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

A forum for all topics related to constructed languages
User avatar
Aevas
admin
admin
Posts: 1445
Joined: 11 May 2010 05:46
Location: ꜱᴇ

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

Post by Aevas »

Someone can correct me if I'm wrong, but I'm not certain Finnish actually had θ > ts. As I understand it, [θ] was spelled <tz> and later <ts>, and this became the basis for a spelling pronunciation in the standard language, which was influenced by Swedish speakers. The sound [θ] had previously developed from and older [ts].
User avatar
sangi39
moderator
moderator
Posts: 3024
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

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

Post by sangi39 »

Aevas wrote: 14 Mar 2023 19:54 Someone can correct me if I'm wrong, but I'm not certain Finnish actually had θ > ts. As I understand it, [θ] was spelled <tz> and later <ts>, and this became the basis for a spelling pronunciation in the standard language, which was influenced by Swedish speakers. The sound [θ] had previously developed from and older [ts].
I swear this change has been discussed in quite some depth, with some sources, recently over on the ZBB, but I'm having trouble finding it, and it's been discussed here on the CBB as well. From what I can tell, it's a bit of a mixed case, e.g. <tz> as /ts/ was a spelling pronunciation in some areas (influenced by Swedish speakers), but it might also have been a retention in others (where older *(c)c didn't become /(θ)θ/), which might have further influenced the spelling pronunciation. I think I'd agree, though, that it's less "θ > ts" and more "/ts/ as the resulting form of *(c)c replaced /(θ)θ/"
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.
Ælfwine
roman
roman
Posts: 940
Joined: 21 Sep 2015 01:28
Location: New Jersey

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

Post by Ælfwine »

Fascinating stuff about Finnish.

Judging by one of those posts, I assume affricating a cluster like /ts/ to /t͡s/ is perfectly attested as well. Im just brainstorming ways to get this phoneme without palatalization necessarily.
My Blog

A-posteriori, alternative history nerd
User avatar
GoshDiggityDangit
greek
greek
Posts: 549
Joined: 18 Dec 2018 21:27
Location: Oakwood OH, USA
Contact:

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

Post by GoshDiggityDangit »

It's probably been discussed here before, but what is a vertical vowel system? And how does it work? And how would it evolve, and how would it affect derived languages?
“Like billowing clouds, Like the incessant gurgle of the brook,
The longing of the spirit can never be stilled.” ― St. Hildegard von Bingen
User avatar
loglorn
mayan
mayan
Posts: 1728
Joined: 17 Mar 2014 03:22

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

Post by loglorn »

GoshDiggityDangit wrote: 05 Apr 2023 19:25 It's probably been discussed here before, but what is a vertical vowel system? And how does it work? And how would it evolve, and how would it affect derived languages?
A vertical vowel system is a system in which backness is not phonemic but height is, the most usual example being /a ə ɨ/. I'm not sure vertical systems with four or more vowels are at all attested and systems with two or less vowels are usually the result of some analytical acrobatics so you can safely say thats the most stable one. A variant where they are all analyzed as front (/i ɛ a/) is found in Whichita.

They are not very stable overall though, as they leave lots of room for allophonic backness which readily phonemicizes, and they usually occur in languages where backness is supplied elsewhere (i.e. there is a really pervasive palatalization or labialization contrast in consonants).

They arise by neutralizing backness contrasts, you can be creative there, and for deriving from a vertical vowel system my advice is capitalizing on the amount of allophony they tend to have (e.g. Arrernte /ə/ has the range [ɪ ~ e ~ ə ~ ʊ]), and i'd say the tendency is for them to redevelop a backness contrast as some of that allophony is phonemicized.
Diachronic Conlanging is the path to happiness, given time. [;)]

Gigxkpoyan Languages: CHÍFJAEŚÍ RETLA TLAPTHUV DÄLDLEN CJUŚËKNJU ṢATT

Other langs: Søsøzatli Kamëzet
Salmoneus
MVP
MVP
Posts: 3030
Joined: 19 Sep 2011 19:37

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

Post by Salmoneus »

A vertical vowel system is one that only has one spatial dimension, not two (and where that dimension is height - although I don't think 'horizontal' systems are actually attested, so that's kind of just a theoretical issue!).

They work by... well, having only one spatial dimension in their vowels. Not having a back vs front contrast. The term is specifically more likely to be used when the vowel phonemes have no consistent frontness value - that is, where front and back allophones are common. [so, a system like /e a u/ is in one sense 'vertical' (there's no phonemic backness), but if /e/ is always [e] and /u/ is always [ u ] then it's less likely (though not impossible) that the system will be called 'vertical].

Since these systems have many allophones, and frequently show extensive interaction between vowels and surrounding consonants (or other vowels and semivowels), it's often controversial how many vowel phonemes exist in a vertical system, and even whether a system is vertical. Or even whether 'phonemes' exist at all. In the case of Irish short vowels the system has been described by some as featuring 'quasiphonemes' - because on the one hand you can't always predict the phonetic vowel in a word unless you know whether it is phonemically front or back; and yet there may not be any phonemic minimal pairs between front and back vowels - that is, the quasiphonemes are distinctive but not contrastive. [Irish short vowels have apparently been analysed with at least seven different systems in the literature, ranging from a three-vowel vertical system up to three different six-vowel systems].



How does it evolve? Well, assuming it didn't evolve from a vertical vowel system, it must have its vowel system become vertical - i.e. lose the contrast between back and front vowels. I can think of three or four methods for this:
- vowels can simply move. Eg, a /i a u/ system could become /e a u/ could become /e a i/, leaving no back/front contrasts. Some American dialects of English are actually moving in this direction, though fronting or lowering/derounding of all back rounded vowels.

- features of the vowels can move to surrounding consonants, removing the contrasts from the vowels (though not producing any new minimal pairs necessarily)

- features of surrounding consonants can 'dominate' vowel contrasts (eliminating old minimal pairs).

- vowel attraction and harmony can remove contrasts from individual vowels, leaving only a suprasegmental featue (which could be neutralised or transformed in some way) or leaving all such features predictable from a single (first or last) vowel that could then have its features neutralised in some way.

Verticality can happen in one fell swoop (eg /i a u/ can become /i a/ simply by saying 'consonants preceding /u/ become labialised and /u/ merges with /i/'. Or, it can occur as the result of multiple shifts conspiring. Quasiverticality in Irish short vowels is an example of the latter: vowel frontness has resulted in palatalisation of adjacent consonants; palatalisation of adjacent consonants has dominated the realisation of adjacent vowels, as has the primary POA of some consonants; and remaining vowel contrasts have been eliminated though subsequent shifts and mergers (eg /e/ and /o/ would have been left contrastive before non-palatalised coronal consonants, but aren't, because /e/ subsequently 'coincidentally' merged with /a/ in that environment)


--------

As for what happens to them: they tend to become non-vertical. This can be because allophones are phonemicised by the loss of their generating contexts, or because loanwords have broken the allophony system, or simply because vowel shifts have occured.
User avatar
Acipencer
sinic
sinic
Posts: 327
Joined: 27 Jun 2021 08:39
Location: Somewhere

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

Post by Acipencer »

Hi, I have a question about a feature in one of my conlangs. This language has two separate genitive cases; one which is used when the possessee is masculine, and one which is used when the possessee is feminine.

The thing is, I am almost certain this is ANADEW and I have been looking online for similar features in natural languages so I can gain some more insight into how natural languages use them. But I haven't found much online so far. Do you know of any natural languages where this occurs or if there is a specific linguistic term for these gender-agreeing genitives?
User avatar
VaptuantaDoi
roman
roman
Posts: 1067
Joined: 18 Nov 2019 07:35

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

Post by VaptuantaDoi »

Acipencer wrote: 07 Apr 2023 09:28 Hi, I have a question about a feature in one of my conlangs. This language has two separate genitive cases; one which is used when the possessee is masculine, and one which is used when the possessee is feminine.

The thing is, I am almost certain this is ANADEW and I have been looking online for similar features in natural languages so I can gain some more insight into how natural languages use them. But I haven't found much online so far. Do you know of any natural languages where this occurs or if there is a specific linguistic term for these gender-agreeing genitives?
While I'm not sure if there are any good examples IRL, there is some vaguely similar precedent. Some linguists make the distinction between clausal cases, which mark the type of argument a noun is while retaining its status as an NP head (i.e. nominative/accusative/dative); and phrasal cases, which turn nouns into modifiers which can be used as dependents of other nouns (hence including the genitive). Phrasal cases can behave less like case suffixes and more like derivational suffixes; e.g. in Australian languages the genitive can be stacked on top of clausal case markers. I think it's pretty plausible that a language could develop gender marking on the genitive on this basis.

A bit more research shows that this is indeed ANADEW, attested in Romani (here's a kinda messy-looking but vaguely reliable source)

Me phraleskere khereskoro vudar.
me
1SG.POSS
phral
brother
-es
OBL
-kere
GEN.NEUT
kher
house
-es
OBL
-koro
GEN.MASC
vudar
doot

'The door of my brother's house'

Note also the stacking with the oblique. Apparently in this case, they're grammaticalised clitics which derive ultimately from the participle of Sanskrit kṛ "do, make".
Salmoneus
MVP
MVP
Posts: 3030
Joined: 19 Sep 2011 19:37

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

Post by Salmoneus »

A comparison there would be a relative clause with a SAE possessive perfect.

We could imagine expressing "My papers" as "the papers (that) I have owned". In early forms of the SAE perfect, the participle, "owned", agreed in gender with the head, "papers". If such a relative clause were fossilised as a genitive construction, the gender marking could get embedded into it.


A completely different route might be through a sort of classifier or appositive: "My thing, the house". We could imagine the 'classifier' matching the true head in gender. So for an English example we could use mass vs count nouns as our 'gender' and contrast "my thing, the house" with "my stuff, the petrol". This 'classifier' could then be fossilised into an affix.

[Why would that construction be used? Well, who knows. But, for example, perhaps there's originally a distinction between direct and indirect possession markers, with only some nounds able to take the former ("my thing" vs "the house belonging to me"), and then the latter is stylistically avoided in favour of appositive constructions using the former?]
User avatar
Omzinesý
mongolian
mongolian
Posts: 4079
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

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

Post by Omzinesý »

Acipencer wrote: 07 Apr 2023 09:28 Hi, I have a question about a feature in one of my conlangs. This language has two separate genitive cases; one which is used when the possessee is masculine, and one which is used when the possessee is feminine.

The thing is, I am almost certain this is ANADEW and I have been looking online for similar features in natural languages so I can gain some more insight into how natural languages use them. But I haven't found much online so far. Do you know of any natural languages where this occurs or if there is a specific linguistic term for these gender-agreeing genitives?
Hindi का kā is like that https://en.m.wiktionary.org/wiki/%E0%A4%95%E0%A4%BE
My meta-thread: viewtopic.php?f=6&t=5760
User avatar
Sequor
sinic
sinic
Posts: 438
Joined: 30 Jun 2012 06:13

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

Post by Sequor »

Acipencer wrote: 07 Apr 2023 09:28 Hi, I have a question about a feature in one of my conlangs. This language has two separate genitive cases; one which is used when the possessee is masculine, and one which is used when the possessee is feminine.

The thing is, I am almost certain this is ANADEW and I have been looking online for similar features in natural languages so I can gain some more insight into how natural languages use them. But I haven't found much online so far. Do you know of any natural languages where this occurs or if there is a specific linguistic term for these gender-agreeing genitives?
Note that the term "genitive" is usually used in grammars for the possessor of something, not the possessed entity (the possessee). For the possessee, you may want to talk about a "construct" case (or state) as in the grammars of Semitic languages, or "possessed" case as in Nahuatl grammars (and maybe other Uto-Aztecan languages). In both Semitic and Nahuatl, the non-possessed form is referred to as the "absolut(iv)e".

The Romani and Hindi examples above are actually about the possessor.

I think an example of what you're asking about may actually be your typical modern dialect of Arabic, where masculine nouns are zero marked in the construct, but feminine nouns are usually marked with -a in the absolutive which becomes -at in the construct.

ʔesˤbɑʕ - finger

ʔesˤbɑʕ hodæ
Hoda's finger

sˤɑħebæ - female friend

sˤɑħebæt hodæ
Hoda's (female) friend
hīc sunt linguificēs. hēr bēoþ tungemakeras.
User avatar
qwed117
mongolian
mongolian
Posts: 4094
Joined: 20 Nov 2014 02:27

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

Post by qwed117 »

Sequor wrote: 07 Apr 2023 23:07
Acipencer wrote: 07 Apr 2023 09:28 Hi, I have a question about a feature in one of my conlangs. This language has two separate genitive cases; one which is used when the possessee is masculine, and one which is used when the possessee is feminine.

The thing is, I am almost certain this is ANADEW and I have been looking online for similar features in natural languages so I can gain some more insight into how natural languages use them. But I haven't found much online so far. Do you know of any natural languages where this occurs or if there is a specific linguistic term for these gender-agreeing genitives?
Note that the term "genitive" is usually used in grammars for the possessor of something, not the possessed entity (the possessee). For the possessee, you may want to talk about a "construct" case (or state) as in the grammars of Semitic languages, or "possessed" case as in Nahuatl grammars (and maybe other Uto-Aztecan languages). In both Semitic and Nahuatl, the non-possessed form is referred to as the "absolut(iv)e".

The Romani and Hindi examples above are actually about the possessor.

I think an example of what you're asking about may actually be your typical modern dialect of Arabic, where masculine nouns are zero marked in the construct, but feminine nouns are usually marked with -a in the absolutive which becomes -at in the construct.

ʔesˤbɑʕ - finger

ʔesˤbɑʕ hodæ
Hoda's finger

sˤɑħebæ - female friend

sˤɑħebæt hodæ
Hoda's (female) friend
from what I can tell Acipencer means genitive ie. that "the mans' woman" would be glossed as MAN-GEN.F woman. This *feels* ANADEW to me, but I don't really have any examples. It's at least pretty common in pronominal constructions, at least in what European languages I know. For example, Sardinian (from the Lord's Prayer)

Code: Select all

s-u   rennu    to-u
DET-M reign(M) 2S.GEN-M
"your reign"

s-a   boluntade  tu-a
DET-F will(F)    2S.GEN-F
"your will" 
This compares to the English use of his/her which relates to the gender of the possessor. I want to say that German does double marking, but I don't know for sure.
Spoiler:
My minicity is [http://zyphrazia.myminicity.com/xml]Zyphrazia and [http://novland.myminicity.com/xml]Novland.

Minicity has fallen :(
The SqwedgePad
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

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

Post by Creyeditor »

Okay, so as mentioned before possessor agreement is a thing in natlangs. The example here seems to involve posessee agreement, which is rare. I think it is rare for independent reasons though. Just like subject agreement is vastly more common than object agreement, possessor agreement is vastly more common than possessee agreement, but in addition, argument indexing on the verb is more common than agreement on nouns in an adnominal possession construction. So a lot left for possesse agreement. Closest thing that comes to my mind is different genitive cases that code noun classes. This is sometimes called possessive classification and there is a WALS chapter on it.

Of course, one could argue that colloquial German marks possesse gender (and possessor gender) in adnominal possession, I am just not too sure about a genitive case being involved. Have some glosses examples:

dem Vater sein Hund
the.DAT.M father(M) his.M dog(M)
'the father's dog'

dem Vater seine Katze
the.DAT.M father(M) his.F cat(F)
'the father's cat'

der Mutter ihr Hund
the.DAT.F mother(F) her.M dog(M)
'the mother's dog'

der Mutter ihre Katze
the.DAT.F mother(F) her.F cat(F)
'the mother's cat'
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]
Salmoneus
MVP
MVP
Posts: 3030
Joined: 19 Sep 2011 19:37

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

Post by Salmoneus »

Creyeditor wrote: 08 Apr 2023 00:44 Okay, so as mentioned before possessor agreement is a thing in natlangs. The example here seems to involve posessee agreement, which is rare. I think it is rare for independent reasons though. Just like subject agreement is vastly more common than object agreement, possessor agreement is vastly more common than possessee agreement, but in addition, argument indexing on the verb is more common than agreement on nouns in an adnominal possession construction. So a lot left for possesse agreement. Closest thing that comes to my mind is different genitive cases that code noun classes. This is sometimes called possessive classification and there is a WALS chapter on it.
I would just tangentially interject, however, to be aware that most examples of what the rest of the world calls possessive classification wouldn't qualify as possessive classification by WALS' standards, or be relevant to the question per se. [and I suspect that some of the ones they include they probably shouldn't]

[WALS (and OP) are talking about lexical classification, whereas the famous possessive classification systems of Oceanic, while originally considered lexical, are actually essentially semantic - the apparent lexical class of some nouns is just an artifact of the difficulty in finding them with different semantics. But, for instance, Blust gives the example of Kwaio susuna, "her breast" (part of her body), vs susu ana, "it's breast" (the breast a baby is suckling at). What is being encoded is the relationship between possessor and possessed, not really a lexical property of the possessed noun. Similarly, as I understand it, North American possessive classification by shape can sometimes by 'defeated' (made to use non-normal values) when items appear that are not the shape usually associated with that thing (that is, for at least some of these languages the classifier is still semantically marking the shape of the object, rather than just lexically marking the class of the noun - it's just difficult to tell because nouns refer to objects that tend to have certain shapes!).]
teotlxixtli
cuneiform
cuneiform
Posts: 85
Joined: 05 Jan 2021 04:37

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

Post by teotlxixtli »

What vocabulary is almost never derived?
I know that demonstratives are pretty much never derived, and personal pronouns are only occasionally so, but since adpositions can come from verbs or possessive phrases, and conjunctions like "and" (at least in English, anyway) derive from adjectives or adverbs like "abutting" or "facing", what parts of the lexicon are basically always root words?
Not knowing the answer to a question like this has tripped me up while devising lexicons for my conlangs and it might make that process easier.
User avatar
Omzinesý
mongolian
mongolian
Posts: 4079
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

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

Post by Omzinesý »

I decided that Dlor demonstratives are
sóon 'this'
and
sen 'that'.

They consist of the definite article s- and demonstrative parts óon and en. What could óon and en mean without the article? I would not like to make them mean 'here' and 'there' cos its very European.
My meta-thread: viewtopic.php?f=6&t=5760
User avatar
sangi39
moderator
moderator
Posts: 3024
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

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

Post by sangi39 »

Omzinesý wrote: 14 Apr 2023 12:43 I decided that Dlor demonstratives are
sóon 'this'
and
sen 'that'.

They consist of the definite article s- and demonstrative parts óon and en. What could óon and en mean without the article? I would not like to make them mean 'here' and 'there' cos its very European.
Could be that the definite article used to mean "this" or "that" (like in a number of languages), and then have óon and en be fossilised forms meaning something like "at me" and "at you", either surviving into the modern language (as stand-alone words as they used to be, as bound morphemes, maybe marking possession, many options), or not surviving into the modern language at all outside of the demonstratives
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
sangi39
moderator
moderator
Posts: 3024
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

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

Post by sangi39 »

teotlxixtli wrote: 12 Apr 2023 02:30 What vocabulary is almost never derived?
I know that demonstratives are pretty much never derived, and personal pronouns are only occasionally so, but since adpositions can come from verbs or possessive phrases, and conjunctions like "and" (at least in English, anyway) derive from adjectives or adverbs like "abutting" or "facing", what parts of the lexicon are basically always root words?
Not knowing the answer to a question like this has tripped me up while devising lexicons for my conlangs and it might make that process easier.
I could be wrong, but I feel nouns that aren't geographically, culturally, or technologically specific tend to be more resistent to being derived (so things like body parts, the sun, the moon, clouds, stone, etc.), at least in that, as languages change, they tend to be replaced less (for example, when words become homophonous, it's maybe more likely that the "basic" word remains, while the "less basic" word gets replaced), which from what I can remember is the concept behind the creation of the Swadesh List. Similarly for verbs as well, on that note, like "eat", "drink", "walk", etc.

On the other hand, however, it does seem that all words can be derived, given the right pressures to do so (whether those pressures be the result of sound change, culture, history, etc.)

But, yeah, off the top of my head, I can't think of any particular class of words, or subset of a class, that are always, or almost always, roots
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.
BarkMiner
rupestrian
rupestrian
Posts: 13
Joined: 30 Mar 2023 17:28

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

Post by BarkMiner »

Quick question, if my speakers interpret a consonant cluster as a unique phoneme, does it make sense to Romanize it to the same characters as the consonants in the cluster, or use a modified character to match with how my speakers write and think about it. Specific example: the speakers of my language have separate characters for the consonants and consonant cluster /x/, /w/, /xw/. Would it make sense to write /xw/ simply as <xw>, or to mark it with a diacritic or something like <x́>?
User avatar
Omzinesý
mongolian
mongolian
Posts: 4079
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

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

Post by Omzinesý »

sangi39 wrote: 14 Apr 2023 13:35
Omzinesý wrote: 14 Apr 2023 12:43 I decided that Dlor demonstratives are
sóon 'this'
and
sen 'that'.

They consist of the definite article s- and demonstrative parts óon and en. What could óon and en mean without the article? I would not like to make them mean 'here' and 'there' cos its very European.
Could be that the definite article used to mean "this" or "that" (like in a number of languages), and then have óon and en be fossilised forms meaning something like "at me" and "at you", either surviving into the modern language (as stand-alone words as they used to be, as bound morphemes, maybe marking possession, many options), or not surviving into the modern language at all outside of the demonstratives
Possessive pronouns could be an interesting alternative. My understanding is though that it's somewhat 'unnatural'/rare that demonstratives and posessive/personal pronouns are somehow related.
My meta-thread: viewtopic.php?f=6&t=5760
Post Reply