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

[VaptuantaDoi]

Syllable boundaries are not a priori detectable. There is no difference between [ai] and [a.i]. However, phonemically they can be distinct, so that /ai a.i/ might be realised as [aĭ̝ aˑiˑ] or similar. So /ɹiakt/ is [ɻʷɪˈjækt̚ˀ] with a drop in sonority between /i/ and /a/ marking a syllable boundary.

In some languages, it makes sense to analyse diphthongs as vowel sequences because they are realised close to the cardinal values of their components (e.g. /ai/ → [ăĭ] rather than [ɑˑɪ̯]) and they pattern as such. In other languages like English, diphthongs are best thought of as distinct nuclei in their own right; for instance in Australian English the diphthong /ɔy̯/ has a distribution much closer to say /ʊː/ than say /ia/. /ɔy̯/ is also not made up of two monophthongs – neither /ɔ/ nor /y/ occur on their own.

[Ahzoh]

Can certain grammatical voices take different alignments with respect to more-than-monotransitivity?

Normal ditransitives in Vrkhazhian take a secundative alignment where the recipient is marked in the accusative while the theme is marked in the instrumental (much akin to "I supplied X with Y")

But it feels very awkward to mark the causee in the accusative and the actual affected object in the instrument. Especially given that causees are rather similar in the semantic space to an instrument, which is what is typically marked by the instrumental.

So I am thinking that it would make more sense if the causative construction took on a double object construction (AKA "neutral" alignment) where both causee and actual object are both in the accusative. Perhaps because the causative was originally an instrumental applicative and perhaps other applicatives can take neutral alignment as well.

So in sum:
Basic ditransitives take secundative alignment while applicative (and by extension causative) ditransitives take neutral alignment

[Creyeditor]

That is definitely possible. Some languages even have special morphology to change between dative alignment construction and secundative-like alignment morphology. German be- comes close, though it's not very productive.

Ich schenke dem Kind einen Ball.
1SG.NOM give.as.gift the.M.DAT child a.M.ACC
'I give the child a ball as a gift.'

Sie be-schenkte die Kinder mit Nüssen.
3SG.F.NOM ???-give.as.gift.PST the.PL.ACC with nuts.DAT
'She gave the children nuts as a gift.'

[Ahzoh]

That is definitely possible. Some languages even have special morphology to change between dative alignment construction and secundative-like alignment morphology. German be- comes close, though it's not very productive.

Ich schenke dem Kind einen Ball.
1SG.NOM give.as.gift the.M.DAT child a.M.ACC
'I give the child a ball as a gift.'

Sie be-schenkte die Kinder mit Nüssen.
3SG.F.NOM ???-give.as.gift.PST the.PL.ACC with nuts.DAT
'She gave the children nuts as a gift.'

Yes, the Dative Shift, as in English, though I thought it only turned dative alignment to neutral alignment.

[Creyeditor]

Which would be even closer to your idea, right?

[Ahzoh]

Which would be even closer to your idea, right?

Yes. Just with different default alignments. Though I came to learn that there are quite a lot of constructions regarding ditransitives and it's quite possible to have secundative alignment expressed in a double object construction.

I don't know. I don't consider "I give John a boat" a secundative alignment because there is no case marking. Rather I'd call it neutral.

I swear syntax is like entering a different universe.

[thethief3]

Syllable boundaries are not a priori detectable. There is no difference between [ai] and [a.i]. However, phonemically they can be distinct, so that /ai a.i/ might be realised as [aĭ̝ aˑiˑ] or similar. So /ɹiakt/ is [ɻʷɪˈjækt̚ˀ] with a drop in sonority between /i/ and /a/ marking a syllable boundary.

In some languages, it makes sense to analyse diphthongs as vowel sequences because they are realised close to the cardinal values of their components (e.g. /ai/ → [ăĭ] rather than [ɑˑɪ̯]) and they pattern as such. In other languages like English, diphthongs are best thought of as distinct nuclei in their own right; for instance in Australian English the diphthong /ɔy̯/ has a distribution much closer to say /ʊː/ than say /ia/. /ɔy̯/ is also not made up of two monophthongs – neither /ɔ/ nor /y/ occur on their own.

Many Polynesian languages contrast a lack of syllable boundaries with ones bound by a glottal stop. How do they fit into this?

[VaptuantaDoi]

Syllable boundaries are not a priori detectable. There is no difference between [ai] and [a.i]. However, phonemically they can be distinct, so that /ai a.i/ might be realised as [aĭ̝ aˑiˑ] or similar. So /ɹiakt/ is [ɻʷɪˈjækt̚ˀ] with a drop in sonority between /i/ and /a/ marking a syllable boundary.

In some languages, it makes sense to analyse diphthongs as vowel sequences because they are realised close to the cardinal values of their components (e.g. /ai/ → [ăĭ] rather than [ɑˑɪ̯]) and they pattern as such. In other languages like English, diphthongs are best thought of as distinct nuclei in their own right; for instance in Australian English the diphthong /ɔy̯/ has a distribution much closer to say /ʊː/ than say /ia/. /ɔy̯/ is also not made up of two monophthongs – neither /ɔ/ nor /y/ occur on their own.

Many Polynesian languages contrast a lack of syllable boundaries with ones bound by a glottal stop. How do they fit into this?

They have a contrast between /aʔe/ and /ae/ (etc.) although I believe in general there's no particular restriction on two-vowel sequences so there's no reason to class them as diphthongs. In such a case /ʔ/ patterns as a normal consonant.

[TBPO]

Byntšá is based entirely on phonology from Romanization Game, and I have to create morphology and syntax, basing only on one phonological sample. I analysed it and it has 3 one-syllable words, 3 two-s. w., 3 three-s. w., 2 four-s. w., no five-s. w. and 1 six-s. word, so Number of Syllables in Words (NSW) of 1, 2, 3, 4, 5 and 6 will occur in ratio 3:3:3:2:1:1 in lexicon.

1. How to create lexicon that preserves that ratio?
2. Should this language be analytic? Synthetic? Or maybe 1-syllable-per-morpheme?
3. Should this language be isolating, agglutinative or fusional?

[Keenir]

Byntšá is based entirely on phonology from Romanization Game, and I have to create morphology and syntax, basing only on one phonological sample. I analysed it and it has 3 one-syllable words, 3 two-s. w., 3 three-s. w., 2 four-s. w., no five-s. w. and 1 six-s. word, so Number of Syllables in Words (NSW) of 1, 2, 3, 4, 5 and 6 will occur in ratio 3:3:3:2:1:1 in lexicon.

1. How to create lexicon that preserves that ratio?

one option would be individual word by individual word.

though not every part of your conlang needs to obey that ratio.

3. Should this language be isolating, agglutinative or fusional?[/quote]

a six-syllable isolating language would be...interesting.

[LinguoFranco]

Does pre-fortis clipping occur in any natlangs outside of English?

[Creyeditor]

I don't know and I would really like to know.

[LinguoFranco]

I don't know and I would really like to know.

I think it's what English does where vowels are lengthened before voiced consonants.

[Creyeditor]

According to Wikipedia, French does something similar with lengthening before some consonants.

[VaptuantaDoi]

Pretty much the same thing happens in Friulian; basically CVP CVB → CVP CVːP

[Khemehekis]

And when the (English) words are borrowed into Japanese, some vowels before voiced consonants become long vowels, while vowels before unvoiced consonants remain short, and the consonant gets doubled!

Steven -> Sutiibun
tube -> chuubu

Big Mac -> Biggu Makku or Bikku Makku
hip-hop -> hippu-hoppu

[thethief3]

I don't know and I would really like to know.

I think it's what English does where vowels are lengthened before voiced consonants.

I might be misremembering but it might be universal.

[VaptuantaDoi]

I can't detect it in my own dialect of English except with diphthongs.

[LinguoFranco]

I don't know and I would really like to know.

I think it's what English does where vowels are lengthened before voiced consonants.

I might be misremembering but it might be universal.

Does the rule tend to apply to all vowels before a voiced consonant, or only the ones in stressed syllables?