Lwaitel phonology


Phoneme inventory

Lwaitel has a small number of phonemes, which are shown in the following charts in IPA. For those who don't know IPA, all of these consonants are as in English except /j/ which is the y in yes, and /ŋ/ which is the ng in sing; the vowels are a as in father (though a bit farther forward in your mouth), i as in ski, and u as in haiku, and ə is the first sound in about. The letters between angle brackets are the romanization.

Stops/p/ 〈p〉/t/ 〈t〉(/ʧ/ 〈ch〉)/k/ 〈k〉 audio
Nasals/m/ 〈m〉/n/ 〈n〉(/ɲ/)/ŋ/ 〈ng〉 audio
Fricatives/s/ 〈s〉(/ʃ/ 〈sh〉, /ç/)/h/ 〈h〉 audio
Laterals(/lʷ/ 〈lw〉)/l/ 〈l〉(/lʲ~l̠ʲ~ʎ/ 〈ly, yl〉) audio
Approximants/w/ 〈w〉/j/ 〈y〉 audio

The palatal consonants are mostly allophones of the alveolar consonants (see the section on allophony below); however, /lʷ/ and /lʲ/ are distinguished from /l/ when they're syllabic. (/lʲ/ is like saying /l/ and /i/ at the same time, /lʷ/ is /l/ with rounded lips.)

High/i/ 〈i〉/u/ 〈u〉
Mid(/ə/ 〈e〉)
Low/a/ 〈a〉

/ə/ is an allophone of /a/ (/a/ only occurs in stressed syllables, /ə/ in unstressed), but I'm treating them as separate phonemes here because that's how I originally wrote it.

〈i u e a ai au iu ui em en eng el eyl elw〉::


Palatal consonants

Many consonants have palatal variants in certain positions:

/n/  audio [ɲ]  audio (ñ in Spanish, gn in Italian and French)
/t/  audio [ʧ]  audio (ch in English) or [ʨ]
/s/  audio [ʃ͡ç], [ɕ], or [ʃ]  audio (sh in English)
/h/  audio [ç]  audio
/l/  audio [lʲ] or [l̠ʲ] or [ʎ]  audio

These variants are used when a consonant is adjacent to /i/ (including in diphthongs), /j/, or /əlʲ/. When two alveolar consonants are adjacent, and one becomes palatal by this rule, the other one becomes palatal as well. However, /t/ is only affected by, and only affects, sounds that come after it, not sounds that come before it.

When a consonant with a palatal form comes before /j/, the /j/ is generally not pronounced, except for causing the palatal allophone to be used.


Since only three vowels are distinguished, vowels tend to have a large range of possible realizations: /i/ can be pronounced [i], [ɨ], [e], [ɪ]; /u/ can be pronounced [u], [ʉ], [o], [ʊ]; /a/ can be pronounced /æ/, /a/, /ɑ/, /ɜ/; and /ə/ can be pronounced [ɪ], [ə], [ʊ], [ɐ]. Vowel realizations tend to be closer to the previous vowel or glide, particularly in unstressed syllables or directly following phonemic glides, and tend to be higher after /k/ and /ŋ/. Stressed vowels not following glides or velars are pronounced as cardinal [a], [i], [u]. For instance, súshli /ˈsus.li/ is pronounced [ˈsuʃ.ʎɨ]  audio , with the stressed vowel pronounced as a cardinal [u] and the unstressed vowel pronounced farther back to assimilate with the /u/, and hyángu /ˈhjaŋu/ is pronounced [ˈçæŋu]  audio , with the /a/ being pronounced farther forward due to the /j/. The first component in the diphthongs /iu/ and /ui/ is mid-centralized, so those diphthongs are pronounced [ɪʉ̯] and [ʊɨ̯], respectively.

/ə/ tends to be pronounced extremely short, especially before a consonant. When /ə/ is followed by a consonant, it tends to lengthen the consonant. When /ə/ is followed by /m/, /n/, /ŋ/, /l/, /lʲ/, /lʷ/, the /ə/ is not pronounced and the consonant becomes syllabic, except when the consonant is nasal and it's preceded by a stop or nasal at the same place of articulation (e.g., /pəm/, /nən/).


Non-syllabic /l/ may optionally be pronounced as something near [ɰ] before /k/, especially when it's also after /a/.


Syllable structure: (C)(C)V(V)(C)(C).

The following rules tend to hold within words, but may sometimes be broken:

Stress and timing

Depending on how it's analyzed, either stress in unpredictable (with /a/ and /ə/ being allophones based on stress, and restrictions on what can appear in an unstressed syllable), or stress is mostly predictable based on the vowels and syllable codas.

Lwaitel is stress-timed; that is, the time between stressed syllables is roughly the same.

The person/proximity indicators, the basic verbs, the adadjectives, and a few prepositions have a stressed syllable, or at least treated as stressed with regards to their vowel quality. However, these syllables tend to be given slightly less stress than other stressed syllables, and may act as an unstressed syllable with regards to timing, particularly if the syllable is immediately followed by another stressed syllable in the same phrase.

Words cannot have more than two (rarely three) syllables after the stressed syllable, and cannot have more than two (and usually one or zero) syllables before the stressed syllable. A small number of words have more than one stressed syllable (particularly those that have the prefix ún– "not"); the first stressed syllables of these words are similar to the stressed function words described in the previous paragraph.


Declarative sentences end in a falling tone. Questions end in a rising tone, including content questions. (TODO more?)

Other things affecting pronunciation

Pronunciation of the vowels may change with the person's mood, with excitedness or happiness causing vowels to move forward ([a~æ, i, y~ʉ~ʏ]) and sadness causing vowels to move down and backward ([ɑ, ɪ, ʊ~o̝]).

When imitating foreigners:

Phonological processes

While the language was not derived historically, there are certain phonological processes that occur when words are derived from other words. Most of these changes are not reflected in the writing system or sign language.