17-09-2017, 10:31 PM
Quote:David, since Emma brought up the point about how certain letters are associated with ng (in English it is always preceded by a vowel) in response to my post about positional rigidity in the VMS, I was mainly thinking about pair behavior in the VMS (in terms of glyphs, not necessarily sounds, which is also a good discussion, but wasn't specifically what I was responding to when I saw Emma's post).
OK JPK, I see where we are diverging. I'm purely talking about phonemes, you're talking about glyph placement.
(Although the seperation of the two would only really occur to speakers of a natural language where pronunciation / spelling has diverged, such as English. This discussion wouldn't really make sense in, say, Spanish.)
As you say, we should diverge the two discussions. Although this is fascinating. But the /ŋ/ sound being preceded by a vowel is an English phenomenon. Why are we discussing it? Look at the same sound in Welsh:
Quote:Ng as the 'ng' in finger:
Welsh words Yng Nghaerdydd (ung hire deethe); Yng Nghymru (ung Humree); nghaerusalem; ngyvyrgoll [no idea, but it's in 15th century corpus]
Anton -
Quote:and that there are no underlay spaces where we don't observe them?Timeout for bringing in the spaces rule. I'm still not sure whether we're talking about saying or spelling
