(L&N) Q&A Thread - Quick questions go here

[eldin raigmore]

I have found one or a few natural language(s) with terms for
a consanguine relative’s spouse’s consanguine relative’s spouse’s consanguine relative:

Has anyone seen or heard about any natlang or conlang that has (a) kin-term(s) with meaning(s) roughly similar to one or more of:
[sibling’s or child’s or parent’s] spouse’s [sibling’s or parent’s or child’s] spouse’s cousin
or
[sibling’s or child’s or parent’s] spouse’s cousin’s spouse’s [sibling or parent or child]
or
cousin’s spouse’s [child’s or parent’s or sibling’s] spouse’s [sibling or parent or child]
?

I take “cousin” to be a third-degree blood relative (i.e. parent’s sibling’s child), while “parent or child or sibling” I take to be the first-degree blood relatives.

[Akubra]

Hello everyone!

I'm new here, so greetings to all!

I’m currently in the (very) early stages of designing my second conlang, which I have provisionally named K'aach (/kʔaːt͡ʃ/). I’m focusing on the phonology, which is loosely inspired by Maya languages, Hindi, and the Yolŋu languages spoken in northern Australia.

I'm considering introducing /ɹ/ and /ɾ/ as separate phonemes, despite being aware of how rare (or possibly non-existent) this distinction is. According to ChatGPT, Kalabari (an Ijo language spoken in Nigeria) features this distinction, but I haven’t been able to find sources to confirm this. I also haven’t found any other languages with this characteristic.

Can anyone verify ChatGPT's claim, or provide information on other natlangs that have this feature? Thanks in advance!

[Creyeditor]

Welcome
You could use PHOIBLE to try to verify the claim: https://phoible.org/

[Akubra]

Thanks a lot!

I have discovered that Kalabari does not have /ɹ/ and /ɾ/ as separate phonemes, but Alawa, an Australian language, apparently does. I plan to continue my search to see if there are other languages with this feature. Even if it's extremely rare, that's enough for me to include it in K'aach.

[Akubra]

For those who might be interested, I have found a total of six languages that distinguish /ɹ/ and /ɾ/:

• Alawa, Mangarrayi-Maran family, northern Australia
• Isoko, Atlantic-Congo family, southern Nigeria
• Jebero, Cahuapanan family, northern Peru
• Ngalak(g)an, Gunwinyguan family, northern Australia
• Sedik/Seediq, Austronesian family, Taiwan
• Weh, Atlantic-Congo family, western Cameroon

As I suspected, this feature is extremely rare, but it does exist, so I'm keeping it in K'aach.

[Arayaz]

Welcome to the board, Akubra!

First up: Probably don't trust ChatGPT on its own, especially for less-well-known fields like linguistics. Ideally, use Google or Wikipedia (and don't trust Google's AI search overviews, either). And ideally, check the sources for it.

That said, the distinction between /ɾ/ and /ɹ/ is not at all problematic. I'm pretty sure a majority of Australian languages distinguish /ɾ/ and /ɻ/, which suggests that it's not at all difficult to distinguish two rhotic consonants.

What was your source for this being a rare/non-existent distinction, by the way?

(And as for Kalabari, ChatGPT seems to indeed have been in error. Every source I've found suggests that it has only one sound spelled with an <r>, and this seems to be /ɾ/, not [ɹ]. This gives a phoneme list and wordlist, which have /ɾ/ and no /ɹ/; this shows the Kalabari writing system, which has only one spelling that seems to be a rhotic consonant.)

[Khemehekis]

That said, the distinction between /ɾ/ and /ɹ/ is not at all problematic. I'm pretty sure a majority of Australian languages distinguish /ɾ/ and /ɻ/, which suggests that it's not at all difficult to distinguish two rhotic consonants.

And, lest anyone forget, Spanish distinguishes /ɾ/ vs. /r/. The pero-perro distinction!

[Akubra]

Thanks for the welcome, Arayaz!

Probably don't trust ChatGPT on its own, especially for less-well-known fields like linguistics. Ideally, use Google or Wikipedia (and don't trust Google's AI search overviews, either). And ideally, check the sources for it.

Good advice! I'll keep it in mind.

What was your source for this being a rare/non-existent distinction, by the way?

Years ago, while I was working on Rautahi, my first conlang, I was active on the ZBB for a brief period. Rautahi also included both /ɹ/ and /ɾ/, and someone commented that "/ɾ/ vs. /ɹ/ is not a very strong contrast; either they will merge, or one or both of them will become something more easily distinguishable from the other (e.g. /ɾ/ strengthening to /r/)."
I remembered those words and suspected that if this was the case, such a distinction must be very uncommon. It turns out I wasn't far off, as I found only 6 out of the 2186 languages in PHOIBLE that have this feature.

Every source I've found suggests that it has only one sound spelled with an <r>, and this seems to be /ɾ/, not [ɹ].

PHOIBLE mentions /r/ instead of /ɾ/, but that's not important. What matters is that there's only one and it's not /ɹ/, as you say.

[Salmoneus]

And English - at least, US English - distinguishes [ɾ] and /ɹ/. The former isn't conventionally considered phonemic yet, but its a very well-established allophone.

[Creyeditor]

Is the combination rarer than expected? Let's do some quick and dirty statistics

languages with only the tap/flap: 768
languages with the approximant: 54
languages with both: 6
If there are 2286 laguages in total, that means 2286-768-54-6=1458 languages have neither.

Doing a chi-square test (dirty, I know) yields a p-value of .000076. This could mean that the combination is rarer than expected judging from the frequency of the combined segments.

[Arayaz]

And English - at least, US English - distinguishes [ɾ] and /ɹ/. The former isn't conventionally considered phonemic yet, but its a very well-established allophone.

Well, in this case they're more distinct than that, since the latter is at the least retracted to /ɹ̠/, and more than often labialized /ɹ̠ʷ/.

[Visions1]

….
In a monotheistic religion, where it’s believed only one god exists at all (rather that only one is served or worshipped, as in monolatry or henotheism), why does such a god need a name? (Especially in the beginning?)
….

…. For example, Elohim and the Four-Letter Name symbolize His Judgement (restraint from being too nice) and Mercy (being nice) respectively. Shaddai is apoprotraic.) ….

I’m pretty sure you meant apotropaic. Right?

Yep. Sorry about that.

[Omzinesý]

In English conversion is usually marked so that nouns have an article and verbs don't.
How do you distinguish between nouns and verbs in other analytic languages that have conversion but no articles?

[Omzinesý]

Does Danish have geminates i.e. distinction between consonant lengths? I assume it does not but I haven't seen it said anywhere.

[Creyeditor]

In English conversion is usually marked so that nouns have an article and verbs don't.
How do you distinguish between nouns and verbs in other analytic languages that have conversion but no articles?

In Indonesian you have conversion but verbs take morphology in certain contexts whereas nouns usually don't.

[Omzinesý]

In English conversion is usually marked so that nouns have an article and verbs don't.
How do you distinguish between nouns and verbs in other analytic languages that have conversion but no articles?

In Indonesian you have conversion but verbs take morphology in certain contexts whereas nouns usually don't.

But I mean how do you tell them apart if they don't have morphology?

[Creyeditor]

Hard to say. I think David Gil argues that you just don't, at least in some Indonesian varieties.

[sangi39]

Does Danish have geminates i.e. distinction between consonant lengths? I assume it does not but I haven't seen it said anywhere.

A few sources I found online state that Danish doesn't have geminate consonants:

https://en.wikipedia.org/wiki/Compariso ... nd_Swedish

"Segmental Phenomena in Germanic: Consonants" by S. Litty and J. Salmons (2023)

"Gemination, Lenition, and Vowel Lengthening: On the History of Quantity in Germanic" by K. Goblirsch (2018)

I found one paper that assumes underlying geminates to explain some stuff to do with lenition, but even that says that Danish has no surface level geminate consonants

[Omzinesý]

Does Danish have geminates i.e. distinction between consonant lengths? I assume it does not but I haven't seen it said anywhere.

A few sources I found online state that Danish doesn't have geminate consonants:

https://en.wikipedia.org/wiki/Compariso ... nd_Swedish

"Segmental Phenomena in Germanic: Consonants" by S. Litty and J. Salmons (2023)

"Gemination, Lenition, and Vowel Lengthening: On the History of Quantity in Germanic" by K. Goblirsch (2018)

I found one paper that assumes underlying geminates to explain some stuff to do with lenition, but even that says that Danish has no surface level geminate consonants

Thank you. It will take time to go them through. The Wikipedia article had new info too. What a strange title to put the info.

According to Wikipedia also:
"A sonorant consonant (/ð, l, m, n, ŋ/) and /ə/, in either order, become a syllabic consonant [ð̩, l̩, m̩, n̩, ŋ̍].[54]
It is longer after a short vowel than after a long one: /ˈbaːðə/ → [ˈpæːð̩] bade 'bathe', /ˈhuːlə/ → [ˈhuːl̩] hule 'cave', /ˈsbiðə/ → [ˈspiðð̩] spidde 'spear', /ˈkulə/ → [ˈkʰull̩] kulde 'cold'.[55]"
It might be possible to interpret those final consonants as geminates too. I don't know Danish enough to know if there are evident reasons to interpret them as consonant + syllabic consonant.

The Wikipedia article on Danish phonology seems to be incoherent with respect to stød bases too. There is just much I don't know.

[Omzinesý]

Hard to say. I think David Gil argues that you just don't, at least in some Indonesian varieties.

I'm just wondering how you parse clauses if the same words/phrases could be either nouns or predicates/verbs.