I recently came up with a conlanging idea I don't really know how to continue developing. Basically, either nouns and verbs have an unaltered form, and undertake reduction into a contracted form when inflected (though it has already other uses, not just any sort of inflections).
In verbs, this may result into a perfective vs imperfective dual root system appliable to almost every way.
There is a big problem though. The language's phonoaesthetics look like the eroded result of a whole loot of vowel loss, consonant gradation, metathesis, stress and vowel alterations, etc. Which results on the conlang's phonotactics to look halfway between Georgian and Biblaridion's Edun (https://m.youtube.com/watch?v=y9K1gegu-vg
). Here's an example on two words which had a common incorporated element in the proto-language:
(lit. "High-place" (mountains, highlands)) and kethaana
(lit. "person of the heights" (highlander)).
Then dental continuants turned into alveolar sibilants before palatals and remained the same everywhere else; Ksaanu, kthaana
Sibilants metathesized, and later became postalveolars when preceding velar stops;
And finally, vowel loss in unstressed syllables;
By the way the second word is the endonym for the speakers of this language, which I'm naming Ketanian for now. And since their endonym and their word for "mountain" are etymologically related, it's logical for me to think that my fictional speakers will inhabit highlands or mountains.
Returning to the original topic, the most reliable explanation to why do most inflected words in the language take a contracted root might be stress; a strict stress patern may cause stress in a word to shift when an affix is added, perhaps in reduplication as well (although that wouldn't mean much for non-monosyllabic words), so that, in later stages of the evolution towards the modern lang, vowel loss in unestressed syllables might cause the vowel in the root to disappear. For example:
(I don't know which one is easier for me to pronounce, since the first one might involve a syllabic velar nasal and the second one contains its preglottalized voiceless equivalent) "mountain.ERG".
As for verbs, roots like dak
"to hit" get reduced to -thk
-, which is already so difficult to pronounce that it might metathesize into kth-
if there's an absence of a preffix... It's really confusing, I don't know what's easier to pronounce for me, so discerning among different options resulted from evolution is a nightmare. And I don't even know if any natural language features an even slightly similar system, so that I can't figure out whether these features can pass as naturalistic or not. Phonotactics and the phonoasthetic look of the mothern language become a real problem regarding the fact that I even want postpositions to get incorporated to inflected nouns, so what do you think?