Unit 3: Stacked letters

In this unit we will learn about superscript letters and subscript letters. We will also learn how to spell all the parts of a syllable, and will discuss other rules governing Standard Tibetan pronunciation. By the end of this unit you will be able to pronounce words like དྲ་ཐོག་ (“online”), བོད་སྐད་ (“the spoken Tibetan language”), and སློབ་ཚན་ (“lesson”) .

Unit 3 Sections:

1. 1. Recap
2. 2. Superscript letters
   1. 2.1. Other effects on voicing
3. 3. Subscript letters
   1. 3.1. Subscript ཝ་
   2. 3.2. Subscript ཡ་
   3. 3.3. Subscript ར་
   4. 3.4. Subscript ལ་
   5. 3.5. Resolving ambiguities
4. 4. Spelling the parts of a syllable
   1. 4.1. Spelling prefix letters
   2. 4.2. Spelling suffix and post-suffix letters
   3. 4.3. Spelling superscript and subscript letters
   4. 4.4. “Lazy” pronunciation
5. 5. Rules affecting tone
   1. 5.1. The high-tone second syllable rule
   2. 5.2. The low-tone second syllable rule
   3. 5.3. Syllable chunking
   4. 5.4. Intonation
   5. 5.5. Tone sandhi
6. 6. Other pronunciation rules
   1. 6.1. Consonant restoration
   2. 6.2. Non-first syllables
   3. 6.3. Vowel harmony
   4. 6.4. Rime deletion
7. 7. Terminology
   1. 7.1. Vocabulary
   2. 7.2. Jargon

1. Recap

Let’s recap what we’ve learned about Tibetan syllables. Tibetan syllables are made up of up to seven letters:

1. Prefix letter (སྔོན་འཇུག་ ngönjuk)
2. Superscript letter (མགོ་ཅན་ gochän)
3. Main letter (མིང་གཞི་ mingshi)
4. Subscript letter (འདོགས་ཅན་ dokchän)
5. Vowel letter (དབྱངས་ yang)
6. Suffix letter (རྗེས་འཇུག་ jenjuk)
7. Post-suffix letter (ཡང་འཇུག་ yangjuk)

In this unit we will learn about stacked letters, which include superscript letters and subscript letters:

We can summarize their pronunciation as follows:

1. Superscript letters are usually silent, and have the same effects on pronunciation as prefix letters.
2. Subscript ཝ་ is silent, subscript ཡ་ adds a y sound or turns the main letter into a ch sound, subscript ར་ turns the main letter into a tr sound, and subscript ལ་ turns the main letter into an l sound.

Let’s start by learning about superscript letters.

2. Superscript letters

There are 3 superscript letters: ར་ལ་ས་. Superscript letters are written above the main letter:

རྙ་ལྔ་སྣ་

The superscript letter ར་ has a special conjunct form when attached to most main letters, in which it loses its bottom stroke:

ར་ + ཀ་ = རྐ་

Note: a conjunct is a combination of two or more stacked consonant letters.

The superscript letters are all silent by default, and have the same effects on pronunciation as prefix letters do. For example:

• Effect on tone:
  • རྔ་ nga
  • ལྔ་ nga
  • སྔ་ nga
• Effect on aspiration and/or voicing:
  • རྡིག་ tik / dik
  • ལྡིག་ tik / dik
  • སྡིག་ tik / dik
• No effect on pronunciation:
  • རྐུ་ ku
  • ལྐུ་ ku
  • སྐུ་ ku

Note that the main letters ཡ་ཞ་ཟ་ can have prefix letters but not superscript letters.

2.1. Other effects on voicing

The superscript ས་ written above any nasal letter or ལ་ may cause the main letter to become voiceless (which I show with a preceding “h” in the phonetic transcription) in addition to raising its tone:

• སྔོན་པོ་ hngö(n)po
• སློབ་དཔོན་ hloppön

Also, the superscript ལ་ plus the main letter ཧ་ is pronounced like a high-tone voiceless l:

• ལྷ་ hla

3. Subscript letters

There are 4 subscript letters: ཝ་ཡ་ར་ལ་. Subscript letters are written beneath the main letter. When added to a main letter, ཝ་ཡ་ར་ have special conjunct forms and ལ་ is written slightly smaller than normal, as follows:

ཀྭ་ཀྱ་ཀྲ་ཀླ་

3.1. Subscript ཝ་

The subscript letter ཝ་ (called ཝ་ཟུར་ wasur) has no pronunciation:

• རྩྭ་ tsa (“grass”)
• རྭ་ ra (“horn”)

This letter used to be pronounced in older forms of the language, and is still pronounced in some other Tibetic languages.

3.2. Subscript ཡ་

The subscript letter ཡ་ (called ཡ་བཏགས་ yata) is usually pronounced like a y sound when added to the ཀ་སྡེ་. For example:

ཀྱ་ཁྱ་གྱ་ (ངྱ་ does not exist)

kya, kyha, kyha

However, some speakers will instead pronounce these syllables like the corresponding member of the ཅ་སྡེ་:

ཀྱ་ཁྱ་གྱ་ (ངྱ་ does not exist)

cha, chha, chha

As before, we could generalize this distinction to talk about Kya lects and Cha lects. However, speakers often use both pronunciations. More research is needed to determine the contexts in which each form is preferred.

When a ཡ་བཏགས་ is added to the པ་སྡེ་, it causes the syllable to be pronounced like the corresponding member of the ཅ་སྡེ་. For example:

པྱ་ཕྱ་བྱ་མྱ་

cha, chha, chha, nya

Also, the cluster དབྱ་ is pronounced ya. (Note: a cluster is a group of consonants with no intervening vowel sound.)

3.3. Subscript ར་

The subscript letter ར་ (called ར་བཏགས་ rata) causes the following main letters to be pronounced like a “tr” sound, while retaining their original voicing and aspiration:

ཀྲ་ཁྲ་གྲ་	tra, trha, trha
ཏྲ་དྲ་	tra, trha
པྲ་ཕྲ་བྲ་	tra, trha, trha

It is important to note that these sounds are not exactly like Indian retroflex consonants such as ट or ठ. In careful speech, you can clearly hear a subtle “r” sound in the Tibetan conjuncts. The best practice, as always, is to listen to native speakers and follow their pronunciation.

When combined with མ་ and ས་, the subscript letter ར་ is silent:

• མྲ་ ma
• སྲ་ sa

Also, the cluster དབྲ་ is pronounced ra.

3.4. Subscript ལ་

The subscript letter ལ་ (called ལ་བཏགས་ lata) causes almost all main letters to be pronounced like a simple high-tone ལ་:

• ཀླ་ la
• བླ་ la
• རླ་ la

The one exception to this is the conjunct ཟླ་, which is pronounced da.

3.5. Resolving ambiguities

In certain conjuncts it can be unclear which letter is the main letter. For example, in སླ་, is ས་ a superscript letter and ལ་ a main letter? Or is ས་ a main letter and ལ་ a subscript letter? Typically, སླ་ and རླ་ are considered to be conjuncts with a subscript letter ལ་. This is a minor point, though, because the pronunciation would be the same either way.

4. Spelling the parts of a syllable

In Unit 1 we learned how to spell main letters and vowels using Tibetan’s native spelling system, སྦྱོར་ཀློག་ jorlok. We will now learn how to spell the remaining parts of the syllable.

4.1. Spelling prefix letters

Prefix letters are spelled by saying the prefix letter followed by the word འོག་, o(k) which means “after”. The prefix letter is then incorporated into the spelling of the rest of the word (see below). For example:

• ཀུ་ (no prefix letter)
  • Spelling: ཀ་ཞབས་ཀྱུ་ཀུ། (ka shapkyu ku)
• དཀུ་ (prefix letter ད་)
  • Spelling: ད་འོག་དཀའ་ཞབས་ཀྱུ་དཀུ (t^ha-o ka shapkyu ku)

Note that the འ་ in དཀའ་ is only there because དཀ་ by itself is not permitted in Tibetan writing. See section 5.3 above.

Sound change incorporation in jorlok:

When spelling, any sound change that a letter causes is generally incorporated into the spelling. For example, prefix letters cause main letters such as ག་ to become voiced. Therefore, when spelling the word དགུ་, the main letter ག་ can be pronounced as a voiced ga rather than as k^ha. To illustrate:

• དགུ་
  • Spelling: ད་འོག་དགའ་ཞབས་ཀྱུ་དགུ་ (t^ha-o ga shapkyu gu)

4.2. Spelling suffix and post-suffix letters

Suffix and post-suffix letters are spelled by 1) saying the name of the suffix letter (and the post-suffix letter, if there is one), and then 2) pronouncing the resulting syllable. When spelling, post-suffix ས་ is often said as “se” rather than “sa”. For example:

• ཐག་
  • Spelling: ཐ་ག་ཐག་ (t^ha k^ha t^hak)
• འགོས་
  • Spelling: འ་འོག་འགའ་ན་རོ་འགོ་ས་འགོས་ (a-o ga naro go sa gö)
• ཐུགས་
  • Spelling: ཐ་ཞབས་ཀྱུ་ཐུ་ག་སེ་ཐུགས་ (t^ha shapkyu t^hu k^ha se t^huk)

4.3. Spelling superscript and subscript letters

Superscript and subscript letters are both spelled using the word བཏགས་ tā, which means “attached (below).”

Superscript letters are spelled by 1) saying the name of the superscript letter, and then 2) saying the name of the main letter followed by བཏགས་, and then 3) saying the resulting syllable. For example:

• ལྔ་
  • Spelling: ལ་ང་བཏགས་ལྔ་ (la ngata nga)
• སྐོར་
  • Spelling: ས་ཀ་བཏགས་སྐ་ན་རོ་སྐོ་ར་སྐོར་ (sa kata ka naro ko ra kor)

Subscript letters are spelled by 1) saying the name of the subscript letter followed by the word བཏགས་, and then 2) saying the name of the resulting syllable. For example:

• བླ་
  • Spelling: བ་ལ་བཏགས་བླ་ (p^ha lata la)
• སྐྱོ་
  • Spelling: ས་ཀ་བཏགས་སྐ་ཡ་བཏགས་སྐྱ་ན་རོ་སྐྱོ་ (sa kata ka yata kya naro kyo)
• བསྒྲིམས་
  • Spelling: བ་འོག་བསའ་ག་བཏགས་བསྒ་ར་བཏགས་བསྒྲ་གི་གུ་བསྒྲི་མ་ས་བསྒྲིམས་ (p^hao sa gata ga rata dra k^hik^hu dri ma se drim)

4.4. “Lazy” pronunciation

In general, unstressed words in Standard Tibetan tend to be pronounced 1) low tone and 2) in their weak form. Therefore, when doing jorlok, the words འོག་ and བཏགས་ are typically pronounced simply as “o” and “ta”, and བཏགས་ is typically pronounced low-tone. Likewise, the vowels, suffix letters, and post-suffix letters such as ས་ are unstressed in jorlok, so they tend to be pronounced low-tone as well.

5. Rules affecting tone

There are a variety of other rules affecting the tone of syllables in Standard Tibetan. In this section we will learn some of the most important ones.

5.1. The high-tone second syllable rule

In most two-syllable words the second syllable is pronounced high-tone, even if it would naturally be low-tone. Look at the following words and listen to the audio:

• ལིམ་ཇ་ limch^ha (“lemon tea”)
• ཞྭ་མོ་ shamo (“hat”)

This means that 2-syllable words must have one of two basic tone formats:

• high-high (if first syllable is high)
• low-high (if first syllable is low)

5.2. The low-tone second syllable rule

The exception to the above rule is that in two-syllable verbs, the second syllable is typically pronounced low-tone, even if would naturally be high-tone. This principle is used to help distinguish between nouns and verbs with the same spelling:

• ཐུག་པ་ t^hukpa (the noun “noodle soup”)
• ཐུག་པ་ t^hukpa (the verb “to meet”)
• འགྲོ་བ་ drowa (the noun “transmigrator”)
• འགྲོ་བ་ drowa (the verb “to go”)

Verbs usually behave quite differently from nouns in Tibetan grammar, so it is usually easy to tell whether a word is acting as a noun or as a verb. In cases of ambiguous written sentences (e.g. ཁོ་འགྲོ་བ་རེད།, ཁ་ལག་བཟོ་ས་རེད།), we must rely on context to determine the intended reading. However, ambiguity is uncommon.

5.3. Syllable chunking

Syllable chunking is when words longer than two syllables are broken up into smaller “chunks” through the use of tone and rhythm. Each chunk is kept to only one or two syllables long, and follows all the rules for pronunciation that we’ve learned so far. This allows longer words to be spoken more easily.

Often in three syllable words, the first syllable is treated as one chunk, and the second and third syllables are treated as another chunk. For example, the word for “pretty” (སྙིང་རྗེ་མོ་) is read as nying-jemo:

It would be incorrect to pronounce the tones of each syllable individually (*nying-je-mo), and it would also be incorrect to chunk the word differently (e.g. *nyingje-mo). However, other three-syllable words chunk differently.

This use of tone and rhythm can be very useful for distinguishing between ambiguous written sentences. As we will see in unit 10, chunking is used in the speech to differentiate when words like རན་ are being used as nominalizers (in which case they are chunked together with the verb) or as secondary verbs (in which case they are chunked separately from the verb):

• ཁ་ལག་འདི་ཟ་རན་འདུག k^halak di sarän du: “This food is worth eating.”
• ཁ་ལག་འདི་ཟ་རན་འདུག k^halak di sa rändu: “It’s time to eat this food.”

5.4. Intonation

Like English, Tibetan uses intonation (pitch in a sentence) for emotional and communicative effect. Intonation may exaggerate or violate established tone rules, and is most often used to make a statement stronger or more emphatic. One important difference to note is that unlike English, Standard Tibetan does not use a rising pitch to mark questions. Also, sometimes low-tone syllables are pronounced high-tone when spoken in isolation, or when emphasized at the beginning of a sentence.

5.5. Tone sandhi

Tone is not actually read at a robotic high-or-low binary in Tibetan. Tibetan tone flows naturally and interacts with intonation in complex ways. All of the tone rules covered so far apply mainly to individual words spoken clearly or in isolation, but actual speech is complex and may differ from this standard.

In the middle of a sentence, tones may flow in unexpected ways, with some high tones sounding higher than others, and some low tones sounding lower than others. Tones change their pitch due to the influence of other adjacent tones, which is a process called tone sandhi.

I will not discuss tone sandhi in detail here, and instead I encourage learners to copy the patterns of native speakers. For a discussion of tone sandhi in Lhasa Tibetan speakers in the Canadian Tibetan diaspora, see Jonathan Lim’s dissertation, The Tonal and Intonational Phonology of Lhasa Tibetan.

6. Other pronunciation rules

6.1. Consonant restoration

Consonant restoration is when a normally silent consonant gets pronounced in a word. Consonant restoration typically happens when the first syllable of the word is an open syllable, meaning that it ends in a vowel sound. However, we will also look at one kind of consonant restoration that occurs after closed syllables, which end in a consonant sound.

Consonant restoration with adjacent letters:

Here are the usual restorations of the five prefix letters:

• བཅུ་གཅིག་ chukchik (restored “k” from the prefix letter ག་)
• ཉི་ཤུ་རྩ་དགུ་ nyishu tsargu (restored “r” from the prefix letter ད་)
• ས་བཀྲ་ saptra (restored “p” from the prefix letter བ་)
• ཨ་མདོ་ amdo (restored “m” from the prefix letter མ་)
• དེ་འདྲ་ t^hendra (restored “n” from the prefix letter འ་)

After an open syllable, a silent letter at the beginning of the second syllable has the space to reappear (or be “restored”) in the pronunciation. The restored consonant is generally pronounced as if it were the suffix letter of the first syllable, even though it’s not.

When the first syllable is an open syllable, the restored prefix letter འ་ is pronounced like the nasal of the same row of the second syllable’s main letter, so ནད་འབུ་ (“virus”) is pronounced like nämbu, and མི་འགྱུར་ (“unchanging”) is pronounced like ming-gyur. Note that syllables ending in consonant letters still count as open syllables if their suffix letter is silent, so ནད་ counts as an open syllable.

When the first syllable is a closed syllable, the prefix letter འ་ can make the first syllable’s suffix letter be pronounced like the nasal of its same row:

• ཞབས་འདེགས་ shamdek (“a service”)
• ཞབས་འདྲེན་ shamdren (“a disservice”)
• རྟགས་འཇུག་ tangjuk (“The Application of Signs”, a grammar text)
• ཚོགས་འདུ་ ts^hongdu (“a meeting”)

The restored prefix letter མ་ causes the same kinds of changes as འ་ after both open and closed syllables:

• གཟའ་མཇུག་ sanjuk
• ས་མགོ་ sang-go
• ཡིག་མགོ་ ying-go

Consonant restoration often triggers further sound changes. For example, the word བཅུ་གཉིས་ chung-nyi undergoes two shifts: 1) the prefix ག་ is restored to a “k” sound, and 2) the restored “k” sound changes to a “ng” sound under the influence of the following ཉ་. This second shift is the same phenomenon as when the suffix letter ག་ is pronounced like ང་ before a nasal sound, as seen in Unit 2 §4.2. I will use the term suffix nasalization to refer to the transformation of syllable-final consonant sounds into nasal sounds in any context.

Consonant restoration is often optional. For example, རྟགས་འཇུག་ can be read either as tangjuk or as takjuk. Different speakers use consonant restoration to different degrees. Some words even have alternate spellings, one subject to consonant restoration and the other not. For example, both ས་བཀྲ་ and ས་ཁྲ་ mean “map”.

Consonant restoration with stacked letters:

Consonant restoration also affects the superscript letters ར་ and ལ་, as well as the three conjuncts ལྡ་ལྗ་ཟླ་:

• རྡོ་རྗེ་ dorje (restored “r” from the superscript letter ར་)
• བཅོ་ལྔ་ chölnga (restored “l” from the superscript letter ལ་)
  • (note that this restoration also triggers vowel raising due to the influence of ལ་ after a vowel)
• དཔལ་ལྡན་ pändän (restored “n” from the conjunct ལྡ་)
• ཡུལ་ལྗོངས་ yünjong (restored “n” from the conjunct ལྗ་)
• སྤྱི་ཟླ་ chinda (restored “n” from the conjunct ཟླ་)

“N” intrusion:

The sound “n” can also be inserted unexpectedly, which I call the intrusive “n”:

• སྒྲ་གདངས་ dradang or drandang
• ཕྱི་དྲིལ་ chitril or chindril (possibly influenced by ཕྱི་འབྲེལ་)
• ང་ཚོ་ ngats^ho or ngants^ho
• འདོད་དཀྱིལ་ dökyil or dönkyil

More research is needed to determine the contexts in which “n” intrusion occurs, so examples of “n” intrusion can be treated as exceptions and learned by rote. It seems to happen most often around letters of the ཏ་སྡེ་.

6.2. Non-first syllables

A non-first syllable is any syllable in a word other than the first one.

1) པ་སྡེ་ syllables:

The syllables བ་ and བོ་ are pronounced as “wa” and “wo” respectively when they’re not the first syllable of a word:

• ཇོ་བོ་ ch^howo
• བྱ་བྲལ་བ་ ch^ha tr^hälwa

When པ་ is not the first syllable of a word and occurs after a vowel sound, it is often pronounced with the lips barely touching, a sound that I will transcribe as β̞. For example:

• ག་པར་ག་པར་ k^haβ̞a: k^haβ̞a
  • Note: this slightly exasperated interjection is a good example of how intonation can affect pronunciation. The first syllable is lengthened for emphasis, which I show with a colon (:), and the tone of the fourth syllable unexpectedly falls.
• གྲྭ་པ་ tr^haβ̞a

2) Aspirated syllables:

The pronunciation of aspirated non-first syllables needs more research. In careful speech, aspirated non-first syllables tend to stay aspirated, but in some contexts they become unaspirated (and possibly voiced). As always, it is best to follow the pronunciation of a native speaker.

6.3. Vowel harmony

Vowel harmony is when the vowel in one syllable becomes more similar in sound to the vowel in an adjacent syllable. Vowel harmony is mainly triggered by the vowels ཨི་ and ཨུ་. For example, we might expect the word གོམས་གཤིས་ to be pronounced k^homshi, but it’s typically pronounced k^humshi. The “o” in the first syllable changes to “u” due the influence of the second syllable’s “i”.

Below I list some common types of vowel harmony. I’ve named each type for the vowel sounds that are in each syllable before vowel harmony is applied:

• i-o harmony (i-o → i-u):
  • སྐྱིད་གཤོངས་ kyishung
  • སྲིད་སྐྱོང་ sikyung
• u-o harmony (u-o → o-o):
  • བུ་མོ་ phomo
  • མཆུ་ཏོ་ ch^hoto
• ü-o harmony (ü-o → ü-u):
  • དུད་འགྲོ་ t^hündru
  • ཀུན་སློང་ künlung
• o-i harmony (o-i → u-i):
  • ཁོ་གཉིས་ k^hunyi
  • བློ་འཛིན་ lundzin
  • གོ་སྒྲིག་ k^hudrik
  • སློབ་ཁྲིད་ luptri
• o-u harmony (o-u → u-u):
  • ཤོག་གུ་ shuku (proper written form: ཤོག་བུ་)
  • གོ་བསྡུར་ k^hudur
  • གོ་རྒྱུ་ k^hurgyu
  • ལྗོངས་རྒྱུ་ jungju

Sometimes vowel harmony seems to operate in different ways. For example, anecdotally I have heard t^huktru for ཐུགས་སྤྲོ་ and ngüldö for དངུལ་རྡོ་, both of which differ from the patterns outlined above.

6.4. Rime deletion

In linguistics, the term rime refers to the part of the syllable that includes the vowel and any final consonants. In a Tibetan syllable, the rime consists of the vowel, the suffix letter, and the post-suffix letter.

In Standard Tibetan, many words beginning in ཞ་, ཟ་, ར་, ཤ་, ས་ are subject to rime deletion, which means that they lose their rime entirely. This tends to happen to unstressed syllables. For example:

• ལབ་དགོས་རེད་ lap gör (instead of lap göre)
• འདུག་སེ་ duks (instead of dukse)
• ཁ་པར་སླེབས་སོང་ k^hapar leps (instead of k^hapar lepsong)

This may trigger other changes. For example, it may causes a preceding suffix letter ང་ to convert to a nasal vowel (n):

• མཐོང་སོང་ t^ho(n)s

Rime deletion is more common with some syllables than others. For example, it seems to be mandatory for སེ་ but generally optional for སོང་.

The sound “s” (derived from ཟེར་, “to say”) is often attached to the end of an sentence when quoting or repeating what someone else has said.

---

The rules learned in this unit will serve as our baseline for Standard Tibetan pronunciation. This pronunciation will help you understand and be understood by most Tibetan speakers in the diaspora. Remember that so-called “Standard Tibetan” is just a particular dialect, and is only called “standard” because people treat it as such. It doesn’t mean that some kinds of Tibetan are better than others. Tibetan is Tibetan no matter what dialect someone speaks.

7. Terminology

7.1. Vocabulary

Audio:

7.2. Jargon

Audio: