Twin Aster

[Darren]

At minimum I’d expect something like /əj jə əw wə/ → [e e o o]. Come to think of it, I’m not sure there’s any two-vowel system where this doesn’t happen.

I guess there's Nuxálk with allegedly just /a/ that doesn't display notable allophony (and syllabic /j w/). Maybe Arrernte too with /ɪ~ʊ a/ where /ɪ~ʊ/ variation is fairly unconditioned? And I'm tempted to say some stage of PIE (if at one point it had just */a aː/ or */æ ɒ/ or something)? I think with the syllabic semivowels it's reasonable enough.

I usually take posited "syllabic semivowels" to mean that the people doing the analysis are being too clever.

This is usually true, although I can see a justification for it in languages like PIE where ablaut plays a major role.

[Travis B.]

I guess there's Nuxálk with allegedly just /a/ that doesn't display notable allophony (and syllabic /j w/). Maybe Arrernte too with /ɪ~ʊ a/ where /ɪ~ʊ/ variation is fairly unconditioned? And I'm tempted to say some stage of PIE (if at one point it had just */a aː/ or */æ ɒ/ or something)? I think with the syllabic semivowels it's reasonable enough.

I usually take posited "syllabic semivowels" to mean that the people doing the analysis are being too clever.

This is usually true, although I can see a justification for it in languages like PIE where ablaut plays a major role.

Thing is, a vowel system of, say, */i æ ɒ u/ or */i a aː u/ is much more sane than one of just */æ ɒ/ or */a aː/ (after all, real extremely small vowel inventories are typically vertical, e.g. /a ɨ/). I feel that sometimes the people doing analyses forget that what they are analyzing were real languages spoken by real people at one point, and not just figments of linguists' imaginations.

[Darren]

Thing is, a vowel system of, say, */i æ ɒ u/ or */i a aː u/ is much more sane than one of just */æ ɒ/ or */a aː/ (after all, real extremely small vowel inventories are typically vertical, e.g. /a ɨ/). I feel that sometimes the people doing analyses forget that what they are analyzing were real languages spoken by real people at one point, and not just figments of linguists' imaginations.

Yeah I've got no argument with that. But it's also true that PIE */i u/ and */j w/ alternate in the same way that */m̩ n̩/ and */m n/ do, so there isn't really a good reason to say that */i u/ are vowels and */m̩ n̩/ aren't. It's equally as sane as claiming that say Ubykh has only two vowels /ə a/ when all of [i y ɨ u e ø ə o ɛ œ a ɔ] or whatever appear with high frequency. Sure, if you were writing a layman's grammar of PIE, you'd say it has the vowels */i u e o a/, but we do gain some understanding of the language from recognising that */i u/ (and */a/ for that matter) are not like the other vowels.

(sorry for the derailing of your thread Mr Space)

[Man in Space]

Verbal Mood in Tm

-Ø- INDICATIVE
-w- POTENTIAL CONDITIONAL
-y- INTERROGATIVE
-ə- OPTATIVE
-a- VERIDICAL CONDITIONAL

So for these morphemes, the potential conditional and interrogative are always word-final (even after the locative). The optative and veridical conditional affixes have some behavioral peculiarities:

- The vowel will precede a final -w -y following a nonvowel: /haspkj/ [ŋãspti] ngaspti 'cut(s)' ~ /haspk<ə>j/ [ŋãspkəj] ngaspkəy 'may cut' ~ /haspk<a>j/ ngaspnay 'would certainly cut'; /kjØwsw/ [tjuɹu] tyuru 'slather, wet, moisten, dampen, cover in fluid; paint, write' ~ /kjØws<ə>w/ [tjuɹəw] tyurəw 'may paint' ~ /kjØws<a>w/ [tjuɹãw] tyuraw 'would certainly write'.
- If the vowel in the final is the opposite of the morphemic vowel, it ablauts: /jaksa/ [jãksã] yaksa ~ /jaks<ə>/ [jãksə] yaksə; /skasap/ [ʃnãɹãm] snaram ~ /skas<ə>p/ [ʃnãɹəm] snasəm; /ajkə/ [ãjkə] aykə ~ /ajk<a>/ [ãjnã] ayna; /ssapəss/ [^tsãpət] tsapət ~ /ssap<a>ss/ [^tsãmãt] tsamat.
- If the vowel in the final is identical to the morphemic vowel, it instead becomes a preposed epenthetic /-h-/ ([-ŋ-] before /a/): /jaksa/ [jãksã] yaksa ~ /jaks<a>/ [jãksŋã] yaksnga; [ʃnãɹãm] snaram ~ /skas<a>p/ [ʃnãsŋãm] snasngam; /ajkə/ [ãjkə] aykə ~ /ajk<ə>/ [ãjkhə] aykhə; /ssapəss/ [^tsãpət] tsapət ~ /ssap<ə>ss/ [^tsãphət] tsaphət.
- In cases where an illegal cluster would form, it is generally realized as null.

/pjə kahak-Ø/ [pjə nãŋãn] 'he guts'
/pjə kahak-w/ [pjə nãŋãku] 'he might gut'
/pjə kahak-j/ [pjə nãŋãti] 'does he gut?'
/pjə kahak-ə/ [pjə nãŋãkə] 'may he gut'
/pjə kahak-a/ [pjə nãŋãnã] 'he would certainly gut'

/pjə haspkj-Ø/ [pjə ŋãspti] 'he cuts'
/pjə haspkj-w/ [pjə ŋãsptju] 'he might cut'
/pjə haspkj-j/ [pjə ŋãsptji] 'does he cut?'
/pjə haspk<ə>j/ [pjə ŋãspkəj] 'may he cut'
/pjə haspk<a>j/ [pjə ŋãspnãj] 'he would certainly cut'

/pjə haspkØi/
[pjə ŋãspti]
pyə ngaspti

/pjə haspk<a>j/
[pjə ŋãspnãj]
pyə ngaspnay

pyə

3SG

haspkØi

cut

<a>

VC

'(that) he would/will (certainly) cut'

Atyə ksəskyə?
/akjə ksəskjə/
[ãtjə ksəʃtjə]

atyə

house

k-

INC

s-

PASS

əskə

build

<j>

INT

'has the house begun to be built?' (the builders are relevant somehow)

Atyə ksastyə?
/akjə ksaskjə/
[ãtjə ksãʃtjə]

atyə

house

k-

INC

a-

PSV

əskə

build

<j>

INT

'has the house started being built?' (focus is on the house)

Nouns

[bradrn]

Out of curiosity, precisely how does /haspkjw/ turn into [ŋãsptju]?

[Man in Space]

Out of curiosity, precisely how does /haspkjw/ turn into [ŋãsptju]?

- /a/ [ã] obligatorily
- /h/ [ŋ] / _a (rhinoglottophilia)
- /k/ [t] / _j
- /w/ [u] / in nucleic position (/j/ [j] in this particular syllable shape)

[Raholeun]

While scrolling through this thread earlier, there were a couple of evocatively named items in your glossary:

Soldiers’ sickness. Perceived illness or culture-bound disorder whose cause is the breakdown of the bicameral mind in one so afflicted and the emergence of an analog consciousness.

Can you elaborate? What do you mean with "analog consciousness" in this context?

Stavewort. Black plant genus found below the Messerini line with powerful contraceptive properties.

I might want to coin that for the language I am working on. Wasn't rue used in ancient times to induce miscarriage?

Weird, to. To read or forecast someone’s destiny; to tell someone’s fortune. In Caber practice, often accompanied by the administration of hallucinogens.

So in this case, do I as fortune-seeker administer these to myself, so that in the trip I can foresee my own future? Or rather, does a third party, like a shaman-scientist of sorts, take it and then tell my future?

3. I'm trying to write a basic reference grammar of CT for Fiat Lingua. What sorts of things should a good grammatical sketch include/would you like to see in such a sketch?

Presuming that this project is still ongoing, you could aim to have the grammar reflect the level and detail of worldbuilding. I am sure your choice of example phrases and notes can add flavoring. If you have something up for review, I'd like to see it.

A CT sentence:
[image]

I saw the slang?

[Man in Space]

While scrolling through this thread earlier, there were a couple of evocatively named items in your glossary:

That is a compliment, that you scroll through it.

Soldiers’ sickness. Perceived illness or culture-bound disorder whose cause is the breakdown of the bicameral mind in one so afflicted and the emergence of an analog consciousness.

Can you elaborate? What do you mean with "analog consciousness" in this context?

Jayesian bicameralism is a theme in Twin Aster. The adasar themselves were initially wholly predisposed to a bicameral sort of cognition. It's been quite some time since I actually read Jaynes' book, but the upshot is that according to him, self-awareness as we understand it is a recent phenomenon; prior to developing self-awareness (the "analog I" in Jaynes' terminology, and hence mine), humans had a dual sort of awareness (the "bicameral I") in which each hemisphere of the brain was unaware that it was part of the same thing as the other. This supposedly led to misinterpreting trains of thought as external sources—e.g. a god, a messenger thereof, a deceased family member, whatever—commanding them to do things as opposed to them thinking of it themselves. Per Jaynes, the Bronze Age Collapse catalyzed the development of an analog consciousness, though whether that is likely is an exercise for the reader.

I am simplifying (and half-remembering) considerably here, leaving much from the books out. Suffice to say, one of the themes of Twin Aster is the sort of clash of the bicameral/analog mindsets.

Soldiers' sickness is what happens when someone undergoes a transition into (or has fully developed) an analog consciousness. One of the most "effective" treatments is essentially psychological and physical brain damage with the goal of inducing schizophrenia-like symptoms. I mention the possibility of it being culture-bound because it largely is due to its inherent postulates.

Stavewort. Black plant genus found below the Messerini line with powerful contraceptive properties.

I might want to coin that for the language I am working on. Wasn't rue used in ancient times to induce miscarriage?

That, I do not know. Stavewort isn't an abortifact, though; it is used to interrupt the menstrual cycle.

Whether one of stavewort or matriarchy itself begat the other or if they were independent developments that happened to have success in convergance is a bitterly-disputed topic.

Weird, to. To read or forecast someone’s destiny; to tell someone’s fortune. In Caber practice, often accompanied by the administration of hallucinogens.

So in this case, do I as fortune-seeker administer these to myself, so that in the trip I can foresee my own future? Or rather, does a third party, like a shaman-scientist of sorts, take it and then tell my future?

Most mainstream practitioners would tell you to go the third-party route. Self-practicing, while (as you would expect) does happen occasionally, it's almost always by a) fringe groups and heretics, b) genuinely curious but clueless people, c) madmen, or d) edgy teenagers trying to be "dark".

3. I'm trying to write a basic reference grammar of CT for Fiat Lingua. What sorts of things should a good grammatical sketch include/would you like to see in such a sketch?

Presuming that this project is still ongoing, you could aim to have the grammar reflect the level and detail of worldbuilding. I am sure your choice of example phrases and notes can add flavoring. If you have something up for review, I'd like to see it.

Stand by on that. (Alternatively, DM me your e-mail if you're interested in the whole thing.)

A CT sentence:[image]

I saw the slang?

"I saw the stars".

[Man in Space]

A thought occurs. Now we get to play "How much of this will be recorded before I pass out and how much will be comprehensible".

A tricon, but with vocalic roots and consonantal transfixes.

This has undoubtedly been done before, but I thought of some ways I might handle it.

This language is going to be crazy vowel-heavy. However, I saw some descriptions of an Afroasiatic language that had roundedness of vowels as a suprasegmental affix. So, dimensions I can think of to play with:

- Place and height. Seems a little too unnaturalistic for these to be separated.
- Roundedness, as above.
- Stress placement.
- Length, though this one, I'm not so fond of (albeit for aesthetic reasons), but I guess we'll see.
- Phonation.
- Tense/lax. Hmm…
- Tone. This one's probably going to have to do some heavy lifting.

The Unusual Suspects: The Vowels

Ignoring other factors for the time being, these are the basic vowel phonemes.

/æ e i ɑ ʌ ɯ ə¹ ɜ² æ̃ ĩ ɑ̃ ɯ̃/ ä e i a v eu ê er ę į ą ęu
+ rounding = [œ ø y ɔ o u o ɜ œ̃ ỹ ɔ̃ ũ] äu oe u oa o ou ô or ǫe ų ǫ ǫu

/æⁱ eⁱ ɑⁱ ʌⁱ ɯⁱ æ^u æ̃^u ɑ^u ɑ̃^u ʌ^u/ äi ei ai vi eui äou ęou aou ąou vou
+ rounding = [œⁱ øⁱ ɔⁱ oⁱ uⁱ œ^u œ̃^u ɔ^u ɔ̃^u] äui oei oai oi oui oeu ǫeu oau ǫau au

1. Surfaces as [e] unless rounded, when it's [o]
2. Always has effects of rounding on adjacent consonant(s) and instigates consonant harmony: /ɬ~l/ [θ~ɹ]

There will be few consonants (where a colon appears, the form to the right surfaces when rounding is applied; a hyphen, the effects of ê/ô):

/t~d~dɹ:dw k:p~g:b~n:m ʔ:ŋ/ t~d~dr:dw k:p~b:g~n:m ':ng
/s x:h̪͆ h~ɦ/ s ḫ:f h
/ʜ~ʡ/ ḥ
/ɬ-θ~l-ɹ/ l
/j:w/ y:w

For convenience, we can refer to the phonemes in the abstract:

/T K ʔ/
/S X H/
/R/
/L/
/Y/

Rounding itself can be denoted º.

So, you could have, say, a root √eu-ęu-ê, a transfix Ø-K-H, and rounding º. This could yield, say, 'ounǫuhô.

Not sure what happens when the null phoneme /Ø/ causes hiatus.

[Travis B.]

You could make /ʔ:ŋ/ be the same phoneme as the null phoneme, making glottal stop or nasal realizations be the surface forms when in hiatus.

[TomHChappell]

A tricon, but with vocalic roots and consonantal transfixes.

This has undoubtedly been done before, but I thought of some ways I might handle it.

.

I have tried this before, and couldn’t make it work. Too few roots, too many transfixes.
If you can make it work, I really want to read about it!
So I’m going to lurk and watch. I’ll probably post something long once you succeed.
Good luck!

[Man in Space]

You could make /ʔ:ŋ/ be the same phoneme as the null phoneme, making glottal stop or nasal realizations be the surface forms when in hiatus.

Oh, good idea!

I have tried this before, and couldn’t make it work. Too few roots, too many transfixes.
If you can make it work, I really want to read about it!
So I’m going to lurk and watch. I’ll probably post something long once you succeed.
Good luck!

Thank you!

My idea is to try to hack the phonology such that there are a stupid number of vowels but still make it naturalistic. There's a considerable amount of allophony among the consonants, to which I will get later, but right now we're talking about the vowels. There're 22-ish¹ at base, all of which have a rounded counterpart (rounding is a suprasegmental operation in this language), plus the null vowel. The paradigm is as below.

Unrounded

ä

ę

a

ą

e

v

i

į

eu

ęu

ê2

er3

äi

ęi

ai

ąi

ei

vi

eui

aou

ąou

veu

–

Vª

æ

æ̃

ɑ

ɑ̃

e

ʌ

i

ĩ

ɯ

ɯ̃

əª

ɜª

æi

æ̃i

ɑi

ɑ̃i

ei

ʌi

ɯi

ɑu

ɑ̃u

ʌɯ

Ø

↓

↓

↓

↓

↓

↓

↓

↓

↓

↓

↓

↓

↓

↓

↓

↓

↓

↓

↓

↓

↓

↓

↓

↓

Vº

œ

œ̃

ɔ

ɔ̃

ø

o

y

ỹ

u

ũ

əº

ɜº

œy

œ̃y

ɔy

ɔ̃y

øy

oy

uy

ɔu

ɔ̃u

ou

Ø

Rounded

äu

ǫe

oa

ǫ

oe

o

u

ų

ou

ǫu

ô2

or3

eü

ęü

oaü

ǫaü

oeu

oü

ouü

oau

ǫau

oöu

–

1. "-ish" because there's some crossover in how the vowels surface.
2. /əª əº/ [ɛ~e ɔ~o].
3. /ɜª ɜº/ are both generally realized as either [ɜ] or [ə], with no actual rounding on the vowel (unless a labial precedes). The adjacent consonant(s) nonetheless surface as their rounded forms when applicable.

Now we get onto the consonants. I group these into four MOAs:

Stop	T	K	7
Continuant	S	X	H
Resonant	L	Y	2
Null		Ø

These have a stupendous range of realizations:

• /T/ is realized:
  • As [t] when word-initial, following another voiceless phone, and finally following a non-nasal phone
  • As [d] between a voiced oral consonant and an oral vowel
  • As [dɹ] between a voiced phone and an unrounded nasal vowel
  • As [dw] between a voiced phone and a rounded nasal vowel
• /K/ is realized differently depending on whether the following sound is rounded or not.
  • If /K/ does not precede a rounded vowel, it surfaces:
    • As [k] before a voiceless sound when initial or following a voiceless phone
    • As [g] between a voiced phone and an oral vowel
    • As [n] when standing adjacent to a nasal vowel
  • If /K/ does precede a rounded vowel, it surfaces:
    • As [p] before a voiceless sound when initial or following a voiceless phone
    • As [b] between a voiced phone and an oral vowel
    • As [m] when standing adjacent to a nasal vowel
• /7/ is realized:
  • As [ʔ] initially before an oral vowel or in hiatus between two oral vowels
  • As [ŋ] initially before a nasal vowel, in hiatus between two vowels of which at least one is nasalized, and finally
  • As [ː] (i.e., consonant length) when following a consonant
  • As [Ø] when preceding a consonant
• /S/ is realized:
  • As [s] when initial, when final, and when adjacent to a voiceless phone
  • As [z] when adjacent to a voiced consonant and when intervocalic
• /X/, like its occluded counterpart /K/ above, surfaces differently depending on the roundedness of its environs.
  • If /X/ does not precede a rounded phone, it surfaces:
    • As [x] when initial or when standing next to a voiceless sound
    • As [j] when intervocalic or when final
  • If /X/ does precede a rounded phone, it surfaces as [h̪͆].
• /H/ is realized:
  • As [h] initially or when following a voiceless sound
  • As [ː] (i.e., vowel length) when following a vowel
  • As voicelessness when following a resonant
• /L/ is realized:
  • As [ɬ] adjacent to a voiceless sound (or as [θ] if /ɜ/ is in the word)
  • As [l] when initial, final, or standing next to a voiced consonant (or as [ɹ] if /ɜ/ is in the word)
• /Y/ is realized:
  • As [j] when next to an unrounded phone (except when final)
  • As [w] when next to a rounded phone or when final
• /2/ is realized:
  • As [ʜ] when initial or when next to a voiceless phone
  • As [ʢ] when next to a voiced phone or when final
• /Ø/ is the absence of a vowel.

Now that that's out of the way, let me commit this so I don't lose my work.

[bradrn]

Of course, I expect a full diachronic justification for this system.

[Man in Space]

I should probably show how I romanize the consonants, yeah?

• /T/ – [t] t, [d] d, [dɹ] dr, [dw] dw
• /K/ – [k] k, [g] g, [n] n, [p] p, [b] b, [m] m
• /7/ – [ʔ] ’, [ŋ] ng, [Cː] CC, [Ø] –
• /S/ – [s] s, [z] z
• /X/ – [x] ḫ, [j] y, [h̪͆] f
• /H/ – [h] h, [Vː] V·, [Rh] Rh
• /L/ – [ɬ] ł, [l] l, [θ] đ, [ɹ] r
• /Y/ – [j] y, [w] w
• /2/ – [ʜ] ḥ, [ʡ] ṛ

Phonological structure

Roots

The basic root is a trivocalic skeleton of the form √X-Y-Z. The X position almost always takes a consonant [EDIT: Vowel, vowel, a thousand times vowel, whether the error be fair or foul], with a handful of exceptions; Y and Z may be null more or less without restriction, so long as a) at least one of them bears a consonant [EDIT: Vowel; I’m used to dealing with tricons] and b) Y must be present in those exceptional cases in which X is null.. Somewhat rarely, roots of the form √W+X-Y-Z or √X-Y-Z+Q occur; only one of the extra slots W or Q can be filled at any time (i.e., roots of the form ^X√W+X-Y-Z+Q cannot occur).

Personal pronouns

1SG känęi /KæKæ̃ⁱª/ [kænæ̃ⁱ]
1PL.INCL nąi /Kɑ̃ⁱª/ [nɑ̃ⁱ]
1PL.EXCL mǫau /Kã^uº/ [mɔ̃^u]
2 ’owoṛ /7ʌYʌ2º/ [ʔowoʢ]
3 poe· /KeHº/ [pøː]

On roots and patterns

The main constructs here are:

• The root, typically¹ denoted √A-B-C;
• The pattern, which is usually given in the format ^X-Y-Z-;
• Roundedness, denoted² as Xº if present and Xª if absent; and
• Any relevant affixes.

1. Typically. There's a few quadriliteral exceptions (roots that look biliteral generally have Ø as a consonant [EDIT: Vowel; I plead the Fifth]).
2. The º typically is notated on the relevant affix that triggers the harmony. The ª is typically implicit to the root, but does show up in a few affixes.

Let's do some samples, shall we?

× ^H-K-Y-
× ^S-7H-T-
× ^T-H-K7-
× ^Y-2-SL-
× ^2-H2-7-
× ^7-L-2-X

Quadriliteral Roots

Quadriliteral roots do sometimes occur. These typically feature an extra vowel at either the beginning or end of the root (not both), and the extra vowel is signified by using a + instead of a - to join the bonus vowel to the rest of the traditional root structure.

√i+į-v-ê, √e+i-ęi-ä

√e-eu-a+į, √eui-er-a+a

Verbnouns and Nounverbs

Most roots can be split into two categories: Those coming from verbnouns and those coming from nounverbs.

Verbnouns

Kind of like how the maṣdar is considered the "starting point" of Arabic roots (at least sometimes), there are a few common "verbnouns" that serve as a starting point for deriving trivocalic roots. Some of the more common patterns are:

• ^7-27-L-K (√äi-er-eu > ’äiḥḥerreuk, √ąi-äi-ęi > ngąiḥḥäilęik, √äi-ęi-vi > ’äiḥḥęilvik, √į-aou-veu > ’įḥḥaoulveuk)
• ^X-KL-T- (√į-ei-ą > ḫįnleidrą, √v-ai-eu > ḫvkłaideu, √vi-er-aou > ḫvipđerdaou, √ęu-Ø-a > ḫęunla)
• ^Y-Y-S- (√ęu-er-ê > yęuwerze, √ęu-vi-ä > yęuyvizä, √eu+ęu-ai-ą > ’euyęuyaizą, √äi-ęi-ąi > yäiyęiząi)
• ^H-H-L- (√aou-äi-ą > haouhäilą, √aou-äi-eu+e > haouhäileu’e, √ę-ai-ei > hęhailei, √į-ai-ąi > hįhailąi)
• ^H-T-YL-S (√i-ê-ą > hidêyląs, √į+ęu-Ø-ą > ngįhęudiląs, √v-ä-aou > hvdäylaous, √ęu-er-ä > hęuderyräs)
• ^T-X-7 (√į-aou-ê > drįyaou’ê, √ęu-e-Ø > dręuyeng, √v-äi-ą > tvyäingą, √veu-Ø-ą > tveuḫḫą)

It's useful to split these up into classes based on the shape of the obligate transfix:

• Class Ia: ^X-X-X- / Class Ib: ^X-X-X-X
• Class IIa: ^X-XX-X- / Class IIb: ^X-XX-X-X
• Class IIIb: ^X-X-XX- / Class IIIb: ^X-X-XX-X

Nounverbs

We also get some verbs that come from basic nominal templates, some common ones of which are provided below. These tend to be fewer in number and less intuitive.

• ^K-Y7-H-K (√į-ai-ä > nįyyaihäk, √i-äi-aou > kiyyäihaoup, √ęu-veu-ê+eu > nęuyyveuhêgeu, √v-Ø-ä > kvyy^ahäk)
• ^T-K-H7- (√į-ê-ą > drįnê·’ą, √į-veu-ä > drįnveu·’ä, √į+ęu-ai-ei > ngįdręunai·’ei, √v-ą-ą > tvną·’ą)
• ^X-7-X-L (√v-Ø-ê > ḫvyêl, √į+į-äi-ą > ’įyį’äilą, √a-aou-ą > ḫa’aouyąl, √aou-a-aou > ḫaou’ayaoul)
• ^2-T-X- (√v-a-ą > ḥvdayą, √v-veu-ê > ḥvdveuyê, √ęu-er-ä > ḥęuderyä, √ä-ai-eu > ḥädaiyeu)
• ^7-T-T- (√ęu-er-aou > ngęuderdaou, √eu-a-ą > ’eudadrą, √eu-ai-ä > ’eudaidä, √e-er-ä > ’ederdä)

As with the verbnouns, it's useful to split these up by class:

• Class A1: ^X-X-X- / Class A2: ^X-X-X-X
• Class B1: ^X-XX-X- / Class B2: ^X-XX-X-X
• Class C1: ^X-X-XX- / Class C2: ^X-X-XX-X

…I was going somewhere with this. Now I'm having trouble figuring it out. Hmm. Maybe after I eat dinner I'll be more creative.

---

gen dump:

More: show

Categories

More: show

W=JJJJJJJJį5ąa1iä
X=1į234evaä56ę789AąBEêDi
Y=EA3427Dąäa8ê5Ø1eØv9iįØB6ę
Z=ąäê69D45a7Ø1e2įAEv3ęiB8
Q=JJJJJJJJJJJeą5iäįv1aę

Rewrite rules

More: show

J×||
×J||
1|ęu
2|veu
3|äi
4|aou
5|eu
6|ąi
7|ei
8|ąou
9|vi
A|er
B|eui
E|ai
D|ęi
Ø||

Syllable types

More: show

W×X-Y-Z
X-Y-Z×Q

Settings

More: show

Dropoff: Medium
Monosyllables: Always

TODO: Nouns (atelic º/telic ª)

[Man in Space]

Y’know…the more I think about it…why not have /u ũ/ < *a^u *ã^u, such that it creates a round vowel that patterns with the unrounded ones? EDIT: Though it’d have to do something like the inner-outer flip if I’m to preserve /ɔ^u ɔ̃^u/ as such, say, something like *ə^u vel sim. as an intermediary to keep them phonologically distinct enough to undergo that shift while having the desired results…

[Man in Space]

Of course, I expect a full diachronic justification for this system.

I was going to carefully avoid that, but the challenge has been broached. I do have some ideas, at least…

[Man in Space]

EDIT: It cut off the quote from bradrn's post that started it all, so here it is—

I’m very surprised that /ə/ has no allophony here. At minimum I’d expect something like /əj jə əw wə/ → [e e o o]. Come to think of it, I’m not sure there’s any two-vowel system where this doesn’t happen.

I guess there's Nuxálk with allegedly just /a/ that doesn't display notable allophony (and syllabic /j w/). Maybe Arrernte too with /ɪ~ʊ a/ where /ɪ~ʊ/ variation is fairly unconditioned? And I'm tempted to say some stage of PIE (if at one point it had just */a aː/ or */æ ɒ/ or something)? I think with the syllabic semivowels it's reasonable enough.

I usually take posited "syllabic semivowels" to mean that the people doing the analysis are being too clever.

This is usually true, although I can see a justification for it in languages like PIE where ablaut plays a major role.

…this chain of conversation just turned an annoyance of Tm into a non-issue: How to romanize the schwa.

Originally I just had /ə/ <ə>, using <Ǝ> in the majuscule, which is ugly as sin. Unfortunately there's not much I can do with my choice of font, but that is largely due to a deliberate stylistic constraint: I'm writing in the style of ye olde linguistics monographs of yesteryear, so I'm using one of those old-time typewriter-daisy-wheel fonts and the character set is appropriately minimal. It is, at least, the case that I can hack a few things together in LaTeX to give, say, a schwa where it's really needed—in that case, just \T{e}.

Now, originally I had not much by way of allophony in the vowels, and, as bradrn pointed out, that's likely unrealistic, if not uninteresting. However, /a/'s hat is that it is always nasalized…so allophony would make perfect sense.

Something like:

əw əy > o(w) e(y)
wə yə > wo ye

And then just write them as such; plain /ə/ which remains gets the customary â of which I so enjoy the use (though I don't know whether I want variation or conditioned allophony where [ɨ] appears if an /a/ is near…?).

/kwəskyə/ [kwoštye] kwostye, /pəwka/ [ponã] pona, /kəykyə/ [ketye] ketye, /həwa/ [howã] howa, /sswəy/ [tswe] tswe, /hayəw/ [ngãyo] ngayo, /sakyəy/ [sãtye] satye, /ksyə/ [kšye] ksye, /pakə/ [mãkə ~ mãkɨ] makâ