Skip navigation

An Essay on the Dragon Language

17 replies [Last post]
Joined: 01/25/2012

 A Preliminary Examination of the Tongue of Alduin’s Kin


An Early Treatise on the Language of Dragons


Apprentice Historian M. Khan



It is worth beginning this examination by saying that this small work is not meant to analyze the “written form” of Dragon Language, as of yet unnamed, out of care for my own scholarly credentials. I have neither the time or the funds, as of yet, to do as I would like and charter an adventurer to find inscriptions of all the Dragon Walls over the province of Skyrim. Even with the defeat of Alduin in Sovngarde, travel is still dangerous, and so as an Apprentice Historian I shall remain safely ensconced in the halls of the college.


The Dragon Language is a particularly interesting study. It’s words can easily be spoken by a human mouth as much as by the jaw of a dragon, and the sounds are said to have a great power in and of themselves. I have taken the liberty of using the lexicon of known dragon terms to define the speech of Dragons into two forms: Thu’umzul and Dovahzul. The first is the “Shout voice” more commonly known as simply the Voice, and practiced by the Greybeards of High Hrothgar.  The latter, Dovahzul, or “dragon voice” is the term I use for the language as a whole. The distinction is purely academic, as the lexicon is the same, but there is a big difference in the use of the words as a “power” and their use purely for the conveyance of thought. For dragons, I hasten to add, there distinction is even more immaterial. Both are one for the Dovah, as are they for the Dovahkiin.


In the analysis of dovahzul there are several facts that we must wrestle with. First, the language is incomplete. Much can be said in it as it currently stands, but it is a tongue for use by dragons in their everyday lives and so will not be a replacement for Tamrielic speech yet! In translation, one often has to determine the “concept” being conveyed, rather than translate the actual physical words. Second, the grammar of the speech has to be determined backwards… I have two examples for this exercise, but until I manage my full exploration of written examples this is all that I have.


Let us concern ourselves with the first passage as a starting excercise. This purports to be a record of a conversation with Alduin himself, along with a translation. I doubt the veracity of that statement, but it comes from an unimpeachable, excellent source provenance (UESP as we Apprentices are wont to say), and the words are certainly Dovahzul! Let us examine the first words:


Alduin: Sahloknir, ziil gro dovah ulse!

Sahloknir, ever-bound dragon spirit!

(Spirit bound dragon eternity-of)


Clearly Sahloknir (a name which literally translates as “phantom-sky-hunt”, a concept that thus far eludes me to translate well into Tamrielic. The best I can do is “Ghost Hunter”, clearly a deadly but stealthy predator on the wing.) is here receiving an address from Alduin (destroyer-devourer-master, far easier to translate the concept!) Why that would be the case is unclear, because the rest of the phrase seems to suggest that Saloknir is well beyond mortal hearing. The rest of the translation is above, along with the necessary conceptual transfer, gives the impression that the dragon has been bound eternally. (NB. My lexicon does not have the word “gro” included, so the translation of “bound” is dubious, assumed, or my lexicon is incomplete. The “ziil” too is confusing, as the word ‘spirit’ lacks the final ‘l’ when spelt phonetically. Again the scholarly provenance is not doubted, so it may be that my lexicon contains the mistake. Or, in revisiting this passage, the “l” represents “your”, so “eternally bound dragon spirit of yours”? This is mere conjecture, but I will make many more before I am done!) The only other point worth noting is that the adjective “ever-bound” is placed at the end of the sentence, so the translation would be “bound dragon spirit” without the “ulse” at the end, itself a conjunction of “ul” eternity, and “se” of. That latter word “ulse” may be better translated as “eternally” rather than ever-bound, but the meaning is largely the same and a question of preference. Therefore we learn two things: first, adjectives come at the end in dovahzul, perhaps because to add them is mere rhetoric amongst dragons, and second that dragon words are oft conjoined to bring these adjectives into play.


Alduin: Slen Tiid Vo!

Let your flesh be unrotten!

(Flesh time opposite-of)


This passage is relatively simply, and reflects a use of Thu’umzul rather than dovahzul. The speech here is imperative, and so lacks any grammatical extras except the simple command of the words’ literal translation. This is characteristic of Thu’umzul. The conceptual leap in the above phrase (I have edited the translation of the word “vo” to match my lexicon) is indicative of the mental leaps that have to be made. The “opposite of” time is naturally to travel backwards in time, and since it is flesh that is traveling backwards in time, un-rotten is the result. Another example of the latter word changing the meaning of the sentence as a whole. We need spend no further time on this passage beyond incredulous eyebrow raising at the apparent nature of the command.  (Perhaps I should begin to doubt the translator? This is either a practical joke, or it is a metaphorical tale.)


Sahloknir: Alduin, thuri! Boaan tiid vokriiha suleyksejun kruziik?

Alduin, my overlord! An age past, did you not destroy the power of the ancient kings?

(Alduin, lord! Age-an time opposite-of-kill-[something] power-of-kings ancient)


And yet our poor brains shall not receive any more respite. For here the apparently long dead arise and address Alduin. Double checking with my lexicon, it appears more translational leaps have been made. “Boaan” must be “Bok-aan” a conjunction of “age” and “an” to form “an age.” Tiid we know to mean time. The next word eludes me, “vo” is “opposite-of” and “krii” means kill, but the “ha” eludes me. I have no idea how or why this word is joined together. The opposite of kill is obviously resurrect, but that changes the apparent meaning of the passage. A tentative translation to achieve the desired affect (the conceptual translation above, if we take it as correct) is to assume that the “ha” indicates that it is a question. The question mark added is a human and superfluous addition, so the logical leap is that the “question signifier” is added to a verb to make it an “un-surity” (forgive the word!) much as we would say “did you not?”


Alduin: Geh, Sahloknir, kaali mir.

Yes, Sahloknir, my trusted ally.

(Yes, Sahloknir, champion-[something] allegiance)


Alduin’s response requires another small leap of deduction. The literal translation makes the conceptual translation obvious, except that “kaal” has an ‘I’ added to it. This, as far as I can deduce, therefore indicates possession. Sahloknir is Alduin’s champion, much as we place an apostrophe and an “s” after the person who owns an object, or “my” before a noun. In dovahzul, the “my” comes after the noun, much as “of” comes after an adjective.


Alduin: Ful, losei Dovahkiin? Zu'u koraav nid nol dov do hi.

So, my false Dragonborn? Your voice has nothing of the Dragon about you.

(So, decieve-of-[my] Dragonborn? Voice-[something] [something]-join no from dragon about [something])


Here Alduin addresses the apparently present Dovakhiin. Once again, my conjecture as to the meaning of ‘I’ to a word to indicate possession appears to be correct on the “losei.”  The “’u” eludes me on the voice, replacing the “l” of the usual word. “Kor” is likewise not in my lexicon, although the translator determines it to be “in” which seems acceptable. “Hi” at the end of the sentence appears to be a contraction of “hin,” “your”, perhaps “you” would be suitable. The end result perhaps is to say that “zu’u” stands for “language” or indicates power or something similar (Taking the “‘u” in “Thu’um” to indicate power? Perhaps “dovahzul” is wrong then and it should be “dovahzu’u”? I need to address this to some other scholars for their opinions.) So the ending literal translation would be “language in-join no from dragon about you.” I have edited the conceptual translation above to reflect my findings, if you wish to look at the original please do, but I seriously believed the previous translation to be lacking (the translation of “do” as “of” was a serious one. I do not see why it would replace “se” and so assumed it must be “about” only, and not both as my lexicon informed me.)


Alduin: You do not even know our tongue, do you? Such arrogance, to dare take for yourself the name of Dovah.


A Tamrielic interruption of what is in all else a purely dovahzul work. I must presume it was meant to stand out when this now certainly metaphorical passage was written down. There is nothing to comment on here.


Alduin: Sahloknir, krii daar joorre.

Sahloknir, kill these mortals.


This instruction is clear, and the words are in fact identical in dragon and tamrielic. Therefore I shall not elaborate further.



Thus I come to the end of this first elaboration on the construction of dovahzul. A lot of my findings are conjecture and guesswork based on previous translations, but do someway enhance our understanding I hope. My next aim will be too examine a more common example of dovahzul, which can be found in the revised version of the Songs of Skyrim. It is what appears to be a paean to the Dragonborn. I shall not attempt an examination of it now, for fear of overextending my abilities, and therefore, I present this treatise first for scholarly perusal. If it appears that my findings and work are of use, then I shall attempt further work on the subject. Finally, I must give a vote of thanks, first to the UESP of my analysed source, and secondly to the Great Creators of Nirn.


  I rest my pen: M. Khan.


            Aal fin suleyk ahrk dun do pah bormah aak hi.



Joined: 01/25/2012

Any feedback anyone? Comments? 

Proweler's picture
Joined: 06/14/2010

Not yet. There are five Usability Issues papers waiting before of you. :)

Joined: 10/11/2011

Very interesting article, well written too.

Some of my thoughts:

I agree with your theory of the -ha ending marking a question.

Could the -i ending in "kaali mir" and "losei Dovahkiin" be a kind vocative ending as in Latin? In English this could be translated as "my" as a form of reference as English lacks grammatical vocative.

In "zu'u" the ending -u is a strange one, but I guess it is some kind of derivative suffix forming a noun out of a root word that cannot work as a noun on its own, a bit like in English where you can't say "hairy" as a noun, but you have to say "hairy one". There could be more of these derivative suffixes, but this might be one of them. In Thu'um the -um ending could be a referential form of this u-suffix.

"Ful, losei Dovahkiin? Zu'u koraav nid nol dov do hi" is an interesting and complex example of the language. "Hi" could be a dative form of second person singular personal pronoun. The word "koraav", seemingly being derived from in+join could be a positional word with the meaning of something like "among, on."

dinmenel's picture
Joined: 09/11/2011

Is there a source for "zu'u"? You have it as a form of "zul", but both wikis have it as "I".

Joined: 01/25/2012

Thanks very much for your comments both!

Hrafnir, I agree with your suggestions. Calling the "i" ending a vocative ending certainy fits the appearance of the language. "Zu'u" really was a difficult one, thanks for your thoughts. Again thanks for your input!

Dinmenel, I believed the translation of "zu'u" on the wiki to be wrong, simply because the meaning of the sentence simply wouldn't work otherwise. Besides, there are other words that can provide the neccessary "I" if that were the intended meaning.

All in all I think it will be very difficult ever to get a complete picture of the language however. The Song of Skyrim, which i had intended to examine next, is simply a cipher. And without getting transcripts of spoken dragon tongue or analysing the dragon walls (if they are different, they may just all be identical I don't know) that's where I have to stop. It appears this essay will be both preliminary and last word!

Thanks for the comments, I'd be grateful for any more!

Aal fin suleyk ahrk dun do pah bormah aak hi.

Joined: 01/25/2012

Just thought I'd add to that to say it is all just theory, so feel free to call me wrong!

All the best everyone!

Aal fin suleyk ahrk dun do pah bormah aak hi.

Joined: 10/11/2011

Keep up the good work! Analyzing the song of the Dragonborn (search uespwiki for Songs of Skyrim) could reveal more of the syntax and grammatical rules of the language, and I wish to see your analysis. I can spend some time deciphering the word walls while playing Skyrim this weekend. It will be interesting to see if they all have the same text, and see a partial or complete translation of the walls. If you have time, you can also check my Yoku Language article and tell me if you have any ideas on it.


I was also told that there is a whole wall written in Falmeris in Skyrim, it just can't get to see it. I wish someone would post me either the original text with the Elven alphabet or a good, large screentshot or two taken directly from the front. I'd start analyzing the Falmeri grammar and vocabulary immediately. Anyone?

dinmenel's picture
Joined: 09/11/2011



EDIT: Well, fuck you too Dropbox. Here, just use this.

Joined: 10/11/2011

Did you post some screenshots? I can see only three red X's, but I guess they are pictures.


Anyway, I also found some Dragon language material for you. Look for a in-game book from Skyrim here in the Imperial Library by the name of "Dragon Language: Myth no More" which contains many good examples of the language; perhaps new words and at least more examples of grammar.


I also had time to check (and decipher and translate) one Word Wall, the one near the Shrine of Meridia. It says in Dragonic "Het nok Fjalmar bein [unknown word] pook ol pogaas nau gol ol ok kopraan dreh nuk golt" and translated literally with the words from Uespwiki Dragonic vocabulary with the same word order as in the original text, it means "Here lies Fjalmar foul [unknown word]; stink as much on earth as his body does in ground." But you might be able to give better and more precise translation.


The undeciphered word was something like SUW- [four letters long], but since I found no such word anywhere on the known Dragonic material, I started to doubt my translation of that word. I have to go back and check that word again.

dinmenel's picture
Joined: 09/11/2011


Fiore1300's picture
Joined: 07/24/2011

I remember Calcemo's wall. Looks like a Rosetta stone of sorts. Dwemeris and Falmeris, right?

Joined: 10/11/2011

Thank you! I'll start working with these as soon as I get home from work. It is indeed the Rosetta stone of Elven languages, with Dwemeris and Falmeris. Still the translation will not be easy as both languages are unknown. Have to start by analyzing the grammatical features, then start using everything that is known from Ayleid and the other Mer languages, which is not too much.

Dont Forget This's picture
Joined: 10/02/2010

Dwemeris IS ayliedic in so much as the letters are the same. It can be assumed they have the same attached meaning as well. As such Dwemeris is a know script, but the words themselves are unknown. If is likely simialer to Ayliedic, as they would both be VERY close to Aldmeri. Falmeri should also be close to Aldmeri. The thing is we don't know Aldmeri. 

Joined: 10/11/2011

All Elven languages and Dwemeri seem to use more or less the same alphabet, but that does not mean the languages are similar or even related. I have analyzed the texts (the upper part is in Dwemeri and the lower is in Falmeri) and I have partial translations and some features of grammar for both of them, but I\'m not at all sure the Dwemeri is related to the other Elven languages, not closely at least. It has some words that can perhaps be congates to words in the other languages, but I have to do a lot more research on the Dwemeri before I can say anything.


Falmeri and Ayleidic are both very early Elven languages and at that early stage they are basically dialects of Aldmeri, with small differences in sounds, grammar and vocabulary. But as can be seen from the known examples of place names, personal names, vocabulary and grammar from the modern Elven languages (Bosmeri, Altmeri, Dunmeri, Ashalander) the Elven languages have changed a lot during the many millenia after the early First Era. Although my analysis into Altmeri names etc. give a hint that their language has changed very little from the Aldmeri, perhaps a consious desicion by the Altmer.


The interesting thing is that using the grammar of Ayleidic and Falmeri and what little is known of the modern Elven languages, I can "reconstruct" the Aldmeri grammar as fan fiction, filling in the details with my imagination. After my fan fiction version of Aldmeri grammar and vocabulary is ready, I can create the modern languages for my roleplaying purposes.

Joined: 07/13/2012

Does the Dragon tongue have a bearing on modern day Tamerialic languages and culture. We Dragons were major influences on the Akvair. I personally think it may have some impact on portions of Tamerial that were ruled by the Akvair. 

Joined: 02/04/2013

I agree with the translation of the first phrase, but perhaps instead of "Let your flesh be un-rotten", it could be "Let your flesh be restored."  Perhaps in the phrase "Ziil gro dovah ulse", Ever-bound may mean Immortal, making the entire phrase "Ghostly/Ghastly Sky Hunter, immortal dragon spirit, let your flesh be restored."

the u'u does, in fact, indicate power in Thu'um whilst Dovahzul didn't demonstrate any such thing. Thus I believe Dovahzul is the proper term for Dragonspeak

Kondar95's picture
Joined: 11/02/2016

Sorry for necroïng the post. I just thought I'd mention to anybody who's interested and doesn't know about it, a site for learning Dovahzul. is the name, and it's the best and only (as far as I know) site for learning the language, though there is a Memrise course for it as well.