Meskangela: Difference between revisions

From Linguifex
Jump to navigation Jump to search
Tags: Mobile edit Mobile web edit
Line 345: Line 345:


===Syllable structure===
===Syllable structure===
The typical Proto-Meskangela root syllable consisted of the following structural elements: an onset consisting of a root initial consonant '''C<sub>i</sub>''', optionally followed by a liquid '''L''' or semivowel glide '''G''' (either "j" or "w"); and a vocalic nucleus consisting minimally of a simple vowel '''V''', followed by a final consonant '''C<sub>f</sub>'''. The semivowels could also occur postvocalically, forming falling diphthongs in  "-w"  and "-j", thus belonging to the inventory of '''C<sub>f</sub>'''. Unlike word roots, prefixes and suffixes followed a different pattern, consisting of a single consonant followed by a vowel '''PV<sub>p</sub>''' or '''SV<sub>s</sub>''' (in case of prefixes and suffixes respectively). Only root vowels could carry vowel length (:) and tone ('''t'''), the latter being a consonantal feature at the Proto-Meskangela stage. There was no contrast between zero-initial  *VC  and glottal-initial  *ʔVC, in such cases the second variant is reconstructed with '''C<sub>i</sub>''' being an obligatory element. of the root syllable. Two non-syllabic suffixes are reconstructible for Proto-Meskangela, *-s and *-n. When added they could have resulted in a forbidden postvocalic sequence of two consonants ( e.g. -C<sub>f</sub>s or C<sub>f</sub>n). For these specific instances the suffixes are instead reconstructed as '''ə<sub>s</sub>S''' with "ə" being a short semi-syllabic element that disappeared in Classical Meskangela and later dialects. In other cases, where a single final consonant suffix is expected, a "hollow" consonant is reconstructed (which in some cases gives rise to tone H<sub>t</sub>, or disappear without a trace H<sub>0</sub>), as in  *rjaH<sub>0</sub>ən  “to laugh”. In Classical Meskangela the suffix became phonetically identical the C<sub>f</sub>, becoming a part of the root ( e.g. ''gësata'' ← ''*gVsaH<sub>0</sub>t-ʔa''  “he/she kills”. Thus a potential fully inflected word consists of the following elements:
The typical Proto-Meskangela root syllable consisted of the following structural elements: an onset consisting of a root initial consonant '''C<sub>i</sub>''', optionally followed by a liquid '''L''' or semivowel glide '''G''' (either "j" or "w"); and a vocalic nucleus consisting minimally of a simple vowel '''V''', followed by a final consonant '''C<sub>f</sub>'''. The semivowels could also occur postvocalically, forming falling diphthongs in  "-w"  and "-j", thus belonging to the inventory of '''C<sub>f</sub>'''. Unlike word roots, prefixes and suffixes followed a different pattern, consisting of a single consonant followed by a vowel '''PV<sub>p</sub>''' or '''SV<sub>s</sub>''' (in case of prefixes and suffixes respectively). Only root vowels could carry vowel length (:) and tone ('''t'''), the latter being a consonantal feature at the Proto-Meskangela stage. There was no contrast between zero-initial  *VC  and glottal-initial  *ʔVC, in such cases the second variant is reconstructed with '''C<sub>i</sub>''' being an obligatory element. of the root syllable. Two non-syllabic suffixes are reconstructible for Proto-Meskangela, *-s and *-n. When added they could have resulted in a forbidden postvocalic sequence of two consonants ( e.g. -C<sub>f</sub>s or C<sub>f</sub>n). For these specific instances the suffixes are instead reconstructed as '''ə<sub>s</sub>S''' with "ə" being a short semi-syllabic element that disappeared in Classical Meskangela and later dialects. In other cases, where a single final consonant suffix is expected, a "hollow" consonant is reconstructed (which in some cases gives rise to tone H<sub>t</sub>, or disappear without a trace H<sub>0</sub>), as in  *rjaH<sub>0</sub>ən  “to laugh”. In Classical Meskangela the suffix became phonetically identical the C<sub>f</sub>, becoming a part of the root ( e.g. ''gësata'' ← ''*gasaH<sub>0</sub>t-ʔa''  “he/she kills”. Thus a potential fully inflected word consists of the following elements:
: {|
: {|
|PV<sub>p</sub>—C<sub>i</sub>—L—G—V(:)—C<sub>f</sub>/H<sub>0/t</sub>—SV<sub>s</sub>—(ə<sub>s</sub>S)
|PV<sub>p</sub>—C<sub>i</sub>—L—G—V(:)—C<sub>f</sub>/H<sub>0/t</sub>—SV<sub>s</sub>—(ə<sub>s</sub>S)
Line 354: Line 354:
|}
|}
Later dialects generally follow the model above, modifying some individual elements, such as adding more permissible clusters, or merging the clusters into single consonants, thus retaining all the elements only nominally. This is especially true for the Southern dialects, most of which became fairly analytic and lost most of their suffixes and prefixes in the process, as well as tone and contrastive vowel length.
Later dialects generally follow the model above, modifying some individual elements, such as adding more permissible clusters, or merging the clusters into single consonants, thus retaining all the elements only nominally. This is especially true for the Southern dialects, most of which became fairly analytic and lost most of their suffixes and prefixes in the process, as well as tone and contrastive vowel length.
==Grammar==
==Grammar==
In this subsection only the grammar of Classical Meskangela is discussed, considering the amount of variation among different dialect groups and uncertainty of the Old Meskangela morphological structure. Classical Meskangela as well as most of its dialects are [[w:Agglutinative language|agglutinative]] or weakly [[w:Fusional language|fusional]], the individual morphological elements are not easily segmentable, due in large part to the presence of [[w:Portmanteau|portmanteaux]] morphemes and [[w:Allomorph|allomorphy]].
In this subsection only the grammar of Classical Meskangela is discussed, considering the amount of variation among different dialect groups and uncertainty of the Old Meskangela morphological structure. Classical Meskangela as well as most of its dialects are [[w:Agglutinative language|agglutinative]] or weakly [[w:Fusional language|fusional]], the individual morphological elements are not easily segmentable, due in large part to the presence of [[w:Portmanteau|portmanteaux]] morphemes and [[w:Allomorph|allomorphy]].

Revision as of 06:09, 19 September 2022

Meskangela language
མསྐཾངེལཿ
Mëskaŋelā
Meskangela dialects.png
Created byRaistas
Settingplanet Earth (Europe)
Sino-Tibetan
Early form
*Proto-Himalayan
  • Proto-Meskangela (4000–1300 BNE)
  • Old Meskangela (1300–300 BNE)
  • Classical Meskangela (300 BNE–0 NE)
This article contains IPA phonetic symbols. Without proper rendering support, you may see question marks, boxes, or other symbols instead of Unicode characters. For an introductory guide on IPA symbols, see Help:IPA.

Meskangela (Classical Meskangela: མསྐཾངེལཿ mëskaŋelā; Western མསྐཾངེལཿ mëskaŋela; Eastern མསྐཾངེལཿ mëskåŋeła; Southern མཁཾངྃག măkhoŋäg) is a Himalayan language of an unknown origin. The most common hypothesis suggests its origin in the West Himalayan region and migrated westward over the period of three thousand years. For a millennium, Old Meskangela served as a language of public life and administration as well as a language of divine worship. The Classical Meskangela (also known as “Meskangela Proper” མསྐཾངེལཿ རནཏཾཀེ Meskangela Rántake) was a standardised dialect that emerged from Old Meskangela in approximately 300 BNE and remained spoken until the New Era, after which it remained only a written standard, as local dialects gained more recognition and prominence. Seven dialects are still spoken, but only the Southern variety diverges from the classical spelling and uses its own modified version of the Meskangēl script.

New Meskangela dialects are written in the Meskangēl script, a descendant of the Ancient Himalayan script, and the classical variety remains most prominent. The earliest inscriptions date from 13th century BNE, although such inscriptions remained scarce until approximately 500 BNE, when first religious texts were written. However, the linguistic history of Meskangela prior to the appearance of such textual sources remains unknown.

Name and history

The name of the language was coined during the classical period from the word སྐཾང་ skaŋ “mountain” and means “pertaining to the mountains”, since the land where it was traditionally spoken is mountainous. Other groups used different terms to refer to themselves and their languages: མྸཾཨྃལཿ mágailā “Southern”, སྱ྅རེལཿ syörilā “Western”, ཀལོནེལཿ këlónelā “Plain dialect”, ཁམེལཿ khëmelā “Coastal dialect”.

Historically Meskangela had always been the language of the mountainous islands. Its origin, however, is obscure, as all documentations of the previous eras were lost, and local folklore only briefly mentions an ancient journey to the west, called ཨཱགརྭཾཏ Āgërwat. Little is known about the language of that period itself, its phonology is the only part that is well understood, which allows to reconstruct many Proto-Meskangela words.

The period, named Old Meskangela, is marked by the reappearance of writing, although the records from that period are still rare. The earliest attestation of Old Meskangela dates back to approximately 2000 BNE and is an inscription on a temple on top of the Lhaidërmū Mountain (lit. “Pillar of the sky”). It is written in the early logosyllabic version of the Meskangel script and it reads: [PN] *imē-hwa wa-bu-*sVkwal, hwi-rai wa-ha-*kēn – “[PN] set a stream into motion, filled up a lake.” The first logogram likely represents a personal name (PN), the pronunciation of which is unknown. The syllables marked with an asterisk (*) were reconstructed. The corresponding Classical Meskangela sentence is: (PN)-is lwī sëkwaltekyi, eju dëgenkyi.

By around 1300 BNE, the Meskangel people had a many chiefdoms in all of the three main islands. During that period the written language rose to prominence and was standardised for the first time in 300 BNE. Later, Meskangela Proper became a prestige language after being adopted as a lingua franca between its various dialects, as well as between speakers of the former and the Ilain people. The Ilain languages adopted many loanwords from Meskangela, particularly its Southern dialects. The dialects themselves had already developed their distinctive features by the classical period, and Meskangela Proper was not a common ancestor of those dialects, instead it was a standardised variety of the Central Syörilā, which comprises Western and Eastern Tūŋëdēla (“Innersea”) group. During the New Era Meskangela is still often referred as a single language, even though by the end of the classical period it had already been a group of closely related languages. The most accurate term to describe Meskangela as a whole is a dialect continuum.

External relation

The only neighbouring family to Meskangela is the Ilain languages in the southeast. There is no agreement whether the Ilain family is genetically related to Meskangela. The common reconstructed features of Proto-Meskangela and Proto-Ilai are spurious. A large amount of words that seem to be cognates are likely borrowings into Proto-Ilai or individual Ilain varieties from Meskangela, and inconsistent sound correspondences support that idea. The similarity in grammar, especially verbal morphology, is also likely due to later Meskangela influence, rather than a genetic similarity between the two branches. Some ancient morphological elements, such as the nominalising suffix -in, agentive -uma/-upa, optative-imperative p- prefix, the negative-prohibitive particle ma; as well as some words belonging to core vocabulary: Il. cai – Cl.Mes. dzān “to eat”, Il. mecuki “full” – Cl.Mes. cyok “enough”, W.Il. līŋ “soul, heart”, E.Il paluŋ “stomach” – Cl.Mes. luŋ “heart” point out to a distant genetic relationship with a possible common ancestor breaking into the two branches no later than 8000 BNE.

The table below represents numerals in the two families along with the reconstructed forms for Proto-Meskangela and Proto-Ilai:

Numeral Proto-
Meskangela
Classical
Meskangela
Western
Meskangela
Eastern
Meskangela
Southern
Meskangela
Proto-
Ilai
Western
Ilai
Eastern
Ilai
1 *tyik dan, acyik tik acyik lăŋē *aˀtyi asa áci
2 *nis ni gënis nis gne aˀ(a)li āli ára
3 *sum sum gësum sum gsō *pei pei phe
4 *liH meli *liˀ rarí~lalí
5 *luŋa lëŋa eŋa łuŋa băŋa *lu pēli lu
6 *dVkruk khuk khuk lëruk luk *anöm enem anom
7 *sVni(s) sënis stris hëni enit *peˀtu pētu petho
8 *biryat bëryat, ryit bërkyat rit mejat meja mia
9 *kū dëkū, kwau hëkwa lagu *ciwa siva siwa
10 *cyai cyai gip līni si *tyāhi sahi cāwi
12 *nistVkruk nikhuk niskuk nirhuk niluk nileu phesî
20 *mVkul mëkul mëkul mëkū mukog *hul(i~u) huli mokō
60 *rVgya rëgai rëge rëja răsē pēhuli cāwia anom

Ilain word for “eight” (as well as Western Ilai nileu “twelve” and Eastern Ilai mokō “twenty”) are obvious borrowings from Southern Meskangela. The word *li “four” may also be an old borrowing, although Eastern Ilai lalí~rarí can be derived from “two pairs” (ra “pair” is likely present in *ára “two”). Other words, such as “one” and “ten” show possible cognation, and Eastern Ilai lu “five” being similar to first element in *lu-ŋa “five”, though the evidence is inconclusive. The similarities might have arisen from borrowing into Ilain from Meskangela (or vice versa) during prehistoric times.

Geographic distribution

The Meskangela language is spoken on three main landmasses, which are used to group its dialects: འས྅རིཀཿ Hasörikā “Westland”, སྱཾརལིངཿ Syarëliŋā “Eastland” and མྸཾགཾརིཀཿ Mágårikā “Southland”, as well as on many islands surrounding the landmasses. There are also some small communities in the southern continent Lyökimëranā, but most of those settlements are recent, and for the most part originate from the Southland.

Dialects

Seven main dialects of Meskangela:
  Outer Syörilā
  Inner Syörilā
  Western Tūŋëdēllā
  Eastern Tūŋëdēllā
  Outer Khëmelā
  Inner Khëmelā
  Southern
  sparsely populated or uninhabited

Linguistically speaking, Meskangela is not a single language, but a group of closely related vernacular languages, divided into three subgroups: Western, Eastern and Southern, based on their relative location. Some dialects within one subgroup may differ more from each other than other dialects belonging to different subgroups, which mostly depends on their geographic isolation and influence of the standard language. Some dialects are mutually intelligible, whereas others are not, not unlike the situation of a typical dialect continuum. Several varieties still use different names for themselves: a relatively divergent far Eastern variety is called Khīmła [ˈkʰiː.wɑ], Southeastern dialects – Majäg [ˈma.jɛɣ], Southwestern – Gakhō Łatem [ɣa.ˈkʰy. ʟa.ˈteʊ]. Most dialects can be described as either "Western", "Eastern" or "Southern", which corresponds to three main islands. It is also important to draw a distinction between later dialects of the New Era (often called "New Meskangela"), the classical variety used as a written, but not a spoken language, and those that are extinct. Thus, it is convenient to classify these dialects as "Modern", "Middle" or "Classical", and "Old", alongside the classification, based on geograhic areas.

Writing system

“Mëskangelā” in the cursive script.

The Meskangela script was based on the Ancient Himalayan script. With time, it developed its distinctive style and characters, specifically vowel markers, which were seldom used in writing Old Meskangela. The later varieties adopted this script with minor modifications. The other main writing system used for Meskangela, mostly after the classical period, was a cursive script.

Consonants
Aspirated Unaspirated Fricative Voiced Nasal Lateral
approximant
Glottal
letter sound rom. letter sound rom. letter sound rom. letter sound rom. letter sound rom. ལྷ​
Velar [kʰ] kh [k] k ཨ​ [ɣ~∅], *[x] g ​ག​ [ɣ] g ​ང​ [ŋ] ŋ [ɬ]
Palatal [t͡ɕʰ] ch [t͡ɕ] c ​ས​​ [s] s ​ཡ​ [j] j lh
Dental [tʰ] th [t] t ཞ​ [θ] z ཨ​ [ð̞~ɹ] d ​ན​ [n] n
Labial ཕ​​ [pʰ] ph [p] p ཧ​​​​ [h~ɦ], *[f] h བ​ [w] w ​མ​ [m] m [l] [a], *[aʱ]
Trill རྷ​​​​ [r̥] rh [r] r རྲ​​​ [d͡z] ʒ l a
Vowels and semivowels
vowel sound rom. vowel sound rom. vowel sound rom. vowel sound rom. vowel sound rom. vowel sound rom.
Short vowels ◌ི [i] i ◌ུ [u] u ◌ེ [e] e ◌ོ [o] o ◌ཾ [a] a [ə] ë
Long vowels ◌ཱི [iː] ī ◌ཱུ [uː] ū ◌ཻ [eː] ē ◌ཽ [oː] ō ◌ཱ [aː] ā ◌྅ [øː] ö
Diphthongs ◌ྃ [aɪ] ai ◌ྂ [aʊ] au ◌ཿ *[aʱ] ā ◌ྸ [˥] ◌́ ◌ྱ [ʲ] y ◌ྭ [ʷ] w

The table above demonstrates all characters of the Meskangēl script and their classical pronunciation. Individual dialects may vary greatly over the way certain characters are pronounced, and some dialects do not use certain letters, while others still retain the Classical Meskangela spelling, even though the pronunciation is different. For example, the Eastern variety does not use "◌ྂ" (au), replacing it with "◌ཱུ" (ū) everywhere, and uses "ཨ" to represent [l], while the Southern variety does not use letters "ཨ" and "ཞ​", the tone symbol "◌ྸ" and the "◌ཿ " ending and ignore those characters while reading.

Phonology

Each dialect of Meskangela has its own distinctive pronunciation, and it would not be feasible here to go into all these properties. Classical Meskangela has 60 distinct phonemes, while Old Meskangela likely had up to 66 phonemes. All later dialects have smaller consonant inventories, but some have more vowel phonemes, than the Classical variety.

Vowels

The table below represents vowels of Classical Meskangela:

Short Long
Front Central Back Front Central Back
Close i u
Close-mid e o ɵː
Open-mid ə (ɔ) ɑʊ
Open a ɑ(ː)

The vowel [ɔ] was an allophone of short /a/ in open syllables. Whether it contrasted with the open vowels remains a matter of debate, however, in most later dialects it became a separate phoneme. The same is true for [ɑ(ː)], the quality of which is not certain, because it did not merge with [ɔ] in the Eastern group, but likely was short. The exact quality of Meskangela diphthongs is uncertain as well, "ai" remains a diphthong only in the far Western dialects, where it is refected as [eɪ] or [ɛɪ], while "au" becomes [øʏ] in the Outer and [ɶ] in the Inner Western dialects, so it could have been slightly fronted or centralised in the proto-language.

The high long vowels are the most stable among the Meskangela dialects (with "ū" fronting to [y(ː)] only in the Inner Eastern variety), while the mid vowel group is the most variable. Eastern and Southern varieties exhibit metaphony affecting both short and long vowels to a different degree. The short vowels in the Southern dialect are frequently affected by syncope, thus creating new vowel contrasts.

Consonants

Labial Dental Alveolar Palatal Velar Glottal
plain labialised plain palatalised labialised plain palatalised labialised
Nasal m n ɲ ŋ ŋʷ
Plosive aspirated (pʰ) tʷʰ t͡sʰ t͡ɕʰ t͡sʷʰ kʲʰ kʷʰ
tenuis p t t͡s t͡ɕ t͡sʷ k
Fricative θ θʷ s ɕ h~ɦ
Approximant Voiceless ʎ̥ l̥ʷ
Voiced β~ʋ ð̞ d͡ʒʷ~r̝ʷ l ʎ j ɣ̞ ɥ w
Trill Voiceless r̥ʲ r̥ʷ
Voiced r
  • The exact value of a phoneme, denoted with the character "རྲ" is uncertain. It was likely a voiced affricate /d͡ʒʷ/ or a fricative /r̝ʷ/ (considering that it is represented with a modified letter "r") during the classical period. In later dialects it became [pʰ], [r], [θ] or [ʃ].
  • /pʰ/ was a marginal phoneme in Classical Meskangela, it became more common in the Western dialects and disappeared in Southern Meskangela.
  • /ɥ/ was a separate phoneme in Old Meskangela and had remained distinct from /j/ at least by the early classical period. In dialects with metaphony it likely survived longer, resulting in the rounding of preceding vowels.
  • The /s/ series likely had an aspirated allophone [sʰ] word-initially and between vowels. In the Eastern dialects it debuccalised to [ɦ] and had likely been pronounced this was already by the classical period.
  • The consonant /β/ was a separate phoneme from /w/ in Old Meskangela, but both merged into /w/ by early Classical period at least in writing, both being represented with "བ". The Southern dialects, however, preserved the distinction with /β/ becoming [b] initially and /w/ medially and the original /w/ disappearing: S. ē “to be” ← C. wai “to become”, but S. “to give” ← C. bëjan.

The aspirated consonant series likely developed from certain consonant clusters in Proto-Meskangela. Classical Meskangela allows very few initial consonant clusters, which may be expained by their merging into single consonants, thus making the aspiration contrast phonemic. The Inner Eastern dialect later lost this distinction, instead adding a high tone contrast to following vowels. Although a full set of aspirated consonants is shown in the table above, it was likely that some of these phonemes were marginal, appearing only in few words or under exceptional conditions. Certain morphological alternations gave rise to a contrast between plain and aspirated series (as well as voiced-voiceless contrast in the approximant series), but most dialects lost this feature mostly due to later morphological levelling and analogy.

Palatalised and labialised consonsonants were separate phonemes in Classical Meskangela, but the Western dialects lost the former series and most Eastern and Southern dialects lost the latter. Dental and alveolar consonants likely both had palatalised counterparts, although these two series merged in Classical Meskangela. The Western dialects, however, retain the distinction: tik ("one") compared to Eastern and Classical acyik.

Prosody

Classical Meskangela is a pitch-accent language. Its prosodic system is characterized by free accent. In lexical words, only one syllable can be tonically prominent. A heavy syllable – that is, a syllable containing a long vowel, diphthong, or a sonorant coda – may have one of two tones, level tone (unmarked) or rising tone (marked). Light syllables (syllables with short vowels) can only be marked (having hugh tone) or unmarked (having low tone, which is considered neutral). Stress is fixed on the root syllable, but words having more than three syllable receive a secondary stress. Such words follow a trochaic pattern, for example: སིནྣཾནངཾཏཾ sinnanëŋata [sin.ˈna.nə.ˌŋa.ta] “I have been reading it”.

The pitch accent remained contrastive in most dialects of Meskangela, as it was in the Classical language: ཅྱེལཾན​ cyelan [ˈt͡ɕe.lan] “to spread” ཅྱཱེལཾན​ cyélan [ˈt͡ɕe˥.lan] “to scold”, which also contrasts with ཅྱཻལ cyēl [t͡ɕeːl] “front”. The Southern dialects lost the pitch accent completely: both words became ཅེལེ cele [ˈt͡ʃe.lə] (however, the former is not used without a fused prefix གཆེལེ gchele [ˈt͡ʃʰe.lə] “to spread” which was an intransitive verb “to spread out”, but became ambitransitive – a characteristic feature of the Southern dialects).

Syllable structure

The typical Proto-Meskangela root syllable consisted of the following structural elements: an onset consisting of a root initial consonant Ci, optionally followed by a liquid L or semivowel glide G (either "j" or "w"); and a vocalic nucleus consisting minimally of a simple vowel V, followed by a final consonant Cf. The semivowels could also occur postvocalically, forming falling diphthongs in "-w" and "-j", thus belonging to the inventory of Cf. Unlike word roots, prefixes and suffixes followed a different pattern, consisting of a single consonant followed by a vowel PVp or SVs (in case of prefixes and suffixes respectively). Only root vowels could carry vowel length (:) and tone (t), the latter being a consonantal feature at the Proto-Meskangela stage. There was no contrast between zero-initial *VC and glottal-initial *ʔVC, in such cases the second variant is reconstructed with Ci being an obligatory element. of the root syllable. Two non-syllabic suffixes are reconstructible for Proto-Meskangela, *-s and *-n. When added they could have resulted in a forbidden postvocalic sequence of two consonants ( e.g. -Cfs or Cfn). For these specific instances the suffixes are instead reconstructed as əsS with "ə" being a short semi-syllabic element that disappeared in Classical Meskangela and later dialects. In other cases, where a single final consonant suffix is expected, a "hollow" consonant is reconstructed (which in some cases gives rise to tone Ht, or disappear without a trace H0), as in *rjaH0ən “to laugh”. In Classical Meskangela the suffix became phonetically identical the Cf, becoming a part of the root ( e.g. gësata*gasaH0t-ʔa “he/she kills”. Thus a potential fully inflected word consists of the following elements:

PVp—Ci—L—G—V(:)—Cf/H0/t—SVs—(əsS)

A word could have more than one suffix and prefix. In Classical Meskangela this system was altered, allowing open root syllables, as well as consonant clusters within syllables, where the first element was "s" followed by a consonant, usually an aspirated plosive, which could precede the third element "l". Individual consonants could be geminated depending on their position in the string of morphemes. The semivowel phonologically had become a part of the initial consonant or cluster by the classical period, but in the model it is still convenient to analyse it as a separate element. Certain suffixes underwent syncope of their reduced vowels, thus allowing more consonant clusters outside the root. Thus, during the classical period the inflected word typically followed this structure:

PVp—(s)Ci(:)—(l)—G—V(:/t)—Cf(:)—(Vs1)—S(:)(Vs2)

Later dialects generally follow the model above, modifying some individual elements, such as adding more permissible clusters, or merging the clusters into single consonants, thus retaining all the elements only nominally. This is especially true for the Southern dialects, most of which became fairly analytic and lost most of their suffixes and prefixes in the process, as well as tone and contrastive vowel length.

Grammar

In this subsection only the grammar of Classical Meskangela is discussed, considering the amount of variation among different dialect groups and uncertainty of the Old Meskangela morphological structure. Classical Meskangela as well as most of its dialects are agglutinative or weakly fusional, the individual morphological elements are not easily segmentable, due in large part to the presence of portmanteaux morphemes and allomorphy.

Although attempts have been made to reconstruct a quasi-regular “ablaut” system for Proto-Meskangela, the vowel gradation in the Classical Meskangela variety is sporadic and irregular, especially in case of open-syllable roots. Some verb conjugational pattern may be attributed to the Proto-Meskangela affixes that later merged with the verb stem, such as the "a-ö" alternation, which often shows up in the derivational morphology of Meskangela. Pairs, such as, khitan “to rub” and khutan “to scratch”, rum “darkness” and rim “evening” are relatively common with back vowels often representing a more "internalised" process or abstract phenomenon, than their front counterparts which are more concrete and "external". Certain prefixes have active and passive counterparts, such as si- (active transitive) and su- (mediopassive): sinnaŋan “to read (something)” and sunnaŋan “to read (in general)/ to be read” (hence Western sunnaŋ “book”, but Eastern abirai from bërëjan “to write”).

Nouns

The following parts of speech are viewed as nominals in terms of their morphology: nouns, pronouns and numerals. There are unambiguous morphological criteria for distinguishing between nominals and verbs. In Classical Meskangela, as well as in its later dialects, nouns can be marked for case, person (possession) and number. There is no grammatical gender in any Meskangela variety, instead nouns are differentiated by their animacy with humans, mythological and folktale figures being animate and everything else being inanimate; the animacy is not marked on the noun itself, but is instead reflected by verb conjugation. Nouns can be focused or neutral, depending on their importance in the conversation. Meskangela is a topic-prominent language, the topicalised nouns are often marked with demonstratives, and a topic is placed first in the sentence, for example: ལཾཧྸཾང་ཏྂ རྲཾནཾ མཾཏྭིབཾ laháŋ-tau dzana matwiwā “cooked rice is a delicious food” (laháŋ-t-au dzan-a matwi-wā “cooked.rice-Abs-that-Top food-Foc be.tasty-3sg.Inan”

Classical Meskangela has thirteen cases. Case endings are attached to nouns with or without the non-singular suffix. Allomorphy of case endings depends on whether the noun ends in a vowel or consonant. Below the case endings of the noun ཀྱིམ​ kyim “house” are presented:

Case Suffix Non-focused Focused Translation
absolutive -∅-a kyim kyima house (patient)
ergative -i-s kyimi kyimis house (agent)
instrumental -o-n kyimo kyimon by means of the house
vocative á- (-o) á kyimo oh house
locative 1 -na/-mpa kyimma kyim in the house (neutral/same level)
locative 2 -ta-u kyimëta kyimëtau in the house (at a higher level)
locative 3 -tuh(e)-u kyimtuhe kyimtuhu in the house (at a lower level)
comitative -d/-llō kyimëd kyimëllō with the house
ablative -gi-s kyimëgi kyimëgis from the house
essive -lwa-s kyimëlwa kyimëlwas as the house
allative -l-as kyimël kyimëlas to the house

The dual and plural numbers are denoted, when required, by adding the resepective suffixes: -seŋ for dual and daŋ for plural.

Case Dual Plural
suffix non-focused focused suffix non-focused focused
absolutive -s-a/e-ŋ kyimse kyimsaŋ -(e)d-a-ŋ kyimeda kyimëdaŋ
ergative -s-ī-s-ŋ kyimsi kyimissīŋ -(is)tī-ŋ kyim kyimistīŋ
instrumental s(ë)n-o-ŋ kyimsën kyimësnoŋ -(o)nī-ŋ kyim kyimonīŋ
vocative hí- (-o) hí kyimo í- -oda í kyimoda
locative 1 -sin/-pa-ŋ kyimsin kyimpaŋ -din/-pi-ŋ kyimdin kyimpiŋ
locative 2 -t(a)s-u-ŋ kyimëtas kyimëttuŋ -(e)tū-ŋ kyimetū kyimëtūŋ
locative 3 -ëth-(is)-u-ŋ kyimëthis kyimëthuŋ -(e)tud-a-ŋ kyimetud kyimëtudaŋ
comitative s-(i)d-aŋ kyimsid kyimissaŋ -(e)dal-a-ŋ kyimedal kyimdalaŋ
ablative -ski-ŋ kyimëski kyimëskiŋ -(e)kk-a-ŋ kyimekka kyimëkkaŋ
essive -si-lwa-ŋ kyimsilwa kyimsilwaŋ -(e)dz-a-ŋ kyimedza kyimëdzaŋ
allative -sl-a-ŋ kyimësla kyimëslaŋ -(e)ll-a-ŋ kyimella kyimëllaŋ

Possession is marked with possessive prefixes, attached to the possessed noun:

Prefix Singular Dual Plural
1st person ŋo- kas- kai-
2nd person na- cu- kye-
3rd person ka- khu- me(k)-
Inanimate a- hu- me(j)-

For example: མེཀྱིམཾ སྱེ mekyima sye “this is their house”; ནཾཁཾཏྟཾ སྤཾཏཾངེཧཿ nakhātta spataŋehā “we found your coat”.

Nominal derivation

The most common way to form new noun is derivation from verbs with special suffixes. Such forms are treated as nouns, but some may still possess verb-like qualities, such as conjugation. For example ལཏེམཾཏཾ​ lëtemata “sightful” is a verb with its infinitive being ལཏེསཾཏཾན lëtematan.

Suffix Category Example
-t Deverbal action or process nouns ལཏཾཏ​ lëtat “sight” ལཏེམཾཏཾ​ lëtemata “sightful” ལཏཾསཏ​ lëtasët “appearance (poetic)”
-mā Resultative and manner nouns ལཏཾམཿ lëtamā “scene” ལཏེཏཾམཿ lëtetamā “observation”
-s Object nouns ལཏཾས​ lëtas “glass” ལཏཾམཾས​ lëtamas “sign” ལཏཾཏས​ lëtatës “character”

To create new agent nouns Meskangela has various agentive suffixes. The most common are: -is (bërëjis “writer” from bërëjan “to write”), -cis (lëtacis “witness” from lëtan “to look”), -ikā (punikā “carrier” from punan “to carry by hand”), -wa (biliŋwa “forest dweller” from biliŋ “forest”), -pa (këmōpa “worker (Western)” from këmowan “to work”). The instrumental suffixes -(a)zā and -r can also possess agentive meaning (khazā “walker” from khadan “to walk”).

Deixis

Topographical deixis refers to a variety of spatial-environmental deixis, in which typically distal reference to entities is made in terms of a set of topographically-anchored referential planes: most often, upward, downward,or on the same level. Topographical deixis is a pervasive feature in Meskangela dialects. The primary frame of reference of deixis in Meskangela is egocentric, in that the speaker is a deictic centre. It manifests in the form of demonstrative pronouns, motion verbs modifiers and certain nominals, most commonly locatives.

Demonstratives are the most common category of topographic deixis in Meskangela. The spatial orientation is defined in three dimensions: distance from the deictic centre (or depth), position upward or downwards from said centre (verticality) and position relative towards a certain geographical feature (horizontality). These demonstratives transparently derive from the motion verb modifiers, for instance, the upward-position modifier twa- comes from the verb twajan “go up, ascend”.

The table below shows the simplest demonstratives in two dimensions: depth and verticality. The third dimension is added as a suffix to the demonstratives, and are highly varied. The most common are topographic features: -mo “surface”, -(s)ki/-gi ‘’hill, slope’’, -le “river, valley”, -ju “bank, shore”; focus: -no/-nū “already mentioned”, -kha “any, newly introduced, one of several”; position above or below: -tu/-syu “on top, above, overhead”, -ma “below, under water”; side, relative to the speaker: -bi(j) “left”, -ra “right”, -(së)ŋa “front”, -phir “back, behind”. Southern Meskangela has two additional widely used deictic modifiers: -naŋ “inside”, phënaŋ “outside” that are present in other dialects as temporal deictics: saŋas “before”, phis “after”.

Proximal Mesioproximal Mesiodistal Distal
Up tut twadi twas tup
Down jaut jawi jause jopa
Same level ta(u) dëgi segi pagi

These demonstratives frequently serve role of endophoric reference and act as topic markers.

Evidentiality

Evidentiality in Classical Meskangela is an optional category that indicates evidence for a statement. This category is indicated with evidential particles, or copulas. The unmarked verb is assertive – representing a simple fact or general truth. The corresponding copula is , the negative copula is mīn/mīd, likely from *ma-rijan “being absent”.

A similar copula uré and its negative counterpart mórid are basic locative copulas that mark possession (maroŋ uréŋ “I have a cat”) or location (tau uréŋi “we are here”).

The postverbal particle lak is testimonial, it represents witnessed, first-hand experience. It also acts a locative copula with the meaning “there is”. Its negative form milak has only a locative meaning.

The paticle and its negative counterpart mód are egophoric copulas. It expresses a speaker’s acquaintance with something. On some occasions, it represents an intention of the speaker nalha-ta dihö “have some tea, which I made”. The locative copula marks possession (“I have”) or location (“I am at”) of the speaker and has no negative counterpart. In Classical Meskangela it expresses the speaker’s opinion towards a certain statement: ta lyaŋa wö “it is nice, in my opinion”.

“so, it seems” is a very common postverbal particle that expresses inferred or assumed information or probability of an event. It is mainly used with irrealis verbs for emphasis, for example: dispha lú “it is likely going to snow”. This particle often indicates that an event will happen in the future: gëspha lú “it will snow (eventually)”.

Verbs

Classical Meskangela has five verb classes, based on their stem ending:

  • class I: stem ends in a vowel (including diphthongs) – ལྭའཾན།ལྭའཾཡཾན lwān/lwajan “to be easy”, བཡཾན bëjan “to give”;
  • class II: stem ends in -n – ཪྱུནཾན ryunan “to flow”, ཁཾནཾན khanan “to shine”, ཝིནཾན།བིནཾན​ wīnan “to be far”;
  • class III: stem ends in -l – སྟཾལཾན stalan “to acquire”, ཟཻལཾན zēlan “to be surrounded”, སཀིལཾན sëkīlan “to bind”;
  • class IV: stem ends in -s or -r – གུནཾསཾན gunasan “to rest”, ཁོརཾན khoran “to cry”, བསྐྱུརཾན bëskyuran “to interpret/to translate”;
  • class V: stem ends in other consonants – ཅྱུཀཾན cyukan “to trust”, ལཏཾཏཾན lëtatan “to observe”, ཆོཏཾན chōtan “to be pierced”.

Each class has its own conjugation pattern.

Stem

The verb stem includes the basic root as well as optional affixes and auxiliaries. Certain nouns or verbs can be incorporated into the stem with the main verb following the incorporated part. Negation is also marked inside the stem with an infix -ma-, which appears either after the prefixes (e.g. ཪཾམཾཏུལྷ ramatulh “do not steal”, or between two consitual roots of the stem (ལརྲཾམྨཾཏེངིཏཾ lëdzammateŋitā “it was given back to the owner”), and is the only infix besides the old progressive infix -en-, which was substituted by the bi- prefix in all but few irregular verbs. The dictionary form has the ending -an and is considered the infinitive of the verb.

The most common Classical Meskangela prefixes and suffixes are listed in the table below:

Prefixes Suffixes
prefix Ci- V- suffix class I class II class III class IV class V
Future di- d- Imperative -∅ -no -h -os -o
Optative (tr.) bo- p- Optative (intr.) -˥s -wod -od -sos -os
Terminative
Perfect
N- m- Adverbial
Participle
-sē
Continuous bi- b- Stative -z -cya -ha -ta -ha
Inchoative ha-/*pa- h-/*bam- Cislocative -ruŋ -oŋ -huŋ -ruŋ/-yuŋ -ruŋ
Middle voice më- m- Translocative -soŋ -noŋ -loŋ -soŋ -hoŋ
Iterative mih- m- Supine -mi -umi
Detransitive g- Adjectival
Participle
-d -ëd -ud
Causative së-/s- s- Progressive -*jen -en *V-nCfe
Directive ra- r- Non-volitional -tha -tha -utha

The deictic suffixes attract the negative infix, indicating that they might have originated from an incorporated element: ཁཾམཾསོངིཀྐཾ khamasoŋikka “he didn’t walk away from me”.

The causative, detransitive and iterative prefixes are used mostly for verb-derivation (the iterative prefix lost its conjugational meaning in the Western and Southern dialects, where it is only used for derivation). These prefixes often create a string of derived forms of simple root verbs, for example: koŋan “to carry” → goŋan “to wear (to carry on ones’ body)” → skoŋan “to dress” → mikhoŋan “to continue (to carry on)” → sëmikhoŋan “to prolong”; koran “to turn” → goran “to be late” → *skoran “to cause turning (verb is possible, but not used)” → sikoran “make it turn around” → sukoran “to turn around” → misphikoran “to overturn”.

The derivational prefixes can change valency of the verb, and the marking of arguments. The prefix thus determines the role of all arguments in the sentence alongside verbal agreement:

Agent prominence
Rëmin-is=sëgi rossē bërëlai-kyi partës-pā parcis-ë-pagi.
man-ERG.FOC=that.PROX apple.ABS buy.PT-3.sg.AG market-LOC1.FOC trader-ABL-that.DIST
"The man bought an apple from a trader at the market."
Patient prominence
Rossej-a=sëgi rëminë-las a-brulai-kka-hā partës-pā parcis-ë-pagi.
apple-ABS.FOC=that.PROX man.ALL 3sg.IND.OBJ-DETR-be.bought.PT-3.sg-PAT.FOC market-LOC1.FOC trader-ABL-that.DIST
"The apple was bought by a man from a trader at the market."
Locative prominence
Partës-a=sëgi rossē a-rgëbrëlai-kka-pā rëminë-las parcis-ë-pagi.
market-ABS.FOC=that.PROX apple.ABS buy.PT-3.sg.AG-LOC.FOC man-ALL trader-ABL-that.DIST
"The market was where a man bought an apple from the trader."
Oblique argument prominence
Parcis-is=sëgi rëmin-a=pagi bësmërëlai-kyi partës-pā rossej-o.
trader-ERG.FOC=that.PROX man.ABS.FOC=that.DIST CAUS.buy.PT-3.sg.AG market-LOC1.FOC apple-INST
"The trader made the man buy an apple at the market."

The first sentence is considered the most semantically neutral, the latter three sentences are not commonly found in colloquial speech. Later dialects generally dropped the valency changing operations as a mean to mark argument prominence, relying instead solely on topic-focus markers.

Intransitive verbs

The table below represents personal endings of intransitive and reflexive verbs.

Simple Nonpreterit Preterite
singular dual incl. dual. excl. plur. incl. plur. excl. singular dual incl. dual. excl. plur. incl. plur. excl.
1st person -ŋa -ci -ce -ŋi -ŋe -iŋka -isti -icce -inkyi -inkye
2nd person -na -nsi -ni -inta -issyi -intyi
3rd person -ka -ŋki -kyi -ikka -iski -ikkyi
Inanimate -(w)ā -swe -(w)ai -itā -istwe -itwai
Reflexive Nonpreterit Preterite
singular dual incl. dual. excl. plur. incl. plur. excl. singular dual incl. dual. excl. plur. incl. plur. excl.
1st person -siŋa -sica -sice -sijaŋ -sijeŋ -siŋka -stasa -stise -staŋi -stiŋe
2nd person -sine -sissi -sine -siste -stasi -stani
3rd person -sai -ska -skyi -stai -staka -stakyi
Inanimate -swā -swīs -saja -stwā -stwīs -staja

Transitive verbs

The personal suffixes of transitive verbs differ from the intransitive verb paradigm, as it incorporates both subject and direct object of a clause (as well as the indirect object in case of ditransitive verbs, in which case it is represented with a prefix). The table below represents the general structure of transitive verbs:

Indirect Object Negation STEM Agreement Suffixes Comitative
p2 p1 0 s1 s2 s3 s4 s5 s6 s7 s8 s9
1sg.
ka-
1du.
kha-
1pl.
kyi-
negation
ma-/∅-
stem nonpreterit
-∅
1→3.NPT
-ŋa
1du.incl
-is/-s(i)
1Pat
-eŋ/-n
3pl
-ū/-*wi
3Ag
-kyi
1pl.incl
-(ŋ)i
3Pat.focus
-hā
1sg.
-laŋ
1du.incl
-lhaŋ
1du.excl
-lheŋ
1du.incl
-liŋ
1du.excl
-leŋ
2sg.
ta-
2du.
sta-
2pl.
cyi-
preterit
-∅/-t(e)/-Cːf
1→3.PT
-taŋ
1du.excl;2du
-(i)sya
1pl.excl
-(ŋ)e
Instr.focus
-on
2sg.
-len
2du.
-lhen
2du.
-lin
3sg.
a-
3du.
o-
3pl.
ā-
2→3.NPT
-na
2pl
-ni
Loc.focus
-pā
3sg.
-lo
3du.
-lho
3du.
-lai
2→3.PT
-nta
1du;2du.Pat
-es
1→2.NPT
-nya
1→2.PT
-kta

Verbs do not fill every slot of the table, a typical transitive verb have two or three slots filled at a time, for example: སླིཀཾ ཀུཀནཾ ཀོངསྐྱི slika kukëna koŋëskyi “they two carry fruit in baskets” (sli-ka “fruit-Pl” kuk-na “basket-Loc” koŋ-s-kyi “carry-Du-3Ag”); ཀཾམཱིཀྐྱི kamīkkyi “he has given it to me” (ka-m-bī-t-kyi “1sg.IndObj-Perf-give-Pret-3Ag”).

Commitative markers are used to mark two subjects of the same verb: Nicwatis Gësalpis girālkyilo – “Nitswata and Gesalpa are fighting (together)”; or to indicate an action that is reciprocal: Nicwata Gësalpa girālsailo “Nitswata and Gësalpa are fighting each other”. Since both arguments are equal, neither can be a direct object, even though the verb girālan “to fight” is transitive. In the first case the direct object is omitted, yet implied, while in the second case the verb becomes reflexive.

Syntax

Meskangela is an ergative language. Grammatical constituents in most Meskangela dialects broadly have head-final word order (SOV, or "subject-object-verb"). There are some general tendencies:

  • adjectives generally follow nouns in Classical Meskangela. In the Southern dialect opposite is true, unless the two are linked by a genitive particle e;
  • objects and adverbs precede the verb, as well as adjectives in copular clauses (which are stative verbs morphologically);
  • marked (focused or topicalised) nouns precede the unmarked nouns. The Eastern dialects often do not follow this rule;
  • unless topicalised, oblique nouns tend to follow both agent and patient: ནིཅྭཾཏཾལ མརཱིངི གསཾལཔཾལ ཨཾགྂཏཾང Nicwata Mërīngi Gësalpal agautaŋ “Mering called Nicwat to Gesalpa”. Only the Western dialects strictly follow this rule;
  • subordinate clauses generally precede the main clause in Classical Meskangela and most later dialects, apart from the Western group, in which both subordinate and relative clauses are internally headed and long-distance dependent clauses are common;
  • demonstratives and numerals follow the noun they modify. In the Western and Eastern groups demonstratives usually precede the noun, if another modifier is present;
  • in attributive clauses the head is followed by its atribute, unless the latter is topicalised;
  • in adpositional phrases postposition follows its noun;
  • interrogative clauses are signalled by the interrogative particle mai (Western lai), interrogative pronouns: kha “who”, kai “how”, ba “what”, kaima “when”, baima “where”; or the emphatic particle ŋé, which follow the head of the clause: མཾཔཱཡིནཏཾཔཿ་ངྸེ mapājintapā-ŋé? “you didn’t go there, did you?”;
  • the particle tai (and its synonym lhot, more typical to the South) are used to set off qoutations, they follow the phrase they modify: kyénna-tai cyenikka “he said "you are pretty"”. Eastern dialects use a quotative particle cai instead, which might have been an allomorph of Western tai;
  • the negative particle mu/mau precedes its modifying noun and attracts possessive prefixes: ངོམུ་ཀྱིམཾ ŋomu-kyima “not my house”. When the negative particle acts as a predicate, it follows the noun instead: ངོཀྱིམ མྂ ŋokyim mau “this isn’t my house”. In the Eastern dialects the negative predicate is substituted with a verbal negative infix, which precedes its noun.

The example below shows some differences between the three main dialect groups:

  • English: “I saw the bird that was singing on a tree”.
  • Classical: མསངཾཀ་ཏྂ སིངཏཾ བཾ་བིམྨཀའཾརིཀྐཾ མརཾནཏཾང​ mësëŋak-tau siŋëta wa-bimmëkārikka mërantaŋ (bird=TOP tree-LOC2 REL=CONT-sing-3sg.PRT see-1→3.PRT).
  • Western: མརཾནིཀྐྃ ཏའིབཾ ཨེ་བིམྨཀའར྅ཀ ཐཾངྟཾ་པཾ mëranikke tîwa e-wimmëkārök thaŋta-pa (see-1sg.PRT bird-DEF REL=CONT-sing-3sg.PRT tree-LOC2=on).
  • Eastern: ཧིནྟུ མརེམྨཾ་ཏྂ ཨཀྱཽརོད​ ཧེནྟཾན hiŋtu mëremmå-tā̊ akyōrol hentaŋ (tree-LOC2.FOC bird=TOP 3sg-sing-APRT see-1→3.PRT).
  • Southern: ངི སྱེན་པཾཏ ཨངྐོཏ མསྔཾཀ་ཏཾ མྱཾནྟ ŋi sjeŋ-pat ăŋkōt măsŋak-tå mjant (I.ERG tree=on-LOC sing-APRT bird.ABS=TOP see-PRT).