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

[Zekoslav]

A video by Biblaridion talks about this in some depth. The fact that multiple languages turned some kind of perfect marker into an evidential was surprising to me (I thought the situation in Bulgarian and Macedonian was due to Turkish influence, but Turkish seems to have done the same by itself... otherwise personal evidence past combining with hearsay evidence past to form a pluperfect makes absolutely no sense!).

[CarsonDaConlanger]

How naturalistic is it for a language to have an auxiliary (derived from a verb meaning do) that takes person/maybe tense (its an agglutinative language so the two are completely separate) that comes after a serial verb construction? It would replace person/tense on the meaning verbs in the construction.

[DesEsseintes]

How naturalistic is it for a language to have an auxiliary (derived from a verb meaning do) that takes person/maybe tense (its an agglutinative language so the two are completely separate) that comes after a serial verb construction? It would replace person/tense on the meaning verbs in the construction.

That sounds entirely reasonable. Although a serial verb interpretation may well be the best analysis for your language (as this depends on other aspects of grammar and syntax), I could also see this as a kind of nominalisation of the preceding verb string.

A: What did you do yesterday?
B: Oh, I just did some go.out-buy.stuff-ing.

[Creyeditor]

How naturalistic is it for a language to have an auxiliary (derived from a verb meaning do) that takes person/maybe tense (its an agglutinative language so the two are completely separate) that comes after a serial verb construction? It would replace person/tense on the meaning verbs in the construction.

If you wonder about the position of the auxiliary, I think there is a crosslinguistic tendency. SOV languages tend to be head final. In these languages auxiliaries tend to follow verbs. So a possible order would be Subject - Object - Verb (-Verb) - Auxiliary. Was that also what you were asking for?

[Ahzoh]

I don't know how naturalistic it is that the gender endings tend to be phonological opposites to the voice vowels.

e.g. /a u/ is basically the ablaut vowel for tenses in the active voice while /e i/ are the ablaut vowels for tenses in the passive voice. [saras vs sares, sarus vs saris]
But /a u/ are also the case endings for nouns in the feminine gender while /e i/ are the case endings for nouns in the masculine gender. [haze/hazi "man" vs sama/samu "woman"]

To compound the weirdness, deverbal nouns derived from verbs in the passive voice tend to have masculine gender while deverbal nouns derived from verbs in the passive voice tend to have feminine gender. Thus, it creates this weird alternation in words like such as a-mlek-a (passive deverbal) and a-mlak-e (active deverbal) which feels artificial.

[Dormouse559]

To compound the weirdness, deverbal nouns derived from verbs in the passive voice tend to have masculine gender while deverbal nouns derived from verbs in the passive voice tend to have feminine gender. Thus, it creates this weird alternation in words like such as a-mlek-a (passive deverbal) and a-mlak-e (active deverbal) which feels artificial.

This sounds like the sort of unplanned order emerging from interlocking systems that would give me the most exquisite of conlanging satisfaction. I will gladly take it off your hands if you don't want it.

[kiwikami]

Given all the changes I've made to Alál in the last few years, I'd like to get back to putting together proper documentation for it. Since I'll have to rewrite almost all of what I wrote in the existing Alál thread to bring it up to date, do y'all think it'd be more efficient / less messy to just start a new thread for it? I'm not certain if that'd be a forum faux pas (you'd think I'd pick up on these things after seven years here, but no), but it'd be nice to have a clean slate since I'll be starting the write-up from scratch.

[Dormouse559]

Given all the changes I've made to Alál in the last few years, I'd like to get back to putting together proper documentation for it. Since I'll have to rewrite almost all of what I wrote in the existing Alál thread to bring it up to date, do y'all think it'd be more efficient / less messy to just start a new thread for it? I'm not certain if that'd be a forum faux pas (you'd think I'd pick up on these things after seven years here, but no), but it'd be nice to have a clean slate since I'll be starting the write-up from scratch.

I don't believe there'd be a problem with that. It's been done before without issue.

[DesEsseintes]

Given all the changes I've made to Alál in the last few years, I'd like to get back to putting together proper documentation for it. Since I'll have to rewrite almost all of what I wrote in the existing Alál thread to bring it up to date, do y'all think it'd be more efficient / less messy to just start a new thread for it? I'm not certain if that'd be a forum faux pas (you'd think I'd pick up on these things after seven years here, but no), but it'd be nice to have a clean slate since I'll be starting the write-up from scratch.

I don't believe there'd be a problem with that. It's been done before without issue.

I think it’s more efficient and less confusing for others to start a new thread when languages have been overhauled, especially when said langs haven’t been posted on for a while.

I plan to start a new Híí thread if I ever start posting on it again.

As a rather keen Alál aficionado, I look forward to seeing the new stuff.

[yangfiretiger121]

Are rhotacized long monophthongs better transcribed as [Vːʴ] or [Vʴː]?

[dva_arla]

Is a gap in the voiceless and voiced plosives/affricates natural/realistic in a language with the 3-way distinction? My naturalistic conlang has:

/p ph b/
/t th d/
/k kh g/
/ts tsh dz/
/t̠ʃ t̠ʃh d̠ʒ/
/ʈʂh/

i.e. no /ʈʂ/ or /ɖʐ/.

I also plan to neutralise the distinctions between voiceless and aspirated plosives (but not affricates) word-medially i.e. word medially only the following sounds are permitted :

/p b/
/t d/
/k g/
/ts tsh dz/
/t̠ʃ t̠ʃh d̠ʒ/
/ʈʂh/

Plausible?

If possible, do give real-language examples (past or present) to support your stand.

N.B. the letter h I use to mark aspiration is unsuperscripted, because I am currently on a phone, and superscripting them would be quite tedious.

[Zekoslav]

Are rhotacized long monophthongs better transcribed as [Vːʴ] or [Vʴː]?

I'd choose [Vʴː]. Quality modifiers before quantity modifiers.

[Ser]

Are rhotacized long monophthongs better transcribed as [Vːʴ] or [Vʴː]?

I'd choose [Vʴː]. Quality modifiers before quantity modifiers.

It's always pretty awkward though. This works well for vowels, but for affricate consonants it gets trickier, e.g. Italian razzo 'rocket' and ragazzo 'boy' [ˈradzːo raˈgatsːo] don't look like what they are, even with that joiner on top, [ˈrat͡sːo raˈgad͡zːo]... So [ˈradːzo raˈgatːso] is better, or also the usual convention, [ˈraddzo raˈgattso].

[Zekoslav]

Chechen or some related Caucasian language which has phonemic long consonants in both onsets and codas, specifically has long /t͡sː/ with a long [s ] rather than a long [t]. Croatian /tt͡s/ and /dt͡s/ are realised as [t͡s] with a long [t], while /ts/ and /ds/ are realised as [t͡s] with a long [s ], and in careful speech they both contrast with /t͡s/ [t͡s] with both parts short. So marking length probably has to be adjusted based on what suits the language best. In languages where geminates are heterosyllabic, i.e. [at.ta], I prefer to just write both consonants and put the stress mark between them if needed.

[Ahzoh]

My conlang's stress rules are that stress always falls on the last heaviest syllable. Pretonic short vowels syncopate except in CVCVC syllables.
So, this creates syllables (a and ā here represent short and long vowels, respectively) like CaC.Ca(C) and CCā.Ca(C) and Cā.Cā.Ca(C) but what natural processes (aside from simple vowel coalescence) can I do so that CāC.Ca(C) and CCa.Ca(C) Cā.Ca.Ca(C) are also valid syllable shapes.

Additional information:

• It has a triconsonantal root system, not sure if that makes a huge difference for this question
• Inventory:
  /a e i u /<a e i u>
  /aː eː iː uː/<ā ē ī ū>
  /aj əj aw əw/<ay ey aw ew>
  /aːj əːj aːw əːw/<āy ēy āw ēw>
  
  /m n ŋ/
  /p pʼ b t tʼ d k kʼ g ʔ/
  /s sʼ z ɬ ɬʼ ɮ x xʼ ɣ h/
  /r j w/

[Zekoslav]

I'd love to help, but I can't accurately picture your conlang's syllable structure. Is it underlyingly CVC or CV with a coda consonant allowed word-finally? Where does stress fall if there are no long vowels in the word?

In an underlyingly CVC structure I see no reason why CāC.Ca(C) and Cā.Ca.Ca(C) wouldn't be there from the get go, especially if only pretonic, but not posttonic vowels are syncopate (syncope of posttonic vowels would turn Cā.Ca.Ca(C) into CāC.Ca(C) so it's good it's not there). CCa.Ca(C) is trickier to derive from an underlying CVC structure with the stress rules you've given and assuming in words without long syllables the initial syllable would be stressed. Maybe if it used to be CCā.Ca(C) and then the vowel was shortened for whatever reason.

[Ahzoh]

I'd love to help, but I can't accurately picture your conlang's syllable structure. Is it underlyingly CVC or CV with a coda consonant allowed word-finally? Where does stress fall if there are no long vowels in the word?

The syllable is CV(:)(C) and classed into light (CV), heavy (CVC or CV:) and superheavy (CV:C) but the language does not like clusters of more than two consonants but also doesn't like word-final clusters or CCV(:)C monosyllables. If the word consists entirely of light open syllables, the stress is either always on the last vowel (the last heaviest vowel) or always on the second last vowel. I rather like the idea of having both stress patterns.

If stress is always on the last vowel than I can explain CV́C.CV(C) by way of CV.CV.CV́ > CVC.CV́ with a stress shift, but if stress is penultimate then I can explain CCV́.CV by way of CV.CV́.CV > C.CV́.CV. Neither stress patterns can, however, explain CV́:C.CV.

[Salmoneus]

I'd love to help, but I can't accurately picture your conlang's syllable structure. Is it underlyingly CVC or CV with a coda consonant allowed word-finally? Where does stress fall if there are no long vowels in the word?

The syllable is CV(:)(C) and classed into light (CV), heavy (CVC or CV:) and superheavy (CV:C) but the language does not like clusters of more than two consonants but also doesn't like word-final clusters or CCV(:)C monosyllables. If the word consists entirely of light open syllables, the stress is either always on the last vowel (the last heaviest vowel) or always on the second last vowel. I rather like the idea of having both stress patterns.

If stress is always on the last vowel than I can explain CV́C.CV(C) by way of CV.CV.CV́ > CVC.CV́ with a stress shift, but if stress is penultimate then I can explain CCV́.CV by way of CV.CV́.CV > C.CV́.CV. Neither stress patterns can, however, explain CV́:C.CV.

I'm not sure what you mean. Given your rules, CV:C.CV shouldn't need any explaining? Superheavy syllable plus light syllable, cluster of only two consonants. What's the problem?

[Pabappa]

i think he wants a diachronic solution from a previous CV-only stage. in which case, i dont see one either.

[Zekoslav]

There could be a situation, after syncope, that vowels are lengthened before some coda consonants, but not before others, resulting in CV.CV.CV́ > CVC.CV́ or CVːC.CV́ > CV́C.CV or CV́ːC.CV. The relevant consonants could then merge, making this distinction unpredictable.

If you want to have both penultimate and ultimate stress patterns for light roots while keeping predictable stress, you could make them depend on vowel quality or the presence of a coda consonant (possibly later lost, making this distinction, once again, unpredictable).

CV.CV́.CV > CCV́.CV
CV.CV.CV́C > CV́C.CV / CV́ːC.CV

Make this potentially lost coda consonant a suffix and voilà, ablaut!