ma prefix as those with an ‘a prefix, or syllables with a ra superscript or a da prefix as those with a ga prefix. For example, they make errors like ‘go bo for མགོ་བོ་ or ‘kha’ for མཁའ་; or like rtong ba for གཏོང་བ་ or rga ba for དགའ་བ་, and so on.
[…]
Furthermore, there are really very few suffix letters (མཐའ་རྟེན་) in the lects of the commoners in the central region of both dBus and gTsang; for example, ga la ‘khod for གང་ན་འཁོད་ or sgrub pa for བསྒྲུབ་པ་ or dba’ phyug for དབང་ཕྱུག་. That kind [of pronunciation] does not exist in the periphery of Tibet, but it does exist in the centre of the region, and so if the children of the central region are studying, then when they gather, they should be careful about suffix letters and give rise to diligence.”
Discussion
This text shows that the following sound changes were either complete or underway in Tibet before the 13th century:
Unspecified location:
- zla > lda (Note: this is consistent with Bialek’s description of the onset /sl-/ undergoing metathesis, then epenthesis, then reduction.)
- bsgru- > sgyu
- kra > tra
In dBus:
- rta > sta
- sta > ta (Note: this and the preceding change appear to be chain shift.)
- -ra > -ya
The changes in dBus affect stacked letters.
In gTsang:
- m- > ‘-
- g- > r-
- d- > r-
The changes in gTsang affect prefix letters. I am not sure why he describes r and d turning being confused for g, when the examples he gives involve g and d being confused for r.
In the centre of dBus-gTsang, but not the periphery:
- གང་ན་ > ག་ལ་
- b- > [silent]
- དབང་ཕྱུག་ > དབའ་ཕྱུག་
I am not sure why he only describes a loss of suffix letters here, when the second example given is the loss of a prefix letter.
There is not enough data to determine the precise environments that these changes occur in, if they are conditioned. Also, I don’t recognize the -ra > -ya change in modern Lhasa dialect, so I wonder whether it was a fad, or if it affected some other dialect, or something else.
