Talk:आंबट

@माधवपंडित IDK anything about Konkani phonology, but are you sure the /b/ is dropped? It should use [] phonetic brackets if this is just a regular outcome of /m.b/. —Aryaman^{A (मुझसे बात करें • योगदान)} 02:42, 29 November 2017 (UTC)

@AryamanA: In Konkani, when a stop follows a nasal, the stop is dropped. This is why I have added pronunciation to pages like this & उंदिर (undir). This may be just colloquial though. -- माधवपंडित (talk) 02:45, 29 November 2017 (UTC)

@माधवपंडित: Interesting! Now I wonder if a MOD:kok-IPA is possible... —Aryaman^{A (मुझसे बात करें • योगदान)} 02:46, 29 November 2017 (UTC)

Very... though i'm prettt inept at phonolgy. In some places the schwa of the Konkani is a bit different from that of Hindi. Which is why I put an elongated schwa here. Another example would be भय (bhay, “fear”) which is neither pronounced like hindi nor skt. To avoid confusion i just made भैय (bhaiy) (an alternate spelling). I will try to gather more on this subject. -- माधवपंडित (talk) 02:52, 29 November 2017 (UTC)

@AryamanA: and of course the biggest challenge would be /ts/ & /z/ against /ch/ & /j/. No idea how you and Kutchkutch dealt with it for Marathi. -- माधवपंडित (talk) 03:00, 29 November 2017 (UTC)

@माधवपंडित: Yes, it's quite a challenge. In case you don't know already, we discussed it in a lot of detail at Module talk:mr-translit/testcases, and even User:Wyang participated. Wyang and AryamanA came up with very good ideas, but they probably haven't had the time to test them out yet. It's interesting that Konkani has the same phenomenon so when the magical duo User:Wyang and User:AryamanA (and perhaps even User:DerekWinters) have the magical solution it will useful for Konkani as well. There was an attempt in the 1800s according to old Google Books to create a diacritic to distinguish the two, but it never caught on. Such a diacritic would have been immensely useful.

Their current project appears to be Module:lif-translit. Kutchkutch (talk) 03:25, 29 November 2017 (UTC)

Now that you've mentioned it, Dhongde & Wali says 'Marathi does not allow word final consonant-clusters except in words borrowed from English'. So आंबट (āmbaṭ) in Marathi doesn't qualify for the rule you presented above since 'word-medial non-geminates never occur in the same syllable', but other words such as कोंबणे (kombṇe) do qualify. The णे isn't part of the stem so the [b] is dropped from the stem-final /mb/ just like Konkani. In कोमजणे (komajṇe), the /b/ already dropped out of the spelling. Konkani phonology looks detailed but it doesn't appear to mention this. Kutchkutch (talk) 06:07, 29 November 2017 (UTC)

@Kutchkutch: True, which is why I believe this assimilation occurs in fast colloquial speech. Konkani does not spell words differently than how she wants them said. If the /b/ were to be dropped, it would not be there. So I'm guessing in an older form of Konkani this word was pronounced with a labial stop, and later on dropped. Same with उंदिर (undir), आंबो (āmbo, “mango”), कांदो (kāndo, “onion”). I really do wish the diacritic were included to record the /z/ and /ts/ sounds absent in Old Indo Aryan and the rest of Modern Indo Aryan. I guess the want for conformity has compromised accuracy. -- माधवपंडित (talk) 12:46, 29 November 2017 (UTC)

@माधवपंडित: The problem with diacritics is that no one uses them consistently. Even the nuqta hasn't caught on for Hindi despite it being used since at least the 1800s. Kashmiri recently tried to start using च़ (ċ) for /ts/ but again, standardization is slow. I think /z/ and /ts/ will have to given to the module in a phonetic transcription. —Aryaman^{A (मुझसे बात करें • योगदान)} 12:56, 29 November 2017 (UTC)

@AryamanA: This is possible if there's a rule as to when च & ज are ts & z. If there is one, we haven't discovered it yet. -- माधवपंडित (talk) 13:18, 29 November 2017 (UTC)