[A Query from Reddit] Could someone grammar check this poem?

[shanoxilt]

I wrote it for a school assignment. The teacher probably doesn't care if it's proper Lekh Dothraki, but I do. Here's what I have. (It's a [villanelle](http://en.wikipedia.org/wiki/Villanelle))

Ven dothrak, arakh anni affisak.

Az hranni vithale hatif Chafis

Qoyi arakh anni eqorasak.


Hrazef ann' najahak esittesak,

Fin dothraki qafa mae Qafis?

Ven dothrak, arakh anni affisak.


Fin vilajero nakhalot vorsakh?

Vilajero finne anh' najah dis!

Qoyi arakh anni eqorasak.


Anha vi Havazh Dothraki dothrak

Ma filkaki azhi azho kemis.

Ven dothrak, arakh anni affisak.


Anha nhara dozgoon filkakoon riss'k

Jahak anhi vitisheratneak kis

qoyi arakh anni eqorasak.


Anha she hrazef anhi ahhasak

arakh anhi, affin 'jjin "Sorohis!"

Ven dothrak, arakh anni affisak.

Qoyi  Arakh anni eqorasak.

[Qvaak]

Oh my! That's a grand order attempt. My hat's off. I'll answers as if to OP.
Grammatically, and considering poetic freedoms, the language is pretty good looking. I'll try to take my usual pedantical/meditative long-form walk through.

It seems you're going for pentameter, but not worrying about following a fixed pattern for stresses or heavy syllables. I read this is normal for Villanelles. Lemme take a look, how the prosody (if that's the word) looks, anyway. I'll mark heavy syllables with an underline and stress with bold. Heavy syllables are mostly stressed, so there'll be a plenty of bold underlined syllebles.

Ven dothrak, arakh anni affisak.
Az hranni vithale hatif Chafis
Qoyi arakh anni eqorasak.

Hrazef ann' najahak esittesak,
Fin dothraki qafa mae Qafis?
Ven dothrak, arakh anni affisak.

Fin vilajero nakhalot vorsakh?
Vilajero finne anh' najah dis!
Qoyi arakh anni eqorasak.

Anha vi Havazh Dothraki dothrak
Ma filkaki azhi azho kemis.
Ven dothrak, arakh anni affisak.

Anha nhara dozgoon filkakoon riss'k {12 syllables}
Jahak anni vitisheratneak kis      {11 syllables; anhi -> anni}
Qoyi arakh anni eqorasak.

Anha she hrazef anni ahhasak      {anhi -> anni}
arakh anni, affin 'jjin "Sorohis!"      {anhi -> anni}
Ven dothrak, arakh anni affisak.
Qoyi arakh anni eqorasak.

Corrected some most obvious missprint-looking "anhi -> anni" typos where mentioned, but largely left the text 'as is'.

----

Ven dothrak, arakh anni affisak.

If I get this right, it's roughly: like a Dothrak, arakh of mine {I} clean.

A solid start. It's breaks Dothraki's syntactic rules, but only in ways I'd expect from a poem.
- Compared to English, Dothraki is less into adverbial stuff thrown to the beginning of a sentence. It's not a definite no-no, but still more markedly unusual word order.
- Dothraki is a non-pro-drop language, even when it marks person in the verb. So dropping the subject anha is definitely taking a poetic freedom.
- Putting the object before the verb is pretty normal thing, but still against the basic SVO word order Dothraki follows.

So completely flat version would be
Anha affisak arakh anni ven dothrak. ~ I clean my arakh like a Dothrak.

Az hranni vithale hatif Chafis

If I get this right, it's roughly: Blades of grass flow facing a wind.

Still not bad, but things are a little less solid. Some stuff is just unsure because we're operating without full knowledge; some stuff looks wrong. I don't think poetic licence is much help here.
- OK, so our knowledge on some of the words here is a bit lacking. We don't have animacy for az. Az is probably inanimate, and thus makes no difference between plural and singular, so I went with plural in my translation as per the context. But if az happens to be animate, then we must go with singular interpretation as plural would be azi.
- Also, there is no supporing past singular conjugation for vithalat. It's pretty safe bet the stem is vitha and the suffix /-lat/, so the present third person should be vithae. If the stem would happen to be vithal, then the verb conjugation would mark the difference between plural and singular, no matter the animacy of az. Plural would be vithali and singular vithala. Vithale is wrong which ever way I look at it, but all the alternatives follow the same three light syllable pattern (Dothraki never melds vowels into diphthongs, so even vithae is spelled vi-tha-e), so I don't think that's a big problem.
- Prepositions always determine the case of the following noun. No known preposition allows the following word to be in accusative. Here the correct case would be genitive, chafi. That of course fails the rhyme, so something would need to change. Vithalat seems intransitive (and if it were not, the line would be just more broken), so getting an accusative there takes some serious changes.

Anyway, the grammar-corrected version would be something like
Az hranni vithae hatif chafi

Qoyi arakh anni eqorasak.

This is perplexing me a bit. A translation closest to the words would be I let go of my bloody arakh, but that is a strange line for the poem. Or maybe not? Maybe the idea that blood never cleans off the blade is precisely the point of the poem.
- Eqorasalat is very much a "letting go" word. If blood were the subject, it might make sense to say, with a bit of a metaphore, that blood is letting go of the sword. As the verb is in first person singular, it seems the "I" subject is holding something (in hand or metaphorically) and now letting go of it.
- If arakh is the object and qoyi is meant to be an adjective modifying it, qoyi should follow arakh. I think shuffling this word order is stretching poetic licence and not needed.
- Qoyi is not an adjective anyway. The genitive marks possessor, so arakh qoyi should mean "arakh of blood" (or "blood's arakh"). I guess poem might make do with double possessor like arakh qoyi anni - "arakh of blood of mine", but even at best that's pretty iffy syntax.
- This is of course again pro-dropping and OV, but these things are still IMO good form poetic freedom.

Flattened attempt on making sense of this line:
Anha eqorasak arakh saqoya anni. ~ I let go of my blood-covered arakh.

Hrazef ann' najahak esittesak,

Again I'm unsure of the intented meaning, but maybe the easiest translation attempt would be taking this as a complex zero-copula sentence: Horse of mine is a winner is a braggart. I have a feeling this is not what's meant here.

- Part of the problem is probably that pro-dropping that's going on. Probably in this sentence too at least one of the /-ak/ words is meant to be a verb, but it's hard to grasp how the syntax is meant to flow. If this is meant to be a curious two-way zero-copula sentence, then the earlier pro-dropping is setting up for a misunderstanding here.
- Could it be that the sentence is meant to continue on the next line?
- If esittesak is meant to be a verb, it's actually likely it does not use accusative as an object case (though it's possible that it might). "I brag about my horse" is probably Anha esittesak hrazefi anni.
- If najahak is meant to be a noun and esittesak a verb, I think the sentence falls apart pretty bad.
- Ann' is, in a way, pretty good shortening scheme. The nominative anha breaks down a-nha, so a heavy syllable an kinda informs that we have an accusative or genitive case. I'm less convinced that the lenghtened word ending 'n' works for Dothraki phonotactics. There's always a syllable boundary in the middle of geminate. If you remove a syllable from anni, you'll probably end with an'. Not that it's impossible to pronounce a lenghtened 'n'; we're talking completely speculative here. You might even get a syllabic 'n'

I think I need to know more about what is intented before making any corrected or flattened or what-have-you attempts on this one.

Fin dothraki qafa mae Qafis?

If I get this right, it's roughly: Who dothraki ask about him/her a question?

- In my analysis fin is a modifier and dothraki is plural subject, so qafa should be conjugated in plural, qafi.
- Qaf is an inanimate noun, so the accusative should be qaf, not qafis. Maybe you might try a poetic licence and wordsmith an animate extension of qaf, but that's very very iffy.
- I took mae to be topical genitive, but it's likely the intented meaning was "ask a question from him". I don't know how qaf assigns cases/prepositions, but I don't think it likely that both "from who the question is asked" and "what is asked" are both in accusative. I think accusative is always reserved for one specific function and everything else is dealt with prepositions and cases. A sentence "I asked her a question about you." would most likely be Anha qaf qaf yeri moon, though to my intuition allative works almost as well as ablative, and if there is any preferred order to complimentary objects (or if the object in accusative has a definite first place), I can't tell.

So perhaps the meaning would be more as intented as Fin dothraki qafi qaf moon. Dunno.

Fin vilajero nakhalot vorsakh?

Perhaps this might be meant to be "What battle is to end the flame?" - or maybe it's just feeling the easiest interpretation.

- Nakhalot is surely a spelling-erred nakholat.
- Nakholat is an intransitive word, "to end" in a sense "the road ends there" or perhaps "my life ended that day". The word being in infinitive and all, it's hard to tie the vorsakh to the syntax. Most likely nakholat should be changed to annakholat so that there would be space for subject ("battle" or "he") and object ("flame").
- Though perhaps the infinite is left there by an accident? "What battle finishes the flame" seems pretty sensible line. With the infinitive we're left with either the grammatically iffy "What battle is to end the flame?" or the contextually iffy "Who started to battle to end the flame?" I'm sure somehow a infinite verb construct like this would work, just not sure how.
- This is of course also pretty impure rhyme, as [kh] versus [k] is comparable to [t] versus [th] in English - close enough to sell the rhyme, probably, but definitely dirty.

Dunno. Perhaps Fin vilajero annakholat vorsakh? would work (grammatically). Perhaps it should be Fin vilajero haji annakholat vorsakh?, or perhaps the syntax is actually just too English and should be thoroughly reformed.

Vilajero finne anh' najah dis!

I'm guessing roughly: A fight where I am a plain winner!

- Finne probably works, even though I don't think I've ever seen it (or related word affin) tied to a noun phrase like a relative pronoun introducing a relative clause.
- Relative clauses follow VSO word order, and that's kinda their thing, so it might be best putting a verb first in the sub-clause, maybe?
- najah and dis are both used in adjective form. That is a stretch on najah (zero copula adjectives are for universal qualities and here you're going for a definite situational quality) and won't IMO work at all in dis. As an adjective dis inevitably ties to anha and thus should be interpreted as a parallel quality: you are victorious and simple - which isn't terribly flattering.
- Making the sentence regular (and to make it work better in the context), you'd need a stative verb, najahat, conjugated najahak. Of course if past tense was meant to be, then najah is actually a proper verb form. The difference to the noun najahak would of course disappear, but that's how it goes.
- Dis would probably work best as disse, ie. in an adverb form. I think it would miss a bit the intented meaning, but I dont know how you'd exaclty do it closer to how I read the intention.
- Would anh' work in Dothraki phonotactics? I think /-nh/ ending is a bit outside the regular options, but could probably be managed.

So perhaps grammatical clean up might give us Vilajero finne najahak anha disse!

Anha vi Havazh Dothraki dothrak

roughly: I through Dothraki Sea ride

- Putting an adverbial phrase in the middle is anything but normal, but should be perfectly fine poetically.
- The preposition vi assigns allative, so Havazh should be Havazhaan.

Flatly put: Anha dothrak vi Havazhaan Dothraki.

Ma filkaki azhi azho kemis.

Roughly: And cowards give dried fig gifts.

- I'd like to see more ma-pairs. While this singular use is passable, it always feels like non-Dothraki way.
- Azho is in accusative azh.
- Azho kemis looks like an attempt for a word compound, and I don't think that works. Making up a proper compound, azhkemisi, might work better, and in accusative it would be azhkemis, so that would be almost perfect. Otherwise you'll need to tie kemis to the syntax. For "gifts of dried figs" azh kemisi might work. Or you might try to sell one or another word as an adjective: azh kemisa ~ "dried-figgy gifts" or kemis azha ~ "giftous dried figs"

So, yeah. Here my grammar correction proposition would be Ma filkaki azhi azhkemis. ...only a syllable off from the original.

Anha nhara dozgoon filkakoon riss'k

Roughly: I the head of an enemy of a coward cut.

- Nhare is an animate noun, and as such in in accusative nharees.
- It's wonderful to see someone remember to use inalienable possession (I usually forget). As far as I understand this, though, dozgoon and filkakoon are meant to be parallel, both tied to nharees. How would this work? I guess fine, but I'd at least use puncturation to clear the syntax a bit. There isn't much to the Dothraki puncturation, illiterate barbarians and all, so this is largely up to your judgement.
- Riss'k looks super dubious shortening to me. Here I'm quite certain the gemination of 's' would be lost - there's no way you'll otherwise sell it as one syllable. And even if we'd buy this "ris'k", it would be pronounced pretty much exactly like english word "risk" and would rhyme with /-sak/ pretty much as well as English "risk" rhymes with "suck".
- Dothraki has no long vowels, it just has vowels wihout consonats in the middle. Dozgoon breaks down doz-go-on, and filkakoon breaks down as fil-ka-ko-on. This means you have too many syllables.
- The SOV word order is more unusual than the most unusual word orders you've used, but I guess that's still no problem.

Dull and grammaticalized: Anha rissak nhare dozgoon, filkakoon.

Jahak anhi vitisheratneak kis

Roughly: braid of mine long-grow tries

- Anhi should, of course, be anni.
- Are you really trying to form a compound verb or is there just a space missing? There must be some compound verbs in Dothraki, but I'm not sure if we've ever seen one or how it'd work - and I'm pretty sure there won't be an infinite suffix in the middle of the compound. Let's just take this as a missing space, yes.
- Kis is one of them curious Dothraki particles that are not really verbs, but still work a bit like English auxiliary verbs. It's probably a bit too much poetic freedom to throw that around. And it won't do to sell kis as a verb, leaving vitisherat in infinite.
- How to bring neak to the syntax? Now there's a question. "Long" is used as an adverb there, and of course our Dothraki word is an adjective. I have very little idea, how that would be solved. The same no-fuzz flexibility as in English should not work.

Perhaps the best grammatically promising version I can come up with would be Jahak anni kis vitishera athneakaan. ~ "My braid tries to grow to longness."

Anha she hrazef anhi ahhasak

Roughly (or quite exactly): I on horse of mine sharpen...

- Anhi should, of course, be again anni.

...

arakh anhi, affin 'jjin "Sorohis!"

Roughly: ...arakh of mine, when now "Halt!"

- Anhi should, of course, be again anni.
- How would 'jjin be anything but jin? I guess it kinda sorta might, but I'm not convinced.
- "When now" makes remarkably much sense in English. It might well work for the same effect in Dothraki. You bring the storytelling from vague whenever into virtual 'now', which is still actually whenever, but feels more immediate, zooming to moment. I think there's a whole lot of back framework of storytelling tradition in work here, so it's not a certain to translate sensibly.

----

Anyway, I'm sure I missed some stuff and went way off in my interpretation, but I hope this is a good commentary nonetheless. Come here to Dothraki forums (or should I come there?) and tell what you think, ask questions. I reiterate: there aren't many this ambitious Dothraki writing projects. You are awesome.

Edit: I managed to ask David about az and vithalat. No surprises. Az is inanimate and /-lat/ is the suffix on vithalat. And while I'm at it, let me also apologize some more. My commentaries tend to erode in quality the longer they go. It's just a big task (even if mostly a pleasant one) to check and comment through a long text like this. I tend to get a bit exhausted. I may also grow unnecessarily harsh... or be unnecessaruly harsh to begin with.

[Havazhyol]

poem Qvaaked ^^.

I envy you, the fact that you can use or are asked to use or study dothraki in school.