The mind of the Dark Lord

Since I had so little direct linguistic information about Black Speech to go on other than what could be gleaned from the Ring-inscription (object suffixes -ul, -ulûk; verbal infinitive (perhaps) ending -at; abstract ending -um in burzum “darkness”, containing the same burz element seen in Lugbúrz “Dark Tower”; postposition –ishi “in”) I had to go on à priori notions of what a language such as Black Speech might be like — I had to get inside the mind of Sauron, and try to figure out what somebody like the Dark Lord of Mordor might put into his language.

As a matter of fact, this is something I had thought about some years before. As an undergraduate in college, I had contributed to a set of ongoing stories, where each participant wrote additional chapters and introduced characters and events as he or she pleased. Into one of these stories I introduced the character (played partly for humor, partly tragically) of a misfit Orc who, sometime after the fall of Mordor, had found himself transported through time and space into a new environment. On introducing this Orc, I thought it would add a touch of realism to let him speak in his own language; so I sketched the outline of what I imagined Black Speech might be like, and wrote a couple of paragraphs in it.

I have no idea if any copy of this text survives somewhere in my files. At any rate, I made no direct use of it, except for one small element that I retained in memory, the 1st person pronoun za — possibly suggested by Avestan azəm.

What I did retain, however, was the overall notion of Black Speech as a complex but consistent language, rich in affixation and inflection, but with a wholly transparent morphology. Indeed, the transparency of the morphology, the lack of any phonetic alterations between morphemes that could obscure the structure, would help explain the prevalence of clashing consonant clusters; morphemes ending in one consonant were jammed up against morphemes beginning in another, with nothing to ease the transition.

Sauron, I imagined, was an enormously practical person, who would have made the Black Speech as “perfect” (according to his notions of perfection) as he could make it, with a rigorous consistency and logic, but without making any allowance for æsthetics. It would not eschew borrowings from other languages of Middle-earth, but it would adapt them to its own style. It would in fact have been, as my friend Helge Fauskanger terms it, Sauron’s Esperanto.

Whether I actually managed to capture this vision of Sauron’s mind in my version of Black Speech may be doubted. I largely lack that sort of mental rigor, and when it comes to language I often prefer sloppiness to tight organization. Nonetheless, I think I succeeded in making my version of Black Speech a bit more consistent than most of the other languages of this milieu, though not without its own quirks.

In sharp contrast to the relatively tight organization of Black Speech, the Orkish languages were to be simple, disorganized, and inconsistent, the result of years of rapid and ungoverned evolution. There would be grammar, of course, but also a fair amount of toleration of variation, and a continuous tendency to proliferate new words and abandon old ones. They would show a strong influence from Black Speech, at various stages in their development; but they would fail to adhere by its rules.

For the Lord of the Rings films I intended to come up with three Orkish dialects in addition to Black Speech: one, spoken by the Orcs of Mordor, would be closely based on Black Speech, but spoken in a more casual and clipped manner. The other two, the Orkish of Isengard and the Orkish of Moria, ideally should have been developed as wholly independent languages, touched by Black Speech only through borrowings. Faced with rapidly approaching deadlines, however, I cheated; I made all of them descendants of Black Speech, via a hypothetical Proto-Orkish, with the language of Isengard showing several distinguishing sound changes, and the language of Moria showing an even more advanced set of sound changes, intended to give the Moria-goblins a hissing, sibilant sound.

The last was not actually my own idea; I received a message (February 20, 2001) from writers Philippa Boyens and Fran Walsh, asking for the following characteristics of the Orcs:

Mordor Orcs – harsh and guttural, reflecting the barrenness of Gorgoroth.
Moria Orcs – more secretive, whispery, hissing, reflecting the darkness of the tunnels.
For the Uruk-hai we want powerful sounds that reflect aggression and hatred, also a discipline to the language not shown by the lesser Orcs.

Since “harsh and guttural” and “aggression and hatred” were likely to be associated with these languages in any case, I didn’t have to do much with regard to the dialects of Mordor and Isengard other than to render them distinct; but for Moria I had to implement sound-changes (of a somewhat improbable nature) that considerably increased the proportion of fricatives.

By way of illustration, I copy a list of words which I wrote out at the beginning of this task. From *lutu- down this is actually a rationalization of miscellaneous jottings, not necessarily consistent with the rest of this list or even with themselves; they were probably added later. You can also observe that I was not very consistent with my choice of symbols.

Meaning Proto-Orkish Mordor Isengard Moria
sky *kilmi kilᵊm kʰīm ʃīm
sun *ūru ūr ūr ūʒ
tree *turu tur tʰur suʒ
foot *tulu tul tʰul sū(l)
tooth *nakṛ nakur nagū pl. nāgā noɣ
flesh *marna marn mān mān
bite *naka- naka- naga- naɣa-
beat *bada- bada- bara- vaza-
whip *badgu badg bāg vāɣ
fight *kutkuta- kukut- kūkda- χūχza-
kill *daka- daka- daga- zaɣa-
war *kutmu kutum kūm χūm
die *guru- guru- gura- ɣur-
death *gurutu gurut gūt ɣūʃ
heave *giba- giba- giva- ʒīa-
pull *tugu- tugu- tʰuɣu- sū-
hack *skada- skada- kʰara- χaza-
slash *kliza- kliza- kʰiza- χliza-
command *durbu- durbu- dūbu- zū-
light *gāra gār gār ɣāʒ
day *garmu gārum gām ɣām
to daw [sic] *garmuza- garmuza- gāmza- ɣānza-
dawn *garmuzata- garmuzat gāmzad ɣānzaz
arise *huru- huru- huru- uʒu-
burn *laka- laka- laga- laɣa-
burn (caus) *lakja- lača- laiga- laʒa-
feel, touch *maka- maka- maga- maɣa-
want/will *hizi- hizi- hizi- iʒi-
hurry *klikja- kliča- kʰīga- χīɣa-
come *lutu- lutu- ludu- ruzu-
rest luzu-
close *karba- kāba- χaχāv-
flee *drigi- drigi- diɣi- dī-
work *bulu-
wait, stay *dara- dara- dara- zara-
feast *mamata- mamata- manda-
scream *skriki- kʰigi-
screaming *skrikikai χriχa
crawl *smugu-  smugu- muɣu- šmū-
crawling *smuguku muɣgu šmūɣ
gnash *karka- kāka- χāχa-
see *guglu- gūgu- ɣūɣu-
more *tʰag
get *snaba- nava- ʃnā-
mountain *urudu
come *nakʰ
return *agnakʰ
army *kʰotʰ
destroy *gutja- gūda-
hard *kraka kʰag

Isengard kūm is doubtless a mistake for kʰūm. There are likely other errors as well.

I did not write down any details of the sound changes, since I presumably made them up as I went along; nonetheless they appear to be fairly regular. Mordor-Orkish loses the final vowels in nouns, and adds an epenthetic vowel in some final consonant clusters. Isengard-Orkish aspirates initial voiceless stops and voices medial ones, while turning medial voiced stops to voiced fricatives (but -d- to -r-), drops preconsonantal r with compensatory lengthening, and has a variety of other assimilations. Moria-Orkish assimilates like Isengard-Orkish, but also turns almost all stops to fricatives.

24 Responses to “The mind of the Dark Lord”

  1. evenstar

    This was a very interesting article to read, and I would like to know how you were able to figure out all these words if Tolkien hadn’t developed this language to the same extent as Sindarin.

    • David Salo

      I figured them out by making them up. Language creation is a game anybody can play, after all.

  2. david

    Could it not be that the language of Sauron was Númenorean and also for the Orcs, because his most loyal servants were Nazgûl? He came probably from Númenor before the Fall. Sauron spoke with Númenoreans in the Akallabêth.
    This could explain why Gandalf knew few words of the Orkish. Gandalf never had contact with Orcs but with mankind. Also, a clue is that Gandalf was called Incánus which means something like “spy.” He must have been once in the far south of Harad at least.
    The name Sauron was given by the Sindar in Sindarin. Also, in The Lord of the Rings most Orcs spoke Westron because they did not understand each others’ language.

    • David Salo

      It can hardly be doubted that Sauron knew the Adûnaic language spoken by the Númenorean people — he spent years on the island of Númenor, after all — and it is possible that he used elements of it in the Black Speech; however, the two are clearly distinct. The phonetic styles of the extant examples are quite different, and Tolkien is quite clear that Sauron invented the Black Speech and did not merely adapt an existing language to his uses.

      Three of the Nazgûl are said in the Akallabêth to have been “great lords of Númenórean race”; but that implies that six of them weren’t. The Nazgûl as a group seem to have adopted the Black Speech, as they are said to have been the only ones to remember it during the period of Sauron’s disembodiment in the early Third Age.

  3. Mad Latinist

    Well, I had been wondering about the Morian Goblin language in the Hobbit. From what little I had heard, I thought you had invented it entirely for the new films, but I guess you started on it a long time ago.

    Also, ahem, *turu?

    OMG OMG! Proto-Orkish borrowed a word from Caryatic!!

    • David Salo

      You hadn’t heard wrong. The goblin-language in the film of The Hobbit is a new invention. It’s still supposed to have a relationship to the Black Speech (which is the same Black Speech) but is more remote and the relationship is more complex than that between Black Speech and the Orc-languages in The Lord of the Rings. In other words, it’s more like what I should have done, and would have done with the Orc-languages in the earlier films if I had had the time.

      The theoretical justification for this procedure is that the goblins of the northern end of the Misty Mountains belong to a different branch of the family, less influenced, perhaps, by Black Speech, or at an earlier phase in their evolution. Their language is presumably closer akin to that of the Orcs of the Ered Mithrin (of which they may be regarded as a southern branch), while the Orcs who infested Moria in the time that the Fellowship visited it had evidently repopulated it from the South some time after the War of the Dwarves and Orcs, their language being essentially an eccentric variant of the Orkish spoken in Isengard, which was doubtless a standardized synthesis (under Saruman’s guidance) of some closely-related Orkish dialects of the southern Hithaeglir.

  4. Nicolás

    I like it. It seems fitting that Moria-Orkish should be the most divergent of the three. I would have thought all Isengard-Orcs were Uruk-hai, as he created them there, maybe the other orcs in Isengard were sent from Mordor directly. While creating the dialects did you have in mind the Mordor-Orc’s curse, the one of “pushdug “? Also have you incorporated the sharkû/Sharkey thing into the dialects?

    • David Salo

      I did consider the “bagronk” curse, but made little use of it for two reasons. First, it consists almost entirely of expletives, and has little communicative value, which is a problem when you’re being called on to translate lines that convey a bit more meaning. Second, Tolkien gave it three largely contradictory and in some places incommensurate readings: for instance, bagronk is translated as “torture,” “cesspool,” and “dungpit”; but “dung-heap” in one version is búbhosh, “dungfilth” in another is pushdug, and “dung-pit,” as mentioned, is bagronk in a third. That’s a lot of ways to say “dung”! With Tolkien himself apparently of three minds on the question, I felt pretty safe in largely ignoring it. I think I may have used skai (a meaningless interjection in all three renderings) somewhere, and possibly –glob, but I am not sure about either.

      • Nicolás

        Haha, true, don’t you hate those multiple contradicting translations? But certainly it’s interesting as to word-structure and phonotactical inspiration, right?

  5. H.K. Fauskanger

    You refer to the “abstract ending –um in burzum ‘darkness'”.

    Well, that is what everybody thought before the publication of Tolkien’s notes in Parma 17. According to PE17:12, this –um is actually a suffixed article, or “particularizing” element. Hence burzum = “the dark”. Both Westron and the Black Speech have a suffixed article, then.

    • David Salo

      There are two different sets of glosses or interpretations of the Ring-inscription, and they don’t agree with each other, at this point and at one other*. In the first burzum is simply glossed “darkness (cf. búrz, dark, adjective).” This gloss does not remotely suggest definiteness or any kind of particularity. The second calls –um “a particularizing suffix or ‘article'” (which are not exactly the same thing), and the scare quotes around ‘article’ don’t inspire confidence! Moreover, the “debased” version of Black Speech used by the Mordor-orc in his curse has no –ums, despite (in some of its translations) having one or more articles.
      However, functionally I am not sure there’s much difference. At any rate, in my versions of Orkish –um (or its descendants) does function as an abstract suffix; not an unreasonable historical development.

      *The other point is about subject affixes; the first set of glosses says that “particles indicating ‘subject’ were usually prefixed,” while the second refers to “pronominal suffixes expressing the object, as well as those indicating the subject,” which would seem to indicate that subject-agreement was marked by suffixes. The latter is the way that my version of Black Speech turned out.

  6. Demosthenes

    I’ll preface my comment by IANAL(inguist) but in my very inexpert wanderings through Tolkien’s languages I did feel there was some surface resemblance (in look/sound?) between the fragments of Valarin that exist, and the Black Speech. Neither are “pretty” in terms of how they sound to the ear.

    Is there actually any linguistic basis to this? I mean, as a point of logic I imagine it might be feasible for Sauron to take the language of the Valar and running with it to create his own little God-tongue… I think this would be, internally, more likely than, say, Adunaic as a precursor language… but if the linguistic evidence isn’t there, then it isn’t there…

    • David Salo

      There are some resemblances, as indeed there might be between any two languages, randomly chosen, and doubtless more because of Tolkien’s tendency to work within a restricted range of styles. An obvious similarity is the presence of “clashing” consonant groups which might be simplified or otherwise resolved in an Eldarin language: šk, lg, and so forth; also such sounds as z and š, not found in Noldorin Quenya or Sindarin (both are in Vanyarin Quenya, however).
      However, Valarin seems to have a much wider range of sounds than Black Speech: notably voiceless aspirates as distinct from fricatives (e.g. both tʰ and θ, kʰ and χ) and perhaps even voiced aspirates, if that is how we are to interpret “bh”. Front vowels are more common than in Black Speech, and there are æ and ǭ, which don’t exist in Black Speech at all. Open syllables are far more common in Valarin; final consonant clusters (like -zg) don’t exist at all in Valarin, exemplified finals being vowels and d, l, m, n, r, s, š, th, z. These are for the most part coronals; there are no examples of the final velars so common in Black Speech.

      The most likely connection between Black Speech and Valarin is in the Valarin word (a)naškād, probably = “ring,” which can be compared with Black Speech nazg. There is a rival etymology of nazg from an Eldarin root √nasag– > *nazgā, Quenya naxa “bond, fetter”; but strictly speaking these two etymologies are not necessarily inconsistent, as the Eldarin root could have been borrowed from Valarin (though this is not actually stated anywhere that I know of).

      Probably the Black Speech was — where not wholly original — made of whatever materials came to hand and appeared suitable (phonetically or structurally) for the purpose; in the early Second Age, that could include Quenya, Sindarin, the Mannish languages of West, East, and South, and Khuzdul — as well as whatever remained of the Orkish languages of the First Age (I assume that words like golug “elf,” itself probably a modification of Sindarin golodh, were retained). Presumably whatever Sauron knew of Valarin could be an influence as well, but as Sauron entered the service of Morgoth very early, long before the awakening of the Quendi, it is possible that the Valarin language as the Eldar later knew it had not yet been developed and was largely unknown to him; though presumably Morgoth himself would have had to learn it, and Sauron could have then learned it at second hand, if Morgoth were disposed to teach him.

  7. Rebecca

    Did you ever think about making an alphabet like some have done with Elvish? I’m merely just curious. This article was very interesting to read and I’m 100% interested in LotR language, made up by fans or made up by Tolkien.

    • David Salo

      I have invented many alphabets in my life, though never an “elvish” one. For writing Quenya or Sindarin, if I’m not using Tolkien’s romanization (or a more phonetic transcription), I make use of one or another of Tolkien’s own modes for writing these languages, or sometimes the runic Certhas Daeron for Sindarin. For writing Khuzdul I have used either the Angerthas Moria or its Ereborian modification. I did invent a writing system for one Orkish dialect of my devising, but I have no idea if it will ever see the light of day. At the moment I can only describe its style as not particularly beautiful, though perhaps not lacking in all interest.

  8. Tigran

    Have you ever thought of making a full Black Speech grammar and a vocabulary? Judging by the films, you’ve accumulated a lot of material.
    Could you analyze in detail the Ring verse, the orc cry from “Two Towers” (za dashu snaku…), and the rest phrases on BS? Especially I’m interested in postpositions you’ve invented, like “-izin”, “-uk”, “-shu”…
    And on what principle do you create new words?

    • David Salo

      I’ll be writing a bit more about Black Speech in future blog posts, including all of the elements you’ve mentioned.

      I’ve talked a bit about my language-creating processes in earlier posts. I suppose I could say (with Tolkien) that “when you invent a language you more or less catch it out of the air. You say boo-hoo and that means something.” But in fact any method of creating vocabulary that is not totally randomized, and even more so selecting appropriate words out of the various sound-groupings that are possible according to the rules you’ve made, cannot be done without some influence from one’s own language or other languages that you know. You can try to suppress it, or deliberately alter reminiscent words so that they no longer look too much like their sources; but I find that too much of that tends to dry up inspiration while leaving me just as conflicted as before over the “proper” word to choose. So when looking for a sound to fit a particular meaning, I let various sound-groups float into mind — some of whose sources are obvious to me, others of which are more opaque (which, however, doesn’t mean they don’t have a source). Then I pick one (or more) of them; sometimes I find I’m not quite happy with the result, and I change a vowel or a consonant here and there until it feels more natural to me (and I can hardly define what “naturalness” means in that context). As far as real-world influences go, the only countermeasure I take is to try not to let a word look too obviously like a borrowing from a well-known real-world language at first glance; I don’t much care if the influence can be detected at 2nd or 3rd glance, as I figure that’s all part of the fun. In fact, I’ve gone to some pains to point out the sources of real-world influence on this blog, when I can remember them, or deduce them after the fact.

      In these languages, there’s also a lot of influence from other languages invented by Tolkien; but as I think there’s some in-world justification for this, and even some canonical examples (like Sindarin golodh > Orkish golug; perhaps even Valarin uruš > Orkish ghâsh), I haven’t tried to avoid it at all.

    • David Salo

      Yes, I will be making some posts about Black Speech in the future (I hope the near future).

      As for other fans’ inventions, as far as I am concerned they are welcome to come up with their own version of Black Speech (or anything else) just as I came up with mine, and I hope that such exercises in language invention give satisfaction to their creators. But otherwise I think it would be very bad manners for me to comment on their work.

    • Brian

      The biggest glaring thing I’m seeing is that they aren’t using any form of phonetic alphabet to clarify any possible confusion over pronunciation or place of articulation.

  9. Ovi


    In the movie there is a scene in which a vexed Sauron passes through Azog saying “Death will come to all”. I think I can hear quite clearly the word snabalak or snabanakh (probably means “will get”). The other words are harder to get. Death is “gurutu”, but I’m quite certain that Sauron doesn’t utter that word. So what exactly does he say?

    • David Salo

      At the moment I can’t find anything quite like that in my notes. I’ll make a note if anything turns up as I continue to go through everything.

    • David Salo

      He doesn’t. The usual word that’s used is zidgu. The only place where a word related to sharkû appears in my translations, it means “old man.”


Leave a Reply

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>