(Conlangs) Q&A Thread - Quick questions go here [2010-2020]

[Shemtov]

Is there a way that leaves the genitive untouched, but changes the Nom. and Acc. to Erg. and Abs., and if so, which case becomes which?

[Davush]

There are a good number of monosyllabic words, although they could be any of L, M, or H. This is where tonal harmony comes in; the tones in an affix would have to change if necessary to fit a contour, usually, but not always, in the direction of the tone of the root:

L + LL > LML > LHL
M + HH > MMH > LLH
M + LH > MMH > LLH
M + HL > MML > HHL
H + MM > HML or HHL

Does that mean that the first tone spreads to the second syllable?

I guess so. But the mid-tone should not be perceived as atonal.

Is it possible that the peak of the high toned syllable will be perceived as 'stressed' and the system ends up being a gradual up-shift in pitch towards the 'stress' or gradual down-shift away from it? I may be getting this completely wrong, but that is how I perceived it when I said those examples out loud.

mùpuná - pitch lowest on mu, bit higher on pu, and highest on na.
múpunà - same but reversed.

Would something line MMH contrast with LMH and HMH? If not it seems to resemble a pitch accent more like the Japanese down step maybe?

I didn't think tonal languages would have stress, but it would probably have stress follow the highest syllable.
Yes, it constrasts MMH and MML, but I think such a contrast would be quickly assimilated into LLH and HHL. Same goes for MLM and MHM.

I think what I'm trying to say is, if the contrast ends up basically being L vs H, and how they interact with suffixes is predictable, then it seems more like a pitch-accent system than purely tonal? I.e. in those examples, you could use the Japanese notation of downstep, where the syllable before the down step is accented (and receives the highest pitch I think), and everything before it (within a word) gradually rises to that pitch, then falls:

HHL > kakáka
LLH > kakaká

If monosyllabic do appear as monosyllables and contrast LMH, though, I think that would count as 'properly' tonal or a mixed system which is actually pretty interesting. Tonal monosyllables, pitch accented otherwise.

[Creyeditor]

Is there a way that leaves the genitive untouched, but changes the Nom. and Acc. to Erg. and Abs., and if so, which case becomes which?

I think the easiest way would be from a marked nominative language (where accusative case is not marked by any affix) to a ergative language, where you would just have to change the marking of intransitive subjects from the old marked nominative case, to the now emerging absolutive case (which was an unmarked accusative case). This would make the older marked nominative case become an ergative automatically.
So maybe if you start with a language where both accusative and nominative case are marked, you could just erode the accusative case first. So let's schematize that. We start with a language were the intransitive subject (S) and the subject of a transitive verb (A) are marked with a marker -i, whereas the object of a transitive verb (O) is marked with a marker -j:

Stage I:
S: -i
A: -i
O: -j

Now in the second stage the accusative case gets eroded, i.e. the object is now unmarked.

Stage II:
S: -i
A: -i
O: -0

In the third stage, the subject of intransitive verbs loses its overt nominative marker, because in intransitive clauses there is no need to distinguish it from any other argument.

Stage III:
S: -0
A: -i
O: -0

Viola, an ergative language

[Ahzoh]

I think what I'm trying to say is, if the contrast ends up basically being L vs H, and how they interact with suffixes is predictable, then it seems more like a pitch-accent system than purely tonal? I.e. in those examples, you could use the Japanese notation of downstep, where the syllable before the down step is accented (and receives the highest pitch I think), and everything before it (within a word) gradually rises to that pitch, then falls:

HHL > kakáka
LLH > kakaká

If monosyllabic do appear as monosyllables and contrast LMH, though, I think that would count as 'properly' tonal or a mixed system which is actually pretty interesting. Tonal monosyllables, pitch accented otherwise.

It can't be pitch accent because in a pitch accent only one syllable in a word ever is phonemically tonic, whereas this language, all syllables are phonemically tonic, and there's stress (which causes allophonic lengthening).

[Davush]

I think what I'm trying to say is, if the contrast ends up basically being L vs H, and how they interact with suffixes is predictable, then it seems more like a pitch-accent system than purely tonal? I.e. in those examples, you could use the Japanese notation of downstep, where the syllable before the down step is accented (and receives the highest pitch I think), and everything before it (within a word) gradually rises to that pitch, then falls:

HHL > kakáka
LLH > kakaká

If monosyllabic do appear as monosyllables and contrast LMH, though, I think that would count as 'properly' tonal or a mixed system which is actually pretty interesting. Tonal monosyllables, pitch accented otherwise.

It can't be pitch accent because in a pitch accent only one syllable in a word ever is phonemically tonic, whereas this language, all syllables are phonemically tonic.

I may be completely wrong, so apologies if so (and someone please correct my understanding), but if words (other than monosyllables) all look like LLH HLL LHL HLH, isn't that more or less like a downstep system except some words of the HLH shape might have two downsteps? I think I am thinking if e.g. LLH is realised as a low tone, mid tone and high tone - would the high tone be perceived as 'stress' or the 'tonic' syllable as a higher pitch is often associated with stress and isn't contrastive with, say, LLM which would mean the relative tone becomes more of a determining factor? I am just basing this on my own perception so I could be entirely wrong.

[Ahzoh]

I may be completely wrong, so apologies if so (and someone please correct my understanding), but if words (other than monosyllables) all look like LLH HLL LHL HLH, isn't that more or less like a downstep system except some words of the HLH shape might have two downsteps? I think I am thinking if e.g. LLH is realised as a low tone, mid tone and high tone - would the high tone be perceived as 'stress' or the 'tonic' syllable as a higher pitch is often associated with stress and isn't contrastive with, say, LLM which would mean the relative tone becomes more of a determining factor? I am just basing this on my own perception so I could be entirely wrong.

https://en.wikipedia.org/wiki/Downstep

...and LLH wouldn't be realized as Low-Mid-High (˩ ˧ ˥), it would be realized as Low-SemiLow-High (˩ ˨ ˥).

[Davush]

I may be completely wrong, so apologies if so (and someone please correct my understanding), but if words (other than monosyllables) all look like LLH HLL LHL HLH, isn't that more or less like a downstep system except some words of the HLH shape might have two downsteps? I think I am thinking if e.g. LLH is realised as a low tone, mid tone and high tone - would the high tone be perceived as 'stress' or the 'tonic' syllable as a higher pitch is often associated with stress and isn't contrastive with, say, LLM which would mean the relative tone becomes more of a determining factor? I am just basing this on my own perception so I could be entirely wrong.

https://en.wikipedia.org/wiki/Downstep

...and LLH wouldn't be realized as Low-Mid-High (˩ ˧ ˥), it would be realized as Low-SemiLow-High (˩ ˨ ˥).

Thanks - for some reason I just couldn't see how it was different, but after a bit of reading it looks like your lang was more inspired by a Bantu type system? I am always interested in tonal systems outside of Chinese, so please do post more on that language if you get a chance!

[DesEsseintes]

I’m not gonna fret over this unduly, but at the mo I’ve got /jy ɥi/ → wi ju in Núta diachronics. This seems somewhat counterintuitive at first glance, so I’m wondering what I could use as a justification.

How about a triphthongisation stage where sth like this happened:

jy → ɥuɪ̯ → ʊ̯i → wi
ɥi → ɥiʊ̯ → ɪ̯u → ju

There must be a better way... Any ideas?

[sangi39]

I’m not gonna fret over this unduly, but at the mo I’ve got /jy ɥi/ → wi ju in Núta diachronics. This seems somewhat counterintuitive at first glance, so I’m wondering what I could use as a justification.

How about a triphthongisation stage where sth like this happened:

jy → ɥuɪ̯ → ʊ̯i → wi
ɥi → ɥiʊ̯ → ɪ̯u → ju

There must be a better way... Any ideas?

Hmmm, how about:

jy > ji > i: > wi
ɥi > ɥy > y: > ju

/ɥ/ labialises the following /i/ while /j/ delabialises the following /y/ (alternatively, you could go through a stage in which /y/ and /i/ merge into /i/, with /ɥ/ then labialising the following /i/ but I guess that's just a question of timing), then it's just a case of diphthongisation.

[DesEsseintes]

I’m not gonna fret over this unduly, but at the mo I’ve got /jy ɥi/ → wi ju in Núta diachronics. This seems somewhat counterintuitive at first glance, so I’m wondering what I could use as a justification.

How about a triphthongisation stage where sth like this happened:

jy → ɥuɪ̯ → ʊ̯i → wi
ɥi → ɥiʊ̯ → ɪ̯u → ju

There must be a better way... Any ideas?

Hmmm, how about:

jy > ji > i: > wi
ɥi > ɥy > y: > ju

/ɥ/ labialises the following /i/ while /j/ delabialises the following /y/ (alternatively, you could go through a stage in which /y/ and /i/ merge into /i/, with /ɥ/ then labialising the following /i/ but I guess that's just a question of timing), then it's just a case of diphthongisation.

I’m afraid that doesn’t work, as I’ve got /i y/ contrasting with each other and /jy ɥi/ at the stage where this happens. That’s why I was straining things a bit. Thanks got the suggestion though.

Any other ways?

[sangi39]

I’m not gonna fret over this unduly, but at the mo I’ve got /jy ɥi/ → wi ju in Núta diachronics. This seems somewhat counterintuitive at first glance, so I’m wondering what I could use as a justification.

How about a triphthongisation stage where sth like this happened:

jy → ɥuɪ̯ → ʊ̯i → wi
ɥi → ɥiʊ̯ → ɪ̯u → ju

There must be a better way... Any ideas?

Hmmm, how about:

jy > ji > i: > wi
ɥi > ɥy > y: > ju

/ɥ/ labialises the following /i/ while /j/ delabialises the following /y/ (alternatively, you could go through a stage in which /y/ and /i/ merge into /i/, with /ɥ/ then labialising the following /i/ but I guess that's just a question of timing), then it's just a case of diphthongisation.

I’m afraid that doesn’t work, as I’ve got /i y/ contrasting with each other and /jy ɥi/ at the stage where this happens. That’s why I was straining things a bit. Thanks got the suggestion though.

Any other ways?

Do you have them contrasting with /ji/ and /ɥy/?

[Davush]

I’m not gonna fret over this unduly, but at the mo I’ve got /jy ɥi/ → wi ju in Núta diachronics. This seems somewhat counterintuitive at first glance, so I’m wondering what I could use as a justification.

How about a triphthongisation stage where sth like this happened:

jy → ɥuɪ̯ → ʊ̯i → wi
ɥi → ɥiʊ̯ → ɪ̯u → ju

There must be a better way... Any ideas?

Hmmm, how about:

jy > ji > i: > wi
ɥi > ɥy > y: > ju

/ɥ/ labialises the following /i/ while /j/ delabialises the following /y/ (alternatively, you could go through a stage in which /y/ and /i/ merge into /i/, with /ɥ/ then labialising the following /i/ but I guess that's just a question of timing), then it's just a case of diphthongisation.

I’m afraid that doesn’t work, as I’ve got /i y/ contrasting with each other and /jy ɥi/ at the stage where this happens. That’s why I was straining things a bit. Thanks got the suggestion though.

Any other ways?

How about /jy ɥi/ > /wy ɥy/ > /wi jy/ > /wi ju/? j>w because of rounding harmony before y, i>y for similar reason, then dissimilation?

[DesEsseintes]

I’m not gonna fret over this unduly, but at the mo I’ve got /jy ɥi/ → wi ju in Núta diachronics. This seems somewhat counterintuitive at first glance, so I’m wondering what I could use as a justification.

How about a triphthongisation stage where sth like this happened:

jy → ɥuɪ̯ → ʊ̯i → wi
ɥi → ɥiʊ̯ → ɪ̯u → ju

There must be a better way... Any ideas?

Hmmm, how about:

jy > ji > i: > wi
ɥi > ɥy > y: > ju

/ɥ/ labialises the following /i/ while /j/ delabialises the following /y/ (alternatively, you could go through a stage in which /y/ and /i/ merge into /i/, with /ɥ/ then labialising the following /i/ but I guess that's just a question of timing), then it's just a case of diphthongisation.

I’m afraid that doesn’t work, as I’ve got /i y/ contrasting with each other and /jy ɥi/ at the stage where this happens. That’s why I was straining things a bit. Thanks got the suggestion though.

Any other ways?

How about /jy ɥi/ > /wy ɥy/ > /wi jy/ > /wi ju/? j>w because of rounding harmony before y, i>y for similar reason, then dissimilation?

Hmm, let me think. I guess if /jy/ → /wy/ occurs first...

Lemme test this. Thanks!

[Ahzoh]

Revisiting my tone question, how naturalistic is it to have tonal harmony? As I said before, entire words can have one of a set of contours spread throughout the entire word and the addition of certain morphemes can even modify one contour into another, as shown below by the ergative marker, which heightens tone, and the genitive marker, which lowers tone:
HLH > HLM (downstep) > HML (tonal metathesis)
LHL > LHM (upstep) > LMH (tonal metathesis)

Relatedly, I don't know what possible contours these could become (the ones on the left and the ones above are the only allowed contours):
LMH > LMM (downstep) >
HML > HMM (upstep) >
LLH > LLM (downstep) > LHL?
HHL > HHM (upstep) > HLH?
MMH > MMM (downstep) >
MML > MMM (upstep) >

[Creyeditor]

To answer your first question: Yes, this is possible, even though your descriptions and the terms you use might seem a bit unorthodox.

Related to your second question. You should keep in mind that downstep is a global phenomenon, i.e. it applies to whole utterances. If a high tone is downstepped all following high tones are realized a bit lower. This makes it different from usual lowering, which is local and leaves any following high tones intact. This is why people often differentiate the two as M vs. ↓H or even as M vs. !H

[Davush]

I've become fixated on toponyms and their etymology...to the detriment of other aspects of my languages. I have a few questions: Can endonyms ever become toponyms, and are there any real word examples of this? Do toponyms ever show resistance to sound change or maybe behave in unexpected ways?

Is the following plausible:
I am thinking for a name for the region of Qutrussan's sister language, and I have come up with:

Qanshar /qanʃar/ (in Qutrussan)
Għāx /ʕa:ʃ/ (in the native sister language)

This derives from a non-Qutrussan word which was something like /ʁa:nʂ/ when the first Qutrussic people arrived, probably meaning 'desert' or similar. This was loaned into proto-Qutrussic as *ʕa:naʃ or *ʕa:ʃ (as the final cluster was maybe disputed?). Proto-Qutrussic *ʕ usually went to > /h/ in Qutrussan, but became /q/ due to analogy with Qutrus and Qashrus giving the three Qs (which I like): Qutrus, Qashrus, Qanshar. The -ar suffix remains to be explained. In the sister language /ʕ/ remains and sound change hasn't effected the word much, giving the name Għāx /ʕa:ʃ/.

[Ahzoh]

Why not?:
/ʁ/ (as a "loaneme") > /ɢ/ (word-initial fortition) > /q/
/ʁ/ > /ʕ/

---

To answer your first question: Yes, this is possible, even though your descriptions and the terms you use might seem a bit unorthodox.

The problem is I don't know how to describe it.

Related to your second question. You should keep in mind that downstep is a global phenomenon, i.e. it applies to whole utterances. If a high tone is downstepped all following high tones are realized a bit lower. This makes it different from usual lowering, which is local and leaves any following high tones intact. This is why people often differentiate the two as M vs. ↓H or even as M vs. !H

Right, I had mistakenly thought these were words for simply lowering or raising a tone. But, anyways, since there are a certain number of allowed tone "patterns" for a word, I don't know what to do about illegal patterns.

I'm also thinking of having these tonal patterns on a morpheme-level instead of a word-level.

[Davush]

Why not?:
/ʁ/ (as a "loaneme") > /ɢ/ (word-initial fortition) > /q/
/ʁ/ > /ʕ/

Thanks, the same occurred to me too, but proto-Qutrussic doesn't have /q/, this was a later development in Qutrussan. I suppose I could make that a *very* early change, though, i.e. more or less as soon as the first Qutrussic peoples arrived.

[Creyeditor]

To answer your first question: Yes, this is possible, even though your descriptions and the terms you use might seem a bit unorthodox.

The problem is I don't know how to describe it.

I think reading up on autosegmental representations of tone might really help you here, I feel that's what you are going for anyway. I might pm you something in a few days, if that's okay.

[Davush]

I have a few suffixes in Qutrussan that I am finding difficult to create a suitable etymology for.

The -nóya suffix is for people involved in some sort of agriculture or farming, but the form of this word actually suggests a verbal-noun 'farming, cultivation' as a common method of derivation is to add -a to the stem. How could this end up denoting the person doing the job, when such words are usually formed differently? I.e. a farmer would be nóyansa with the usual -ansa suffix comparable to English -er.