Chlouvānem: Difference between revisions

Jump to navigation Jump to search
mNo edit summary
Line 101: Line 101:
==Phonology - Yāṃstarlā==
==Phonology - Yāṃstarlā==
===Consonants - Hīmbeyāṃsai===
===Consonants - Hīmbeyāṃsai===
Chlouvānem has a large consonant inventory, with 47 different consonants, divided into seven categories: labials, dentals, palatalized dentals, retroflexes, palatals, velars, and laryngeals. The Chlouvānem term for "consonant" is ''hīmbeyāṃsa'', a compound of ''hīmba'' (colour) and ''yāṃsa'' (sound).
Chlouvānem has a large consonant inventory, with 47 different consonants, divided into seven categories: labials, dentals, palatalized dentals, retroflexes, palatals, velars, and laryngeals. The Chlouvānem term for "consonant" is ''hīmbeyāṃsa'', a compound of ''hīmba'' (colour) and ''yāṃsa'' (sound). The following table organizes consonants by their behaviour - thus, for example, the treatment of the phonetic affricates /c͡ɕ(ʰ) ɟ͡ʑ(ʱ)/ as stops.
{| class="wikitable" style="text-align: center;" |
{| class="wikitable" style="text-align: center;" |
|-
|-
! rowspan=2 colspan=2 | → PoA <br/> ↓ Manner !! rowspan=2 | Labials<br/><small>hærṣoke</small> !! colspan=2 | Dentals<br/><small>aṣṭrūke</small> !! rowspan=2 | Retroflexes<br/><small>āḍhyāsūke</small> !! rowspan=2 | Palatals<br/><small>dehāṃlūdvyūke</small> !! rowspan=2 | Velars<br/><small>bhyodilūdvyūke</small> !! rowspan=2 | Laryngeals<br/><small>diṇḍhūke</small>
! rowspan=2 colspan=2 | → PoA <br/> ↓ Manner !! rowspan=2 | Labials<br/><small>hærṣoke</small> !! colspan=2 | Dentals<br/><small>aṣṭrūke</small> !! rowspan=2 | Retroflexes<br/><small>āḍhyāsūke</small> !! rowspan=2 | Palatals<br/><small>dehāṃlūdvyūke</small> !! rowspan=2 | Velars<br/><small>bhyodilūdvyūke</small> !! rowspan=2 | Laryngeals<br/><small>diṇḍhūke</small>
|-
|-
! <small>Plain<br/>''pūdre''</small> !! <small>Palatalized<br/>''pindehāṃlūdvyūke''</small>
! <small>Plain<br/>''pūdre''</small> !! <small>Palatalized<sup><small>1, 2</small></sup><br/>''pindehāṃlūdvyūke''</small>
|-
|-
! colspan=2 | Nasals
! colspan=2 | Nasals<sup><small>3</small></sup>
| m mʲ || n || nʲ || ɳ || ɲ || || ɴ
| '''''m''''' m<br/>'''''mь''''' mʲ || '''''n''''' n<sup><small>4</small></sup> || '''''nь''''' nʲ || '''''ṇ''''' ɳ || '''''ñ''''' ɲ || ''l'' [ŋ~ɴ]<sup><small>5</small></sup> || '''''ṃ''''' ɴ<sup><small>6</small></sup>
|-
|-
! rowspan=2 | Stops !! <small>Unvoiced</small>
! rowspan=2 | Stops !! <small>Unvoiced</small>
| p pʰ || t̪ t̪ʰ || tʲ tʲʰ || ʈ ʈʰ || || k || ʔ
| '''''p''''' p<br/>'''''ph''''' pʰ || '''''t''''' <br/>'''''th''''' t̪ʰ || '''''tь''''' <br/>'''''thь''''' tʲʰ || '''''ṭ''''' ʈ<br/>'''''ṭh''''' ʈʰ || '''''c''''' c͡ɕ<br/>'''''ch''''' c͡ɕʰ || '''''k''''' k~q<br/>'''''kh''''' kʰ~qʰ || '''''ɂ''''' ʔ
|-
|-
! <small>Voiced</small>
! <small>Voiced</small>
| b bʱ || d̪ d̪ʱ || dʲ dʲʱ || ɖ ɖʱ || || g gʱ ||
| '''''b''''' b<br/>'''''bh''''' bʱ || '''''d''''' <br/>'''''dh''''' d̪ʱ || '''''dь''''' <br/>'''''dhь''''' dʲʱ || '''''ḍ''''' ɖ<br/>'''''ḍh''''' ɖʱ || '''''j''''' ɟ͡ʑ<br/>'''''jh''''' ɟ͡ʑʱ || '''''g''''' ɡ~ɢ<br/>'''''gh''''' ɡʱ~ɢʱ ||  
|-
! rowspan=2 | Affricates !! <small>Unvoiced</small>
| || ||  ||  || c͡ɕ c͡ɕʰ ||  ||
|-
! <small>Voiced</small>
|  ||  || ||  || ɟ͡ʑ ɟ͡ʑʱ ||  ||  
|-
|-
! colspan=2 | Fricatives
! colspan=2 | Fricatives
| f || s || sʲ || ʂ || ɕ ||  || ɦ
| '''''f''''' f~ɸ<sup><small>7</small></sup> || '''''s''''' s || '''''sь''''' sʲ || '''''ṣ''''' ʂ || '''''š''''' ɕ ||  || '''''h''''' ɦ<sup><small>8</small></sup>
|-
|-
! colspan=2 | Approximants
! colspan=2 | Approximants
| ʋ || ||  || || j ||  || ʀ ʀʲ ɴ̆ ɴ̆ʲ
| '''''v''''' ʋ<sup><small>9</small></sup> || ||  || || '''''y''''' j ||  || '''''r''''' ʀ<sup><small>10</small></sup><br/>'''''rь''''' ʀʲ<br/>'''''l''''' ɴ̆<br/>'''''lь''''' ɴ̆ʲ
|}
|}
Table notes:
# The orthographic palatalization marker is transliterated as '''i''' before vowels and '''ь''' before consonants or at the end of a word.
# All consonants which have a non-palatalized/palatalized phonemic pair are always palatalized before /i iː i̤/, thus the contrast is neutralized there.
# All nasals except /ɴ/ may be said to be a single phoneme /N/ when preceding another consonant (except /j/), as they assimilate to the following consonant's PoA.
# Realized as [ŋ] word-finally after high monophthongs (/i iː i̤ u uː ṳ/) and as vowel nasalization after the others, including diphthongs.
# The realization of the sequences orthographically marked as '''lk lkh lg lgh''' varies regionally and, therefore, the '''l'''-phoneme can in these contexts be realized as either [ŋ] or [ɴ]. In most local Chlouvānem pronunciations, these sequences are [ŋk(ʰ) ŋɡ(ʱ)] but, in areas including notably Līlasuṃghāṇa, most of the southern Jade Coast, and the South, they are [ɴq(ʰ) ɴɢ(ʱ)].
# /ɴ/ contrasts with other nasals only before /d dʱ dʲ dʲʱ/, where it is realized as nasalization of the preceding vowel.
# Both allophones are found, in free variation in some pronunciations (e.g. Ilėnimarta, Galiākina, Western Plain), as conditional allophones in others (e.g. Līlasuṃghāṇa, coastal Jade Coast), while some only use the [f] one (e.g. Cami and most of the Far East). For this reason, the phoneme is usually transcribed /f/.
# /ɦ/ may be realized in various ways, including uvular or velar voiceless fricatives, especially when at the end of a word.
# /ʋ/ may be realized as [f] before voiceless consonants; this is <small>NOT</small> reflected orthographically.
# /ʀ/ is often realized as [ʁ] after consonants, especially after coronal stops, and as [ɽ] or [ɻ] adjacent to retroflex consonants. In coda it is usually vocalized to [ɐ̯], except when before a retroflex consonant.


There are only a few instances of consonant allophony, mostly due to the large number of phonemic consonants. The following ones apply to standard Chlouvānem:
Some allophonic variations not proper of standard Chlouvānem but widespread in many areas:<br/>
* All dentals are allophonically palatalized before /i iː i̤/, thus the palatalized/plain contrast is neutralized there.
'''f''' is realized as [ɸ] in those few words where it occurs before /p/ (e.g. ''kafpa'' (a sardine-like fish) [ˈkaɸpa]) or word-finally (all interjections, such as ''hųf!'' "phew!" [ɦṳɸ]); in many pronunciations (notably in Yāmbirhālih, Galiākina, and across Lgraṃñælihaikā diocese, and also increasingly found among young speakers in Līlasuṃghāṇa) it is also realized this way before /u uː ṳ/ (e.g. ''maifu'' "enough" [ˈmaɪ̯ɸu]).
* Coda /ʀ/ is diphthongized to [ɐ̯].
* /ɴ/ has two different realizations depending on context: [ɴ] before other laryngeals, and nasalization of the preceding vowel anywhere else.
* Word-final /n/ is realized as [ŋ] after high vowels, and as vowel nasalization after the other ones.
* Nasals, except /ɴ/ before non-laryngeals, assimilate to the PoA of the following consonant, except for /j/ (note that /ɴj/ does not exist).
 
'''v''' may also allophonically be [f] before voiceless consonants; unlike other cases of saṃdhi, this is ''not'' reflected in the orthography (cf. ''chlævprauda'' [c͡ɕʰɴ̆ɛfˈpʀaʊ̯da] "TV news broadcast" (< ''chlæviṭa'' "TV" + ''prauda'' "newspaper").<br/>'''f''' itself is realized as [ɸ] in those few words where it occurs before /p/ (e.g. ''kafpa'' (a sardine-like fish) [ˈkaɸpa]) or word-finally (all interjections, such as ''hųf!'' "phew!" [ɦṳɸ]); in many pronunciations (notably in Yāmbirhālih, Galiākina, and across Lgraṃñælihaikā diocese, and also increasingly found among young speakers in Līlasuṃghāṇa) it is also realized this way before /u uː ṳ/ (e.g. ''maifu'' "enough" [ˈmaɪ̯ɸu]).


Not proper of standard Chlouvānem but so widespread it is now by far the most common pronunciation is also the fall of /j/ and /ʋ/ before /i iː i̤/ or /u uː ṳ/ respectively, e.g. in ''yinām'' /jinaːm/ [iˈnaːm] (protection, refuge) or ''vurāṇa'' /ʋuʀaːɳa/ [uˈʀaːɳa] (a kind of small-sized reptile)<ref>Many pronunciations, including the common Līlasuṃghāṇi and Galiākñi ones, keep /ʋ/ word-initially in words like ''vurāṇa''. It does however fall in other widespread pronunciations like in most of the Far East, including Cami, as well as parts of the Jade Coast like in Līlta and Ilėnimarta.</ref>. This also leads to phonetic hiatuses, like in ''Kāyīchah'' /kaːjiːc͡ɕʰaɦ/ [kaːˈiːc͡ɕʰaɦ] (an insular diocese between Mārṣūtram and Vedren) or the common given name ''Martayinām'' /maʀtajinaːm/ [maɐ̯ta.iˈnaːm].<br/>A similarly widespread but not standard allophony is the use of either [ɻ] or [ɽ] for /ʀ/ before /ɖ ɖʱ/, as in ''larḍhīka'' (girl, maid) /ɴ̆aʀɖʱiːka/ [ɴ̆aɻˈɖʱiːka].<br/>
Not proper of standard Chlouvānem but so widespread it is now by far the most common pronunciation is also the fall of /j/ and /ʋ/ before /i iː i̤/ or /u uː ṳ/ respectively, e.g. in ''yinām'' /jinaːm/ [iˈnaːm] (protection, refuge) or ''vurāṇa'' /ʋuʀaːɳa/ [uˈʀaːɳa] (a kind of small-sized reptile)<ref>Many pronunciations, including the common Līlasuṃghāṇi and Galiākñi ones, keep /ʋ/ word-initially in words like ''vurāṇa''. It does however fall in other widespread pronunciations like in most of the Far East, including Cami, as well as parts of the Jade Coast like in Līlta and Ilėnimarta.</ref>. This also leads to phonetic hiatuses, like in ''Kāyīchah'' /kaːjiːc͡ɕʰaɦ/ [kaːˈiːc͡ɕʰaɦ] (an insular diocese between Mārṣūtram and Vedren) or the common given name ''Martayinām'' /maʀtajinaːm/ [maɐ̯ta.iˈnaːm].<br/>A similarly widespread but not standard allophony is the use of either [ɻ] or [ɽ] for /ʀ/ before /ɖ ɖʱ/, as in ''larḍhīka'' (girl, maid) /ɴ̆aʀɖʱiːka/ [ɴ̆aɻˈɖʱiːka].<br/>