Tibetan Alphabet Table
| Letter | Says (IPA) | Column / register | Note |
|---|---|---|---|
| ཀ | ka | unaspirated - high tone class | |
| ཁ | kʰa | aspirated - medium tone class | |
| ག | ɡa | voiced - low tone class | historical voiced - devoiced in modern Lhasa Tibetan |
| ང | ŋa | nasal/final - low tone class | |
| ཅ | tʃa | unaspirated - high tone class | |
| ཆ | tʃʰa | aspirated - medium tone class | |
| ཇ | dʒa | voiced - low tone class | historical voiced - devoiced in modern Lhasa Tibetan |
| ཉ | ɲa | nasal/final - low tone class | |
| ཏ | ta | unaspirated - high tone class | |
| ཐ | tʰa | aspirated - medium tone class | |
| ད | da | voiced - low tone class | historical voiced - devoiced in modern Lhasa Tibetan |
| ན | na | nasal/final - low tone class | |
| པ | pa | unaspirated - high tone class | |
| ཕ | pʰa | aspirated - medium tone class | |
| བ | ba | voiced - low tone class | historical voiced - devoiced in modern Lhasa Tibetan |
| མ | ma | nasal/final - low tone class | |
| ཙ | tsa | unaspirated - high tone class | |
| ཚ | tsʰa | aspirated - medium tone class | |
| ཛ | dza | voiced - low tone class | historical voiced - devoiced in modern Lhasa Tibetan |
| ཝ | wa | nasal/final - low tone class | glide w - closes the affricate row, no tone-class twin |
| ཞ | ʒa | additional low register row | |
| ཟ | za | additional low register row | |
| འ | ɦa | additional low register row | the a-carrier: breathy h onset, name a (Hill 2005) - bases every vowel sign |
| ཡ | ja | additional low register row | |
| ར | ra | additional medium register row | |
| ལ | la | additional medium register row | |
| ཤ | ʃa | additional medium register row | |
| ས | sa | additional medium register row | |
| ཧ | ha | additional high register row | |
| ཨ | a | additional high register row |
| Vowel sign | Demo on ཀ | Says | Position |
|---|---|---|---|
| ི | ཀི = ki | i | above |
| ུ | ཀུ = ku | u | below |
| ེ | ཀེ = ke | e | above |
| ོ | ཀོ = ko | o | above |
| Tibetan | Devanagari | Arabic | Fraction glyph | Fraction value |
|---|---|---|---|---|
| ༠ | ० | 0 | ༬ | -0.5 |
| ༡ | १ | 1 | ༭ | 0.5 |
| ༢ | २ | 2 | ༮ | 1.5 |
| ༣ | ३ | 3 | ༯ | 2.5 |
| ༤ | ४ | 4 | ༰ | 3.5 |
| ༥ | ५ | 5 | ༱ | 4.5 |
| ༦ | ६ | 6 | ༲ | 5.5 |
| ༧ | ७ | 7 | ༳ | 6.5 |
| ༨ | ८ | 8 | ༴ | 7.5 |
| ༩ | ९ | 9 | ༵ | 8.5 |
The Tibetan alphabet carries thirty consonant radicals, and it wears its phonological history on its sleeve: the classic teaching grid lays them out as five places of articulation times four tone classes, because when the script was devised in the 7th century Old Tibetan still distinguished plain, aspirated and voiced stops - and the tone system of modern Lhasa speech later grew out of exactly those contrasts. Spelling froze around the 9th century; the spoken language kept moving. That is why every letter in the voiced column here - ག ga, ཇ ja, ད da, བ ba, ཛ dza, plus ཞ zh and ཟ za - carries a historical voiced value that Lhasa mouths now say plain and voiceless. Spell one way, hear another, and know that the dictionary is keeping score.
Four vowel signs do all the vowel work: ཨ carries the inherent /a/, while ི i, ེ e and ོ o are written above the letter and ུ u hangs beneath. There is no long-short distinction on the page (Sanskrit loanwords aside), and famously no tone marks at all: modern Tibetan is tonal, but each tone is predictable from the archaic spelling, so the script never needed them. Between syllables sits the tshek (་), a dot doing the work spaces do in English, because Tibetan packs each word into stacked syllable blocks.
How to use
- Read the main table by row group: the five classic rows (guttural ཀ-ང, palatal, dental, labial, affricate ཀ) each walk the four tone classes left to right - unaspirated high, aspirated medium, voiced low, nasal low. Note the two seat-swap specials: the affricate row's fourth seat belongs to ཝ w, a glide with no tone-class twin, and the a-carrier འ sits in the low row with a breathy h-onset. The three added rows at the bottom (zh-za-a-ya, then ra-la-sha-sa, then ha-a) complete the thirty. Filter by group to drill one row at a time.
- Learn the vowel demo column as a chant: ka, ki, ku, ke, ko. The base letter always says its inherent a; the three top-hung signs (i, e, o) and the one bottom-hung sign (u) do the rest. Because every letter secretly ends in a, a bare consonant letter in running text never ends a syllable unwarned - either you wanted the a, or a vowel sign is coming to replace it.
- Use the digit table for dates, prices and chapter numbers: Tibetan runs ༠0-༠9 exactly parallel to Devanagari, and the last two columns are the party trick - a full set of half-number glyphs (༬-༵) dedicated to the values -0.5 through 8.5, one per digit, coined for old astronomical and medical texts so a scribe never had to write a half as two characters.
Frequently asked questions
Why does Tibetan spelling look nothing like modern pronunciation?
Orthographic freeze. The script was designed in the 7th century for Old Tibetan, and spelling was standardized around the early 9th - then the spoken language spent a thousand years collapsing clusters and devoicing the voiced stops. A famously ivory word spelled with seven letters comes out as one syllable in Lhasa. The frozen spelling keeps morphology visible - related words share their silent letters - which is why dictionaries still teach the classical letters, and why the tone of every modern syllable can be predicted from the historical consonant column.
If Tibetan is tonal, why are there no tone marks?
Because the tones arrived after the spelling was frozen, and they never needed their own symbols: each modern tone is the regular reflex of an old initial consonant class plus syllable shape. High-tone syllables descend from the high and medium columns; low-tone syllables descend from the voiced and nasal columns. Learners who memorize the tone-class grid are effectively memorizing the tone rules of modern Lhasa speech with a thousand years of head start.
What is the small dot inside Tibetan text?
The tshek (་), the syllable separator. Tibetan writes whole syllables as stacked blocks - base letter, prefixes, sub-joined letters, vowel signs, suffixes - and sets a tshek after each block. Spaces exist but they separate phrase-level units, not words, so the tshek is what you scan when reading aloud. It also closes each syllable unambiguously, which matters because a single stack can carry up to seven written parts.
What are the fraction glyphs for?
A dedicated half-number series: one glyph per value from -0.5 to 8.5 (༬ ༭ ༮ ༯ ༰...), so 'half past three' or an astronomical reading could be written as a single character. They are the Tibetan answer to a problem other scripts solve with two symbols. Everyday arithmetic never used them - you will meet them in calendars, astrology charts and medical treatises - but Unicode keeps the full set at U+0F2C-U+0F35, and spotting one is the fastest way to identify a classical text.
Who invented the Tibetan script?
Tradition says Thonmi Sambhota, a minister sent by King Songtsen Gampo to India in the 7th century, modelled it on late Brahmi-descended scripts - a story celebrated in Tibet even as historians debate its details. What is certain is the lineage: the letterforms, the inherent-a principle and the digit row descend from the Indic family, making Tibetan the high-altitude anchor of the Brahmi line that runs through Myanmar, Khmer, Thai and Lao.