Modern Khotanese -- "reviving" an extinct Eastern Iranian language (advice needed)

A forum for all topics related to constructed languages
Post Reply
dva_arla
cuneiform
cuneiform
Posts: 75
Joined: 25 Oct 2019 21:03
Location: Realm of Ideas

Modern Khotanese -- "reviving" an extinct Eastern Iranian language (advice needed)

Post by dva_arla » 30 Oct 2019 22:44

Modern Khotanese (hazirje zamo fano-pha) is a conlang that I have been working for some time on. It is meant to be a revival of the not-so-ancient Khotanese "Saka" language (hvatanau / hvanau), *here* becoming extinct during the 11th century. It had a substantial corpus, easing for research, and the pholonogical development from proto-Iranian to Old Khotanese then to Late Khotanese is unique among Iranian languages.

I have already worked out some parts of the morphology and begun on word-building (having finished up with the numbers and some basic words), however I have reached a deadlock in some aspects of the phonology, which is where I need your help.

Geography

Khotanese would be spoken mainly in "Khotan Country" (Fanocha): an autonomous province stretching from Guma/Pishan to Niya (you can find all of the places in Google Maps). There would be diaspora communities; based on *real* Khotanese texts it can be surmised that there were a substantial amount of Khotanese migrants in Shachu/Dunhuang (currently in Gansu Province) since the 9th century. Khotanese communities would also be found in Kashgar (possibly the capital of my alt-history), Yarkend (former capital), and Dzungaria.

Influences

Khotanese would of course be influenced by "Eastern Turki"/Chagatai/Uyghur, the language being spoken in areas surrounding Khotan Country. The influence would mostly take place in grammar and vocabulary, though phonology will see some quite subtle influence (vowel reduction etc.) Persian and Perso-Arabic loans would pervade the language, especially in technical terms.

It would also be a good idea to use Sarikoli, a Pamiri language used in the mountains to the west of East Turkestan/Xinjiang, as some sort of reference, although the language itself doesn't have any influence on Khotanese (the two being geographically separated). For Sarikoli, like Khotanese, is a *real* Eastern Iranian language still spoken today that is also influenced considerably by Turkic languages and Persian.

dva_arla
cuneiform
cuneiform
Posts: 75
Joined: 25 Oct 2019 21:03
Location: Realm of Ideas

Re: Modern Khotanese -- "reviving" an extinct Eastern Iranian language (advice needed)

Post by dva_arla » 30 Oct 2019 22:55

Phonology

Consonants of the latest stage of RTL Khotanese (left = standard transliteration of Khotanese, right = phonetic value)

p [p] or {b} ?*
ph [pʰ]
b [β], {b} before consonants
m [m]
v [w]

t [t]
th [tʰ]
d [ð], [d] before consonants
n [n]
rr [r]

ṭ [ʈʂ] or [ʈ]**
kṣ [ʈʂʰ] or [ʈʰ]
jṣ [ɖʐ]
ṇ [ɳ]
ṣ [ʂ]
ṣ' [ʐ]
r [ɹ]

tc [ts]
ts [tsʰ]
js [dz]
ñ [ɲ]
s {s}
ys [z]
l [l]

c [tʃ]
ch [tʃʰ]
j [dʒ]
ś [ʃ]
ś' [ʒ]
ḍ [ʎ]

k [k]
kh [kʰ]
g/gg [g]
y [j]

h: [χ]***
h: [ʁ]***
hv: [f]
[ʔ]****
h [h]
hv [hʷ]***


Notes:
* the phonetic value of p is still debated. Hitch suggested {b}, other linguists before him proposed [p]. Either value would cause a gap in the inventory of labials (/* p pʰ/ or /* pʰ b/, while /t tʰ d k kʰ g/)
** very marginal; occuring alone only in three or four words. Usually as allophone of /t/ following /ṣ/
***only in foreign loans
****as a result of consonant loss; may have been replaced by /w/ or /j/, or only appearing in hiatus, in Later RTL Khotanese

Its vowels were of the usual five-vowel system /a e i o u/, with ə and several other allophones in unstressed position. The length distinction found in Old Khotanese had disappeared by the time.

Phonology of Sarikoli : https://www.omniglot.com/writing/sarikoli.htm
Phonology of Uyghur : https://en.wikipedia.org/wiki/Uyghur_phonology
(note : modern standard Uyghur is based on Urumchi. Here it would be wise to refer to the Uyghur dialect of Khotan instead, as the vowels and morphology, from what I read, are slightly different).

P.S. is there a way to create tables in this board? I tried entering CBB codes for the consonant inventory table, however in preview the code didn't tur into table, so that I am forced to go along with this utterly un-aesthetic list.

dva_arla
cuneiform
cuneiform
Posts: 75
Joined: 25 Oct 2019 21:03
Location: Realm of Ideas

Re: Modern Khotanese -- "reviving" an extinct Eastern Iranian language (advice needed)

Post by dva_arla » 31 Oct 2019 00:20

Phonological change (a.k.a) the part that troubles me currently

I am troubled with deciding certain sound changes, so please do give me your feedbacks.

Certain :

1. /hw/ -> /f/, /h/ -> /f/ _u

There are two sounds "begging" to be changed to /f/: the one stated above and /pʰ/. After experimenting with my lips, I found out that /hw/ tends to turn to /f/ earlier then /pʰ/ does.

2. The letter transcribed p will most likely become b in the modern language, for the simple reason that it would look good on comparison charts : compare Persian panj vs. Pashto pindzə vs. Mod. Khotanese badz(ə).[/s]

The letter transcribed p will sound as /p/ in the modern language. b will sound as /b/ (or /m/ in the environment of nasals).

3. Elision of final vowels in most clusters e.g. guśt 'meat' > *goś / guś


Uncertain / doubtful :

1. Merge /w/ and /v/ <- /β/?

I can't decide whether to go on with Uyghur or Sarikoli on this. Sarikoli (and other Pamiri language) differentiates /w/ and /v/ (the latter coming from proto- Iranic /β/), while Uyghur lacks phonemic distinction between the two (there w is usually fricativised before front vowels).

According to Hitch, Old Turkic b became /β/ at one time (probably around the 10th century) in all positions, including initial positions, but then in initial positions b reverted to /b/. It would go like this:
/b/ > /β/
/β/ > /m/ #_ in the environment of nasals
/β/ > /b/ #_

This explains why proto-Turkic b is vulnerable to changing to m in most Turkic languages (*ben > men "I" in allTurkic language except Anatolian Turkish). However, I have reserved /b/ for p, so I won't probably have /β/ > /b/.




2. Merge /ʒ/ and /dʒ/?

Again, Uyghur and Sarikoli differs on this one. Sarikoli differs between the two, Uyghur doesn't.

An argument for going with Uyghur on the two changes is that Khotan is situated more openly/urbanly, unlike Sarikoli whose speakers live in the Pamir mountains, thereby "shielding" them from Turkic influence in phonology, and bringing their phonology more in line to other Pamiri languages (Wakhi, Shughni, etc.) However, idk... Internal developments do play part in the development of phonology.

By the way, is the fortition /ʒ/ > /dʒ/ in all positions plausible/common?

3. /ʂ/ --> /χ/ (uvular) and
/ʐ/ --> /ʁ/?

Both Uyghur and Sarikoli doesn't have /ʂ/ and /ʐ/; the latter having turned them into the "softer" velar /x/ and /ɣ/. This also goes for foreign loanwords; according to the glossary I read Persian "Padishah" (king) is read in Sarikoli as /pa:dxa:(h)/

However, I have several issues with this sound change:
  1. *Real* Khotanese has a particle āṣṭa 'in the direction of; -wards" which would develop into the directional dative suffix (to) -oχ. I find /-oχ/ to be somewhat un-euphonic; "Klingon-ish".
  2. The copula of *real* Khotanese is ṣte (sing.) or ṣto (pl.). I don't want the language to have too much /χ/sounds, especially due to the copula having the sound.
  3. /ṣt/ on the onset is unlikely to turn into /χt/ (or can it?).
I shall list some more problems as they come, so do leave your feedbacks; they would most certainly be appreciated. Cheers!
Last edited by dva_arla on 11 Nov 2019 12:11, edited 2 times in total.

User avatar
samsam
hieroglyphic
hieroglyphic
Posts: 32
Joined: 18 Jul 2018 16:40

Re: Modern Khotanese -- "reviving" an extinct Eastern Iranian language (advice needed)

Post by samsam » 08 Nov 2019 16:46

I love the idea of a Modern Khotanese! Eastern Iranian language are immensely interesting, I'm also working on an(albeit fictional) Indo-Iranian language that would be spoken between the Ferghana Valley, Estearn Xinjiang and Northeastern Afghanistan. in the Oxus (Amudarya) and Jaxartes (Syrdarya) valley and Pamir.
However, I have several issues with this sound change:
*Real* Khotanese has a particle āṣṭa 'in the direction of; -wards" which would develop into the directional dative suffix (to) -oχ. I find to be somewhat un-euphonic; "Klingon-ish".
The copula of *real* Khotanese is ṣte (sing.) or ṣto (pl.). I don't want the language to have too much /χ/sounds, especially due to the copula having the sound.
/ṣt/ on the onset is unlikely to turn into /χt/ (or can it?).
My idea is that /χ/soften to h in codas and/or when difficult to pronounce; in result, āṣṭa→/oχ/ becomes /oh/ (and could possibly become /o:/) and ṣto→/χto/ becomes /hǝto/.

I'd be really grateful if you had a link to the ressources you used for Modern Khotanese !
Last edited by samsam on 11 Nov 2019 20:40, edited 1 time in total.

dva_arla
cuneiform
cuneiform
Posts: 75
Joined: 25 Oct 2019 21:03
Location: Realm of Ideas

Re: Modern Khotanese -- "reviving" an extinct Eastern Iranian language (advice needed)

Post by dva_arla » 11 Nov 2019 11:57

samsam wrote:
08 Nov 2019 16:46
I'd be really grateful if you had a link to the ressources you used for Modern Khotanese !
I can only name the titles. I have to dig all over the Web to find the books and journals; precise links are hard to list. Happy hunting! (should you need them)

Bibliography

A Guide to the Literature of Khotan by Emmerick. A catalog of Khotanese texts, grammars, dictionaries, and journals pertaining to Khotanese as of 1992.

Lexicon/Dictionary

The only dictionary I use is Prof. Bailey's Dictionary of Khotan Saka. The thing to be lamented about this dictionary is that 1) it does not provide index, making searching for a word a pain and 2) it is quite old, so that there is no pdf or doc of it with selectable text. It is however the completest of its kind.

The dictionary, moreover, does not separate Old (~6 to 8th cent.) and Late Khotanese (9 to 11th cent.) forms of a lemma. A substantial part of my efforts relies on collecting the latest possible forms of a lemma or grammatical form.

Grammar

Saka Grammatical Studies by Emmerick; a thick book focusing on verbal and substantival morphology. This book however wants treatment of pronouns, prefixes, suffixes, etc.
Jatakastava or "Praise of the Buddha's former Births" by Dresden; contains grammar of (not-so) Late Khotanese.
Chapter 7 Khotanese and Tumshuqese by Emmerick . In The Iranian Languages. Newest and most updated of any descriptions of the language, however only a summary.
and many others...

Corpus

Khotanese Texts I-VII by Bailey. As of now I am only able to access KT I-IV, and only IV comes with an English translation. Some of the texts in KT I to III, such as the Lyrical love Poem, the Khotanese report about Turks, are translated in different journals.

Saka Documents Text Volume III : The St. Petersburg Collections by Emmerick. Of course, if there is an SDTV III, there must be too SDTVs I and II, however these I haven't been able to get my hands on yet [:(].

and many others...

Considerations of Phonology (besides those given in the grammars)

The Vowel phonemes of Khotanese by Emmerick
The Consonant phonemes of Khotanese by Emmerick (I haven't been able to get my hands on this yet [:(] )
Late Khotanese Numbers from 1 to 36 in Tibetan by Maggi
A Chinese Text in Central Asian Brahmi Script by Emmerick and Pulleybank

This list isn't exhaustive, of course. I only pick the more "major" and influencing of the books, and I shall ammend it when I have the engergy to do so.

By the way, do you study in a university with access to journals? Or does your university has any materials pertaining to the Khotanese language? My university does not provide access to journal collectors (such as Brill), and some of the more interesting journals I cannot access because they are not scanned on the Internet yet (i.e. I'd have to travel ~3000 km away to access them).


Any help would be appreciated!

P.S. how do you create tables in this thread?
samsam wrote:
08 Nov 2019 16:46
I'm also working on an(albeit fictional) Indo-Iranian language that would be spoken between the Ferghana Valley, Estearn Xinjiang and Northeastern Afghanistan.
Anything remarkable about your conlang? (sound change, history, external influence, grammatical feature, etc.)

User avatar
Isfendil
greek
greek
Posts: 709
Joined: 19 Feb 2016 03:47

Re: Modern Khotanese -- "reviving" an extinct Eastern Iranian language (advice needed)

Post by Isfendil » 11 Nov 2019 16:51

I just want to say that that's a lovely bibliography you've assembled and its a shame some of the sources aren't as well constructed as they could be.

User avatar
samsam
hieroglyphic
hieroglyphic
Posts: 32
Joined: 18 Jul 2018 16:40

Re: Modern Khotanese -- "reviving" an extinct Eastern Iranian language (advice needed)

Post by samsam » 11 Nov 2019 20:53

dva_arla wrote:
11 Nov 2019 11:57
I can only name the titles. I have to dig all over the Web to find the books and journals; precise links are hard to list. Happy hunting! (should you need them)
Thank you ! That's for sure a lot of sources. It'll definitely help me a lot to comparate it with the grammar of my conlang, as the only langages with good documentation that I've used are Old Avesta and Vedic Sanskrit.
Anything remarkable about your conlang? (sound change, history, external influence, grammatical feature, etc.)
For now all I can say is that it's a member of the iranian language branch and that it surely will have borrowings from Sanskrit, Tocharian and other Iranian languages (even maybe from Middle Chinese), a lot of which will be calques.

User avatar
samsam
hieroglyphic
hieroglyphic
Posts: 32
Joined: 18 Jul 2018 16:40

Re: Modern Khotanese -- "reviving" an extinct Eastern Iranian language (advice needed)

Post by samsam » 12 Nov 2019 18:08

dva_arla wrote:
11 Nov 2019 11:57
By the way, do you study in a university with access to journals? Or does your university has any materials pertaining to the Khotanese language? My university does not provide access to journal collectors (such as Brill), and some of the more interesting journals I cannot access because they are not scanned on the Internet yet (i.e. I'd have to travel ~3000 km away to access them).
I still study in High School so you can imagine that there isn't the word "Khotanese" printed somewhere in the whole building, but my main teacher learned about my naturalistic and realistic conlang project for my "Travail de Maturité" (High-School diploma side work in Switzerland) that I'll have to do in 2 years and suggested that I go to the local University's Library

Zythros Jubi
sinic
sinic
Posts: 430
Joined: 24 Nov 2014 17:31

Re: Modern Khotanese -- "reviving" an extinct Eastern Iranian language (advice needed)

Post by Zythros Jubi » 21 Jan 2020 09:45

In this alternate timeline, what is the political scenario like in this language's area of distribution? I assume nothing much changes; perhaps one more autonomous region in China.

One thing I don't understand is the different usage between New Persian and Chagatai languages in Transoxiana. In Transoxiana the Tajik-speaking population had undergone Turkic language assimilation for centuries, but in Kokand and Bukhara courts Persian still has a high prestige (it's mostly limited to literature in Ottoman Empire, but I don't know the situation for Kokand and Bukhara), let alone there were and still are millions of Tajiks speakers in Transoxiana. On contrary, Iranian language speakers are few in China. Why is that?
Lostlang plans: Oghur Turkic, Gallaecian Celtic, Palaeo-Balkanic

dva_arla
cuneiform
cuneiform
Posts: 75
Joined: 25 Oct 2019 21:03
Location: Realm of Ideas

Re: Modern Khotanese -- "reviving" an extinct Eastern Iranian language (advice needed)

Post by dva_arla » 21 Jan 2020 16:44

Zythros Jubi wrote:
21 Jan 2020 09:45
In this alternate timeline, what is the political scenario like in this language's area of distribution? I assume nothing much changes; perhaps one more autonomous region in China.
I've been toying for a while with alt-history scenarios involving an independent East Turkestan, actually. With a revolution in the 1933 (there was a very inspiring graffito in RTL Khotan, written while the battle rages on, that said
Revolution is an edifice built of many bricks
Each brick is an injustice
Blood is Mortar
Each wall is a mountain of sorrow
The foundation is most important
Alone, it must sustain the structure
Martyrdom is the Excellent Foundation!
(hoping to translate it into Khotanese one day!) And perhaps a civil war between the Bughras and E. Qasim, a language/ethnic/religious strife, an expedition to Tibet, etc. But all of that history can surely wait later, after I've finished with the language? I don't want to shun certain people from aiding with this project due to assumed political disagreements...

The history of the Tarim Basin is somewhat tumultous and "interesting"... you have Yaqub Beg, the Bughras, the Khojas, Dungan cavalrymen, a band of Swedes who set up a printing press in Yarkend and tried to standardise "Eastern Turki" orthography, &c.
One thing I don't understand is the different usage between New Persian and Chagatai languages in Transoxiana. In Transoxiana the Tajik-speaking population had undergone Turkic language assimilation for centuries, but in Kokand and Bukhara courts Persian still has a high prestige (it's mostly limited to literature in Ottoman Empire, but I don't know the situation for Kokand and Bukhara), let alone there were and still are millions of Tajiks speakers in Transoxiana. On contrary, Iranian language speakers are few in China. Why is that?
Short answer: because the Samanids were based to the south of Transoxiana.

Somewhat speculative answer : the cities of the Tarim Basin had a relatively small size (in comparison to, say, Merv or Chang'an), making assimilation much faster.

Long answer : *takes a deep breath* the Samanids expanded to Central Asia and salvaged as well as elevated the Persian language for centuries to come, "stamping" out Sogdian (at least in Sogdiana) in the process. If you're wondering why Sogdian didn't get salvaged unlike its West Iranian counterpart, it was because of the fact that the Sogdians themselves failed to create any consolidated state; they are merchants who would gladly assimilate into any society they touch! Anyways, Persian had a high prestige not only in the two cities-- but everywhere in Central Asia, even in the Tarim Basin (Yaqub Beg knew the language, educated Turkis and perhaps Dungans were learning it)... until 1920 in Transoxiana & Ferghana and a bit earlier/later than that in the Tarim Basin. Stalin, due to his Georgian origins, had somewhat an enmity towards the Armenians, and by extension, Persians, which was why the Tajiks were allocated a small strip of mountaneous land, and their language in an official capacity limited thither. Source? The caption of this map: https://www.reddit.com/r/MapPorn/commen ... 5987x4192/
On contrary, Iranian language speakers are few in China.
Keep in mind that the Iranian language spoken by "Tajiks" in China isn't Persian. Shaw's "On the Ghalchah [Sarikoli] Language" (published in 1876) states that the literates of the Sarikolis used Persian as a written language, though.

Zythros Jubi
sinic
sinic
Posts: 430
Joined: 24 Nov 2014 17:31

Re: Modern Khotanese -- "reviving" an extinct Eastern Iranian language (advice needed)

Post by Zythros Jubi » 21 Jan 2020 19:03

Well, what religion do you think is the majority among the Khotanese, perhaps Ismaili like in RTL Pamir, or some syncretic religion like Druze and Yazidi?
Lostlang plans: Oghur Turkic, Gallaecian Celtic, Palaeo-Balkanic

dva_arla
cuneiform
cuneiform
Posts: 75
Joined: 25 Oct 2019 21:03
Location: Realm of Ideas

Re: Modern Khotanese -- "reviving" an extinct Eastern Iranian language (advice needed)

Post by dva_arla » 22 Jan 2020 00:34

Zythros Jubi wrote:
21 Jan 2020 19:03
Well, what religion do you think is the majority among the Khotanese, perhaps Ismaili like in RTL Pamir, or some syncretic religion like Druze and Yazidi?
Perhaps an even mixture of Muslims and Buddhists, with some of the "syncretics" you just mentioned.

dva_arla
cuneiform
cuneiform
Posts: 75
Joined: 25 Oct 2019 21:03
Location: Realm of Ideas

Re: Modern Khotanese -- "reviving" an extinct Eastern Iranian language (advice needed)

Post by dva_arla » 22 Jan 2020 13:58

Starting off with the numbers -- probably among the most semantically-stable of the lexicon.

1 şo
2 do
3 dre (dialectal dray / drey)
4 tso
5 badz
6 čha / čhe? (ʈʂʰa / ʈʂʰe)
7 ho / hod?
8 haʂ / aʂ / aχ
9 no
10 das


۱ شو
۲ دو
۳ دری
۴ څو
۵ بذ
۶ څه / څی
۷ هو / هود؟
۸ اښت / اخت / اخ
۹ نو
۱۰ دس


I'm still unsure of whether to implement the /ʂ/ > /χ/ or /x/ shift (and if implemented, to what extent / in which environments).

Khotanese has a the three way distinction /k kʰ g/ /p pʰ b/ /t tʰ d/. The first one will be represented as ک گ ݢ. ط ت would do for /t tʰ/. I want to put the madda of alif آ on top of پ to represent /pʰ/, but a stand-alone madda "subscript" cannot be found in unicode...

Zythros Jubi
sinic
sinic
Posts: 430
Joined: 24 Nov 2014 17:31

Re: Modern Khotanese -- "reviving" an extinct Eastern Iranian language (advice needed)

Post by Zythros Jubi » 22 Jan 2020 14:48

I suggest hod and hax for 7 and 8 respectively.

PS: please have a look at this post viewtopic.php?f=31&p=298157#p298157, I'm seeking for advice
Lostlang plans: Oghur Turkic, Gallaecian Celtic, Palaeo-Balkanic

Post Reply