|Region||Ekuosia, Puzimm, and Lahan|
Proto-Ekuo-Lahiri, often abbreviated PEKL or PEL, is the reconstructed Proto-Language of the Ekuo-Lahiri language family, which arose around BCE 6500 in the Kasingadh Lakes of north-central Ekuosia, and spread across much of Ekuosia, through Puzimm, and into Lahan. Ekuo-Lahiri is one of the widest-spoken language families on Sahar with over ?00 million native speakers of its various languages, the most populous being Neviran.
PEKL had a fairly small inventory, especially its vowels, which formed a three-height vertical vowel system with no back/front or rounding contrast — although it possibly featured allophonic rounding in at least some situations.
Since the 1760s when EKL was first theorized as a family there have been attempts to reconstruct the parent language. The first and most widely-accepted reconstruction was forwarded primarily by Sendap Atirewwa and Nain Ashedeg in 19??.
|Plosive||p b||t d||k g||ʔ¹|
¹ Only some scholars support the presence of */ʔ/.
² */ʍ/ is broadly accepted to be a fricative or approximant with rounding influence that was distinct from the also-accepted */w/. The most common interpretations are /ʍ/ or /xʷ/ due to its reflexes as velar sounds (/x k/) in some daughter languages, but some scholars also posit /ɸ~f/.
Linguist Qirouhem Kozmek in 19?? posited a more robust inventory with consonant coarticulation series, principally to allow for simpler syllable construction in proto-words, reducing /wj/ and /jw/ clusters.
|Plosive||pˠ bˠ||pʲ bʲ||tˠ dˠ||tʲ dʲ||kʲ gʲ||k g|
|Fricative||fˠ||vʲ ³||sˠ||sʲ ¹||ç ¹||x||h|
¹ Most linguists only assume the presence of one member of each of these pairs.
² The presence of a velar nasal is not widely accepted.
³ */vʲ/ is also sometimes considered /fʲ/ or /ɥ/.
Other scholars have posited similar coarticulation, sometimes substituting Cˠ with Cʷ, or including a third (plain) series as well. Several other reconstructions exist but few are widely-accepted among linguists. Notably, some expand on the Kozmek Reconstruction by adding a third, plain series in addition to the palatalized-velarized contrast.
As with the consonants there are alternative reconstructions for the vowels, the most popular being a two-tier system with three length gradations (/ɨ ɨ: ɨ:: ä ä: ä::/), and several where depth and rounding are phonemically contrastive, but these are less widely-accepted.
The Practical orthography is typically just the IPA, with the exception that stressed vowels are written as í é á.
PEKL allowed moderate to complex syllables depending on the reconstruction method used. The classical reconstruction allows for at least CCCV:V:CC syllables with clusters of four consonants (or more) at syllable boundaries; the Kozmek reconstruction simplifies this to CCV::CC. Both forms allow null onsets and codas, therefore making cross-syllabic diphthongization possible.
Stress was distinctive in PEKL, most strongly reflected in consonant gradations in the Halarian branch. The majority of words had final stress, but it could occur anywhere in the word.
Like the majority of its daughter languages, PEKL is posited to have used an animacy hierarchy. At the least robust it would simply have been a noun classification system, and at its most it would resemble that of the Adzo-Neviric branch, using the animacy hierarchy as the basis for word order and thematic role assignment.
The exact number and formalization of animacy categories is unclear; they likely developed from simple classifiers in an earlier stage of the language. The typically-accepted hierarchy is: Sentient, Sacred, Higher animals, Lower animals, Moderate, Instrumental, Natural, Construct, Location, and Uncouth. The Sacred and Sentient tiers may have occurred in either order. The presence of many additional tiers in the Lahiri languages indicates that there could have been a much more expansive hierarchy/ classification system.
PEKL seems to have had fairly bare nouns and verbs, or to have only had very weak affixes that were lost in all but some irregular descendant words. Instead words were marked by a complex series of clitics that could clearly occur in at least some different orders. For example, the animacy/number clitics developed into suffixes in Adzo-Neviric, but prefixes in Lahiri.
Nouns were also marked with articles (definite, indefinite, and construct state). The order of the nominal clitic and the article also seem to have varied.
Verbs were marked with a variety of auxiliaries and clitics. The auxiliaries indicated tense/aspect/location, the clitics negation and modality. Voice seems to have been marked with a suffix.
The major word order was free, usually SVO or SOV, but in OS-ordered phrases a special 'nominative article' was used to mark the agentive noun. This later became an inverse marker on the verb in Adzo-Lahiri.