On Monday, right after I posted my last entry, I got to do another IRC session with Mr. Peterson, and we learned a good deal.
First of all, when do adjectives go before or after the noun?
<DavidJPeterson> All adjectives can be postpositive if they want, but especially those that are more determinative in nature generally come before the noun, unless you want to sound...official?
Next we have a discussion of verbs, including the following points:
<DavidJPeterson> Here's probably why. In my dictionary, I do forms like this:Let's also throw in this bit, from the discussion of my last entry's title:
<DavidJPeterson> iōr·agon, -on, -ilen, -tan
<DavidJPeterson> That's "to be standing".
<DavidJPeterson> The first form is the infinitive.
<DavidJPeterson> The other suffixes are: (1) the first person singular subjunctive present; (2) the first person singular imperfect; (3) the first person singular perfect.
<DavidJPeterson> Here's the entry for smile upon:
<DavidJPeterson> ilīri·gon, -on, ilīrīlen, -tan
<DavidJPeterson> And I would've used the aorist here as well.From this we finally learn the official names of tenses, and also the principal parts of a High Valyrian verb:
- The -il- form I've been calling "timeless" is the "imperfect." (Assuming that -ilen is the same as -ilas etc.)
- The -t- form I've been calling "past" is the "perfect."
- The -ss- form I've been calling "frequentative" is the "aorist."
Critically, although I colloquially call these "tenses," they are more properly "aspects." The usage will be something like this:
- The perfect emphasizes the end of the action, and so will generally indicate past events.
- The imperfect does not emphasize the end. Critically, unlike the "imperfect" of most European languages (more like Biblical Hebrew), this will not necessarily refer to the past, and in fact most commonly will refer to the future or present.
- The present refers to things that are happening right now.
- The aorist refers to general truths—things that in English get put in the present tense, but aren't literally happening right now. Although I previously referred to the imperfect as "timeless," that name would have been more applicable to this form. This is not quite how the term "aorist" is used in Greek grammar, but it is exactly how it is used in Demotic Egyptian.)
As for the principal parts, they are: the infinitive, the present subjunctive, the imperfect, and the perfect. Note that principal parts are generally chosen as the minimum amount of information you need to conjugate a regular verb it in all its forms. So the implication would seem to be that you cannot necessarily predict the present subjunctive, the imperfect or the perfect, but you apparently can always predict the aorist.
Incidentally, the example verb Mr. Peterson gave, iṓragon "to be standing," is presumably the etymon of the AV word I transcribed as mejorezlívas "they will stand." (For the change of i to j, compare iā > ja, both meaning "or.")
In another discussion (which was partially by comments lost dothraki.com's mysterious crash last week, partially by IRC, and partially by email—none of which is publicly available!), Peterson informs me that the verb iderḗbagon "to choose" (and yes, that is the correct infinitive, which is interesting, because we know it's a -za verb) should theoretically take a dative object (e.g. gláesot iderḗptot daor; gáomoti iderḗbzi), but when he wrote the dialog for "Kissed By Fire," he accidentally let a few accusatives slip by (jénti ... iderēbilātā́s; brṓza ... iderēbilātā́s; brōzi, iā mirre tolie iderēbās)—this is very important to know for the declension chart I'm working on, so my thanks to him for this clarification. Apparently iderḗbagon can take either a dative or a genitive object, depending on the exact meaning, but the closely related uderḗbagon must take a dative. (By the way, did anyone else notice that in the review scenes before this episode, the subtitle switched back to "You will choose a leader..."?)
In other news, in a recent interview, Mr. Peterson also gives us the following phrase: Dorior dorion udrirzi mijessis,
meaning "There’s no country without a language." I take this to literally translate as "No country with no language" (Latin nulla nullō sermōne nātiō or the like)—if this correct, dorior modifies mijessis, and dorion udrirzi. I'm guessing the latter is the comitative form of *udrir, the collective of udra "words" (so literally "vocabulary.") The only other known comitative we've seen so far is valoma "with a man," but we know that the comitative has a tendency to merge with the instrumental, and at least some of the instrumentals we've seen (namely valosa and ēngoso) look at least slightly more like this.
Speaking of *udrir, here are some random vocabulary items we have gained:
- At last we know what syt means: "Syt‘s just a postposition that assigns the genitive. A number of uses all stemming from a general benefactive." In other words it means something like "for," and goes after a noun in the genitive case— in these respects it is very similar to the Latin words causā and grātiā (or the Greek ἕνεκα). This word probably survives in AV as postposition zy, as in zýa rúo murghí zy "for her dead baby." Syt- can also be appended as a prefix to verbs, in which it adds a sense of "for" (e.g. sytivīlī́bilāt "you will fight for..."), but does not (so far as we know) alter the case the verb would otherwise call for. By the way, Dinok totally called this one.
- We now know that "son" and "daugher" are respectively trēsy and tala. And that Peterson likes to "create pairs like that (both lunar; one -a, other -y)."
- Silver is gēlion, genitive gēlio, but Mr. Peterson has not yet decided how best to translate "silver" when it is an adjective. (Oh, and Najahho, might I suggest you just go with *Gēlia for now, like Qelbria from qelbar?)
- Astapori Valyrian has not been forgotten either: we now know the nominative for the 2s pronoun: o. This makes it the first Valyrian pronoun for which we have every form: nom. o, acc. av, dat. o, possessive adjective oa, oo.
- We can now confirm that the AV word for "no" is indeed do. "Yes" turns out to be kiz (presumably from kizy "this.")
This week's episode was interesting in that, while we got only a couple of lines spoken in Valyrian, there is actually a lot to cover.
The first Valyrian occurs in the scene between Robb and Talisa (NSFW: I had thought this would get banned from YouTube vasko v'uvar ez zya gundja, but there it is!) We see Talisa writing a letter, which begins with Muñus jorraeli.... She explains that it's to her mother, and is in Valyrian. Over the course of the conversation we learn two words: "yes" and "hello." Now, of course, there is no real word for "yes" in High Valyrian (GRL), much as in Latin or Mandarin. David J. Peterson wisely ruled that whatever it was she said was the word for "yes" in Volantene Low Valyrian. Unfortunately we can't quite make it out: Peterson heard it as dha (GRL), Najahho hears wa, gwa, or va, the subtitles apparently give it as "Gaaa", and I... am still not sure.
As for "hello," it is rýtsas. Very glad to know this, as valar morghūlis was beginning to get morbid.
There is no more Valyrian in that scene. However, an image of the completed letter was released, from which we can get much of the text. A good deal of speculation has resulted from this image, but I have been avoiding it for fear of spoilers (recall that I have not read the books.) That goes even for discussions that explicitly cite me(!) such as this one and this one (OK, I admit I peeked a little.)
By the way, Zhalio plans to release a font based on Talisa's hand, which will be very cool. I am indebted to The Dragon Demands for getting the ball rolling on transcribing the letter (on the other hand, in the same message he does makes potentially spoilery speculations in Latin, which I found it impossible not to read!) Here is my own transcription:
Muñus jorrāeliarzus,Múñus jorrāeliárzus: Something like "Dearest Mother." We have seen "muñar" in the sense of "parents," and this is presumably the collective form. In the singular, though, it means "mother." Muñus is the vocative form, and the nominative is muña—which means it probably declines like vala "man," for which we did not previously know the vocative. This vocative is, I suppose, consistent with dovaogēdys being the voc.pl. of dovaogēdy. Muñar "parents" seems to be literally "the set of mothers"—the opposite of Spanish, where "parents" is padres "fathers." Jorrāeliarzus I'm mostly getting from context, but, as Zhalio suggests, it is very likely a form of the verb rāelagon (which occurs later in this text) with the continuative prefix jo-.
]e hen embaro tolmiot nẏkēlot avẏ ivestragon issa[
n]ykēlo sẏt undon daor luo valzẏro ñoghossi oress oressiks
darẏs issa vestris, se prūmio ñuho konir drepor issa.
Aghapī īlōn rāelza, kesro sẏt lanta iksan, rūso zȳhosy
]zyo sẏt pyghas lue prūmie:
]za, ẏn aderī, mōrī, aōt māzīli s[
hen émbaro tólmiot nykḗlot ávy ivéstragon íssa: "There is something for me to tell you from across the water." Íssa "(there) is (something)." Nykḗlot "for myself" that is, nýke "I," plus the emphatic/reflexive suffix seen in pōntā́lī/pṓntalo and jemḗle, and the dative ending -ot. Ivéstragon "to tell"; ávy "you," acc.s. Hen émbaro tólmiot: "from across the water." This one is tricky, and somewhat speculative. *Émbar almost certainly means "water," on the basis of embōñor qogror "the aquatic gender," so this phrase must mean "from the tólmiot of water." Zhalio made the brilliant proposal that tólmiot (or whatever that word looks like in the nominative) means "the other side."
n]ykḗlo syt úndon daor luo válzyro ñoghóssi oress oressiks: "of a ... which I am not seeing(?) for my own sake" Unfortunately I have no guess what ñoghóssi is, not oress and oressiks (which don't even look like proper Valyrian), which is a shame because they would likely clarify the rest of the sentence. I don't know what válzyro is either, but it is at least clear it a noun in the genitive. The rest of the line is a relative clause dependent on válzyro: luo "which." Úndon daor "I don't úndon." Úndon daor might be the present subjunctive of the aorist subjunctive ūndéssun daor, if Talisa (or rather the art department) forgot to make the u long (and, as we will later see, we know for a fact they left off a macron at least once)—of course it is questionable that "I don't see" really fits in this sentence. Nykḗlo syt "for my own sake." One possible way to make sense of this this sentence is to assume válzyro means something like "land," "territory," or "country" (though contrast tegōñor "terrestrial" and mijessis "country, neither of which looks related), and the sense of the sentence is "(he is a man) of a country which I still have not seen for myself."
dárys íssa véstris: "they are saying he is king." Vestris "they are saying," 3pl. Íssa "he is." Dárys "king." Peterson has confirmed that this is a scribal error for dā́rys "king" (which we have already seen in the acc.s., dā́ri. Note also the female equivalent dā́ria "queen") Notice that we have an indirect statement here... I had already wondered how subordinate clauses of this sort are marked, and speculated that the zūgusy form was used for that purpose. But according to Peterson, reported speech is merely marked by the subjunctive, and even that is. Perhaps zūgusy is just a present subjunctive, with some sort of -y augment? I'm still very much unclear about this. In any case, since the subjunctive is optional, it is reasonable to assume that you use the indicative when you agree with the statement, and the subjunctive when you are merely reporting it without asserting that it's true. If that's the case, then the sense cannot be "they say he is king, but he's a total loser, I can't believe he's falling for this shit," as "candygram4mongo" quipped.
sé prū́mio ñúho kónir drépor íssa: "And such a drepor is my husband's(?)": se "and." Kónir "such, this" but no drépor. Íssa "is." Ñúho "of my." Prū́mio a noun in the genitive. It shows up again later, and I speculate it means "husband."
Aghápī ī́lōn rāélza: Very difficult. Aghapī could also possibly be Oghapī, but I have no guess as to the meaning. Īlōn could possibly be related to ilas as in geros ilas "fare well," or AV íles, the past tense of las ("That I was Aghapī?" Of course I don't believe we've seen any instances of this verb in HV yet!) Rāelza seems to be a -za verb in the 3s. If it is indeed connected to jorrāeliarzus from the salutation, it might mean "loves," "esteems," or even "misses."
késro syt lánta íksan: "On account of this, I am lánta." I am guessing that késro is the genitive of késir "this; here." I don't know what lánta means, unless it is somehow connected to AV nalánta "twenty" (an idea which also occurred to "Eli"), but this does not seem like a good fit semantically.
rū́so zȳ́hosy: uncertain, probably not a full and connected phrase anyway. Rū́so looks like a noun in the genitive, unless it is simply in the āéksio declension (it looks vaguely like AV rúo "baby," a word we would like to find in this letter, but the connection seems tenuous.) Zȳ́hosy must surely be some form of the 3s possessive, but what? An instrumental? A -y augment?? The point is moot, though, as possessives are normally prepositive, so we expect the noun it modifies to be the next word, whatever that may be.
]zyo syt pýghas lúe prū́mie: "... a husband(?) who pyghas for ...zyo...." I previously speculated that prū́mi- meant "husband." So far the only noun form we've seen that ends in -e is the accusative of the vála declension, so perhaps that is what we have here (in which case the nominative would be *prū́mia)... but so much of the declension chart has yet to be filled in that we cannot be sure. Lúe "who." Pýghas, a 3s verb in the present tense. "Fights" fits the context, but the root we currently know for "fight" is vīl-, as in "Ýne sytivīlī́bilāt?" Syt "for, on account of": we expect this to be preceded by a genitive, and it would indeed seem that the previous word—which seems to end with zyo or gyo (and given the space this could be the whole word) is indeed a genitive.
]za, yn adérī, mṓrī, aṓt māzī́li s[ "...]s, but sooner(?) or later(?) we will come to you a[nd" ...za might possibly be the 3s ending of a -za verb, but even if this is correct we can only guess. Yn "but." Adérī and mṓrī could be acc.pl. forms, or 2s. forms of an i-stem. Zhalio ingeniously speculates that these words are some sort of expression for "sooner or later"—he guesses that these are locatives, which is plausible, but we don't yet have any confirmed locatives of that form. Māzī́li 1pl imperfect of māzigon "to come." Aṓt 2s pronoun in the dative "to you." There is a curl to the right of māzīli that looks like it is maybe the top of an s. Since we have already had the verb for this sentence, it's hard to imagine what could follow it other than se "and."
I don't want to get too much into the speculation about Talisa (and if you want to comment on it, please avoid spoilers!), but it seems to me that this letter nearly exactly matches what you would expect it to say, based on her conversation with Robb. Neither does it appear to contain anything of military value. It is true that a spy is going to use codes to disguise her meaning, but I can't see the producers going trouble to write the letter in Valyrian AND obscure the meaning as well. Time will tell.
Our second Valyrian scene takes place just outside of Yunkai:
Daenerys: Va óktio remȳ́ti vále jikā́s.Va óktio remȳ́ti vále jikā́s "Send a man to the gates of the city." Jikā́s "send," imp.s. Vále "a man," acc.s. Va "to" (HV va "to, for.") Remȳ́ti "the gates," though what case this is is open to question: we've already seen that prepositions indicating location, and motion from take the locative, but this is our first preposition indicating motion towards. In Indo-Eruopean languages, such prepositions often take accusatives, but there's no guarantee Mr. Peterson would follow that pattern. Remȳ́ti could easily be an accusative plural (with an -i ending like ménti), or a locative plural (with a -ti ending, like gláesoti)... we'd have to see more cases to even guess. Óktio "of the city," gen.s.
Subtitle: “Send a man to the city gates.”
D: Belmúrtī ivestrā́s késir pṓnte jiōrínna se pṓjon obū́ljarion mazōrī́nna. Lodaor hḗnkos vḗjose hae Ástaprot Yúnkai botílza.Belmúrtī ivestrā́s: "Tell the slavers." Ivestrā́ "tell," imp.s. Belmúrtī "slavers," acc.pl., very likely a Ghiscari borrowing. Note that, unlike indirect statements, indirect commands seem to follow the verb (we've already seen this in Méntyri ídañe jévi ivestrilātā́s keskydóso gáomagon). Of course this could be just a military convention, to make sure the soldier knows what the actual order is, before having to memorize details. (The ancient military writer Asclepiodotus, on the other hand, did say that commands should go from specific to general, the exact opposite! But then he was talking about drills, not situations like this.)
S: “Tell the slavers I will receive them here and accept their surrender. Otherwise, Yunkai will suffer the same fate as Astapor.”
késir pṓnte jiōrínna: "I will receive them here." Jiōrínna "I will receive," 1s, -za verb. This certainly looks like a present tense at first glance, but I suspect it is secretly an imperfect: the form we would expect is *jiōrílna, but, as Mr. Peterson has told us, "l and n don’t play nice together," so I suspect the l here assimilates into the n. Pṓnte "them" acc. (AV pon.) Késir "here, this" (AV kizir).
se pṓjon obū́ljarion mazōrī́nna: "and I will accept their surrender." Mazōrī́nna "I will accept." Pṓjon "their," thus confirming the reading of poja (as in nágizi pojá pitenkávesa "... such is their loyalty"), way back in my first entry. Obū́ljarion "surrender," acc.s.
Lodaor hḗnkos vḗjose hae Ástaprot Yúnkai botílza: "Otherwise Yunkai will suffer the same fate as Astapor." Lodaor "otherwise," from lo, which presumably means "if" (AV lu), and daor "not." Yúnkai botílza "Yunkai will suffer," 3s imperfect, -za verb. Hḗnkos vḗjose "the same fate," acc.s.; I'm pretty sure hḗnkos is "same," and vḗjose is "fate" (cf. Qupténkos Ḗngoso "in the common tongue"—which I think is ins.s.) I have to wonder if vḗjose is somehow related to vejotréi from my first entry... but given how uncertain that is, and how little obvious semantic connection there is to be found, it seems unlikely. Hae Ástaprot "as Astapor"; hae takes the locative, as you will recall, and so *Astapor becomes Astaprot, just like lentor "house" becomes lentrot.
Next we see Dany's meeting with "Razdal" mo Eraz. Unfortunately this is all shown in English (i.e. the Common Tongue?) except for the words Razdal mutters on his way out:
Razdal mo Eraz: Inkán úndagho karakésh…Or as David J. Peterson wrote it: Inkan undagho buna gundjabo jorydrare evi rungo pulgarinko. Now, technically we should put this in Wise Master blue, and call it "Yunkish Valyrian" (YV), but since Mr. Peterson did not have time to create the whole Yunkish dialect just for this one throw-away line, we may as well treat it as AV until such time as Peterson can elaborate.
I confess, I cannot get much of this phrase. Here's what we do have:
- Inkán: a verb in the 1s, but the meaning is not obvious. From context we would expect something like "I want," "I'm going to," etc.
- úndagho: a verb in the infinitive. Probably "to see," like HV ūndéssun.
- karakésh: probably an improv by George Georgiou, who played Razdal. Perhaps Mr. Peterson will retcon this into a Yunkish word.
- buna: unknown.
- gundjabo: certainly related to gúndja "ass," but hard to say what the -bo is. It could be an enclitic postposition (Peterson has said that most AV adpositions go after the noun.) It also reminds me of gelébo "silver coin," from HV gḗlion "silver." If a gelébo is a "silver coin," then what is gundjabo? A "piece of ass"??
- jorydare: maybe a participle, or a passive form?
- evi: Perhaps e·vi "on the," like é·ji tóvi "on the day." It certainly cannot mean "we want," since we now know that's ébi
- rungo: unknown, other than it's almost certainly a noun. If evi really means "on the" then we might expect a unit of time (even though I want units of time to start with t-)
- pulgarinko: Probably an adjective, like nedhínki "brave," and HV Qupténkos "Common."