Let's create a good language!

[Schiaparelliorbust]

By inter-syllable cluster I mean a cluster involving a coda and an onset (eg "a*nt*i"), and I didn't include a rhotic in my proposal because there's a lot of them and I didn't want to choose. I don't think a constrastive postalveolar stop would be a good idea, but /ʃ/ should work. The only problems with that are that it might mess up the sandhi system (although that might already have problems with the lack of phonemic /ɲ/) and that we don't have a Latin letter for it (<q> is free, but that would probably look really weird).

How does our sandhi system work? What is the contrastive postalveolar stop? None of the sounds I mentioned are plosives.

[fluffykitty]

How does our sandhi system work?

I dunno. Maybe changing manner of articulation?

What is the contrastive postalveolar stop?

A postalveolar stop that is distinguished from an alveolar stop.

None of the sounds I mentioned are plosives.

Yeah, that's true. If we do have /ʃ/ we could use <z> for it since we don't have /z/.

[Schiaparelliorbust]

I dunno. Maybe changing manner of articulation?

I think sandhi should just be for pronunciation. Something like French liaison perhaps? I also think there should be sandhi for two consonants.

A postalveolar stop that is distinguished from an alveolar stop.

I just realized then, we don't have /t/ or /d/?

Yeah, that's true. If we do have /ʃ/ we could use <z> for it since we don't have /z/.

Ok, that's fine by me.

[fluffykitty]

I think sandhi should just be for pronunciation. Something like French liaison perhaps?

I think we should use it in some grammatical affixes, like verb conjugation and stuff.

I also think there should be sandhi for two consonants.

Stuff like /mk/->/ŋk/ or something else?

I just realized then, we don't have /t/ or /d/?

/t/ exists, but it doesn't contrast with a postalveolar stop.

[Schiaparelliorbust]

I think we should use it in some grammatical affixes, like verb conjugation and stuff.

Sounds good. Should there be sandhi at a distance, i.e. sounds not right next to each other will still influence each other?

Stuff like /mk/->/ŋk/ or something else?

Yes, that's exactly what I had in mind.

/t/ exists, but it doesn't contrast with a postalveolar stop.

Should it be an allophone then?

[fluffykitty]

Sounds good. Should there be sandhi at a distance, i.e. sounds not right next to each other will still influence each other?

Maybe.

Yes, that's exactly what I had in mind.

Okay.

Should it be an allophone then?

[t̠] will probably be an allophone of /t/ next to /ʃ/.

Should we finalize the phonology/phonotactics/orthography as my previous post, but with /ʃ/ <z> with affricate form /tʃ/? Also, I just noticed that I accidentally allowed coda nasal+fricative rather than nasal+affricate there. Also, what do you think of

Other possible ideas are allowing (potentially non-matching) fricative+stop clusters in the coda as well, and permitting /l/ as a syllable nucleus.

?

[Schiaparelliorbust]

[t̠] will probably be an allophone of /t/ next to /ʃ/.

AFAIK, the [t̠] is the dental t, and on the Wikipedia pulmonic consonant chart, /t/ sounds almost indistinguishable from /d/. [t̠] still sounds like normal t, as in take. What do you mean by "next to"?

Should we finalize the phonology/phonotactics/orthography as my previous post, but with /ʃ/ <z> with affricate form /tʃ/? Also, I just noticed that I accidentally allowed coda nasal+fricative rather than nasal+affricate there. Also, what do you think of

Other possible ideas are allowing (potentially non-matching) fricative+stop clusters in the coda as well, and permitting /l/ as a syllable nucleus.

?

So <z> is /ʃ/ and tz is /tʃ/? I'm kind of neutral on allowing fricative + stop clusters. I assume by non-matching you mean their place of articulation doesn't need to be the same. /l/ as a syllable nucleus sounds good. Good thing you you realized that allowed coda nasal + fricative instead of nasal + affricate. The former sound weird.

[fluffykitty]

AFAIK, the [t̠] is the dental t, and on the Wikipedia pulmonic consonant chart, /t/ sounds almost indistinguishable from /d/. [t̠] still sounds like normal t, as in take.

I copied it from the symbol for the voiceless postalveolar plosive on https://en.wikipedia.org/wiki/Voiceless ... r_plosives. I'm pretty sure the chart /t/ and /d/ being similar is just a quirk of how the voicing distinction works in English.

What do you mean by "next to"?

Immediately following or preceding.

So <z> is /ʃ/ and tz is /tʃ/?

Yes.

I'm kind of neutral on allowing fricative + stop clusters. I assume by non-matching you mean their place of articulation doesn't need to be the same.

Yes.

/l/ as a syllable nucleus sounds good.

Okay, so that's a legal nucleus, and syllabic /l/ cannot be immediately followed or preceded by another /l/. Should we move on to making words and stuff now?

Good thing you you realized that allowed coda nasal + fricative instead of nasal + affricate.
The former sound weird.

Yep.

[Schiaparelliorbust]

I copied it from the symbol for the voiceless postalveolar plosive on https://en.wikipedia.org/wiki/Voiceless ... r_plosives. I'm pretty sure the chart /t/ and /d/ being similar is just a quirk of how the voicing distinction works in English.

How does it work in English?

Immediately following or preceding.

Could you give an example please?

I'm kind of neutral on allowing fricative + stop clusters. I assume by non-matching you mean their place of articulation doesn't need to be the same.

Yes.

Do you want them or not?

Okay, so that's a legal nucleus, and syllabic /l/ cannot be immediately followed or preceded by another /l/. Should we move on to making words and stuff now?

I think we should do a complete recap first.

[fluffykitty]

I copied it from the symbol for the voiceless postalveolar plosive on https://en.wikipedia.org/wiki/Voiceless ... r_plosives. I'm pretty sure the chart /t/ and /d/ being similar is just a quirk of how the voicing distinction works in English.

How does it work in English?

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

Immediately following or preceding.

Could you give an example please?

<aztetze> has both /t/s moved to postalveolar position.

I'm kind of neutral on allowing fricative + stop clusters. I assume by non-matching you mean their place of articulation doesn't need to be the same.

Yes.

Do you want them or not?

Yes.

Should we move on to making words and stuff now?

I think we should do a complete recap first.

Sounds are /aeiouy/ /ptck/ /fsʃçx/ /mnŋ/ /lwj/. Legal diphthongs are /ai/ /ei/ /oi/ /au/ /eu/.
/ŋ/ is written as <ng> when not followed by /k/ and <n> otherwise. /ç/ is written with <h>, /ʃ/ is written with <z>, and everything else is written with its IPA character.
Syllable onsets can be:

• Any single consonant
• Any affricate (/pf/ /ts/ /tʃ/ /cç/ /kx/)
• Stop or fricative followed by /l/
• Fricative or nasal followed by /w/

Syllable codas can be:

• Any single consonant other than /w/ or /j/
• Any affricate (/pf/ /ts/ /tʃ/ /cç/ /kx/) (I also forgot this one in the last summary post)
• Nasal followed by matching stop or fricative (this can produce [ɲ] but only non-contrastively)
• Fricative followed by stop (not necessarily matching)

Restrictions on inter-syllable clusters:

• Affricates cannot be followed by stops
• Nasals followed by stops must match the stop ([ɲ] appears again)

Syllable nuclei can be any vowel or diphthong, or /l/ if the nucleus is not adjacent to another /l/ or another syllable's nucleus.

If we're moving on to syntax and stuff, here's some things I think we should do:

• Make the word for "good" the same as the word for "edible" (and maybe have more food metaphors while we're at it)
• SOV word order
• Split case marking across determiner and noun (like in German)
• Use auxiliary verbs
• Allow omission of subject pronouns by marking person and number on verbs
• Sandhi with grammatical affixes
• Affixes for deriving agent nouns and similar from verbs (eg English "-er", "-ee", but for more arguments like "receiver", "tool", "source", etc)

[Schiaparelliorbust]

Sounds are /aeiouy/ /ptck/ /fsʃçx/ /mnŋ/ /lwj/. Legal diphthongs are /ai/ /ei/ /oi/ /au/ /eu/.
/ŋ/ is written as <ng> when not followed by /k/ and <n> otherwise. /ç/ is written with <h>, /ʃ/ is written with <z>, and everything else is written with its IPA character.
Syllable onsets can be:

• Any single consonant
• Any affricate (/pf/ /ts/ /tʃ/ /cç/ /kx/)
• Stop or fricative followed by /l/
• Fricative or nasal followed by /w/

Syllable codas can be:

• Any single consonant other than /w/ or /j/
• Any affricate (/pf/ /ts/ /tʃ/ /cç/ /kx/) (I also forgot this one in the last summary post)
• Nasal followed by matching stop or fricative (this can produce [ɲ] but only non-contrastively)
• Fricative followed by stop (not necessarily matching)

Restrictions on inter-syllable clusters:

• Affricates cannot be followed by stops
• Nasals followed by stops must match the stop ([ɲ] appears again)

Syllable nuclei can be any vowel or diphthong, or /l/ if the nucleus is not adjacent to another /l/ or another syllable's nucleus.

If we're moving on to syntax and stuff, here's some things I think we should do:

• Make the word for "good" the same as the word for "edible" (and maybe have more food metaphors while we're at it)
• SOV word order
• Split case marking across determiner and noun (like in German)
• Use auxiliary verbs
• Allow omission of subject pronouns by marking person and number on verbs
• Sandhi with grammatical affixes
• Affixes for deriving agent nouns and similar from verbs (eg English "-er", "-ee", but for more arguments like "receiver", "tool", "source", etc)

All sounds good to me, but I need clarification on some things.
[*]Make the word for "good" the same as the word for "edible" (and maybe have more food metaphors while we're at it)
Why this?
[*]Split case marking across determiner and noun (like in German)
Will /l/ as a nucleus b like, say, ablto?

[Hunting]

I like "I CA investigate" word order, because stack-based languages.

[fluffykitty]

All sounds good to me, but I need clarification on some things.
[*]Make the word for "good" the same as the word for "edible" (and maybe have more food metaphors while we're at it)
Why this?

Why not?

[*]Split case marking across determiner and noun (like in German)

German doesn't (usually) completely mark case on either the article or the noun, but the combination of both (usually) completely determines the case.

Will /l/ as a nucleus b like, say, ablto?

Yes.

I like "I CA investigate" word order, because stack-based languages.

That's SOV.

We really should have an off-thread place to track grammar and words and stuff. If we do that, I think we should use userpages on Lifewiki, since everyone here either has or can get an account.

[Schiaparelliorbust]

German doesn't (usually) completely mark case on either the article or the noun, but the combination of both (usually) completely determines the case.

I'm pretty sure I understand you, but I'd still like an example. I take mandatory German classes at school, but they started teaching me at level A2 last year, when all I'd known from my previous two years was almost completely vocabulary. I don't understand anything.

We really should have an off-thread place to track grammar and words and stuff. If we do that, I think we should use userpages on Lifewiki, since everyone here either has or can get an account.

I agree. Then Hunting can also see our progress.

[fluffykitty]

German doesn't (usually) completely mark case on either the article or the noun, but the combination of both (usually) completely determines the case.

I'm pretty sure I understand you, but I'd still like an example. I take mandatory German classes at school, but they started teaching me at level A2 last year, when all I'd known from my previous two years was almost completely vocabulary. I don't understand anything.

I just checked the Wikipedia page about German noun declension, and apparently that doesn't actually happen. Any case information on the noun is also coded on the article. Maybe we could do that, but we should also have more significant inflection on the noun than in German.

We really should have an off-thread place to track grammar and words and stuff. If we do that, I think we should use userpages on Lifewiki, since everyone here either has or can get an account.

I agree. Then Hunting can also see our progress.

Okay. Whose userspace will we use?

[Schiaparelliorbust]

I just checked the Wikipedia page about German noun declension, and apparently that doesn't actually happen. Any case information on the noun is also coded on the article. Maybe we could do that, but we should also have more significant inflection on the noun than in German.

The should we have suffixes? Or maybe changing a short bit at the end of a word? Like there could be a "plain" form and every other case form is based on that.

Okay. Whose userspace will we use?

Maybe everyone can explain their opinions on their own pages. Alternatively, people could tell their opinions on this thread and that could be written down by a single user. I could do it, but if anyone else wants to, go ahead.

[fluffykitty]

I just checked the Wikipedia page about German noun declension, and apparently that doesn't actually happen. Any case information on the noun is also coded on the article. Maybe we could do that, but we should also have more significant inflection on the noun than in German.

The should we have suffixes? Or maybe changing a short bit at the end of a word? Like there could be a "plain" form and every other case form is based on that.

Probably a mix of the two because of sandhi. Should we have a small case system like German or Latin or a big case system like Finnish or Hungarian? I'm leaning towards a small case system.

Okay. Whose userspace will we use?

Maybe everyone can explain their opinions on their own pages. Alternatively, people could tell their opinions on this thread and that could be written down by a single user. I could do it, but if anyone else wants to, go ahead.
[/quote]
I'm fine with you doing it.

[Schiaparelliorbust]

Probably a mix of the two because of sandhi. Should we have a small case system like German or Latin or a big case system like Finnish or Hungarian? I'm leaning towards a small case system.

By big or small I guess you mean many or few cases. I think small would be better. 4-6 cases should be okay.

I'm fine with you doing it.

Ok then, list your opinions and I'll write them down. I'll also keep track of progress.

[fluffykitty]

By big or small I guess you mean many or few cases. I think small would be better. 4-6 cases should be okay.

Okay. According to https://en.wikipedia.org/wiki/Case_hierarchy, a natural six-case system would be nominative, accusative, genitive, dative, locative, and instrumental. Thoughts?

I'm fine with you doing it.

Ok then, list your opinions and I'll write them down. I'll also keep track of progress.

• Make the word for "good" the same as the word for "edible" (and maybe have more food metaphors while we're at it)
• SOV word order
• Split case marking across determiner and noun (like in German)
• Use auxiliary verbs
• Allow omission of subject pronouns by marking person and number on verbs
• Sandhi with grammatical affixes
• Affixes for deriving agent nouns and similar from verbs (eg English "-er", "-ee", but for more arguments like "receiver", "tool", "source", etc)
• Six-case system as mentioned above (NAGD, locative, instrumental)

[SquishyBoi]

hi
i'm too lazy to read the entire thread so would it be a bother if you could make a post showing the current phonology for this language/conlang/whatever?

ok thanks

[Schiaparelliorbust]

hi
i'm too lazy to read the entire thread so would it be a bother if you could make a post showing the current phonology for this language/conlang/whatever?

ok thanks

Look a few posts above.

[fluffykitty]

Ok then, list your opinions and I'll write them down. I'll also keep track of progress.

Link?

We should start making actual words and stuff. Who wants to start? Also, here's a table we can use for sandhi variations: (Parenthesized entries only occur as mutations, so fricative /w/ is /f/ not /x/)

Code: Select all

nasal       m   n   (n) (n) ng
approximant w   l   (l) j   (w)
fricative   f   s   z   h   x
stop        p   t   (t) c   k

[Schiaparelliorbust]

Ok then, list your opinions and I'll write them down. I'll also keep track of progress.

Link?

We should start making actual words and stuff. Who wants to start? Also, here's a table we can use for sandhi variations: (Parenthesized entries only occur as mutations, so fricative /w/ is /f/ not /x/)

Code: Select all

nasal       m   n   (n) (n) ng
approximant w   l   (l) j   (w)
fricative   f   s   z   h   x
stop        p   t   (t) c   k

I need to get a trusted flag? Correct me if I'm wrong. I just asked for it. Could you also explain what exactly the tables says please?

[fluffykitty]

I need to get a trusted flag? Correct me if I'm wrong. I just asked for it.

Yep.

Could you also explain what exactly the tables says please?

It lists phonemes and their analogs in other manners of articulation (ish), since we'll probably have stuff like "change to fricative" or something for sandhi.

[Schiaparelliorbust]

Yep.
It lists phonemes and their analogs in other manners of articulation (ish), since we'll probably have stuff like "change to fricative" or something for sandhi.

What words should we focus on now?