Ȧbhannı

A forum for all topics related to constructed languages
vo1dwalk3r
cuneiform
cuneiform
Posts: 108
Joined: 17 Mar 2014 22:37
Location: dʰǵʰémi
Contact:

Ȧbhannı

Post by vo1dwalk3r »

LaTeX grammar: here!

Ȧbhannı is just an a-priori artlang, with aesthetic inspiration from mostly Basque and Japanese.

Phonological inventory (updated)

/m n̪/ ⟨m n
/t̪ ɟ̟ k/ ⟨t ȷ k
/β θ̱ ɕ ʑ x/ ⟨b z s ȷh h
/ɺ j/ ⟨r y

/i e ɔ a/ ⟨ı e o a
/í é ɔ́ á/ ⟨i ė ȯ ȧ
/eɪ̯ ɔʊ̯ ɔɪ̯ aɔ̯ aɪ̯/ ⟨eı ou oı ao aı
/éɪ̯ ɔ́ʊ̯ ɔ́ɪ̯ áɔ̯ áɪ̯/ ⟨ėı ȯu ȯı ȧo ȧı

Progressive vowel harmony exists between the two mid vowels, i.e. roots are either e-grade or o-grade.

old stuff below

Phonotactics

Maximal syllable structure is (C)V(C), with the following constraints
  • ȷh⟩ becomes ⟨r⟩ word-finally
  • ȷ⟩ becomes ⟨n⟩ word-finally
The following (intervocallic) clusters are valid:
  • mm⟩, ⟨nn⟩, ⟨tt⟩, ⟨kk⟩, ⟨zz⟩, ⟨ss⟩, ⟨hh⟩, ⟨rr
  • zm⟩, ⟨sm⟩, ⟨hm⟩, ⟨rm⟩, ⟨hn
  • nt⟩, ⟨zt⟩, ⟨st⟩, ⟨ht⟩, ⟨rt⟩, ⟨nk⟩, ⟨rk
  • rb⟩, ⟨nh⟩, ⟨th⟩, ⟨bh⟩, ⟨rh
  • kr⟩, ⟨br⟩, ⟨hr
Suffixes will place an epenthetic -a- after the root in order to prevent the formation of illegal clusters.

Prosody

Stress is dactylic in roots, that is, it follows a repeating stressed-unstressed-unstressed pattern. Only stressed vowels may take a high tone. However, normal-tone vowels in word-final syllables are always unstressed. Abhanȷı is roughly stress-timed.

Allophony

Ȧbhannı has fairly minimal allophony (suggestions welcome). This includes:
  • /a/ → [ɑ] in o-grade words
  • /k x n̪k/ → [k̟ x̟ n̪k̟] when adjacent to /i/
  • [k x n̪k] → [ḵ x̱ n̪ḵ] in o-grade words, when not adjacent to /i/ or a consonant adjacent to containing /i/
  • [n̪t̪ n̪k̟ n̪k n̪ḵ] → [n̪d̪ ŋ̟k̟ ŋg ŋ̱g̱]
  • Plosives are unreleased word-finally
Last edited by vo1dwalk3r on 24 Mar 2018 04:19, edited 10 times in total.
vo1dwalk3r
cuneiform
cuneiform
Posts: 108
Joined: 17 Mar 2014 22:37
Location: dʰǵʰémi
Contact:

Re: Abhanȷı

Post by vo1dwalk3r »

Check out the edit of my first post where I linked my WIP grammar typeset in LaTeX! I don't have too much else for this post other than a bit of typology. I have a lot planned out for verbal and nominal morphology and other grammatical features but they're very incomplete and I want to get down the basics before moving on. I'm curious to hear what you think!

Morphological Typology

Synthesis in Abhanȷı is tends to fall into three categories: verbal morphology, marking tense; nominal morphology, marking case and number; and extensive derivational morphology. These functions of synthesis are almost exclusively agglutinating and suffixing, apart from stem modification and suppletion being very common for the irregular formation of past tense verbs. Abhanȷı is also predominantly head-marking; for example, in the case of possession, the object possessed is marked.

Other grammatical functions are marked using isolating morphological processes.

Syntactic Typology

The clause order of Abhanȷı is a strict VSO except in negative clauses and questions, where it is SVO. Questions are marked only through word order, unless they ask a negative in which case a special negative particle is used. Abhanȷı also tends to be head-initial; for example, adpositions are prepositions and occasionally circumpositions, and modifiers (including relative clauses and auxiliary verbs) follow what they modify.

Morphosyntactic Alignment

Abhanȷı follows roughly a split-S active-stative alignment, marking subjects of volitional verbs and objects of non-volitional verbs in the active case and other nouns in the stative case, where each verb belongs to a fixed “volitional” or “non-volitional” class. However, it has similarities with a fluid-S alignment, as morphological processes can be used to transform a verb between cases. However, this process is often irregular and generally non-functional, so that younger words cannot generally undergo this process.
Iyionaku
mayan
mayan
Posts: 2102
Joined: 25 May 2014 14:17

Re: Abhanȷı

Post by Iyionaku »

Nice to see that you use LateX for your grammar. I do that too (although I lack motivation after having written two subsequent scientific works for university with it atm).

Actually, your grammar reads very well and way better than your post that I found very confusing at first sight. I recommend everyone who is interested in the language to directly read the LateX document.
Wipe the glass. This is the usual way to start, even in the days, day and night, only a happy one.
User avatar
Frislander
mayan
mayan
Posts: 2088
Joined: 14 May 2016 18:47
Location: The North

Re: Abhanȷı

Post by Frislander »

I do really rather like the look of that phonology, though the voicing contrast in the palatal fricatives only strikes me as a little weird. I can't wait for more of the grammar!
vo1dwalk3r
cuneiform
cuneiform
Posts: 108
Joined: 17 Mar 2014 22:37
Location: dʰǵʰémi
Contact:

Re: Abhanȷı

Post by vo1dwalk3r »

Thanks guys!
Iyionaku wrote:Nice to see that you use LateX for your grammar. I do that too (although I lack motivation after having written two subsequent scientific works for university with it atm).
I finally convinced myself to start using it now that I'm in an "actual" math course and realized it would be cool to use for conlanging. I'm really enjoying it so far, the results are so aesthetically pleasing.
Frislander wrote:I do really rather like the look of that phonology, though the voicing contrast in the palatal fricatives only strikes me as a little weird.
Yeah the phonology's a wierd one. There are a lot of quirks to it that I can come up with historical explanations for, but the palatals are giving me a hard time. If only it had some living relatives I could make a more informed reconstruction...
vo1dwalk3r
cuneiform
cuneiform
Posts: 108
Joined: 17 Mar 2014 22:37
Location: dʰǵʰémi
Contact:

Re: Ȧbhannı

Post by vo1dwalk3r »

I've been working a lot but I tend to work very non-linearly, so it's hard to organize it into separate posts. I've uploaded my current LaTeX grammar and dictionary here. I also renamed it to Ȧbhannı because I realized I don't allow the ⟨⟩ cluster, haha. I imagine you guys would like to see some text—here's the opening of the Odyssey.

Mıȯuzko mȧȷharra erekı e, om ȧr e ȧnnır
[miˈɔ̂ʊ̯θ̱.ḵɔ ˈmá.ʑɐr.ra ˈɛ.ɺe.k̟i e, om áɺ e án̪.n̪iɺ]
muse-VOC sing from.within 1sg, and with 1sg tell.story
Sing in me, Muse, and through me tell the story

rah ȧrkon a sinıhratı mȯrakoztı inok,
[ɺɐx ˈáɺ.ḵɔn̪ a ˈɕí.n̪ix̟.ɺɐ.t̪i ˈmɔ́.ɺɐ.ḵɔθ̱.t̪i ˈí.n̪ɔḵ̚]
man with-be-PST REL.sg skill-ACT-POSS-pl method-all-POSS-pl fighting,
of that man skilled in all ways of contending,

tanarahȧıbasan a, ėzehamrahamı roȷoran barı a,
[ˈt̪a.n̪ɐ.ɺɐˈxáɪ̯.βɐ.ɕɐn̪ a, ˈɛ́.θ̱e.xɐmˈɺa.xɐ.mi ˈɺɔ.ɟ̟ɔ.ɺɐn̪ ˈβa.ɺi a,]
wander-er-be-PST REL.sg, for.year-many-REDUP-PL request-make-PST 3-ACT-pl REL.sg,
the wanderer, harried for years on end,

ȯrı ėkken bar nȧntar kı ȧztennaket
[ˈɔ́.ɺi ˈék.kɛn̪ βɐɺ ˈn̪án̪.d̪ɐɺ k̟i ˈáθ̱.t̪ɛn̪.n̪ɐ.kɛt̪̚]
after take-PST 3-ACT wealth out.from strong-place
after he plundered the stonghold

hın a zabokotarat Tarȯı.
[x̟in̪ a ˈθ̱a.βɔ.ḵɔˈt̪a.ɺɐt̪ t̪ɐˈɺɔ́ɪ̯]
be.on-PST REL.sg high-place-ACT-POSS Troy
on the proud height of Troy.

Here's another text, a short poem I wrote and translated, which I set a song to for my composition class. It's not an example of "in-universe" poetry, with respect to my unmade conworld, but just an experiment.

ke ȧıno saratı e,
ke ȧıno saratı e

zaratar e ommıratı
ere zin akamı hȧtousırı—
bȯu issat ȯ tan
emık e zabor—

ȯ ȷano zarat er,
ȯ tan zȧn asorat e


Curious what you guys think. Thanks for reading!
User avatar
DesEsseintes
mongolian
mongolian
Posts: 4331
Joined: 31 Mar 2013 13:16

Re: Ȧbhannı

Post by DesEsseintes »

Your grammar doc makes me want to learn LaTeX.

The language looks pleasant. You make use of pretty much all my favourite features, namely overdots on vowels, ı, and gemination.
Last edited by DesEsseintes on 29 Dec 2017 07:35, edited 1 time in total.
vo1dwalk3r
cuneiform
cuneiform
Posts: 108
Joined: 17 Mar 2014 22:37
Location: dʰǵʰémi
Contact:

Re: Ȧbhannı

Post by vo1dwalk3r »

Beautiful, right? It can be tricky sometimes though.
User avatar
Thrice Xandvii
runic
runic
Posts: 2698
Joined: 25 Nov 2012 10:13
Location: Carnassus

Re: Ȧbhannı

Post by Thrice Xandvii »

(You formatted your link in the OP incorrectly. In the opening [ it needs to be url= and then the URL, not just the URL.)
Image
User avatar
Frislander
mayan
mayan
Posts: 2088
Joined: 14 May 2016 18:47
Location: The North

Re: Ȧbhannı

Post by Frislander »

Interesting grammar, looks good, I especially like the use of auxilliaries for aspect (very Tibeto-Burman) and the the different varieties of adjectives.

I'm slightly confused by the vocative case though, which may be a feature of the lack of examples, but what does the difference between "positive" and "negative" vocatives mean?
vo1dwalk3r
cuneiform
cuneiform
Posts: 108
Joined: 17 Mar 2014 22:37
Location: dʰǵʰémi
Contact:

Re: Ȧbhannı

Post by vo1dwalk3r »

Ah, right. The vocative has a normal (i.e. Latin-esque) use, but when the subject of a clause is put in the vocative case and placed before the verb, this forms an imperative mood clause. Now, instead of negating the verb in this case, you use the negative vocative ending.
User avatar
Frislander
mayan
mayan
Posts: 2088
Joined: 14 May 2016 18:47
Location: The North

Re: Ȧbhannı

Post by Frislander »

vo1dwalk3r wrote: 30 Dec 2017 17:45 Ah, right. The vocative has a normal (i.e. Latin-esque) use, but when the subject of a clause is put in the vocative case and placed before the verb, this forms an imperative mood clause. Now, instead of negating the verb in this case, you use the negative vocative ending.
Oh my that is an interesting construction, and a good way to get more use out of the vocative (which seems to be a rather weird case that doesn't seem to do anything else in natlangs). This would be good to put in your syntax section. I don't think it's attested in natlangs at all but I like it all the same (though languages having separate forms for the prohibitive vis-à-vis the normal negative is extremely common).
vo1dwalk3r
cuneiform
cuneiform
Posts: 108
Joined: 17 Mar 2014 22:37
Location: dʰǵʰémi
Contact:

Re: Ȧbhannı

Post by vo1dwalk3r »

Okay, I think I've gotten far enough on Ȧbhannı that it's time to start over. :D

But to be serious, this is a problem of mine. If I work too long on a conlang, particularly if I'm going for naturalism, it starts to feel "topheavy" and frustrating and unmotivating. So I want to start over (not entirely from scratch of course), this time with Ȧbhannı's structure in mind and try to give it a more solid linguistic base. I may go through this process multiple times, but I think eventually I'll get it in a state that I like.

This time though I want to focus a lot more on the worldbuilding aspect of it, since Ȧbhannı is set in a conworld. So, for reference, here's a map of the region, with the Ȧbhannı region roughly in the blurry red part. The bottom of this image is the equator and the top is roughly the 75º; that puts the Ȧbhannı region at approximately the latitude of Georgia/the Carolinas, the northernmost land around mid-south Scandinavia, and the whole continent (what can be seen) maybe a bit smaller than South America. This map is fairly tentative, but should be a decent idea of the geography of the region.

Now, Ȧbhannı is actually not itself a language isolate as I previously stated, but a member of the Abhannic language family along with another, smaller language spoken a bit north (and some dialects of the two). These split off maybe 1000 or so years before modern Ȧbhannı as described above. The first major stage of the Abhannic languages comes probably a few centuries before, with the emergence of a complex society by speakers of "Old Ȧbhannı," or Hȧṕuxańńı, an isolate. A harsher climate lead to the shrinking of this civilization and the split of the northern group. The northern language remains fairly conservative, while Ȧbhannı underwent some major phonological developments during this time. Around 500 years BP, Ȧbhannı was in its middle stage and pretty close to its modern stage, and Ȧbhannı society had begun to recover and spread again.

The following is a tentative sketch of Old Ȧbhannı phonology. It's tricky working backwards, especially because Ȧbhannı phonology is so wierd and I wanted to explain both the palatal series and the vowel harmony which were hard to do together.

/mʲ mˠ nʲ nˠ/ ⟨ḿ m ń n⟩
/pʲ pˠ tʲ tˠ kʲ k ʔ/ ⟨ṕ p t́ t ḱ k ʔ⟩
/sʲ sˠ xʲ x h/ ⟨ś s x́ x h⟩
/rʲ rˠ j ɰ/ ⟨ŕ r ȷ w⟩

/i ɯ ǝ a/ ⟨ı u ǝ a⟩
/əi əɯ ai/ ⟨ǝı ǝu aı⟩

Syllables are CV(R), where R is a nasal, rhotic or approximant. Stress is exclusively root-initial, where there may be a pitch accent (high tone, marked as overdot). /h/ and /ʔ/ are found only (almost only?) word-initially.

Old Ȧbhannı had a topic marker wǝ/jǝ, which was always unstressed and preceded the noun; which one was used was based on noun class, which was either animate or inanimate (roughly).

More soon, hopefully...
vo1dwalk3r
cuneiform
cuneiform
Posts: 108
Joined: 17 Mar 2014 22:37
Location: dʰǵʰémi
Contact:

Re: Ȧbhannı

Post by vo1dwalk3r »

Ok ok, I've finally figured out Old Ȧbhannı (now Haʔpuhannı) phonology and the diachronics to modern Ȧbhannı:

/mʲ mˠ nʲ nˠ/ ⟨ḿ m ń n
/pʲ pˠ kʲ k ʔ/ ⟨ṕ p ḱ k ʔ
/sʲ sˠ h/ ⟨ś s h
/ɾʲ ɾˠ j w/ ⟨ŕ r w y

/i ɯ ǝ a/ ⟨ı u ǝ a
/ǝɪ̯ aɪ̯/ ⟨əı aı

Stress is exclusively word-initial.

Glides occur exclusively word-initially.

Syllable structure is CV(N), where N can be a nasal or glottal consonant. Valid clusters are homorganic nasal+plosive, and glottal+consonant (except ʔ). In addition, all consonants except ʔ, j and w can be geminated.

Coincidentally, the consonant inventory is identical to Abau's, plus the palatalized/velarized contrast and the glottal stop.

The diachronics are kinda messy, and still subject to change, so here they are in action. The words and are animacy-based topic markers which were dropped but effected vowel harmony:

/wǝ jaɪ̯ʔnˠa/ wǝ yaıʔna → /áɪ̯na/ ȧına 'rain'
/hɯmˠa/ huma → /xama/ hama 'to eat'
/wǝ hiɾˠǝ/ wǝ hırǝ → /xiɺɔ/ hıro 'sky'
/wǝ yihsʲǝ/ wǝ yıhśǝ → /íɕ.ɕa/ issa 'breath'
/yǝ kʲǝsʲiɾˠɯ/ wǝ ḱǝśıru → /keɕiɺ/ kesır 'storm'
/wǝ yǝʔhanˠ/ wǝ yǝʔhan → /ɔ́.xan/ ȯhan 'snow'
/ɾapˠǝɪ̯/ rapǝı → /ɺaβɔʊ̯/ rabou 'to fall'
/pʲǝʔkaɪ̯/ ṕǝʔkaı → /tákaɪ̯/ tȧkaı 'to run'
/pʲaɾˠapˠɯ/ ṕarapu → /taraβ/ tarab 'clear'
/wǝ sˠapˠǝ/ wǝ sapu → /θaβɔ/ zabo 'wind'
/yǝ sˠɯhɯʔ/ yǝ suhuʔ → /θax/ zah 'horizon'
/sˠaʔnʲɯ/ saʔńu → /θán/ zȧn 'to flee'
/sˠaɾˠapʲɯ/ saraṕu → /θaɺat/ zarat 'to see'

I'm really glad I have a decent outline of Ȧbhannı diachronics because now I can make a much more historically informed (and hopefully self-consistent) grammar.
vo1dwalk3r
cuneiform
cuneiform
Posts: 108
Joined: 17 Mar 2014 22:37
Location: dʰǵʰémi
Contact:

Re: Ȧbhannı

Post by vo1dwalk3r »

Agh, I'm way to wishywashy about this kind of stuff. I've decided not to develop Ȧbhannı has an conlang for a conworld but just as a personal language. Maybe once I'm more confident in my skills I'll try my hand at "true" diachronic conlanging. Regardless, my reconstruction of Old Ȧbhannı is still going to inform my construction of Ȧbhannı, but I won't focus very hard on it.

Here's my new LaTeX grammar, featuring an attractive sans-serif font and more details and examples than previously. I would greatly appreciate any feedback or criticism you have about formatting/writing/organization/etc.

Anyways, this post is going to be about a very important topic for Ȧbhannı, its morphosyntactic alignment, which I've overhauled somewhat.

Morphosyntactic Alignment

Morphosyntactic alignment is expressed through four "core" cases for which verbal arguments are marked: stative, active, dative, and vocative. The following will not explore the vocative case, which is outlined in the linked grammar.

The case used for an argument is dependent on whether the verb is of the volitional or non-volitional class. Since these classes are fixed, i.e. a verb cannot change its class, Ȧbhannı could be classified as a split-S active-stative language. However, this is not entirely accurate, as will be seen below, since the volitionality class also applies to transitive verbs (active-stative languages are also called split-intransitive, since the S argument, the subject of an intransitive clause, can take different markings).

Active and Stative

The stative case is the default case, taking a null ending. It is generally associated with objects of volitional verbs and subjects of non-volitional verbs. In contrast, the active case is associated with subjects of volitional verbs and objects of non-volitional verbs. Moreover, in contrast to the dative case, the stative case analyzes verbs as being merely events or actions, rather than experiences. The following examples demonstrate the stative case (henceforth unmarked) and the active case in their uses as verbal arguments:

hama ȧhıro-n
eat cat-ACT
‘the cat is eating’

mikkar bırama
rock boat
‘the boat is rocking’

ohta-n rah-ar okon-nı
cut_down-PST man-ACT tree-PL
‘the man cut down the trees’

ėhaȷ-an enık sabah-ar
wash_away-PST water dust-ACT
‘the water washed away the dust’

Note that in the fourth example, the agent is marked stative, because the verb is non-volitional.

Dative

The dative has two main uses in Ȧbhannı. The first as an indirect object, and the second is as a replacement for the stative case. Both of these uses function as verbal arguments. The use as an indirect object is seen on trivalent verbs, which take active and stative arguments as usual. Many occurences of the dative as an indirect object are a result of valence-increasing operation. The dative argument is classically associated with an allative meaning (‘to’), but can take others depending on the verb, especially given different valence-increasing operations. The following example shows its use as the direct object of a “true” (i.e. underived) trivalent verb:

amȷat ȷȧo abır berrak-ı-m
sent_to food 3pl.ANIM.ACT village-PL-DAT
‘they have sent food to the villages’

Dative-substitution

However, the dative case can also replace the stative (but not active) argument of a monovalent or divalent verb. This usage focuses on the experience impacted upon the stative argument, particularly the emotional state caused. For example:

tezka-n e ar
hit-PST 1sg 3sg.ANIM.ACT
‘he hit me’ (a retelling of events, I don’t care anymore, etc.)

tezka-n em ar
hit-PST 1sg.DAT 3sg.ANIM.ACT
‘he hit me’ (it hurt, I cried, he should be punished, etc.)

Note that (c) is emotionally charged, effectively blaming the agent (the hitter), while (b) takes a (relatively) neutral stance on the occurence. The following examples show this dative-substitution used with a monovalent verb:

hȯzo-n a
fall-PST 3sg.ANIM
‘she fell’

hȯzo-n am
fall-PST 3sg.ANIM.DAT
‘she fell’

As before, the second example conveys emotional meaning. However, since there is no agent and the subject is not the first person, the use of the dative conveys empathy toward the subject rather than emphasizing the experience. In a case where the subject is not an experiencer, the dative substitution can be used to express empathy towards those effected by the situation.

Here, with the dative-substitution, we see another way in which Ȧbhannı differs from a standard split-S ACT-STAT language, in that the dative can freely replace the stative case regardless of the verb in question; this bears resemblance to a fluid-S ACT-STAT alignment. However, as already seen, it again differs in that the argument split occurs with transitive verbs as well.
vo1dwalk3r
cuneiform
cuneiform
Posts: 108
Joined: 17 Mar 2014 22:37
Location: dʰǵʰémi
Contact:

Re: Ȧbhannı

Post by vo1dwalk3r »

Okay, this has taken a while, but I'm gonna talk about the last two cases (skipping the vocative for now I guess) which I call the "peripheral" cases (since they don't mark arguments, unlike the "core" cases, and are suffixed after those cases).

Possessive

The possessive case differs from the other cases in that it is head-marking, that is, it is placed on the head noun of a genitive phrase. Because of this, nouns marked with the possessive case will often also be marked with one of the core cases.

Possession

The main use of the possessive case is to mark ownership or possession and similar relationships. Possession which is deemed alienable is marked with -z after the genitive ending. Each noun is inherently either alienable or inalienable and is always (un)marked as such when possessed, although they are only marked when the possessive case is used in this sense (semantically).

kouma-bıto-kke-z izzan-ı
egg-DIM-POSS-AL snake-PL
‘the snakes' eggs’

The following shows the possessive used with a core case:

tare tenı-kke-z Yȧra ar
hold gift-(PAT)-POSS-AL Yara 3sg.HUM.AGT
‘she has Yara's gift’

on sȧmar-ra-kko-ka em
talk brother-AGT-POSS-2sg 1sg.DAT
‘your brother was talking with me’

Parts of a Whole

A closely related function of the possessive is to represent constituents of a whole which are in some way distinguished from the rest. As stated above, this does not use the alienable ending.

hȧıra-kke-ı
arm-POSS-1sg
‘my arm’

hatem-akke ȧyı
table-POSS ground
‘the flat part of the ground’

Location

The final basic function of the possessive is as a locative case, conveying that the possessed noun is at the location of the possessor. This usage is only used to convey temporary locative states, i.e. those which are not inherent to the possessive-marked noun.

Moreover, the use of the possessive as a locative has a very general meaning. This contrasts with the locative preposition za which conveys the default, most contextually salient relation with the noun (this seems to be roughly analogous to English dropping the determiners of nouns in prepositional phrases, i.e. ‘he's in the court’ vs. ‘he's in court’). The use of za also draws greater attention to the spatial relation between the nouns. For example:a-kke hatasakkıma
3sg.HUM-POSS school
‘he's at the school’ (conveying only locational information)

a za hatasakkıma
3sg LOC school
‘he's at school’ (implying he is occupied with school-related activities)

Partitive

The partitive case, unlike the possessive, is dependent-marking. The partitive is marked by the suffix -tat, which can be reduced to -tt- if preceding a vowel-initial determiner suffix.

Subsets

The stereotypical use of the partitive is to mark subsets of a group of entities or subsets of a mass noun which are relatively homogenous with respect to the whole. Determiners which mark the size of a subset may be suffixed to the partitive ending; in these cases, the noun it describes acts as the head of the genitive clause, and can take a core case.

ha ti aı-tt-os
there.MED stand 3sg.HUM(.PAT)-PTV-none
‘none of them were there’

imat amrah ishar-atat
dry.up much(-PAT) blood-PTV
‘much of the blood has dried’ (yes I shamelessly stole ishar from PIE *h₁esh₂r̥)

Note the similarity between the previous example and ‘the flat part of the ground’ example; both distinguish some portion of a mass noun from the rest of it. The difference between these two is that, in the former, the head of the genitive phrase, ‘some’, does not make this distinction, while in latter it does, with ‘flat part’.

Prepositional

The partitive has a number of closely related prepositional functions (which, of course, may only modify nouns). These all have to do with expressing some intrinsic property of the genitive phrase head.

The first of these is as an “intrinsic” locative, in which the location begin expressed somehow expresses some inalienable quality of the referent. For example:

zabo rabke-tat
wind north-PTV
‘the north wind’

The second use is as an ablative. Specifically, this is used for describing past locations.

ėmenkı haz-tat
cup there.DIST-PTV
‘the cup (which came/you brought/etc.) from over there’

A very closely related use is for place of origin, which can often overlap semantically with the above two functions.

mıyo-u ahme-tat
face-PL ocean-PTV
‘the people of the ocean’

Note that many of the above functions can also be done by compounding, i.e. zaborabko ‘north wind’ and mıyohmou ‘ocean people’.
vo1dwalk3r
cuneiform
cuneiform
Posts: 108
Joined: 17 Mar 2014 22:37
Location: dʰǵʰémi
Contact:

Re: Ȧbhannı

Post by vo1dwalk3r »

Blech, ok I'm starting over again. Well, I'm keeping most of what I did since the first restart but overhauling it quite a bit. Not so much a restart as simply reorienting myself.

But first I'd like some advice from you guys on how most efficiently to make my grammar. The way I've been doing it since the first restart is trying to very methodically and thoroughly go through describing the grammar, but this was kind of frustrating and not super fun. It just seemed like way too much work until I'd have something useful, I guess. The first time through I started with a very rough grammar sketch and then started translating sentences, but this didn't work out so well either as the grammar felt a bit patched-together and topheavy, if that makes sense, since I didn't have a solid base. Just curious what works for you guys in this respect.

Anyways, I'm going back to the idea of Ȧbhannı being an in-world conlang, but I'm going to be going about it differently. First, I'm still going to focus on making it as a personal language and save the conworld-related conlinguistics for later. Also, Ȧbhannı is now going to be a proto-language of my conworld. Old Ȧbhannı, as described above, is still its precursor but now buried deep in the past.

In-world, an early form of Ȧbhannı underwent influence by a substrate language with the following phonology:

/m n ŋ/
/p b t d k g ʔ/
/w l j ɦ/

/i iː í íː u uː ú úː/
/e eː é éː o oː ó óː/
/a aː á áː/

It had a (C₁)(C₂)V(C₃) syllable structure, where C₁ was a non-approximant, C₂ was a non-glottal approximant, and C₃ was a glottal. Alveolar-/l/ sequences were disallowed. Vowel hiatus was common.

This language contributed some vocabular into that early form of Ȧbhannı but then went extinct; however, a dialect or possibly closely related language in a different region did not undergo contact with Ȧbhannı, surviving to become a different proto-language or language family by the time of modern Ȧbhannı.

Anyways just a quick post. Any advice on my situation would be appreciated. Thanks for reading [:)]
vo1dwalk3r
cuneiform
cuneiform
Posts: 108
Joined: 17 Mar 2014 22:37
Location: dʰǵʰémi
Contact:

Ábhanni

Post by vo1dwalk3r »

Well, it's been a while. I haven't done a whole lot of conlanging since I was last here but I have some motivation to get back into it. I have a lot of material on Ábhanni (plus a little bit of orthography change lol) but most of it is scattered across many different revisions that didn't get very far. So I sort of want to collect together all the stuff that I like into one coherent conlang which will hopefully get developed a lot further than any of my previous attempts.

I'll talk here a bit about verbs, which are significantly different than anything I presented above.



There are two types of verbs: active verbs and stative verbs. They have different morphology and syntax and so will be described separately. In order to describe the function of tense and aspect in Ábhanni verbs, we will use the following definition: the temporal focus of a verb is the point of time in which the information that is meant to be conveyed takes place. For example, with a perfective verb such as 'I have eaten', the temporal focus is on the present even though the action of eating takes place in the past.

We now begin by describing active verbs.

Active verbs

Active verbs are, put vaguely, those which have some sort of internal structure. These can be divided into two categories, events and processes. Events are occurrences which are presented as instantaneous and occupying a single location in time (so perhaps "internal structure was not the best terminology), while processes are dynamic actions or states presented as having duration and time-varying internal structure. Unconjugated active verbs by default represent events, but can be transformed into processes by aspectual endings.

Morphology

In this section we briefly describe the basic tense and aspect markers that an active verb may take (when I develop the verbal system further there will likely be more that will go here). The verb can take up to three endings; the first (tense) must be the future tense ending, the second (perfectivity) either the perfective or aperfective ending, and the third (telicity) either the progressive, telic, or atelic ending. Of course any of the endings can be absent, but at most one from each group can occur on any given verb.

Future tense (FUT): -n, -na,1 -nn-2

Perfective aspect (PFV): -riyá, -niyá3
Aperfective aspect (APFV): -riyóh, -niyóh3

Progressive aspect (PROG): -si
Telic aspect (TEL): -ti
Atelic aspect (ATEL): -ttame

1 This allomorph is used for verbs whose unconjugated form ends in -n.
2 This is used when the future tense allomorph -n is followed by a (possible epenthetic) vowel (either as part of an ending or clitic); for example, -FUT-PROG becomes -nn-asi (the a here is epenthetic), or -FUT=1sg becomes -nn=e.
3 These allomorphs are used following the future tense allomorph -n.

Tense

Ábhanni verbs are marked for two tenses, nonfuture (NFUT) (the default unmarked tense) and the future. With events, these simply mark verbs which take place either in the past or the future. (Here, the present is excluded since events, which are viewed in their entirety and as instantaneous, generally do not coincide perfectly with the present moment.) With processes, the use of the tense marker places the temporal focus either in the future or the nonfuture. For example, a process might extend into the future but be marked by the nonfuture tense if the temporal focus is in the past.

The following shows tense marking on the verb háza, '(to) fall':

háza-∅
fall-NFUT
'it fell'

háza-n
fall-FUT
'it will fall'

Notice the lack of a possible reading 'it falls' in either of the above.

Perfectivity

The perfectivity ending has two forms, perfective and aperfective. The perfective aspect places the temporal focus after the event or process, and emphasizes the resulting state of the focus of the phrase. The aperfective does the opposite of the perfective, placing the temporal focus before the event.

haza-riyá
fall-PFV
'it has/had fallen'

haza-n-niyá
fall-FUT-PFV
'it will have fallen'

haza-riyóh
fall-APFV
'it had/has not yet fallen (but will, or did, eventually fall)'

haza-n-niyóh
fall-FUT-APFV
'it will not yet have fallen (but will eventually fall)'

We see that nonfuture event verbs are compatible with (in some sense) a present tense reading when a perfectivity ending is present, since the temporal focus is no longer on the event and so can be in the present.

Telicity

The telicity ending has three forms, progressive, telic, and atelic. These all interpret the verb as a process and differ in their treatment of the endpoint of the event. The telic aspect marks the process as completed; atelic marks it as not being completed; and progressive is used if the endpoint is ambiguous or irrelevant. Without a telicity ending, event verbs may be either completed or uncompleted.

háza-si
fall-PROG
'it is/was falling' (may or may not fall/have fallen completely)

háza-ti
fall-TEL
'it fell down'

háza-ttame
fall-ATEL
'it was falling' or 'it stopped falling' (but didn't fall completely)

Note that, like event verbs but unlike the progressive aspect, the telic and atelic aspects are incompatible with a present tense reading, since the endpoint (which is in temporal focus) of the process is treated as an event.

Stative verbs

Stative verbs, in contrast with active verbs, represent constants, sort of like an adjective, which are stretched over a period of time but do not have internal structure. Each stative verb is obligatorially followed by a stative tense particle, and most stative verbs are actually adjectives or nouns which are accompanied by such a particle. These stative tense particles are the following, which immediately follow the stative verb:

Past tense (PST): áka
Present tense (PRS): í
Future tense (FUT): áo

This particles can often be translated as 'is', although it does not act as a verb in the sense seen above, and often has very different meanings. For example, it can be used to describe states which are canonically or metaphorically associated to the stative verb (generally noun in this case), as the following demostrates:

rósmi í
fire PRS
'it's on fire'

íshar í=i
blood PRS=1sg
'i'm alive'

As with active tenses, the stative tense describes when the temporal focus is. Gnomic statements and other statements which lack a well-defined temporal focus are used with the present tense marker.

Important stative verbs are áat 'stand' and áappo 'sit', which are used to translate the phrases 'there is/are/were/etc.'; the former is used for living creatures which are standing, objects oriented vertically or which are upright, while the latter is used for objects/beings which are laying, sitting, or otherwise horizontally oriented. For abstract situations, áat is the default, although áappo is used for emotional states. For example:

áat áka n⟩aji⟨ík ha
stand PST ⟨COLL⟩person FOC
'there was a group of people (standing there)'

The stative verbs áat and áappo may be omitted, in which case the orientation of the referent is ambiguous; this is the only case in which the stative tense particles can appear without a corresponding stative verb. Also, these verbs may take a dative argument, which is the location. The following examples shows both of these:

(áappo) í mintá zi ka em
(sit) PRS love for 2sg 1sg.DAT
'I love you' (lit. 'I have love for you' or 'love for you sits inside me')



That's it for now. Thoughts are welcome, thanks for reading!
User avatar
Omzinesý
mongolian
mongolian
Posts: 4079
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

Re: Ȧbhannı

Post by Omzinesý »

I'm sorry. I'm again fucking with terms.

I think that what you call "perfective" is actually "perfect". In linguistics they are two different things, although in Romance languages they are expressed with the same form. Perfect can also be called "retrospectative" if you want to emphasize that it is an aspect, not a tense. Usually it's something in between.
I think its complemental antonym should be non-perfect.

Telicity is rather a semantic feature with which you describe aspects. I would call them perfective, imperfective, and progressive aspects.
My meta-thread: viewtopic.php?f=6&t=5760
vo1dwalk3r
cuneiform
cuneiform
Posts: 108
Joined: 17 Mar 2014 22:37
Location: dʰǵʰémi
Contact:

Re: Ȧbhannı

Post by vo1dwalk3r »

Omzinesý wrote: 04 Jan 2020 12:03 I'm sorry. I'm again fucking with terms.

I think that what you call "perfective" is actually "perfect". In linguistics they are two different things, although in Romance languages they are expressed with the same form. Perfect can also be called "retrospectative" if you want to emphasize that it is an aspect, not a tense. Usually it's something in between.
I think its complemental antonym should be non-perfect.

Telicity is rather a semantic feature with which you describe aspects. I would call them perfective, imperfective, and progressive aspects.
Thanks for the advice. I'll definitely change it around to be more standard. It also seems that the event/process distinct is somewhat similar to the imperfective/perfective distinction of Ukrainian (and probably Russian), especially in how the tenses are divided up (particularly, the lack of a true "present" for the perfective). I was also inspired by Ukrainian to add a /ɦ/, so now we have /x ɦ/ ⟨ḫ h⟩.

Anyways, I wanted to make a quick post to sketch some of the ideas that I've been having recently and keep myself motivated. Since I want to try to do some historical development of Ábḫanni, right now I just have glosses, but hopefully soon I'll figure out the phonological development.

So the first thing is that I want to emphasize the fundamentality of the active/stative distinction. So much so in fact that I'll change my terminology (to something probaby a bit nonstandard): what I previously called "active verbs" I will now call "actives", and what I previously called "stative verbs", "adjectives", and "nouns" will all fall under the category of "statives". The first reason is that statives move fluidly between functioning as stative verbs, adjectives, and nouns, and in fact there really won't be much distinction at all between adjectives and stative verbs. Second, the way active verbs function from stative verbs syntactically and morphologically is so different that I think it makes sense to not call them both "verbs".

A bit of historical background: the actives are the old "verbs" of Ábḫanni, and their morphology is also very ancient, so it will end up with a lot of irregularity and complexity (something which I have yet to really develop). The statives derive from non-verbs of pre-Ábḫanni, and their use as verbs (i.e. with the "stative tense particle" and all that) is much newer, so that their syntax and morphology is much simpler and more analytic than that of the actives.

The nominal morphology corresponding to the actives (i.e., the morphology used for the arguments of actives) is also very old (as is the number morphology), about as old as the morphology of the actives. This is the ergative-absolutive distinction seen above. The "peripheral cases" are maybe a bit less old but still an old development.

However, the nominal morphology in "stative clauses" is again much newer. In pre-Ábẖanni, "stative clauses" weren't really a thing, and so the "arguments" of statives didn't really have their own morphology (perhaps it would be more accurate to say that "statives" didn't really have arguments in a verbal sense). But in the development to modern Ábẖanni, statives began taking on verbal roles, and so the Ábḫanni speakers innovated a nominative-accusative system in stative clauses. The accusative case, by virtue of being much newer than the ergative, works differently, and is maybe closer to a clitic construction than a case. (I have yet to figure out how exactly the accusative is derived, which will of course inform the exact details of its morphology, but this is just a sketch of how it is used in the modern language).

Here are some idiosyncracies of the accusative. It generally looks as follows:

hang PRS painting wall-ACC
'the painting hangs on the wall'

In the main clause, the order of the arguments is variable, so also "hang PRS wall-ACC painting" is valid.

One case that works differently is if the accusative argument is a cliticized pronoun. In this case, the nominative argument takes a special ending (historically, this probably means that the cliticization of pronouns precedes the development of the accusative, which was somehow incompatible or preferably not used with the clitics):

see PRS=1sg cat-NOM
'the cat sees me'

If one of the arguments of a stative clause is itself a member of a subclause, it must directly follow the head of its subclause:

see PRS=1sg [hang PRS wall-ACC painting]
'i see the wall on which the painting hangs'

In the above case, "wall" is the accusative argument of both the main clause and its subclause. I'm not sure about this yet, but I think that if these differ, then the head of the subclause should take some marking, as follows:

see PRS=1sg [hang PRS-ACC painting wall-ACC]
'i see the painting hanging on the wall'

The accusative marking on the subclause head will probably be different than the accusative marking on nouns (I'm thinking maybe there would be special tense particles for this). However, subclauses cannot have an active head; in this case, a relative clause-like would be necessary. I have ideas for how this will work but I'll leave it for later.
Post Reply