[YG Conlang Archives] > [romconlang group] > messages [Date Index] [Thread Index] >
On 2011-08-12 00:38, davidjohnmccann wrote:
--- In romconlang@yahoogroups.com, "thomasruhm"<thomas@...> wrote:There is something else, I am wondering lately. Did 'j', I mean like in 'jocus' develop quickly as quickly to 'dZ' as 'g'?{je} is treated exactly like {ge} in languages that palatalise, so I assume that they merged as /je/ quite early.
They may well have merged by //j// becoming [J\] (voiced mediopalatal stop), and then on to [dz\] either at the same time as //g// or somewhat later, rather than //g// becoming [j]. In early Middle Indo-Aryan old /j/ merged with old /J\/ at the same time as there developed a new /j/ next to front vowels in hiatus, which never ever became [\J] so there is no question that a /j/--/\J/ merger can go towards the obstruent. IIRC misspellings with G for I seem to be rather more frequent than the other way around, which probably amounts to something. Note also the quite early Greek spellings with zeta for Latin //j//; they would hardly have occurred if //g// > [j] rather than //j// > [J\]. Neither of course is it certain that they merged in the same way everywhere and at the same time. /bpj