PART THREE. THE ETYMOLOGIES. THE ETYMOLOGIES. The mode of my father's linguistic construction, which as is well known was carried on throughout his life and in very close relation to the evolution of the narratives, shows the same unceasing movement as do they: a quality fundamental to the art, in which (as I believe) finality and a system fixed at every point was not its underlying aim. But while his 'language' and his 'literature' were so closely interwoven, to trace the history of the literary process through many texts (even though the trail might be greatly obscured) is of its nature enormously much easier than to trace the astounding complexity of the phonological and grammatical evolution of the Elvish languages. Those languages were conceived, of course, from the very beginning in a deeply 'historical' way: they were embodied in a history, the history of the Elves who spoke them, in which was to be found, as it evolved, a rich terrain for linguistic separation and interaction: 'a language requires a suitable habitation, and a history in which it can develop' (Letters no. 294, p. 375). Every element in the languages, every element in every word, is in principle historically 'explicable' - as are the elements in languages that are not 'invented' - and the successive phases of their intricate evolution were the delight of their creator. 'Invention' was thus altogether distinct from 'artificiality'. In his essay 'A Secret Vice' (The Monsters and the Critics and Other Essays, 1983, p. 198) my father wrote of his liking for Esperanto, a liking which, he said, arose 'not least because it is the creation ultimately of one man, not a philologist, and is therefore something like a "human language bereft of the inconveniences due to too many successive cooks" - which is as good a description of the ideal artificial language (in a particular sense) as I can give.' The Elvish languages are, in this sense, very inconvenient indeed, and they image the activities of countless cooks (unconscious, of course, of what they were doing to the ingredients they had come by): in other words, they image language not as 'pure structure', without 'before' and 'after', but as growth, in time. On the other hand, the linguistic histories were nonetheless 'images', invented by an inventor, who was free to change those histories as he was free to change the story of the world in which they took place; and he did so abundantly. The difficulties inherent in the study of the history of any language or group of languages are here therefore compounded: for this history is not a datum of historical fact to be uncovered, but an unstable, shifting view of what the history was. Moreover, the alterations in the history were not confined to features of 'interior' linguistic development: the 'exterior' conception of the languages and their relations underwent change, even profound change; and it is not to be thought that the representation of the languages in letters, in tengwar, should be exempt. It must be added that my father's characteristic method of work - elaborate beginnings collapsing into scrawls; manuscripts overlaid with layer upon layer of emendation - here find their most extreme expression; and also that the philological papers were left in the greatest disorder. Without external dating, the only way to determine sequence (apart from the very general and uncertain guide of changing hand- writing) is in the internal evidence of the changing philology itself; and that, of its nature, does not offer the sort of clues that lead through the maze of the literary texts. The clues it does offer are very much more elusive. It is also unfortunately true that hasty handwriting and ill- formed letters are here far more destructive; and a great deal of my father's late philological writing is, I think, strictly unusable. It will be seen then that the philological component in the evolution of Middle-earth can scarcely be analysed, and most certainly cannot be presented, as can the literary texts. In any case, my father was perhaps more interested in the processes of change than he was in displaying the structure and use of the languages at any given time - though this is no doubt due to some extent to his so often starting again at the beginning with the primordial sounds of the Quendian languages, embarking on a grand design that could not be sustained (it seems indeed that the very attempt to write a definitive account produced immediate dissatisfaction and the desire for new constructions: so the most beautiful manuscripts were soon treated with disdain). The most surprising thing, perhaps, is that he was so little concerned to make comprehensive vocabularies of the Elvish tongues. He never made again anything like the little packed 'dictionary' of the original Gnomish language on which I drew in the appendices to The Book of Lost Tales. It may be that such an undertaking was always postponed to the day, which would never come, when a sufficient finality had been achieved; in the meantime, it was not for him a prime necessity. He did not, after all, 'invent' new words and names arbitrarily: in principle, he devised them from within the historical structure, proceeding from the 'bases' or primitive stems, adding suffix or prefix or forming compounds, deciding (or, as he would have said, 'finding out') when the word came into the language, following through the regular changes of form that it would thus have undergone, and observing the possibilities of formal or semantic influence from other words in the course of its history. Such a word would then exist for him, and he would know it. As the whole system evolved and expanded, the possibilities for word and name became greater and greater. The nearest he ever came to a sustained account of Elvish vocabulary is not in the form of nor intended to serve as a dictionary in the ordinary sense, but is an etymological dictionary of word-relationships: an alphabetically-arranged list of primary stems, or 'bases', with their derivatives (thus following directly in form from the original 'Qenya Lexicon' which I have described in I. 246). It is this work that is given here. My father wrote a good deal on the theory of sundokarme or 'base- structure' (see suo and KAR in the Etymologies), but like everything else it was frequently elaborated and altered, and I do not attempt its presen- tation here. My object in giving the Etymologies * in this book is rather as an indication of the development, and mode of development, of the vocabularies of the Elvish languages at this period than as a first step in the elucidation of the linguistic history; and also because they form an instructive companion to the narrative works of this time. It is a remarkable document, which must be reckoned among the most difficult of all the papers containing unique material which my father left. The inherent difficulties of the text are increased by the very bad condition of the manuscript, which for much of its length is battered, torn, crumpled at the edges, and discoloured (so that much that was very lightly pencilled is now barely visible and extremely hard to decipher). In some sections the maze of forms and cancellations is so dense, and for the most part made so quickly, that one cannot be sure what my father's final intention was: in these parts he was working out potential con- nections and derivations on the spot, by no means setting down already determined histories. There were many routes by which a name might have evolved, and the whole etymological system was like a kaleidoscope, for a decision in one place was likely to set up disturbing ripples in etymological relations among quite distinct groups of words. Moreover, complexity was (as it were) built in, for the very nature of the 'bases' set words on phonetic collision courses from their origin. The work varies a great deal, however, between its sections (which are the groups of base-stems beginning with the same initial letter). The worst parts, both in their physical condition and in the disorganisation of their content, are the central letters of the alphabet, beginning with E. As the text proceeds the amount of subsequent alteration and addition, and resultant confusion, diminishes, and when P and R are reached the etymologies, though rough and hasty, are more orderly. With these groups my father began to use smaller sheets of paper which are much better preserved, and from S to the end the material does not present serious difficulty; while the concluding section (W) is written out very legibly in ink (in this book the last section is Y, but that is not so in the original: see p. 346). These relatively clear and orderly entries are found also in the A-stems, while the B-stems are distinct from all the rest in that they were written out as a very finished and indeed beautiful manuscript. The entries under D are in two forms: very rough material that was partly overwritten more legibly in ink, and then a second, much clearer and more ordered version on the smaller sheets. (* On a covering page to the manuscript is written Etymologies, and also Beleriandic and Noldorin names and words: Etymologies.) I have not been able to reach any certain interpretation of all this, or find an explanation that satisfies all the conditions in detail. On the whole I am inclined to think that the simplest is most likely to be right in essentials. I have little doubt that the dictionary was composed pro- gressively, through the letters of the alphabet in succession; and it may be that the very making of such a dictionary led to greater certainty in the whole etymological system, and greater clarity and assurance in its setting-out, as the work proceeded - but this also led to much change in the earlier parts. Having reached the end of the alphabet, my father then turned back to the beginning, with the intention of putting into better order the sections which had been first made and had suffered the most alteration; but this impulse petered out after the entries under D. If this were so, the original A and B entries were subsequently destroyed or lost; whereas in the case of D both survive (and it is noticeable that the second version of the D-entries differs from the former chiefly in arrangement, rather than in further etymological development). Turning now to the question of date, I give some characteristic examples of the evidence on which I think firm conclusions can be based. The original entry ELED gave the meaning of the stem as 'depart', with a derivative Elda 'departed'. Since this was the interpretation of Eldar in the Lhammas $2 and in QS $23 as those works were originally written, and first appears in them, the original entries under E clearly belong to that time. This interpretation was replaced in both the Lhammas and QS by carefully made emendations changing the meaning to 'Star-folk', and introducing the term Avari, with the meaning 'Departing'. Now the meaning 'Star-folk' appears in a second entry ELED replacing the first (and to all appearance made not long after); while the stem AB, ABAR bore, as first written, the meaning 'depart', and the derivative Avari was defined as 'Elves who left Middle-earth'. Thus the original A-entries and some at least of the alterations under E belong to the phase of the earliest alterations to QS. In QS the meaning of Avari was then changed to 'the Unwilling' (see p. 219), and at the same time the root-meaning of AB, ABAR in the Etymologies was changed to 'refuse, deny' and the interpretation of Avari to 'Elves who never left Middle-earth or began the march.' This change can be dated from the note of 20 November 1937 (given on p. zoo) in which my father said that Avari was to replace Lembi as the name of the Elves who remained in the East, while Lembi were to be 'Ilkorin Teleri', i.e. the Eldar who remained in Beleriand (see QS $$29-30 and p. 219). These changes were incorporated in the typescript of QS, which seems to have been in being by the beginning of February 1938 (p. 200). (The additional entry LEB, LEM shows this development, since Lembi is there translated as 'Elves remaining behind = Telerin Ilkorins'.) In the note dated 3 February 1938 (p. 200) my father said that while Tintalle 'Kindler' could stand as a name of Varda, Tinwerontar 'Queen of Stars' must be changed to Elentari, because 'tinwe' in Qenya only = spark (tinta- to kindle).' In the entry TIN the names Tinwetar and Tinwerontar of Varda were struck from the original material, and in the margin was written: Tintanie, Tintalle Kindler = Varda; Q tinta- to kindle, make to spark'. Original T-entries can therefore be dated before February 1938. Under the stem MEN appears the form harmen 'south', which was not subsequently changed, and again under the (additional) entry KHAR, but in this case the base-stem was afterwards changed to KHYAR and harmen to hyarmen. The insertion of y in this word was one of the alterations required in the note of 20 November 1937. Putting these and a number of other similar evidences together, it seems to me clear that despite their very various appearance the Etymologies were not spread over a long period, but were contemporary with QS; and that some of the additions and corrections can be securely dated to the end of 1937 and the beginning of 1938, the time of the abandonment of QS and the beginning of The Lord of the Rings. How much longer my father kept the work in being with further additions and improvements is another question, but here also I think that an answer can be given sufficient for the purpose. This lies in the observations that there are relatively few names that belong specifically to The Lord of the Rings; that all of them are quite clearly additions to existing entries or introduce additional base-stems; that almost all were put in very hastily, mere memoranda, and not really accommodated to or explained in relation to the base-stems; and that the great majority come from the earlier part of The Lord of the Rings - before the breaking of the fellow- ship. Thus we find, for example, Baranduin (BARAN); the imperative daro! 'stop!' (DAR; this was the sentry's command to the Company of the Ring on the borders of Lothlorien); Hollin added under EREK; the scrawled addition of a base ETER with the imperative edro! 'open!' (the word shouted by Gandalf before the doors of Moria); Celebrimbor (KWAR); Caradras (RAS; replacing in the original draft of the chapter The Ring Goes South the name Taragaer, itself found in the Etymologies under the added base TARAK); Celebrant (RAT); Imladris (RIS). The words caras (KAR) and naith (SNAS), both of them additions, probably argue the existence of Caras Galadon and the Naith of Lothlorien, and the added rhandir 'pilgrim' under RAN, taken with the added mith 'grey' under MITH, shows Mithrandir. Clear cases of names from later in The Lord of the Rings do occur (so Palantir under PAL and TIR, Dolbaran under BARAN), but they are very few. I conclude therefore that while my father did for two or three years more make rather desultory entries in the Etymologies as new names emerged in The lard of the Rings, he gave up even this as the new work proceeded; and that the Etymologies as given here illustrate the develop- ment of the Quenya and Noldorin (later > Sindarin) lexicons at the decisive period reached in this book, and provide in fact a remarkable point of vantage. The Etymologies, then, reflect the linguistic situation in Beleriand envisaged in the lhammas (see especially the third version, Lam- masethen, p. 194), with Noldorin fully preserved as the language of the Exiles, though profoundly changed from its Valinorian form and having complex interrelations in respect of names with 'Beleriandic' (Ilkorin), especially the speech of Doriath. Afterwards my father developed the conception of a kind of amalgamation between Noldorin and the in- digenous speech of Beleriand, though ultimately there emerged the situation described in The Silmarillion (p. 129): the Noldor abandoned their own tongue and adopted that of the Elves of Beleriand (Sindarin). So far-reaching was this reformation that the pre-existent linguistic structures themselves were moved into new historical relations and given new names; but there is no need here to enter that rather baffling territory. The presentation of such a text as this can obviously not be exact: in the most chaotic parts a degree of personal interpretation of what was meant is altogether inevitable. There is in any case a great deal of inconsistency in detail between the different parts of the manuscript- for example, in the use of marks expressing length of vowel, which vary unceasingly between acute accent, macron (long mark), and circumflex. I have only 'standardized' the entries to a very limited extent, and only in so far as I have felt confident that I ran little risk of confusing the original intention. In particular, I have done nothing to bring divergent forms, as between one part of the Etymologies and another, into accord, seeing that the evolution of 'bases' and derivative words is an essential part of the history; and indeed in the most complex parts of the manuscript (initial letters E, G, K) I have attempted to distinguish the different 'layers' of accretion and alteration, though elsewhere I have been very selective in pointing out additions to the original list. I have 'standardized' the entries to the extent of giving the 'bases' always in capitals, and of using the acute accent to signify long vowels in all 'recorded' forms (as opposed to 'hypothetical' antecedent forms), with the circumflex for long vowels in stressed final syllables in Exilic Noldorin and Ilkorin, as is largely done in the original. I use y for j of the original throughout (e.g. KUY, DYEL for KUJ, DJEL), since this is less misleading and was my father's own practice elsewhere (found in fact here and there in the Etymologies); the stems with initial J, becoming Y, are moved forward from their original place before K to the end of the list. I print the back nasal (as in English king) with a Spanish tilde (n), again following my father's frequent practice, though in the Etymologies he used special forms of the letter n. His grammatical abbreviations are retained, as follows: adj. adjective. g.sg. genitive singular. adv. adverb. inf. infinitive. cpd. compound. intr. intransitive. f. feminine. m. masculine. pa.t. past tense. q.v. quod vide, 'which see'. pl. plural. sg. singular. p.p. past participle. tr. transitive. prep. preposition. The sign + means 'poetic or archaic'. The abbreviations used for the different languages are as follows (there is no explanatory list of them accompanying the manuscript): Dan. Danian. Dor. Doriathrin. Eld. Eldarin. EN. Exilic Noldorin (also referred to as 'Exilic', but most often simply as N) Ilk. Ilkorin. L. Lindarin. N. Noldorin. ON. Old Noldorin (i.e. the Korolambe' or Kornoldorin, see the Lhammas $5). Oss. Ossiriandeb (the name in the Lhammas, where however the form Ossiriandic is also found). PQ. Primitive Quendian. T. Telerin. An asterisk prefixed to a form means that it is 'hypothetical', deduced to have existed from later, recorded forms. My own contributions are always enclosed within square brackets. A question mark standing within such brackets indicates doubt as to the correctness of my reading, but in other cases is original. Where I have found words totally illegible or can do no better than a guess (a very small proportion of the whole, in fact) I have usually omitted them silently, and so also with scattered jottings where no meaning is attached to forms, or where no clear connections are given. I have kept my own notes to a minimum, and in particular have very largely eschewed the temptation to discuss the etymologies in relation to earlier and later Elvish forms published elsewhere. On the other hand, while my father inserted many internal references to other stems, I have substantially increased the number (those due to me being enclosed within square brackets), since it is often difficult to find an element when it had been greatly changed from its ultimate 'base'. The Index to the book is further designed to assist in the tracing of name-elements that appear in the Etymologies. A. AB-, ABAR- refuse, deny, * abaro refuser, one who does not go forth: Q Avar (or Avaro), pl. Avari = Elves who never left Middle-earth or began the march; N Afor, pl. Efuir, Efyr (ON aboro). Cf. AWA. [This entry as first written gave the root-meaning as 'go away, depart', translated * abaro as 'departer, one who goes forth', and defined Avri as 'Elves who left Middle-earth' (see p. 344). An additional entry seems to allow for both developments from the root-meaning: 'AB- retreat, move back, refuse'.] AD- entrance, gate, * adno: Q ando gate; N annon, pl. ennyn great gate, Q andon (pl. andondi). AIWE-. (small) bird. Q aiwe, N aew. Cf. Aiwenor 'Birdland' = lower air. [For Awenor(e) see the Ambarkanta and diagrams, IV. 236 etc.] AK-. narrow, confined. * akra: Q arka narrow; N agr, agor. Cf. N Ag- lond, Aglon defile, pass between high walls, also as proper name; cf. lond, lonn path [LOD]. Q aksa narrow path, ravine. AKLA-R-. See KAL. Q alka ray of light; alkar or alkare radiance, brilliance; alkarinqa radiant, glorious. N aglar glory, aglareb glorious. ALAK-. rushing. * alako rush, rushing flight, wild wind: N alag rushing, impetuous; alagos storm of wind. Cf. Anc-alagon dragon- name [NAK]. Related to LAK(2). (* alk-wa swan: Q alqa; T alpa; ON alpha; N alf; Ilk. alch; Dan. ealc. Cf. Alqalonde Swan-road or Swan-haven, city of the Teleri [LOD].) ALAM-. elm-tree. Q alalme, also lalme; N lalf (lelf) or lalven, pl. lelvin; Ilk. lalm, pl. lelmin; Dan. alm. The stem is perhaps LALAM, q.v., but some hold it related to ALA since the elm was held blessed and beloved by the Eldar. [The end of this entry, from 'but some hold it', was an addition. Probably at the same time a stem AL- was added, with derivatives alma 'good fortune', alya 'rich', etc.; but this entry was struck out. The same derivative words are found under GALA.] ALAT- large, great in size. Q alta (...J alat- as in Alataire = Belegoer [AY]. AM(1)- mother. Q anil or amme mother-, Ilk. aman, pl. emuin. (N uses a different word, naneth, hypocoristic [pet-name form] nana [NAN]). AM(2)- up: usually in form amba-. Q prefix am- up; amba adv. up(wards); amban upward slope, hill-side; ambapenda, ampenda uphill (adj.); see PEN. N am up; am-bend, amben uphill; amon hill, pl. emuin, emyn; am-run uprising, sunrise, Orient = Q ambaron (g.sg. ambaronen) or Ambarone. ANA(1)- Cf. NA(1). to, towards. * anta- to present, give: Q anta- give; anna gift; ante (f.), anto (m.) giver. Cf. Yav-anna [YAB]; Aryante [AR(1)]. N anno to give; ant gift. [Added:] Q anta face. ANA(2)- Cf. NA(2). be, exist. [Added:] anwa real, actual, true. ANAD-, ANDA-. * anda long: Q anda; N and, ann. Cf. names Andram long-wall [RAMBA], Andfang, Anfang Longbeard, one of the tribes of Dwarves (pl. Enfeng) [SPANAG]. ANAK-. Cf. NAK bite. Q anca jaw; N anc; cf. Ancalagon [ALAK]. ANAR-. sun; derivative of NAR(1).* anar-: Q Anar sun; EN Anor. ANGA-. iron. Q anga; N ang. Q angaina of iron; N angren, pl. engrin. ANGWA- or ANGU-. snake. Q ango, pl. angwi; N am- in amlug dragon: see LOK. AP-. apsa cooked food, meat. N aes; Ilk. ass. AR(1)-. day. * ari: Q are, pl. ari; N ar- only in names of week-days, as Arvanwe [see LEP]. Cf. name Aryante Day-bringer [ANA'], N Eriant. Q arin morning, arinya morning, early; arie daytime; ara dawn; Arien the Sun-maiden. N aur day, morning; arad daytime, a day (= Q arya twelve hours, day). AR(2)-. Q ara outside, beside; also prefix ar- as in Arvalin (= outside Valinor). In Q this is purely local in sense. So also in Ilkorin, cf. Argad place 'outside the fence', or Argador (in Falathrin dialect Ariad, Ariador) lands outside Doriath (in Ilkorin Eglador), especially applied to West Beleriand, where there was a considerable dwelling of Dark-elves. In N ar- developed a privative sense (as English without), probably by blending with * al, which is only preserved in Alchoron = Q Ilkorin [LA]. Thus arnediad without reckoning, = numberless [NOT]. In this sense Q uses ava-, as avanote (see AWA). Hence Q ar and. AS-AT-. Q asto dust; N ast. ATA-. father. PQ * atu, * atar: Q atar, pl. atari; hypocoristic atto. N adar, pl. edeir, eder; ada. Cf. Iluv-atar. Ilk. adar, pl. edrin; adda. AT(AT)- again, back. Q ata again, ata-, at- back, again, re-; N ad. Cf. TAT, ATTA = two; Q atta two, N tad. N prefix ath- on both sides, across, is probably related; athrad ford, crossing (see RAT). Ilk. adu, ado double; cf. Adurant, a river in Ossiriand which for a distance has divided streams. [Ilk. adu, ado 'double' and the explanation of Adurant was an addition; this shows the conception of the island of Tol-galen (see the commentary on QS $ 114). Other additions made at different times to this entry were Q atwa double, and N eden new, begun again.] AWA- away, forth; out. Q ava outside; Avakuma [KUM] Exterior Void beyond the World; au-, ava- privative prefixes = N ar (see AR(2)), as avanote without number, numberless [NOT]. [Added:] Avalona, cf. lona [LONO]. AY- * ai-lin- pool, lake: Q ailin (g.sg. ailinen); N oel, pl. oelin; cf. Oelinuial Pools of Twilight [LIN(1); YU, KAL]. AYAR-, AIR- sea, only used of the inner seas of Middle-earth. Q ear (earen) and aire (airen); N oear, oer. Cf. Earrame, a Q name = Wings of the Sea, name of Tuor's ship. Belegoer 'great sea', name of Western Ocean between Beleriand and Valinor, Q Alataire (see ALAT). AYAK- sharp, pointed. Q aika sharp, aikale a peak; N oeg sharp, pointed, piercing, oegas (= Q aikasse) mountain peak. Cf. N Oeges engrin Peaks of Iron, oeglir range of mountain peaks. ?Related is Q aiqa steep, cf. Ilk. taig deep (blended with tara, see TA). AYAN- See YAN. * ayan- holy: Q Ainu, f. Aini, holy one, angelic spirit; aina holy; Ainulindale Music of the Ainur, Song of Creation, B. [On the distinctive manuscript of the B-entries see p. 343. The following entries were added in pencil: BAD, BARAN, BARAT, BARATH, BEN, and at the same time certain changes were made to existing entries. In this section I give the original entries as they were written, and note the alterations.] BAD- * bad- judge. Cf. MBAD. Not in Q. N baud (bad-) judgement; badhor, hadron judge. [Pencilled addition.] BAL- * bala: Q Vala Power, God (pl. Valar or Vali = PQ * bal-i formed direct from stem, cf. Valinor); there is no special f. form, where necessary the cpd.Valatari 'Vala-queen' is used, f. of Valatar (g.sg. Valataren) 'Vala-king', applied only to the nine chief Valar: Manwe, Ulmo, Aule, Mandos, Lorien, Tulkas, Osse, Orome, and Melko. The Valatari were Varda, Yavanna, Nienna, Vana, Vaire, Este, Nessa, Uinen. T Bala. ON Bala, and Balano m., Balane f.; EN Balan m. and f., pl. Belein, Belen. In Ilk. torin 'kings' was used, or the cpd. Balthor, Balthorin (* bal'tar-). Q valya having (divine) authority or power; valaina of or belong- ing to the Valar, divine; valasse divinity. Q Valinor, for * balf-ndore, reformed after the simplex nore 'land', also in form Valinore, land of the Gods in the West; ON Balandor (* bala-ndore), EN Balannor. Cf. also ON Balthil one of the names of the White Tree of Valinor, usually named in Q Silpion; EN Belthil, but this was usually applied to the image of the divine tree made in Gondolin, the tree itself being called Galathilion. Related is probably the name Balar of the large island at Sirion's mouth, where the Ilkorins long dwelt who refused to go West with Ulmo; from this is named Beleriand which they colonized from the island in the dark ages. Balar is probably from * balare, and so called because here Osse visited the waiting Teleri. [The explanation of Balar, Beleriand given here is not necessarily at variance with the story told in QS $35 that the Isle of Balar was 'the eastern horn of the Lonely Isle, that broke asunder and remained behind, when Ulmo removed that land again into the West'; but it can scarcely be brought into accord with the story (QS $36) that 'the Teleri dwelt long by the shores of the western sea, awaiting Ulmo's return', and that Osse! nstructed the waiting Teleri 'sitting upon a rock nigh to the margin of the sea.' Moreover, the 'colonization' of Beleriand from Balar seems to take no account of Thingol, and those of his people 'that went not because they tarried searching for Thingol in the woods'. 'and these multiplied and were yet scattered far and wide between Eredlindon and the sea' (Lhammas $6). More must be meant than simply that Elves from Balar removed to the mainland, for this 'colonization' from Balar is here made the very basis of the name Beleriand.] BAN- * bana: Q Vana name of the Vala, wife of Orome, and sister of Varda and Yavanna; ON and T Bana; in ON also called Bana-wende, whence EN Banwend, Banwen (see WEN). * banya: Q vanya beauti- ful; EN bein. Cf. Q vanima fair; Vanimo, pl. Vanimor 'the beautiful', children of the Valar; Uvanimo monster (creature of Melko); EN uan (* ubano) monster; uanui monstrous, hideous. BAR- Original significance probably 'raise'; cf. BARAD, MBAR. Hence uplift, save, rescue(?). * barna: Q varna safe, protected, secure; [struck out: varne protection;] varnasse security. * barya-: Q varya- to protect; EN berio to protect. [The removal of varne 'protection' was due to the emergence of BARAN 'brown' with the derivative Q varne 'brown'.] BARAD- [Added: is blended with BARATH, g-V.] * barada lofty, sublime: [added: ON barada, EN baradh, steep;] Q Varda, chief of the Valatari, spouse of Manwe; T Barada [> Baradis]. [Struck out: ON Bradil, EN Bredil (* b'radil-).] * b'randa lofty, noble, fine: T branda; ON branda, EN brand, brann (whence brannon lord, brennil lady); cf. name Brandir (brand-dir: see DER). BARAN- Q varne (varni-) brown, swart, dark brown. ON barane, EN baran. Cf. river name Baranduin, Branduin. Dolbaran. [Pen- cilled addition. On Dolbaran (probably a further addition) see p. 345.] BARAS- Stem only found in Noldorin: * barasa hot, burning: ON barasa, baraha; EN bara fiery, also eager; frequent in masculine names as Baragund, Barahir [KHER], etc. * b'ras-se heat: ON brasse, white heat, EN brass: whence brassen white-hot. BARAT- N barad tower, fortress. [Pencilled addition.] BARATH- Probably related to BAR and BARAD. * Barathi spouse of Manwe, Queen of Stars: ON Barathi(l); EN Berethil and El-bereth. Q Varda, T Baradis show influence of barada lofty. [Pencilled addition. The application of the name Elbereth to Varda seems to have arisen in the hymn of the Elves to the Goddess in the original second chapter (Three is Company) of The Lord of the Rings, written early in 1938 (where in rough workings for the song the name appears as Elberil). Concomitant with this the Ilkorin names Elbereth (of dif- ferent meaning) and Elboron were removed from the original entries BER and BOR. These were the names of Dior's sons in AB r and 2 (annal 206/306), replaced in AB 2 (note 42) by Eldun and Elrun (which were added also to Q $14); Elrun appears in the Etymologies in an addition to stem RO.] BAT- tread. * bata: ON bata beaten track, pathway; EN bad. * batta- (with medial consonant lengthened in frequentative formation): ON battho- trample, EN batho. ON tre-batie traverse, EN trevedi (pa.t. trevant) [see TER]. Cf. Q vanta- to walk, vanta a walk. BEL- strong. Cf. BAL(?). Stem not found in Q. T belle (physical) strength, belda strong. Ilk. bel (* bele) strength, Beleg the Strong, name of Ilkorin bowman of Doriath. * belek, * beleka, ON beleka mighty, huge, great; EN beleg great (n.b. this word is distinct in form from though related to Ilk. name Beleg); cf. EN Beleg-ol [GAWA] = Q Aule; Belegoer Great Sea [Av], name of sea between Middle-earth and the West; Belegost Great City [os], name of one of the chief places of the Dwarves. T belka 'excessive' is possibly from ON; ON belda strong, belle strength (EN belt strong in body, bellas bodily strength) are possibly from T. Cf. name Belthronding of Beleg's yew- bow: see STAR, DING. BEN- corner (from inside), angle. N bennas angle [NAS]. [Pencilled addition.] BER- valiant. * berya-: Q verya- to dare; verya bold; verie boldness. ON berina bold, brave; bertha- to be bold; EN beren bold, bertho dare; cf. proper name Beren. Ilk. ber valiant man, warrior (*bero); bereth valor; [struck out: cf. Ilk, name El-bereth.] Danian beorn man; this is probably blended with * besno: see BES. [On the removal of El- bereth see BARATH.] BEREK- * bereka: Q verka wild; EN bregol violent, sudden, cf. proper name Bregolas fierceness; breged violence, suddenness; breitho (*b'rekta-) break out suddenly. Cf. Dagor Vregedur [UR] Battle of Sudden Fire (EN bregedur wild-fire). [See MEREK.] BERETH- T bredele beech-tree; Ilk. breth (*b'retha) beech-mast, but the beech was called galbreth [CALAD] in Falasse, and neldor in Doriath (see NEL). The beech-tree was probably originally called *pheren, Q feren or ferne (pl. ferni), ON pheren; but in EN fer pl. ferin was usually replaced by the Ilk. breth mast, whence EN brethil beech-tree; cf. Brethiliand, -ian 'Forest of Brethil' [see PHER). BES- wed. *besno husband: Q verno; ON benno, EN benn man, replacing in ordinary use the old word dir (see DER); hervenn, herven husband (see KHER). Ilk. benn husband; Danian beorn man, blended with *ber(n)o: see BER. *besse wife: Q vesse; ON besse, EN bess woman, replacing old words di, dis (see NI(1), NDIS); herves wife (see KHER). In the f. the shift of sense in ON was assisted probably by blending with *dess young woman, ON dissa. *besu dual, husband and wife, married pair: Q veru. Cf. Q Arveruen third day (of the Valinorian week of 5 days) dedicated to Aule and Yavanna [LEP]. *besta: Q vesta matrimony; vesta- to wed; vestale wedding. BEW- follow, serve. *beuro follower, vassal: ON biuro, bioro, EN bior, beor; cf. proper name Beor. *beuya- follow, serve: ON buiobe to serve, follow, EN buio serve, hold allegiance to. T bum vassal, bua- serve. [On the name Beor see the commentary on QS $ 128.] BIRIT- Stem only found in Ilkorin. *b'ritte: Ilk. brith broken stones, gravel. Cf. river name Brithon (whence is named Brithombar) 'pebbly'. Late Exilic brith gravel is from Ilkorin. BOR- endure. Q voro ever, continually; prefix vor, voro- as in voro- gandele 'harping on one tune', continual repetition; vorima continual, repeated. "boron-: ON boron (pl. boroni) steadfast, trusty man, faithful vassal; EN bor and pl. byr for older berein, beren; Ilk. boron, pl. burnin. Cf. N names given to the 'Faithful Men': Bor, Borthandos, Borlas, Boromir. Borthandos = Borth (see below) [but this element is not further mentioned] + handos (see KHAN). Borlas = Bor + glass joy (see GALAS). Boromir is an old N name of ancient origin also borne by Gnomes: ON Boronmiro, Boromiro: see MIR. [Struck out: Cf. also Ilk. boron in Dor. name El-boron.' On the removal of El-boron see BARATH.] BORON- extension of the above (originally a verbal form of the stem seen in *boron- above). Q voronwa enduring, long-lasting; eoronwie endurance, lasting quality; cf. name Voronwe = ON Bronwega, EN Bronwe [WEG]. ON bronie last, endure, survive; EN bronio endure, brono last, survive; bronadui enduring, lasting. *b'rona: ON bruna that has long endured, old (only used of things, and implies that they are old, but not changed or worn out); EN brun old, that has long endured, or been established, or in use. Brodda -. Name of a man in Hithlum. He was not one of the Elf-friend races, and his name is therefore probably not EN or Ilkorin. BUD- jut out. Cf. MBUD. D. [A very rough pencilled list was for most of its length overwritten in ink, and nearly all these entries appear in a second, pencilled list, the differences between the two being largely a matter of arrangement; see p. 343.] DAB- give way, make room, permit, allow. Q lav- yield, allow, grant. N daf permission. DAL- flat (variant or alteration of LAD). Q lara 'flat' may derive from *dala or *lada. EN dalw flat; dalath flat surface, plane, plain [see TIR], ON dalma (probably = dal + ma hand) palm of hand; EN dalf. Ilk. dol flat, lowlying vale. DAN- Element found in names of the Green-elves, who called them- selves Danas (Q Nanar, N Danath), Cf. Dan, Denethor and other names. See NDAN? DAR- stay, wait, stop, remain. N deri, imperative daro! stop, halt; dartha wait, stay, last, endure. DARAK- *