Zompist Bboard Again

/m n/ <m n>
/pʰ tʰ qʲʰ qʰ qʷʰ/ <p t ky k kw>
/ɖ ɢʲ ɢ ɢʷ/ <d gy g gw>
/bʱ ɖʱ ɢʲʱ ɢʱ ɢʷʱ/ <bh dh ghy gh ghw>
/t͡ɬ t͡ʃ q͡χʲ q͡χ q͡χʷ/ <th ch khy kh khw>
/d͡ʒ/ <j>
/d͡ʒʱ/ <dj>
/ʂ xʲ x xʷ/ <s xy x xw>
/v ɣʲ ɣ ɣʷ/ <v qy q qw>
/l r j w/ <l r y w>

/i u/ <i u>
/e eː o oː/ <e ee o oo>
/m̩ n̩ l̩ r̩/ <m n l r>, with an apostrophe afterwards if needed to disambiguate
/é é:/ <é ée>

Wow, two sane proposals for an insane (consonantal) phonology, one of which is courtesy of Nort.

This lang has already been cyrillised here, but I do want to see it romanised.

/sexretʰʲ nẽwaː/
m mʲ n nʲ ɲ ŋ ŋʷ
b bʲ d dʲ dʒ ɡ ɡʷ
pʰ pʰʲ tʰ tʰʲ tʃʰ kʰ kʰʷ
f s ɕ ʑ ʂ x xʷ h
w j ɾ l ʎ

Palatalised consonants are always found before front vowels in onset, labialised ones - before back vowels. In essence, the palatalised/nonpalatalised and labialised/nonlabialised distinction is limited to coda. Likewise, nasal consonants in onset are limited to before nasal vowels. /ʑ/ arose from palatalised /ɾ/.

a e i o u +length +nasalisation +high tone
Language also distinguishes nasalised vowels and VN/VNC clusters.

And another 'lang of mine, in close contact with the previous one
/iwəskɪ ɪzak/
m mʲ n nʲ ŋ
pʰ pʰʲ b bʲ tʰ tsʰʲ d dzʲ kʰ ɡ
tɕʰ dʑ
f fʲ s sʲ z zʲ ɕ ʑ ʂ ʐ x h
ɾ l j w ɥ

a e i o u + reduced ɪ ə ʊ

Nobody even knows what the syllable structure is, the clusters are prodigious and prolific both. /ʑ/ arose from palatalised /ɾ/.

Sejreti Nęvà
/sexretʰʲ nẽwaː/
/m mʲ n nʲ ɲ ŋ ŋʷ/ m m n n ń ñ ñ
/b bʲ d dʲ dʒ ɡ ɡʷ/ b b d d ʒ g g
/pʰ pʰʲ tʰ tʰʲ tʃʰ kʰ kʰʷ/ p p t t c k k
/f s ɕ ʑ ʂ x xʷ h/ f s ś z š j j h
/w j ɾ l ʎ/ v y r l

/a e i o u ʲ ʷ/ a e i~ie o u~ue i u
final or pre-consonantal /i u/ are <ie ue> and if you have final /ie ue/ they are <iee uee> and if you have final /iee uee/ then they're <ieee ueee> and so on and so forth
/a aː ã ãː á áː ã́ ã́ː/ a à ą ą̀ á â ą́ ą̂



Iwaske Ezák
/iwəskɪ ɪzak/
/m mʲ n nʲ ŋ/ m mi n ń ñ
/pʰ pʰʲ b bʲ tʰ tsʰʲ d dzʲ kʰ ɡ/ p pi b bi t c d ʒ k g
/tɕʰ dʑ/ ć ǵ
/f fʲ s sʲ z zʲ ɕ ʑ ʂ ʐ x h/ f fi s si z ri š ž j h
/ɾ l j w ɥ/ r l ĭ v w

/a e i o u/ á é i~y ó u
/ɪ ə ʊ/ e a o

/bʲi bʲe bi be/ <bi bie by be> etc.

With a name like /ɪzak/ that's gotta be a posteriori, right?

Darren wrote: ↑Sat Dec 28, 2024 3:06 pm With a name like /ɪzak/ that's gotta be a posteriori, right?

It's got to be a Slavic-lang IMO.

Darren wrote: ↑Sat Dec 28, 2024 3:06 pm With a name like /ɪzak/ that's gotta be a posteriori, right?

I took Russian and put it through some hypothetical questionably polonic future sound changes. But not too many, given how conservative it is.

reˈhɑlː ˈkɑːɸe

I want to see what alternative analyses you guys can come up with for this:

There are the following long vowel phones:

[iː eː ɛː æː ɑː ɔː oː uː]

There are the following short vowel phones:

[i e ɛ æ ɐ ɑ ɔ o u]

There is a stress accent, which is on one of the last three syllables of a word.

Only zero or one long vowel can appear in a word, including clitics. It can also only be in one of the last three syllables, the one on which the stress accent falls.

Long vowels in stressed syllables correspond to the equivalent short vowels in unstressed syllables if not indicated otherwise.

There are the following consonant phones:

[m n ɲ ŋ ɴ]
[b d dz]
[t ts c k q]
[tʼ tsʼ cʼ kʼ qʼ]
[ɸ f s ʃ ç x χ ħ h]
[β ð z ʒ ɣ ʁ]
[ɾ r l ʎ]
[w j]

The syllable structure is CV(ː)(C)(C) and geminates are allowed for all consonants, except that the initial consonant is not required in initial syllables. However, [ɛː ɔː ɛ ɔ] can only occur before one consonant in the same syllable.

The following consonants only occur when intervocalic and non-geminate:

[ɸ β ð ɾ]

The following consonants never occur when intervocalic and non-geminate:

[b d dz f ħ r]

The following consonants only occur before homorganic consonants:

[ŋ ɴ]

The following consonants never occur before [ɔː oː uː ɔ o u] or before [ɑː ɑ] not before [k q kʼ qʼ x χ ɣ ʁ]:

[ɲ dz ts c tsʼ cʼ ʃ ç x ʒ ɣ ʎ]

These consonants also always occur if a consonant in the same series of consonants before the next vowel is in it, except that [dz ts tsʼ] never occur before another consonant, remaining [d t tʼ]. These consonants also never occur in word-final consonant clusters.

The following consonants do not occur in a series of consonants containing the above set excluding consonants that cannot be intervocalic in the first place:

[n ŋ ɴ q qʼ s χ ħ h z ʁ l]

The following vowels never occur before [q qʼ x χ ɣ ʁ]:

[iː æː uː i æ u]

If [q qʼ x χ ɣ ʁ] are attached afterwards these are replaced with:

[eː~e ɑː~ɑ oː~o e ɑ o]

The following consonants:

[n ŋ ɴ d t k q tʼ kʼ qʼ s χ ħ h z ʁ l]

must become:

[ɲ ɲ ɲ dz ts c k tsʼ cʼ kʼ ʃ x ç ç ʒ ɣ ʎ]

when [j] or one of the following vowels are attached afterwards:

[iː eː ɛː æː i e ɛ æ]

This may also occur if [ɐ] is attached afterwards.

There is the exception that if [dz] were to be intervocalic it becomes [z].

The following consonants:

[n ŋ ɴ k q tʼ kʼ qʼ s χ ħ h z ʁ l]

must become:

[ɲ ɲ ɲ c k cʼ kʼ ʃ x ç ç ʒ ɣ ʎ]

if they are in a final consonant cluster but not final when [j] or one of the following vowels are attached afterwards:

[iː eː ɛː æː i e ɛ æ]

This may also occur if [ɐ] is attached afterwards.

The following vowels when stressed:

[i u]

become in derived forms when unstressed:

[e o]

The following vowels when stressed:

[æ ɑ]

become [ɐ] in derived forms when unstressed, unless they alternate with long vowels, except if followed by [q qʼ x χ ɣ ʁ], where then they become [ɑ] in derived forms when unstressed.

The following vowels:

[ɛː ɔː ɛ ɔ]

become:

[æj~ɐj ɑw~ɐw æj~ɐj ɑw~ɐw]

when a vowel is attached afterwards in derived forms.

The following vowels:

[eː oː]

are only found before [k q kʼ qʼ x χ ɣ ʁ].

(Hint ─ when dividing consonants into sets, [k kʼ z] may be in more than one set.)

Examples

[ˈtʼɑːtsekɐ ceɲˈtsɛːn zɑk kɑˈɸitsecʼe dɔː ˈkɑjcæɸe]

[ˈtɔːsokɐ ˈtɑːçenhɐ fɑˈkitːo χɑˈkuɲːɐje ʃæ ˈræːtsɛhɐ ˈsuhkʼɛɾe]

[kʼɑˈʎæsosɐ cenˈtsɛːnhɐ ˈʒiːnhɐn rɑj ˈɑrːɐ]

[ˈtɔːtseʁɐ ˈtɑːçen tsʼɐjɐʎeˈqɑn wɐʎɐˈzizdze ʃæ toˈɸɔːr ˈçɛːce]

[tsehɐˈtæw mɑlː ˈqʼɑrːɐt ˈmɑːtjɐ ʃæ χɑholˈlɑjetse ɣæːwb]
[ˈbɑjejɐ ʃæ ɣæːwb ˈmɑːtɐ ˈlɑwɐ ˈrɑçʎecʼɐ ʃæ mɑlː]
[ɑˈqutd ɑˈqutd ɑˈqutd ɑˈqutd ɑˈqutd ɑˈqutd ɑˈqutd]

[ˈquːrnɐβɐ kɐˈɾistmɐs lɔː]

[nɐjeðɐkʼɑˈwizd nonɐkɐˈcʼæːm ceˈtɑːb ˈʃæːβɐ ʃæ ʎin]

[iː eː ɛː æː ɑː ɔː oː uː] <ii ii aai ee aa~ee oo uu uu>
[i e ɛ æ ɐ ɑ ɔ o u] <i i ai e a~e~ee a~aa~e~ee o u u>
[m n ɲ ŋ ɴ] <m n n n n>
[b d dz] <b d d>
[t ts c k q] <t t k k kh>
[tʼ tsʼ cʼ kʼ qʼ] <t' t' k' k' kh'>
[ɸ f s ʃ ç x χ ħ h] <f f s sh x~h x~xh xh q h>
[β ð z ʒ ɣ ʁ] <b d z zh g gh>
[ɾ r l ʎ] <r r l ly>
[w j] <w y>

[ˈtʼɑːtsekɐ ceɲˈtsɛːn zɑk kɑˈɸitsecʼe dɔː ˈkɑjcæɸe]
[ˈtɔːsokɐ ˈtɑːçenhɐ fɑˈkitːo χɑˈkuɲːɐje ʃæ ˈræːtsɛhɐ ˈsuhkʼɛɾe]
[kʼɑˈʎæsosɐ cenˈtsɛːnhɐ ˈʒiːnhɐn rɑj ˈɑrːɐ]
[ˈtɔːtseʁɐ ˈtɑːçen tsʼɐjɐʎeˈqɑn wɐʎɐˈzizdze ʃæ toˈɸɔːr ˈçɛːce]
[tsehɐˈtæw mɑlː qʼɑrːɐt mɑːtjɐ ʃæ χɑholˈlɑjetse ɣæːwb]
[ˈbɑjejɐ ʃæ ɣæːwb ˈmɑːtɐ ˈlɑwɐ ˈrɑçʎecʼɐ ʃæ mɑlː]
[ɑˈqutd ɑˈqutd ɑˈqutd ɑˈqutd ɑˈqutd ɑˈqutd ɑˈqutd]
[ˈquːrnɐβɐ kɐˈɾistmɐs lɔː]
[nɐjeðɐkʼɑˈwizd nonɐkɐˈcʼæːm ceˈtɑːb ˈʃæːβɐ ʃæ ʎin]

Taatika kintaain zak kaafitik'i doo kaykefi.
Toosuka taahinha faakitto xhakunnyayi se reetaiha suhk'airi.
K'aalesusa kintaainha ziinhan ray arra.
Tootigha taahin t'eyalikhan walezizdi se tufoor haaiki.
Tihatew mall kh'arrat maatya se xhahullayiti geewb. Bayiya se geewb maata lawa rahlik'e se mall. Akhutd.
Khuurnaba Karistmas loo.
Nayidak'aawizd nonakac'am kitaab seeba se lin.

Because Cyrillic probably fits it more than any Latin orthography.

/mʲ mˠ n̪ˠ nʲ ɲ ŋ/ <м м н н ӈ ӈ>
/pʲ pˠ t̪ˠ tʲ c k/ <п п т т к к>
/bʲ bˠ d̪ˠ dʲ ɟ g/ <б б д д ґ ґ>
/fʲ fˠ sˠ ʃ ç x h/ <ф ф с с х х ҳ>
/vʲ w l̪ˠ lʲ j ɣ/ <в в л л г г>
/ɾʲ ɾˠ/ <р р>

/a e i o u ə/ <а э ы о у ӕ> after broad, <я е и ё ю і> after slender. Double for length.

/əi əu iə uə/ <аи-яи ау-яу ые-ие уо-юо>

<ъ ь> indicate unexpected broad- or slenderness.

Sample sentence from Wikipedia (Aran dialect, but with /N L/ merged as in the Dingle Peninsula):
/vʲiː ʃeː əɟ ˈafˠəɾˠk əˈmˠax asˠ ə ˈwɪnʲoːɡ nuəɾʲ ə vʲiː ˈmʲɪʃə ɡɔl haɾˠt̪ˠ/
<Вии сее ӕґь афӕрк ӕмах ас ӕ вынёёґ нуорь ӕ вии мисі ґол ҳарт.>

[iː eː ɛː æː ɑː ɔː oː uː] <í ía aí/eí é á aú ó ú>

[i e ɛ æ ɐ ɑ ɔ o u] <ì i/ìe ai/ei e a o å u/ùo ù>

[m n ɲ ŋ ɴ] <m n n(j) n n>
[b d dz] <b d d(j)>
[t ts c k q] <t t(j) k(j) k/q q>
[tʼ tsʼ cʼ kʼ qʼ] <th th(j) kh(j) kh/qh qh>
[ɸ f s ʃ ç x χ ħ h] <p f s s(j) h(j) x x ħ h>
[β ð z ʒ ɣ ʁ] <b d z z(j) g g>
[ɾ r l ʎ] <r rr l l(j)>
[w j] <v/u j/i>

Thátika kinteín zak kopìtikhi daú koikefi.
Taúsuka táhinha foqìttu xakùnnjaji se réteiha sùhkhairi.
Kholesusa kinteínha zínhan roi òrra.
Taútiga táhin thjajaliqòn valjazizdji se tupaúr heíki.

Tihatèu mall qhòrrat mátïa se xohullájiti gaíwb.
Bòjija se gaíwb máta lòva rahlikhja se moll.
Oqùtd. Oqùtd. Oqùtd. Oqùtd. Oqùtd. Oqùtd. Oqùtd.

I've got an idea for a phonemic inventory but i'm not entirely certain how I want to romanize it. here it is:

Consonants:

	Labial	Alveolar	Post Alveolar	Labio-Palatal	Palatal
Stop	p	t		pʲ	c
Fricative	ɸ	s	ʃ		ç
Affricate		ʦ	ʧ
Nasal	m	n		mʲ	ɲ
Approximant		ɹ			j
Lateral Approximant		l			ʎ
Tap/Flap		ɾ

Vowels:

	Front	Center	Back
High	i	ɨ	u
Middle	e		o
Low	a

while this would normally be fairly easy to romanize, i'm seeking something that favors aesthetics over readability. (things like using <c> for /s/ and <x> for /ʃ/, with possibly a complex set of rules determining what letter is used for what phoneme based on its phonemic environment.)

any ideas?

edit: fixed a spelling error

Where are the velars?

WeepingElf wrote: ↑Mon Mar 10, 2025 10:05 am Where are the velars?

Seconded -- having at least one velar consonant is practically a universal.

WeepingElf wrote: ↑Mon Mar 10, 2025 10:05 am Where are the velars?

this conlang has none, actually! might be kinda odd, but the aesthetic I want this conlang to have suits better if there are none. although I could probably add /w/.

Travis B. wrote: ↑Mon Mar 10, 2025 10:40 am

WeepingElf wrote: ↑Mon Mar 10, 2025 10:05 am Where are the velars?

Seconded -- having at least one velar consonant is practically a universal.

There's a surprising amount of languages in New Guinea and South America without velars (or at least without the main ones, they often have /w/ or something boring like that)

Ok, I added a few velarized phonemes, along with a few others. here's the new phonemic inventory:

Consonants:

	Labial	Alveolar	Post Alveolar	Labio-Palatal	Palatal	Labio-Velar
Stop	p	t		pʲ	c	pʷ
Fricative	ɸ	s	ʃ		ç
Affricate		t͡s	t͡ʃ		c͡ç
Nasal	m	n		mʲ	ɲ	mʷ
Tap/Flap		ɾ
Approximant		ɹ		ɥ	j	w
Lateral Approximant		l			ʎ

Vowels:

	Front	Center	Back
High	i	ɨ	u
Middle	e		o
Low	a

now, back to what was my original question. How should I go about making a romanization method for this conlang that highly favors aesthetics over pronounceability?

edit: added a phoneme I missed earlier

Darren wrote: ↑Mon Mar 10, 2025 3:35 pm

Travis B. wrote: ↑Mon Mar 10, 2025 10:40 am

WeepingElf wrote: ↑Mon Mar 10, 2025 10:05 am Where are the velars?

Seconded -- having at least one velar consonant is practically a universal.

There's a surprising amount of languages in New Guinea and South America without velars (or at least without the main ones, they often have /w/ or something boring like that)

But none with as many consonants as this language! The closest is probably Wutung, with 15 consonants and no velars. (In fact even this is debatable, since it has a single word with [k]; but it seems that this may be a loanword.)

The revised version makes more sense with basically a contrast of front vs. rounded velar, like Nuxalk (29 consonants and no plain velars), only the rounded velars have prominent labial articulation. And anyway if he wants a conlang without velars that's absolutely fine.


In any case:
/p pʲ t c pʷ/ <p pi t c qu>
/ts tʃ cç/ <ç cz çh>
/ɸ s ʃ ç/ <f s sz hi>
/m mʲ n ɲ mʷ/ <m mi n ñ mu>
/ɹ ɾ l j ɥ ʎ w/ <ř r l y yu ł u>

/i ɨ u e o a/ <i î u e o a>

Darren wrote: ↑Tue Mar 11, 2025 4:32 am Nuxalk (29 consonants and no plain velars)

‽

Well, I mean, it makes perfect sense in historical terms, but I’m slightly shocked to discover there’s a language which actually does that. (Come to think of it, there’s probably at least one Australian language which does that too.)

Anyway, here’s my attempt:

/p pʲ pʷ t c/ ⟨p pj pv t k⟩
/ɸ s ʃ ç/ ⟨f s x h⟩
/t͡s t͡ʃ c͡ç/ ⟨z c g⟩
/m mʲ mʷ n ɲ/ ⟨m mj mv n ñ⟩
/ɾ ɹ l ʎ/ ⟨r rr l ll⟩
/j ɥ w/ ⟨y ÿ w⟩

/a e i ɨ o u/ ⟨a e i ï o u⟩

bradrn wrote: ↑Tue Mar 11, 2025 4:40 am

Darren wrote: ↑Tue Mar 11, 2025 4:32 am Nuxalk (29 consonants and no plain velars)

‽

Well, I mean, it makes perfect sense in historical terms, but I’m slightly shocked to discover there’s a language which actually does that. (Come to think of it, there’s probably at least one Australian language which does that too.)

It's areally fairly common to have /c kʷ/ or even /tʃ kʷ/ without /k/; cf. Squamish, Klallam, Chemakum (do not look at the 19th-century orthography for this one), Kwakʼwala, Snchitsu’umshtsn, Montana Salish - the latter of which has 37 consonants with no plain velars, but with all three of /k̟ʷ q qʷ/.