Ai Se Sa Nimama? Interlinear - Part 2

This is the conclusion of the interlinear gloss and commentary for my Koa translation of Are You My Mother. Part 1 is here, and the full text is here.

Ka ame-nene i luvu ve cu-meti.
DEF bird-baby VB.CL cease ADV.CL IRR-think
The baby bird ceased (in order) that he might think
"The baby bird stopped to think."

Directly inspired by Nahuatl, here we have an irrealis adverbial clause ve cumeti "in order that he might think." This is one of two primary ways of showing purpose (more about this in an upcoming post on adverbial clauses); the other would be the nonfinite la ko meti, literally "for thinking."

Ka séne-to e ka móa-he i na-ta-mama.
DEF cat-CHILD and DEF chicken-FEM VB.CL NEG-3SG-mama
The kitten and the hen are not his mama
"The kitten and the hen were not his mother."

In re the little argument I had with myself in Part 1, here's yet more internal evidence that inalienably possessed nouns can be predicates on their own...

Ka kunu e ka léma-e i na-ta-mama.
DEF dog and DEF cow-FEM VB.CL NEG-3SG-mama
The dog and the cow are not his mama
"The dog and the cow were not his mother."

Ni-ku-me-mama, — ka ame-nene i sano,
1SG-OLD-COM-mama DEF bird-baby VB.CL say
I clearly am with mother, the baby bird said
"I have a mother, said the baby bird."

I originally translated this as ni-ia-me-mama with the ia "yes/certainty" viridical showing verum; in this interpretation, the baby bird was pushing back against a universe he felt was suggesting he might not have a mother. I don't think that's positively wrong, but as I thought about it, in this moment I don't think he's focusing on the truth of his statement so much as establishing a base of known information. Ku is much better, a particle which is almost impossible to translate succinctly and fully into English (Polish has a perfect translation, przecież: Matkę przecież mam... "I [obviously] have a mother..."). It indicates that the information being conveyed is old, and is already (or should already be) known. English can kind of get at it with "clearly," "obviously," "of course," or "you know."

Ni-ilo, ia-ki-tai.
1SG-know VERUM-DEB-be
I know, it must be
"I know I do."

...on the other hand, here he definitely is asserting truth (just like I am in this moment), so ia is the most appropriate translation.

Ni-lu-lúta-ta. Ni-cu-lúta-ta. IA-CÚ!
1SG-VOL-find-3SG 1SG-IRR-find-3SG VERUM-IRR
I want/intend to find her. I will/would find her. Will/would!
"I will find her. I will. I WILL!"

I'm attempting here to convey the baby bird's escalating commitment to the task at hand, though using somewhat different words than the English. I continue to be a little uncertain about full extension of contextual meaning available to each of these particles lu "volitive" and cu "irrealis"; I'm hoping that lu can convey intention in the first clause, and then that cu could be a prediction of the future in the others. Now I wonder if all three clauses are really more about determination than fortune telling, though...should I have done lu for all three? Nilulútata. Ialú. IALÚ!

I guess there are two issues here: the accuracy of the translation of the English meaning on the one hand, and the naturalness of the Koa on the other. When I read just the Koa, this feels like a reasonable thing for the bird to say in this moment, in line with the vibe of the English if not exactly the same. For the moment I'll leave things as they are with the lu version as an understudy...and I'd really better think more about distinguishing intentionality versus desire with lu, and futurity versus imagination with cu.

Ie-he-toa ka ame-nene vo nae a mea iso.
JUST-TEMP-that DEF bird-baby PRES.CL see INDEF thing big
Just at that time, behold, the baby bird sees a big thing
"Just then the baby bird saw a big thing."

Should I have thought more about this calque of the English "just then"? It might very well be idiomatic Koa, but it's also not a phrasing I've ever actually used before. On the other hand, the presentative vo is doing great work in this sentence.

SEKA sa ni-mama! — ta-sano.
2SG.EMPH FOC 1SG-mama 3SG say
You are the one who is my mama! he said
"You are my mother! he said."

Noting in passing the first recorded use of the emphatic/predicative version of a personal pronoun. The bird could have said just se sa nimama, but not with the degree of emphasis we see here.

Ka mea iso i sano — HUAUU!
DEF thing big VB.CL say snort
The big thing said SNORT
"The big thing said, 'SNORT!'"

I have no idea how to actually say "snort" in Koa, so I tried to imagine how one might write the sound of a steam shovel in Koa phonology...

Ii, na! — ka ame-nene i sano.
EMOT NEG DEF bird-baby VB.CL say
Yikes, no! the baby bird said
"Oh, no! said the baby bird."

Hidden away in the "Particles" tab of the Koa lexicon and never before mentioned aloud is a list of emotive noises. I was glad I had something ready to translate that "oh, no!" intentionally! The full list is:

aa understanding, surprise
ee uncertainty; filler (um, uh, er)
ei calling attention
eu disgust
ii pain, dislike or nervousness
oi request for repetition or confirmation
oo understanding, confirmation
ui regret, commiseration
uu excitement, pleasure

Na ni-mama sa se. Huauu mu-pato sa se!
NEG 1SG-mama FOC 2SG Snort CAUS-fear FOC 2SG
It's not my mama that you are. Snort cause-fear is what you are
"You are not my mother. You are a scary Snort!"

Ka Huauu i nose ka ame-nene
DEF Snort VB.CL raise DEF bird-baby
The Snort lifted the baby bird
"The Snort lifted the baby bird"

i la-nomu, la-nomu, la-nomu.
VB.CL DAT-upper DAT-upper DAT-upper
to-upper, to-upper, to-upper
"up, up, up."

Sii hua i osu.
next something VB.CL occur
Next something happened
"Then something happened."

Ka Huauu vo ie-pane ka ame-nene
DEF Snort PRES.CL JUST put DEF bird-baby
Behold, the Snort just put the baby bird
"The Snort put the baby bird"

ie "just" is translating the sense of "right back in the tree" in the next line.

i lai la ka puu.
VB.CL return DAT DEF tree
returns to the tree
"right back in the tree"

Ka ame-nene i ho-ne-koto!
DEF bird-baby VB.CL NEW-LOC-home
The baby bird is wow in home!
"The baby bird was home!"

Ho is exactly the opposite of ku, marking information that's new, previously unknown, or surprising. This could also have been a strategy for two clauses above, as in Ka Huauu vo hoiepane ka amenene. This particle's usage is still in the process of being fully understood so I'm not sure whether two ho's in quick succession like that would be too much stylistically; in case it would be, I thought it should be reserved for the conclusion of that series of dramatic occurrences.

Ie-he-toa ka ame-mama vo lai.
JUST-TEMP-that DEF bird-mama PRES.CL return
Just at that time, behold, the mama bird returned
"Just then the mother bird came back."

Ni-ilo ka imi SEKA, ka ame-nene i sano.
1SG-know DEF self 2SG.EMPH DEF bird-baby VB.CL say
I know the equals-YOU one, the baby bird said
"I know who you are, said the baby bird."

Na séne-to, na móa-he, na kunu sa se.
NEG cat-CHILD NEG chicken-FEM NEG dog FOC 2SG
It's not kitten, not hen, not dog that you are
"You are not a kitten or a hen or a dog."

Na léma-e, na Huauu sa.
NEG cow-FEM NEG Snort FOC
It's not cow, not Snort
"You are not a cow or a snort."

Ame sa se, e mama nika sa!
bird FOC 2SG and mama 1SG.EMPH FOC
A bird is what you are, and it's my mama!
"You are a bird, and you are my mother!"

No further commentary here other than that I particularly like the translation of the last five sentences. On the whole, warts and all, I'm pretty happy with this inaugural translation in general; I feel like a Koa-speaking toddler would find it fun, natural and idiomatic. I suppose now I ought to try another genre...maybe a short news story? That would maintain the presentative nature of the information structure, but flip to the far other side of formality of language: maybe an interesting exercise. I'll think about it.

Finite clause types at last

In the winter of 2012, while on a dog walk in Washington Manor, out of the blue I had an idea about Koa clause marking. This idea felt crazy and revolutionary, but also beautiful, and possibly even ingenious. I dismissed it as too revolutionary and let it go for almost a decade, but never forgot it; and in the past few years of feverish Koa development I haven't been able to stop thinking about it.

The idea concerns the i that precedes the VP when it lacks a pronominal subject. Prior to 2012 I assumed the rule was simply that verbs have a pronoun slot that must be filled, so if an overt pronoun is absent, i is required as a placeholder. This is the case (more or less) in Toki Pona, which originally inspired this marker.

The revolutionary idea was that perhaps I had misanalyzed my own syntax. What if, instead of being a  pronoun, i is marking the clause type? And if that's the case, what if the ko I had previously regarded as a specifier that nominalizes clauses, or a complementizer, is actually just marking another type...and the u I thought was a relativizer was marking yet another? And what if there could be other markers too?

Borrowing from the original post in January 2012 in which I introduced and then immediately discarded this idea, this would give us the following marking for finite clauses:

i - verbal
ko - nominal
u - adjectival

...and the following template for the VP:

(SUBJECT) TYPE (PRONOUN) (TAM) VERB (OBJECT)

There's an important bit of subtlety here that I need to make sure to touch on before continuing. When languages nominalize clauses they usually become non-finite, sort of by definition; but as I said above, clauses with these markers are all finite. What one has to remember is that every Koa predicate can be verbal, nominal, or adjectival depending on its syntactic position; in this system, clauses are no exception. These clause type markers identify what the clause is doing in syntax -- i.e. whether the clause is acting like a verbal, nominal, or adjectival predicate -- but the clauses are still finite regardless.

Clause nominalization, the nonfinite variety, is an entirely different topic...though it does also exist in Koa, and would also be able to used for similar purposes! This is all very easy to confuse, and I think I've been consistently inconsistent or imprecise in my language here in the past: in Koa there's an extremely important distinction between a nominal clause and a nominalized clause. I should possibly ideally find a different term for one of these!

Anyway, those who have been following the plot closely may have noticed that I've been nonchalantly slipping structures like these into example sentences and translations for more than a year. After I discovered last year that Macedonian does "nominal clauses" this way too, e.g.

не сакам [ Јуле да знае ]
NEG want.1SG [ Julie NOM.CL know.3SG ]
"I don't want Julie to know"

...for which the precisely parallel Koa translation would be...

ni-na-halu [ le Iúliki ko ilo ]
1SG-NEG-want [ NAME Julie NOM.CL know ]

...and also discovered that Basque does nominal (and conditional, and adverbial) clauses this way:

[ bere aita Californian dago-ela ] esan_du Mikelek
[ his father California.LOC be.3S-NOM.CL ] say.3SG Michael.ERG
"Michael says his father is in California"

le_Mikele i sano [ taémaka ko ne le_Kalipónia ]
Michael VB.CL say [ his.father NOM.CL LOC California ]

...and that Swahili does relative clauses this way:

kitabu [ a-li-cho-ki-soma mtoto ]
book_β [ 3SG.SUBJ_α-PAST-REL_β-3SG.OBJ_β-read child_α ]
"the book that the child read"

ka tusi [ ka toto u luke ]
DEF book [ DEF child REL.CL read ]

...I was convinced that this isn't a completely outlandish, typologically ridiculous strategy at all. It's certainly not augmenting Koa's IAL-worthiness, but as I keep pointing out, in some cases the art is starting to feel more important than the charter, and these structures feel right for Koa. They're simple and elegant, and Koa syntax as a whole actually makes more internal sense when they're handled this way.

And so, by the power vested in me, I hereby admit these structures into the Koa canon. As ever I have the right to change my mind, but I'm going to give them their fair shot and see how Koa feels in practice when it makes full use of them.

I alluded above to the possibility that there might even be more than the three clause types referenced so far, and indeed many more have surfaced. Here are the categories and subcategories of clause types as identified so far:

i - verbal: neutral unmarked
vo - verbal: presentative
oe - verbal: obligative
ea - verbal: hortative
ko - nominal
u - adjectival
ve - adverbial
ha - conditional

Notes about distribution:

1) Verbs with pronominal subjects have some special properties, which affect the use of these clause markers. i is not used (or is extremely marked, TBD); ko is optional; and u is also optional. The remaining markers are required in all circumstances regardless of clause type.

2) A clause carries one and only one clause marker; BUT

3) Serial verbs to the right of the main verb are marked by i regardless of the clause type of the main verb (see example below).

In any event, this is really just an introduction and opportunity for me to welcome this new(ish) syntax out loud. Many of these types will need a post of their own to demonstrate usage; first up and currently in progress is a discussion of the newly discovered and rather exciting conditional structures, which have been patiently waiting for someone to notice them for 25 years.

Oh, and about those actually legitimately nominalized clauses I mentioned earlier...it turns out I got those right in that same 2012 post as well, actually. They're formed with specifier version of ko, and all clause markers -- as markers of finiteness -- are omitted. One example for now, pending a full treatment in a later post:

le Mia i sano ko [ ka moa Ø ma-lalu i poli ]
NAME Mia VB.CL say NOM [ DEF chicken NONFIN IMPF-sing VB.CL much ]
"Mia said the chickers were singing a lot"

Days of the week

At the same time that I was working on the Koa dictionary and phrasebook that ended up being a Solstice gift for my girls in December 2022, I figured out a little corner of the Koa lexicon that had waited a long time for attention. It subsequently got buried in 2023's avalanche of important progress, so let's finally document this.

The names of days of the week had always felt like one of those areas whose development was bound to be marred by arbitrariness, something I felt powerless to take on in earlier Koa: there really is no reason -- as in derivation from core Koa principles -- that any one system would be superior to any other. Perhaps the most obvious option would be reference the days by number, sort of like in Portuguese, like

énapai "Monday" (one-day)
lúapai "Tuesday" (two-day)
tátupai "Wednesday" (three-day)

...and so on. There are two issues with that approach, though: (1) Which day should be day 1, Monday or Sunday?? I have my own feeling about this (as above), but Portuguese does it the other way. Where Koa usage varied from what felt sensible or natural to a learner, would using numbers actually make it feel less universal? (2) I just don't like the number days aesthetically. Sometimes I just have to grit my teeth and accept decisions that feel right even though they aren't my preference, but in this case I didn't feel that there was a strong reason to do so. Days of the week were one of the first places where the conceptual territories of Koa were being named; if the choices were going to be arbitrary, I wanted them at least to have poetry!

I liked the basis in nature of Sunday (sun) and Monday (moon), but not the rest of the Roman planet-based naming scheme. I looked for other nature words that might work, with the additional design goal of having every weekday name start with a unique letter for ease of abbreviation, and this just sort of tumbled out almost fully-formed:

E - ésipai "Monday" (moon-day)
A - ánupai "Tuesday" (water-day)
M - múnupai "Wednesday" (middle-day)
L - lúlupai "Thursday" (tree-day)
K - kívipai "Friday" (stone-day)
V - válopai "Saturday" (light-day)
S - súapai "Sunday" (sun-day)

(I suppose, thinking about it now, that I could have followed Koa alphabetical order with the assignment...though then I would again have had to grapple with which is the "first" day of the week! And more importantly, the poetry...)

Obviously this is a question of my own aesthetics, but I really, really, really like these. There have rarely been design decisions that have felt so "right" to me so easily.

In syntax, just like any other time words (like tana "today") these need a specifier if talking about them, or a temporal particle he if referencing the temporal position of an event:

po ésipai i hive
GEN Monday VB.CL awful
"Mondays are awful"

ai nu-ma-puhu pe ka lúlupai ai ka kívipai?
QU 1PL-CONT-speak BEN DEF Thursday QU DEF Friday
"Are we talking about Thursday or Friday?"

he ánupai ni-me-lía-pi níke-te
TEMP Tuesday 1SG-COM-too-QUANT meet-INSTANCE
"On Tuesday I have too many meetings"

As with other time words like viko "week," "last" is vime and "next" is sii: ka/he múnupai vime "last Wednesday," ka/he súapai sii "next Sunday."

Before coming up with these day names I had actually suggested names for the months of the year as well, and initially really liked them too...but there was a major problem that will be very obvious:

válokuu "January" (light month)

lúmikuu "February" (snow month)

vúakuu "March" (rain month)

lúlukuu "April" (flower month)

náukuu "May" (birth month)

késakuu "June" (summer month)

kúmakuu "July" (hot month)

kócukuu "August" (fruit month)

mákekuu "September" (harvest month)

hísikuu "October" (mist month)

túlikuu "November" (wind month)

tévikuu "December" (winter month)

I still feel embarrassed that it took a friend (thanks, Kate) to point out that these names are highly specific to the northern hemisphere above a certain latitude...not exactly appropriate for a project with any pretensions to IALdom. Though I've workshopped this a bit I have yet to come up with an alternative system that feels sensible to me: whether using numbers or some kind of nominal compounds, whether semantically meaningful or not. Oh well, one of these years.

Special predicates

This is only an idea, not a certainty, but I've gotten the sense over the past decade of consideration that predicates with initial pronouns may have some special properties that need to be fully described. These predicates fall into two classes unified by the presence of an initial personal pronoun; though formally similar (or even identical), the two types should not be confused.

1) Inalienably possessed nouns: nimama "my mom," semolo "your hand," talae "his voice," etc.

Inalienably possessed nouns are special in that, although semantically the initial pronoun clearly makes them definite, formally they're not used with a particle considered a specifier (e.g. ka, a, ti, po, etc.). This makes them capable of occupying syntactic positions that other definite nouns cannot. For example,

1A. ni-imi se-mama
1SG-self 2SG-mom
"I am your mom"

2A. ni-imi ka cími-ci pe-tia
1SG-self DEF language-ist BEN-this
"I'm the linguist around here"

In the examples above, semama "your mom" is a special predicate and ka címici "the linguist" is an ordinary predicate, but in these structures there is no difference in usage. However, semama is permissible in verbal position where ka címici is not (I know I literally said the exact opposite yesterday about nimama, but things change fast around here). If we remove the dummy predicate imi in 1-2A, the clause with the special predicate remains grammatical (1B) but the one with the ordinary definite predicate does not (2B):

1B. ni-se-mama
1SG-2SG-mom
"I'm your mom"

2B. *ni ka cími-ci pe-tia
1SG DEF language-ist BEN-this
"*I'm the linguist around here"

Similarly, when the pronoun or predicate are focalized, the special predicate is fine (3A, 4A) where the ordinary definite predicate becomes marginal (3B, 4B):

3A. se-mama sa ni
2SG-mom FOC 1SG
"I'm your mom"

3B. ?ka címici sa ni
DEF linguist FOC 1SG
"?I'm the linguist"

4A. ni sa se-mama
1SG FOC 2SG-mom
"I'm your mom"

4B. ?ni sa ka címici
1SG FOC DEF linguist
"?I'm the linguist"

The second special structure is of much greater import, and is therefore also more controversial...

2) Verbs with pronominal subjects: nimama "I'm a mom," selóhani "you love me," tunuku "they're asleep," etc.

Here's the setup. We've known for many years that clauses with a pronominal subject omit the main clause marker i, thus

5. Ø ta-ma-viti po-vihu
MAIN.CL 3SG-CONT-cook GEN-mushroom
"she's cooking mushrooms"

6. ni-mama i ma-viti po-vihu
1SG-mom VB.CL CONT-cook GEN-mushroom
"my mom is cooking mushrooms"

(Note: Our awareness of the fact that the Ø main clause marker in 5 is actually outside of the verb word is a more complicated story that I'll get into later...)

If 5-6 become dependent clauses, objects of sesano "you said," let's say, the main clause marker i/Ø must change to the dependent clause marker ko. Similarly, if the dependent clause becomes a modifier of another predicate (i.e. a relative clause), the clause marker changes from i/Ø to u.

The difference in usage between the ordinary predicate maviti po vihu "is cooking mushrooms" and the special pronoun-initial predicate tamaviti po vihu "she's cooking mushrooms" is that clauses with pronominal subjects may omit the dependent clause markers ko and u, whereas ordinary clauses may not. This is the same pattern we see with the main clause marker i. Thus

7A. se-sano ko ta-ma-viti po vihu
2SG.say DEP.CL 3SG-CONT-cook GEN-mushroom
"you said that she's cooking mushrooms"

7B. se-sano Ø ta-ma-viti po vihu
2SG.say DEP.CL 3SG-CONT-cook GEN-mushroom
"you said she's cooking mushrooms"

8A. ka vihu u ta-ma-viti i cuti
DEF mushroom REL.CL 3SG-CONT-cook VB.CL lovely
"the mushrooms that she was cooking were lovely"

8B. ka vihu Ø ta-ma-viti i cuti
DEF mushroom REL.CL 3SG-CONT-cook VB.CL lovely
"the mushrooms she was cooking were lovely"

In 7-8, with special pronoun-initial predicates, the "B" version without a clause marker is a grammatical alternative to the "A" version with the marker. Below in 9-10, however, omitting the clause marker from an ordinary predicate yields an unacceptable result:

9A. se-sano ni-mama ko ma-viti po vihu
2SG-say 1SG-mom DEP.CL CONT-cook GEN-mushroom
"you said that my mom is cooking mushrooms"

9B. *se-sano ni-mama Ø ma-viti po vihu
2SG-say 1SG-mom DEP.CL CONT-cook GEN-mushroom
"*you said my mom is cooking mushrooms"

10A. ka vihu ni-mama u ma-viti i cuti
DEF mushroom 1SG-mom REL.CL CONT.cook VB.CL lovely
"the mushrooms that my mom was cooking were lovely"

10B. *ka vihu ni-mama Ø ma-viti i cuti
DEF mushroom 1SG-mom (REL.CL) CONT.cook VB.CL lovely
"*the mushrooms that my mom was cooking were lovely"

Again, I'm still feeling pretty cautious about all of this! I resisted for a long time because I'd never seen a language outside of English and Scandinavian that could do something like this, but recently I discovered with great surprise that Macedonian can as well with object clauses:

се надевам Ø таа ќе дојде
REFL hope.1S (COMP) she FUT come.3S
"I hope she'll come"

At this point I can't think of a strong reason not to allow them. I've been hemming and hawing since the idea first occurred to me in 2012, so we might as well go ahead and codify the phenomenon and try it out in practice. We'll see how it feels in actual usage!

In terms of the difference in feeling between the two alternatives, I'm vacillating between on the one hand imagining it to be perhaps slightly less formal to omit the clause marker (as in English), and on the other hand wondering whether the style is instead more elevated with more compact syntax. Time will tell.

Ai Se Sa Nimama? Interlinear - Part 1

This is the first half of the detailed interlinear gloss of the Koa translation of Are You My Mother promised in this introductory post. For fans of the work, note that this is the abridged edition, which happens to be the one I read to my girls when they were little:

Eastman, P. D. Are You My Mother? New York: Random House Children's Books, 1988.

Note that these glosses will have a slightly different structure than is usual for my examples: here I include an additional line, italicized, containing a literal translation into English to give a clearer sense of how Koa approaches the meaning expressed in the original text excerpted in each gloss.

Ai Se Sa Ni-mama?
QU 2SG FOC 1SG-mama
Is it you who is my mama?
"Are You My Mother?"

I went for Koa's "mama" word, mama, rather than directly translating the full, formal "mother" as émae. I guess this issue exists even in English, but I just couldn't imagine a newly-hatched baby bird asking where his émae was...that and ameémae for "mother bird" is hideous. I'm pretty sure Koa uses mama for those kinds of constructions wherever they exist: molamama "mama bear," manomama "mama shark," etc.

Also perhaps worth mentioning overtly that mama most naturally takes preposed, inalienable possession marking: thus nimama "my mama" rather than ka mámani "the mama of mine."

A ame-mama vo éki-ne pe ka múna-ta.
INDEF bird-mama PRES.CL sit-LOC BEN DEF egg-3SG
Behold, a mama bird sits there with respect to her egg
"A mother bird sat on her egg."

Obviously "behold" is too strong a word here. The thing is that English doesn't have a great way of saying this kind of thing overtly: we can't say ecco like Italian or jen like Esperanto. Another approach would be "look, there's a mama bird..." or "here we have a mama bird..."

Note that the tense of ékine is translated as present "sits there" rather than past as it appears in the English text. With statives like this Koa isn't as sensitive to tense as English, so though we could have sait siékine "sat there," it wouldn't be the most natural thing to say in Koa, especially in a children's book. See here for an old but in-depth discussion of default TAM status of different types of verbs.

Ka muna i hupa.
DEF egg MAIN.CL jump
The egg jumped
"The egg jumped."

Again no tense marking on the verb, because the unmarked tense of active (i.e. non-stative) verbs defaults to past.

Ni-ki-kíi-pa la ka ame-néne-ni ko suo! — ta-sano.
1SG-DEB-get-INDEF DAT DEF bird-baby-1SG NOM.CL eat 3SG-say
I must get stuff for my baby bird to eat! She said
"I must get something for my baby bird to eat! she said."

My first translation of "I must get something" was a direct calque: Nikikii hua, literally "I have to get something." Much later I realized that hua for "something" in this context too heavy-handed, and maybe not so idiomatic in Koa. I ended up using the "indefinite object" -pa suffix (discussion here), which deemphasizes the particular nature of the "something," just leaving the sense that she needs to engage in getting stuff generally.

I also went back and forth on whether "I must get" should be translated with a pronoun as nikikíipa, or if it was enough to say simply kikíipa with no pronoun. This is something Koa can do -- pronouns aren't positively required, just available for clarity -- so leaving it out would give the sense of "Gotta get something!" rather than "I must get something!" I did end up choosing the pronoun because the English text is a little formal too, but I continue to be torn.

Laa vo ta-la-poi.
therefore PRES.CL 3SG-DAT-away
So behold, she to-awayed
"So away she went."

Ne-sala ka pesa, ka muna i hupa.
LOC-inside DEF nest DEF egg VB.CL jump
In inside of the nest, the egg jumped
"Inside the nest, the egg jumped.

I fret about "inside," wondering if just ne ka pesa "in the nest" would have been entirely sufficient. I guess it's probably not of tremendous consequence.

Hupa, hupa, hupa. Ata...
jump jump jump reach
Jump, jump, jump. Reach...
"It jumped and jumped and jumped. Until..."

...a ame-nene vo tule la-pole!
INDEF bird-baby PRES.CL come DAT-outside
behold, a baby bird came to outside!
"...out came a baby bird!"

Ne-kea sa ni-mama? — ta-sano.
LOC-what FOC 1SG-mama 3SG say
What is my mama in? he said
"Where is my mother? he said."

Ta-na-si-náe-ta ne-naa.
3SG-NEG-ANT-see-3S LOC-nothing
He didn't see her in nothing
"He did not see her anywhere."

Nope, I still don't know what should happen with strings of negative words in negative sentences; for the moment I'm not issuing a verdict, and just treating them the same way that Spanish or Polish would. But maybe tanasináeta nehua "I don't see here anywhere" is better? I'm just not sure how to make this decision.

Vo ni-vi-háke-ta, — ta-sano.
PRES.CL 1SG-IMP-search-3SG 3SG-say
Behold, let me look for her, he said
"I will go and look for her, he said."

A more literal translation of "I will go and look for here would be" Vo nicuháketa, "Behold, I will/would look for her." But cu is more about futurity or possibility, not intention as I think is going on here. The volitive lu was another possibility, but again too strong on the wanting. I ended up settling on the imperative vi as a kind of personal exhortation.

O-sala ka pesa vo ta-lahe. La-lovo, la-lovo, la-lovo! Popo!
ABL-inside DEF nest PRES.CL 3SG-leave DAT-lower DAT-lower DAT-lower plop
From inside of the nest, behold he left. To lower to lower to lower! plop
"Out of the nest he went. Down, down, down! Plop!"

Ka ame-nene i na-voi ko lehu.
DEF bird-baby VB.CL NEG-able NOM.CL fly
The baby bird can't fly
"The baby bird could not fly."

Ala ko kave, ta-voi.
but NOM.CL walk 3SG-able
But walking, he is able
"But he could walk."

I'd just like to say I'm proud of this one. It's not anything like the English structure, but feels like it's getting at the meaning in a solidly Koa way.

He-sena ni-lu-luta ni-mama, — ta-sano.
TEMP-now 1SG-VOL-find 1SG-mama 3SG-say
At now I will/would find my mama, he said
"Now I will go and find my mother, he said."

Here I went with the volitive niluluta "I want/intend to find," but I have some anxiety that I messed it up and should again have chosen the imperative niviluta "let me find" instead. I'm counting on lu for as much "I mean to do this" energy as it can give me here.

Ai se sa ni-mama? — ka ame-nene i kusu a séne-to.
QU 2SG FOC 1SG-mama DEF bird-baby VB.CL ask INDEF cat-CHILD
Is it you who is my mama? the baby bird asked a kitten
"Are you my mother? the baby bird asked a kitten."

Ka séne-to i ie-nae, nae. Naa sa ta-sano.
DEF cat-CHILD VB.CL JUST-see see nothing FOC 3SG-say
The kitten just sees, sees. nothing is what it said
"The kitten just looked and looked. It did not say a thing."

Ie "just" is a more recent addition to Koa, inspired/informed by Nahuatl a couple years ago. The full form of "only," mono, would have been okay here too -- Ka séneto i nae, nae i mono -- but ie is quicker and I think a better fit to the intended semantics.

Ai se sa ni-mama? — ka ame-nene i kusu a móa-he.
QU 2SG FOC 1SG-mama DEF bird-baby VB.CL ask INDEF chicken-FEM
Is it you who is my mama? the baby bird asked a hen
"Are you my mother? the baby bird asked a hen."

Na, — ka móa-he i sano.
NEG DEF chicken-FEM VB.CL say
No, the hen said
"No, said the hen."

Ai se sa ni-mama? — ka ame-nene i kusu a kunu.
QU 2SG FOC 1SG-mama DEF bird-baby VB.CL ask INDEF dog
Is it you who is my mama? the baby bird asked a dog
"Are you my mother? the baby bird asked a dog."

Na se-mama sa ni.
NEG 2SG-mama FOC 1SG
It is not your mama that I am
"I am not your mother."

Kunu sa, — ka kunu i sano.
dog FOC DEF dog VB.CL say
It is a dog, the dog said
"I am a dog, said the dog."

Idiomatic pronoun omission again...

Ai se sa ni-mama? — ka ame-nene i kusu a léma-e.
QU 2SG FOC 1SG-mama DEF bird-baby VB.CL ask INDEF cow-FEM
Is it you who is my mama? the baby bird asked a cow
"Are you my mother? the baby bird asked a cow."

Ni mo-kea sa ni-cu-te-imi se-mama?
1SG SIM-what FOC 1SG-IRR-ABIL-self 2SG-mama
Me how is it that I could equal your mama?
"How could I be your mother?"

There's a bunch going on here. First of all, there's the very odd-to-English imi, which means "self" but also "identity" in a mathematical sense, an equals sign (more about that here). Theoretically I think nisemama could mean "I'm your mom," so one could pile that whole clause into a single verb phrase nicutesemama "I could be your mom, but...I'm a little uncertain and therefore uncomfortable with statements of identification as VP's in general, which may merit its own post to figure out. Basically, can definite phrases themselves be predicates? Is "she's the woman of my dreams" takamina ka móeni or does it have to be taimi ka mina ka móeni? The syntax of the former is breaking all kinds of rules and I don't at all care for it, so maybe this is an open and shut case. We had to use imi in the translation above, because semama, the predicate, is definite -- voilà. [Update a week later: on further reflection I came to a different conclusion with regard to verbs with pronominal subjects, but I'll let this stand as part of the process.]

...and that means that the title of the story could never have been Ai Se Nimama? like I was saying yesterday even if there had been no formal focus: it would have to have been Ai Seimi Nimama? (And the final title, Ai Se Sa Nimama?, is in fact a shortening of an underlying ai se sa nimama i imi? "is it you that my mama equals?")

Secondly, I think this may be the first recorded example of both topicalization and focalization in the same sentence. Ni out front there is establishing itself as topic: "Me? If we're talking about me, I can't be your mother..." That's followed by a typical fronted/focalized question word, and then the VP which still has a 1SG pronoun. The topicalization was necessary because from a pragmatically sensitive standpoint, the question isn't just "How could I be your mother?", it's "How could I be your mother?" As such, it's necessary to emphasize the topicality of "I." I'm glad a sentence like this happened to turn up, and rather chuffed with how well the Koa syntax ended up handling it.

ka léma-e i sano. — Léma-e sa ni.
DEF cow-FEM VB.CL say cow-FEM FOC 1SG
The cow said. Cow is what I am
"said the cow. I am a cow."

All for today -- second half coming soon. I'm not sure what's going on but I appear to have suddenly been hit by a full-speed Koa train; I must have spent four hours today writing, and probably that many yesterday too. I hope it sticks around for a while! I'm going to have to fire up my vocab database and refresh my memory, I'm getting pretty rusty on my newer vocabulary.

Conventions with quantities

We have a decision to make regarding writing conventions with pi. Should "some water" be…

1. nai pi anu "some of water"
2. náipi anu "some-of water"
3. nai pianu "some of-water"

Right off the bat, I was initially going to say that I feel like it should not be 3, because pianu would literally never appear without a preceding quantity word…but on reflection, I’m not sure why that should necessarily matter. Maybe more telling is the fact that pi can appear without a following noun:

ai sehalu náipi/nai pi?
QU 2S-want some-QUANT
do you want some of?
“do you want some [of this]?”

It is rather convenient to say that -pi is actually a suffix that turns its head into a kind of adverb of quantity. BUT…what if the quantity word is a complex NP? Let’s see…

[[sata lúa-ku tátu]-pi sého]-pi alu ne ka lise...
100 2-x10 3-QUANT bottle-QUANT beer LOC DEF wall
123 of bottle of beer on the wall
"123 bottles of beer on the wall…"

Aside: Wait. Is that how we do partitives?? Bottles of beer, slices of cake? I actually don’t know. Could we have a compound with no particle instead, like Turkish or Macedonian – seho alu? Gotta come back to this!

Anyway, that long number phrase is certainly rather ungainly. But outside of big numbers, I can’t come up with a complex NP as a quantity word that isn’t marginal at best, so it seems that the scope of the pi is just everything, moving backwards, until it hits another phrase with a preposed particle. I think that's okay.

So given that the phrase-final accentuation would be e.g. náipi and not naipí, it does rather seem to make sense to spell things that way. And also, didn't we just announce that -pi turns its host into a kind of pronominal? That might have been the place to start with this whole topic.

An interesting consequence to keep in mind about these theoretical emergent conventions, though, is that the dependent of such pronominal quantifiers would usually be without specifier. This otherwise never happens with NPs in syntax unless they're set off by an immediately preceding particle of some other kind: nekoto "at home," for example. This may have been the reason why pianu felt appealing: that and the fact that it felt like partitive marking in a language like Finnish (lasi vettä "glass of water") which is comfortable, and mirrors the dependent marking which is typical in Koa in other kinds of structures.

I suppose we could say that the -pi suffix kind of turns its host into a giant particle, which we write separately for visual clarity because lásipianu is clearly absurd.

Let's check one thing, though. Imagining a conversation while setting the table...

Ai ípu-pi anu ai lele sa se-si-halu me ka líla-suo?
QU cup-QUANT water QU milk FOC 2SG-ANT-want COM DEF evening-eat
Was it or cupsworth of water or milk that you wanted with the evening meal?
"Did you want a cup of water or milk with dinner?"

Anu, noi-a.
water ask-HON
Water, ask nicely
"Water, please."

I guess that makes sense. But it bugs me that in some other situations, repeating the particle to avoid a bare-stem noun seems to be required:

Ai ne-nomu ai ne-lovo sa se-sano ko se-si-nae ka sími-ni?
QU LOC-upper QU LOC-lower FOC 2SG-say NOM.CL 2SG-ANT-see DEF phone-1SG
Was it in upper or in lower that you said that you saw the phone of mine?
"Did you say you saw my phone upstairs or downstairs?"

...but

*ai ípu pi-anu ai pi-lele
QU cup QUANT-water QU QUANT-milk
Was it a cup of water or of milk
"was it a cup of water or milk"

feels...wrong. Or unnecessary. Or like it's missing the point. Because in the exactly parallel construction of the phone question above, both options actually would be without specifier:

Se-sano...ai Ø-anu ai Ø-lele sa se-halu?
2SG-say QU Ø-water QU Ø-milk FOC 2SG-want
You said...was it water or milk that you want?
"Did you say you wanted water or milk?"

And as I think about it, my example question of "Did you want a cup of water or milk..." was pretty contrived: you really would never say that. You might say "Do you want a glass or a bottle?", but "Do you want a glass or a bottle of beer?" is weird and stilted. These "quantity phrases" (QPs?) really are tight little units, actually quite different from locatives, which is another point in favor of marking them differently.

Also! I almost forgot that the quantified words actually can have specifiers of their own!

Ána-Ø-ni sého-pi ka álu-a-so ce-koa!
give-IMP-1SG bottle-QUANT DEF beer-HON-2PL SUP-good
Give me bottlesworth of the honorable beer of yours most good
"Give me a bottle of your finest ale!"

Compared to e.g. the infamous dependent clauses, this isn't an area I've really considered much through my career in philology, linguistics, or conlinguistics; as such, reaching an coherent understanding of how they would surface in Koa is taking some effort, and I feel clumsy at it. For the moment, though, I think these conclusions are correct, and fairly inescapable once we see that phrases like this are grammatical:

ána-a-ni líma-pi
give-HON-1SG 5-QUANT
kindly give me five of
"please give me five [of them]"

This really is an enclitic, not proclitic, particle.

Back to Turkish and Macedonian: after further reflection, as much as bare nouns may be the default in many languages, including probably the creoles on whose structures we most want to lean, I think we do need to continue to prescribe ípupi anu rather than just ipu anu for "glass of water." Since in N1 N2 structure N2 becomes adjectival and descriptive of N1, the least marked meaning of ipu anu is not "glass of water" but "water glass." This type of ambiguity would cause pandemonium in my household full of literal-thinking neurodivergent people, and is not present in e.g. Macedonian in which we have

чаша вода
glass water
"a glass of water"

but

чаша за вода
glass for water
"a water glass"

On the other hand, returning to the 123 bottles of beer on the wall, eliminating the second pi might sort of make sense given that the focus here is very much on the quantity rather than the nature of the container:

[sata lúa-ku tátu]-pi sého alu ne ka lise...
100 2-x10 3-QUANT bottle beer on the wall
123 of beer bottle on the wall
"123 bottles of beer on the wall…" OR "123 beer bottles on the wall"

The ambiguity feels reasonable in that context. I almost want to ask...though I wince as I do so...whether nested pi-phrases should in fact be forbidden. It may take a little time to answer this, as these kinds of phrases simply don't come up very often in the kind of writing I naturally tend to do. Maybe I need to construct some market haggling dialogs.

Incidentally, the above discussion of accentuation raises the question of whether postverbal locator particles should be written separately after all: ai sehalu mene mé "do you want to come with?", reflecting actual pronunciation, rather than ai sehalu méneme as we currently have the convention. *sigh*.

Learn your ACE's, kids

When I was assembling the concise Koa dictionary and phrasebook for my girls at the end of 2022 I had to finally decide what the names of the letters were going to be. This was something I'd never had cause to think about in a conlang before. I decided that there was an opportunity here for the letter names to help avoid the kinds of acoustic pitfalls we see in some other languages -- on the phone, for example, where in English we have to say "S as in 'Sam'" and "F as in 'Frank'" because there's so much less acoustic differentiation at that level of compression.

The vowels were easy -- just the sounds themselves -- and for consonants I was fine with /e/ as the base vowel. A few other choices:

• Velar and glottal sounds (i.e. K and H) have an /a/ vowel both for ease of pronunciation and to differentiate K from P
• M uses the /o/ vowel to distinguish from N
• C uses /i/ to distinguish from S in speakers who pronounce C as [ts]

The whole alphabet (cóepo "collection of letters" in Koa), then, ends up looking like this:

A  C  E  H  I  K  L  M  N  O  P  S  T  U  V
a  ci  e  ha  i  ka  le  mo  ne  o  pe  se  te  u  ve

This may not be perfect. I sometimes sweat a bit about whether P should have been po to unify the bilabials, and also whether L and V could be confused in speakers who pronounce V as [w]. Should V be vi? I guess there's a part of me that starts to like the aesthetics less for some reason as an increasing percentage of the letters deviate from the base vowel. Let's leave an even more acoustically distinguished set of letter names on the table, should it seem necessary someday:

a  ci  e  ha  i  ka  le  mo  ne  o  po  se  te  u  vi

In the mean time, though, I've been marching on with the original 2022 letter names.

Note that instead of "ABC," Koa speakers have an ACE! This would be pronounced acié as an acronym. (In fact, are acronyms always accented on the final syllable like strings of particles? ETM, the acronym for e tei motoa "et cetera," could be shortened to etemó in speech; OVN, oo vala ni "oh my god" becomes ovené "omg." Seems solid!) Anyway, letter names and acronyms probably need the naming article le when integrated into syntax: opi le ACE! "learn your ABC's!", Ai »mene« me le M sa toa, ai »nene« me le N? "Was that 'mene' with an M, or 'nene' with an N?"

Last October I also did my best to compose an alphabet song for Koa: not usually my province, but it was kind of fun. There was a bit of a challenge in getting a 15-letter alphabet to fit nicely into a song, and I ended up adding five extra syllables at the end -- vo ka cóepo "that's the alphabet!" -- to play nice with the meter. The melody is very simple and also pentatonic*, on the grounds that as a scale it's comfortable pretty much anywhere; the range is a 9th, which is a full step wider than I had intended, but one could move the vo note up a 5th to A if necessary.

At first I wasn't sure how much I loved it but at this point I'm pretty happy with it -- I think it's just the right amount of cute for a Kindergarten classroom.

*I have a little anxiety that everyone's first impression will be that I'm trying to make it sound stereotypically East Asian. That is very much not the intention: pentatonic scales are about as international as you can get musically.

Are you, in fact, my mother?

Around November 1st last year I finally managed to translate a substantive text into Koa for the first time. Well, "substantive" may be a little subjective. But children's books -- while superficially simple -- actually access core pragmatics at a pretty deep level, to the point that I found this specific exercise completely impossible even to begin when I first tried five years ago.

As just one example, the title of this work, "Are You My Mother?" by P. D. Eastman, could be rendered  into Koa in many different ways; if one were doing a mechanical translation exercise, this is probably how it would come out:

Ai Se-ni-mama?
QU 2SG-1SG-mama

But that translation would be completely pragmatically neutral, and as Allison pointed out to my consternation at the time, it doesn't really make sense in the context of the story. I tried to argue that, this perhaps not being the most astute baby bird we could ever imagine, he might genuinely have been asking the same sort of meaningless, vapid question over and over, and I even recorded audio of the whole story with that emphasis to try to prove it.

But she was right: as the baby bird wanders around making inquiries of various creatures, the information he seeks to get back is not equally weighted in terms of priority. For example, he's not looking for the answer "No, I'm your father." And if I imagined someone writing the story in Koa from scratch, with the intention of it being read to children, the bird's catchphrase would certainly not have been Ai Se Nimama?, but rather

Ai Se Sa Ni-mama?
QU 2SG FOC 1SG-mama

...in other words, "are you my mama? Is it you who is my mama?" And it was necessary to tune into that level of pragmatic sensitivity throughout the translation.

Another thing that frustrated this project in 2020 was that the narrative structure of young children's books tends to be very blow-by-blow: "Look, this is happening! Now look, that's happening!" Koa's presentative structures were in a prolonged state of infancy at that point, so it was very perplexing to go through the available options and find that none of them worked.

A good case -- and this illustrates why I got so impassably frustrated so quickly -- is the very first line of the story, "A mother bird sat on her egg." Again the mechanical translation would be

A ame-mother i eki pe ka múna-ta
INDEF bird-mother MAIN.CL sit BEN DEF egg-3SG

...but I had the strong sense that something presentative was going on here -- like Polish Szła dzieweczka do laseczka "Went a girl into the forest..." -- and I kept trying to make focus do a job it wasn't intended to:

*Eki sa a ame-mama...
sit FOC INDEF bird-mother

Despite all my hopes, that doesn't in fact mean "Hey check it out, so this mother bird was sitting on her egg, [and then the craziest thing happened!]" It means the nonsensical "It's sitting that a mother bird was doing on her egg," especially silly dropped out of the blue onto the first page of the story. What I needed was this:

A ame-mama vo éki-ne pe ka múna-ta
INDEF bird-mother PRES.CL sit-LOC BEN DEF egg-3SG
lit. "Behold, a mother bird sits there on her egg"

But this kind of structure where vo replaces i in presentative contexts didn't develop until fall 2024, and even now deciding when to use it feels like it requires a thrilling/terrifying amount of trust in my own intuition. This was also and especially the case with less textbooky ways of expressing desired meaning, terribly important in a children's book that isn't intended to sound like a stuffy drawing room. This kind of vernacular Koa style had simply never existed before, and I couldn't derive it from first principles: I just had to invent.

I like my translation, though it's not perfect -- there are still places where my choices feel too heavily influenced by English structure, semantics or idiom -- but I feel like it fulfills its purpose in showing what a Koa children's book might look and sound like. At some point perfection may not really exist, and one could tweak ad infinitum without really improving anything.

So then, below I present to you my translation, Ai Se Sa Nimama?* Here I'll reproduce only the Koa; in order not to make this post unmanageably long, this will be the first of a series, and I'll follow this up with interlinear and commentary.

Ai Se Sa Nimama?
ci le P. D. Eastman, palili ci le Iúliki Líkololu

A amemama vo ékine pe ka múnata.
Ka muna i hupa.
— Nikikíipa la ka amenéneni ko suo! — tasano.
Laa vo talapoi.

Nesala ka pesa, ka muna i hupa.
      Hupa,
          hupa,
               hupa.
Ata...
     ...a amenene vo tule lapole!

— Nekea sa nimama? — tasano.
Tanasináeta nenaa.
— Vo niviháketa, — tasano.

Osala ka pesa vo talahe.
Lalovo, lalovo, lalovo!
Popo!

Ka amenene i navoi ko lehu.
Ala ko kave, tavoi.
— Hesena niluluta nimama, — tasano.

— Ai se sa nimama? — ka amenene i kusu a séneto.
Ka séneto i ienae, nae. Naa sa tasano.

— Ai se sa nimama? — ka amenene i kusu a móahe.
— Na, — ka móahe i sano.

— Ai se sa nimama? — ka amenene i kusu a kunu.
— Na semama sa ni. Kunu sa, — ka kunu i sano.

— Ai se sa nimama? — ka amenene i kusu a lémae.
— Ni mokea sa nicuteimi semama? — ka lémae i sano, — Lémae sa ni.

Ka amenene i luvu ve cumeti.
Ka séneto e ka móahe i natamama.
Ka kunu e ka lémae i natamama.

— Nikumemama, — ka amenene i sano, — niilo, iakitai.
» Nilulútata. Niculútata. IACÚ!

Iehetoa ka amenene vo nae a mea iso.
— SEKA sa nimama! — tasano.
Ka mea iso i sano — HUAUU!

— Ii, na! — ka amenene i sano, — Na nimama sa se.
» Huauu mupato sa se!
Ka Huauu i nose ka amenene i lanomu, lanomu, lanomu.

Sii hua i osu.
Ka Huauu vo iepane ka amenene i lai la ka puu.
Ka amenene i honekoto!

Iehetoa ka amemama vo lai.
— Niilo ka imi SEKA, — ka amenene i sano.
» Na séneto, na móahe, na kunu sa se.
» Na lémae, na Huauu sa.
» Ame sa se, e mama nika sa!

*What about capitalization conventions??? Another topic for the inexhaustible list...

Temporal conjunctions when I still haven't forgotten again

At some point in the early 2010's, while reading a grammar, I made a discovery that I found fascinating and considered for Koa. I can't remember what language it was, and I'm not sure whether I consciously decided against this construction or just forgot about it for a decade. If I didn't write about it then, I should have; there is a possibility that this post may even be an unintentional duplicate.

What we're talking about here are temporal relationships between clauses -- or perhaps more accurately, relationships of subsequence. Koa now has its own particular way of doing this via the verbs koe "precede" and hala "follow," thus:

koe ka súo-nose
precede the sun.rise
"before sunrise"

hala ka ívo-munu
follow the night-middle
"after midnight"

One could make the mistake of thinking that koe and hala were prepositions of some kind, but their verbal force becomes clear when they govern clausal relationships, in which case they're integrated into syntax as serial verbs:

ni-lopu ka ámu-suo i koe ko se-vene
1SG-finish DEF morning-eat MAIN.CL precede NOM.CL 2SG-awaken
"I finished my breakfast before you woke up"

vi-komo ka ívo-puku i hala ko se-mie-nio
IMP-put.on DEF night-suit MAIN.CL follow NOM.CL 2SG-wash-teeth
"Put on your pajamas after you brush your teeth"

It took me a lot of years to figure this out -- I think in the neighborhood of 2021 -- and I've always been kind of proud of it. I don't intend to make a change, but I wonder if that anonymous language (could it have been Hawai‘ian??) might offer an option for a second strategy for clausal relationships.

This language, as I recall, entirely lacked words for "before" and "after" in this context, relying instead on the conjunction "when." Thus "before X" is "when X hadn't happened yet," and "after X" is "when X had already happened." Translated into Koa, this looks like

ni-lopu ka ámu-suo he ko se-ca-na-vene
1SG-finish DEF morning-eat TEMP NOM.CL 2SG-CONT-NEG-awaken
"I finished my breakfast before you woke up," lit. "I finished my breakfast when you still weren't awake"

vi-komo ka ívo-puku he ko se-io-mie-nio
IMP-put.on DEF night-suit TEMP NOM.CL 2SG-TRANS-wash-teeth
"Put on your pajamas after you brush your teeth," lit. "Put on your pajamas when you've already brushed your teeth"

...or should the "before" structure be with naio "not-already" = "not yet," rather than cana "still not"?

ni-lopu ka ámu-suo he ko se-na-io-vene
1SG-finish DEF morning-eat TEMP NOM.CL 2SG-CONT-NEG-awaken
"I finished my breakfast before you woke up," lit. "I finished my breakfast when you still weren't awake"

Either way, I certainly see no reason not to import this structure as an option -- there is such a beautiful, logical simplicity to it that feels like it suits Koa's style, as well as brilliantly following the charter in avoiding IE calques.

It occurs to me that there would also be a longer form, if more wordiness is desirable, using kei "extent." Usually we see this in e.g.

he-kei ko se-nuku
TEMP-extent NOM.CL 2SG-sleep
"while you were sleeping"

Thus in addition to "when X hadn't happened yet," we could have "while X hadn't happened yet":

ni-lopu ka ámu-suo he-kei ko se-ca-na-vene
1SG-finish DEF morning-eat TEMP-extent NOM.CL 2SG-CONT-NEG-awaken
"I finished my breakfast before you woke up," lit. "I finished my breakfast while you still weren't awake"

vi-komo ka ívo-puku he-kei ko se-io-mie-nio
IMP-put.on DEF night-suit TEMP-extent NOM.CL 2SG-TRANS-wash-teeth
"Put on your pajamas after you brush your teeth," lit. "Put on your pajamas while you've already brushed your teeth"

I'm relieved to have finally remembered this, whatever muse it was who recalled it to me last night -- while I was trying to figure out these constructions a few years ago I recalled that there was something I'd once come across that I'd liked, but had no idea what it was.

Incidentally, I snuck in "MAIN.CL" and "NOM.CL" as the glosses of i and ko above, which are anticipating a pretty major topic in subordinate clauses (yes, yet again: buckle up) that I've been sitting on for the last several months. Still waiting for the muses to help me with that one...