(Conlangs) Q&A Thread - Quick questions go here

A forum for all topics related to constructed languages
User avatar
WeepingElf
sinic
sinic
Posts: 214
Joined: 23 Feb 2016 18:42
Location: Braunschweig, Germany
Contact:

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by WeepingElf »

yangfiretiger121 wrote: 05 Jun 2020 13:41 I describe the vowel system of my Elvish language's abugida as follows: "[t]he native script is an abugida with an inherent [​i] and the remaining seven vowels arranged from first high-to-low, then front-to-back as diacritics—[y, u, e, ɛ, œ, ɔ, ɑ]". Are the descriptors in the correct order for the provided collation pattern?
The correct order, according to the description you give, would be [e ɛ y œ u ɔ ɑ]. Your order is more like first front-to-back, then high-to-low, i.e. the priorities reversed.
... brought to you by the Weeping Elf
shimobaatar
korean
korean
Posts: 7580
Joined: 12 Jul 2013 23:09
Location: PA → IN

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by shimobaatar »

Based on the order "[y, u, e, ɛ, œ, ɔ, ɑ]", I thought yangfiretiger121 might have intended "first high-to-low, then front-to-back" to be interpreted as "vowels are arranged by hight, from high to low, and vowels of the same height are arranged from front to back", if that makes sense. So, the two high vowels are listed before all non-high vowels, but [y], being front, is listed before [u]. Of course, I can't speak for yangfiretiger121, though.

In any case, I think I agree that "first front-to-back, then high-to-low" seems like a better way to describe "[y, u, e, ɛ, œ, ɔ, ɑ]".
Salmoneus
MVP
MVP
Posts: 2021
Joined: 19 Sep 2011 19:37

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Salmoneus »

A better option would be to drop the description altogether - if someone doesn't already know that /a/ is lower than /e/, they won't understand you telling them it anyway.
yangfiretiger121
sinic
sinic
Posts: 324
Joined: 17 Jun 2018 03:04

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by yangfiretiger121 »

Thanks for the help, guys.
Alien conlangs (Font may be needed for Vai symbols)
yangfiretiger121
sinic
sinic
Posts: 324
Joined: 17 Jun 2018 03:04

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by yangfiretiger121 »

As you know, the language's base vowel system is [ɑ, e, ɛ, i, œ, ɔ, u, y]. Considering the language is meant to sound like French, is a vowel harmony system likely to develop? If so, how?
Alien conlangs (Font may be needed for Vai symbols)
User avatar
jimydog000
sinic
sinic
Posts: 281
Joined: 19 Mar 2016 04:14
Location: Australian Country

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by jimydog000 »

Has nasalisation of vowels ever been a product of prosody?
User avatar
Omzinesý
runic
runic
Posts: 2847
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Omzinesý »

jimydog000 wrote: 08 Jun 2020 10:03 Has nasalisation of vowels ever been a product of prosody?
Guaraní has this nasal harmony. https://en.wikipedia.org/wiki/Guarani_l ... al_harmony
yangfiretiger121
sinic
sinic
Posts: 324
Joined: 17 Jun 2018 03:04

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by yangfiretiger121 »

My setting's main language has the beginnings of a two-tiered vowel harmony system containing front/+atr [æ,i], back/-atr [ɔ, ʊ̠], and central/neutral [ə]. If full front/back harmony develops, is [ɑ, ɪ̟, ø, u] or [ɑ, e, o, ʏ̟] the more likely set of allophones to arise? Mind you, I am just asking because of the language's current vowel alignment but don't think I'm using vowel harmony at all.
Alien conlangs (Font may be needed for Vai symbols)
User avatar
jimydog000
sinic
sinic
Posts: 281
Joined: 19 Mar 2016 04:14
Location: Australian Country

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by jimydog000 »

yangfiretiger121 wrote: 14 Jun 2020 23:54 My setting's main language has the beginnings of a two-tiered vowel harmony system containing front/+atr [æ,i], back/-atr [ɔ, ʊ̠], and central/neutral [ə]. If full front/back harmony develops, is [ɑ, ɪ̟, ø, u] or [ɑ, e, o, ʏ̟] the more likely set of allophones to arise? Mind you, I am just asking because of the language's current vowel alignment but don't think I'm using vowel harmony at all.
The difference between an Advanced Tongue Root harmony and Front/Back harmony may occasionlly be up to what analysis you make of the language (Kazakh), but they are not interchangeable.

I think you made a few typos or don't fully understand harmonic systems did you actually mean:

+ATR [æ, ʊ̠], -ATR [i, ɔ], and neutral [ə]. ?

That would make more sense and put the vowels in the right order. Although /æ/ alternating with /i/ with a neutral schwa is a bit out there. Alternatively you could do:

+ATR [æ, ʊ̠], -ATR [ə, ɔ], and neutral [i].

Or:

+ATR [æ, ʊ̠], -ATR [ɑ, ɔ], and neutral [i, ə].

Or even simpler (and follows what Kazakh apparetly does)

+ATR [æ, i, ʊ̠], -ATR [ɑ, ə, ɔ].

And to answer your question about the result of allophones becoming front back or something. I would suggest a combination of the two as:
-Back[e, ø] ~/~ +Back[ɑ, o]
Last edited by jimydog000 on 15 Jun 2020 20:21, edited 1 time in total.
yangfiretiger121
sinic
sinic
Posts: 324
Joined: 17 Jun 2018 03:04

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by yangfiretiger121 »

jimydog000 wrote: 15 Jun 2020 18:31
yangfiretiger121 wrote: 14 Jun 2020 23:54 My setting's main language has the beginnings of a two-tiered vowel harmony system containing front/+atr [æ,i], back/-atr [ɔ, ʊ̠], and central/neutral [ə]. If full front/back harmony develops, is [ɑ, ɪ̟, ø, u] or [ɑ, e, o, ʏ̟] the more likely set of allophones to arise? Mind you, I am just asking because of the language's current vowel alignment but don't think I'm using vowel harmony at all.
The difference between an Advanced Tongue Root harmony and Front/Back harmony may occasionlly be up to what analysis you make of the language (Kazakh), but they are not interchangeable.

I think you made a few typos or don't fully understand harmonic systems did you actually mean:

+ATR [æ, ʊ̠], -ATR [i, ɔ], and neutral [i]. ?

That would make more sense and put the vowels in the right order. Although /æ/ alternating with /i/ with a neutral schwa is a bit out there. Alternatively you could do:

+ATR [æ, ʊ̠], -ATR [ə, ɔ], and neutral [i].

Or:

+ATR [æ, ʊ̠], -ATR [ɑ, ɔ], and neutral [i, ə].

Or even simpler (and follows what Kazakh apparetly does)

+ATR [æ, i, ʊ̠], -ATR [ɑ, ə, ɔ].

And to answer your question about the result of allophones becoming front back or something. I would suggest a combination of the two as:
-Back[e, ø] ~/~ +Back[ɑ, o]
Thanks for the help. However, I ordered them like that because I interpreted this to mean that near-high vowels, such as [ʊ̠], will be -atr no matter their backness, which may have been incorrect. The neutral [ə] originated through consolidation of Elvish's [e, ɛ, œ] and Phoenixtongue's [ø, œ, ə]. That said, I understand vowel harmony just fine and was mainly referring to the fact that the current system has to front vowels that can be described as +atr in [æ,i], two back vowels that are, presumably, -atr in [ɔ, ʊ̠], and a central vowel that would be neutral in [ə]. Perhaps, my disclaimer should've been clearer and said that the language (a) doesn't and (b) is exceedingly likely to never have actual vowel harmony.
Alien conlangs (Font may be needed for Vai symbols)
User avatar
Creyeditor
mongolian
mongolian
Posts: 3930
Joined: 14 Aug 2012 19:32

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Creyeditor »

yangfiretiger121 wrote: 14 Jun 2020 23:54 My setting's main language has the beginnings of a two-tiered vowel harmony system containing front/+atr [æ,i], back/-atr [ɔ, ʊ̠], and central/neutral [ə]. If full front/back harmony develops, is [ɑ, ɪ̟, ø, u] or [ɑ, e, o, ʏ̟] the more likely set of allophones to arise? Mind you, I am just asking because of the language's current vowel alignment but don't think I'm using vowel harmony at all.
If [+ATR] becomes [+front] and [-ATR] becomes [-front], you simply get [æ,i], back/-atr [ɔ, ʊ̠], where [æ] alternates with [ɔ] and [ i ] alternates with [ʊ̠]. Am I missing something here?

If you are asking generally, which of the the two system presented is more likely, I would guess the first one, but I don't see how you could go from your five vowel system to such a four vowel system with a set of simple changes.

Just a nitpick. Schwa is usually considered [+ATR].
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
yangfiretiger121
sinic
sinic
Posts: 324
Joined: 17 Jun 2018 03:04

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by yangfiretiger121 »

Creyeditor wrote: 15 Jun 2020 20:02
yangfiretiger121 wrote: 14 Jun 2020 23:54 My setting's main language has the beginnings of a two-tiered vowel harmony system containing front/+atr [æ,i], back/-atr [ɔ, ʊ̠], and central/neutral [ə]. If full front/back harmony develops, is [ɑ, ɪ̟, ø, u] or [ɑ, e, o, ʏ̟] the more likely set of allophones to arise? Mind you, I am just asking because of the language's current vowel alignment but don't think I'm using vowel harmony at all.
If [+ATR] becomes [+front] and [-ATR] becomes [-front], you simply get [æ,i], back/-atr [ɔ, ʊ̠], where [æ] alternates with [ɔ] and [ i ] alternates with [ʊ̠]. Am I missing something here?

If you are asking generally, which of the the two system presented is more likely, I would guess the first one, but I don't see how you could go from your five vowel system to such a four vowel system with a set of simple changes.

Just a nitpick. Schwa is usually considered [+ATR].
That's nice to know, and you're not missing anything. Honestly, I wasn't aware how close the language is to having a working vowel harmony system, which changes things to make one developing slightly more likely. While that's still highly unlikely. what's a traditionally neutral vowel I can slot in with the four current peripherals because I like the idea of one vowel being eligible for every word?
Alien conlangs (Font may be needed for Vai symbols)
User avatar
Creyeditor
mongolian
mongolian
Posts: 3930
Joined: 14 Aug 2012 19:32

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Creyeditor »

In ATR vowel harmony systems the low vowel (/a/) is often neutral, because it is inherently RTR and does not have an ATR counterpart in most languages. In backness harmony systems /i/ or /e/ are often neutral, because there are no back unrounded vowels. This fact however mostly holds for languages where rounding and backness are contrastive/distinctive independtly. In the system I proposed in the last post, rounding and backness covary. There are neither front rounded nor back unrounded vowels. I cannot guarantee that this system itself is natural as a regular productive vowel harmony system. I have seen such systems mostly in restricted systems, where harmony only applies to some affixes. Anyways, neutral vowels are a difference between ATR harmony and backness harmony systems, at least frequently.
One sudden idea I just had is to introduce a diphthong /ai/. This could be neutral in both systems. Also, some diachronic ideas from the main language's system to an ATR or a backness harmony system.

Proto-L:
/i ʊ̠/
/æ ɔ/
/ai/

Daughter ATR:
/æ/ > /e/
/ai/ > /a/
We get the common neutral /a/ for ATR systems.
/i ʊ̠/
/e ɔ/
/a/

Daughter Backness:
i > e
ai > i
We get the common neutral /i/ for backness systems.

Just for completeness sake. Another, less spontaneous, change leading to vowel harmony is the phonologization of overlapping articulatory gestures. At least that's what I have heard. This would mean that you could start out with a very basic vowel system, like the following, and still get a nice ATR or backness harmony system.

/i u/
/e o/
/ɑ/

After a back rounded vowel, the non-rounded vowels could develop a rounded allophone and after front vowels, the back vowels could develop a front allophone. After /ɑ/ all vowels could develop a back/central unrounded allophone. This would yield the following allophones.

/i/ [i y ɨ]
/u/[u y ɨ]
/e/[e ø ɤ]
/o/[o ø ɤ]
/ɑ/[æ ɑ ɒ]

Words are now generally harmonic, but they start with one of the five vowels from the first stage. The new allophones have to become phonemic. One way to do this is by monophthongizing diphthongs.

ɑi > ɑ
ɑu > ɒ
oi > ø
ui > y

This would mean that only non-low back unrounded vowels are non-phonemic.You would get the following phonemic system with rounding and backness harmony.

/i/ /y/ [ɨ] /u/
/e/ /ø/ [ɤ] /o/
/æ/ /ɑ/ /ɒ/

You could apply a similar strategy to your main languages vowel (either for backness or analogically for ATR/RTR harmony) and adjust the result. I am not a 100% sure about the result though.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
User avatar
CarsonDaConlanger
sinic
sinic
Posts: 233
Joined: 02 Nov 2017 20:55

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by CarsonDaConlanger »

In languages with case prefixes, is number also often prefixed as well?
User avatar
eldin raigmore
korean
korean
Posts: 5566
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by eldin raigmore »

CarsonDaConlanger wrote: 20 Jun 2020 04:21 In languages with case prefixes, is number also often prefixed as well?
I bet WALS.info can tell us. I’ll go look.

See
https://wals.info/combinations/51A_33A#2/6.3/146.6

757 of their languages don’t have plural prefixes and don’t have case prefixes
51 of their languages have plural prefixes but don’t have case prefixes
14 of their languages have plural prefixes and case prefixes
13 of their languages have case prefixes but don’t have plural prefixes.

So about 21.5% of their languages with plural prefixes also have case prefixes;
while about 1.7% of their languages without plural prefixes have case prefixes.

And about 51.9% of their languages with case prefixes also have plural prefixes;
while about 6.3% of their languages without case prefixes have plural prefixes.

So it seems languages with case prefixes probably have plural prefixes.

Languages probably don’t have case prefixes, even if they do have plural prefixes;
but languages with plural prefixes are more than 12.5 times likelier to have case prefixes, than other languages.

__________

HTH!
Salmoneus
MVP
MVP
Posts: 2021
Joined: 19 Sep 2011 19:37

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Salmoneus »

Not very meaningful, though, without distinguishing the questions of "does this language have prefixes for X?" and "where do the prefixes for X go?" In particular, there'll be a huge chunk of languages that just aren't inflecting at all.
User avatar
sangi39
moderator
moderator
Posts: 2701
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by sangi39 »

So, of the languages that WALS gives as having both (14 languages), half of them are Bantu languages, which are included because of their locative noun classes.

Nias is included because it marks case via initial consonant mutation, but number marking on nouns appears to be relatively rare. You can use a collective prefix, which precedes the root (and thus precedes the mutated consonant), or you can reduplicate the noun as a whole, but that's noted as rare. The plurality of a noun seems to be left up to context and marking elsewhere in the sentence most of the time, though.

There appears to be some debate over what's going on in Enggano. From what I can tell, there are human nouns, which take singular and plural number prefixes which are different, and non-human nouns which take the same prefix and singular human nouns (e-), but do not distinguish between singular or plural. This e- is then replaced by u- when, for example, the noun is an object, and by i- when in the locative (plurality in nouns can also be marked by reduplication, but apparently that functions more like a collective).

Marra is... interesting. It has number prefixes which differ depending on the case the noun is in (nominative vs. non-nominative) but the cases themselves are marked with suffixes (except the nominative and ergative/instrumental, which are marked solely by use of the nom vs. non-nom prefixes respectively), so that's cool.

I can't find anything yet on Ocuilteco or Mitla Zapotec that gives much more than their Wikipedia articles, but it seems like the situation is similar to Nias in the latter, i.e. there is plural marking, but it appears rarely, and the only case marking I can see is a prefix which appears on possessed alienably possessed nouns (so I guess a sort of possessed case prefix), but I can't see whether that comes before or after the number prefix.

So, at least on that front, it doesn't look like there's much of a trend either way, In Nias, it's number then case or case then number, depending on how its marked, in Enggano it's a single case/number prefix that can precede a reduplicated noun, and in Marra it's a case/number prefix before the root, with the root then taking case suffixes.
You can tell the same lie a thousand times,
But it never gets any more true,
So close your eyes once more and once more believe
That they all still believe in you.
Just one time.
User avatar
eldin raigmore
korean
korean
Posts: 5566
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by eldin raigmore »

Salmoneus wrote: 20 Jun 2020 15:57 Not very meaningful, though, without distinguishing the questions of "does this language have prefixes for X?" and "where do the prefixes for X go?" In particular, there'll be a huge chunk of languages that just aren't inflecting at all.
Many of their 757 languages with neither kind of prefix actually have neither kind of affix. So your last sentence is supported by WALS.

OTOH I don’t see how you got the idea that their data doesn’t distinguish between the questions “does this language have prefixes for ....?” and “where do the prefixes for .... go?”.
In particular if only 14 languages in their sample have prefixes for both grammatical case and grammatical number, it wouldn’t be a big enough sample to make any statement about which of those two(?) prefixes “usually” come first, or nearer the root of the noun.
And in languages where one prefix fuses case and number, the question wouldn’t even make sense.

You seem to be responding to things posted before the question by Carson that I was replying to.
I don’t think your comment was relevant to my post at all.
Edit: I apologize for those last two sentences. I thought Salmoneus’s post and sangi’s post were by the same author. They’re not. The stuff I just struck out was based on that error, and I’d rather I hadn’t posted it. It’s not really responsive to either’s post if they didn’t also post the other one.
In fact, I regret the tone of my third sentence (“ ... I don’t see how etc.”). I’m not striking it out because I still want to express the purely factual content and I can’t think of a way to re-word it with a more appropriate tone. I apologize for that lack of skill on my part.
Last edited by eldin raigmore on 20 Jun 2020 23:57, edited 1 time in total.
User avatar
sangi39
moderator
moderator
Posts: 2701
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by sangi39 »

eldin raigmore wrote: 20 Jun 2020 23:23
Salmoneus wrote: 20 Jun 2020 15:57 Not very meaningful, though, without distinguishing the questions of "does this language have prefixes for X?" and "where do the prefixes for X go?" In particular, there'll be a huge chunk of languages that just aren't inflecting at all.
Many of their 757 languages with neither kind of prefix actually have neither kind of affix. So your last sentence is supported by WALS.

OTOH I don’t see how you got the idea that their data doesn’t distinguish between the questions “does this language have prefixes for ....?” and “where do the prefixes for .... go?”.
In particular if only 14 languages in their sample have prefixes for both grammatical case and grammatical number, it wouldn’t be a big enough sample to make any statement about which of those two(?) prefixes “usually” come first, or nearer the root of the noun.
And in languages where one prefix fuses case and number, the question wouldn’t even make sense.

You seem to be responding to things posted before the question by Carson that I was replying to.
I don’t think your comment was relevant to my post at all.
Edit: I apologize for those last two sentences. I thought Salmoneus’s post and sangi’s post were by the same author. They’re not. The stuff I just struck out was based on that error, and I’d rather I hadn’t posted it.
In fact, I regret the tone of my third sentence (“ ... I don’t see how etc.”). I’m not striking it out because I still want to express the purely factual content and I can’t think of a way to re-word it with a more appropriate tone. I apologize for that lack of skill on my part.
I think Sal is right about his two questions. WALS doesn't actually specify in its data, as far as I know, exactly what sort of grammatical number is being encoded (and when it's encoded, as was mentioned in my part about Nias and Enggano, for example, although I think WALS also has a section regarding obligatory vs. option number marking, but I'm not sure how many features you can combine in WALS on one map), and Sal is definitely right about WALS not indicating where certain prefixes go, i.e. while it indicates that they're both prefixes, it doesn't say whether the case prefix or the number prefix comes first (or at least that's the way I read the question).

WALS does have it's limitations, and it's questions like these that show them. It's a nice handy tool to show what sorts of features co-occur in which languages, but it's only really a place to start, and requires some further digging when it comes to specifics (it's also a must to read the articles related to each map, because, as noted by Nias, it groups certain features together for the sake of the maps that would otherwise be kept distinct).

WALS also, on occasion (as I believe is the case for resources like PHOIBLE) is beholden to its resources, and often it seems like the things it says about some languages are either wrong (I think it said at one point that one of the languages that has case prefixes has six grammatical cases, but as far as I could tell, from available free online resources, that language didn't mark grammatical case on nouns at all (I think it was Shuswap)), or at least disputed (as was the case with Mitla Zapotec, where WALS mentioned number prefixes, but Wikipedia calls it a proclitic).
You can tell the same lie a thousand times,
But it never gets any more true,
So close your eyes once more and once more believe
That they all still believe in you.
Just one time.
User avatar
Creyeditor
mongolian
mongolian
Posts: 3930
Joined: 14 Aug 2012 19:32

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Creyeditor »

sangi39 wrote: 20 Jun 2020 23:53 WALS also, on occasion (as I believe is the case for resources like PHOIBLE) is beholden to its resources, and often it seems like the things it says about some languages are either wrong [...], or at least disputed [...].
That's why I really appreciate PHOIBLE also showing contradictory sources.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
Post Reply