Great Skjajræ [ˈɕcɐ.ɾʲæ] Empire Conlang Scratchpad (currently, Skjajræfæ [ɕcɐˈɾʲæ.fʲæ])

[TurkeySloth]

Thanks, Akangka. I'll think about the final forms.

Before I post Aéhoi Creole's redone inventory, I have a question about phone classification. As commonly-used words can behave unexpectedly (the as [ðə], not [θə]), are <c> unconditionally shifting [p͡ʔ → ʔ] and <h> shifting [ɴ̥͡ɱ̊ → ɴ̥] before <u, ú> in its particles phonemic or allophonic shifts?

"the" is not unexpected. It just that /ð/ and /θ/ is represented with the same graph <th>. Also it's an allophonic shifts, until something is done to the rest so that the appearance of /ʔ/ and /ɴ/ is no longer predictable. (For example, if some vowel becomes /u/, but the shift stops to operate)

Okay. Upon loss of stress, the formerly-stressed vowels shifted as follows (short marked under cover symbols cause tone system), noting the orthographic changes occurred later: [A → O, Á → Ó, E → Y, É → Ý, I → U, Í → Ú].

Are there any natural languages with vowel-inclusive roots? I am considering allowing Aéhoi Creole's roots to contain long high-tone vowels, such as [ə˦ː]. If I do so, Aéhoi's root would be <éh>.

Aéhoi Creole uses particles, similar to East Asian languages. The locative particle's always been oa [œ.ɶ]. Is having the movement particles (lative, ablative, place-to-place coupla) as tonal derivatives of the locative (oxa [œ˩˥.ɶ]; oax [œ.ɶ˩˥]; oxax [œ˩˥.ɶ˩˥] in that order) from the start or having the movements faux-merge into the locative by tonal derivation more natural since all four particles are still in use?

[TurkeySloth]

Are the changes below plausible? I've broken them up because the questions are for specific stages. I'm aware that [ʟ̝̊] is rare, so please don't mention it.

Direct without appropriate approximants
[ɾ~ɺ → ɹ̝̆~ɮ̆]

Direct after devoicing even though they'd been voiced
[ɹ̥̆˔~ɬ̆ → x~ʟ̝̊]

[TurkeySloth]

Do you guys see a box the inverted ts glyph (ƾ), Greek stigma (ϛ), both, or neither? I need to know because I'll transcribe Imperial Creole's moraic consonant as the more widely-seen glyph as their linguists based the phoneme's shape off the local syllabary's character for it, with those two being the closest computer-friendly glyphs to it.

[TomHChappell]

Do you guys see a box the inverted ts glyph (ƾ), Greek stigma (ϛ), both, or neither? I need to know because I'll transcribe Imperial Creole's moraic consonant as the more widely-seen glyph as their linguists based the phoneme's shape off the local syllabary's character for it, with those two being the closest computer-friendly glyphs to it.

I see both of them as correct glyphs, not empty boxes or question-marks.

OTOH I think it might be wise to limit the number of diacritical marks you put on any one glyph.
Maybe two or three should be the max.
Also; are you going to allow all of pre-superscript, right-on-top-of-it-center-superscript, and post-superscript?
If so, is a center-superscript going to rule out the pre-superscript and post-superscript?

Same questions for pre-subscript, right-under-it-center-subscript, and post-subscript.

Are you going to prohibit center-superscripts on glyphs with ascenders?
Are you going to prohibit center-subscripts on glyphs with descenders?
Are you going to prohibit two pre-superscripts at once? Or two center-superscripts at once? Or two post-superscripts at once?
Same questions with various pre- and center- and post- -subscripts.

Are you going to prohibit diacritical marks on diacritical marks, like A^BC?

[akam chinjir]

I see them both too. (I'd be stunned if someone could see ƾ but not ϛ. Incidentally, the unicode description of the first one is "Latin letter inverted glottal stop with stroke," not "inverted ts.")

[bradrn]

Do you guys see a box the inverted ts glyph (ƾ), Greek stigma (ϛ), both, or neither? I need to know because I'll transcribe Imperial Creole's moraic consonant as the more widely-seen glyph as their linguists based the phoneme's shape off the local syllabary's character for it, with those two being the closest computer-friendly glyphs to it.

I see them both correctly as well. In fact, the only problem I’ve had with the forum font so far is with the vowel-caron letters ⟨ǎǐǒǔ⟩, which appear with the caron next to them rather on top. (The font seems to be falling back to Tahoma (I think) with those characters, as Trebuchet MS doesn’t support them.)

I'd be stunned if someone could see ƾ but not ϛ.

I’d be fairly surprised as well.

Incidentally, the unicode description of the first one is "Latin letter inverted glottal stop with stroke," not "inverted ts."

True, but it was originally derived from a t-s ligature, and it was used for the affricative /t͡s/.

[akam chinjir]

Incidentally, the unicode description of the first one is "Latin letter inverted glottal stop with stroke," not "inverted ts."

True, but it was originally derived from a t-s ligature, and it was used for the affricative /t͡s/.
[/quote]

Oh, cool!

[bradrn]

Incidentally, the unicode description of the first one is "Latin letter inverted glottal stop with stroke," not "inverted ts."

True, but it was originally derived from a t-s ligature, and it was used for the affricative /t͡s/.

Oh, cool!

Yep, I think so too! There’s a whole list of obsolete IPA letters on Wikipedia, which includes lots of similarly weird ones.

[TurkeySloth]

Do you guys see a box the inverted ts glyph (ƾ), Greek stigma (ϛ), both, or neither? I need to know because I'll transcribe Imperial Creole's moraic consonant as the more widely-seen glyph as their linguists based the phoneme's shape off the local syllabary's character for it, with those two being the closest computer-friendly glyphs to it.

I see both of them as correct glyphs, not empty boxes or question-marks.

OTOH I think it might be wise to limit the number of diacritical marks you put on any one glyph.
Maybe two or three should be the max.
Also; are you going to allow all of pre-superscript, right-on-top-of-it-center-superscript, and post-superscript?
If so, is a center-superscript going to rule out the pre-superscript and post-superscript?

Same questions for pre-subscript, right-under-it-center-subscript, and post-subscript.

Are you going to prohibit center-superscripts on glyphs with ascenders?
Are you going to prohibit center-subscripts on glyphs with descenders?
Are you going to prohibit two pre-superscripts at once? Or two center-superscripts at once? Or two post-superscripts at once?
Same questions with various pre- and center- and post- -subscripts.

Are you going to prohibit diacritical marks on diacritical marks, like A^BC?

With the vowels having eight marked tones (seven short tones and a long high tone), the syllabary will be limited to one tone per vowel-inclusive glyph. The marked tone will always be on the small glyph of sounds like [çʌ] (<sju>). Keeping that example, if the [çʌ] glyph includes the diacritic for peaking tone—making the result [çʌ²⁷], the following bare [ʌ] (<u>) glyph mustn't carry the same tone. Thus, [çʌ²⁷ʌ] is fine whereas [çʌ²⁷ʌ²⁷] isn't.

[TomHChappell]

With the vowels having eight marked tones (seven short tones and a long high tone), the syllabary will be limited to one tone per vowel-inclusive glyph. The marked tone will always be on the small glyph of sounds like [çʌ] (<sju>). Keeping that example, if the [çʌ] glyph includes the diacritic for peaking tone—making the result [çʌ²⁷], the following bare [ʌ] (<u>) glyph mustn't carry the same tone. Thus, [çʌ²⁷ʌ] is fine whereas [çʌ²⁷ʌ²⁷] isn't.

I understand that your system won’t let what I was afraid of happen;
but I still don’t quite understand how it works!

For instance, how do I decode the title of your thread?

[TurkeySloth]

With the vowels having eight marked tones (seven short tones and a long high tone), the syllabary will be limited to one tone per vowel-inclusive glyph. The marked tone will always be on the small glyph of sounds like [çʌ] (<sju>). Keeping that example, if the [çʌ] glyph includes the diacritic for peaking tone—making the result [çʌ²⁷], the following bare [ʌ] (<u>) glyph mustn't carry the same tone. Thus, [çʌ²⁷ʌ] is fine whereas [çʌ²⁷ʌ²⁷] isn't.

I understand that your system won’t let what I was afraid of happen;
but I still don’t quite understand how it works!

For instance, how do I decode the title of your thread?

As the system doesn't have tone sandhi, the current, albeit out of date, title's [ɔ̥ᵝ, o̥ᵝ, ɨ̥] can be read as being either mid tone and short or toneless with [ə̥˦:] being held slightly longer than its short counterpart, [ə̥˦] (think English sea/see, usually, [si:] v. French si (if) [si]).

[TurkeySloth]

Before I go into as much detail as I have about a language isolate spoken on my setting's home world, I'll provide some detail about the world itself, albeit with outdated names.

"Batíkí" is a massive planet. Although there are several small islands dotting the Nubian Sea, the planet is dominated by the continent of "Balinè". Due to several battles involving powerful mages, "Batíkí" is pot-marked with rifts that, somewhat mysteriously, only seem to affect the land. These rifts are why natives call the plane The Forsaken Riftlands. A substance called mana, which can be semi-solid or liquid in nature, flows within the rifts. Another side effect of these battles is the opening of portals in random locations on the planet.

Before the mage battles began, the Ocean deity, "Meng", grew to favor the inhabitants of the continent "Llanowar," who I've coined the "oíƿkusjaxƿ" ( /ʍi˦ːƿ.kʌ.çɔ˩˥ƿ/ [ʍy˦ːŋ.kʌ.çɔ˩˥ŋ]) in the Lexicon Sculpting thread, and sunk the continent to take them under their protection.

Sadly, I don't know much about the language, itself, other than it combining the broad/slender distinction of the Gaelic languages (Irish, Manx, Scottish) with yet-to-be-determined features of the other three living Celtic languages right now.

[TurkeySloth]

"Llanoran" phonology Parenthetical phonemes share orthographies.
/(mˠ mʲ) (m̥ˠ m̥ʲ) (ŋ ɲ) (n̪ˠ n̪ʲ)/ <m ṁ n ṅ>
/(pˠ pʲ) (bˠ bʲ) (t̪ˠ ʧ) (d̪ˠ ʤ) (k c) (g ɟ)/ <p b t d k g>
/(w vʲ) (fˠ fʲ) (s̪ˠ ʃ) (ɣ ʝ) (x ç) (ʟ ʎ) (ʟ̝̊ ʎ̝)/ <ḃ f s ġ h l ḣ>
/(ɣ˕ ʝ˕) (ɾˠ ɾʲ) ʍ/ <j r ḟ>

Vowels
Front/slender
/æ æː e̞ e̞ː ø̞~y ø̞ː~yː/ <ă ắ e é y ý>

Back/broad
/ɑ ɑː o̞ o̞ː u uː/ <a á o ó u ú>

As the language obeys the "broad with broad, slender with slender" orthographic paradigm, what's the closest pronunciation to [ʍɑ.y] speakers can generate? Currently, the affected name is spelled Ḟaoyṁ.

[Xwtek]

Actually the most likely broad equivalent for /ʎ/ is /ɫ/, with an allophone of /l/ before front vowel if you broad consonant followed with front vowel. Also optional, but I think it's best to have /xʷ/ or /f/ and /ɬ/ instead of /ʍ/ and /ʟ̝̊/

[TurkeySloth]

Actually the most likely broad equivalent for /ʎ/ is /ɫ/, with an allophone of /l/ before front vowel if you broad consonant followed with front vowel. Also optional, but I think it's best to have /xʷ/ or /f/ and /ɬ/ instead of /ʍ/ and /ʟ̝̊/

All of those changes are previous stages in history.
1. [l̪ˠ → ɫ̪ → ʟ]
2a. [xʷ → ʍ], b. [xᶣ → ʍʲ → ʍ]
3. [ɬ̪ˠ → ʟ̝̊]

Does Luhúyrắ or Lauhúyrắ yield [ʟu.xuː.ɾʲæː] seeing as <l, h> are both broad?

[TurkeySloth]

Can fricative consonants coexist with approximants at the same POA? For example, Imperial Creole contrasts voiceless labiodental fricatives with voiceless/voiced labiodental approximants and voiced palatal/velar fricative/approximant consonant ([ʝ̞]) pairs.

[bradrn]

Can fricative consonants coexist with approximants at the same POA? For example, Imperial Creole contrasts voiceless labiodental fricatives with voiceless/voiced labiodental approximants and voiced palatal/velar fricative/approximant consonant ([ʝ̞]) pairs.

It might be better to ask this in the Conlang Random Thread.

[TurkeySloth]

Ăjljykyfă’s [æ.ɿɬ̪ˠɿy.cy.fʲæ] new phonology.

Consonants^{N1, 3, 4}

	Bilabial	Labiodental	Alveolar	Palato-alveolar	Palatal	Velar	Pharyngeal	Glottal
Nasal	m <ṁ, m>				ɲ <ṅ, n>	ŋ <ṅ, n>
Plosive	p b <p, b>		t d <t, d>		c ɟ <k, g>	k g <k, g>
Fricative		f <f>	s <s> s z <ṫ, ḋ>	ʃ ʒ <s, ḋ>	ç ʝ <x, ġ>	x ɣ <x, ġ>	ħ <ṡ>	h <ṡ>
Affricate				ʧ ʤ <t, d>
ApproximantN5		ʋ <ḟ, ḃ>			ʝ <w>	ɣ <w> [x, ɣ] <ṅ, n>
Lateral			ɬ l <l, ŀ>

Vowels^{N2, 6}

	Front	Central	Back
High	I <i, í> Y <y, ý>		U <u, ú>
Mid	E <e, é>	ɿ <j>	O <o, ó>
Low	Æ <æ, ǽ>		A <a, á>

Notes
1. Come in velar(ized) ([ˠ])-palatal(ized) ([ʲ]) pairs depending on adjacency to back vowels, front vowels, and/or the glide vowel (see below)
2. Central glide vowel, IPA [ə̯], used between a velar(izaed) consonant/[ħ] and a front vowel or a palatal(ized) consonant/[h] and a back vowel
3. Dotted orthographies, such as ṁ, are lenited. Of these, <ṁ, ṅ, ḟ> are voiceless [m̥, ɲ̊, ŋ̊, ʋ̥] and both iterations of <ṫ>, broad <ḋ>, and both iterations of <ŀ> are dentialveolar [s̪, z̪, ɫ̪, l̪ʲ]
4. Additionally to the dentialveolars above, broad <t>, broad <d>, and both iterations of <l> are [t̪, d̪, ɬ̪]
5. [ʝ, ɣ, x, ɣ] are [ʝ˕, ɣ˕, x̞̃, ɣ̃˕], with [x̞̃, ɣ̃˕] as allophones of [ŋ̊, ŋ] word-initially
6. [A, Æ, E, I, O, U, Y] represent [ɑ(ː), æ(ː), e̞(ː), i(ː), o̞(ː), u(ː), ʏ(ː)], of which [æ(ː)] are near-low, [e̞(ː), o̞(ː)] are true mid, and [ʏ(ː)] are near-high

[Xwtek]

Your language lacks /i/? I don't think it's realistic. Your /Y/ is more likely to unround instead.

[TurkeySloth]

Your language lacks /i/? I don't think it's realistic. Your /Y/ is more likely to unround instead.

As stated in the post, the vowel was freely variable until I merged it into [ʏ(ː)]. In other news, <i> has returned.