—Treachery of Men, corrupted by Melko, at that battle;
—But the people of Úrin (Húrin) are faithful, and do not survive it;
—The leader of the Gnomes is isolated and slain (in D only);
—Turgon and his host cut their way out, and go to Gondolin;
—Melko is wrathful because he cannot discover where Turgon has gone;
—The Fëanorians come late to the battle (in D only);
—A great cairn is piled (in D only).
These are essential features of the story that were to survive. But the unlikenesses are many and great. Most striking of all is that the entire later history of the long years of the Siege of Angband, ending with the Battle of Sudden Flame (Dagor Bragollach), of the passage of Men over the Mountains into Beleriand and their taking service with the Noldorin Kings, had yet to emerge; indeed these outlines give the effect of only a brief time elapsing between the coming of the Noldoli from Kôr and their great defeat. This effect may be to some extent the result of the compressed nature of these outlines, and indeed the reference in the last of them, D, to the practice of many arts by the Noldoli (p. 240) somewhat counteracts the impression—in any case, Turgon, born when the Gnomes were in Hisilómë or (according to D) when they were encamped by Sirion, is full grown at the Battle of Unnumbered Tears.12 Even so, the picture in The Silmarillion of a period of centuries elapsing while Morgoth was straitly confined in Angband and ‘behind the guard of their armies in the north the Noldor built their dwellings and their towers’ is emphatically not present. In later ‘phases’ of the history my father steadily expanded the period between the rising of the Sun and Moon and the Battle of Unnumbered Tears. It is essential, also, to the old conception that Melko’s victory was so complete and overwhelming: vast numbers of the Noldoli became his thralls, and wherever they went lived in the slavery of his spell; in Gondolin alone were they free—so in the old tale of The Fall of Gondolin it is said that the people of Gondolin ‘were that kin of the Noldoli who alone escaped Melko’s power, when at the Battle of Unnumbered Tears he slew and enslaved their folk and wove spells about them and caused them to dwell in the Hells of Iron, faring thence at his will and bidding only’. Moreover Gondolin was not founded until after the Battle of Unnumbered Tears.13
Of Fëanor’s death in the early conception we can discern little; but at least it is clear that it bore no relation to the story of his death in The Silmarillion (p. 107). In these early outlines the Noldoli, leaving Hisilómë, had their first affray with the Orcs in the foothills of the Iron Mountains or in the pass of the Bitter Hills, and these heights pretty clearly correspond to the later Mountains of Shadow, Ered Wethrin (see p. 158, 238); but in The Silmarillion (p. 106) the first encounter of the Noldor with the Orcs was in Mithrim.
The meeting of Gnomes and Ilkorins survived in the meeting of the new-come Noldor with the Grey-elves of Mithrim (ibid. p. 108); but the Noldor heard rather of the power of King Thingol of Doriath than of the Battle of Palisor.
Whereas in these outlines Maidros son of Fëanor led an attack on Angband which was repulsed with slaughter and his own capture, in The Silmarillion it was Fingolfin who appeared before Angband, and being met with silence prudently withdrew to Mithrim (p. 109). Maidros (Maedhros) had been already taken at a meeting with an embassage of Morgoth’s that was supposed to be a parley, and he heard the sound of Fingolfin’s trumpets from his place of torment on Thangorodrim—where Morgoth set him until, as he said, the Noldor forsook their war and departed. Of the divided hosts of the Noldor there is of course no trace in the old story; and the rescue of Maedhros by Fingon, who cut off his hand in order to save him, does not appear in any form: rather is he set free by Melko, though maimed, and without explanation given. But it is very characteristic that the maiming of Maidros—an important ‘moment’ in the legends—should never itself be lost, though it came to be given a wholly different setting and agency.
The Oath of the Sons of Fëanor was here sworn after the coming of the Gnomes from Valinor, and after the death of their father; and in the later outline D they then left the host of (Finwë) Nólemë, Lord of the Noldoli, and returned to Dor Lómin (Hisilómë). In this and in other features that appear only in D the story is moved nearer to its later form. In the return to Dor Lómin is the germ of the departure of the Fëanorians from Mithrim to the eastern parts of Beleriand (The Silmarillion p. 112); in the Feast of Reunion that of Mereth Aderthad, the Feast of Reuniting, held by Fingolfin for the Elves of Beleriand (ibid. p. 113), though the participants are necessarily greatly different; in the latecoming of the Fëanorians to the stricken field of Unnumbered Tears that of the delayed arrival of the host of Maedhros (ibid. p. 190–2); in the cutting-off and death of (Finwë) Nólemë in the battle that of the slaying of Fingon (ibid. p. 193—when Finwë came to be Fëanor’s father, and thus stepped into the place of Bruithwir, killed by Melko in Valinor, his position as leader of the hosts in the Battle of Unnumbered Tears was taken by Fingon); and in the great cairn called the Hill of Death, raised by the Sons of Fëanor, that of the Haudh-en-Ndengin or Hill of Slain, piled by Orcs in Anfauglith (ibid. p. 197). Whether the embassy to Túvo, Tinwelint, and Ermon (which in D becomes the sending of messengers) remotely anticipates the Union of Maedhros (ibid. p. 188–9) is not clear, though Tinwelint’s refusal to join forces with Nólemë survived in Thingol’s rejection of Maedhros’ approaches (p. 189). I cannot certainly explain Tinwelint’s words ‘Go not into the hills’, but I suspect that ‘the hills’ are the Mountains of Iron (in The Hiding of Valinor, p. 209, called ‘the Hills of Iron’) above Angband, and that he warned against an attack on Melko; in the old Tale of Turambar Tinwelint said: ‘Of the wisdom of my heart and the fate of the Valar did I not go with my folk to the Battle of Unnumbered Tears.’
Other elements in the story of the battle that survived—the steadfastness of the folk of Úrin (Húrin), the escape of Turgon—already existed at this time in a tale that had been written (that of Túrin).
The geographical indications are slight, and there is no map of the Great Lands for the earliest period of the legends; in any case these questions are best left until the tales that take place in those lands. The Vale (or Valley) of the Fountains, afterwards the Valley (or Vale) of Weeping Waters, is in D explicitly equated with Gorfalong, which in the earlier outlines is given as Gorfalon, and seems to be distinct; but in any case neither these, nor ‘the Tumbled Lands’, can be brought into relation with any places or names in the later geography—unless (especially since in D Turgon is said to have fled ‘south down Sirion’) it may be supposed that something like the later picture of the Pass of Sirion was already in being, and that the Vale of the Fountains, or of Weeping Waters, was a name for it.
NOTES
1 Above Turuhalmë are written Duruchalm (struck out) and Halmadhurwion.
2 This paragraph is marked with queries.
3 The word may be read equally well as ‘dim’ or ‘dun’.
4 The original reading here was: ‘and few of his folk went with him, and this Tû forbade to his folk, fearing the wrath of Ilúvatar and Manwë yet did’ (sc. curiosity overcome Nuin, etc.).
5 Earlier in the Tales, ‘the Lost Elves’ are those who were lost from the great journey and wandered in Hisilómë (see p. 118).
6 In the tale the ‘fairies’ of Tû’s dominion (i.e. the Dark Elves) are given the name Hisildi, the twilight people; in outlines A and B, in addition to Hisildi, other names are given: Humarni, Kaliondi, Lómëarni.
7 Cf. also Sador’s words to Túrin in his boyhood (Unfinished Tales p. 61): ‘A darkness lies behind us, and out of it few tales have come. The fathers of our fathers may have had things to tell, but they did not tell them. Even their names are forgotten. The Mountains stand between us and the life that they came from, flying from no man now knows what.’
8 Cf. The Silmarillion p. 104: ‘It is told that ere long they met Dark Elves in many places, and were befriended by them; and Men became the c
ompanions and disciples in their childhood of these ancient folk, wanderers of the Elven-race who never set out upon the paths to Valinor, and knew of the Valar only as a rumour and a distant name.’
9 Above Ermon is written, to all appearance, the Old English word Æsc (‘ash’). It seems conceivable that this is an anglicizing of Old Norse Askr (‘ash’), in the northern mythology the name of the first man, who with the first woman (Embla) were made by the Gods out of two trees that they found on the seashore (Völuspá strophe 17; Snorra Edda, Gylfaginning §8).
10 The text has here the bracketed word ‘(Gongs)’. This might be thought to be a name for the Kaukareldar or ‘false-fairies’, but in the Gnomish word-list Gong is defined as ‘one of a tribe of the Orcs, a goblin’.
11 The cutting out of Nólemë’s heart by the Orcs, and its recapture by Turgon his son, is referred to in an isolated early note, which says also that Turgon encased it in gold; and the emblem of the King’s Folk in Gondolin, the Scarlet Heart, is mentioned in the tale of The Fall of Gondolin.
12 Cf. p. 167: ‘Turondo son of Nólemë was not yet upon the Earth.’ Turgon was the Gnomish name of Turondo (p. 115). In the later story Turgon was a leader of the Noldor from Valinor.
13 After the story was changed, and the founding of Gondolin was placed far earlier, the concluding part of The Silmarillion was never brought into harmony; and this was a main source of difficulty in the preparation of the published work.
APPENDIX
NAMES IN THE LOST TALES-PART I
There exist two small books, contemporary with the Lost Tales, which contain the first ‘lexicons’ of the Elvish languages; and both of them are very difficult documents.
One is concerned with the language called, in the book, Qenya, and I shall refer to this book as ‘QL’ (Qenya Lexicon). A good proportion of the entries in the first half of the alphabet were made at one time, when the work was first begun; these were very carefully written, though the pencil is now faint. Among these original entries is this group:
Lemin ‘five’
Lempe ‘ten’
Leminkainen ‘23’
The choice of ‘23’ suggests that this was my father’s age at the time, and that the book was begun therefore in 1915. This is supported by some of the statements made in the first layer of entries about certain figures of the mythology, statements that are at odds with everything that is said elsewhere, and which give glimpses of a stage even earlier than the Lost Tales.
The book naturally continued in use, and many entries (virtually all of those in the second part of the alphabet) are later than this first layer, though nothing more definite can be said than that all entries belong to the period of (or not long preceding) the Lost Tales.
The words in QL are arranged according to ‘roots’, and a note at the beginning states:
Roots are in capitals, and are not words in use at all, but serve as an elucidation of the words grouped together and a connection between them.
There is a good deal of uncertainty, expressed by queries, in the formulation of the roots, and in the ascription of words to one root or another, as my father moved among different etymological ideas; and in some cases it seems clear that the word was ‘there’, so to speak, but its etymology remained to be certainly defined, and not vice versa. The roots themselves are often difficult to represent, since certain consonants carry diacritic marks that are not defined. The notes on names that follow inevitably give a slightly more positive impression than does the book itself.
The other book is a dictionary of the Gnomish language, Goldogrin, and I shall refer to this as ‘GL’ (Goldogrin, or Gnomish, Lexicon). This is not arranged historically, by roots (though occasionally roots are given), but rather, in plan at least, as a conventional dictionary; and it contains a remarkable number of words. The book is entitled i·Lam na·Ngoldathon (i.e. ‘the tongue of the Gnomes’): Goldogrin, with a date: 1917. Written beneath the title is Eriol Sarothron (i.e. ‘Eriol the Voyager’), who else is called Angol but in his own folk Ottor Wfre (see p. 23).*
The great difficulty in this case is the intensity with which my father used this diminutive book, emending, rejecting, adding, in layer upon layer, so that in places it has become very hard to interpret. Moreover later changes to the forms in one entry were not necessarily made in related entries; thus the stages of a rapidly expanding linguistic conception are very confused in their representation. These little books were working materials, by no means the setting-out of finished ideas (it is indeed quite clear that GL in particular closely accompanied the actual composition of the Tales). Further, the languages changed even while the first ‘layer’ was being entered in GL; for example, the word mô ‘sheep’ was changed later to moth, but later in the dictionary uimoth ‘sheep of the waves’ was the form first written.
It is immediately obvious that an already extremely sophisticated and phonetically intricate historical structure lies behind the languages at this stage; but it seems that (unhappily and frustratingly) very little indeed in the way of phonological or grammatical description now survives from those days. I have found nothing, for instance, that sets out even in the sketchiest way the phonological relations between the two languages. Some early phonological description does exist for Qenya, but this became through later alterations and substitutions such a baffling muddle (while the material is in any case intrinsically extremely complex) that I have been unable to make use of it.
To attempt to use later materials for the elucidation of the linguistic ideas of the earliest period would in this book be quite impractical. But the perusal of these two vocabularies shows in the clearest possible way how deeply involved were the developments in the mythology and in the languages, and it would be seriously misleading to publish the Lost Tales without some attempt to show the etymological connections of the names that appear in them. I give therefore as much information, derived from these books, as is possible, but without any speculation beyond them. It is evident, for instance, that a prime element in the etymological constructions was slight variation in ancient ‘roots’ (caused especially by differences in the formation of consonants) that in the course of ages yielded very complex semantic situations; or again, that an old vocalic ‘ablaut’ (variation, in length or quality, of vowels in series) was present; but I have thought it best merely to try to present the content of the dictionaries as clearly as I can.
It is noteworthy that my father introduced a kind of ‘historical punning’ here and there: so for instance the root SAHA ‘be hot’ yields (beside saiwa ‘hot’ or sára ‘fiery’) Sahóra ‘the South’, and from NENE ‘flow’ come nen ‘river’, nénu ‘yellow water-lily’, and nénuvar ‘pool of lilies’—cf. nenuphar ‘water-lily’, modern French nénufar. There are also several resemblances to early English that are obviously not fortuitous, as hôr ‘old’, HERE ‘rule’, rûm ‘secret (whisper)’.
It will be seen that a great many elements in the later languages, Quenya and Sindarin, as they are known from the published works, go back to the beginning; the languages, like the legends, were a continuous evolution, expansion, and refinement. But the historical status and relationship of the two languages as they were conceived at this time was radically changed later on: see p. 51.
The arrangement of the material has proved difficult, and indeed without a better understanding of relationships and their shifting formulations could scarcely be made satisfactory. The system I have adopted is to give etymologically-connected groups of words, in both Qenya and Gnomish, under an important name that contains one of them; to this entry other occurrences of a word in the group are referred (e.g. glor- in Glorvent, Bráglorin is referred to the entry Laurelin, where the etymological associations of Qenya laurë ‘gold’ are given).* Every name in the Lost Tales of this volume is given—that is, if any contemporary etymological information is to be found concerning it: any name not found in the following list is either quite opaque to me, or at least cannot be identified with any certainty.
Rejected names are also included, on the same basis, but are given under the names that replaced them (e.g. Dor Uswen under Dor Faidwen).
The list of secondary names of the Valar which is written out on blank facing pages in the tale of The Coming of the Valar (see p. 93) is referred to as ‘the Valar name-list’. The sign < is used only where it is used in the Gnomish dictionary, as alfa < alchwa, meaning that the one was historically derived from the other: it is not used in this Appendix to refer to alterations made by my father in the dictionaries themselves.
Ainur Among the original entries in QL are ainu ‘a pagan god’ and aini ‘a pagan goddess’, together with áye ‘hail!’ and Ainatar ‘Ilúvatar, God’. (Of course no one within the context of the mythology can call the Ainur ‘pagan’.) GL has Ain: ‘also with distinctive masc. and fem. forms Ainos and Ainil, a God, i.e. one of the Great Valar’.
Alalminórë See Aldaron, Valinor. In QL Alalminórë is glossed ‘Land of Elms, one of the provinces of Inwinórë in which is situated Kortirion (Warwickshire)’ i.e. Alalminórë="Warwickshire" (see p. 25). Gnornish words are laln or larm, also lalmir, ‘elm’.
Aldaron In QL is a root ALA ‘spread’, with derivatives alda ‘tree’, aldëa ‘tree-shadowed’, aldëon ‘avenue of trees’, and alalmë ‘elm’ (see Alalminórë). In GL this name of Oromë appears as Aldor and Ormaldor (Oromë is Orma in Gnomish); ald ‘wood (material)’, later altered to âl.
Alqaluntë QL alqa ‘swan’ GL alcwi, with the corresponding word in Qenya here given as alqë, alcwi changed later to alfa < alchwa.
QL luntë ‘ship’ from root LUTU, with other derivatives lúto ‘flood’ and verb lutta-, lutu- ‘flow, float’ (cf. Ilsaluntë). GL has correspondingly lunta ‘ship’, lud- ‘flow, stream, float’.