Classical Ngade n Tim Ar (CT: Ĝate n Tim Ar /ŋa˩te˩ n̩ tim˩ aɹ˩/ [ŋa˩de˩ n̩ tim˩ aɹ˩]) is the ancestral language to the various Tim Ar languages and is currently in use as a lingua franca for the Kmtön n Tim Ar.
- 1 Phonology
- 2 Writing system
- 3 Linguistic history
- 4 Grammar
- 4.1 Verbs
- 4.1.1 Voice
- 4.1.2 Aspect
- 4.1.3 Voice
- 4.1 Verbs
|Fricative||θ ð||s · ɬ ł||x ḫ||h|
|Approximant||ɹ r · l||ʕ g|
Voicing of voiceless obstruents (except /h/) between two voiced sounds within the same word was productive in CT—e.g., híðia /hi˥θi˩a˩/ → [hi˥ði˩a˩] 'crab'.
|High||i · y ü||ɯ ï · u|
|Mid||e · ø ö||ɤ ë · o|
- High-tone vowels are written with an acute if the low-tone grapheme has no accent mark or with a circumflex if it has an umlaut, which the circumflex replaces (q.v.).
- Fronting of /a/ to [æ] immediately preceding a nasal within the same word was productive in the CT period, e.g. saman út /sa˩man˩ ut˥/ → [sæ˩mæn˩ ut˥] ' saman úd '
/m̩ n̩ ŋ̩ ɹ̩ l̩/ m n ĝ r l
There are two tones, low and high. Low tone is indicated with either no diacritic or an umlaut; high tone is indicated with an acute accent or a circumflex (replacing an umlaut).
Syllabic consonants cannot take distinctive tone.
When the genitive particle (m, n, or ĝ, as appropriate) precedes a nasal in the next words, that nasal elides.
n noḫo ü /n̩ no˩xo˩ y˩/ → n 'oḫo ü [n‿o˩ɣo˩ y˩] 'of the nojo-plant'
This contrasts with vowel-initial words:
n oḫo ü /n̩ o˩xo˩ y˩/ → n oḫo ü [n̩ o˩ɣo˩ y˩] 'of the washbasin'
CT is natively written using a form of cuneiform called e nákhî ar.
Development from Proto-Tim Ar-O
Phonological situation in Proto-Tim Ar-O
Proto-Tim Ar-O appears to have had only two phonemic vowels, *o and *e. As in CT, several consonants appear to have had the ability to stand without *o *e in the nucleus, i.e. *w *ɹ *l *j *ʁ. Of these, *w *j *ʁ seem to have had syllabic allophones [*u *i *a].
The apparent voicing contrast in Proto-Tim Ar-O seems to have generated a tone split. Voiced onsets created a low tone on a following vowel. Voiceless consonants generated a high tone, with the exception of the glottal stop *ʔ, which patterned with the voiced onsets. This was followed by voiced-voiceless pairs merging to the voiceless consonant, with *h becoming *ʔ. Additionally, original *VʔV and *VhV sequences, where the vowels are identical, yield low-tone and high-tone variants of the vowel.
Cluster resolution and metathesis
Clusters consisting of three consonants resolved as follows:
- If the first consonant was a resonant, it was deleted if it came before two obstruents (the "ROO rule").
- Remaining clusters of three consonants dropped the third consonant.
Syllable-initial clusters of an obstruent plus one of *l *ɹ also simplified. *p was deleted before *l when both occurred in onset position; onset *t *k *s *x combined with this *l to form a lateral fricative *ɬ. All five consonants combined with *ɹ yielding *ʃ when these consonants were in onset position.
After this, sandhi rules changed the plural marker from a standalone word to a proclitic. The eroded plural marker then triggered metathesis of a following CVC sequence to VCC; that is, *#ɹa=CVC > *#aɹCVC > *#əɹVCC > *#VCC. This once more permitted clusters of three consecutive consonants in the language, though these only appear in plural forms (e.g., híðia ~ íhðia). Extensive analogy has converted this to a productive plural pattern.
New syllabic *ʁ appears to have vocalized at this time; this ultimately leads to forms such as łog ~ oła.
Several lenitions appear to have occurred. The uvular plosive *q lenited to /x/. Additionally, a shift of the spirants *s *ʃ forward occurred, ending up as *θ *s. Additionally, pursuant to this process, *p lenited to /h/ through an intermediate *f (as attested by evidence from loanwords). These latter two processes are exemplified in the history of PTO *pjsjʁ 'scorpion' > CT híðia /hi˥θi˩a˩/ 'crab'; cf. PO *pjhjr [pfihja] 'scorpion' > O fư:yè [fɨːjɛ].
Two vowels of unlike frontness/backness (*a was exempt from this process) merged into a vowel with the height, tone, and place of articulation of the second vowel and the roundedness of the first—e.g., *u˩e˥ > ø˥. Following this, the glottal stop *ʔ dropped, leading to hiatus. *w *j near a vowel then gained fully vocalic surface forms, and the uvular *ʁ backed to a pharyngeal resonant /ʕ/.
Consonants, with the exception of *h, voiced when in between two voiced sounds within a word. Additionally, the low vowel *a fronted slightly to [æ] immediately preceding a nasal. These remained a productive allophonic process in CT.
The primary diagnostic for how CT developed into the various Tim Ar dialects involves the evolution of original /θ/. No current idiom preserves /θ/ as /θ/. The current Imperial dialect has merged this with /s/, leading to a number of homophones.
Proto-Tim Ar-O had a nominalizing infix *-e- (the "*e-infix") that appears to have generally been used to derive agentive nouns from verbs. Classical Ngade n Tim Ar, for unknown reasons, eschewed this infix, leading to an abundance of zero-derivation processes, to the point that nouns and verbs often have the same form. Some derivational processes do appear to have arisen or to have been retained.
CT also developed a copula, áge, from a PTO phrasal verb *hʁ ʁe 'stand there'.
CT has four verbal voices: Active, reflexive, reciprocal, and cooperative. (What would be phrased as a passive in English is expressed using the dummy subject lü.)
Voice in CT can be expressed several ways.
. . .lü (dummy subject)
intëĝ. . .lü
mho. . .lü
ágné. . .(siḫna/whatever)