Nisha historically had many consonant clusters just as Latin had, since it is basically a rip-off. Since the beginning of development many consonant clusters disappeared by inserting vowels and in the latest and current stage palatalization set in in order to get the typical CVCVC behaviour. Palatalization is fairly common when /n/ appears before another consonant, i.e. /mundus/ > /mʲud/Chengjiang wrote:@mezziah: What's the distribution of the palatalized consonants? Are they more common than non-palatalized consonants before front vowels but less common elsewhere, to use a fairly common pattern? Do they contrast with sequences of a consonant and /j/?
Also, I'm kind of curious about something. Contrastive palatalization is consistently absent on voiced obstruents. Is there a historical reason for this, by any chance? I say contrastive palatalization because I don't know if, say, /b d g ʒ/ have palatalized allophones.
I never had the need to palatalize voiced consonants so I left them out... that is all. For some reason this kind of development leaves palatalized vowels and I'm struggling with the phonetic realization of them. Maybe they'll ultimately turn out to be lengthened vowels and/or diphtongs /aʲ/ > /ai/, /uʲ/ > /ui/, /iʲ/ > /i:/, /æʲ/ > /æi/
Also, thanks for your romanizations, they're quite inspiring.