/p b ɓ̥ ɓ t d ɗ k g ʔ/
/f s h/
/m̥ m mˀ n̥ n nˀ ɲ̥ ɲ ɲˀ ŋ̥ ŋ ŋˀ/
/l lˀ ç j jˀ ʍ w wˀ/
The consonant /r/ colors preceding /a/ to [ɔɔ], and is realized as [ɔ̈] in syllabic position; it is otherwise merged in realization into /w/.
The bilabial implosives are recent developments from labial-velars.
Nasals are preploded after oral vowels in main syllables.
/a e i ɨ u/
Vowels can be nasal, breathy, or both: e.g. /nˀjnˀã̤h/ 'be inebriated'. Nasal breathy vowels are realized as lengthened sequences of a breathy vocalic onglide and a nasal vowel: [ʔnɪˈʔnɦɑ̤ɑ̃].
All vowels are pronounced by default as lax [ɑ̈ ɛ ɪ ə ʊ]; however, all except /ɨ/ lengthen and tense in certain VR sequences.
Note that all of /i ji ij jij/ contrast:
/tit/ [tɪt] < *tiiti
/tijt/ [tiːt] < *taati
/t~jit/ [tɕɪt] < *sita + reduplicative prefix
/t~jijt/ [tɕiːt] < *hihata + reduplicative prefix
The same is probably true of /u wu uw wuw/.
The default What word structure is a minor syllable, maximally CRC-, followed by a major syllable, maximally CRVRCF, where R is one of /l r j w/, and F is one of /f s/.
The major syllable can be divided into the CR- onset, the -VR- vocalic, the -C- coda, and the -F- postcoda. However, resonants are preferentially syllabified into the coda for the purposes of reduplicative morphology, so the resonantal element of the vocalic can only be present if there's a coda.
Minor syllables can take epenthetic schwa; however, postconsonantal resonants in minor syllables are made syllabic:
/tjmah/ [tɪmɑ̈] < *tejuma
/mwma̤h/ [mʊmɑ̤̈] < *mufume
/t-r~nẽr/ [tɔ̈nɛ̃w] < *damzo
Major syllables must have a coda, but don't necessarily have an onset. If there's no onset, a glottal stop is epenthesized. This shows up in reduplication:
/uh/ [ʔʊ] > /h~uh/ [hʊ]
cf. /ʔuh/ [ʔʊ] > /h~ʔuh/ [həʔʊ]
Coda /h/ is also often realized phrase-finally.
Implosives and voiceless resonants can't appear in minor syllables.
It is, of course, also possible for words to consist of multiple syllabic groups; e.g. /ʔɨhhat/. These are stressed on the first major syllable, I guess.
The consonants are written straightforwardly.
/p b ɓ̥ ɓ ⁿb ⁿɓ̥ ⁿɓ t d ɗ ⁿd (ⁿɗ) k g ⁿg ʔ/ <p b ph bh mb mph mbh t d dh nd (ndh) k g nk x>
/f s h/ <f s xh>
/m̥ m mˀ n̥ n nˀ ɲ̥ ɲ ɲˀ ŋ̥ ŋ ŋˀ/ <mh m mx nh n nx ñh ñ ñx ngh ng ngx>
/l lˀ ç j jˀ ʍ w wˀ/ <l lx hi~ih i ix hu~uh u ux>
The vowels are not.
/a e i ɨ u/ <a e i w u>
/ar ej ij uw/ <â ê î û>
/-j- -w- -r- -0-/ <i u a y>
/ju iw/ <iu iù>
Nasal finals are written -m -nn -ñ -ng after nasal vowels, otherwise -bm -dn -dñ -gn. (Voiceless nasals are -pm etc.) Final -h is unwritten; -n -h are used for nasal and breathy vowels. For example:
- Warning (octavia.php:3212): A non-numeric value encountered