Wiktionary talk:About Ter Sami

Orthography

@Tropylium I've been working on "standardising" the orthography of our entries somewhat, by making it more consistent in how it represents individual phonemes, and also by using characters that are more standard in Sami orthography in general. There are some issues that I'm not sure about, though.

Firstly, the voiced affricates, which occur in mostly the same environments that cause voicing of stops. Northern Sami uses z and ž. Ter Sami has actual /z/ and /ʒ/ as phonemes though (they are the weak grades of the single sibilants) so we must use z and ž for those. We can't use single c and č either, because voicing is contrastive at the end of a cluster (it reflects the single-geminate distinction). Skolt Sami solves the problem by introducing two entirely new letters, ʒ and ǯ. While this is a neat solution, these letters are probably not very well known and thus not easily understood. At the very least, though, Skolt could serve as a model for using these in Ter. Another option would be to write dz and dž. This is how Kildin handles it in Cyrillic, which also lacks letters for voiced affricates.

Secondly, the strong grade single stops/affricates. These seem to be transcribed variously as plain geminates (pp, tt, cc, čč, kk) and partly voiced geminates (ʙp, ᴅt, ᴅc, ᴅč, ɢk). The latter style is used in Kildin Sami orthography, while the former conforms with Skolt and Inari. I would be inclined to transcribe them as plain geminates, leaving the pre-stopping to a phonetic detail, but I don't actually know if Ter Sami has actual plain geminates that contrast with these. I think if there is no contrast, we should probably go with the plain geminates.

Thirdly, there's the palatalization. Ter Sami didn't lose the conditioning final *ē in most cases, so it seems that the palatalization is purely allophonic in those environments. It only needs to be indicated when there is no following front vowel. However, this all hinges on whether there are also instances of e with a preceding nonpalatalized consonant. Since Ter is heavily influenced by Russian, this is a real possibility, but I can't say. —Rua (mew) 19:30, 22 November 2017 (UTC)