The Treasure of the Wastes

So back when I announced the annual Dothraki haiku contest, I thought it would be fun to see if anyone could do something with High Valyrian. Then this thing basically became all about High Valyrian. Yikes!

All right, so let’s deal with that first. Since Japanese originally used mora counting for its haiku, I thought it would be cool to do that for High Valyrian, since it also had long and short vowels. Clearly I did not think this through. High Valyrian words are way too big for a haiku. The form just doesn’t make sense. If anything, one should only pay attention to syllables. That might make haiku possible for High Valyrian; it just makes the practice a little less interesting. Haiku seem to work very well for Dothraki, but it’s just not going to work for High Valyrian.

In discussing this with my wife, she had an idea: What about limericks? Kind of sillier, but I think it could work, because three of the lines are usually quite longer. I think of the classic limerick as being 9-9-5-5-9 (syllable count) with an AABBA rhyme scheme. However true limericks often will have more syllables than that (or fewer, as the case may be), which I think would suit High Valyrian quite well.

So this is what I want to try. Those who were trying to do High Valyrian haiku, try a limerick. Give it an AABBA rhyme scheme and try to make the B lines shorter, but there will be no strict syllable counts. We all know what limericks sound like, so you should try to make it sound like that. Use the heavy syllables to your advantage. If you want, you can have long vowels count for more than one syllable, if it makes sense in your schema, but you’ll be in charge of coming up with that schema (the poem itself will, essentially, argue for a meter). Anyway, once you’ve tried it out, if you think it’s doable, I’ll announce a separate High Valyrian limerick contest at some point in time later on. You’ll have more time than the Dothraki haiku contest, since the form is longer and a bit more complicated. Let me know what you think in the comments.

Now, unfortunately because of new work that has come up, I’m not going to be able to review as many submissions as I wanted to. If you’re new here, go check out the comments on the announcement post, because there’s some great material there. For this post, first let’s look at Joel W’s Valyrian haiku:

Māzīlzi
ōrbar ñuqīr
jelmyssi

The intended meaning is “Smoke and ash will come with the winds”. Very elegant! I like the use of the coordination strategy to stretch out the second line (i.e. lengthening the last vowel of ñuqir). Very well done! Technically it should be jelmȳssi, but that doesn’t change the mora count. Also the second line is one mora short if you discount codas. If you count the final r of ōrbar, though, it works, so I will count it. I like it! This is probably my favorite High Valyrian offering of the bunch.

This is another good one from Zhalio:

Gō ropatas
Valyria, yn vēzos
josīmonis.

That is “Valyria fell before, but the sun continues to rise”. That’s the literal translation. Discounting final consonants, that does work. Nice job! In the comments, Mad Latinist suggested that it should be ropetas? It should not: ropatas is correct. This is because the stem is ropa-, not rop-. Easy mistake to make, though.

Honorary mentions go out to Zhalio and Joel W who tried to translate the Pater Noster, despite lacking most of the necessary words! You can see Zhalio’s translation here, and Joel W’s translation here. I don’t have time to review them, but will look into coining some of that vocabulary.

And before leaving Valyrian, I definitely want to mention Mad Latinist’s opening to the Dæneryd, which sounds like an awesome subject for an epic poem. Mad Latinist wrote up this post on his LiveJournal discussing and presenting two lines he wrote in epic Graeco-Roman hexameter in High Valyrian. The form is, indeed, much better suited to High Valyrian than a haiku is, and the result is incredible. The lines are here:

Ābre se zaldrīzī bone ivāedan hen Essot jitte
ēlī Pento se Dothrakoti Embraro rȳ ondoso vējo…

He didn’t attempt a fluent English translation, but I will: “Dragons and that woman I sing, from Essos sent / First through Pentos, then the Dothraki sea, by the hand of fate…” Sounds awesome. Sounds like something that should be attempted after the series has completed (I promise High Valyrian will have enough words to handle it at that point). It’d require GRRM to sign off on it, but wouldn’t that be awesome? After all, all the old myths are told and retold; they’re not made up whole cloth. Daenerys would be an outstanding subject for an epic poem (or I’m assuming. I too don’t know how it ends). You can hear Mad Latinist’s friend pronounce it here (good reading!).

If there is one quibble I’d have, it’s with ivāedan. Since the oblique applicative is being used, it should be standing in for some sort of adpositional phrase which is appropriate to the oblique applicative. Unfortunately if you want to say something “about” something, the postposition you’d use is , which is technically a locative postposition, so it should probably be uvāedan (and the cases would have to change accordingly). But maybe you could get away with ivāedan.

Okay, enough Valyrian. On to the Dothraki!

Let’s start with Hrakkar’s:

Me zheanalat
Chaf hol she mae noreth
Me davra hrazef

The intended meaning is, “She is beautiful, wind blew on her hair, she is a good horse”. Of course, “she” is just a translation choice; it could be “he” or “she” in Dothraki. There are a couple of things that need fixing. First, zheanalat is the infinitive; it should be zheanae. Next, the possessor comes after the thing it possesses, so it would be noreth mae, but also since “hair” is inalienably possessed, it should be moon, or just not expressed. I might also have used vi instead of she for “through her hair”. So it would be Chaf hol vi norethaan, which would indeed be seven syllables. It’s debatable, though. She is supposed to serve as the locative preposition that “makes sense”, so it could work here. In the last line, it should be hrazef davra (noun-adjective word order), but otherwise this is pretty good! I like it!

Here’s The Majesty’s submission:

Athkisar notat
Lirof mra lekhofaan
Noreth nem jesa

I think the intended meaning is “Trying to turn a great piece of writing into a great language is hair being pulled”. I’ll give you an A for effort here, The Majesty, but this doesn’t really work. Neither kis nor notat can be used in that way. But you did get the message across! Yeah, I gave up on trying to translate the prologue for the first book after sentence one.

Next we have Zhalio’s entry:

Vezh ahajana
Vosma mra noreth anni
Ale ayena.

A good translation of this is “The stallion is stronger, but my hair has more bells in it”. A nice one! Two things are standing in the way of this one being great, though. The first is that “hair” is inalienably possessed, so it should be noreth anhoon. That’d put it one syllable over, but you could do vosm’mra (it is poetry, after all). Second, adjectives follow nouns, so it should be ayena ale. I could see how you’d get a determiner reading for this, though. If you were to put it in front, I would say it has to be ale ayeni—maybe alikh ayeni, “a surplus of bells”. The content is terrific, though, and I really like the use of mra here as “have”. Ordinarily it’s just mra qora which is kind of used as “to have”, but it makes sense to use it with noreth here. Great job!

Now we move to Qvaak. This year Qvaak did a cycle of poems switching between High Valyrian and Dothraki. It was a bold attempt! You can see the whole thing here. I’ll only discuss the two Dothraki haiku here.

First, it begins with this:

Mra qevir noreth
fenoe hatifaan;
azho qosari.

My translation is, “In the forest, hair clings to one’s face: a gift of the spiders.” My only complaint is with the punctuation: I would’ve used a colon rather than a semi-colon. Otherwise, this is good Dothraki! Excellent choice of adding the inchoative -o suffix to fenat (an invention of Qvaak’s; wholly appropriate). I might also have said azho qosaroon, given where it comes from. Otherwise, very good—and certainly a feeling we all know, if you’ve ever run into a spiderweb.

But, of course, no poem with spiders in it is going to win the Mawizzi Virzeth! No, that honor goes to this haiku:

Mas athasari

tolorro mahrazhoa

finis adakh me.

My translation is “The treasure of the wastes is the bones of men whom it has devoured.” Qvaak translated this as “desert”, but there actually is a Dothraki word for “desert”: zelatha (inanimate, Class A). I think it’s also the mark of a good poem when the translation doesn’t do the original justice, and I think that is the case with this poem. I like that on account of the relative clause the subject is forced to go last. Gives it kind of a stinger at the end. Also, if you wanted to switch to “desert”, it’d be an easy fix: Just change it to masar zelathi. I like it the way it is, though. Very nice!

Here’s my rendition of it:

And, yes, this means that, three years running, the Mawizzi Virzeth goes to the evidently unbeatable Qvaak.

You’re a machine, Qvaak! A soulful, artistic machine. Hajas, zhey Qvaak!

The Red Rabbit, 2014, awarded to Qvaak!

Thank you to all who submitted haikus this year, and thank you to all those who ventured into Valyrian territory. Let me know what you think about my idea and we’ll see about starting up another competition. A different option might be two do a hexametrical couplet like Mad Latinist did, but I thought this might be too difficult. Thoughts? I’m open to either. Mad Latinist’s was outstanding.

Posted on January 31, 2014, in Community and tagged , , , . Bookmark the permalink. 51 Comments.

  1. Rytsas! Very excited to see this. Sorry you didn’t have to time to go into the depth you would have liked (and, obviously, so would we.) It would have been especially cool to see what you thought of the various HV coinings, and dangerous grammatical experiments ;)

    Yeah, I gave up on trying to translate the prologue for the first book after sentence one.

    At first I took this to mean that the The Majesty’s submission was somehow referring to something GRRM said in some sort of forward to A Game Of Thrones (since I have not yet read the books)… but I was probably on the wrong track there. Did you just mean that you tried to translate the opening scene into Dothraki, and found it was like hair being pulled?

    I’m glad you enjoyed my hexameters. That was a lot of fun, when it wasn’t totally frustrating. Now you know why I was asking all those questions about (hen) lento(t) ((hen)u)jitta—the first draft I got to scan said hen Esso henujitte, but then I realized that hen does not take a genitive (what do I think this is, Greek?), and if I put it in a dative/locative the scansion would fall apart. That, and you told me that double-hen structure wouldn’t work. In the end the only way to salvage the line was to accept a spondaic fifth foot (which, as I explained in the entry is less than ideal, especially in the very first line of an imaginary epic.)

    My entry was long, though, and you missed a couple minor details because your time was so short this week:

    He didn’t attempt a fluent English translation,

    I did, actually, I just didn’t try to keep the word order of the original (Incidentally, I set up ID-tags so I could link directly to sections, e.g. http://jdm314.livejournal.com/199528.html#compositio, but that apparently didn’t work for whatever reason):

    I won’t even try to do it in order in English:I sing of that woman, and dragons, sent from Essos
    by the hand of fate, first through Pentos and the Dothraki sea…I think “I sing of a woman and dragons, the one sent…” would be a fair translation too, possibly even better for English, but it would also be less literal

    The second thing you missed is a little more important:

    You can hear Mad Latinist pronounce it here (good reading!).

    That is not me reading the poem in the first (nor, indeed, the second) sound file, but a friend of mine who’s good at reading hexameters. Easy to miss that, because I didn’t give his name (he asked to remain anonymous.)

    Sounds awesome.

    Thanks!

    Sounds like something that should be attempted after the series has completed (I promise High Valyrian will have enough words to handle it at that point). It’d require GRRM to sign off on it, but wouldn’t that be awesome?

    It would! But it’s a little daunting: I’ve never tried composing anything longer than a couple of lines in dactylic hexameter … and once you get past the first two-or-so people stop focusing on the flashy meter, and it becomes more about your actual skills as a poet ;)

    I’d certainly consider it, though, and, to be honest, it’s very unlikely I won’t be composing lines now and then as the series continues.

    Of course, even if I could write something the length of the Aeneid in just a couple years, the series not being over yet wouldn’t be a huge proble: I could just say the poet died before finishing the poem ;)

    After all, all the old myths are told and retold; they’re not made up whole cloth. Daenerys would be an outstanding subject for an epic poem (or I’m assuming. I too don’t know how it ends).

    I doubt it ends in any way that makes her an inappropriate topic. Her deeds so far have been plenty epic, if it ends embarrassingly it is, at worst, a question of where to stop the poem (believe me, this was often true for Greek heroes!). It seems most likely her story will end either in honor (reign, or… well you know), or glory (death). But, then again, never trust GRRM to do what it looks like he’s going to do.

    If there is one quibble I’d have, it’s with ivāedan. Since the oblique applicative is being used, it should be standing in for some sort of adpositional phrase which is appropriate to the oblique applicative. Unfortunately if you want to say something “about” something, the postposition you’d use is bē, which is technically a locative postposition, so it should probably be uvāedan (and the cases would have to change accordingly).

    This did all occur to me (changing it to uvāedan comes so close to working, btw… the only issue is zaldrīzoti which cannot be made to scan in the position it’s in)… as you no doubt surmised, I was trying to reproduce the effect of ἀείδω/canō in Greco-Roman epic, which are used, at least in that context, to mean “I sing of,” with an accusative object. Using bēbvāedan (is that the right allomorph?) surely makes more sense on a purely semantic level, but, for me at least, somewhat weakens the allusion.

    (Thanks, btw, for that term “oblique applicative.” I was looking for a good name to distinguish i- from u-, and that’s perfect.)

    But maybe you could get away with ivāedan.

    Yes, I was hoping I could pursuade you that this eccentric usage was common in HV poetry ;)

    On that note, when/if you get around to designing in-world poetic forms or HV, I hope you’ll consider quantitative verse. It just seems so logical!

    but there actually is a Dothraki word for “desert”

    Funny you should mention that ;)

    A different option might be two do a hexametrical couplet like Mad Latinist did, but I thought this might be too difficult. Thoughts? I’m open to either. Mad Latinist’s was outstanding

    Thanks again. I think several people here would find it fun to attempt. But hexameters are complicated, especially if you include all the stylistic rules. Given all the trouble people had counting morae for HV haiku, I would almost certainly find myself nitpicking everyone to death if they did hexameters. Still, if anyone’s willing to brave it, I would love to see it! (And I’d also be willing to help with technical questions, if anyone wants to get in touch with me.)

    But as you’ve said, part of the point is to use a poetic form that’s easy, and to that goal Limericks may well be the best idea. I’ll certainly have to try one or two myself, and see. Of course the subject matter might fit Astapori Valyrian better.

    Oh, on that note, you haven’t discussed the possibility of using AV at all, but I think that would be promising for haiku as well. You did see my (somewhat limerickish) attempt? I hope I can get away with the lack of indefinite articles there.

  2. Yay for me, and also yay for the other Dothraki haikuists. I wish we had more participants, more beginners testing their ability.

    —–

    Ya. That semicolon is totally not an integral part of the spider poem. As I comment on the forum, I rather like the way the haiku rolls stickily from one line to another.
    I knew that with alienability issue and all qosaroon would be a good alternative (not for the syllable count, but otherwise), but the genitive didn’t seem out of place given the couple of times we’ve seen “azho anni yeraan” type of syntax. And the way the last line sticks a bit to the “hatifaan” of the line before works perhaps even better asymmetrically.

    “Wastes” is perfectly fine. I never trust that the word has an exact counterpart in another language, so I like to try the boundaries in tranlation, but now that I know Dothraki make the difference, I would not change to desert. What is the difference, though, exactly? In scientific vocab “desert” is pretty wide top category term, I think, and would cover pretty much all dry desolate envoronments.

    —–

    If anyone wants to know more about grammatical problems of their Dothraki entrys, you can always come to the forum and I’ll be sure to give my unauthoritative opinion :P

    • So far as I know, “waste” (often plural) is largely a synonym for “desert.” If there’s any difference, it would be that a waste doesn’t have to be arid. If there’s a technical definition used by scientists I haven’t been able to find it.

      Of course you’re more likely asking about the diference between athasari and zelathi, about which I know nothing.

      • Yeah, of course the difference in Dothraki is what I’m looking for. “Wasteland” can even denote to uncultivated and undeveloped land (“wastes” probably can’t?). Or devastated, after-a-catastrophe land. But at least the first one would make no sense in Dothraki and the second does not fit the etymology.

        • Could it be that the difference is in how definite a place is. Zelatha sounds like a certain place, the kind that might have a name and all. Athasar in it’s group plurality sounds like a waguely demarcated environs.

  3. A little question for you Dave
    i know its not realy the place to ask this but
    are you the one who created the language of the dark elves in thor 2 because in an interview asaw this : ” I’m afraid I don’t know his name, [but] the gentleman who created it worked for Alan on “Games of Thrones.” sow i thought it might be you and if so are you planning to share this because it sound lovely

  4. In the comments, Mad Latinist suggested that it shold be ropetas? It should not: ropatas is correct. This is because the stem is ropa-, not rop-. Easy mistake to make, though.

    Nice of you to say, but it’s only an easy mistake if you think the verb is *ivestra·gon (rather than ivestr·agon). I have no idea where I got the idea that it was—I even marked it as C-fin in the vocabulary. So, pretty bad fumble on my part.

  5. I’ll give you an A for effort here, The Majesty, but this doesn’t really work. Neither kis nor notat can be used in that way.

    Yeah I felt like I was stretching it a bit there. Out of interest, could you give the words for “to try” and “to translate between languages”?

    • That’s the right word for “try”, but it’s not a verb: it’s an auxiliary. The construction athkisar is a very meta construction, and probably wouldn’t be natural to Dothraki. And there isn’t a word for “translate between languages”. You’d just say “astat ki X-i”, where X is a language.

  6. Wow, thanks for the honorary mention! I suppose one could also go with Māzīlzi // ōrbar se ñuqir // jelmȳssi which makes the morae count 5-7-5, no? Maybe loses some of its poetic simplicity then though (no hubris here). Congratulations to Qvaak as well. I don’t know very much about Dothraki, but from the translation it seems very elegantly put. I especially like the phrasing “treasure of the wastes”, both in English and Dothraki.

    I think you should keep High Valyrian in the haiku competition next year – it was challenging, but very fun, once you understand the way morae are counted. I am going to have a try at dactylic hexameter, but I think that format for a competition will probably scare off quite few people. Limericks seem like a wonderful possibility for you to to expand the vocabulary of High Valyrian, so I’ll be sure to compete if that ends up being the subject matter.

    Regarding our discussion of the Lord’s Prayer, does the subjunctive actually have any function on its own? If so, then what would that be?

    • I need to update http://wiki.dothraki.org/High_Valyrian_Verb_Moods, but uses we know so far are

      • With daor to indicate a negative.

      • In reported speech, apparently indicating uncertainty (the way I read it is that if the speaker wishes to commit to the truth of the statement, he or she will use the indicative. If not, it defaults to subjunctive. But no guarantee that’s right!)

      • With lo to indicate less certainty (from what DJP said it sounds a bit like the Latin “future less vivid” conditional, but, again, no guarantee.)

      • In embedded questions, to indicate potential? (“To boldly go where no man would go before.”)

      Of course all these options so far require a marker or a main clause. So I don’t know how, say, iksos would be understood on its own.

      • With lo to indicate less certainty (from what DJP said it sounds a bit like the Latin “future less vivid” conditional, but, again, no guarantee.)

        OK, looks like I remembered this incorrectly, here’s what DJP actually said:

        If there’s a controversial bit in that translation, it’s the choice of verb and tense in lo ziry arlī jaelāt—i.e. “if it again you want”. There are a couple of ways I could’ve gone. One would be, for example, to use the verb emagon, “to have”, in the subjunctive. I felt that was too hypothetical. This translation I felt was more direct (i.e. using the indicative rather than the subjunctive and using the verb for “to want”), and I liked it better for the content. It was a choice, though, so feel free to skewer me in the comments.

        So, if I understand right, he’s saying that he could have written either lo ziry arlī jaelāt (“If you want it back”), or lo ziry arlī emoty (cf. literary English “If you would have it back.”)

        So I guess it’s not so much a “future less vivid” construction (~ “If you should want it back…”), as using the subjunctive to express a desire. But that meaning doesn’t help you, because you don’t want to say “Your kingdom wishes to come….”

        • Yes, it does seem a bit unnecessary to inform God about the desires of his name and kingdom, doesn’t it? Well, even if the subjunctive doesn’t carry the sense of a third person command, it seems to me it should have some sense on its own. But given that there hasn’t been any official use of it it yet, our hopes appear slim. I am still curious for an answer from āeksio Peterson though!

          • The lives of subjunctives often do not start with independent meanings. Languages start without subjunctives, but then develop them naturally as old forms are fossilized in subordinate constructions, with other forms evolving away from them. At some point in time later, this new subjunctive form can then develop non-subordinate uses, but that will very highly from language to language. I’d say High Valyrian proper hasn’t developed those uses explicitly at this stage, and that it’s very much open to interpretation or individual variation.

            • Hmm, it will be interesting to see how this fits together with your earlier statement about “affective” vs. syntactic usages. I wasn’t sure what that meant, but my guess was somewhat at odds with what you say here.

            • Is that always how it works? Could a subjunctive also “devolve” from a more marked verb mood?

              Do you have a conhistory for the development of HV’s subjunctive?

              Incidentally, where does Latin’s subjunctive come from?

            • Well, it does check out, in Latin the Subjunctive forms are the ones that use the athematic form: amo/amem. In Greek the difference between the present of -ω verbs and the Optative seems to be also that of thematic/athematic respectively.

        • Oh! Another possible use of the HV subjunctive! We have the expression kostilus, which looks like a future subjunctive of kostagon, literaly meaning something like “Be it possible.”

          In other words, it appears to be using a subjunctive in place of a conditional. Who knows if we can generalize from this, though, since it is a fixed expression.

  7. It looks like my Dothraki is getting rusty ;) In any case, thanks for the good mention, zhey David!

  8. Congrats to everyone, especially Qvaak for his well-deserved prize!

    I should have seen the possession issue coming; it’s been my Achilles heel in Dothraki before… meh, blame it on Vacation Brain (my ELO in WordFeud has been taking a nose dive for the past two weeks, too).

    I tend to think of “more” in “more bells” as the phrase head, but I guess adjectives just don’t work that way in Dothraki.

    I’ll have to try the hexameter at some point. Definitely not with Vacation Brain, though.

    • By the way, the gō in my Valyrian haiku was intended to mean “down” rather than “before”; is that a possible reading?

      • I also took it to mean “before”!

        My understanding from Elat k’Athivezhofari… was that could indeed mean “downwards” or “below.” In particular see this comment thread between me (“locative/allative”) and DJP (“semantic/metaphorical”).

        But, to repeat a point I made there, if you wanted to be less ambiguous, gōvropatas (or is it gōbropatas? Still don’t know the rules for that allomorphy) might do the trick.

  9. The Dragon Demands

    A question has developed over on Game of Thrones Wiki about when Low or High Valyrian is being spoken: in the TV series, from your comments in the episode-by-episode blog postings, Kraznys is speaking Ghiscari Low Valyrian (in which case, is “the Climb” the only time we hear High Valyrian? — the language wiki does not yet have a by-episode guide for Valyrian as it does for Dothraki).

    Anyway, in the books, A Storm of Swords Daenerys II describes Kraznys’s speech as:

    “Even the Ghiscari tongue was largely forgotten; the slave cities spoke the High Valyrian of their conquerors, or what they had made of it.”

    Emphasis on “or what they had made of it” makes me think that this is a confusing way of saying that Low Valyrian, derived from High Valyrian, is what they actually speak.

    More importantly:

    “Kraznys’s High Valyrian was twisted and thickened by the characteristic growl of Ghis, and flavored here and there with words of slaver argot. Dany understood him well enough, but she smiled and looked blankly at the slave girl, as if wondering what he might have said.”

    So it says his High Valyrian is “twisted” and has some slaver-specific jargon thrown in there, but grammatically, etc. is book-Kraznys fundamentally speaking Low Valyrian or High Valyrian?

    While I’m on the subject: Daenerys can speak High Valyrian and most if not all of the Low Valyrian varieties of the Free Cities (she’s lived in most of them at one time or another, grew up in Braavos, “Valyrian” is her mother-tongue)……so, did Daenerys have any difficulty adapting to Ghiscari Low Valyrian?

    Problem is academic because it’s related enough to the Valyrian she knows that by Meereen, I think she’d have picked it up.

    But the scenarios we’re looking at are:

    1 – Book-Kraznys speaks High Valyrian, which Daenerys already knows. Daenerys doesn’t initially know Ghiscari Low Valyrian but knows Free Cities Valyrian already so it’s a quick study.

    …in which case, the TV show officially changed this TV-Kraznys speaks Ghiscari Low Valyrian….and Daenerys was just able to pick it up very quickly because she already knows the other Valyrian variants as a mother-tongue.

    Scenario 2: Book-Kraznys was actually speaking Ghiscari Low Valyrian, and the text is just confusing in describing it as “twisted High Valyrian peppered with slaver argot”.

    Does anyone else have quotes about this?

    • I hate to be that guy, but all of these questions have already been answered on this blog and elsewhere—and decisively, in my mind.

      To summarize:

      (1) Kraznys speaks Astapori Valyrian—in the book and in the show. Astapori Valyrian is descended from High Valyrian, but descended directly after a process of creolization and de-creolization with the Ghiscari language. This development was different from the development of the other Low Valyrian languages, but given that the source material is the same, it’s understandable that Dany gets it pretty well, despite the fact that she’ll miss a word here or there that’s of Ghiscari origin.

      (2) Dany only speaks High Valyrian on the show. Thoros and Melisandre only speak HIgh Valyrian on the show. Every episode that has them speaking Valyrian of any kind has High Valyrian in it.

      (3) The fictional narrator of GRRM’s books is clearly someone from the same time period, and not someone from a modern perspective. Consequently, the descriptions you quote pretty much describe the linguistic situation I’ve described before to a T. In other words, just because the Slavers say they’re speaking High Valyrian doesn’t mean they’re actually speaking High Valyrian.

      • The Dragon Demands

        Thank you and I apologize – I’m “that guy” asking questions again – but the backlog and overlap made it confusing. Time spent answering our repetitive Q&A is time that could be spent working on HV, I understand. I’ll update the wiki with this…

    • “[L]argely forgotten”? That could be read as meaning it’s not quite dead yet. (But it could also mean that it is dead and only a few people can even read it, or somethign like that, which I assume is your official interpretation.)

    • Actually at the Valyrian Language wiki we have the whole Corpus of all the High Valyrian dialogue in Season 3 by episode. So you can also use that to get the languages right. Hope it helps!

  10. It just occurred to me: if khaleesi comes out as just *khaleesi in HV (which, based on what we’ve seen of Dothraki → HV loanwords so far, seems very likely)… I’ll never be able to fit it into a hexameter (because it would scan as four consecutive short syllables—the last one could be elided, but that still wouldn’t solve the problem.) A shame, that. Maybe some people pronounce it **khalēsi instead? ;)

    • The syllable separation between two same vowels is quite apparent, when the latter syllable is stressed (like with all /-aan/ and /-oon/ endings), but rather subtle otherwise, IMO. In khaleesi neither syllable gets major stress, so I’d actually expect the syllable boundary to be lost in the loan process. It would be odd for Valyrians to be that exact in loaning when the long vowel interpretation fits so well to their phonotactics.

      Finnish sometimes drops consonants in declination (words have weak and strong form, that is), so sometimes vowel collisions like koko (size) -> koot (sizes) happen. I was well into my twenties in non-aware belief that I and people in general respected the syllable boundary and pronounced koot in two syllables. When the topic came up, it turned out almost noone thought they did, and I’m not at all sure if I really do either – in regular speech. ych! I’ve read an article on diphthongs in the same situation, and evidently we have a lot of grey area between definitely two syllables and definitely one syllable.
      …have I written this same anecdote before?

      You’d kinda expect Dothraki to invent long vowels in some hundread years, but who knows, could they go some other way, like just dropping repeated vowels?

      • Pretty much everything Qvaak said above is spot on. And I would expect Valyrians to borrow it in as khalēsi (pronouncing it either kalēsi or halēsi mostly, I’m guessing).

        • Yay, that will scan! Do you expect that firmly enough for me to add it to the vocab page?

          • You know I haven’t added it to my own dictionary, but it does seem stable and obvious enough to add it, yes. And we can retain the spelling khalēsi, even if the kh isn’t always pronounced [x].

            Might as well add khal too. It’d be the same borrowed class as buzdar, with a floating i ending that some may use and some may not. Be sure to mark them as in-universe borrowings, though.

            • Actually, khal(i) is already there, on the authority of this IAmA thread.

              Shall I assume that khalēsi declines the same way, or does the fact that it already ended in -i change anything? And that being the case, is it ever *khalēs in the nominative?

              [Come to think of it, THAT would have solved my problem if you had gone with the double e form: use *khalees followed by a consonant. Mind you, as soon as it gets out of the nominative the problem would be back…]

        • This also answers a long-standing question I have had, and even asked David in person about. I have always expected the /ee/ in khaleesi to break into separate syllables as usually seems to be the case in Dothraki. This gives this word four syllables and a very different pronunciation that the usual 3 syllable KhalEEsi. But when asked David to pronounce khaleesi, he pronounced it the 3 syllable way, too, and I have never understood why Now, I know why. (and it shows that I still have much to learn!)

  11. Ok, so I’ve been sick and have had some fun translating two texts which you might find interesting: The North Wind and the Sun (orginally) by Aesop, and The Sheep and the Horses, aka. Avis akvāsas ka, by August Schleicher. The first is in my understanding sometimes used as a parallel text, especially in phonetics, to compare different languages, and the second was written as to give a sample of what Proto-Indo-European might have look liked. My source is Wikpedia; make of that what you will. This page also has translations of the second fable into Latin, Ancient Greek and Sanskrit. I had to make quite few changes, since I could not find words for north, strong, to argue, cloak, close (adj.), to wrap or to fold, sheep, wool, horse, wagon, load and garment. Also sort of hard to find were to carry (on one’s back), to drive (of animals) and plain (I used the collective of grass, parmon). I present the translations sentence by sentence, to give a better overview.

    Jelmio Sōnaro Vēzōs

    Jelmio sōnaro vēzōs rovyktī ondori emilis sparo bēvīlīptis, skorī vala ñellyrry bānoti mastas.
    The winter wind and the sun were fighting about which had the greater power, when a man in a warm skin came.

    Vali ñellyri nādīnagon ēlī verdis lue ziry rovyktī ondori emagon otāpilzi vestretis.
    They said that they will think the one who first arranged the man to take off the skin to have greater power.

    Jelmio sōnaro tolvomy zȳho ondondo jēlēbetas, yn vala ñellyri va zirȳlā mērī mazēdan.
    The winter wind blew with every one of his powers, but the man only took the skin towards himself.

    Mōrī jelmio sōnaro zirȳle rūdas. Vēzos bānī jehittas, se vala ñellyri nādīnetas.
    At last the winter wind quit. The sun shone warmly, and the man took of the skin.

    Sesīr vēzos rovyktī ondori emiles vestragon sōnaro jelmio bēvilza.
    And so it was incumbent upon the winter wind to say that the sun had the greater power.

    Epses Nuspessē

    Dōrior ōghri ēdas lȳs epses nuspī undas, ēlos pōnto kempi hakore, tȳnos rōvi ōrere se saelos vale aderī ōrere.
    A goat that had no hair saw cows, the first pulling a heavy thing, the second carrying a big thing and the third carrying a man quickly.

    Epses nuspī ivestretas: “Prūmia aōha avy ōdriksa, vale nuspī memēbare urnere.”
    The goat said to the cows: “My heart pains me, seeing a man driving cows.”

    Nuspesse vestretas: “Ryptēs, epsys, prumī īlvi ōdriksi skorī kesir urnēt: vala, aēksio, zirȳlo syt ōghrosa epso bāni mazverdas. Sesīr epses dorior ōghri ēza.”
    The cows said: “Listen, goat, our hearts pain us when we see this: a man, the master, makes a warm thing with goat hair for himself. And so the goat has no hair.”

    Kesir rypta, va parmorro epso hembistas.
    Having heard this, the goat left for the plain.

    • Oh well, a mess-up of the HTML tags was bound to happen I guess. Please fix?

    • Correction: it should of course say prumī īlvi īlon ōdriksi in the third sentence of the second fable.

    • I’m on my way out the door, so I’ll have to go over this later, but cool!

      I know Esploranto/Najahho was planning to do the latter tale eventually. Cf Dārys se Jaes (which he’s also been planning to revise and bring up to date).

      • Oh yes, I did read that! It was kind of what inspired me to try to translate something with a more natural language, since the Lord’s Prayer was quite a piece of work.

    • Impressive! I too don’t yet have the time to go through this, but an amazing attempt whatever the result! Well done!

    • Great stuff Joel! Haven’t had time for an in-depth analysis, but the glossing looks convincing. “Warm skin” and “winter wind” should work well. Not so sure about “goat” and “cow”, but it beats “bear” and “dragon”! ;o)

      This would also make a great text for pronunciation practice, but again I’d better wait for the final version.

    • You’ve done a great job, generally. Here are the problems (or at least questions) I’ve noticed for the first text.

      Jelmio sōnaro has a following genitive, which actually I’m thinking is kind of cool here (I normally default to a preceding genitive on HV, myself.)

      sparo bēvīlīptis “were fighting over who…” I have no idea how to say this in HV, I think it’s an issue of “embedded questions” (certainly in Latin it’s an “indirect question,” which is roughly equivalent). But whatever it is, I really doubt it’s this. Even if this construction is correct, we still have some problems.

      1. The root for fight appears to be vīlīb- To add bē- you need to make it a locative applicative. Problem is, we don’t know what that would look like in this phonetic environment: *bēvvīlīb- does not seem likely, nor does *bēuvīlīb-. Bēbvīlīb- seems unlikely as well, but is the least problematic of the three. I suspect you knew about this issue, and just decided that bē- + u- + vīlīb- would just resolve into bēvīlīb- with only one -v-… but I kind of doubt that. Only DJP can say.

      2. The tense here appears to be either perfect (“…have fought…” “…fought…”), past habitual (“…used to fight…”), or pluperfect (“…had fought…”)—so far as I can tell -tis could be any of those three, but that may very well turn out to be wrong (would love some clarification on this), In any case, surely the most logical form would be imperfect (“…were fighting…”)?

      skorī “when”, problem is, skorī may only be usable in the interrogative sense (“when is a man coming?”), and not the relative sense (“…when a man came”). Hopefully DJP will cover this when he gets to his post on HV relative clauses, which he’s been promising will come out very soon (I’m thinking a week from today is most likely).

      mastas should be māstas.
      I’m not even going to try to tackle the next sentence up to ziry, but I think it’s good. Verdagonis a clever choice, by the way. But ziry … emagon otāpilzi appears to be an accusative-infinitive construction, something we haven’t seen in HV so far. I would have assumed it was ungrammatical, but come to think of it, I may just be overcorrecting (since Latin uses that construction almost exclusively, and as we all know, I usually have Latin in mind).

      tolvomy zȳho ondondo “with each of his powers”: the partitive genitive is clever, since we don’t officially know the instrumental of ondor (ondorzo?) But thus far we’ve only seen tolvie used attributively with the noun it quantifies. It will be interesting to see if this is OK or not (my money is on “yes,” but we’ll see.)

      jēlēbetas “blew”, nādīnetas “took off”: these forms are incorrect, but the fault is entirely mine. First of all, I listed jēlēbagon as V-fin, when in fact it is C-fin, thus the perfect should be *jēlēptas. Furthermore, I listed the regular perfect of a-stem verbs as -etan, but this was a (fairly stupid) mistake… it’s -atan. But, while we don’t know whether nādīnagon is vowel or consonant final, consonant final is significantly more likely (it almost always is), so my guess would be *nādīntas.

      • I corrected the errors you mentioned. For the embedded question and relative clause, I left them as they were but marked them with brackets. Do you think it’s possible that the relative clause (“when a man came”) could somehow be constructed with lua but with some special affix? I am thinking skoros/skorion “what one, what thing, what” gives us skorī, so lua could perhaps render luī. We’ll see anyway, and I will make the corrections to the second fable soon.

        Jelmio sōnaro vēzōs [rovyktī ondori emilis sparo bēbvīlībilis], [skorī vala ñellyrry bānoti māstas].

        Vali ñellyri nādīnagon ēlī verdis lue ziry rovyktī ondori emagon otāpilzi vestretis.

        Jelmio sōnaro tolvomy zȳho ondondo jēlēptas, yn vala ñellyri va zirȳlā mērī mazēdan.

        Mōrī jelmio sōnaro zirȳle rūdas. Vēzos bānī jehittas, se vala ñellyri nādīntas.

        Sesīr vēzos rovyktī ondori emiles vestragon sōnaro jelmio bēvilza.

    • Forgot to mention, I’m working on a major overhaul of the verb conjugation page, which will fix that -etan error (among other things). I did also update the subjunctive page with those other uses I listed for you before, and DJP’s recent pronouncement on the topic.

      Moving on the the Goat and the Cows… hahaha, I knew we didn’t have horse, but I could have sworn we had sheep! Hmm, Zhalio’s joke about dragons… I dunno, that COULD work, given that the Valyrians were famous for domesticating them! Of course if a goat spoke to a dragon that way, I kind of doubt the dragon would have bothered to retort before eating her.

      For the most part, I’m not going to bother comparing the original to see where you could have gotten the grammar closer, but that might be something we should do in the future.

      • General comment: in many cases it might have made sense to refer to “man” as valar.

      ōrere will definitely do for “carrying” the first time. I briefly considered going the Hebrew rout, “cause to come,” but then I realized we don’t have a causative construction yet. The second time, not so much. Not sure what to use instead.

      aōha should of course be ñuha.

      • I know we have no word for “drive,” but memēbare “advance” doesn’t seem likely to cut it. Maybe valo syt or ondoso nuspī memēbari urnere? That also solves the problem of the unclear antecedents for those last two participles.

      urnēt should be urnī. And of course the skorī problem.

      • Since rȳbagon is transitive, rypta will mean “having been heard” not “having heard.” You might try kesrioso ryptroso or the like.

      parmor is a great solution. I don’t think I would have thought of it myself, but now that you mention it I would not be at all surprised if that turns out to be the official word.

      • The new verb conjugation page looks great, and is a lot easier to overview now. Great job! It will be a lot of help in writing my dactylic hexameter, which I have begun writing ever so slowly. Some revisions, as promised:

        Dōrior ōghri ēdas lȳs epses nuspī undas, ēlos pōnto kempi hakore, tȳnos rōvi ōrere, saelos zȳho inko bē vale ōrere se naejot aderī dekurūbare.

        Quite a circumlocution, “the third on its back carrying a man and walking forward quickly”, but I think should be grammatically correct and also more suitable than the previous, since (I realize now) vale aderī ōrere would more likely be interpreted as “soon holding a man” (!).

        Epses nuspī ivestretas: “Prūmia ñuha yne ōdriksa, vale ondoso nuspī memēbari urnere.”

        So this would essentially mean “… seeing cows advancing at the hand of a man”? I like this better than valo syt, which seems a bit too abstract, “advancing for (the sake of) a man”.


        Nuspesse vestretas: “Ryptēs, epsys, prumī īlvi īlōn ōdriksi skorī kesir urni: vala, aēksio, zirȳlo syt ōghrosa epso bāni mazverdas. Sesīr epses dorior ōghri ēza.”

        Kesrioso ryptroso, va parmorro epso hembistas.

        The last bit would mean literally “by means of this having been heard”, no? How does that make sense exactly? Also, is not kesrȳsi more likely? I am thinking that perhaps it would be more natural to use a different construction, like tolī kesir rȳptas, although the meaning is slightly different.

        On another note, I realized that a possible word for “clothes” could be ñepety, i.e. “that which has been sewn”. Funny how ñepegon appears right after ñellyr in the vocabulary! Drējī, ñepety ñellyri tȳniar issa!

        (Oh, and I am curious to know what *ñellys could possibly mean. Skin cells, perhaps?!)

        • The new verb conjugation page looks great, and is a lot easier to overview now. Great job!

          Thanks! Have you looked at http://wiki.dothraki.org/High_Valyrian_Verbal_Prefixes yet? It’s still extremely rough, but has some really important stuff in it that was really overdue for a write-up.

          It will be a lot of help in writing my dactylic hexameter, which I have begun writing ever so slowly.

          Get in touch with me if you need any help. Should be easy to track down one way or other to contact me.

          Quite a circumlocution, “the third on its back carrying a man and walking forward quickly”, but I think should be grammatically correct and also more suitable than the previous, since (I realize now) vale aderī ōrere would more likely be interpreted as “soon holding a man” (!).

          I like it!

          So this would essentially mean “… seeing cows advancing at the hand of a man”? I like this better than valo syt, which seems a bit too abstract, “advancing for (the sake of) a man”.

          Whichever you prefer. Yeah, ondoso means literaly “at the hand of,” but it’s also used for agents. And of course cf. ondor “power.”

          Kesrioso ryptroso, va parmorro epso hembistas.
          The last bit would mean literally “by means of this having been heard”, no? How does that make sense exactly?

          OK, what I’m going for is a “{case} absolute” construction here. Latin has an “ablative absolute,” Greek has a “genitive absolute,” Gothic has a “dative absolute” and so on… a remnant of the construction survives in English expressions like “that being said.” Now, I took rūso zȳhosy gōvilirose to be possible evidence of an instrumental absolute in HV, but of course since rūs is 3rd declension there’s no way to be sure it isn’t just a comitative, exactly equivalent to the English “with his child beneath…,” I don’t know. I would be surprised if HV had no “absolute” construction at all, but then again I was also surprised to find it had no “accusative infinitive” construction, so what do I know?

          Also, is not kesrȳsi more likely?

          Oops, yeah.

          I am thinking that perhaps it would be more natural to use a different construction, like tolī kesir rȳptas, although the meaning is slightly different.

          Yes, that would probably work.

          On another note, I realized that a possible word for “clothes” could be ñepety, i.e. “that which has been sewn”.

          Nice.

          (Oh, and I am curious to know what *ñellys could possibly mean. Skin cells, perhaps?!)

          Or possibly *ñelly, since so far as I know we don’t know the gender of ñellyr. Perhaps ñellyr is “skin” as an organ, and *ñelly(s) is “a skin” as a material? But if so that would be a problem for you. Maybe ñellyr is just a “collectivum tantum?”

  12. Kesrioso ryptroso, va parmorro epso hembistas.

    The last bit would mean literally “by means of this having been heard”, no?

    Comitative, maybe? Something like kesrȳmi ryptromo “along with this heard”?

Leave a Reply

Your email address will not be published. Required fields are marked *

%d bloggers like this: