[YG Conlang Archives] > [Latejami group] > messages [Date Index] [Thread Index] >
Lirakdom uses triliteral roots. Core roots have the shape CCC while secondary roots have the shape CCCVC. Words of the form CVCVC are verbs and CVCCVX are other parts of speech, where X is the part of speech consonant. The first vowel is a matrix which indicates dynamics and +/- focus: i y u = no focus e a o = focus i e = static u o = dynamic And the second vowel indicates the other semantic roles: e = 0 a = P i = A o = A/P u = AP y = do not apply Voices, as well as non-linking argument structures, are marked with additional affixes though. A single transfix combined with a normal argument structure gives the non-linking argument structure; P + this affix = 0/P. -- Veoler (Soon will this list be able to parse the word "lirakdom" itself) The Shadow wrote: > For quite a while I've had it in my mind to produce a language based > more or less on Latejami grammar (which impresses the heck out of me) > but using triliteral roots. (And perhaps some longer and shorter > roots as well.) > > I think the triliterals will have several advantages. By implementing > the verb types (ie, A/P/F-s and so on) and probably also the most > common voice changes as infixes, I hope to increase brevity and get > rid of memorizing default verb types. Basically, every verb will be > marked for type. > > I've been playing around with it for some time, but haven't yet > settled on the best way forward. Time is an issue. >