Ayakadiya

Substantial postings about constructed languages and constructed worlds in general. Good place to mention your own or evaluate someone else's. Put quick questions in C&C Quickies instead.
Post Reply
sanskacharu
Sanci
Sanci
Posts: 29
Joined: Mon Feb 13, 2017 8:15 am

Ayakadiya

Post by sanskacharu »

Herein is contained a compilation thread for Ayakadiya. This language is not technically spoken by humans, but the speakers inhabit bodies humanoid enough that there are no relevant differences to be noted. General phonetic capabilities for speech are identical -- general musical capabilities, on the other hand, are not.

The speakers of the language call themselves aya, which can technically be translated 'person', but it's slightly more accurate to translate it as 'aya', once one has an understanding of the language's gender system. Non-aya species don't count as aya, despite any level of sentience (if present). The aya themselves are best thought of as a spiritual sort of creature, a soul if you will. They are capable of manufacturing flesh-bodies (and bodies made of other substances) in order to inhabit them, and by means of so doing gain a physical presence. They are effectively the gods of their world, having created it out of base matter via song. Their world's physics and magic are identical, being governed by a complex elemental-magic-esque system which manifests itself as music. Each aya has one, most likely multiple, songs (that is, specific elements) associated with their person. These songs compose their metabolism. They can get more by training themselves to hear and deeply understand the new song, usually helped by a strong similarity between the one they seek to learn and one they already have. (For instance, going from fog to smoke is easy -- but 'smoke' doesn't necessarily include 'smoke from a fire'.) They can also lose them, usually only by extreme physical (to the soul, but physical enough for them) trauma. Silence and discordant music are anathema to them.

All told, the deep connection they have with physics/magic effectively makes every aya a potent mage, or a potent reality-warper, depending on which term you want to use.

Last edited by sanskacharu on Tue Feb 14, 2017 9:05 pm, edited 5 times in total.
My conlang family:
  • Ukumusi & Mupuasa -- Two peas in a pod. Tired of your nonsense.
  • Ku Ṣili -- Lonely Misfit. Can't make up its mind.
  • Ayakadiya -- Standoffish, self-important. Needs More Lexicon.

sanskacharu
Sanci
Sanci
Posts: 29
Joined: Mon Feb 13, 2017 8:15 am

Basic Typology

Post by sanskacharu »

I'm leaving the phoneme inventory and the phonology until I figure out how to make those pretty pretty tables I see in the phonology thread. Anyway.

Basic Typology
Ayakadiya is an agglutinative language, with a heavy focus on verbal morphology. A number of the example sentences I have been translating have been coming up glossed something like 'plain.noun - plain.noun - morphologically.buried.verb', which leads me to make this generalization. Depending on your definition of polysynthesis, and how relaxed you are with it, then Ayakadiya may or may not count as both agglutinative and partly polysynthetic. Take the name of the language itself:

aya-ka-di-ya.
person-all-speak-3.INAN
'All people speak it.'

Here we have an incorporated agent. Note well that aya 'person' is the ONLY noun which can be so incorporated, although it can be so as either an agent or a patient.

aya-ka-we-duu.
person-all-1.SG-rule
'I rule all people.' 'I rule everyone.'

(You have a question about how patient/agent is determined, because you can't tell from these examples? Good. Yes. Of course you do. We'll get to that.)

Other than aya incorporation, the only polysynthesis test that Ayakadiya meets is "can express full sentences as one word", and even then, it can meet this criterion only by using a sentence which involves only pronominal agents and patients, both of which are marked on the verb, and possibly an aspect marker.

we-reṭ-ya-re.
1.SG-eat-3.INAN-PERF
'I ate it.'
(tense unspecified)

I tend to shy away from calling it polysynthetic for this reason. I don't feel that the justification is sufficient -- certainly if this were a real language I was describing in the field, I wouldn't claim in my eventual paper(s)/dissertation/reference grammar that it was polysynthetic. I think my advisor would eat me for suggesting it.
Last edited by sanskacharu on Tue Feb 14, 2017 2:29 pm, edited 1 time in total.
My conlang family:
  • Ukumusi & Mupuasa -- Two peas in a pod. Tired of your nonsense.
  • Ku Ṣili -- Lonely Misfit. Can't make up its mind.
  • Ayakadiya -- Standoffish, self-important. Needs More Lexicon.

sanskacharu
Sanci
Sanci
Posts: 29
Joined: Mon Feb 13, 2017 8:15 am

Overview of Language Features

Post by sanskacharu »

Overview of Features
Roughly speaking, Ayakadiya includes the following features. I will explain these in a truncated, sketch-like manner.

Agentivity Hierarchy
The key to comprehending agent/patient assignments in any sentence. Agent/patient is the most saliant binary by which Ayakadiya determines its case roles, such as they are, so I continue to use it. Effectively, this means a decision on whether or not to use an inverse marker on the verb to redirect the action in the unexpected direction. The hierarchy goes like so:

1st person > 2nd person, animate > 3rd person, animate > 2nd person, inanimate-but-alive > 3rd person, inanimate-but-alive > 3rd person, inanimate-and-unalive

If 1st person and 3rd animate are both marked, it is assumed that 'I' am the agent, and the third person the patient. To say that I am the patient and the third person the agent who acts upon me, the inverse suffix -cu must be used.

Gender System
Helping the agentivity hierarchy do its job is the gender system. Nouns are either animate, or inanimate. This is largely semantically predictable, and is not explicitly marked. Animate nouns include roughly the following:
  • aya 'person'
  • derivational nouns based on aya
  • nouns which denote roles people take on, such as 'leader', 'parent', 'crafter', 'soldier'. An exception: 'slave' is inanimate.
  • nouns denoting musical terminology, such as 'note', 'melody', 'harmony', 'crescendo'
Everything else is considered inanimate. Note well that even a sentient species which is not an aya will be inanimate -- this is why there are second-person inanimate pronouns, which I'm aware are highly unusual. Two such species exist, the hyaay and the siin. They were created by aya to be pets and slaves, respectively, and the Ayakadiyan species names also mean 'pet' and 'slave' respectively.

Possessive System
There are two means of describing possession. One, the alienable, is most common, and is used as a default. Inalienable possession, the second system, is used in a limited number of situations:
  • parts of your own body or soul
  • your biological children (or your adopted children, if you wish to deeply express your love for them)
  • things you have made, during the making of which you have poured significant time, energy, and emotional investment into them, such that they have left a mark on your psyche. Think an artist's masterwork. It would be appropriate for me to describe any of my conlangs with the inalienable possessor, as well as my thesis, although the two evoke very different reactions.
In other situations, alienable possession is used.

Some Basic Word Orders
  • N Det
  • N Adj
  • N Num
  • Poss N
  • Rel N
  • Obl N
  • V Adv
  • O S V
'OSV' is better described as either 'patient-agent-verb' (PAV), or as "highest animacy constituent goes closest to verb". Because the higher animacy is assumed to be the agent, OSV is the most likely word order. Another argument for OSV word order:

duu taḳḳa a-mel-le.
Lord child 3.AN-strike-PERF
'The child struck the Lord.'

To say the Lord struck the child, the two nominal constituents must be reversed. taḳḳa duu amelle. You'll also note that the patient is not marked on amelle. When both the agent and patient would require the same pronoun agreement on the verb, only the subject's agreement is marked, and the subject is confined closest to the verb.

That's all the interesting features I can come up with off the top of my head. It's very sparse and I'm sure I'll remember something I left out tomorrow, but I need sleep for now.
Last edited by sanskacharu on Tue Feb 14, 2017 2:33 pm, edited 1 time in total.
My conlang family:
  • Ukumusi & Mupuasa -- Two peas in a pod. Tired of your nonsense.
  • Ku Ṣili -- Lonely Misfit. Can't make up its mind.
  • Ayakadiya -- Standoffish, self-important. Needs More Lexicon.

User avatar
Frislander
Avisaru
Avisaru
Posts: 836
Joined: Mon Feb 29, 2016 6:34 am
Location: The North

Re: Ayakadiya

Post by Frislander »

sanskacharu wrote:The speakers of the language call themselves aya, which can technically be translated 'person', but it's slightly more accurate to translate it as 'aya', once one has an understanding of the language's gender system. Non-aya species don't count as aya, despite any level of sentience (if present). The aya themselves are best thought of as a spiritual sort of creature, a soul if you will. They are capable of manufacturing flesh-bodies (and bodies made of other substances) in order to inhabit them, and by means of so doing gain a physical presence. They are effectively the gods of their world, having created it out of base matter via song. Their world's physics and magic are identical, being governed by a complex elemental-magic-esque system which manifests itself as music. Each aya has one, most likely multiple, songs (that is, specific elements) associated with their person. These songs compose their metabolism. They can get more by training themselves to hear and deeply understand the new song, usually helped by a strong similarity between the one they seek to learn and one they already have. (For instance, going from fog to smoke is easy -- but 'smoke' doesn't necessarily include 'smoke from a fire'.) They can also lose them, usually only by extreme physical (to the soul, but physical enough for them) trauma. Silence and discordant music are anathema to them.

All told, the deep connection they have with physics/magic effectively makes every aya a potent mage, or a potent reality-warper, depending on which term you want to use.
Interesting!
sanskacharu wrote:I'm leaving the phoneme inventory and the phonology until I figure out how to make those pretty pretty tables I see in the phonology thread. Anyway.

Basic Typology
Ayakadiya is an agglutinative language, with a heavy focus on verbal morphology. A number of the example sentences I have been translating have been coming up glossed something like 'plain.noun - plain.noun - morphologically.buried.verb', which leads me to make this generalization. Depending on your definition of polysynthesis, and how relaxed you are with it, then Ayakadiya may or may not count as both agglutinative and partly polysynthetic. Take the name of the language itself:

aya-ka-di-ya.
person-all-speak-3.INAN
'All people speak it.'

Here we have an incorporated agent. Note well that aya 'person' is the ONLY noun which can be so incorporated, although it can be so as either an agent or a patient.

aya-ka-we-duu.
person-all-1.SG-rule
'I rule all people.' 'I rule everyone.'

(You have a question about how patient/agent is determined, because you can't tell from these examples? Good. Yes. Of course you do. We'll get to that.)
You know if there were ever found to be a natlang which permitted agentive incorporation, I imagine generic functions would be the main context for its usage.
sanskacharu wrote:Other than aya incorporation, the only polysynthesis test that Ayakadiya meets is "can express full sentences as one word", and even then, it can meet this criterion only by using a sentence which involves only pronominal agents and patients, both of which are marked on the verb, and possibly an aspect marker.

we-reṭ-ya-re.
1.SG-eat-3.INAN-PERF
'I ate it.'
(tense unspecified)

I tend to shy away from calling it polysynthetic for this reason. I don't feel that the justification is sufficient -- certainly if this were a real language I was describing in the field, I wouldn't claim in my eventual paper(s)/dissertation/reference grammar that it was polysynthetic. I think my advisor would eat me for suggesting it.
And that's fine: if you feel the polysynthetic label is unjustified then you don't need to use it.
sanskacharu wrote:Overview of Features
Roughly speaking, Ayakadiya includes the following features. I will explain these in a truncated, sketch-like manner.

Agentivity Hierarchy
The key to comprehending agent/patient assignments in any sentence. Agent/patient is the most saliant binary by which Ayakadiya determines its case roles, such as they are, so I continue to use it. Effectively, this means a decision on whether or not to use an inverse marker on the verb to redirect the action in the unexpected direction. The hierarchy goes like so:

1st person > 2nd person, animate > 3rd person, animate > 2nd person, inanimate-but-alive > 3rd person, inanimate-but-alive > 3rd person, inanimate-and-unalive

If 1st person and 3rd animate are both marked, it is assumed that 'I' am the agent, and the third person the patient. To say that I am the patient and the third person the agent who acts upon me, the inverse suffix -cu must be used.
Yay, non-egalitarian animacy hierarchy!
sanskacharu wrote:Possessive System
There are two means of describing possession. One, the alienable, is most common, and is used as a default. Inalienable possession, the second system, is used in a limited number of situations:
  • parts of your own body or soul
  • your biological children (or your adopted children, if you wish to deeply express your love for them)
  • things you have made, during the making of which you have poured significant time, energy, and emotional investment into them, such that they have left a mark on your psyche. Think an artist's masterwork. It would be appropriate for me to describe any of my conlangs with the inalienable possessor, as well as my thesis, although the two evoke very different reactions.
In other situations, alienable possession is used.
Interesting how music and the arts are important enough to warrant inalienable classification in certain circumstances.
sanskacharu wrote:Some Basic Word Orders
  • N Det
  • N Adj
  • N Num
  • Poss N
  • Rel N
  • Obl N
  • V Adv
  • O S V
'OSV' is better described as either 'patient-agent-verb' (PAV), or as "highest animacy constituent goes closest to verb". Because the higher animacy is assumed to be the agent, OSV is the most likely word order. Another argument for OSV word order:

duu taḳḳa a-mel-le.
Lord child 3.AN-strike-PERF
'The child struck the Lord.'

To say the Lord struck the child, the two nominal constituents must be reversed. taḳḳa duu amelle. You'll also note that the patient is not marked on amelle. When both the agent and patient would require the same pronoun agreement on the verb, only the subject's agreement is marked, and the subject is confined closest to the verb.
Interesting agreement deletion there. So there's no passive/inverse morphology to help disambiguate? No obviative marker? It's all dependent on word order?
https://frislander.tumblr.com/

First known on here as Karero

User avatar
mèþru
Smeric
Smeric
Posts: 1984
Joined: Thu Oct 29, 2015 6:44 am
Location: suburbs of Mrin
Contact:

Re: Ayakadiya

Post by mèþru »

I wish most people here, including me, could put this much detail in their conlangs.
ìtsanso, God In The Mountain, may our names inspire the deepest feelings of fear in urkos and all his ilk, for we have saved another man from his lies! I welcome back to the feast hall kal, who will never gamble again! May the eleven gods bless him!
kårroť

sanskacharu
Sanci
Sanci
Posts: 29
Joined: Mon Feb 13, 2017 8:15 am

Re: Ayakadiya

Post by sanskacharu »

mèþru wrote:I wish most people here, including me, could put this much detail in their conlangs.
Aaah, thank you so much for this. I'm really glad you appreciate it, and I hope it's interesting.

Thanks to Frislander for your comments as well! I hope you enjoyed the read -- I've responded to some of your comments below.
Frislander wrote: You know if there were ever found to be a natlang which permitted agentive incorporation, I imagine generic functions would be the main context for its usage.
This is probably right. To be honest, I started out not intending nominal incorporation to even be possible, but when I was reading example sentences out loud to myself, the 'ayaka' and 'aya' ones started melding right in. (I do live reading fairly often, and much of my phonology is motivated by it in all my conlangs.) I figured I'd just let them meld, because eh. Why not?
Frislander wrote:And that's fine: if you feel the polysynthetic label is unjustified then you don't need to use it.
Hehe. A little idiosyncrasy of mine is that I tend to describe my conlangs as though they were languages I was describing in the field, even though with the conlangs, I do have the authority to make sweeping typological judgements. Again, I think I blame my advisor for this.
Frislander wrote:Yay, non-egalitarian animacy hierarchy!
Now you've got me interested. I've always thought of animacy hierarchies as inherently non-egalitarion -- what would an egalitarian animacy hierarchy look like? Is this something you've seen in conlangs and/or maybe a natlang before?
Frislander wrote:Interesting agreement deletion there. So there's no passive/inverse morphology to help disambiguate? No obviative marker? It's all dependent on word order?
Well, it's a little complex. There is an inverse marker, -cu, but its purpose is to redirect action backwards up the agentivity hierarchy. In the sentence duu taḳḳa amelle, both duu 'Lord' and taḳḳa 'aya child' are animate nouns, and both duu and taḳḳa should receive agreement marking with a-. -cu has nowhere to redirect to in this case, so it cannot be used. Further, Ayakadiya verbs have a slightly templatic nature (though I'm starting to hate that term because of my current class on Navajo, and may change up some things). Regardless, at the moment, it's not possible to double-mark the same person on a verb, because the agreement affixes all go in a fixed spot. You can't have 3.AN-3.AN-VERB -- you have to just do 3.AN-VERB, in which case, the marker that is present is assumed to be the agent. The actual noun denoting the agent moves closer to the verb to do this, because in the moment it's being slightly more animate, and as stated, Ayakadiya likes to have its animate agents closer to the verb.

And now I make a big caveat that I forgot to mention last night. You can't do this whole agreement-deletion-and-word-order-grounding when only pronouns are available. So, there is another way to say all of those conflicting sentences, because Ayakadiya does ALSO have an Oblique.

aa=ċɛ a-mel-le.
3.AN=OBL 3.AN-strike-PERF
'Ey struck em.'

(The pronouns I use to translate Ayakadiya's 3rd person animate singular are nonbinary 'ey' because aya don't have any distinguishable biological (or otherwise) gender. It might also be accurate to call them all 'she', because they're all capable of giving birth, but that hinges femaleness on birthgiving, which is icky to me.)

duu=ċɛ taḳḳa a-mel-le.
Lord=OBL child 3.AN-strike-PERF
'The child struck the Lord.'

So there are two ways to get around the presented problem of agentivity overlap in the hierarchy. One is to use word order -- the other is to background the patient argument with the oblique. In both cases the agent noun is still closer to the verb, but the oblique version gives slightly less ambiguity about which interpretation is intended, so I'd say that the oblique version is probably more common/favored. I could still see word order being used for stylistic concerns, and I think it sounds more old-fashioned.
Last edited by sanskacharu on Tue Feb 14, 2017 2:44 pm, edited 1 time in total.
My conlang family:
  • Ukumusi & Mupuasa -- Two peas in a pod. Tired of your nonsense.
  • Ku Ṣili -- Lonely Misfit. Can't make up its mind.
  • Ayakadiya -- Standoffish, self-important. Needs More Lexicon.

sanskacharu
Sanci
Sanci
Posts: 29
Joined: Mon Feb 13, 2017 8:15 am

Phonemes; Phonotactics

Post by sanskacharu »

I found out that the pretty pretty tables I saw in the phonology thread were images. I feel sad, and fairly stupid.

Phonemes

Image

Image

These clippings are from the odt document where I'm currently keeping my Ayakadiya reference grammar repository. It's possible for consonants to appear doubled in the romanized orthography, ie. /kk jj yy '' ss žž/ &c. This means that they're pronounced as geminates [kː ɟː jː ʔː sː ʒː], but geminates are not actually phonemes. They only appear as a result of phonological processes at syllable boundaries.

/ḥ/ varies freely from the velar [x] to the uvular [χ]. There's also other phonology, but I need to comb through it before I post it.

Phonotactics
Ayakadiya's basic syllable is CV(C). Only vowels can be the nucleus, but any vowel can take the position. By far the most common syllable is CV. Any consonant may be an onset, and any consonant may technically be a coda, although /b d j g z ž/ are made into marginal codas by means of phonological processes which devoice them word-finally, and /h/ is made into a marginal coda by means of a phonological process which deletes it word-finally. Consonant clusters between a coda and onset (syllable one's coda and syllable two's onset) are also prone to various forms of assimilatory change, such that being a coda in Ayakadiya is a fairly unstable pastime.

Codas require an onset to be present -- no VC syllables are allowed. V-only syllables are allowed only word-initially, or as monosyllabic words, such as ururu 'again, repeatedly' and aa '3rd person animate singular' (a free word, not the agreement marker) respectively.

Gotta go to class now. Probably will post phonology table before tomorrow, at least. I need to kick myself into gear and start making more lexicon.

Edit: Augh, I committed a felony misdemeanor. [l] shouldn't technically be a glide, since another word for 'glide' is 'semivowel', and [l] is definitely not that thing. I just have it in that row because it groups with /y/ and /w/ in the phonology. Don't have energy to fix it now, just ignore my flub please.
My conlang family:
  • Ukumusi & Mupuasa -- Two peas in a pod. Tired of your nonsense.
  • Ku Ṣili -- Lonely Misfit. Can't make up its mind.
  • Ayakadiya -- Standoffish, self-important. Needs More Lexicon.

User avatar
Frislander
Avisaru
Avisaru
Posts: 836
Joined: Mon Feb 29, 2016 6:34 am
Location: The North

Re: Ayakadiya

Post by Frislander »

Nice: Indo-Aryan consonants + West African vowels!
https://frislander.tumblr.com/

First known on here as Karero

sanskacharu
Sanci
Sanci
Posts: 29
Joined: Mon Feb 13, 2017 8:15 am

Phonology

Post by sanskacharu »

Finally, something gets done.

Phonology
I would enjoy listing these by the type of sound change they are... but they're also ordered rules, under rule-based SPE phonology. I've briefly considered doing an OT writeup but tossed the idea last semester because of my thesis -- maybe I'll come back to it this year.

Some phonological categories relevant to the below:
  • Consonants (C): /m n ŋ ṗ ṭ ċ ḳ ' p t c k b d j g s š ḥ h z ž r l y w/
  • Vowels (V): /i e ɛ a ɔ o u ii ee ɛɛ aa ɔɔ oo uu/
  • Nasals (N): /m n ŋ/
  • Plosives (P): /ṗ ṭ ċ ḳ p t c k ' b d j g/
  • Fricatives (F): /s š ḥ h z ž/
  • Sibilents (S): /s š z ž/
  • Liquids (L): /r l y w/
  • Glides (G): /y w/
  • Rhotics (R): /r/
The rules can be divided into three basic types -- (1) rules that mess with vowels and glides, (2) rules that affect the word-final consonant, and (3) assimilation nonsense in cross-syllable consonant clusters. I'll label each item with such a number to be more specific about which type of rule it is.
  • If LL, the first L assimilates to the second L. (3)
  • If CG, where the C is not another G, the two coalesce to form an onset [Cʲ] for /Cy/ and [Cʷ] for /Cw/. (1)
  • Voiced P and F become voiceless word-finally. (2)
  • /h/ is deleted word-finally. (2)
  • If NP, the plosive becomes an associated nasal in assimilation. (3)
    • /ṗ p b/ > [m]
    • /ṭ t d/ > [n]
    • /ċ c j/ > [ɲ] (not a phoneme otherwise)
    • /ḳ k g/ > [ŋ]
  • /'/ assimilates the nasal instead, creating a geminate glottal stop.
  • If VV(V...) sequence where one or more V is /i u/, then /i u/ > /y w/ respectively. This is primarily a repair strategy. (1)
  • Any remaining VV(V...) sequence collapses into a single vowel. The initial vowel of the sequence is what remains, invariably long. (1)
  • If PP, the first P assimilates to the second P. If either P initially held aspiration, then the result will be aspirated, unless the second P is voiced or is /'/ -- neither of these can hold aspiration. (3)
  • If PN, N becomes identical to P. (3)
  • If FN, N becomes identical to F. (3)
  • If PR, R becomes identical to P. (3)
  • If FR, R becomes identical to F. (3)
  • If SS, the first S assimilates to the second S. Definitely do note well that /ḥ h/ don't invoke this rule. (3)
  • If NN, the first N assimilates to the second N. (3)
Looking at it now, this list is a goddamn mess and I need to take a scalpel to it. But it's what I have for now. I'll post an updated phonology when I clear it up.
My conlang family:
  • Ukumusi & Mupuasa -- Two peas in a pod. Tired of your nonsense.
  • Ku Ṣili -- Lonely Misfit. Can't make up its mind.
  • Ayakadiya -- Standoffish, self-important. Needs More Lexicon.

User avatar
Chengjiang
Avisaru
Avisaru
Posts: 437
Joined: Sun Feb 26, 2006 4:41 am
Location: Davis, CA

Re: Phonology

Post by Chengjiang »

Nice phonotactics you've got there. Reminds me of Pali.
[ʈʂʰɤŋtɕjɑŋ], or whatever you can comfortably pronounce that's close to that

Formerly known as Primordial Soup

Supporter of use of [ȶ ȡ ȵ ȴ] in transcription

It is pitch black. You are likely to be eaten by a 青.

sanskacharu
Sanci
Sanci
Posts: 29
Joined: Mon Feb 13, 2017 8:15 am

Re: Ayakadiya

Post by sanskacharu »

I'm super stupidly stuck on what to write for my next post, and considering that I got slammed with homework, I might not have time to write anything more significant until the weekend.

In the meantime, have a glimpse into one of the notebook pages that I scribbled on back when I was designing Ayakadiya's vocabulary. At this time, it was called 'Ayadi', and the scribblings were written in a notebook I had originally intended to be for tarot study -- thus the nonsequitor heading.

Image
My conlang family:
  • Ukumusi & Mupuasa -- Two peas in a pod. Tired of your nonsense.
  • Ku Ṣili -- Lonely Misfit. Can't make up its mind.
  • Ayakadiya -- Standoffish, self-important. Needs More Lexicon.

Post Reply