Mon Dec 29, 2014 4:16 pm
Apologies if this question has been asked before, but I was interested in the potential pronunciation of Nynorn words which involve "silent" or "null" <ð> and <g>. The grammar page implies they are unpronounced as in Faroese, but the picture in Faroese is rather complicated, since these will often be realised as glides [j], [w], [v], or sometimes even change the quality of the preceding vowel: compare dagur [dɛa:vʊr] dagin [dɛa:jɪn] dagar [de:ar].
My question is really to open discussion about what we might suggest for Nynorn realisations of these sounds: if we simply assume that it is "as in Faroese", we run into trouble with words like dagar, since in Norn the vowel of the first syllable is presumably not originally a diphthong [ɛa:], but simply [a]. If we simply delete the consonant, we end up with two identical vowels in hiatus: [daar]. Maybe that is fine, but should we then propose a rule to lengthen the vowel as [da:r], or to propose a new diphthong, e.g. [daur]? Or should we just say the Faroese long [e:] is what we end up with even if the original vowel wasn't [ɛa:]? Is there any precedent in ON/Scandinavian for allowing consecutive vowels in hiatus like [daar]?
I have similar questions regarding <ð>, which is straightforward enough at the end of a word, but when between vowels it will also need to be realised sometimes as [v] or [w] if we try to follow Faroese. It may not always make phonological sense to follow Faroese exactly however.
On a side note, if it turns out that <ð> will genuinely always be null, shouldn't we remove it from the orthography? It isn't strictly true that the Faroese <ð> is genuinely null, since it will be realised as glides which in non-Hammershaimb orthographies were spelt <j>, <v> etc. It made sense for Faroese to go for etymological spelling since otherwise it could be harder to recognise that words with <j>, <v> or null were all forms of the same word. But if we decide that <ð> is always null in Nynorn, that difficulty wouldn't arise.