Cissian Revisited

A forum for all topics related to constructed languages
Post Reply
User avatar
ixals
sinic
sinic
Posts: 446
Joined: 28 Jul 2015 18:43

Cissian Revisited

Post by ixals »

After reworking Cissian in private for the second or third time, I thought it would be time to leave the old thread and everything I thought was good back then behind and make a fresh start. So let me introduce you to:

Cissian Revisited #01: Introduction & Phonology

Cissian, the language of the Tisza, is a Slavic language spoken in the region east of the river Tisza in our reality's modern day Hungary and Romania. Cissian is classified as a West Slavic language, however close contact with speakers of Southern dialects early on until the present day led to a big influence in vocabulary and sound changes shared with languages of the South Slavic branch. Other big influences on the language have been Hungarian and Romanian. The country of Cissia is a fairly young nation, first emerging out of the Romanian occupied part of Hungary in World War I and therefore the standardised is quite young, making Cissia a dialect-rich country as well as a country with many regions speaking other languages like Hungarian, Romanian, German, Serbo-Croatian and Slovakian.

Phonology

A big part of Cissian's phonology plays palatalisation. Each consonant technically has a palatalised and an unpalatalised version except the phoneme /j/ which could however be analysed as a palatalised version of nothing, /∅ʲ/. The vowels of Cissian are a standard European five vowel system with an extra /ɘ/. These vowels except for the latter can occur after any consonant, palatalised or unpalatalised (for brevity from now an called soft and hard consonants). Because of the history of the Slavic vowel system in Cissian and the script, the vowels are divided into normal and palatalising vowels which makes for eleven vowel phonemes:

/a e i o u ɘ/ <а е и о у ъ>
/ʲa ʲe ʲi ʲo ʲu/ <я є і ё ю>

If no consonant precedes the palatalising vowels, the results are /ja je ji jo ju/. Cissian also allows diphthongs of every vowel and either /i̯/ or /u̯/. The first is written with the letter <й> whereas the latter can be written in two ways depending on the origin of the word. <ў> is used for words of non-native origin or native words in which a combination of a vowel and /u/ resulted in a diphthong ending in /u̯/. <в> is used in native words in which /u/ is an allophone of coda /ʋ/.

The consonant system can be summarised as having twenty-two consonant (or twenty-three counting /j/. If soft consonants are counted as seperate phonemes, this number increased to forty-five in the standard language, while its actually forty-four for every speaker and forty-three in many dialects. In the following table, /j/ is not listed as a consonant because it does not have its own letter in the Cissian alphabet and is written by using a palatalising vowel or <й> as said before. Cissian also features two consonants that can be syllabic, these are /l/ and /r/.

/p b t d k g/ <п б т д к ѓ>
/t͡s d͡z t͡ʃ d͡ʒ/ <ц s ч џ>
/f s z ʃ ʒ x ɦ/ <ф с з ш ж х г>
/m n ʋ l r/ <м н в л р>

In the spoiler, the whole consonant system with both soft and hard consonants can be found.

Spoiler:
/p pʲ b bʲ t tʲ d dʲ k kʲ g gʲ/
/t͡s t͡sʲ d͡z d͡zʲ t͡ʃ t͡ʃʲ d͡ʒ d͡ʒʲ/
/f fʲ s sʲ z zʲ ʃ ʃʲ ʒ ʒʲ x xʲ ɦ ɦʲ/
/m mʲ n nʲ ʋ ʋʲ l lʲ r rʲ/
There is an additional letter for marking palatalisation if there is no vowel following, which is <ь>.

Exact Pronunciations & Allophones

Of course, the phonemes above are not the most exact, so here is a list of all the exact pronunciations and the allophones of Cissian's consonants:
  • The dental consonants are truly dental, being [t̪ d̪ t̪͡s̪ t̪͡s̪ʲ d̪͡z̪ d̪͡z̪ʲ s̪ s̪ʲ z̪ z̪ʲ n̪]. Exception to this are the soft versions of the dental stops and /n/ which - due to their palatalisation - are palatal consonants instead, [c ɟ ɲ].
  • The dental lateral is not dental, but a velarised [ɫ]. In coda position, this velarisation is strong enough that the lateral vocalalises to [o̯] like in Serbo-Croatian. The soft version is [ʎ] in all positions.
  • The hard palato-alveolars are more accurately described as being laminal retroflex consonants [ʂ ʐ t͡ʂ d͡ʐ], similar to Slovak and Polish. The soft palato-alveolars are instead true palatals [ɕ ʑ t͡ɕ d͡ʑ].
  • Whereas the all the hard velars are actual velars, the soft velar fricative is strongly palatalised as well and is pronounced as [ç]. The voiced version of this, [ʝ], is the allophone of the soft glottal fricative /ɦ/.
  • All the other consonants, which are the labials and labiodentals, are simple labial or labiodental consonants. Their soft versions are either palatalised or dissolve into a combination of the consonant followed by /j/ depending on the speaker. The labio-dental approximant behaves the same, although the soft version is a fricative, not an approximant. So both [pʲ bʲ fʲ vʲ mʲ] or [pj bj fj vj mj] are possible pronunciations. The soft velars also do this, [kʲ gʲ] or [kj gj] are both allowed.
  • Just like /l/, /ʋ/ vocalises in coda position to [u̯], as has been previously mentioned.
  • Cissian's only trill is apical-alveolar [r̺ r̺ʲ] although intervocally it is most commonly pronounced as a simple tap instead. The trill is the only exception to Cissian's soft/hard pairing for every consonant. The soft trill does not exist in any of the dialects in Cissian but the standard language still counts it as a separate phoneme.
  • Syllabic /l/ is not velarised and also pronounced as a apical-alveolar [l̺]. Adjacent to other dentals, this pronunciation changes to being dental ([l̪]) as well.
  • Apart from /lʲ/ [ʎ] and /nʲ/ [ɲ], every consonant depalatalised in coda position. Palatalisation is sometimes written for etymological purposes like in вовьца /ʋoʋt͡sa/ "sheep" and мідь /mʲid/ "copper".
  • The nasals assimilate in some cases. The dental /n/ before velars is velar as well, being [ŋ], while both nasals merge into [ɱ] before the labiodentals.
  • A combination of a voiced consonant and a nasal of the same place of articulation, so /dn/ and /bm/, change into simple nasals, [n̪ m].
  • In the coda, /ɦ/ is deleted.
The vowels have multiple pronunciations as well, but depending on the dialect, there are more or fewer different pronunciations. The following list is standard language only.
  • /a/ is pronounced as [ä] in every position. There are no allophones for these two phonemes.
  • /ɘ/ does not have any allophones either, but it's pronunciation is different from speaker to speaker. /ɘ/ is chosen because it is the most common pronunciation, but pronunciations as /ɵ/, /ɤ/, /ɐ/, /ɨ/ and /ʌ/ exist as well. A lot of non-native speakers use the schwa for this vowel which is heavily looked down upon by native speakers. The schwa was also used in earlier times but not today.
  • Additionally, /u/ does not possess any allophones either and is [​u] everywhere.
  • The high front vowel /i/ is pronounced as [​i] in both stressed and unstressed syllables if preceded by a soft consonant. After hard consonants, it is also [​i] in open stressed syllables, but a more central [ɪ] in unstressed syllables and closed stressed syllables. This is a remnant of the merger of Proto-Slavic *i and *y, the latter changing from /ɨ/ to /ɪ/ before merging with *i. In syllables starting with /r rʲ/, the different pronunciations still exist even after the merger of these two consonants. So /rʲi ri/ in unstressed syllables are [ri rɪ].
  • Unstressed /e/ is generally [e̞] as well as in stressed positions in a multiple of occasions. Soft /e/ - or in other words: /e/ after palatalised consonants - is also [e̞] in stressed syllables. After hard consonants, the vowel is more open than usual and is [ɛ]. There is also again an exception to this rule: If the vowel is followed by a palatalised consonant, it is pronounced as [e̞] again, but closed syllables have the open vowel instead even if a soft consonant follows. A little overview to make everything clearer is found in a spoiler at the end of this list.
  • /o/ is pretty similar to /e/. The standard pronunciation of this phoneme is [ɔ] in unstressed syllables. Should a high vowel (/i/ or /u/) follow, the pronunciation of this vowel is heavily raised and is [ʊ] instead. A preceding soft consonant does not raise the pronunciation in unstressed positions though. In stressed syllables, the pronunciation is also [ɔ], however only after hard consonants. After soft consonants, it is a big higher and realised as [o̞]. Following high vowels also raise the vowel to [o̞], but again, a closed vowel retains the open pronunciation. So the distribution is the same as /e/, except that it is a following high vowel and not a following soft consonant that triggers the raising and /o/ has an additional unstressed pronunciation. Again, the overview is found in the spoiler below.
  • Finally, Cissian also has long vowels in stressed syllables. This will be explained further in a later post. However, what I can already say is, that the long vowels are pronounced as [äː e̞ː o̞ː iː uː]. Of course, /ɘ/'s long version depends on the pronunciation of short /ɘ/.
Spoiler:
/e/:
unstressed - [e̞]
stressed /Cʲe/ - [e̞]
stressed /Ce/ - [ɛ]
stressed /CeCʲ/ - [e̞]
stressed /CeCCʲ/ - [ɛ]
long - [e̞ː]

/o/:
unstressed - [ɔ]
unstressed /oCH/ - [ʊ]
stressed /Cʲo/ - [o̞]
stressed /Co/ - [ɔ]
stressed /CoCH/ - [o̞]
stressed /CoCCH/ - [ɔ]
long - [o̞ː]
Thanks for reading! [<3] :mrgreen:
Native: :deu:
Learning: :gbr:, :chn:, :tur:, :fra:

Zhér·dûn a tonal Germanic conlang

old stuff: Цiски | Noattȯč | Tungōnis Vīdīnōs
Ælfwine
roman
roman
Posts: 940
Joined: 21 Sep 2015 01:28
Location: New Jersey

Re: Cissian Revisited

Post by Ælfwine »

Definitely want to see more of this, Ixals!

You claim it's close to West/South but to me it appears very East Slavic.
My Blog

A-posteriori, alternative history nerd
shimobaatar
korean
korean
Posts: 10373
Joined: 12 Jul 2013 23:09
Location: UTC-04:00

Re: Cissian Revisited

Post by shimobaatar »

ixals wrote: 08 Sep 2018 19:55 The consonant system can be summarised as having twenty-two consonant (or twenty-three counting /j/. If soft consonants are counted as seperate phonemes, this number increased to forty-five in the standard language, while its actually forty-four for every speaker and forty-three in many dialects.
ixals wrote: 08 Sep 2018 19:55
  • Cissian's only trill is apical-alveolar [r̺ r̺ʲ] although intervocally it is most commonly pronounced as a simple tap instead. The trill is the only exception to Cissian's soft/hard pairing for every consonant. The soft trill does not exist in any of the dialects in Cissian but the standard language still counts it as a separate phoneme.
Could you elaborate further on this? Why is the soft trill included in the standard language if it isn't present in any dialect?

Oh, never mind, I reached this part below:
ixals wrote: 08 Sep 2018 19:55
  • The high front vowel /i/ is pronounced as [​i] in both stressed and unstressed syllables if preceded by a soft consonant. After hard consonants, it is also [​i] in open stressed syllables, but a more central [ɪ] in unstressed syllables and closed stressed syllables. This is a remnant of the merger of Proto-Slavic *i and *y, the latter changing from /ɨ/ to /ɪ/ before merging with *i. In syllables starting with /r rʲ/, the different pronunciations still exist even after the merger of these two consonants. So /rʲi ri/ in unstressed syllables are [ri rɪ].
Anway, if the soft trill is the consonant that can be subtracted to get the 44 consonants present in actual speech, which consonant can be subtracted to get the 43 present in some dialects?
ixals wrote: 08 Sep 2018 19:55
  • The dental lateral is not dental, but a velarised [ɫ]. In coda position, this velarisation is strong enough that the lateral vocalalises to [o̯] like in Serbo-Croatian. The soft version is [ʎ] in all positions.
[…]
  • Just like /l/, /ʋ/ vocalises in coda position to [u̯], as has been previously mentioned.
Oh, so [Vo̯] and [Vu̯] contrast?
ixals wrote: 08 Sep 2018 19:55
  • Syllabic /l/ is not velarised and also pronounced as a apical-alveolar [l̺]. Adjacent to other dentals, this pronunciation changes to being dental ([l̪]) as well.
It isn't even velarized when adjacent to velar consonants?
ixals wrote: 08 Sep 2018 19:55
  • The nasals assimilate in some cases. The dental /n/ before velars is velar as well, being [ŋ], while both nasals merge into [ɱ] before the labiodentals.
So, to clarify, /m/ remains [m] before velars? What about before coronals?
ixals wrote: 08 Sep 2018 19:55
  • /ɘ/ does not have any allophones either, but it's pronunciation is different from speaker to speaker. /ɘ/ is chosen because it is the most common pronunciation, but pronunciations as /ɵ/, /ɤ/, /ɐ/, /ɨ/ and /ʌ/ exist as well. A lot of non-native speakers use the schwa for this vowel which is heavily looked down upon by native speakers. The schwa was also used in earlier times but not today.
Is there any pattern to which speakers use which pronunciation?
ixals wrote: 08 Sep 2018 19:55
  • Finally, Cissian also has long vowels in stressed syllables. This will be explained further in a later post. However, what I can already say is, that the long vowels are pronounced as [äː e̞ː o̞ː iː uː]. Of course, /ɘ/'s long version depends on the pronunciation of short /ɘ/.
Does the language contrast between long and short vowels only in stressed syllables, or are all vowels allophonically lengthened in stressed syllables? I guess that will be answered in the aforementioned later post.

Looking forward to more!
User avatar
ixals
sinic
sinic
Posts: 446
Joined: 28 Jul 2015 18:43

Re: Cissian Revisited

Post by ixals »

Ælfwine wrote: 01 Nov 2018 08:03 Definitely want to see more of this, Ixals!

You claim it's close to West/South but to me it appears very East Slavic.
Thanks for reminding me that this conlang still exists [:O] I always try to get rid of the East Slavic as much as I can but it looks like I'll never achieve that because I really want to keep the distinction between hard and soft consonants. Hopefully the grammar and vocabulary will sway you [:P]
shimobaatar wrote: 03 Nov 2018 05:50 Could you elaborate further on this? Why is the soft trill included in the standard language if it isn't present in any dialect?
You already found what you were looking for but I want to add that when the language was standardised, the dialects used for the standard dialect were still having a palatalised rhotic but in the midst of depalatalising it. There is still a distinction between them as you have seen. However, common folk with no linguistic knowledge still quotes stuff like "each of Cissian's consonants can be either hard or soft" although it's not true anymore but they won't budge because "you can clearly see it's written so it does exist" [;)]
shimobaatar wrote: 03 Nov 2018 05:50 Anway, if the soft trill is the consonant that can be subtracted to get the 44 consonants present in actual speech, which consonant can be subtracted to get the 43 present in some dialects?
Due to influence of Hungarian and Romanian, many dialects joined these two languages in deleting /lʲ/ [ʎ]. It merged with /j/ almost everywhere. This leaves most dialects with a palatalisation distinction in every consonant except liquids which makes everything more regular again [:D]
shimobaatar wrote: 03 Nov 2018 05:50 Oh, so [Vo̯] and [Vu̯] contrast?
Yep, they do!
shimobaatar wrote: 03 Nov 2018 05:50 It isn't even velarized when adjacent to velar consonants?
No, it's not [:D] Syllabic /l/ is pretty resistant to velarisation apart from the dialects bordering Serbia which turned pretty much every instance of syllabic /l/ into a back vowel.
shimobaatar wrote: 03 Nov 2018 05:50 So, to clarify, /m/ remains [m] before velars? What about before coronals?
Coda nasals followed by a consonant aren't very common and often noun paradigms have an alternation of /NCV/ and /NVC/ so analogy keeps the original pronunciation pretty much in all cases. Of the two assimilations I explained in the post, the latter is mostly (I'd think) a loanword-only occurence. So there technically is only /n/ changing before velars and nothing more in Cissian.
shimobaatar wrote: 03 Nov 2018 05:50 Is there any pattern to which speakers use which pronunciation?
I haven't put a lot of thoughts into this as of yet. /ɨ/ definitely is a thing in the regions bordering Romania, for the other ones I am still not sure about. I could picture both /ɐ/ and /ʌ/ present in a single dialect but the exact pronunciation depends on class, gender or age. I'll be able to tell you more about this after I put more work in the dialects [:D]
shimobaatar wrote: 03 Nov 2018 05:50 Does the language contrast between long and short vowels only in stressed syllables, or are all vowels allophonically lengthened in stressed syllables? I guess that will be answered in the aforementioned later post.
Yes, the difference between vowel length only exists in stressed syllables so unstressed vowels are short apart from stressed ones, which can be either short or long. The post talking about this has been finished for a long time but I want to post the historical sound changes before that and I have been procrastinating hard on that :roll: [:D]
Native: :deu:
Learning: :gbr:, :chn:, :tur:, :fra:

Zhér·dûn a tonal Germanic conlang

old stuff: Цiски | Noattȯč | Tungōnis Vīdīnōs
Clio
sinic
sinic
Posts: 228
Joined: 27 Dec 2012 23:45

Re: Cissian Revisited

Post by Clio »

I really enjoy this so far! This is definitely one of the most detailed phonology posts I've seen; morsels like nonstandard non-native pronunciations make for a really nice, realistic read.

I was just about to ask if you had a Grand Master Plan to share; I'll be keen to see that post whenever you get around to it. Since I don't know much about the Slavic languages in general or in particular, it would be especially useful if you could make a note of what characteristics make Cissian West Slavic and what sound changes are shared with South Slavic.
Niûro nCora
Getic: longum Getico murmur in ore fuit
scratchpad
Ælfwine
roman
roman
Posts: 940
Joined: 21 Sep 2015 01:28
Location: New Jersey

Re: Cissian Revisited

Post by Ælfwine »

ixals wrote: 04 Nov 2018 16:07
Ælfwine wrote: 01 Nov 2018 08:03 Definitely want to see more of this, Ixals!

You claim it's close to West/South but to me it appears very East Slavic.
Thanks for reminding me that this conlang still exists [:O] I always try to get rid of the East Slavic as much as I can but it looks like I'll never achieve that because I really want to keep the distinction between hard and soft consonants. Hopefully the grammar and vocabulary will sway you [:P]
No problem!

My book, The Pannonian Slavic Dialect of the Common Slavic Proto-Language, insinuates that the likely distribution of the Common Slavic proto-language looked like this:

Code: Select all

	Czech		Slovak		Ukrainian
		Pannonian	Cissian?
	Slovene		Serbocroatian	Bulgarian

According to this theory, Czech is the language closest to Slovene, Slovak is the language closest to Serbocroatian, and Ukrainian is the language closest to Bulgarian. I believe there is some evidence to back this theory up: examine the northern Slovene dialects with West Slavic features such as the preservation of /tl dl/ and the epithensis of [v] before /u/, and Slovak which shares many similarities Serbocroatian (such as the /u/ reflex of the Common Slavic back nasal).

I suspect your language to be somewhere in between my Pannonian and the Slavic dialect spoken in between Bulgaria and Ukraine, or modern day Romania...at least before they were assimilated by the Romance and Hungarian speakers. So I wouldn't totally discount East Slavic influence if I were you.
My Blog

A-posteriori, alternative history nerd
Post Reply