Dharma Mom

Classical Tibetan

Tibetan script


According to the indigenous Tibetan tradition, Tibetan script was invented by Thon- mi Sambhoța during the lifetime of the bean po Khri Sron-rcan (?-649). Whereas the identity of the script inventor (if there was indeed only one person) remains uncertain, it is beyond doubt that the Tibetan script was invented between the 630s and 648.

According to Chinese sources, in 648 Tibetans sent a mission to the Chi- nese court asking for ink and paper manufacturing technology. From this we can infer that by 648 Tibetans already possessed a script. The first mention of a written text (yi ge) in Tibetan sources stems from an Old Tibetan text popularly known as the Old Tibetan Annals. In the annual entry of the year 655/6 we read: བློན་ཆེ་སྟོང་རྩན་གྱྀས། འགོར་ཏཾར། བཀའ། གྲྀམས་གྱྀ་ཡི་གེ་དྲྀས། (PT 1288: 29–30)2 (blon čhe ston rcan gyīs/ ɣgor tīr/ bkay/ grīms gyī yi ge brīs/)
“At Ygor-ti grand councillor [Mgar] Ston-rcan [Yul-zun] wrote down the text of the sovereign laws.” After its primary invention the script underwent a few minor modifications before it reached its final form around the mid-8th century – the form that is still in use today.

The script

The Tibetan script introduced and used throughout the textbook is called dbu čan, lit. “having the head”. The “head” (57 dbu) referred to in the name is the upper horizontal bar characteristic of this script, which is clearly seen above most of the letters of the alphabet, cf.:


DOI: 10.4324/9781003224198-3
Tibetan script 15

Like all the other Tibetan scripts, 5 dbu čan is written from left to right. It is the script traditionally used in woodblock prints and modern printing. It should be learnt as the first Tibetan script because the remaining scripts are either based on it or can be more easily approached with the knowledge of 5 dbu čan.3

Letters versus sounds
At the time of its invention the Tibetan script was phonetic. That means one letter corresponded to one particular sound of the spoken language. As a consequence, learning Tibetan script is almost synonymous with learning the phonetics of the Tibetan language of the 630s. This of course raises the question of how to pro- nounce the letters when reading a Classical Tibetan text. Basically two options are at hand: (1) to read the text as it is written, pronouncing each sound to which a letter is ascribed; or (2) to read the text according to the pronunciation of one of the modern Tibetan dialects. Certainly, for somebody who has just started learning Classical Tibetan and for whom Classical Tibetan is the first Tibetic language learned, the first option would be the easier one. The pronunciation of most of the modern Tibetan dialects has diverged considerably from the written forms of the words. This situation can be compared with French l’eau “water”, which is cur- rently pronounced as [lo], i.e. four letters represent only two sounds, or Eng. plough [plaʊ] – six letters for four sounds. Consequently, the level of difficulty in learning Classical Tibetan with modern pronunciation is much higher for a beginner. Another advantage of the first method is that by learning the classical language in a pronun- ciation distinct from modern ones, students may avoid confusing two (or more) Tibetic languages, in case they should simultaneously study, for example, Classical Tibetan and Lhasa Tibetan. The problematic part of this method of reading texts is that by the period of Classical Tibetan (roughly 9th-19th century; see the chapter Language History) the pronunciation had already deviated from the original, and therefore ascribing Old Tibetan sound values to Classical Tibetan texts might seem anachronistic. However, the second option for reading Classical Tibetan texts (i.e. according to a modern pronunciation) seems even more problematic. Firstly, it is likewise anachronistic because a modern pronunciation is projected on texts written several centuries earlier. Secondly, the pronunciation is arbitrarily chosen and depends on the dialect the teacher herself knows best – usually it is one of the so- called Central Tibetan dialects (Lhasa or the Standard Tibetan), but it may well be any of the modern dialects. Thirdly, classical orthography allowed for consonant clusters that are disallowed in modern written Tibetan (e.g. the onsets – smr- or -sc-); thus the problem arises how to pronounce these. Lastly and most impor- tantly, from a didactic point of view, for students who simultaneously learn Classi- cal Tibetan and a modern Tibetan dialect this option often results in much confusion.
16 Tibetan script
The most typical situation is when a Classical Tibetan word is attributed a meaning from a modern language because in the Classical Tibetan class and in the modern Tibetan class the word is pronounced in exactly the same manner.
Since the main purpose of the textbook is to present the basic grammatical notions of Classical Tibetan in a written form in order to prepare students for read- ing texts, I will restrict the explanations of the script to discussing the letters and the sound values represented by them at the time of the script invention. The most obvious consequence of this approach is the equation:
Tibetan script
བོད་ཡིག་ “written
= Transliteration = Phonetic transcription (= pronunciation)
bod yig
The textbook abstains from introducing modern pronunciation. This approach leaves it to the teacher whether she would like to teach any modern pronunciation to her students, and if so, which of the modern dialects is preferred. The textbook does not impose any dialect and so is more easily accessible to all teachers of Classical Tibetan, disregarding their skills in modern Tibetic languages.”
Alphabet and transliteration
The Tibetan alphabet is called ka kha in Tibetan. The name consists of the first two letters of the alphabet:k and kh. This can be compared with the Eng. alphabet that was borrowed from Latin alphabetum, ultimately derived from Greek aλpáẞηtoç alphabētos that likewise names the first two letters of the Greek alphabet: alpha and bēta.
The Classical Tibetan alphabet consists of thirty letters:7
ཀ ཁ

ག ང க ཇ ཉ ད ན
16 W X 16 M KH R
FD D C 3 q
ཨ ར


5 W
ཤ ས ཧ ཨ
and four additional signs to mark vowels:
a a
Tibetan script 17
As the Tibetan alphabet was based on Gupta alphabets, the arrangement of its let- ters follows (to some extent) the arrangement of the letters in Indian alphabets. The first four rows contain letters that at the time of the script invention represented velar, alveolo-palatal, alveolar, and labial consonants, respectively. The letters within the columns of the first four rows are arranged according to their manner of articulation. The first column contains voiceless unaspirated, the second voice- less aspirated, the third voiced, and the fourth nasal consonants of the respective group. For instance, letter 5 d belongs to the third, alveolar row, and is the third letter in this row – that means it represented a voiced alveolar consonant. Begin- ning with the fifth row, the rules of the arrangement change slightly. They will be explained together with the pronunciation of the letters in the following sections.
Pronunciation of the letters
As mentioned, the letters of the Tibetan alphabet have been systematically arranged according to the original place and manner of articulation of their corresponding sounds.8
Latin letters written beside Tibetan letters in Table III.1 form the transliteration
system used in the textbook.” This system has been adapted from the system used earlier by Jacques Bacot. Its methodological foundations are explained in Bialek (2020b). In the meantime, many scholars and publishing houses around the world have shifted to the so-called Wylie system, which is currently the most frequently applied transliteration system for Classical Tibetan.1o However, due to the serious limitations of Wylie’s transliteration in textual and linguistic studies, the new trans- literation will be used throughout the textbook instead.
Table III.1 provides the phonetic values (in square brackets) of the corresponding Tibetan letters, as well as the transliteration of the latter (in italics).
Table III.1 Consonantal letters








18 Tibetan script
Table III.1 (Continued)












The letters of the first row all represent velar consonants:

voiceless unaspirated [k], like in Eng. scan [skan]
voiceless aspirated [k], similar to but with a strong aspiration voiced [g], like in Eng. go [gǝʊ]
nasal [n], like in Eng. sing [si] or Ger. lang [laŋ] “long”
The letters of the second row represent alveolo-palatal consonants:
3 voiceless unaspirated affricate [tc], like in Pol. éma [tema] “moth”
n | બ
க voiceless aspirated affricate [tch], similar to but with a strong aspiration
voiced affricate [], like in Pol. dźwięk [dzvjɛ̃k] “sound”
3 nasal [n], like in French agneau [ano] “lamb” or Pol. koń [kɔɲ] “horse”
The letters of the third row represent alveolar consonants:
5 voiceless unaspirated [t], like in Eng. star [sta:]
voiceless aspirated [th], similar to 5 but with a strong aspiration
5 voiced [d], like in Eng. doll [dol]
nasal [n], like in Eng. note [noʊt]
The letters of the fourth row represent labial consonants:
པ voiceless unaspirated [p], like in Eng. spear [spi]
24 voiceless aspirated [p], similar to 4 but with a strong aspiration
KY K 16

voiced [b], like in Eng. ball [bɔ:l]
nasal, like in Eng. moth [mv0]
Tibetan script 19
The first three letters of the fifth row follow the pattern of the preceding rows:
voiceless unaspirated alveolar affricate [ts], like in Eng. tsar [tsa:] or Ger. Zeit [tsart] “time”
voiceless aspirated alveolar affricate [th], similar to but with a strong aspiration
voiced alveolar affricate [dz], like in Pol. dzwon [dzvɔn] “bell” or It. zero [dzɛ:ro] “zero”
With the following letter the pattern is broken:

Proudly powered by WordPress