Conlang Random Thread

[akam chinjir]

In human languages at least, affricates often pattern as plosives, and some think that they simply are plosives; like ts is a coronal strident plosive (contrasting with nonstrident t). I think there are supposed to be languages in which they pattern with fricatives on their right hand side only, but none where they simply pattern as fricatives. The Blackwell Companion to Phonology has an article on affricates that you might find useful, fwiw.

[Richard W]

Voicing won't contrast in my setting's Elvish language's plosives but will in its fricatives. Is this attested in any natlang? Does having the language's affricates pattern like its plosives or fricatives sound more natural?

Greek has been pretty close to this, though to varying degrees Classical Greek nasal + voiceless unaspirated stops has yielded modern voiced stops.

Accord to Wikipedia, Bouyei fits this pattern. However, if one includes Chinese loans (and quite possibly some expressive forms), there is a contrasting set of aspirated stops.

I suspect this pattern, namely phonation contrast on fricatives but not plosives, is not very stable even though it does exist.

Affricates generally pattern like plosives rather than like fricatives, though manners of articulation can be more like fricatives.

[Pabappa]

Yupik is doing it too, and on top of that it even has a voicing contrast in its nasals. But still not in its stops. I dont know the details of allophony, however.

[Nortaneous]

I think there are supposed to be languages in which they pattern with fricatives on their right hand side only

For an incredibly cheap example, English affricates pattern with strident fricatives on their right-hand side:

grænt + PL = grænts
dæns + PL = dænsɨz
bræntʃ + PL = bræntʃɨz
flændʒ + PL = flændʒɨz

(But this can be captured with [+strident]: græf + PL > græfs, lɒx + PL > lɒxs.)

And the prints-prince merger:
prins > prints
hæmstər > hæmpstər
grænt > grænt (*græntt)
bræntʃ > bræntʃ (*brænttʃ)

OTOH, I don't think affricates take preglottalization, so they pattern with fricatives there.
pæt > pæˀt
pæk > pæˀk
pæs > pæs
pætʃ ?> pætʃ

For a better example, you'd want a rule where the value of [±strident] determines the realization of something on its left.

[akam chinjir]

grænt + PL = grænts
dæns + PL = dænsɨz
bræntʃ + PL = bræntʃɨz
flændʒ + PL = flændʒɨz

But bæθs, not bæthɨz, dwɔrfs or dwɔrvz but not drwɔrvɨz---your examples here show strident affricates patterning with strident fricatives, and not with non-strident ones---which is to say, as strident, not as fricative.

OTOH, I don't think affricates take preglottalization, so they pattern with fricatives there.
pæt > pæˀt
pæk > pæˀk
pæs > pæs
pætʃ ?> pætʃ

Interesting, if you're right about this.

[Kuchigakatai]

But bæθs, not bæthɨz, dwɔrfs or dwɔrvz but not drwɔrvɨz---your examples here show strident affricates patterning with strident fricatives, and not with non-strident ones---which is to say, as strident, not as fricative.

It is still amusing to be able to bring up cheap examples from English where affricates don't pattern with plosives but with (a subtype of!) fricatives though, precisely on the right side, as you said first.

It reminds me of that one time I realized that the -n of the English article "a" is basically French-style liaison, and could be used as a cheap example to explain what liaison is. Or when I realized that Classical Arabic uses liaison, vowel(!) liaison, to prevent three-consonant clusters. Not the normal way to talk about this at all...

ʕan fasˤli rrabiiʕi 'about the season of spring'
laa tantaðˤir lailan wa-nahaaran 'don't wait day and night'
tuħibbu-hum sirran 'she loves them in secret'
min ʔidriisa 'from Idris (Enoch)'

ʕan-i lfarraani 'about the baker'
laa tantaðˤir-i llailata 'don't wait tonight'
tuħibbu-hum-u ssaariqatu 'the (female) thief loves them'
min-a lʔuulaa 'from this world (as opposed to the hereafter)'

(The default vowel is -i. The exceptions are the verbal inflection -tum and the suffix pronouns -kum and -hum, which take -u, and the preposition min, which takes -a.)

[bradrn]

Voicing won't contrast in my setting's Elvish language's plosives but will in its fricatives. Is this attested in any natlang? Does having the language's affricates pattern like its plosives or fricatives sound more natural?

This famously happens in Standard Mandarin: it has an aspiration contrast in its plosives but a voicing contrast in its fricatives. WALS lists 38 languages with a ‘voicing contrast in fricatives alone’, including Eskimo-Aleut, Na-Dene, and a few other languages elsewhere.

[Pabappa]

Wikipedia says WALS is wrong: https://en.wikipedia.org/wiki/Standard_ ... Consonants . No voiced fricatives. there is a fricated vowel, but thats not really the same as a voicing contrast in fricatives.

[TurkeySloth]

Thanks guys.

My setting's Primordial language has tenuous, aspirated, nasal, and affricated clicks at three POA (cover K) [K Kʰ K̃ K͡χ]. There's at least one language that has [i → e] after [q] (can't remember which). For future reference, is such an alignment possible with clicks? "Future reference" because I'm not changing Primordial, which was used for example purposes.

[bradrn]

Wikipedia says WALS is wrong: https://en.wikipedia.org/wiki/Standard_ ... Consonants . No voiced fricatives. there is a fricated vowel, but thats not really the same as a voicing contrast in fricatives.

Interesting — I had remembered Standard Mandarin as having a voicing contrast. WALS appears to be incorrect for Wu, West Greenlandic and Sgaw Karen as well. But it is correct for Nivkh and Slavey. It’s also vaguely correct for Tsou and Vietnamese as well, as long as you don’t count a pulmonic/implosive contrast as one in voicing.

[bradrn]

Thanks guys.

My setting's Primordial language has tenuous, aspirated, nasal, and affricated clicks at three POA (cover K) [K Kʰ K̃ K͡χ]. There's at least one language that has [i → e] after [q] (can't remember which). For future reference, is such an alignment possible with clicks? "Future reference" because I'm not changing Primordial, which was used for example purposes.

What do you mean by ‘alignment’ here?

[Nortaneous]

Voicing won't contrast in my setting's Elvish language's plosives but will in its fricatives. Is this attested in any natlang? Does having the language's affricates pattern like its plosives or fricatives sound more natural?

It's hard to find examples of languages with a voicing contrast in the fricatives and only one plosive series, but Awa Pit looks like a good example. There are also plenty of Austronesian languages with a voicing contrast only in /f v/, although it's debatable whether that should count.

Interesting — I had remembered Standard Mandarin as having a voicing contrast. WALS appears to be incorrect for Wu, West Greenlandic and Sgaw Karen as well. But it is correct for Nivkh and Slavey. It’s also vaguely correct for Tsou and Vietnamese as well, as long as you don’t count a pulmonic/implosive contrast as one in voicing.

Some Nivkh dialects have a voiced plosive series. Some don't.

[TurkeySloth]

Thanks guys.

My setting's Primordial language has tenuous, aspirated, nasal, and affricated clicks at three POA (cover K) [K Kʰ K̃ K͡χ]. There's at least one language that has [i → e] after [q] (can't remember which). For future reference, is such an alignment possible with clicks? "Future reference" because I'm not changing Primordial, which was used for example purposes.

What do you mean by ‘alignment’ here?

Another bad wording due to autism. I'm trying to ask if it's possible for a language to have [i → e] after [K] or not.

[Whimemsz]

Voicing won't contrast in my setting's Elvish language's plosives but will in its fricatives. Is this attested in any natlang? Does having the language's affricates pattern like its plosives or fricatives sound more natural?

It's hard to find examples of languages with a voicing contrast in the fricatives and only one plosive series, but Awa Pit looks like a good example. There are also plenty of Austronesian languages with a voicing contrast only in /f v/, although it's debatable whether that should count.

There are Eskimo-Aleut examples that are valid too, e.g. Central Alaskan Yup'ik (though that wasn't one of the languages tracked by WALS for this feature), which has a single voiceless plosive series and a large number of fricatives contrasting in voicing. (As Pabappa alluded to above for Yup'ik varieties in general, though I don't know for certain about varieties other than Central Alaskan Yup'ik because Wikipedia can't be trusted.)

[Kuchigakatai]

WALS appears to be incorrect for Wu, West Greenlandic and Sgaw Karen as well.

Shanghainese, which is probably what is meant by "Wu" here (it being the most studied Wu variety*), is often described as technically having a slack-voiced ("murmured") series of stops (even if in Chinese linguistics as a whole these are often referred to as "voiced" and notated with /b d dʑ g/). That could be the reason why it is classified that way, because then its /v z ʑ ɦ/ fricative series would be the only consonants that count as truly voiced. If only WALS actually explained why every choice has been made and what are some alternative views, it'd be so wonderful...

* Wu is actually a very diverse group of mutually unintelligible Chinese varieties; Shanghainese and Suzhounese are nice examples but are not necessarily representative of the whole group. Why, then, is Wu often talked about as a single language, you ask? It is because, as Nort once observed with regard to the Bai family "language", in China, language families are "languages" and languages are "dialects"!

[bradrn]

WALS appears to be incorrect for Wu, West Greenlandic and Sgaw Karen as well.

Shanghainese, which is probably what is meant by "Wu" here (it being the most studied Wu variety*), is often described as technically having a slack-voiced ("murmured") series of stops (even if in Chinese linguistics as a whole these are often referred to as "voiced" and notated with /b d dʑ g/). That could be the reason why it is classified that way, because then its /v z ʑ ɦ/ fricative series would be the only consonants that count as truly voiced. If only WALS actually explained why every choice has been made and what are some alternative views, it'd be so wonderful...

For me, a murmured/voiceless distinction is close enough to voicing that I listed WALS as incorrect.

* Wu is actually a very diverse group of mutually unintelligible Chinese varieties; Shanghainese and Suzhounese are nice examples but are not necessarily representative of the whole group. Why, then, is Wu often talked about as a single language, you ask? It is because, as Nort once observed with regard to the Bai family "language", in China, language families are "languages" and languages are "dialects"!

Sorry! As you have noticed, I did indeed use Shanghainese phonology as a reference, but listed it as ‘Wu’. Mainly this was because WALS just listed Wu without mentioning the language, but also because I was under the impression that Shanghainese was fairly representative of Wu languages.

(BTW, I’m fully aware that Chinese ‘languages’ are families. That’s why I always talk about Standard Mandarin rather than Mandarin (or even just ‘Chinese’, as it so often gets called), since Mandarin is a whole family of mutually unintelligible languages.)

[Knit Tie]

Could a language running a four-way coronal distinction, i.e. /t̪ t ʈ tɕ/ also maintain an i - ɨ distinction for dentals and coronals? Perhaps with /ti/ being /tsi/?

[bradrn]

Could a language running a four-way coronal distinction, i.e. /t̪ t ʈ tɕ/ also maintain an i - ɨ distinction for dentals and coronals? Perhaps with /ti/ being /tsi/?

Is there any reason to think that it couldn’t?

[jal]

Voicing won't contrast in my setting's Elvish language's plosives but will in its fricatives. Is this attested in any natlang? Does having the language's affricates pattern like its plosives or fricatives sound more natural?

I don't know whether any natlang does that, but in my native Dutch, it's the opposite: plosives have voicing contrast, but fricatives have not*. Also, the languages that I do know that have no voicing contrast in the plosives have other contrasts there, like aspirated/non-aspirated.

*Standard Dutch has phonetic contrast between voiced and unvoiced fricatives, but many dialects/accents do not, and even then, minimal pairs are very hard to find.


JAL

[Nortaneous]

Could a language running a four-way coronal distinction, i.e. /t̪ t ʈ tɕ/ also maintain an i - ɨ distinction for dentals and coronals? Perhaps with /ti/ being /tsi/?

Mianchi Qiang: /t̪ ts tʂ tɕ/ + /i ɨ/
Yele: /t̪ t̠/ + /i ɨ/
Bantawa: /t̪ t̺ tɕ/ + /i ɯ/
Toda: /t̪ t ts ʈ tʃ/ + /i y ɯ u/
Lhagang Choyu: /t ts tʃ ʈ tɕ/ + /i ʉ ɯ u ɯˠ/ (also /n n̠ ȵ/)