Ah ok that makes sense now;)
22.02.2026 16:24 โ ๐ 2 ๐ 0 ๐ฌ 0 ๐ 0@learnani.bsky.social
Script and developing conlang, by @kannydennedy.bsky.social. https://ani-language.info
Ah ok that makes sense now;)
22.02.2026 16:24 โ ๐ 2 ๐ 0 ๐ฌ 0 ๐ 0Cool. Wouldnโt it be โmake 3pl machineโ though? Or I have that wrong
22.02.2026 12:31 โ ๐ 2 ๐ 0 ๐ฌ 1 ๐ 0I recently read (ok, mostly skimmed) a pretty dense book called โClassifiersโ by Alexandra Aikhenvald. If I took one thing away, itโs that there are many ways to do classifier systems. I think if itโs fun and makes sense to you, and you can stress test it and it holds up, then itโs probably good!
15.02.2026 17:54 โ ๐ 1 ๐ 0 ๐ฌ 1 ๐ 0I think it would also depend on the way plant words are constructed. Like if rose is plantClass-blood, then swapping the affix would be confusing. So maybe if someone is called โRoseโ, maybe just the grammatical agreement around it changes to animate, but the word itself keeps the plant affix.
15.02.2026 12:17 โ ๐ 3 ๐ 0 ๐ฌ 1 ๐ 0Good question. I only know one language with classes (Kunwinjku), and Iโve not heard of people being named after plants. I feel like nicknames based on doing things with plants would be more common. Eg plantClass-bla = log, femaleClass-foot-plantClass-bla, โthat woman who stubbed her toe on a logโ ๐คท
15.02.2026 12:03 โ ๐ 1 ๐ 0 ๐ฌ 1 ๐ 0Yes I think you can probably allow wild allophonic variation of vowels, or consonants, but not both, otherwise you reduce your available sounds drastically ๐ Sounds like a good idea with the letters. Iโm thinking of just doing little diacritic for voicing like Japanese, since itโs not phonemic.
14.02.2026 13:22 โ ๐ 2 ๐ 0 ๐ฌ 0 ๐ 0Interesting:) I was thinking actually to have no voiced/unvoiced distinction in consonants, so people could freely express with that. Eg if โtekโ is โtapโ, then teg/dek/deg could be onomatopoeic variations depending on the force of tapping. Maybe more free expression in languages is the zeitgeist.
14.02.2026 06:34 โ ๐ 2 ๐ 0 ๐ฌ 1 ๐ 0Pentagonal tile pattern on the Ulugbek Madrasa, Samarkand. Photo by the author
I tried coming up with words for a while, and I canโt help thinking in patterns. Itโs very hard to believe that โkishโ and โkashโ for example would not be different apophonic grades of the same word. E.g. blood/bleed, sit/seat in English. With 5 vowels and monosyllabic roots thatโs 5 grades per set?
13.02.2026 12:05 โ ๐ 5 ๐ 0 ๐ฌ 0 ๐ 0How many options-ish are you thinking for the options created by the vowels? If that makes sense. Iโm thinking something similar, but Iโm only doing (C)V(C) with 5 vowels, so there are only 5 possibilities there
13.02.2026 11:46 โ ๐ 1 ๐ 0 ๐ฌ 1 ๐ 0Nice. It looks like the city that shares my name
13.02.2026 06:27 โ ๐ 1 ๐ 0 ๐ฌ 0 ๐ 0Floral tilework on a madrasa in Samarkand
I like the idea of having a limited vocabulary, which forces metaphor. #TokiPona is a little extreme for me, but something like ~500 base words? An approximately (C)V(C) structure with 10 consonants & 5 vowels gives around that number, a neat 10*5*10. Iโve heard the grid shouldnโt be full, butโฆ ๐ซฃ
06.02.2026 12:29 โ ๐ 3 ๐ 0 ๐ฌ 0 ๐ 0Ani is back, coming to you from the ancient city of Khiva. New version in development.
This new version will be a real #conlang, not only a script. Iโm thinking the back story might be a kind of ancient Silk Road auxlang.
Mostly apriori, but of mysterious origin?