Pronouncing Naga languages: ü, tones and uvulars
All three languages on this site are written in the Roman alphabet, and all three use it to represent sounds English does not have. Here is what the letters are actually doing.
Why spelling varies so much
None of these languages has a fully enforced spelling standard. Nagamese has essentially none — you will find kinika and kineka, bishi and bisi, apni and apuni, para and pora, all in ordinary use, sometimes in the same document. Ao and Sümi are more consistent because they have school textbooks and published literature, but variation persists.
Two forces are behind this. Roman orthography arrived through nineteenth-century missionary printing, which prioritised getting scripture into print over designing a systematic writing system. And there has never been a state language academy with the authority to settle the remaining questions.
The practical upshot for a learner: do not assume a spelling difference is a word difference. When something looks unfamiliar, try saying it out loud — you have often met the word already under another spelling.
The letter ü
ü is by far the most common non-English character in Ao and Sümi text — Sümi, pezü, tenüng, küthü, Zünheboto, tsüngremmong. It represents a central vowel, made with the tongue neither at the front nor the back of the mouth.
It is not the German ü. The closest widely known reference point for an English speaker is the vowel in "bird" said without the r, or the unstressed vowel at the end of "sofa" — but produced with more tension and held as a full vowel. In Sümi the realisation shifts by position: the close central vowel surfaces as a mid vowel word-medially.
Ao: three tones you cannot see
Chungli Ao has three register tones — high, mid and low — and they are contrastive, meaning tone alone can distinguish two otherwise identical words. Linguistic descriptions mark them (pá, aŋó). Everyday writing does not mark them at all.
So written Ao systematically under-specifies the spoken language, and readers reconstruct tone from context. Two consequences follow:
- For learners: you cannot learn Ao pronunciation from text. Working with a speaker, or with audio, is not a nice-to-have.
- For machine translation: the training data has lost information the language carries. This is one of several reasons Ao output needs checking.
There is also downstream tone behaviour worth knowing about: a high tone before a low tone often surfaces as a falling contour, and many apparent high tones are mid tones upstepped because a low tone follows.
Sümi: uvulars, and the letters that mislead
Sümi is the only Naga language with uvular consonants — stops made with the back of the tongue against the uvula, further back than English k. Both voiceless and aspirated uvular stops occur. To an English ear the effect is a k produced deep in the throat, and it is the single most recognisable feature of spoken Sümi.
Sümi also contrasts aspirated and plain nasals and lateral approximants — an m with a puff of air against an m without, and the same for n and l. English does not use this contrast at all, so it takes deliberate practice to hear before it can be produced.
Several Roman letters are doing unfamiliar jobs in Sümi orthography. q, x and gh all represent sounds that have nothing to do with their English values — q in particular is associated with the back-of-throat articulation, as in timiqo "men" and küsüghu "bad". Read them as Sümi letters, not as English ones.
Nagamese: the easy one
Nagamese is comfortably the most approachable of the three for a new speaker. It is not tonal. It has no uvulars and no aspirated nasals. Its sound inventory is close to Assamese, and spelling — inconsistent as it is — is broadly phonetic.
A few pointers:
bh,dh,gh,kh,ph,thare aspirated consonants, one sound each, not two.bhal"good" begins with a single aspiratedb, notb+h.- Vowels are pure and short. The
aofaseis closer to the vowel in "cut" than to the one in "cake". - Word-final
-eis pronounced, not silent:aseis two syllables. aiinkhai,jaiis a diphthong gliding towardsi.
The translator will read Nagamese output aloud, which is the fastest way to check whether your reading of a spelling matches how it is actually said.
Sources and further reading
Grammar descriptions on this page are drawn from the reference material listed below and from the community phrase set behind the translator. Listing a source does not mean every claim here has been checked against it — where they disagree, or where a speaker says otherwise, the speaker is right.
- NagaTranslate Ao and Sümi reference notes — phonological descriptions
- CIIL adult primer in Sema — Sümi orthography and sound inventory
- Ao language — tone system