Appendix talk:Ingrian orthographic variants

From Wiktionary, the free dictionary
Jump to navigation Jump to search

Checking and sourcing[edit]

@KirillW: I've now created this appendix. Could you review it? I'm going to see if I can find any durable review on Chernyavskij (to back the claims I've now given in his section), but I'm afraid it's nonexistant. If you have any more information to elaborate on what we have now, please do add it! Thadh (talk) 10:46, 25 April 2021 (UTC)[reply]

@Thadh: I've corrected a few things where I know for sure. I'm not aware of any response to Chernyavskij's work in academia and I think that Konjkova didn't use Chernyavskij's alphabet, but just a most obvious modern version approach (see my edits). I would say in all cases the alphabets didn't have to be based on any other specific source — all authors mostly followed the trends of their time. — This unsigned comment was added by KirillW (talkcontribs).
@KirillW: Thanks for your edits. I've also made a few changes to WT:AIZH and created the first automatic inflection template ({{izh-decl/kärpäin}}, see my recent changes for more). Thadh (talk) 19:37, 26 April 2021 (UTC)[reply]
@Thadh: This looks great! But here we might have to solve some issues. I'll post them to the relevant template.
@Thadh: Regarding Chernyavskij. It turned out that he had two versions of the book. The one I had is called "Icceopastaja", not "Ittseopastaja"!
And on usage in "Grammatikka" (a quote from it): "Çç — çentra, çirkkuli, çementta. (Vaa sanan alus. Keskeel sannaa kirjutettaa ts). Cc — seemecka, cirkkulain, caaju." — This unsigned comment was added by KirillW (talkcontribs).
Wow, had no idea about Chernyavskij! And yup, I probably should've read the alphabet section of Grammatikka, my bad. Thadh (talk) 20:44, 26 April 2021 (UTC)[reply]

Pronunciation[edit]

@Thadh, one thing that bothers me about actual dictionary articles: while it's relatively easy to specify what is a "standard" language by following Junus, it's really not possible to specify the standard pronunciation. We can't deduct real pronunciation from written forms. For instance, for orava it would be something in the lines of (Soi) /orːaːva/ or (Ala-L) /oravə/. Should we refrain from adding pronunciation for the time being? — I would better approach this later by adding Soikkolan and/or Ala-Laukan pronunciation in the same way as it's done for English, i.e. marking for which dialect some pronunciation is valid. --KirillW (talk) 20:25, 28 April 2021 (UTC)[reply]

@KirillW: I don't think that's accurate: The different pronunciations can exist side by side. For example, English that would be pronounced /ðæt/ in standard RP, yet as /ðæʔ/ in numerous dialects of British English. I believe the same is true for Ingrian: orava is pronounced /ˈorɑˌvɑ/ in Literary Ingrian, /ˈorːɑːˌvɑ/ (respelled: orraava) in Soikkola and /ˈorɑvə/ (not respelled, I suppose) in Ala-Laukaa. Since Literary Ingrian was designed for schools, it's only natural the pronunciation were standardised as well, otherwise the whole point of bringing the dialects together would be lost. Thadh (talk) 22:49, 28 April 2021 (UTC)[reply]
@Thadh: I don't say that different pronunciation can't exist side by side. My point is that "Literary Ingrian" pronunciation doesn't exist. It could have developed if the standard language had been taught. For this reason researchers don't talk about "standard" Ingrian and modern books use phonetic respelling, usually for Soikkola dialect. The standard exists only in the form of Junus' kirjakeeli. In Esisana he himself mostly talks about a common written language that would be understandable by speakers of two main dialects. The bottom line here is that if we would draw pronunciation from spelling, we will end up with something that probably no Ižorian would ever say themself.
Still, having a standard written language is very useful to keep the language alive. That's why I really want to have Junus' work be digitized and available for everyone. What we can do to handle this situation is to follow the example of English, for which the situation is in some way similar: words, written in the same way are pronounced differently in different standard dialects and neither is marked as a "global standard". With this in mind we can build an analogy: written form orava is pronounced /ˈorːɑːˌvɑ/ in Soikkola and /ˈorɑvə/ in Ala-Laukaa, but nobody says /ˈorɑˌvɑ/, in the same way as advertisement is pronounced /ədˈvɜːtɪsmənt/ in UK and /ˈædvɚˌtaɪzmənt/ in US, but it's not a standard to say something like /ˈadvəɹtaɪzmənt/. Also an interesting example would be Finnish, where kirjakieli exists and used, but people don't talk to each other in this language, they use puhukieli. Unless we have Ingrian media, there would be even no use for "standard" pronunciation (that said, there had been an attempt, but in Ala-Laukaa: http://finugor.ru/news/v-peterburge-startovalo-radioveshchanie-na-izhorskom-yazyke).
To wrap up: if we up to record the actual situation, we should use Junus' written forms and common spoken forms for both Soikkola and Ala-Laukaa as different pronunciation norms. In this case a dictionary entry would have Junus' spelling and, if available, multiple dialectal pronunciations without any of which marked as standard. I really hope that we can find common grounds on this because IMO it's a critical issue. --KirillW (talk) 17:39, 29 April 2021 (UTC)[reply]
@KirillW: In that case, if we want to go forward with only dialectal pronunciations, we will need to find a way to be able to retrieve this information from Nirvi. For example, Nirvi gives Ala-Laukaa orava, but this is /ˈorɑvə/ in IPA - what are the rules for this conversion? And what is precisely the phonemic value of e.g. ᴅ, z, ɢ? Because I would really like to give at least some pronunciations to the entries I create. If you have some source on Ingrian phonology that can answer these questions, that's okay as well.
Also, since we're using only dialectal pronunciations, we should create a designated template. I propose making {{izh-IPA}} with the parameters |S=, |A=, |H= and |Y= (the IPA values of the dialects) and |rS=, |rA=, |rH= and |rY= for the phonemic respellings (of course, all of these being optional).
An important question: what do we do with the terms that aren't recorded in Nirvi at all (like nominativa)? Are these actually in use among the population or was this only part of the kirjakeeli? Thadh (talk) 18:34, 29 April 2021 (UTC)[reply]
@Thadh: And this is exactly where I can be of help, at least for Ala-Laukaa :) There's a great thesis from Natalia Kuznetsova: https://iling.spb.ru/dissovet/old-abstracts/kuznecova/abstract.pdf (there's full text available on the web). It covers a lot on Ingrian phonology. I had made a summary on vowel reduction for Ala-Laukaa here: https://github.com/kirill-gr/izhoran-wiki/wiki/%D0%A0%D0%B5%D0%B4%D1%83%D0%BA%D1%86%D0%B8%D1%8F — for this dialect it would be one of the main features of pronunciation. I plan to compile other practical aspects of pronunciation into a single document, which describes, how to pronounce Junus' words in Ala-Laukaa way.
Linguistical terms seem to be mostly neologisms, made by Junus for kirjakeeli. The is one informant in Ala-Laukaa who himself tried described his language formally. I'll find out what terms he used. If we're lucky we could even ask him if he understands this terminology. He's 97 yo and unfortunately this winter I heard he was ill...
On the pronunciation templates I don't understand the implications of this, so can't comment. --KirillW (talk) 21:01, 29 April 2021 (UTC)[reply]
@KirillW See my contribution at hepoin: Am I correct to understand that is what we're going for? Thadh (talk) 09:19, 30 April 2021 (UTC)[reply]
@Thadh: This is excellent! The idea to use Nirvi as a source of pronunciation is very good, though it might be difficult to "parse" these articles correctly, and cumbersome to list all the subdialectal variants.
As described in "Two phonological rarities..." (section 2.1), Soikkola dialect has 3 consonant lengths. So the transcription should be /ˈhep̆poi̯n/, rather than /ˈhepːoi̯n/.
IIRC, hepoin has exceptional declension (not that many of these, luckily). See in Nirvi hevoist. I will do a refresher on this as well and come back with details.
Otherwise great work, thank you! --KirillW (talk) 19:25, 30 April 2021 (UTC)[reply]
AFAIK, geminates in IPA are shown with the long sound mark (ː), while double consonants could be given as just double (if, like in Soikkola, that difference is present, otherwise phonemically that doesn't matter, only phonetically).
By the way, I couldn't find Kuznetsova's full work online (not that that's very important, the abstract gives quite enough information on its own). I'm a bit disappointed Nirvi misses a lot of pronunciations for lesser-used words, mostly he has just Soikkola or even just a mention from a primary source.
Anyway, I'm going to try and go on like this for the time being; still need some decision on pronunciations of technical terms though. Thadh (talk) 21:23, 30 April 2021 (UTC)[reply]
@Thadh: If I understand you correctly, you're comparing geminates and double consonants, but I mentioned a different opposition in Soikkola: full geminates vs "initial-shortened" geminates, which differ in length. From "Two phonological rarities..." you can see that full geminates are 2x long as short consonants, while "initial-shortened" geminates are 1.5x long. What would be the IPA-way to show this 1.5x long sounds?
Here's a link for Kuznetsova's thesis: https://www.academia.edu/8374858/%D0%A4%D0%BE%D0%BD%D0%BE%D0%BB%D0%BE%D0%B3%D0%B8%D1%87%D0%B5%D1%81%D0%BA%D0%B8%D0%B5_%D1%81%D0%B8%D1%81%D1%82%D0%B5%D0%BC%D1%8B_%D0%B8%D0%B6%D0%BE%D1%80%D1%81%D0%BA%D0%B8%D1%85_%D0%B4%D0%B8%D0%B0%D0%BB%D0%B5%D0%BA%D1%82%D0%BE%D0%B2_PhD_diss_ms_%D0%A1%D0%BF%D0%B1_%D0%98%D0%9B%D0%98_%D0%A0%D0%90%D0%9D_2009_313_%D1%81
What are you leaning to with the pronunciation of the "new" words? --KirillW (talk) 19:28, 1 May 2021 (UTC)[reply]
@KirillW: About geminates: we could use the symbol ˑ, which stands for a half-long (1.5x) sound, though I haven't yet seen it used for consonants. Thanks for the link for the thesis (no idea why I couldn't find it; I usually check academia)! About the technical terms: For now I've just removed the pronunciations, and if it seems the terms aren't actually in use among the population we could let it stay that way. Thadh (talk) 13:26, 2 May 2021 (UTC)[reply]
Looks good to me! And that's probably why Kuznetsova calls these 1.5x long consonants a "phonological rarity" :) I'll post an update on the terms if we will be able to get to informants this summer. --KirillW (talk) 14:28, 2 May 2021 (UTC)[reply]