Evolution of Kebreni
Posted: Fri Sep 27, 2024 5:07 am
The recent discussion of Kebreni literature has reminded me of a long-standing issue I have with the language: how, precisely, did such a bizarre verbal system evolve?
To substantiate that word ‘bizarre’, let’s review the Kebreni verbal system. A Kebreni verb root, as I understand it, is composed of several consonants plus at least vowel, thus:
√kan ‘see’
√dir ‘work’
√kuls ‘command’
√pabad ‘kick’
The first vowel can appear before the root or in the first syllable, and similarly the last vowel may appear in the last syllable or after the root. Thus we get:
abebad-u (VOL.IMPF), pabadu (IMPF), pabuda (PFV)
For a root with a single vowel, these processes collectively allow the root vowel to appear in any position whatsoever relative to the consonants:
agenu (VOL.IMPF), kanu (IMPF), kuna (PFV)
ideru (VOL.IMPF), diru (IMPF), duri (PFV)
ugelsu (VOL.IMPF), kulsy (IMPF), kyluste (SUB.PFV), kylsu (PFV)
In these words we see two other vowels. u is an obligatory vowel denoting benefactive state (alongside alternations of the root vowels), while e denotes volition. The grammar analyses the volitive as an underlying prefix triggering obligatory metathesis — but, because perfectivity also triggers metathesis, the so-called prefix can surface as a suffix (agun-e, idur-e, uguls-e). The most consistent marker of volition is therefore the presence of a root vowel before the first consonant.
Meanwhile, the right edge of the root shows metathesis, not only for perfectivity, but also for subordination. And these aren’t the same process: they differ in how they treat root-final consonant clusters. Perfectivity metathesis is particularly strange, because it operates on non-adjacent segments (whereas the other processes can be analysed as operating on CV or VC substrings).
The resulting verbal system is — to put it mildly — considerably more complex than any natlang system of metathesis I’m aware of. For monosyllabic roots the result looks almost templatic, with vowels being inserted within consonantal skeletons, but longer roots show clearly that the effects at the left and right edge are separate.
My question is, therefore: how did this system evolve? (Either in Kebreni or in Proto-Monkhayic.) I’m willing to accept that a human language could have such alternations, but I struggle to imagine any pathway by which they could plausibly evolve.
To substantiate that word ‘bizarre’, let’s review the Kebreni verbal system. A Kebreni verb root, as I understand it, is composed of several consonants plus at least vowel, thus:
√kan ‘see’
√dir ‘work’
√kuls ‘command’
√pabad ‘kick’
The first vowel can appear before the root or in the first syllable, and similarly the last vowel may appear in the last syllable or after the root. Thus we get:
abebad-u (VOL.IMPF), pabadu (IMPF), pabuda (PFV)
For a root with a single vowel, these processes collectively allow the root vowel to appear in any position whatsoever relative to the consonants:
agenu (VOL.IMPF), kanu (IMPF), kuna (PFV)
ideru (VOL.IMPF), diru (IMPF), duri (PFV)
ugelsu (VOL.IMPF), kulsy (IMPF), kyluste (SUB.PFV), kylsu (PFV)
In these words we see two other vowels. u is an obligatory vowel denoting benefactive state (alongside alternations of the root vowels), while e denotes volition. The grammar analyses the volitive as an underlying prefix triggering obligatory metathesis — but, because perfectivity also triggers metathesis, the so-called prefix can surface as a suffix (agun-e, idur-e, uguls-e). The most consistent marker of volition is therefore the presence of a root vowel before the first consonant.
Meanwhile, the right edge of the root shows metathesis, not only for perfectivity, but also for subordination. And these aren’t the same process: they differ in how they treat root-final consonant clusters. Perfectivity metathesis is particularly strange, because it operates on non-adjacent segments (whereas the other processes can be analysed as operating on CV or VC substrings).
The resulting verbal system is — to put it mildly — considerably more complex than any natlang system of metathesis I’m aware of. For monosyllabic roots the result looks almost templatic, with vowels being inserted within consonantal skeletons, but longer roots show clearly that the effects at the left and right edge are separate.
My question is, therefore: how did this system evolve? (Either in Kebreni or in Proto-Monkhayic.) I’m willing to accept that a human language could have such alternations, but I struggle to imagine any pathway by which they could plausibly evolve.