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

A forum for all topics related to constructed languages
yangfiretiger121
cuneiform
cuneiform
Posts: 126
Joined: 17 Jun 2018 02:04

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

Post by yangfiretiger121 » 26 Oct 2018 18:00

Is a medial [sʍ] required to get [st → sw]?
Ddoean's document
Paleofonts (for Linear-B characters)

shimobaatar
darkness
darkness
Posts: 10911
Joined: 12 Jul 2013 22:09
Location: PA → IN

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

Post by shimobaatar » 26 Oct 2018 18:09

yangfiretiger121 wrote:
26 Oct 2018 18:00
Is a medial [sʍ] required to get [st → sw]?
What conditions this change?

yangfiretiger121
cuneiform
cuneiform
Posts: 126
Joined: 17 Jun 2018 02:04

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

Post by yangfiretiger121 » 26 Oct 2018 20:35

shimobaatar wrote:
26 Oct 2018 18:09
yangfiretiger121 wrote:
26 Oct 2018 18:00
Is a medial [sʍ] required to get [st → sw]?
What conditions this change?
As of now, <st> changes as follows: [st → çt → ç] or [st → çθ → ɬ]. However, the only word it effects currently is Styks ([stɪks → çtɪkç → çɪç]; Styks (second) and Sis (third) names), and I don't like the final pronunciation for a name. Granted, Svis [çwɪç] is iffy as a name as well.

Addendum for anyone who wishes to help:
I'd like to do something similar to the Japanese yōon with my conlang's r-series characters (demoed hereafter with Japanese text). Is there a way other than using a sequence similar to ta-ra with a smaller r-series character (たら [tr̥ɑ̟] (correct characters); ちゃ (correct display; Japanese [tɕa]))?
Ddoean's document
Paleofonts (for Linear-B characters)

User avatar
Davush
greek
greek
Posts: 500
Joined: 10 Jan 2015 14:10

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

Post by Davush » 29 Oct 2018 14:11

I want to allow initial /zb zd zg/ and /qz/ in Qutrussan, but I'm trying to explain how this arose as Proto-Q did not have /z/.

Is the following reasonable or too far-fetched?

Unstressed initial *i- was a common prefix. Post-tonic voicing is a regular sound change.

*i-par > *i-bar

This *i- fortifies to /dz/ (or /z/). Giving z-bar.

I also want /qz/, but probably not /bz gz/. I was thinking of *k'i- >/qz/ via fortification of *i after k', but not elsewhere. I.e. *pˈi- > pi/bi. Not sure how reasonable such a conditioned change is though.

User avatar
sangi39
moderator
moderator
Posts: 3134
Joined: 12 Aug 2010 00:53
Location: North Yorkshire, UK

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

Post by sangi39 » 29 Oct 2018 15:15

Davush wrote:
29 Oct 2018 14:11
I want to allow initial /zb zd zg/ and /qz/ in Qutrussan, but I'm trying to explain how this arose as Proto-Q did not have /z/.

Is the following reasonable or too far-fetched?

Unstressed initial *i- was a common prefix. Post-tonic voicing is a regular sound change.

*i-par > *i-bar

This *i- fortifies to /dz/ (or /z/). Giving z-bar.

I also want /qz/, but probably not /bz gz/. I was thinking of *k'i- >/qz/ via fortification of *i after k', but not elsewhere. I.e. *pˈi- > pi/bi. Not sure how reasonable such a conditioned change is though.
Do they contrast as at all with /sb sd sg/, is there a voiced uvular plosive, and does /z/ appear anywhere else outside of these clusters?
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
Pabappa
cuneiform
cuneiform
Posts: 198
Joined: 18 Nov 2017 02:41
Contact:

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

Post by Pabappa » 29 Oct 2018 17:35

It sounds farfetched, yes, but not impossible. Mandarin has syllabic /z/, but it's just an extension of the previous consonant ... there was no direct change of /i/ > /z/. I could see a clearer path if the elided prefix had a consonant.

Post tonic means after stress, so I'm not sure if that's what you meant to say.

I've used the shift /žp žt žk/ > /žb žt žg/ myself, but I'm not aware if any natlang precedent .... usually voicing assimilation favors stops and goes the other direction if it's indifferent to sonority.
Image

User avatar
LinguistCat
cuneiform
cuneiform
Posts: 146
Joined: 06 May 2017 06:48

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

Post by LinguistCat » 29 Oct 2018 18:11

Question moved elsewhere.
Last edited by LinguistCat on 30 Oct 2018 02:36, edited 1 time in total.

User avatar
Davush
greek
greek
Posts: 500
Joined: 10 Jan 2015 14:10

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

Post by Davush » 29 Oct 2018 20:20

sangi39 wrote:
29 Oct 2018 15:15

Do they contrast as at all with /sb sd sg/, is there a voiced uvular plosive, and does /z/ appear anywhere else outside of these clusters?
No they don't. /z/ does appear, although usually as /dz/, developing from fortification of */jj/. It may also be realised as [dz] in these clusters./q/ varies between [q~ɢ] with /qz/ probably being [ɢdz].
Pabappa wrote:
29 Oct 2018 17:35
It sounds farfetched, yes, but not impossible. Mandarin has syllabic /z/, but it's just an extension of the previous consonant ... there was no direct change of /i/ > /z/. I could see a clearer path if the elided prefix had a consonant.

Post tonic means after stress, so I'm not sure if that's what you meant to say.

I've used the shift /žp žt žk/ > /žb žt žg/ myself, but I'm not aware if any natlang precedent .... usually voicing assimilation favors stops and goes the other direction if it's indifferent to sonority.
Yes, I meant pre-tonic, sorry. Pre-tonic vowels cause voicing as a general rule. Perhaps word-initial pre-tonic */i/ > /dz/ is just a weird outlier. Although with */jj/ > /dz/, it might not be that strange.

So the steps would be: *i.ˈpar > *i.ˈbar > zbar /dzbar/

User avatar
sangi39
moderator
moderator
Posts: 3134
Joined: 12 Aug 2010 00:53
Location: North Yorkshire, UK

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

Post by sangi39 » 29 Oct 2018 20:35

Davush wrote:
29 Oct 2018 20:20
sangi39 wrote:
29 Oct 2018 15:15

Do they contrast as at all with /sb sd sg/, is there a voiced uvular plosive, and does /z/ appear anywhere else outside of these clusters?
No they don't. /z/ does appear, although usually as /dz/, developing from fortification of */jj/. It may also be realised as [dz] in these clusters./q/ varies between [q~ɢ] with /qz/ probably being [ɢdz].
Can /zb zd zg/ not just come from older instances of /Vsb Vsd Vsg/, dropping unstressed initial vowels (or even just have some prefix /s-/)? Similarly, if /qz/ is realised as [ɢ(d)z], then it could come from older /qs/ (/q/ could be [ɢ] syllable-finally, causing the following /s/ to voice, then drop any unstressed initial vowels if you want the cluster to appear word-initially).
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
Davush
greek
greek
Posts: 500
Joined: 10 Jan 2015 14:10

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

Post by Davush » 30 Oct 2018 11:21

sangi39 wrote:
29 Oct 2018 20:35


Can /zb zd zg/ not just come from older instances of /Vsb Vsd Vsg/, dropping unstressed initial vowels (or even just have some prefix /s-/)? Similarly, if /qz/ is realised as [ɢ(d)z], then it could come from older /qs/ (/q/ could be [ɢ] syllable-finally, causing the following /s/ to voice, then drop any unstressed initial vowels if you want the cluster to appear word-initially).
Hmm yes, I suppose that would be an easier route to take. /Cs/ clusters don't appear, but /sC/ clusters do in the Proto-language's very early stage, so maybe VsC > zC was an early change. Maybe /qz/ just underwent metathesis but /zb zd zg/ didn't?

User avatar
sangi39
moderator
moderator
Posts: 3134
Joined: 12 Aug 2010 00:53
Location: North Yorkshire, UK

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

Post by sangi39 » 30 Oct 2018 12:53

Davush wrote:
30 Oct 2018 11:21
sangi39 wrote:
29 Oct 2018 20:35


Can /zb zd zg/ not just come from older instances of /Vsb Vsd Vsg/, dropping unstressed initial vowels (or even just have some prefix /s-/)? Similarly, if /qz/ is realised as [ɢ(d)z], then it could come from older /qs/ (/q/ could be [ɢ] syllable-finally, causing the following /s/ to voice, then drop any unstressed initial vowels if you want the cluster to appear word-initially).
Hmm yes, I suppose that would be an easier route to take. /Cs/ clusters don't appear, but /sC/ clusters do in the Proto-language's very early stage, so maybe VsC > zC was an early change. Maybe /qz/ just underwent metathesis but /zb zd zg/ didn't?
Doesn't seem unreasonable. IIRC, metathesis can get really specific.
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.

yangfiretiger121
cuneiform
cuneiform
Posts: 126
Joined: 17 Jun 2018 02:04

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

Post by yangfiretiger121 » 30 Oct 2018 20:35

My conlang linites stops to fricatives before near-back vowels (cf. [bɑ̟ → βɑ̟]. It is plausible for [l] just to lateralize unlinited fricatives (cf. [ɕ]) and [r]? As of now, only [ɕ], [ʑ], [ʝ], and [r] get lateralized.
Ddoean's document
Paleofonts (for Linear-B characters)

User avatar
Creyeditor
mongolian
mongolian
Posts: 4493
Joined: 14 Aug 2012 18:32

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

Post by Creyeditor » 31 Oct 2018 11:11

yangfiretiger121 wrote:
30 Oct 2018 20:35
[...] It is plausible for [l] just to lateralize unlenited fricatives (cf. [ɕ]) and [r]? As of now, only [ɕ], [ʑ], [ʝ], and [r] get lateralized.
In general, this is plausible. [r] is not a fricative, it is a trill, but these can definitely pattern together. I am curious what you mean by lateralized. You are probably not talking about lateral release, because this happens mostly to stops. So my question: Are you talking about total assimilation ([ɕ], [ʑ], [ʝ], [r] -> [l] /[l]_ ) or only assimilation in laterality ([ɕ] ->[ɬ̠ʲ] /[l]_ , [ʑ] -> [ɮ̠ʲ] /[l]_ , [ʝ] ->[ʎ̝] /[l]_ , [r] -> [l]/[l]_ ). In other words: what is the result of this lateralization?
Also to come back to the original question. If some process does only apply to segments that are not derived by a certain process (lenition in your case), this is often called an opaque interaction.
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
LinguoFranco
sinic
sinic
Posts: 395
Joined: 20 Jul 2016 16:49

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

Post by LinguoFranco » 31 Oct 2018 15:48

I have a question about syllable structure and inflection. Let's say I have a conlang where the basic syllable is CV(C) and the (C) stands for a coda consonant, but it can only be /n r s k/ and I make a rule that says all verbs in the infinitive form take the suffix '-eli', Therefore, I could have something like 'koreli' (let's say it means "to sing"). No big deal, right? But what if I have another verb like 'taveli' ("to eat"). You could say it is [ta.ve.li], and the first person conjugation of it is 'tavoro'. However, would I have to remove the /v/ since it isn't normally possible? I'm asking because -'eli' is two syllables, and always means that a verb is infinitive, but /v/ would in 'taveli' would make it [ta.ve.li] instead of just [tav.e.li] or [tav.el.i]. So with the paramaters I have set for the phonotactics, I don't know if conjugation would be possible without dropping the /v/.

I don't know if my question is making sense or not.

yangfiretiger121
cuneiform
cuneiform
Posts: 126
Joined: 17 Jun 2018 02:04

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

Post by yangfiretiger121 » 31 Oct 2018 16:35

Creyeditor wrote:
31 Oct 2018 11:11
yangfiretiger121 wrote:
30 Oct 2018 20:35
[...] It is plausible for [l] just to lateralize unlenited fricatives (cf. [ɕ]) and [r]? As of now, only [ɕ], [ʑ], [ʝ], and [r] get lateralized.
In general, this is plausible. [r] is not a fricative, it is a trill, but these can definitely pattern together. I am curious what you mean by lateralized. You are probably not talking about lateral release, because this happens mostly to stops. So my question: Are you talking about total assimilation ([ɕ], [ʑ], [ʝ], [r] -> [l] /[l]_ ) or only assimilation in laterality ([ɕ] ->[ɬ̠ʲ] /[l]_ , [ʑ] -> [ɮ̠ʲ] /[l]_ , [ʝ] ->[ʎ̝] /[l]_ , [r] -> [l]/[l]_ ). In other words: what is the result of this lateralization?
Also to come back to the original question. If some process does only apply to segments that are not derived by a certain process (lenition in your case), this is often called an opaque interaction.
My earlier question about "lateral mutation/assimilation" originated because combinations of [l] and a fricative, variously all fricatives or unlenited fricatives, (cf. [lɕ] <ls>) were to assimilate to a lateral fricative (cf. [ʎ̝̊]) while [lr] was to assimilate into [ɮ˔]. Then, the lateral fricatives and lateral trill became phonemic through lenition (cf. [ɕɑ̟] → [ʎ̝̊ɑ̟]) before fronting (cf. [ʎ̝̊] → [ɬ]). However, I'd shifted [ʎ] (<j>) to [ʝ] by the time I asked this question. Currently, [ɕ] and [ʑ] lenite into [ç] and [ʝ], respectively, and I'm not sure if [lʝ] becomes [ʎ̝] or [ʝr] becomes [ʝ˔].
Ddoean's document
Paleofonts (for Linear-B characters)

User avatar
Dormouse559
moderator
moderator
Posts: 2650
Joined: 10 Nov 2012 20:52
Location: California

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

Post by Dormouse559 » 31 Oct 2018 17:23

LinguoFranco wrote:
31 Oct 2018 15:48
I have a question about syllable structure and inflection. Let's say I have a conlang where the basic syllable is CV(C) and the (C) stands for a coda consonant, but it can only be /n r s k/ and I make a rule that says all verbs in the infinitive form take the suffix '-eli', Therefore, I could have something like 'koreli' (let's say it means "to sing"). No big deal, right? But what if I have another verb like 'taveli' ("to eat"). You could say it is [ta.ve.li], and the first person conjugation of it is 'tavoro'. However, would I have to remove the /v/ since it isn't normally possible? I'm asking because -'eli' is two syllables, and always means that a verb is infinitive, but /v/ would in 'taveli' would make it [ta.ve.li] instead of just [tav.e.li] or [tav.el.i]. So with the paramaters I have set for the phonotactics, I don't know if conjugation would be possible without dropping the /v/.

I don't know if my question is making sense or not.
I don't fully understand. I'll just write about a few things you said, and we'll see where I end up. [:P]

How a word is syllabified is somewhat language-dependent. You seem to have taken a morphological approach, where a root isn't resyllabified, regardless of the phonological context (presumably koreli would break down as /kor.e.li/). Another option is onset maximization, which says if a segment can be interpreted as an onset, it must be (syllabifying koreli as /ko.re.li/). As you can see, this approach disregards morphology, so the second syllable contains segments from two morphemes; that's not a problem for languages that use this method. A third option is coda maximization, which says if a segment can be interpreted as a coda, it must be (syllabifying koreli as /kor.el.i/). Since this method also ignores morphology, it's certainly possible that parts of more than one morpheme could end up in the same syllable.

All that said, there's a consequence of your syllable structure that I don't think you've taken into account. Given a template of CV(C), every syllable must begin with a consonant. So of the three syllabifications I gave for koreli (kor.e.li, ko.re.li), only the second one, the onset-maximizing one, is valid. Likewise, taveli can only be syllabified through onset maximization.

EDIT: Those conclusions are based on the assumption that both your syllable structure and the words/morphemes are as you intended. You could change either one in such a way that a morphological syllabification becomes valid.

EDIT2: Forgot /l/ isn't a coda consonant. That makes the results of the morphological and coda-maximizing strategies the same in this case, but my point stands.

User avatar
sangi39
moderator
moderator
Posts: 3134
Joined: 12 Aug 2010 00:53
Location: North Yorkshire, UK

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

Post by sangi39 » 31 Oct 2018 17:48

LinguoFranco wrote:
31 Oct 2018 15:48
I have a question about syllable structure and inflection. Let's say I have a conlang where the basic syllable is CV(C) and the (C) stands for a coda consonant, but it can only be /n r s k/ and I make a rule that says all verbs in the infinitive form take the suffix '-eli', Therefore, I could have something like 'koreli' (let's say it means "to sing"). No big deal, right? But what if I have another verb like 'taveli' ("to eat"). You could say it is [ta.ve.li], and the first person conjugation of it is 'tavoro'. However, would I have to remove the /v/ since it isn't normally possible? I'm asking because -'eli' is two syllables, and always means that a verb is infinitive, but /v/ would in 'taveli' would make it [ta.ve.li] instead of just [tav.e.li] or [tav.el.i]. So with the paramaters I have set for the phonotactics, I don't know if conjugation would be possible without dropping the /v/.

I don't know if my question is making sense or not.
I think Japanese does something similar. It's syllable structure is generally (C)(j)V(:/N/Q), but basically the only consonant that can end a syllable is /N/ (excluding geminates for now). However, Japanese verbs are still divided into consonant-final stems and vowel-final stems, so that, for example, "kaku", the dictionary form of "write" is kak-u, not ka-ku.

This would seem to be the case in your conlang, i.e. tav- is a consonantal stem of, say, type 1, while kor- is a consonantal stem of type 2. The additon of -eli just means that the infinitive of either type of consonantal stem ends up fitting the syllable structure.
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.

holbuzvala
cuneiform
cuneiform
Posts: 160
Joined: 01 Jan 2017 14:03

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

Post by holbuzvala » 01 Nov 2018 12:13

Noun Incorporation into Verb Complexes: Cases Maintained?

So my conlang features some noun incorporation into verb complexes. It also features some case-marking. If a given noun is marked when it's the direct object of a verb, do you think it would retain that marking when incorporated into the verb?

For example:
- haulek = fish.NOM
- taulek = fish.ACC
- mata = hunt.INF
- ki_ = human agent verb prefix
- _je_ = animate object verb prefix

1. He is hunting a fish = taulek kijemata

But if we wanted to say 'He is fish-hunting', would it be:

"kihaulekmata" OR "kitaulekmata"?

shimobaatar
darkness
darkness
Posts: 10911
Joined: 12 Jul 2013 22:09
Location: PA → IN

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

Post by shimobaatar » 01 Nov 2018 12:26

holbuzvala wrote:
01 Nov 2018 12:13
If a given noun is marked when it's the direct object of a verb, do you think it would retain that marking when incorporated into the verb?
I think it would not.

User avatar
Pabappa
cuneiform
cuneiform
Posts: 198
Joined: 18 Nov 2017 02:41
Contact:

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

Post by Pabappa » 01 Nov 2018 14:18

In most languages, case marking is dropped because the intended sense is always accusative. Is haulek the unmarked form, or is even this marked with a prefix like h-? The bare stem is what I recommend.

I have violated this rule a few times myself, though, I will admit. Why? Because sometimes keeping the case marker on makes a better sounding word, and that's important to me.
Image

Post Reply