--- In
cybalist@yahoogroups.com, Bhrihskwobhloukstroy <bhrihstlobhrouzghdhroy@...> wrote:
>
> 1) The case of pignora doesn't add anything to the question, becasue I
> had already conceded, for the sake of the argment, that velar nasal
> <gn> may undergo palatalisation. The problem is /ng/, where /g/ is
> different from /n/ for what matters.
The purpose of describing peyndra , etc., is to show that N > y , L > y , k > y all occur, only 1 can be ascribed to k > x, so that simple x > y doesn't work; better all K>KY in 1 env.
That Ng > NYg happened in Sp (as an intermediate before NYg > Ng), or even in some extinct l., is most easily seen by fungus > onddo. In Sp, the 2 outcomes of Ng before a front V (gingi:va > enzia OSp; ringi: > -ere > reñir) show the same alt. as tegno/tengo (but these not before a front V); most likely from NYge > NYgYe ( > NYGY opt.). Sim., ungula > uña shows dif. from amplus > ancho, etc., all most easily accounted for by N>NY first (shown by the outcomes of Nn, among others, as well as onddo).