The Letters of J.R.R. Tolkien

by Humphrey Carpenter

  Investigators seem commonly to neglect this fundamental point, although sufficient evidence of ‘linguistic construction’ is provided in the book and in the appendices. It should be obvious that if it is possible to compose fragments of verse in Quenya and Sindarin, those languages (and their relations one to another) must have reached a fairly high degree of organization – though of course, far from completeness, either in vocabulary, or in idiom. It is therefore idle to compare chance-similarities between names made from ‘Elvish tongues’ and words in exterior ‘real’ languages, especially if this is supposed to have any bearing on the meaning or ideas in my story. To take a frequent case: there is no linguistic connexion, and therefore no connexion in significance, between Sauron a contemporary form of an older *θaurond-derivative of an adjectival *θaurā (from a base √THAW ‘detestable’, and the Greek σαύρα ‘a lizard’.

  Investigators, indeed, seem mostly confused in mind between (a) the meaning of names within, and appropriate to, my story and belonging to a fictional ‘historic’ construction, and (b) the origins or sources in my mind, exterior to the story, of the forms of these names. As to (a) they are of course given sufficient information, though they often neglect what is provided. I regret it, but there is no substitute for me, while I am alive. I have composed a commentary on the nomenclature for the use of translators;1 but this is directed primarily to indicating what words and names can and should be translated into L(anguage) of T(ranslation) which takes over the function from English of representing the C(ommon) S(peech) of the period, it being understood that names not in or derived from mod. English should be retained without change in translation, since they are alien both to the original C.S. and to the L.T. Desirable would be an onomasticon giving the meaning and derivation of all names and indicating the languages that they belong to. Also of interest to some, and agreeable to me, would be an historical grammar of Quenya and Sindarin and a fairly extensive etymological vocabulary of these languages of course far from ‘complete’, but not limited to words found in the tales. But I do not intend to engage in these projects, until my mythology and legends are completed. Meanwhile dealing piecemeal with guesses and interpretations only postpones and interferes with this work. . . . .

  In illustration of my strictures, I will offer some comments on your specific queries and guesses. Theoden and Gimli. The reason for using ‘Anglo-Saxon’ in the nomenclature and occasional glimpses of the language of the Eorlingas – as a device of ‘translation’ – is given in Appendix F. From which it follows that ‘Anglo-Saxon’ is not only a ‘fertile field’, but the solefn112 field in which to look for the origin and meaning of words or names belonging to the speech of the Mark; and also that A-S will not be the source of words and names in any other languagefn113 – except for a few (all of which are explained) survivals in Hobbit-dialect derived from the region (The Vale of Anduin to the immediate north of Lórien) where that dialect of the Northmen developed its particular character. To which may be added Déagol and Sméagol; and the local names Gladden River, and the Gladden Fields, which contains A.S. glcedene ‘iris’, in my book supposed to refer to the ‘yellow flag’ growing in streams and marshes: sc. iris pseudacorus, and not iris foetidissima to which in mod. E. the name gladdon (sic) is usually given, at any rate by botanists. Outside this restricted field reference to A-S is entirely delusory.fn114

  As stated in the Appendices the ‘outer’ public names of the northern Dwarves were derived from the language of men in the far north not from that variety represented by A.S., and in consequence are given Scandinavian shape, as rough equivalents of the kinship and divergence of the contemporary dialects. A-S will have nothing to say about Gimli. Actually the poetic word gim in archaic O.N. verse is probably not related to gimm (an early loan < Latin gemma) ‘gem’, though possibly it was later associated with it: its meaning seems to have been ‘fire’.

  Legolas is translated Greenleaf (II 106, 154) a suitable name for a Woodland Elf, though one of royal and originally Sindarin line. ‘Fiery locks’ is entirely inappropriate: he was not a balrog! I think an investigator, not led astray by my supposed devotion to A-S, might have perceived the relation of the element -las to lassi ‘leaves’, in Galadriel’s lament, lasse-lanta ‘leaf-fall’ = autumn, III 386; and Eryn Lasgalen III 375. ‘Technically’ Legolas is a compound (according to rules) of S. laeg ‘viridis’ fresh and green, and go-lass ‘collection of leaves, foliage’.

  Rohan. I cannot understand why the name of a country (stated to be Elvish) should be associated with anything Germanic; still less with the only remotely similar O.N. rann ‘house’, which is incidentally not at all appropriate to a still partly mobile and nomadic people of horse-breeders! In their language (as represented) rann in any case would have the A-S form ran (
  Nazgul. There is no conceivable reason why a word from the Black Speech should have any connexions with A-S. It means ‘Ring-wraith’, and the element nazg is surely plainly identical with nazg ‘ring’ in the fiery inscription on the One Ring. I do not know any O.E. compound gael-naes, but in any case an inventor, engaged in rational linguistic constructions would not supplement a failure in inventiveness by reversing the order of elements in a word of a totally unconnected language, which had no appropriate meaning!

  Moria. Your remarks make me suspect that you are confusing Moria with Mordor: the latter was a desolate land, the former a magnificent complex of underground excavations. As to Moria you are told what it means, III 415, and that is an Elvish (actually Sindarin) name = Black Chasm. Does it not plainly contain the √MOR ‘dark, black’, seen in Mordor, Morgoth, Morannon, Morgul etc. (technically √MOR: *mori ‘dark(ness)’ = Q. more, S. môr; adj. *mornā = Q. morna, S. morn ‘dark’.) The ia is from Sind. iâ ‘void, abyss’ (√YAG: *yagā > S. iâ).

  As for the ‘land of Morīah’ (note stress): that has no connexion (even ‘externally’) whatsoever. Internally there is no conceivable connexion between the mining of Dwarves, and the story of Abraham. I utterly repudiate any such significances and symbolisms. My mind does not work that way; and (in my view) you are led astray by a purely fortuitous similarity, more obvious in spelling than speech, which cannot be justified from the real intended significance of my story.

  This leads to the matter of ‘external’ history: the actual way in which I came to light on or choose certain sequences of sound to use as names, before they were given a place inside the story. I think, as I said, this is unimportant: the labour involved in my setting out what I know and remember of the process, or in the guess-work of others, would be far greater than the worth of the results. The spoken forms would simply be mere audible forms, and when transferred to the prepared linguistic situation in my story would receive meaning and significance according to that situation, and to the nature of the story told. It would be entirely delusory to refer to the sources of the sound-combinations to discover any meanings overt or hidden. I remember much of this process – the influence of memory of names or words already known, or of ‘echoes’ in the linguistic memory, and few have been unconscious. Thus the names of the Dwarves in The Hobbit (and additions in the L.R.) are derived from the lists in Völuspá of the names of dvergar; but this is no key to the dwarf-legends in The L.R. The ‘dwarves’ of my legends are far nearer to the dwarfs of Germanic [legends] than are the Elves, but still in many ways very different from them. The legends of their dealings with Elves (and Men) in The Silmarillion, and in The L.R., and of the Orc-dwarf wars have no counterpart known to me. In Völuspá, Eikinskjaldi rendered Oakenshield is a separate name, not a nickname; and the use
of the name as a surname and the legend of its origin will not be found in Norse. Gandalfr is a dwarf-name in Völuspá!

  Rohan is a famous name, from Brittany, borne by an ancient proud and powerful family. I was aware of this, and liked its shape; but I had also (long before) invented the Elvish horse-word, and saw how Rohan could be accommodated to the linguistic situation as a late Sindarin name of the Mark (previously called Calenarðon ‘the (great) green region’) after its occupation by horsemen. Nothing in the history of Brittany will throw any light on the Eorlingas. Incidentally the ending -and (an), -end (en) in land-names no doubt owes something to such (romantic and other) names as Broceliand(e), but is perfectly in keeping with an already devised structure of primitive (common) Elvish (C.E.), or it would not have been used. The element (n)dor ‘land’, probably owes something to say such names as Labrador (a name that might as far as style and structure goes be Sindarin). But not to Scriptural Endor. This is a case in reverse, showing how ‘investigation’ without knowledge of the real events might go astray. Endor S. Ennor (cf. the collective pl. ennorath 1250) was invented as the Elvish equivalent of Middle-earth by combining the already devised en(ed) ‘middle’ and (n)dor ‘land (mass)’, producing a supposedly ancient compound Q. Endor, S. Ennor. When made I of course observed its accidental likeness to En-dor (I Sam. xxviii), but the congruence is in fact accidental, and therefore the necromantic witch consulted by Saul has no connexion or significance for The L.R. As is the case with Moria. In fact this first appeared in The Hobbit chap. 1. It was there, as I remember, a casual ‘echo’ of Soria Moria Castle in one of the Scandinavian tales translated by Dasent. (The tale had no interest for me: I had already forgotten it and have never since looked at it. It was thus merely the source of the sound-sequence moria, which might have been found or composed elsewhere.) I liked the sound-sequence; it alliterated with ‘mines’, and it connected itself with the MOR element in my linguistic construction.fn115

  I may mention two cases where I was not, at the time of making use of them, aware of ‘borrowing’, but where it is probable, but by no means certain, that the names were nonetheless ‘echoes’. Erech, the place where Isildur set the covenant-stone. This of course fits the style of the predominantly Sindarin nomenclature of Gondor (or it would not have been used), as it would do historically, even if it was, as it is now convenient to suppose, actually a pre-Númenórean name of long-forgotten meaning. Since naturally, as one interested in antiquity and notably in the history of languages and ‘writing’, I knew and had read a good deal about Mesopotamia, I must have known Erech the name of that most ancient city. Nonetheless at the time of writing L.R. Book V chs. II and IX (originally a continuous narrative, but divided for obvious constructional reasons) and devising a legend to provide for the separation of Aragorn from Gandalf, and his disappearance and unexpected return, I was probably more influenced by the important element ER (in Elvish) = ‘one, single, alone’. In any case the fact that Erech is a famous name is of no importance to The L.R. and no connexions in my mind or intention between Mesopotamia and the Númenóreans or their predecessors can be deduced.

  nazg: the word for ‘ring’ in the Black Speech. This was devised to be a vocable as distinct in style and phonetic content from words of the same meaning in Elvish, or in other real languages that are most familiar: English, Latin, Greek, etc. Though actual congruences (of form + sense) occur in unrelated real languages, and it is impossible in constructing imaginary languages from a limited number of component sounds to avoid such resemblances (if one tries to – I do not), it remains remarkable that nasc is the word for ‘ring’ in Gaelic (Irish: in Scottish usually written nasg). It also fits well in meaning, since it also means, and prob. originally meant, a bond, and can be used for an ‘obligation’. Nonetheless I only became aware, or again aware, of its existence recently in looking for something in a Gaelic dictionary. I have no liking at all for Gaelic from Old Irish downwards, as a language, but it is of course of great historical and philological interest, and I have at various times studied it. (With alas! very little success.) It is thus probable that nazg is actually derived from it, and this short, hard and clear vocable, sticking out from what seems to me (an unloving alien) a mushy language, became lodged in some corner of my linguistic memory.

  The most important name in this connexion is Eärendil. This name is in fact (as is obvious) derived from A-S éarendel. When first studying A-S professionally (1913 –) – I had done so as a boyish hobby when supposed to be learning Greek and Latin – I was struck by the great beauty of this word (or name), entirely coherent with the normal style of A-S, but euphonic to a peculiar degree in that pleasing but not ‘delectable’ language. Also its form strongly suggests that it is in origin a proper name and not a common noun. This is borne out by the obviously related forms in other Germanic languages; from which amid the confusions and debasements of late traditions it at least seems certain that it belonged to astronomical-myth, and was the name of a star or star-group. To my mind the A-S usesfn116,seem plainly to indicate that it was a star presaging the dawn (at any rate in English tradition): that is what we now call Venus: the morning-star as it may be seen shining brilliantly in the dawn, before the actual rising of the Sun. That is at any rate how I took it. Before 1914 I wrote a ‘poem’ upon Earendel who launched his ship like a bright spark from the havens of the Sun. I adopted him into my mythology – in which he became a prime figure as a mariner, and eventually as a herald star, and a sign of hope to men. Aiya Eärendil Elenion Ancalima (II 329) ‘hail Earendil brightest of Stars’ is derived at long remove from Éala Éarendel engla beorhtast. But the name could not be adopted just like that: it had to be accommodated to the Elvish linguistic situation, at the same time as a place for this person was made in legend. From this, far back in the history of ‘Elvish’, which was beginning, after many tentative starts in boyhood, to take definite shape at the time of the name’s adoption, arose eventually (a) the C.E. stem *AYAR ‘Sea’fn117, primarily applied to the Great Sea of the West, lying between Middle-earth, and Aman the Blessed Realm of the Valar; and (b) the element, or verbal base (N)DIL, ‘to love, be devoted to’ – describing the attitude of one to a person, thing, course or occupation to which one is devoted for its own sake.fn118 Earendil became a character in the earliest written (1916–17) of the major legends: The Fall of Gondolin, the greatest of the Pereldar ‘Half-elven’, son of Tuor of the most renowned House of the Edain, and Idril daughter of the King of Gondolin. Tuor had been visited by Ulmo one of the greatest Valar, the lord of seas and waters, and sent by him to Gondolin. The visitation had set in Tuor’s heart an insatiable sea-longing, hence the choice of name for his son, to whom this longing was transmitted. For the linking of this legend with the other major legends: the making of the Silmarils by Fëanor, their seizure by Morgoth, and the recapture of one only from his crown by Beren and Lúthien, and the coming of this into Earendil’s possession so that his voyages westward were at last successful, see I 204–6 and 246–249. (The attempt of Eärendil to cross Ëar was against the Ban of the Valar prohibiting all Men to attempt to set foot on Aman, and against the later special ban prohibiting the Exiled Elves, followers of the rebellious Fëanor, from return: referred to in Galadriel’s lament. The Valar listened to the pleading of Eärendil on behalf of Elves and Men (both his kin), and sent a great host to their aid. Morgoth was overthrown and extruded from the World (the physical universe). The Exiles were allowed to return – save for a few chief actors in the rebellion of whom at the time of the L.R. only Galadriel remained.fn119 But Eärendil, being in part descended from Men, was not allowed to set foot on Earth again, and became a Star shining with the light of the Silmaril, which contained the last remnant of the unsullied light of Paradise, given by the Two Trees before their defilement and slaying by Morgoth. These legends are deliberately touched on in Vol. I as being the chief ones in the background of The L.R., dealing with the relations of Elves and Men and Valar (the angelic Guardians) and
therefore the chief backward links if (as I then hoped) the Silmarillion was published.

  I relate these things because I hope they may interest you, and at the same time reveal how closely linked is linguistic invention and legendary growth and construction. And also possibly convince you that looking around for more or less similar words or names is not in fact very useful even as a source of sounds, and not at all as an explanation of inner meanings and significances. The borrowing, when it occurs (not often) is simply of sounds that are then integrated in a new construction; and only in one case Eärendil will reference to its source cast any light on the legends or their ‘meaning’ – and even in this case the light is little. The use of éarendel in A-S Christian symbolism as the herald of the rise of the true Sun in Christ is completely alien to my use. The Fall of Man is in the past and off stage; the Redemption of Man in the far future. We are in a time when the One God, Eru, is known to exist by the wise, but is not approachable save by or through the Valar, though He is still remembered in (unspoken) prayer by those of Númenórean descent.

  [The text ends with a brief discussion of Númenórean religion.]

  298 To William Luther White

  [This letter was printed, apparently without permission, with Tolkien’s address and private telephone number at the head of it, in White’s book The Image of Man in C. S. Lewis (1969).]