#afata

May 30, 2019

We were practicing Afata the box-shape move from Tahitian dance.

30 Mai 2019

A while back I posted that I had created four languages for the Tencent game Arena of Valor. I also said at that time that I’d post information on them later, because they were going to be put into the game at some later time when the game would be getting an update. I guess they’ve done this update, but not for all servers...? A player who plays on a server from another country emailed me and showed me the video above, so I guess that means it’s out there. Consequently, I’ll be introducing the languages here on my Tumblr in a series of posts over the next few days.

I created four languages for Arena of Valor, three of which are daughter languages of the first. I also got the opportunity to create a writing system for each one, where again the three daughter writing systems are derived from the parent. It was an absolutely extraordinary opportunity, as I’ve never had the chance to control the linguistic narrative from the beginning of time in a given universe. They gave me a detailed history of the lands that went back to the creation of the universe and left the rest up to me with few constraints. The result are the following languages:

The Veda language (for Veda characters).

The Afata language (for Afata characters).

The Gandal language (for human characters).

The G’vunna language (for Lokheim characters).

Here are the translations I did for the Afata language:

SLOGANS

Ho’Delanaamen sezukudhí!

“Remember Tel’Annas!”

How’Athonuzhu ümem shooshamung miwangü.

“We will defend Athanor from all threats.”

How’Athonuzhu ümem shooshamung küwangü.

“I will defend Athanor from all threats.”

Üüdorasi shumimen hoshomeedhen vüroshunu.

“Yggdrasil gives us strength.”

Üüdorasi shungümen hoshomeedhen vüroshunu.

“Yggdrasil gives me strength.”

BIRTH

KRIXI

Mibuthí!

“Let’s fly!”

CHAUGNAR

Shuzhawoodasi küguveeshunu kümana.

“I live to serve the forest.”

TORO

Nen küdana ven zhoro küniine!

“Just woke up and I’m already angry!”

CRESHT

Üshuuzhu kübanüfo.

“I rule the oceans.”

FENNIK

Üshaydho zhuvidhí: küne!

“Lock up the silverware, here I come!”

LUMBURR

Ay’Athonusi Umu, küzhongidhí!

“Mother Earth, make me strong!”

PEURA

Avimaa howüdhen üvame vürorele.

“Nature always finds a way.”

SLIMZ

Shunusu shayuzuuzhe künii!

“I’m open for business!”

KILGROTH

Odh’ Ayü…

“Red Tide…”

TELANNAS

Fünoyo füvuru me künii.

“I am endless and eternal.”

ZUKA

Üyana füyobeedhen künenena!

“Back for more sparring!”

ZILL

Vogo vumeetam…

“A storm is brewing…”

LINDIS

Wele yen ranedama.

“The cycle starts anew.”

TEEMEE

Iizhudhon: Hodhomimeedhen minolo!

“Together: We’ve got to focus, focus!”

ARUM

Ügaakodho minedemidhí.

“Let’s start the hunt.”

BALDUM

Üraazhudhon foothí!

Actually, despite what I said in my last post, I don’t see the point of not putting up all the glyph tables, so here they are!

I’m not sure how often wu or yi gets used, but I know it was more than zero times (I ended up needing them).

As you can see, b, t, and d all have the same connection point.

In this way, k, f, and v all have the bar connector, though with k it’s raised a bit.

Depending on the word, this last row may come out shu, zhu, shu, and zhu. It varies.

The ng series is one of my favorites. As with the last table, the last row may come out shu, mü, nyu, and nyu. (No change to mü.)

These latter two are former Veda q and Veda gh. These are their schematic forms; not necessarily how they’ll end up being pronounced.

Next, there are some characters that connect to either y or w forms. This is a list of them below:

Again, with these you can think of them as specialized ligatures. Not every consonantal glyph will combine; only those where it’s visually convenient.

The sy and zy forms are all sh and zh (bearing in mind the previous comment about the status of ü). For the zh forms, you’ll see they don’t actually combine, but there’s a modified z that’s used specifically before the glide.

You know, I thought more glyphs combined with y... I guess not.

I’m not sure that last column is ever used.

That covers the main glyphs. Here are some miscellaneous symbols used for punctuation:

And finally, if you’ve seen the grammar of Afata, you may be wondering how long vowels are written, as you will have noticed there aren’t any special characters for long vowels. Welp, the truth of the matter is you just have to figure it out. A lot of that goes on. For example, take the word maga, which means “eagle”. It comes from Veda maq. Afata, of course, has no q anymore, and also doesn’t allow words to end in a stop, so the copy vowel a is added, and in between vowels both old k and q become g, giving us the modern word. This is how it’s spelled:

That is, ma-qa. If you’ll recall from the first post, Afata maga has an alternate stem in maa that occurs in a lot of case forms. Here are some of them and how they’re spelled:

As you can see, our old boy q is in there still. Basically, the way this orthography works is speakers just know what clusters are impermissible, and those consonants are treated as length markers—in places where length is permitted. (So the extra h’s on these inflectional forms don’t trigger length because they’re inflection, and everyone knows those vowels are short.)

The orthographic system of Afata is extensive and confusing. For the basic facts, the system that arose from Veda’s is now an abugida, and is called the Thala (or vine) script. The major glyphs are consonantal with little added bits for the vowels. You can see the major correspondences in the table below:

Those are the basic characters. I think you can see mostly how they work. Certain of them developed a visual element that’s used specifically as a landing place for vowels. For example, if you look at k (the first one), f, v, s, z, h, j (the first one) and r, you’ll see a straight line on the right. That’s where vowels attach. (Well, except maybe for h, because it has the whip thingy above it like the second k [former q], and that’s what’s used for vowels on that one). Vowels, of course, are a big deal in this one, so this is basically how they work.

The glyph for a was a carryover from the old glottal stop character from Veda. Similarly, the glyph for i comes from the old glyph for Veda y, and the glyph for u comes from the old glyph for Veda w. They still have semi-vowel forms as well. To make the word-initial forms for e and o, the vowel markers for i and u, respectively, were added to the base a glyph (you can see kind of how they worked based on the old Veda forms). Then their vowel markers (the e and o vowel markers) are simply modifications of the i and u vowel markers.

So compare real quick the main e glyph and the glyph for ki. You can see they have the same modification. Then for ke, you basically start with ki and add a swoosh that lets you know it’s different. With ko the swoosh is already there, so you double the circles. And those are the forms for the vowels.

Now, the trick here is that basically these vowels began attaching on the right, which is, linearly, where they appeared in Veda. As time wore on, though, it became more convenient to attach the vowels at the end point of the last line, wherever it happened to lie. Thus, for some characters, vowels appear below the main glyph, while for others they appear in front, or inside. When there’s a visual element in common, the vowels always appear in the same place. Otherwise, it depends on how the character is written. Here are some key examples (by the way I forgot that yu has its own special vowel attachment, even though it does not have a special base glyph anymore):

(Also, and I know this gets confusing, but what was *bjw, byu in Veda, is now bü. So all those ju forms above are the vowel ü.)

In addition to these basic vowels, certain consonantal glyphs also combine with glides to form even more characters:

It’s not really based on the sounds, per se, so much as the shapes. They determine which ones will combine with glide characters and which won’t.

That’s the basics of the system. You can see the number system below:

And here’s an example of a full sentence:

That’s one of the Afata slogans: How’Athonuzhu ümem shooshamung küwangü. “I will defend Athanor from all threats.” It might be worth it to put up all the consonant + vowel combinations at some point, but this is good enough for introduction.

In creating this script, I was hoping to create something as wild and undisciplined as some of Trent Pehrson’s scripts (like this one). I don’t think I succeeded, but, provided the art department will let me go ahead with it, I think I’m working on one right now that approaches it. Excited about that one. We’ll see!

Oh, in case you want to see this scribbledygook, here you go:

Continuing from the last post, here’s the second half of the introduction to the Afata language from Arena of Valor.

VERBS

There are basically two different things going on with verbs in Afata: argument agreement and conjugation. Both are involved.

CONSONANT MUTATION

Before getting to agreement, Afata features two different types of consonant mutation. The first is called V-mutation (my shorthand for “in between vowels or vowel-like things”) and the second is called N-mutation (consonants occurring after nasals). Starting with V-mutation, the following happened when a consonant appeared in between two vowels (these are Veda consonants):

Voiceless Stops: p t ts k q ʔ > b d dz g g {j/ɾ}

Voiced Stops: b d dz g > v ð z ɣ

Nasals: no change

Voiceless Fricatives: f s h > v z j

Voiced Fricatives: v z > w z

Approximants: no change

After these changes, the Afata changes applied, so ɣ became j; ts became θ; and dz became ð.

The N-mutations look like this:

Voiceless Stops: mp nt nts ŋk ɴq nʔ > b d dz g g n

Voiced Stops: mb nd ndz ŋg > m n n ŋ

Nasals: no change

Voiceless Fricatives: mf ns nh > p ts k

Voiced Fricatives: mv nz > b dz

Approximants: nɾ nl ŋw nj > d n ŋw ɲ

And, again, Afata sound changes applied. Crucial in these latter sound changes is that the clusters themselves merged, making it look like, e.g., f is going to p straight up.

Okay, with that in mind, let’s look at agreement.

AGREEMENT

In Afata, the verb agrees with the subject and object, if present. These are done with prefixes, and so the prefixes differ depending on whether the verb begins with a consonant or vowel. With a vowel, you get a consonant-final prefix; with a consonant, you have the possibility of mutation, depending on the history of the prefix.

Intransitive verbs are simple. The agreement prefixes look like this (always listing the form added to V-initial verbs first then C-initial verbs):

First Person Singular: kün- / kü- (+N-mutation)

Second Person Singular: d- / da- (+V-mutation)

First Person Plural: min- / mi- (+N-mutation)

Second Person Plural: s- / sa- (+V-mutation)

Third Person Sentient: zuh- / zuu-

Third Person Inanimate: ran- / ra- (+N-mutation)

Third Person Animate: vw- / vu- (+V-mutation)

Hopefully that’s clear. (Also be sure to note that there’s no plurality distinction in third person.) What comes next are two tables where you get one of these agreement prefixes after the other for transitive verbs, and I’m kind of lazy, so I’m just going to paste them as images. (Hey, Tumblr! Why can’t we do tables?!) You can probably figure out what the forms would be if you bear in mind that these things are going to be one syllable max (so CVC). Here’s what the forms look like before vowels (no mutations):

The variability in the third inanimate subject/object prefix depends on whether the following vowel is a front vowel or not. If it’s a front vowel, the form is ren-; otherwise, the form is ron-. For the third animate subject/object prefix, I’m actually not 100% clear why there’s variability... I can’t remember what I did. Based on the other table, I think the form vüb- is used for reflexives, but the form vün- is used for two different third person animate arguments. It could be the other way around... Man. I don’t remember what source I used for reflexives. I’m sure I wrote it down somewhere, but I can’t find it. Anyway, here’s the next one:

Pink means it causes V-mutation; yellow, N-mutation. Then the bottom right box is explained below. Notice how I say “for its reflexive form, it causes no mutation” but I don’t say which one is the reflexive form! I have to think it’s the b form, but I can’t say for certain. You see the same thing happening with the other third person forms. Since all the first/second person forms end in o, I’m going to guess it’s that b/o form that’s reflexive, but that’s just an educated guess, since like an idiot I didn’t write this down. Ugh.

Anyway, so verbs start out with one of these agreement prefixes, depending on its arguments (alignment is accusative). Then comes the verb, which changes based on its conjugation.

VERB CONJUGATION

Verbs have four forms—imperfective, perfective, imperative, and infinitive—and two voices, active and passive. I’ll show you three fully conjugated: pü “to drink”, bana “to live”, and adha “to hope”. Starting with the active:

Active Voice

Imperfective: pü, bana, adha

Perfective: püno, bana, aana

Imperative: püdhí, baní, athí

Infinitive: pübe, bame, aape

The accented vowels are stressed. If you’re wondering how these work, C-final forms get copy vowels in the imperfective; the form is otherwise unmodified. In the perfective, that’s just the old nǝkh verb from Veda which has now shortened up to a suffix. Its vowel changes depending on the vowel of the previous word. For front vowels, it’s e; for rounded back vowels it’s o (note that Afata pü was Veda pul); a otherwise. That’s why the perfective and imperfective for bana are identical. (They’re spelled differently in the orthography.) The imperative form comes from adding tsip, “quick”, on the end (why it gets the stress), and the infinitive form comes from Veda pe, which means “type” or “kind”.

Now for the passive:

Passive Voice

Imperfective: übü, üvana, üyadha

Perfective: übüno, üvana, üyaana

Imperative: übüdhí, üvaní, üyathí

Infinitive: übübe, üvame, üyaape

This passive form comes from the old Veda passive/inchoative yu, which has become ü in Afata and engendered V-mutation in C-initial forms.

Aside from this, the only noteworthy thing is that Afata developed a series of verbal augments—basically because verb forms were too short, and sound changes had made a lot of them indistinguishable. They basically enhance the meaning of whatever verb there is, and also give it an ending in common with a lot of different verbs. A nice little way to flesh out the verb inventory.

*

This is the language of the Afata from Arena of Valor. I took some leaps with this one. Like Gandal, it’s descended from Veda. I’m not sure how well or efficiently I’m going to be able to explain this, because I’d frequently get lost with this one. We’ll see how it goes!

PHONOLOGY

The Afata language got all the long vowels and fricatives. Here’s a look at its inventory:

STOPS: p/b, t/d, k/g

FRICATIVES: f/v, θ (=th)/ð (=dh), s/z, ʃ (=sh)/ʒ (=zh), h

NASALS: m, n, ŋ (=ng)

APPROXIMANTS: w, l, ɾ (=r), j (=y)

The vowels are nice and simple:

HIGH: i/iː (=ii), y (=ü)/yː (=üü), u/uː (=uu)

MID: e/eː (=ee), o/oː (=oo)

LOW: ɑ (=a)/ɑː (=aa)

The sound changes that were applied to Veda for this one resulted in most codas getting mowed down. Also, in case you’re wondering, the *ts > θ sound change happened in each of these three languages specifically because the game calls the land Athanor, and I had to make it work. In Veda it’s Atsnor, but it comes out pretty close to Athanor in the three daughter languages. Anyway, *ju became y, most codas were deleted before onset consonants resulting in long vowels, and affricates were simplified.

As a note, this language developed a whole bunch of cases, and in the nominative, many nouns picked up a copy vowel that suffixed to the root. This copy vowel was still bound by the same lowering that happened with “back” consonants.

PRONOUNS

Afata divided nominals into three types: sentient, inanimate, and animate (these are mainly animals that can’t think/talk). These applied to the third person pronouns, giving Afata the following pronouns:

First Person: künu “I”, mini “we”

Second Person: daa “you”, saa “you all”

Third Sentient: zulin “she, he, it”, zume “they”

Third Inanimate: ranin “it”, rame “they”

Third Animate: vün “he, she, it”, vüme “they”

By now you probably recognize the old Veda pronouns plus mem and lin. The third inanimate pronoun comes from Veda rǝn which means “thing”. I don’t actually recognize where the animate pronoun came from (unless it came from vu “cat”, but that seems doubtful—or at least no the full story. This is why it pays to write those etymologies down!). These pronouns all decline for case, but beyond knowing what the stems are, it makes more sense to jump straight into nouns to understand how the cases work.

NOUNS

Afata nouns inflect for case, but, mercifully, not number. There are 18 cases in Afata: nominative, accusative, dative, instrumental, benefactive, malefactive, apudessive, allative, ablative, inessive, illative, elative, superessive, superlative, delative, initiative, perlative, and aversive. Why that many—and those? It just made sense based on the way I was doing things. They all came up pretty naturally.

Basically, aside from the unmarked nominative, Afata took five prepositions from Veda—ag, kho, yu, shun, and ghor—and these fused to the front of nouns. They combined with nouns that had to do with the structure or body of a prototypical sentient, inanimate, or animate noun. So, for example, ag was “come” in Veda. Zukh was a word for a sentient being. To form cases, then, you got ag + a word in the sentient class + a part of the body. So it’s a different thing to come from the chest of a sentient being vs. coming from the foot of a sentient being or the hand. This same process applied to a prototypical inanimate and animate noun (the latter probably was “cat”, come to think of it). The combinations didn’t always yield a usable result. For example, ghor + foot/base doesn’t yield a usable case for any class. On the other hand, kho + top/animal back is a case for all three classes, but they’re different cases: superessive for sentient and inanimate (sitting on the top of), but accusative for animate.

So, while it looks like a large, arbitrary number of cases, what it really is is just preposition + possessor + (body) part. These fused, and the ones that made sense stuck. So, with that said, here’s what this mess looks like (the words are “sentient being”, “shield”, and “eagle”, in that order):

Nominative: zuu, anga, maga

Accusative: hozuumen, howangidhen, homaazhu

Dative: shudhuumen, shunangidhen, shumaasi

Instrumental: ezuumen, ayangidhen, emaasi

Benefactive: shudhuuzhe, shunangun, shumaale

Malefactive: hozuukon, ürangung, homaamas

Apudessive: hozuulu, howangung, homaapa

Allative: üzuulu, üyangung, ümaapa

Ablative: ezuukon, ayaashudhon, emaamas

Inessive: hozuuzhe, howangun, homaale

Illative: üzuuzhe, üyangun, ümaale

Elative: ezuuzhe, ayangun, emaale

Superessive: hozuudho, howaadho, homaazhu

Superlative: üzuudho, üyaadho, ümaazhu

Delative: ezuudho, ayaadho, emaazhu

Initiative: ezuulu, ayangung, emaapa

Perlative: üzuumen, ürangidhen, ümaasi

Aversive: üzuukon, üraashudhon, ümaamas

Now it may be the case that presenting these as cases in this way is not the best way to do it. Maybe it’s a different type of phenomenon that deserves its own format. If there’s a better way, though, I haven’t found it. All these things are single words, and they’re all stressed on the second syllable. Basically you have a series of different prefixes, a series of different suffixes, and a couple of different stem forms (sometimes). You need to know the stems and the affixes and which arrangement gives you something licit for which type of noun, but you can know those things separately. It goes without saying that at this stage, none of these affixes actually have any meaning whatsoever. They simply exist as pieces of a greater whole. It takes some work to figure out, but I feel like once you get the hang of it, it’ll work pretty well.

ADJECTIVES

Adjectives are fairly simple. Adjectives have a comparative and superlative form (so there’s a superlative for nouns and superlative for adjectives and the two have totally different meanings), and that’s it. A dramatic reduction compared to the Veda forms. The form differs depending on what the adjective ends with, so here are some relevant examples (looking at words that, in order, end with a vowel, stop, nasal, and fricative):

Plain: zho “large”, thibi “quick”, shongo “strong”, shuvu “quiet”

Comparative: zhoyu “larger”, thiiyu “quicker”, shoogu “stronger”, shuuyu “quieter”

Superlative: zhoyume “largest”, thiiyume “quickest”, shoogume “strongest”, shuuyume “quietest”

And that’s all there is to that.

*