Amanghu (天語), or Chinese Altaic

A forum for all topics related to constructed languages
pbastronaut
rupestrian
rupestrian
Posts: 14
Joined: 17 Feb 2018 13:02

Amanghu (天語), or Chinese Altaic

Post by pbastronaut »

Hi all,

I've been working on a new Chinese-inspired conlang, just for fun. I found that agglutinative languages just weren't for me, so I've gone hard in the opposite direction. I haven't got that far so far, but I'm really hoping to get some feedback. I've pretty much just figured out some verb stuff, basic grammar and a handful of vocab. As a full disclaimer, I don't speak or read Chinese, so I'm not sure this will be entirely realistic, but I'd much rather make it interesting than realistic anyway.

A few notes:
  • Amanghu is SOV.
  • I use a lot of Cyrillic characters with non-standard accents (or at least, accents that are not standard on the letters I'm using them on) and I don't know if this is the same for everyone, but on my browser they don't work well on this forum.
  • Amanghu is meant to be descended from an ancestor language somewhere between Proto-Sino-Tibetan and Old Chinese. It's Sinitic, but not actually Chinese. Hopefully this will give me some linguistic breathing room.
  • I've tried to be consistent in my use of Chinese characters, but I've had to be creative with my particles. I accept that this probably looks like a mess to any Chinese speakers out there.
  • I'm not a skilled linguist and this is only something like my third proper effort at a conlang, so I'm sorry if my terminology is kinda rough.
Image

I. Introduction
Amanghu (CYRILLIC: ама̎ңюу̎ HANZI: 天語 IPA: amáŋɥɯ́), also known erroneously as Tienic Chinese, Gam Chinese, and Chinese Altaic, is a mostly-analytic, isolating Sino-Tibetan isolate spoken mainly in Altay Prefecture, Xinjiang, in the People's Republic of China, as well as neighbouring regions of Mongolia, Kazakhstan, and Russia. It's the historic language of the Gam people, also known as the Blue People (CYRILLIC: г̆а̎мцѡ̎ıк HANZI: 藍族 IPA: /ɡʰámt͡só̰k/). It's spoken by roughly half a million people as a first language.

Old Amanghu was originally spoken by nomadic tribes in what is now north-western China between c.200 AD and c.800 AD. The ancestor of Old Amanghu is commonly considered to be a Sinitic language with a close relationship to Old Chinese, but this classification remains contentious. Amanghu vocabulary is largely sourced from Old Chinese, with a large number of loan words from Kyrgyz, Mandarin and Cantonese. Although Amanghu is often referred to as a variety of Chinese, it is distinct from "true" Chinese languages as it is not descended from Old Chinese.

Amanghu is written in both Chinese characters and a variety of Cyrillic, and both scripts are frequently used, with Cyrillic being the dominant script outside of China.

II. Phonology and Orthography
Most consonants have phonemically distinct aspirated and non-aspirated forms. Amanghu is a tonal language with two tones: low, or neutral (which is unmarked) and high. Additionally, vowels may be creaky-voiced or clear. These two vowel distinctions combine to give the language four tone-like vowel qualities. There are also three glide consonants: /j, w, ɥ/.

/m, n, ŋ/ <м, н, ң>
/p, b, t, d, k, ɡ/ <п, б, т, д, к, г>
/pʰ, bʰ, tʰ, dʰ, kʰ, ɡʰ/ <п̆, б̆, т̆, д̆, к̆, г̆>
/s, ʃ, ʂ, x, h/ <с, ш, щ, х, һ>
/t͡s, t͡ʃ, t͡ʂ, k͡x/ <ц, ч, ж, ҡ>
/t͡sʰ, t͡ʃʰ, t͡ʂʰ, k͡xʰ/ <ц̆, ч̆, ӂ, ҡ̆>

/j, w, ɥ/ <ь, в, ю>

/i, ḭ, í, ḭ́, ɯ, ɯ̰, ɯ́, ɯ̰́/ <и, иı, и̎, и̎ı, у, уı, у̎, у̎ı>
/e, ḛ, é, ḛ́, o, o̰, ó, ó̰/ <е, еı, е̎, е̎ı, ѡ, ѡı, ѡ̎, ѡ̎ı>
/a, a̰, á, á̰/ <а, аı, а̎, а̎ı>

Amanghu has four diphthongs in each of the four "tones": /ie, ɯe, ei, eɯ/, and their tonal variants.

Amanghu has a simple (C)(G)V(C) structure. Syllables can begin with any consonant or glide except for /ŋ/, and end in any nasal or stop consonant.

III. Numbers
Just for fun, before I get to the complex bit, here's the ordinal one to ten in Amanghu:
  1. 一ки̎г /kíɡ/
  2. 二 ни /ni/
  3. 三 су̎ıм /sɯ̰́m/
  4. 四 ҡеб /k͡xeb/
  5. 五 наı /na̰/
  6. 六 луг /lɯg/
  7. 七 си̎д /síd/
  8. 八 пе̎ıд /pḛ́d/
  9. 九 ку /kɯ/
  10. 十 гьу̎б /ɡjɯ́b/
IV. Verbs
This so far is the only really developed aspect of the language. I'll try and give a brief rundown here, but there are more details in my article on ConWorkShop, here. I use a lot more examples in the article, which I think gives a better feel for how the language is meant to sound.

Tense
Amanghu is distinguishable from varieties of true Chinese by the presence of particles that indicate tense. These particles are placed before the verb they mark. In the majority of sentences, especially simple sentences in the past tense, these particles are not used. This is because tense is generally implied through other means. For example, the perfect aspect suggests that a verb is in the past tense unless context suggests otherwise. Additionally, tense particles are not used where the sentence is placed in a specific timeframe. For example, 「厶走上星期一」 "вие̎ сѡı дьаң се̎ıңг̆у-ки̎г" /wié so̰ djaŋ sḛ́ŋgʰɯkíg/: "I will go on Monday."

Present tense verbs are unmarked, while past tense verbs are preceded by 旦 юѡ̎н /ɥón/, and future tense verbs are preceded by 弗 щѡı /ʂo̰/. Although both particles have their own pronunciations, they are generally not pronounced in speech, except in archaic or emphatic contexts. Instead they affect lenition of the onset of the verb they affect. For example, the verb 吃 к̆а̎ /kʰá/ "to eat" becomes 旦吃 гк̆а̎ /gʰá/ in the past tense, and 弗吃 хк̆а̎ /xá/ in the future tense.

Lenition is very regular and occurs through either sonorization, where a phoneme becomes more voiced, and debuccalization, where a point of articulation of the phoneme moves towards the glottis. Past tense lenitions become more sonorized (that is, more voiced) and more debuccalized (that is, more glottised), while future tense lenitions do the opposite.

Aspect
Amanghu uses verb-ending particles to distinguish between four morphologically-distinct aspects: simple/perfect (see below), prospective, progressive, and gnomic. Simple/perfect verbs are unmarked, whereas the other three are marked by the particles 正 тиең /tieŋ/, 了 веı /wḛ/, and 有 хуе /xɯe/, respectively.

Unmarked: Simple/Perfect
Unmarked verbs conflate simple and perfect aspects. The sentence 「厶上鎮走」 "вие̎ дьаң щаı сѡı" /wié djaŋ ʂa̰ so̰/ "I walked to town" has a more literal reading of "I had walked to town", indicating that the action occurred earlier than the time under consideration. Note that the example sentence does not use the past tense particle 旦 юѡ̎н, as the aspect particle implies that the action took place in the past. 旦 might be used, however, if you wanted to distinguish this sentence from an action in the present tense, or to specify that this action took place in the past.

In simple sentences involving only one main verb, such as the initial example, 「厶上鎮走」, the unmarked verb can usually be considered to take place in the simple aspect, but it is far more common to use the progressive particle when referring to actions which are still ongoing within the sentence's timeframe.

正 тиең: Prospective
The prospective particle 正 тиең /tieŋ/ marks verbs as actions that are imminent or upcoming. It is the equivalent of the English "about to". Unlike the simple/perfect aspect particle, 正 does not imply any particular tense. The sentence 「厶旦吃正」 "вие̎ к̆уд тиең" /wié kʰɯd tieŋ/ "I am about to eat" is steadfastly present tense.

了 веı: Progressive
The progressive aspect is marked with the particle 了 веı /wḛ/. It's notable that this is typologically identical to the Standard Chinese perfective particle 了, with which it is a cognate. Despite their shared etymology, however, the two particles function very differently. The Amanghu 了 functions similarly to the English suffix "-ing", and is used for both progressive and continuous aspects.

了 is generally omitted in simple sentences where there is sufficient context or where it is not considered to be relevant. There is generally little meaningful distinction between the sentences 「厶吃」 "I eat" and 「厶吃了」 "I am eating", for example, if the addressee can see if one is still eating. 了 is still frequently used to distinguish an action from the perfect aspect. Note that the unmarked 「厶吃」 can be read as either "I eat" or "I have eaten", leaving it ambiguous as to whether the act of eating has concluded or not, but strongly implying that it has. 了 clears up this ambiguity and firmly declares the verb as an ongoing action.

有 хуе: Gnomic
The gnomic particle 有 хуе /xɯe/ is used to declare general truths or aphorisms. For example, 「馬跑有」 "маı буı хуе" /ma̰ bɯ̰ xɯe/ "horses run" is a general claim about the nature of all horses. It is also used to declare axiomatic or even hypothetical and assumed truths, such as 「李師父亡老是有」 "Ру-си̎ба ма̎ң руı щуıң хуе" /ɻɯ síba máŋ ɻɯ̰ ʂɯ̰ŋ xɯe/ "Master Ru is not old", in a context where Ru's age or vitality is in question.

有 also functions as a marker for the aorist or preterite, declaring actions as having been fully completed in the past.
User avatar
Frislander
mayan
mayan
Posts: 2088
Joined: 14 May 2016 18:47
Location: The North

Re: Amanghu (天語), or Chinese Altaic

Post by Frislander »

Firstly, I do really like the look of this, I think Sinitic as a family doesn't get enough love, and I would really like to see both a description of your parent language and the changes which came down into the modern language. However I do have a couple of questions. Firstly why do you refer to it as "Chinese Altaic" when 1 Altaic isn't a family, more a linguistic area and 2 the language is Sinitic not any of the traditional "Altaic" families. I assume you call it that because it is in a contact situation, but then the better description would be "Altaic Chinese". Also while I like the throughgoing aspiration contrast, I struggle to see how a /k͡x k͡xʰ/ contrast can be maintained without one or the other collapsing into something else.
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: Amanghu (天語), or Chinese Altaic

Post by Creyeditor »

Frislander wrote: 22 Mar 2018 10:40 Firstly why do you refer to it as "Chinese Altaic" when 1 Altaic isn't a family, more a linguistic area and 2 the language is Sinitic not any of the traditional "Altaic" families. I assume you call it that because it is in a contact situation, but then the better description would be "Altaic Chinese".
pbastronaut wrote: 21 Mar 2018 23:54 I. Introduction
Amanghu (CYRILLIC: ама̎ңюу̎ HANZI: 天語 IPA: amáŋɥɯ́), also known erroneously as Tienic Chinese, Gam Chinese, and Chinese Altaic, is a mostly-analytic, isolating Sino-Tibetan isolate spoken mainly in Altay Prefecture, Xinjiang, in the People's Republic of China,
(italics mine)
Doesn't this at least partly answer your question?
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
pbastronaut
rupestrian
rupestrian
Posts: 14
Joined: 17 Feb 2018 13:02

Re: Amanghu (天語), or Chinese Altaic

Post by pbastronaut »

Frislander wrote: 22 Mar 2018 10:40 Firstly, I do really like the look of this, I think Sinitic as a family doesn't get enough love, and I would really like to see both a description of your parent language and the changes which came down into the modern language. However I do have a couple of questions. Firstly why do you refer to it as "Chinese Altaic" when 1 Altaic isn't a family, more a linguistic area and 2 the language is Sinitic not any of the traditional "Altaic" families. I assume you call it that because it is in a contact situation, but then the better description would be "Altaic Chinese". Also while I like the throughgoing aspiration contrast, I struggle to see how a /k͡x k͡xʰ/ contrast can be maintained without one or the other collapsing into something else.
Thanks for responding! I think you're right that "Altaic Chinese" might be more accurate. I think I was thinking about it from reverse when I came up with the term; that the languge is "Altaic" by proximity, distinguished by its Chinese-ness, rather than "Chinese" by origin, distinguished by its Altaic-ness.

/k͡x, k͡xʰ/ has been a bit of a headache. My current fix is that /k͡xʰ/ would be more commonly realised as [k͡xh] or [k͡xəh]. But in the end I might drop the aspirated affricates, since they're barely turning up in my lexicon so far.

I'll try and write up something about Old Amanghu sound changes.
User avatar
Vlürch
greek
greek
Posts: 452
Joined: 09 Mar 2016 21:19
Location: Finland
Contact:

Re: Amanghu (天語), or Chinese Altaic

Post by Vlürch »

Really interesting! There's always something about both Chinese- and Altaic-influenced conlangs that's nice, so the two influences together is awesome (and something I've also tried to do myself, but always end up failing for various reasons), especially with the alternative Cyrillic orthography giving it some of that flavour that Dungan and Uyghur have, and of course the use of hanzi for conlangs is always cool.

As for the phonemic distinction between /k͡x/ and /k͡xʰ/ (and even just /kʰ/ and /k͡x/), that is pretty weird; maybe it would be conserved simply so that the native speakers could gloat about how hard their language is for others, who struggle to even hear the difference between the obviously different sounds...? [:P]

And yeah, this forum (and forums in general) tend to mess up diacritics.
User avatar
sangi39
moderator
moderator
Posts: 3024
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Re: Amanghu (天語), or Chinese Altaic

Post by sangi39 »

Personally I'd keep the aspirated affricates, and as for /k͡x/ vs. /k͡xʰ/, the distinction could always be diaphonemic, i.e. different dialects have different realisation for these two sounds, but for the sake of representing the majority of dialects, /k͡x k͡xʰ/ is used.

So for example, in one dialect, they might merge into a single phoneme, either /k͡x/ or /k͡xʰ/, while in another /k͡x/ might be maintained as distinct while /k͡xʰ/ merges into /x/ (a similar shift is attested in Sotho, but it lacks /k͡x/). /kʰ/ could similarly merge into /k͡xʰ/ in a third dialect, with /k͡x/ shifts to /x/ (you could have original /x/ either stay where it is, or shift to /h/). So you have Dialect 1, Dialect 2 and Dialect 3 with the following correspondences:

Code: Select all

Diaphoneme - Dialect 1 - Dialect 2 - Dialect 3 - 
   /kʰ/    -   /kʰ/    -   /kʰ/    -   /kxʰ/
   /kx/    -   /kxʰ/   -   /kx/    -   /x/
   /kxʰ/   -   /kxʰ/   -   /x/     -   /kxʰ/
   /x/     -   /x/     -   /x/     -   /h/
   /h/     -   /h/     -   /h/     -   /h/
   
No one dialect contrasts all five sounds, but there's still five different diaphonemes being represented in the modern dialects as a whole
You can tell the same lie a thousand times,
But it never gets any more true,
So close your eyes once more and once more believe
That they all still believe in you.
Just one time.
User avatar
Ghoster
cuneiform
cuneiform
Posts: 152
Joined: 29 Nov 2010 19:30

Re: Amanghu (天語), or Chinese Altaic

Post by Ghoster »

For example, the verb 吃 к̆а̎ /kʰá/ "to eat" becomes 旦吃 гк̆а̎ /gʰá/ in the past tense, and 弗吃 хк̆а̎ /xá/ in the future tense.
This one I like a lot. It looks like analytical Japanese.

Also I especially like the name of the language. There's something very unique about Chinese characters and matching right sounds with right characters seems to be a very interesting area in conlanging. "Amanghu" for 天语 looks like a great match.
Native: :pol:; Advanced: :usa: :chn: :nld:; Intermediate: :esp:; Lower intermediate: :deu: :rus: :fra: :nor:; Beginner: :jpn: :kor: :hkg:
pbastronaut
rupestrian
rupestrian
Posts: 14
Joined: 17 Feb 2018 13:02

Re: Amanghu (天語), or Chinese Altaic

Post by pbastronaut »

Vlürch wrote: 22 Mar 2018 12:43 Really interesting! There's always something about both Chinese- and Altaic-influenced conlangs that's nice, so the two influences together is awesome (and something I've also tried to do myself, but always end up failing for various reasons), especially with the alternative Cyrillic orthography giving it some of that flavour that Dungan and Uyghur have, and of course the use of hanzi for conlangs is always cool.
I'd be really interested to hear about your previous efforts, as well as where you ran into difficulty. It would be interesting to see what kind of things we did similarly, and what we did differently.
sangi39 wrote: 22 Mar 2018 15:40 Personally I'd keep the aspirated affricates, and as for /k͡x/ vs. /k͡xʰ/, the distinction could always be diaphonemic, i.e. different dialects have different realisation for these two sounds, but for the sake of representing the majority of dialects, /k͡x k͡xʰ/ is used.

So for example, in one dialect, they might merge into a single phoneme, either /k͡x/ or /k͡xʰ/, while in another /k͡x/ might be maintained as distinct while /k͡xʰ/ merges into /x/ (a similar shift is attested in Sotho, but it lacks /k͡x/). /kʰ/ could similarly merge into /k͡xʰ/ in a third dialect, with /k͡x/ shifts to /x/ (you could have original /x/ either stay where it is, or shift to /h/). So you have Dialect 1, Dialect 2 and Dialect 3 with the following correspondences:
That's a genius idea, and something I never would've considered. This might actually be helped by the real-world national borders through the territory where Amanghu is spoken. There would be clear Kyrgyz, Mongolian and Russian dialects. I'll sit down and work out how this could be implemented. Thank you.
Ghoster wrote: 22 Mar 2018 22:30 Also I especially like the name of the language. There's something very unique about Chinese characters and matching right sounds with right characters seems to be a very interesting area in conlanging. "Amanghu" for 天语 looks like a great match.
Thank you. I actually need to go back and revise how I've implemented sound changes over time, so I definitely agree that it's hard to match characters up to sounds creatively. In case you're interested in the etymology here, I took "天", "ама̎ң" from the Kyrgyz "асман", "sky", and "語" "юу̎" from the Old Chinese /*ŋaʔ/.
clawgrip
MVP
MVP
Posts: 2257
Joined: 24 Jun 2012 07:33
Location: Tokyo

Re: Amanghu (天語), or Chinese Altaic

Post by clawgrip »

pbastronaut wrote: 23 Mar 2018 00:06In case you're interested in the etymology here, I took "天", "ама̎ң" from the Kyrgyz "асман", "sky", and "語" "юу̎" from the Old Chinese /*ŋaʔ/.
I had just assumed Japanese, because 天 can be pronounced ama in Japanese.
User avatar
Vlürch
greek
greek
Posts: 452
Joined: 09 Mar 2016 21:19
Location: Finland
Contact:

Re: Amanghu (天語), or Chinese Altaic

Post by Vlürch »

Ghoster wrote: 22 Mar 2018 22:30Also I especially like the name of the language. There's something very unique about Chinese characters and matching right sounds with right characters seems to be a very interesting area in conlanging. "Amanghu" for 天语 looks like a great match.
I agree completely. It's probably perfect, or at least I can't think of anything better; sounds like the name of a real language spoken by a real people. That's one of the reasons why I'd be really interested in knowing about the associated conculture, too!
pbastronaut wrote: 23 Mar 2018 00:06I'd be really interested to hear about your previous efforts, as well as where you ran into difficulty. It would be interesting to see what kind of things we did similarly, and what we did differently.
Oh, my problems started right at the beginning: all my attempts at "Sino-Altaic" conlangs were way too Turkic as soon as I began working on them. [:P] Basically, one of them was just "take Kazakh and replace random words with Mandarin ones, then apply random sound changes like making every /p/ an /f/, and that's it". Another one was a bit more coherent, an attempt at making an a priori language that simply had a very basic Turkic-esque phonology with a few Mandarin features like a contrast between retroflex and alveolopalatal sibilant affricates thrown in. The declension of nouns was all typical stuff like a nasal genitive and whatnot while the pathetic start at a lexicon was just a mix of a priori, Turkic and Mandarin words. It, too, had a Cyrillic orthography.

You can probably tell I don't really know the first thing about Sinitic languages other than a handful of Mandarin words and unironically enjoying its phonology (and occasionally looking up Old Chinese reconstructions), and that I actually prefer Mandarin to all the others based on my very limited knowledge of them (as in I've only read a little bit about their phonology on Wikipedia (and watched a few Cantonese-language films)), even though everyone else seems to think it's the worst... oh well.


I don't want to derail this thread, sorry.
pbastronaut wrote: 23 Mar 2018 00:06I took "天", "ама̎ң" from the Kyrgyz "асман", "sky", and "語" "юу̎" from the Old Chinese /*ŋaʔ/.
Heh, I suspected that was what you derived it from but didn't want to ask because that would've meant pointing out that the word асман in Kyrgyz and its cognates in other Turkic languages are all borrowed from Persian, AFAIK meaning their date of borrowing has to have been in the 9th century or later because A) they almost certainly didn't yet exist in Old Turkic since they're not present in any of the Siberian Turkic languages and B) the Islamic conquest of Transoxania was the earliest stage in the Islamisation of Turkestan, but considering how they had Arabic as a formal language rather than Persian, I'd guess the large-scale borrowing of Persian vocabulary began in the 9th or 10th century when the Samanids ruled the region and accelerated under the Ghaznavids. I don't know if that's the case, though, and I'm just going by possibly faulty logic, borderline inexistent historical knowledge and filling the gaps by skimming through Wikipedia...

Hopefully that doesn't count as the start of a derailment, too...


Anyway, it's not like a medieval etymology would be recent or anything. It's just that if it's a compound of a Persian word and an Old Chinese word, that's chronologically impossible unless the Old Chinese word remained unchanged for a long ass time. Maybe that could be explained away as ама̎ң being a conflation of a native word and the Persian loanword or something like that, if it even matters. It's such an awesome name that changing it would be worse than bending spacetime a little. [:)]
clawgrip wrote: 23 Mar 2018 16:23I had just assumed Japanese, because 天 can be pronounced ama in Japanese.
Interesting, and now I'm confused by not knowing that even though I knew Amaterasu was spelled with it. [:x]

...and here the derailing continues? Uh.
pbastronaut
rupestrian
rupestrian
Posts: 14
Joined: 17 Feb 2018 13:02

Re: Amanghu (天語), or Chinese Altaic

Post by pbastronaut »

Vlürch wrote: 23 Mar 2018 20:23 Oh, my problems started right at the beginning: all my attempts at "Sino-Altaic" conlangs were way too Turkic as soon as I began working on them. [:P] Basically, one of them was just "take Kazakh and replace random words with Mandarin ones, then apply random sound changes like making every /p/ an /f/, and that's it". Another one was a bit more coherent, an attempt at making an a priori language that simply had a very basic Turkic-esque phonology with a few Mandarin features like a contrast between retroflex and alveolopalatal sibilant affricates thrown in. The declension of nouns was all typical stuff like a nasal genitive and whatnot while the pathetic start at a lexicon was just a mix of a priori, Turkic and Mandarin words. It, too, had a Cyrillic orthography.
I think I see. I guess if anything I'm going the opposite route. So far the language is very Chinese-like, with a little vocab from other languages, mainly Kazakh. I was originally considering putting a lot more Turkic features in, but the two language families just seem so different that I can't see a way to do much more. But I'm fine with that.
Vlürch wrote: 23 Mar 2018 20:23 Heh, I suspected that was what you derived it from but didn't want to ask because that would've meant pointing out that the word асман in Kyrgyz and its cognates in other Turkic languages are all borrowed from Persian, AFAIK meaning their date of borrowing has to have been in the 9th century or later because A) they almost certainly didn't yet exist in Old Turkic since they're not present in any of the Siberian Turkic languages and B) the Islamic conquest of Transoxania was the earliest stage in the Islamisation of Turkestan, but considering how they had Arabic as a formal language rather than Persian, I'd guess the large-scale borrowing of Persian vocabulary began in the 9th or 10th century when the Samanids ruled the region and accelerated under the Ghaznavids. I don't know if that's the case, though, and I'm just going by possibly faulty logic, borderline inexistent historical knowledge and filling the gaps by skimming through Wikipedia...
How do you know all this stuff? Don't get me wrong; I'm dead impressed. It's just that I found basically no etymological info on асман, so I just assumed it was native to Kyrgyz. Well, you know what the say about assuming.

I'm sure I can find some excuse to salvage the name, but I'm really concerned now that since I don't know really anything about Turkic languages, I might've opened myself up to a lot of potential mistakes like this. Between this and the wreck I've been making of my reconstructed Old Amanghu phonology, I'm starting to get the feeling that maybe I don't know enough about the languages I'm basing this on to make a decent job of it. I think I have a lot of reading to do.
User avatar
Vlürch
greek
greek
Posts: 452
Joined: 09 Mar 2016 21:19
Location: Finland
Contact:

Re: Amanghu (天語), or Chinese Altaic

Post by Vlürch »

pbastronaut wrote: 24 Mar 2018 01:03How do you know all this stuff? Don't get me wrong; I'm dead impressed.
I don't really know anything, but I do spend a lot of time reading about stuff online (usually Wikipedia at least as a starting point) and search for random .pdf documents every day, especially about Turkic languages and history. I've also developed a habit of bookmarking pretty much everything with detailed tags thanks to a lot of times when I've wanted to link something somewhere but haven't been able to find it again. :roll:
pbastronaut wrote: 24 Mar 2018 01:03It's just that I found basically no etymological info on асман, so I just assumed it was native to Kyrgyz. Well, you know what the say about assuming.
Well, the Persian word for sky is آسمان (āsmān); I suppose it could be a coincidence that the words are so similar, but there are three reasons why I think that's really unlikely: 1) there are tons of Persian loanwords in the Turkic languages where the word for sky is similar to the Persian word, 2) the word for sky is different in the Turkic languages that don't have as many Persian loanwords and 3) no alternative etymologies have even been proposed by anyone AFAIK.

For example, the Turkic language most heavily influenced by Persian is Uzbek, and its word for sky is osmon (same as Tajik осмон; Tajik is the variety of Persian that Uzbek has had the most extensive contact with). Uyghur has also been heavily influenced by Persian, although not as much as Uzbek, and its word for sky is ئاسمان (asman). Kazakh, too, has been influenced by Persian to some extent, and its word for sky is аспан (the /m/ > /p/ shift is a thing in Kazakh after voiceless sounds). And as you know, Kyrgyz has асман.

Turkmen has also been quite heavily influenced by Persian and has both gök and asman. Modern Turkish has only gök because of efforts to reduce loanwords, but Ottoman Turkish had آسمان (asman).

The Turkic languages that haven't been influenced by Persian, on the other hand, have entirely different words: Tuvan has дээр, Altai has теҥери, Khakas has тигір, Sakha has халлаан (while таҥара means "god"), etc.

Here's a link to a site with a somewhat in-depth history of Persian-Turkic contacts, although it's kinda annoying in that it uses the word "Turkish" to mean all the Turkic languages.
pbastronaut wrote: 24 Mar 2018 01:03I'm sure I can find some excuse to salvage the name, but I'm really concerned now that since I don't know really anything about Turkic languages, I might've opened myself up to a lot of potential mistakes like this. Between this and the wreck I've been making of my reconstructed Old Amanghu phonology, I'm starting to get the feeling that maybe I don't know enough about the languages I'm basing this on to make a decent job of it. I think I have a lot of reading to do.
If you want to have words of (more or less) definitely Turkic origin, you can be fairly safe with Proto-Turkic. Regardless of whether or not you believe Altaic is a thing, there's this Proto-Altaic dictionary that's a good resource. If you want to have Persian loanwords but they'd be chronologically impossible, you could instead derive them from Sogdian, Parthian or Middle Persian. Sogdian wouldn't really work for the name since the Sogdian word for sky is smān, unless you had some kind of a shift like /s/ > /h/ > /a/, but Parthian had asmān and Middle Persian had āsmān just like modern Persian. You could easily have it be a loanword from the Sasanid period or even as early as the time of the Parthian empire.

In any case, if you want to be extremely realistic, to get aman (or however the name is syllabified), you'd either have to have an intermediary step where /s/ > /h/ > /Ø/ in the source language (meaning the creation of an Iranian conlang; just a set of sound changes from Parthian or whatever would obviously be enough for the purpose of loanwords) or in the proto-language of Amanghu. The former would of course be much simpler since the latter would mean there having been one more "H-like sound" in the past that was lost... then again, I'm sure that could somehow also be used to explain the presence of all the H-like sounds you currently have?

Well, whatever you do, it'll be interesting! [:D]
Ælfwine
roman
roman
Posts: 940
Joined: 21 Sep 2015 01:28
Location: New Jersey

Re: Amanghu (天語), or Chinese Altaic

Post by Ælfwine »

Nice! I always wanted to do something like this, but with Japanese and Austronesian.
My Blog

A-posteriori, alternative history nerd
User avatar
All4Ɇn
mayan
mayan
Posts: 1765
Joined: 01 Mar 2014 07:19

Re: Amanghu (天語), or Chinese Altaic

Post by All4Ɇn »

I'm not sure how to explain it exactly but something about your conlang has a really interesting vibe that I just can't quite put my finger on. Hope to see more!
pbastronaut
rupestrian
rupestrian
Posts: 14
Joined: 17 Feb 2018 13:02

Re: Amanghu (天語), or Chinese Altaic

Post by pbastronaut »

Vlürch wrote: 24 Mar 2018 16:29 If you want to have words of (more or less) definitely Turkic origin, you can be fairly safe with Proto-Turkic. Regardless of whether or not you believe Altaic is a thing, there's this Proto-Altaic dictionary that's a good resource. If you want to have Persian loanwords but they'd be chronologically impossible, you could instead derive them from Sogdian, Parthian or Middle Persian. Sogdian wouldn't really work for the name since the Sogdian word for sky is smān, unless you had some kind of a shift like /s/ > /h/ > /a/, but Parthian had asmān and Middle Persian had āsmān just like modern Persian. You could easily have it be a loanword from the Sasanid period or even as early as the time of the Parthian empire.

In any case, if you want to be extremely realistic, to get aman (or however the name is syllabified), you'd either have to have an intermediary step where /s/ > /h/ > /Ø/ in the source language (meaning the creation of an Iranian conlang; just a set of sound changes from Parthian or whatever would obviously be enough for the purpose of loanwords) or in the proto-language of Amanghu. The former would of course be much simpler since the latter would mean there having been one more "H-like sound" in the past that was lost... then again, I'm sure that could somehow also be used to explain the presence of all the H-like sounds you currently have?

Well, whatever you do, it'll be interesting! [:D]
Thanks for all the help and resources. I'm really impressed by your knowledge here.

So I'm going though reconstructed Proto-Sino-Tibetan roots, trying to lock down sound changes over time (I may end up revising some of my vocab, but so far nothing's so far off that I can't fudge a few of the sounds to match what I already have). And what I'm thinking is ... folk etymology. There's the PST /*mraŋ/, which is the root of the Chinese 矕 "to see". I'm thinking the name was originally something like /maŋ ŋa/ "seen language", as in a written language. Perhaps with the initial /a/ as a now-defunct particle or inflection for the perfective. As the early Gam people settled in northern Xinjiang, they would have eventually encountered the Kazakh аспан (or its historical equivalent), adopted it as a common word for "sky", conflated the two words, and Bob's your uncle. How does this sound?

I'm putting some serious effort into Amagnhu's phonological history, so until I'm done, I'll probably hold off posting on this thread. But once I'm done, I think adjectives are the next step.
shimobaatar
korean
korean
Posts: 10373
Joined: 12 Jul 2013 23:09
Location: UTC-04:00

Re: Amanghu (天語), or Chinese Altaic

Post by shimobaatar »

pbastronaut wrote: 21 Mar 2018 23:54 I've been working on a new Chinese-inspired conlang, just for fun. I found that agglutinative languages just weren't for me, so I've gone hard in the opposite direction. I haven't got that far so far, but I'm really hoping to get some feedback. I've pretty much just figured out some verb stuff, basic grammar and a handful of vocab. As a full disclaimer, I don't speak or read Chinese, so I'm not sure this will be entirely realistic, but I'd much rather make it interesting than realistic anyway.
Oh, I remember seeing the inventory for this in the random phonology thread a little while ago. So nice to see this thread!

And if realism isn't your goal, don't worry about it! [:D]
pbastronaut wrote: 21 Mar 2018 23:54
  • Amanghu is meant to be descended from an ancestor language somewhere between Proto-Sino-Tibetan and Old Chinese. It's Sinitic, but not actually Chinese. Hopefully this will give me some linguistic breathing room.
What a neat idea! I assume you've looked at these languages?
pbastronaut wrote: 21 Mar 2018 23:54
  • I'm not a skilled linguist and this is only something like my third proper effort at a conlang, so I'm sorry if my terminology is kinda rough.
No worries! Most conlangers aren't linguists.

Love the flag, by the way.
pbastronaut wrote: 21 Mar 2018 23:54 Amanghu (CYRILLIC: ама̎ңюу̎ HANZI: 天語 IPA: amáŋɥɯ́), also known erroneously as Tienic Chinese, Gam Chinese, and Chinese Altaic, is a mostly-analytic, isolating Sino-Tibetan isolate spoken mainly in Altay Prefecture, Xinjiang, in the People's Republic of China, as well as neighbouring regions of Mongolia, Kazakhstan, and Russia. It's the historic language of the Gam people, also known as the Blue People (CYRILLIC: г̆а̎мцѡ̎ıк HANZI: 藍族 IPA: /ɡʰámt͡só̰k/). It's spoken by roughly half a million people as a first language.
Right from the start, Amanghu already has such a unique feel to it, both aesthetically and in terms of "worldbuilding", so to speak. The erroneous names really add a lot to the feel of this project, I think. I know you said you're not going for naturalism, but details like that make the language feel more real, at least to me.
pbastronaut wrote: 21 Mar 2018 23:54 Most consonants have phonemically distinct aspirated and non-aspirated forms. Amanghu is a tonal language with two tones: low, or neutral (which is unmarked) and high. Additionally, vowels may be creaky-voiced or clear. These two vowel distinctions combine to give the language four tone-like vowel qualities. There are also three glide consonants: /j, w, ɥ/.
I like the "voiced aspirates", which especially help to set Amanghu apart from other Sinitic languages, at least as far as I know. I also like the velar affricates. Like I said in the random phonology thread, and like others have apparently said here, I don't think those should be a problem. I also like the combination between tone and phonation. Is there any tone sandhi?

Also, out of curiosity, is there a romanization?
pbastronaut wrote: 21 Mar 2018 23:54 Amanghu has four diphthongs in each of the four "tones": /ie, ɯe, ei, eɯ/, and their tonal variants.

Amanghu has a simple (C)(G)V(C) structure. Syllables can begin with any consonant or glide except for /ŋ/, and end in any nasal or stop consonant.
Are these falling or rising diphthongs?

Are all consonant-glide combinations possible? Similarly, are all glide-vowel combinations possible? Are aspirated stops allowed in the coda position?
pbastronaut wrote: 21 Mar 2018 23:54 Just for fun, before I get to the complex bit, here's the ordinal one to ten in Amanghu:
I'm sure Janko will be very appreciative.
pbastronaut wrote: 21 Mar 2018 23:54 Present tense verbs are unmarked, while past tense verbs are preceded by 旦 юѡ̎н /ɥón/, and future tense verbs are preceded by 弗 щѡı /ʂo̰/. Although both particles have their own pronunciations, they are generally not pronounced in speech, except in archaic or emphatic contexts. Instead they affect lenition of the onset of the verb they affect. For example, the verb 吃 к̆а̎ /kʰá/ "to eat" becomes 旦吃 гк̆а̎ /gʰá/ in the past tense, and 弗吃 хк̆а̎ /xá/ in the future tense.

Lenition is very regular and occurs through either sonorization, where a phoneme becomes more voiced, and debuccalization, where a point of articulation of the phoneme moves towards the glottis. Past tense lenitions become more sonorized (that is, more voiced) and more debuccalized (that is, more glottised), while future tense lenitions do the opposite.
Cool!

clawgrip wrote: 23 Mar 2018 16:23
pbastronaut wrote: 23 Mar 2018 00:06In case you're interested in the etymology here, I took "天", "ама̎ң" from the Kyrgyz "асман", "sky", and "語" "юу̎" from the Old Chinese /*ŋaʔ/.
I had just assumed Japanese, because 天 can be pronounced ama in Japanese.
I had the same thought at first, but given this language's geographic distance from Japan, and the fact that that's a kun'yomi reading (right?), I wasn't sure.
Vlürch wrote: 24 Mar 2018 16:29 Sogdian wouldn't really work for the name since the Sogdian word for sky is smān, unless you had some kind of a shift like /s/ > /h/ > /a/
Couldn't you just get an initial /a/ here through epenthesis?
All4Ɇn wrote: 25 Mar 2018 04:40 I'm not sure how to explain it exactly but something about your conlang has a really interesting vibe that I just can't quite put my finger on. Hope to see more!
[+1]
Birdlang
greek
greek
Posts: 631
Joined: 25 Dec 2014 20:17
Location: Virginia

Re: Amanghu (天語), or Chinese Altaic

Post by Birdlang »

Ælfwine wrote: 25 Mar 2018 03:56 Nice! I always wanted to do something like this, but with Japanese and Austronesian.
Same, with different ones though, here.
I did a Japanese Spanish hybrid, but it went too crazy and I couldn’t do the sound changes too well and it was throwing Japanese and Spanish words at random.
But I do want to make a Cantonese, Malay/Javanese, and Hmong like you want to do with Japanese and Austronesian. But I have a feeling that will be hard because I’d want to without tones. And some Romance-Germanic-Slavic mix too.
pbastronaut
rupestrian
rupestrian
Posts: 14
Joined: 17 Feb 2018 13:02

Re: Amanghu (天語), or Chinese Altaic

Post by pbastronaut »

So I've been very busy retroactively evolving Amanghu. I actually never spent a serious amount of time figuring out how Amanghu would have developed from Proto-Sinitic and Old Amanghu, so I found about 100 Proto-Sino-Tibetan roots and went to town on them. I worked out a pre-Chinese root for each one, and went from there, but I haven't gone too crazy applying this to my lexicon because 1) it wasn't fun to do in the first place, and 2) I don't think all that work will show in the finished work anyway. Nevertheless, here's what I worked out:
  • Old Amanghu, from around 800 AD. Old Amanghu is notable for having much more limited toolbox of vowels than either Old Chinese or Modern Amanghu (/i, u, a/), a whole lot of diphthongs and glides, reduplicated vowel codas following pre-aspirated consonants, and few affricates. Old Amanghu is not tonal, but is generally isolating.
  • Middle Amanghu, from around 1200 AD. As Old Amanghu gives way into Middle Amanghu, those vowel codas become vowel prefixes. Words like the Late Old Amanghu /ʂɻaʰma/ become the Early Middle Amanghu /aʈ͡ʂaʰm/. Consonant clusters simplify across the board, mostly becoming affricates -- including affricates that didn't survive into the modern language, such as /p͡f/. I've tried during this period to diversify the vowels quite a bit, but it's honestly a bit of an arbitrary decision in some cases. /o/ barely makes an appearance. Middle Amanghu still uses a lot of /j/ and /w/ glides.
  • Early Modern Amanghu: from about 1600 AD. I wanted EMA to form the basis of Amanghu Cyrillic, because I felt that the alphabet as it stands is way too phonetic to sound natural (which it should, as I based it on my modern phonology), but honestly, Early Modern Amanghu sounds a lot like Late Modern Amanghu, with the exception of lacking any voiced aspirated stops like /bʰ/. EMA erases all of those vowel prefixes and makes Amanghu almost entirely monosyllabic again. All the pre-aspirated stops transform the vowels in those words to creaky voiced vowels. The high tone evolves out of increasingly simplified onsets.
The truth is, I don't know enough about languages to evolve them correctly. Reading extensively on the evolution of Chinese didn't help me one jot. I just can't do it. So for the most part this is all now just background flavour. I'm probably going back to my original method of sourcing vocabulary: mashing up Old Chinese until it sounds good to me.

Still though, I did manage to formulate a classification system for hanzi which has been enormously helpful. Hanzi in Amanghu can be placed in one of six broad categories, as follows:
  1. Class I, pictograms and ideograms. These are simple pictograms and ideograms which usually share an identical meaning with their Chinese counterpart, such as 水 and 三. These would have been sourced from Chinese bronze inscriptions from about 800 BC.
  2. Class II,, ideograms. These are mostly for particles and pronouns, postpositions, conjunctions and auxiliary verbs. Class-II hanzi would have been introduced to Late Old Amanghu to represent words which did not have clear obvious counterparts in Chinese, and would have been selected for the value of their semantic components above all else. As such, they have little relationship in meaning or pronunciation to the Chinese words they were taken from. For example, the plural particle for people, 侮, contains both the radical 人, for people, and 每, "every; each", but it is unrelated to its Chinese meaning: 侮 "to insult", and its pronunciation.
  3. Class III. The most common class, class-III hanzi would have been sourced for Middle Amanghu and are mostly phono-semantic compounds whose original definitions have been preserved between languages. These would have replaced a huge number of now-obsolete class-II hanzi, such as 時 "time", which replaced the previously-used class-II character 志 /djɯ/.
  4. Class IV. These would have been sourced from Middle Chinese, and are related to class-III hanzi. While class-III hanzi were selected on the basis of their semantic value, class-IV hanzi were adopted for phonetic or visually-aesthetic qualities. For example, 天 /amáŋ/ replaced the visually complicated class-II 矕 /amáŋ/ because of its shared phonetic value and visual simplicity. Class-IV hanzi are the source of a great deal of folk etymology in Amanghu, where words have adopted new presumed-origins based on how they are written.
  5. Class V, compounds. These are any previously adopted characters which underwent simplification or differentiation as Middle Amanghu gave way to Early Modern Amanghu. I don't have an example to hand, but these would be instances where one or more components of a compound ideograph would be replaced, removed, or otherwise simplified; or changed to better distinguish them from other characters.
  6. Class VI. These are parts of compound words taken directly from modern Chinese languages with their definitions intact. These are usually modern inventions or concepts, such as 電影 /lḭnkaŋ/ "film".
--
shimobaatar wrote: 30 Mar 2018 16:30 What a neat idea! I assume you've looked at these languages?
Yep, it's one of the things that gave me the idea in the first place. I couldn't find many resources on them, but they were conceptually useful, at least.
No worries! Most conlangers aren't linguists.

Love the flag, by the way.
I'm very glad to hear that. I struggle sometimes to keep up with some people who are much more knowledgeable than me.

I'm glad you like the flag! I love making them.
I like the "voiced aspirates", which especially help to set Amanghu apart from other Sinitic languages, at least as far as I know. I also like the velar affricates. Like I said in the random phonology thread, and like others have apparently said here, I don't think those should be a problem. I also like the combination between tone and phonation. Is there any tone sandhi?
I've played around with things like a loss of the high tone in non-initial syllables in compound words. I'm not sure whether to make that a rule or not, but it does seem to be how I pronounce things when I read words out loud, probably because of my English-speaking biases.
Also, out of curiosity, is there a romanization?
Not yet. I'll write one up some point soon.
Are these falling or rising diphthongs?

Are all consonant-glide combinations possible? Similarly, are all glide-vowel combinations possible? Are aspirated stops allowed in the coda position?
I'll have to come back to you on that first point. Glides can't come after aspirated stops or affricates, but can come after any other kind of consonant or at the start of "null onset" syllables. The only syllable codas allowed are nasals and voiced, unaspirated stops.
User avatar
Vlürch
greek
greek
Posts: 452
Joined: 09 Mar 2016 21:19
Location: Finland
Contact:

Re: Amanghu (天語), or Chinese Altaic

Post by Vlürch »

Interesting, and I especially like the idea that there would be widespread folk etymologies. Will you use complicated and/or rare hanzi at all? Not saying you should/shouldn't or that it'd be better or worse either way, just asking.
clawgrip
MVP
MVP
Posts: 2257
Joined: 24 Jun 2012 07:33
Location: Tokyo

Re: Amanghu (天語), or Chinese Altaic

Post by clawgrip »

I have to say I like how the tenses are represented phonetically through sound changes, but graphically through the use of a separate character.

I did pretty much the same thing in one of my own languages (but for pluralization), but I quite like the way yours works.

Do you have a list of all the sound changes? Are there times when the sound change cannot apply?
Post Reply