Page 1 of 1

Old Arassenian

Posted: 14 Mar 2015 23:22
by Fanael
I guess there's no harm in describing the language I used in CBBCR VI. It gives me an excuse to actually write down the stuff instead of keeping it solely in my brain.

The original reason for making Old Arassenian was to have a protolang to derive daughterlangs from. For an unknown reason I spent more time working on it than on any other of my languages, so I suppose it's fair to say it's become my main conlang. That doesn't mean it's any good, I am a terrible conlanger.

The language presented here is Old Arassenian version 5000, because I rebooted it so many times it's funny how unfunny it is. The old versions were almost always kitchensinky. This time, I actually have an idea.

Without further ado, let's start with, as is this board's tradition,


Phoneme inventory

It's slightly more European that I normally do, so if you're allergic to anything resembling Eurolangs, I'd recommend averting your gaze now.

Code: Select all

|   Consonants   | labial | coronal | palatal | velar | pharyngeal |
| nasal          |   m    |    n    |         ŋ       |            |
| vl stop        |        |    t    |    c    |   k   |            |
| vd stop        |        |    d    |        ɟ~ɡ      |            |
| vl fricative   |   f    |    θ    |    ç    |   x   |            |
| vd fricative   |   v    |    ð    |        ʝ~ɣ      |            |
| vl sibilant    |        |    s    |         |       |            |
| vd sibilant    |        |    z    |         |       |            |
| lateral liquid |        |    l    |         |       |     lˤ     |
| central liquid |        |    r    |         |       |     ʔ      |

|  Vowels  |   front    |     back    |
| high     |  i  |  iː  |   u  |  uː  |
| high-mid |  e  |  eː  |   o  |  oː  |
| low-mid  |  ɛ  |  ɛː  |      |      |
| low      |  a  |  aː  |   ɑ  |  ɑː  |
Yes, it's fricative heavy. Yes, it has /θ ð/. Yes, there's no /p b/ despite labials being fine otherwise. Yes, that's a glottal stop pretending to be a pharyngeal approximant. All these are easily explained if we look at the parent language's stops and glides:

Code: Select all

|          | labial | alveolar | dorsal | guttural |
| ejective |        |    tʼ    |   kʼ   |     qʼ   |
| vl stop  |   p    |    t     |   k    |     q    |
| vd stop  |   b    |    d     |   ɡ    |          |
| glide    |   w    |          |   j    |     ʕ    |
Those familiar with diachronics probably already see what happened. All pulmonic stops became fricatives (except in some environments, such as after nasals), leaving the ejectives as the only stops. This explains lotta fricatives and no /p b/, but raises another question: where does voicing on stops originate from? If you guessed "glides", you're right. Voicing effect was most consitent and prominent on /j/, though, which also fronted vowels, resulting in standalone voiced stops being most common near front vowels.

And that glottal stop? It's a remnant of /ʕ/. In fact, what happened was /ʕ/ → /ʔ/. The language is somewhat inconsistent with regard to its treatment of /ʔ/, it can't decide whether it's a glottal stop or a pharyngeal approximant.

Syllable structure

Onset: (N/L)C. /ʔ/ is not a valid liquid here. Null onset is strictly speaking not allowed unless the nucleus starts with a semivowel, but nobody will look askance at you if you consider sole /ʔ/ a de facto null onset. In fact, that's what I'm doing.
Nucleus: either:
  • short vowels – (W)V(W) – where W is /i/ or /u/ and allowed only when V is not /i/ nor /u/ itself; each vowel is moraic, except the semivowel closer to the onset when it's null; or
  • a long vowel – Vː – always two moras
Coda: any consonant word-finally, any non-stop otherwise. May be null. Is moraic.

No syllable may have more than three moras. Should a four-mora syllable arise, its semivowel closer to the coda is dropped.

When syllabifying, medial clusters are split between syllables even if they form a valid onset. For example, /anta/ is syllabified /an.ta/, not /a.nta/.

Phonological rules

Basically, allophony, but includes some stuff that's not allophony, so I'm not calling it allophony, because it's not only allophony. Hope it's clear.
  • Palatal stops and fricatives are commonly realized as alveolopalatal affricates and sibilants.
  • /k x/ → [q χ] or [ʡ ħ] before /e eː i iː/. Yes, it looks weird, but look at the parent lang table above. They were uvulars that fronted. This is where they didn't front to avoid a merger.
  • /ŋ ɟ~ɡ ʝ~ɣ/ → [ɲ ɟ ʝ] before palatals and /e eː i iː/.
  • /ŋ ɟ~ɡ ʝ~ɣ/ → [ŋ ɡ ɣ] before velars, pharyngeals back vowels.
  • Elsewhere, /ŋ/ is [ŋ], but both /ɟ~ɡ/ and /ʝ~ɣ/ are in free variation.
  • Fricatives become stops after a nasal.
  • Liquids assimilate to a following liquid. /ʔ/ is not a liquid here.
  • /l r/ are contrastive only before /a aː/, back vowels and word-finally; become /r/ before /ʔ/; become /l/ elsewhere.
  • Word-medial stops spirantize before a consonant.
  • Obstruents assimilate in voice to a following obstruent.
  • Nasals and non-sibilant obstruents assimilate in PoA to a following nasal or non-sibilant obstruent.
  • /θ ð/ → /s z/ when near /s z/.
  • /ð/ becomes /l/ near a /l/.
  • Short vowels /i e ɛ a ɑ o u/ become [ɪ ɪ ə̟ ɐ̟ ʌ ɵ̠ ʊ] in closed syllables.
  • Vowels become nasalized after nasals and /ʔ lˤ/.
  • Vowels assimilate in nasalization to a preceding vowel.
  • Two adjacent vowels of the same position become a single long vowel; /ʔ/ is inserted between any other two adjacent syllabic vowels.
  • Tripled consonants become doubled. This obviously concerns only nasals and true liquids.

In an effort to create a nice rhythmic stress patterns that are something more than "every second syllable is stressed", I fear I might've overcomplicated things. Oh well. At least it's regular.

Stress assignment is done in four steps:
  1. Drop prefixes.
  2. Assign stress to (3k + 1)th mora. That is, to the 1st, 4th, 7th, etc.
  3. Remove stress from the lighter syllable if there are two consecutive stressed syllables.
  4. Remove stress from the syllable following another stressed syllable, starting from the beginning of the word, if there still are two consecutive stressed syllables.
The first stressed syllable is where the primary stress falls.


It's useful to be able to write the language in something else than IPA.

Code: Select all

|   Consonants   | labial | coronal | palatal | velar | pharyngeal |
| nasal          |   m    |    n    |         ñ       |            |
| vl stop        |   p    |    t    |    c    |   k   |            |
| vd stop        |   b    |    d    |         g       |            |
| vl fricative   |   f    |    th   |    ch   |   h   |            |
| vd fricative   |   v    |    dh   |         j       |            |
| vl sibilant    |        |    s    |         |       |            |
| vd sibilant    |        |    z    |         |       |            |
| lateral liquid |        |    l    |         |       |     ł      |
| central liquid |        |    r    |         |       |     x      |

|  Vowels  |   front   |   back    |
| high     |  i  |  í  |  u  |  ú  |
| high-mid |  ì  |  î  |  o  |  ó  |
| low-mid  |  e  |  é  |     |     |
| low      |  a  |  á  |  à  |  â  |
Glottal stop can be omitted if its presence can be inferred.

That would be it for now. Feedback welcome!

Re: Old Arassenian

Posted: 15 Mar 2015 00:00
by Creyeditor
I like what you have so far, I'm glad you included some allophony, stress rules and phonotactics.
Is there a reason for the front vs back distinction in open vowels? And where did /lˤ/ come from?
I would also be interested in the exact POA of the coronal consonants.

Re: Old Arassenian

Posted: 15 Mar 2015 01:20
by Fanael
Creyeditor wrote:Is there a reason for the front vs back distinction in open vowels?
It's not a pure front vs back distinction, and it's another example of */ʕ/ being funny. The parent language had three vowels, */a ə ɨ/ and relied on glides for frontness. /a/ is lone */a/, while /ɑ/ is */a/ + */ʕ/. Thanks to that, /ɑ/ is so low and back it probably should be written [ɑˤ], which is distinct enough from [a]. For some speakers it lost all pharyngeal traits and is closer to [ʌ], which again is quite distinct from [a].
Creyeditor wrote:And where did /lˤ/ come from?
From */ʕl/ mostly. The fact that some neighboring languages had a /ɫ/ of some kind stabilized /lˤ/ while other pharyngealized sonorants either shifted quickly after forming or did not form in the first place.
Creyeditor wrote:I would also be interested in the exact POA of the coronal consonants.
/n t d θ ð/ are dental, /s z l r/ are alveolar. I never really cared about dental vs alveolar much to be frank.

Re: Old Arassenian

Posted: 15 Mar 2015 01:34
by Creyeditor
Those are really interesting and inspiring diachronic explanations [:)]
I just wondered about the main difference between sibilants and dental fricatives in your language. I though it might just be the sibilant vs. non-sibilant distinction and not really one of dental vs. alveolar. But maybe using both of them is the best solution.