J. R. R. Tolkien - Author of the Century

CHAPTER V THE SILMARILLION:

THE WORK OF HIS HEART Lost lore and lays

The publication and success of The Lord of the Rings in 1954-5 left Tolkien in much the same position as the publication and success of The Hobbit in 1937. The publishers wanted a sequel, and this time they were seconded, as Stanley Unwin’s son and successor Rayner confirms in a 1995 memoir, by increasingly large numbers of devoted readers. But Tolkien had no sequel ready to hand, or even in mind. What he had was what would be called nowadays (it is a word he would have hated) a ‘prequel’: the ‘Silmarillion’, existing as many manuscripts in many forms. He was never able to prepare this material for publication in a way which completely satisfied him, though he continued working on it for almost twenty years until he died; all the ‘Silmarillions’ now in existence have been published posthumously. Nevertheless it was the work of his heart, which occupied him for far longer than The Hobbit or The Lord of the Rings. The better-known works are in a way only offshoots, side-branches, of the immense chronicle/mythology/legendarium which is the ‘Silmarillion’, and which we have first in the form in which it was published as a connected narrative in 1977 (which I distinguish as The Silmarillion), and then in many of the twelve volumes of The History of Middle-earth published between 1983 and 1996, all thirteen works (as also the volume of Unfinished Tales of Númenor and Middle-earth, from 1980) edited by Tolkien’s son and literary executor Christopher.

Tolkien was working on something which might be seen as the seed of a section of The Silmarillion at least as early as 1913, when he began to write ‘The Story of Kullervo’, a ‘prose-and-verse romance’ never yet published which resembles in outline the story of Túrin, eventually chapter 21 of the 1977 Silmarillion. In late 1916, by now on convalescent leave from the trench fever contracted on the Somme, he was writing a much more extended and continuous account of elvish story, completed (or at least relinquished) by 1920, and published in 1983-4 as the two-volume Book of Lost Tales. During his years at Leeds University (1920-25) he began to versify two main sections of this material, the tales of Túrin and Beren, eventually published as The Lays of Beleriand in 1985. In 1926, when he sent one of these poems to his old teacher R.W. Reynolds, Tolkien also wrote a brief outline or ‘Sketch of the Mythology’ to act as background for Reynolds, which appeared as ‘The Earliest Silmarillion’ in The Shaping of Middle-earth (1986), though as with so much of what he wrote the published version takes in heavy rewriting up to 1930. Between 1930 and 1937, when The Hobbit came out, the ‘Sketch’ was rewritten in expanded form as the ‘Quenta’ or ‘Quenta Nol-dorinwa’, and then rewritten again as the ‘Quenta Silmarillion’ (the first published in The Shaping of Middle-earth, above, the second in The Lost Road, in 1987). It was this latter work, in its original form ‘a beautiful and elegant manuscript’, which was sent to Stanley Unwin in 1937 along with the poem ‘The Gest of Beren and Lúthien’ (the longest of The Lays of Beleriand), as a possible successor to The Hobbit. It met a confused reception from the publisher’s reader, who seems to have seen only the poem and the section of the prose ‘Quenta Silmarillion’ added to explain it, and was gently rejected by Stanley Unwin as ‘a mine to be explored in writing further books like The Hobbit rather than a book in itself (see The Lays of Beleriand, pp. 364-7). Tolkien turned away to write The Lord of the Rings; but once this was over, first after the completion of The Lord of the Rings in 1951, and then after its publication in 1955, he turned back to rewrite his material once more, this time as ‘The Later Quenta Silmarillion’, published in two ‘phases’ in volumes X and XI of The History of Middle-earth, Morgoth’s Ring (1993) and The War of the Jewels (1994). Even this account very much understates the complexity of The Silmarillion’s development (there is a much more extended account by Charles Noad, see ‘List of References’ below), for the versions listed above were often written and rewritten, in some cases becoming ‘a chaotic palimpsest, with layer upon layer of correction’; while Tolkien also produced several sets of ‘Annals’ covering the same material, some of them written in Old English: ‘The Annals of Valinor’, ‘The Annals of Beleriand’, ‘The Grey Annals’, ‘The Annals of Aman’, all to be found in volumes IV-V and X-XI of The History of Middle-earth.

Generalizing successfully about this mass of heterogeneous material, a ‘fixed tradition’, as Christopher Tolkien notes, but never a ‘fixed text’, may seem to be impossible: but some shafts, one hopes of light, may be driven into it. One by this time predictable point is that Tolkien derived some part of his invention, and a vital part, from an entirely novel solution to an old mythological problem. There was no doubt that a belief in ‘elves’ (Old Norse álfar, Old English ylfe) was widespread in Germanic antiquity: but the words used about them seemed curiously contradictory. The Icelander Snorri Sturluson, whose prose account of Norse mythology remains our only half-coherent account, was aware of both ‘Light-elves’ (ljosálfar) and ‘Dark-elves’ (dökkálfar), but he also recognized ‘Swart-elves’ (svartálfar), though the place they lived, ‘Swart-elf-home’ (Svartálfaheim) was also the home of the dwarves. Meanwhile Old English uses words like ‘Wood-elf’ (wuduælf) and ‘Water-elf (wæterælf). How are all these fragments to be fitted together? Are ‘Swart-elves’ the same as ‘Dark-elves’, and both perhaps the same as dwarves? The OED seems to accept this solution, cross-referring ‘dwarf’ and ‘elf’ rather vaguely to each other, but it is a feeble notion. Early accounts distinguish the two species from each other perfectly clearly, the dwarves being associated with mining, smithcraft, and a world underground, the elves with beauty, allure, dancing, and the woodland. Tolkien’s great predecessor, Jacob Grimm, also pondered the problem, but in his Deutsche Mythologie (translated into English in 1884 as Teutonic Mythology) could come to no conclusion, ending another rather vague discussion with the once-more feeble remark that maybe ‘Dark-elves’ were sort of in between ‘Light-elves’ and ‘Swart-elves’, ‘not so much downright black, as dim, dingy’. It would be surprising if Tolkien had not read this passage in youth, and been annoyed by it.

At the heart of his account of the elves is a quite different distinction. The elves are not separated by colour (black, white, and ‘dingy’), but by history. The ‘Light-elves’ are those who have ‘seen the Light’, the Light of the Two Trees which preceded the Sun and Moon, in Aman, or Valinor, the Undying Land in the West; the ‘Dark-elves’ are those who refused the journey and remained in Middle-earth, to which many of the Light-elves however returned, as exiles or as outcasts. The Dark-elves who remained in the woods of Beleriand are also, of course, naturally described as Wood-elves. And as for the connection with dwarves, the two species in Tolkien are quite distinct and never mingle, but they do in some cases associate. Elves may live underground and be given admiring dwarvish names, like Finrod ‘Felagund’ (< Dwarvish felak-gundu, ‘cave-hewer’). It would only be natural, as time went by and memory became blurred, for men to be unsure whether such a character was once elf or dwarf, or what was the difference. A main aim in Tolkien’s creations was always to ‘save the evidence’, to rescue his ancient sources from hasty modern accusations of vagueness or folly. Saving the evidence, moreover, generated story, in this case the complex story of the elves’ wanderings, separations, and returns, summed up as well as anywhere in chapter 8 of The Hobbit:

For most of [the Wood-elves]…were descended from the ancient tribes that never went to Faerie in the West. There the Light-elves and the Deep-elves and the Sea-elves went and lived for ages, and grew fairer and wiser and more learned, and invented their magic and their cunning craft in the making of beautiful and marvellous things, before some came back into the Wide World. In the Wide World the Wood-elves lingered in the twilight of our Sun and Moon, but loved best the stars; and they wandered in the great forests that grew tall in lands that are now lost.

Two less easily graspable points about The Silmarillion tradition concern language, and nationality. Tolkien said, in many ways, as forcefully as he could, and perhaps with a certain defensiveness (for writing fairy-stories was certainly seen by some in authority as a distraction from his proper job of being a language professor), that all his work was ‘fundamentally linguistic in inspiration’ (his emphasis). The ‘authorities of the university’ might well consider his fiction a hobby, more or less pardonable, but to him it was not a hobby ‘in the sense of something quite different from one’s work, taken up as a relief-outlet’. Instead, ‘The invention of languages is the foundation. The ‘stories’ were made rather to provide a world for the languages than the reverse’ (Letters, p. 219). Accordingly, one might well say that the major root of The Silmarillion and all that followed from it was the invention of the elvish languages, Quenya (‘Elf-latin’, the language of the Light-elves), and Sindarin, the language of Beleriand, of the Wood-elves. One would get even closer by saying that the real root was the relationship between them, with all the changes of sound and semantics which created two mutually-incomprehensible languages from one original root, and the whole history of separation and different experience which those changes implied. (The best discussions of these are to be found on Carl Hostetter’s web-site in the ‘List of References’ below). Such developments were Tolkien’s major professional field, like those which generated (for instance) Gothic, Norse and English from one original root, perhaps preserved in the early runic monuments of Scandinavia. He himself suggested that the Quenya/Sindarin relationship was more like that between Latin and Welsh, though there is probably no one alive with the knowledge to appreciate it. Still, recondite though Tolkien’s linguistic interests were, he could claim, and several times did claim in the lectures of his later years, that he had made his point by demonstration if not by argument: rooting story in language had worked, even for those who did not care, or did not know they cared about language.

Tolkien’s views on nationality may be even more idiosyncratic, though they are straightforward and logical. His family name was German-derived, as he knew perfectly well, being a re-spelling of the nickname tollkühn, ‘foolhardy’. He saw himself, however, as being ‘far more of a Suffield (a family deriving from Evesham in Worcestershire)’, and being like his family ‘intensely English (not British)’. The trouble with this, as no one was in a better position to appreciate, was that native English tradition, following the Norman Conquest and the take-over by French and Latin learning, had in England been very largely, if not quite completely, suppressed. Jacob and Wilhelm Grimm, searching for relics of their country’s past in the children’s tales of the nineteenth century, had come up with quite a respectable haul, as had their followers recording Gaelic, or Irish, or Welsh. But the obverse of the domination of English as a language within the British Isles and elsewhere was that it had become international, multicultural, the language of the educated with no time for fooleries and fairy-tales. As a result, in spite of the very early start made on literacy by the Christianized English, native tradition petered out. The Welsh continued to tell tales of King Arthur, but there are (almost) no native stories of Hengest and Horsa; nineteenth-century English fairy-tale collections are among the weakest in Europe.

One thing Tolkien accordingly set himself to do – it will be remembered that his first extensive composition was called The Book of Lost Tales – was to reverse this decline, and restore to England something like the body of lost legend which it must once have had. His project is discussed by Carl Hostetter and Arden Smith in the volume of Centenary Conference Proceedings, but one may say briefly that this is why Tolkien spent much effort in writing ‘The Annals of Beleriand’ and some of ‘The Annals of Valinor’ in Old English: to provide a chain of communication between the imagined far past and the first beginnings of English history. It was, on a larger scale, the same sort of activity as that mentioned on p. 26 above, writing an Old English riddle to act as a reconstructed ancestor for modern nursery-rhyme. In early versions of the ‘Silmarillion’, furthermore, the stories are passed on by an early Englishman, or ‘Anglo-Saxon’, who has been stranded among the elves and learned their history directly from them. He, Eriol, or Ælfwine, is accordingly a witness to ‘the true tradition of the fairies’ (The Book of Lost Tales II, p. 290), not the ‘garbled things’ told by other nations. Tolkien’s painstaking attempts to develop this idea have been discussed most recently by Verlyn Flieger, in her article ‘The Footsteps of Ælfwine’ in the recent collection Tolkien’s ‘Legendarium’. However, Tolkien also did rather more than flirt with the idea – though in the end he found it untenable – that Elvenhome had survived as England, with England as formerly Tol Eressëa, the Lonely Isle, Warwick as the elvish city Kortirion, and the Staffordshire village of Great Haywood, where he spent some of his convalescence, as Tavrobel, where Eriol learned the ‘lost tales’ of elvish myth (The Book of Lost Tales I, pp. 24-5). The theory could not work. For one thing, as Tolkien knew perfectly well, the English were themselves immigrants, who had come into Britain (not at that time Eng-land, the land of the English, in any sense) some fifteen hundred years ago; and though like the hobbits in the Shire they ‘fell in love with their new land’ and indeed forgot that they had ever had another one, it was impossible for a real historian to imagine a continuous tradition, in the same place, lasting from before the Romans and the ancestors of the Welsh through to the arrival of those whom Tolkien reckoned as his ancestors. Nevertheless, Tolkien would have liked to create a ‘mythology’ for his own people, to anchor it in the counties of the West Midlands, and simultaneously to preserve in it what scraps remained of the myths and legends there must once have been.

One of the clearest signs of Tolkien’s overall intention is his use of the word ‘lays’, as in The Lays of Beleriand. ‘Lay’ is now an unfamiliar term with no precisely accepted meaning, just an old word, it seems, for ‘poem’. Tolkien, however, did not think of it like that. What he did mean by it can be seen in another famous work from a century before his time, Lord Macaulay’s The Lays of Ancient Rome. Many people have come across at least one of the poems from this set, the famous ‘Horatius’, which tells the story of ‘Horatius at the Bridge’, but few readers nowadays realize what Macaulay intended to do in it. Macaulay’s ‘Preface’ makes it quite clear. Before his time (1842), one could say that works like Livy’s Roman Histories had been perfectly familiar, indeed set reading for centuries of schoolboys. They had been accepted, though, just as histories, and even if one had one’s doubts about the stories they contained, there seemed no prospect of correcting them or getting behind them to whatever earlier sources Livy must have used (all of them long vanished from the world). With the coming of the ‘higher criticism’ in Germany, however, methods were developed (largely subjective, but sometimes linguistic) for disentangling earlier from later strata of story, and genuine old tradition from contemporary faking. It became widely believed that behind the extensive epics of Homer, and Virgil, and the Histories of Livy, and Beowulf, and even the accounts of the Old Testament, there must have been early preliterate traditions which were used by the later writers – traditions probably expressed in short poems composed at or near the time of the events they commemorated. Germans called these almost entirely hypothetical poems Lieder, while English-speaking authors divided between calling them ‘ballads’ and calling them ‘lays’: the OED defines ‘lay’, in this technical sense, as:

the appropriate term for a popular historical ballad such as those on which the Homeric poems are supposed by some to have been founded. Some writers have misapplied it to long poems of epic character such as the Nibelungenlied or Beowulf.

It is clear that the OED editor here is unconvinced by the whole theory, with his ‘misapplied’ and ‘supposed by some’, but Lord Macaulay at least believed it. In the’ Preface’ to his Lays of Ancient Rome he put forward the argument that Rome, like England, had had a native tradition of balladic verse; but, like England, had been intellectually colonized by a culture felt by the educated classes to be superior (Greek culture for Rome, French culture for England); and had accordingly suppressed or abandoned its deepest roots. In England and Scotland (Macaulay said) the position had been partly rescued at almost the last possible moment by the activity of antiquarians like Thomas Percy and Sir Walter Scott, but Rome had not been so lucky. Still, something like the ballads of the Anglo-Scottish border must have existed in ancient Rome; that ‘lost ballad-poetry of Rome’ must have been transformed by the likes of Virgil and Livy into epic and history; and ‘To reverse that process, to transform some portions of early Roman history back into the poetry of which they were made, is the object of this work.’ So Macaulay wrote, not only ‘Horatius’ but three other ballads, ‘The Battle of the Lake Regillus’, ‘Virginia’, and ‘The Prophecy of Capys’.

One attraction of this process, furthermore, and another one which could not have been aimed at without the work of the ‘higher critics’, was that even in these supposedly early ballads one could see some indication of date. German critics had become extraordinarily astute (usually too astute) in picking out anachronisms within the works they studied. They could tell (or thought they could tell) the difference between material which was original, which went right back to whatever historical event was being commemorated – in the case of Beowulf, say, the death of Beowulf’s uncle in battle in the early sixth century – and material being inserted maybe two hundred years later, like the many Christian references in Beowulf which could only come from a time after the English had become Christians. One very bad result of this, to which Tolkien put a firm and complete stop with his 1936 lecture, was that much of Beowulf was effectively thrown away as ‘phoney’ by over-astute dissectors. But one good result was that people learned to read histories and historical poems with a kind of double vision, to see both the event being described and the context in which it was described. Macaulay built this kind of vision into ‘Horatius’ (and pointed out that he had done so in the ‘Preface’) by including evidently nostalgic remarks about ‘the brave days of old’, which show that his feigned ‘lay’ is deliberately looking backward from some historical distance. It has two dates in it, event, and record. This is the kind of thing flattened out by the treatments of Virgil or Livy.

‘Alas for the lost lore, the annals and old poets that Virgil knew, and only used in the making of a new thing!’ wrote Tolkien in his discussion of Beowulf (Essays, pp. 27-8), and in that context what he meant above all was that one should concentrate on the ‘new thing’ (the poem that survived) and not mope after all the hypothetical ones that hadn’t. Just the same, he meant the ‘alas’ as well, felt the tragedy of the ‘lost lore and annals’, wished above all to create the sense of age, of antiquity with yet greater antiquity behind it, which was theorized by the ‘higher critics’, counterfeited by Lord Macaulay, and which Tolkien in his turn thought he could recognize in poems like Sir Gawain and the Green Knight. It was his quest for ‘this flavour, this atmosphere, this virtue that such rooted works have’ (Essays, p. 72), which led Tolkien to spend so much time and effort in creating different sets of ‘annals’, in different languages, imagined as the sources for the ‘Silmarillion’; why the 1977 Silmarillion itself is studded with references to poems on which the imagined compiler is drawing, ‘the Noldo-lantë…that Maglor made before he was lost’, the ‘Lay of Leithian’, the ‘Laer Cú Beleg, the Song of the Great Bow’, to name only three of many; and why in the end and in some cases, like the ‘Lay of Leithian’, he went on to write the ‘lay’ itself, creating his own historical tradition like Lord Macaulay.

The effort of doing all this was extremely great, and the returns perhaps for many people rather minimal, for the sense of depth and age, the ability to read a work on two chronological levels at once, are rather recondite matters. Still, the sense that there was a deep and old tradition behind it, surfacing in the poems recited by Aragorn or Bilbo or Sam Gamgee, had been a major part of the texture even of The Lord of the Rings. Perhaps that success could be repeated in the ‘Silmarillion’. In any case, one may feel, this was what Tolkien all his life most wanted to do. The first poem he ever published, in the King Edward’s School Chronicle for 1911, was an account of an inter-house rugby match, called ‘The Battle of the Eastern Field’ (the school rugby pitches being off Eastern Road). As Jessica Yates has pointed out, it is quite clearly written, if in mock-heroic spirit, in a style which closely imitates Macaulay’s Lays. A parallel mythology

Commenting briefly on a tradition as complex and developed as this one is hard to manage with perfect accuracy. However, it can be said that in broad outline Tolkien’s image of the history of the First Age remained relatively stable. It can be divided, if arbitrarily, into three main sections, indicated here according to the titles and chapter-numbers of the published 1977 Silmarillion.

The first ‘section’ consists of the ‘Ainulindalë’, the ‘Valaquenta’, and chapters 1-2 of the 1977 ‘Quenta Silmarillion’ itself. They deal with the creation of the world, the rebellion of one of the Creator’s subordinate spirits, Melkor, and the decision by some of these subordinate spirits (the Valar), including Melkor, to bind themselves within the world, Earth (though to them Earth includes Aman, the Undying Lands, as well as Middle-earth, the lands of mortality).

Chapters 3-8, and 11, deal with the appearance of the elves, the decision of the other Valar to imprison Melkor in order to protect the elves, the migration of the Light-elves from Middle-earth to Aman, and the unrest and destruction caused there by the released Melkor and an elvish faction. Here the ‘Silmarils’ appear. They are jewels made by the greatest of the elvish smiths, Fëanor, and they contain within them the light of the Two Trees of Valinor, the trees which lit the world before the rising of Sun and Moon. Once the Trees are poisoned by Melkor and his spider-ally Ungoliant, their light survives only in the Silmarils. But when Fëanor is called on to give them up, to be broken to bring the Trees back to life, he refuses – only to find that Melkor has already stolen them. Fëanor, his sons and his adherents (mostly from his own tribe of the Noldor) then decide to leave Aman to pursue Melkor and regain the Silmarils, and he and his sons swear an oath of vengeance on anyone, ‘Vala, Demon, Elf or Man as yet unborn…whoso would hold or take or keep a Silmaril from their possession’. In pursuit of this oath they commit two initial acts of violence or treachery: they steal the ships of the elves who dwell on the shores of Aman (the Teleri), killing many in the process; and then having crossed back to Middle-earth, from which they had been brought by the Valar, they burn the ships and refuse to return for those of their supporters (including Galadriel) left behind. The latter group reach Middle-earth only by marching across the ice of the North. Meanwhile the Valar, dismayed by the loss of the Trees and the defection of Fëanor, create the Sun and Moon to replace the Trees, but cut off communication between Middle-earth and Aman.

The third and longest section, effectively chapters 10 and 12-24 of the 1977 Silmarillion, deals with the wars in Middle-earth between the elves and Melkor (renamed Morgoth), and the ill-fated attempts to regain the Silmarils. Into these wars are drawn both men, who appear in chapter 12, and dwarves, while Morgoth deploys ores, balrogs, and dragons. They are marked also by internal feuding and treachery, while the two longest chapters deal with the human heroes Beren, who regains a Silmaril at the cost of his hand, and Turin the ill-fated: it is these stories which Tolkien versified in The Lays of Beleriand. The Silmaril recaptured by Beren goes from one owner to another, always bringing disaster with it. In the end Eärendil, a hero of mixed elvish/human ancestry, sails with its aid to Aman, to beg the Valar in Valinor for forgiveness and assistance to Middle-earth. This is granted; Morgoth is overthrown; and the two remaining Silmarils are recaptured, only to be lost again by the final workings-out of the oath of Fëanor and his sons. Eärendil’s Silmaril, however, shines from the prow of his ship, which has been set in the sky as a star and a sign of hope to Middle-earth.

Even from this summary one can see several things. The Silmarillion bears a kind of relationship to Christian myth. The rebellion of Melkor, and his subordinate spirits, is analogous to the Fall of Lucifer and the rebel angels. Lucifer is by tradition princeps huius mundi, ‘the prince of this world’, and Melkor calls himself, perhaps truthfully, ‘Master of the fates of Arda’. The origin of the fall is also the same in both cases, for the sin of Lucifer was (according to C.S. Lewis) the urge to put his own purposes before those of God, and that of Melkor was ‘to interweave matters of his own imagining’ with the ‘theme of Ilúvatar [the Creator]’. This ‘fall of the angels’ also leads in both mythologies to a second fall: the Fall of Man and the exile from the Garden of Eden in the Book of Genesis, the loss of elvish innocence and the emigration from Aman (which becomes an exile) in The Silmarillion. Finally Eärendil, the half-human emissary who obtains forgiveness and rescue from the Valar, bears a more distant similarity to the Incarnation of Christ, and the Christian promise of salvation.

That said, there are of course very marked differences, especially centring on the Silmarils, to which there is no close Christian parallel. These differences shed more light on the nature of The Silmarillion, for after all it would seem pointless – it might even be thought presumptuous – just to rerun the Christian myth, in which Tolkien devoutly believed, merely in a work of human imagining. There is in fact an ambiguity running through The Silmarillion, which is this. The four most powerful of the Valar are clearly the spirits of earth, water, air and fire, respectively Aulë, Ulmo, Manwë, and Melkor: Melkor is the spirit of fire. Meanwhile Feanor, the maker of the Silmarils and the elf responsible for the second fall, is actually a nickname and means again ‘Spirit of Fire’. But Fëanor is an ambiguous character, proud, selfish, vengeful – but also skilled, ambitious, demanding justice. At the heart of his fall is the refusal to give up the works of his own craft, crying out bitterly, when he is asked to surrender the Silmarils:

‘For the less even as for the greater there is some deed that he may accomplish but once only; and in that deed his heart shall rest.’

(The Silmarillion, chapter 9)

Tolkien clearly had more than a certain sympathy with this view. It was something he felt himself (only his Silmaril was The Silmarillion). In his essay ‘On Fairy-Stories’ he protested strongly, even passionately, that there was a right to create fantasy, even if, even though, fantasy could be abused, could become the making and worshipping of false gods, whether literally (like Beelzebub, the ‘Lord of the Flies’ of Golding’s fantasy), or politically, in the shape of ‘social and economic theories’ also demanding ‘human sacrifice’. But fantasy was a human desire which could not be taken away:

At the heart of many man-made stories of the elves [surely, in this case, The Silmarillion] lies, open or concealed, pure or alloyed, the desire for a living, realized sub-creative art, which (however much it may outwardly resemble it) is inwardly wholly different from the greed for self-centred power which is the mark of the mere Magician [or, one might say, the ‘Necromancer’, Sauron]. Of this desire the elves, in their better (but still perilous) part are largely made…

(Essays, p. 143)

The ‘sub-creative’ desire, then, is legitimate – Tolkien goes on to say, in a fragment of verse, it is ‘our right…That right has not decayed: / we make still by the law in which we’re made’.

But if it is legitimate for Tolkien, is it for Fëanor? And how does it relate to the urge to manufacture which creates not only the Silmarils but also the invention of weapons:

Fëanor made a secret forge, of which not even Melkor was aware; and there he tempered fell swords for himself and for his sons, and made tall helms with plumes of red.

(The Silmarillion, chapter 7)

One Old English poem, which Tolkien certainly knew (it gave him the word éored), seems to locate the Fall of Man not in Adam and Eve and the Garden of Eden, but in Cain and Abel and the invention of metallurgy: ‘A state of violence came into being for the race of men, from the moment when the earth swallowed the blood of Abel…So the inhabitants of Earth endured the clash of weapons widely through the world, inventing and tempering wounding swords’. And one might recall that Fëanor is not the only dangerous maker in Tolkien’s work. Saruman too is a forger and creator and a user of fire, whose name could indeed be translated ‘Artificer’, or even ‘Engineer’, see pp. 169-71 above. Thorin Oakenshield’s disastrous fascination with the Arkenstone parallels the disastrous quests for the Silmarils, but is also only a normal dwarvish urge raised to a higher power, the urge even Bilbo feels for a moment, ‘the love of beautiful things made by hands and by cunning and by magic…the desire of the hearts of dwarves’. In The Silmarillion too Aulë, the earth-spirit of the Valar and patron of ‘all craftsmen’, makes the dwarves against the wishes of Ilúvatar, and weeps when he is detected and must offer to destroy them. Tolkien’s work in fact gives a continuum of creative urges, from wholly evil (Melkor’s, the ‘self-centred’ urge of the ‘Magician’), to wholly legitimate (his own, the right to fantasy and to ‘sub-creation’); but they shade into each other, and it is not easy to see always why they should be distinguished. Some of the tension of The Silmarillion comes from sympathy with the sons of Fëanor and their disastrous oath, and with those who reject Aman and unchanging immortality for Middle-earth, creation, independence, and death. And, one might add, entirely seriously, for linguistic change, which happens only in Beleriand: Tolkien once applied the term felix peccatum, ‘fortunate sin’, not to the Fall of Man (which was made ‘fortunate’ by the Incarnation) but to the Tower of Babel, the presumption which by tradition created the multiplicity of human languages from a single root (see Essays, p. 194).

Tolkien indeed built the concept of the felix peccatum into his own mythology, when Ilúvatar declares that even Melkor in his sin ‘shall prove but mine instrument in the devising of things more wonderful’. And in a sense he also built in, or rather left a space for, the traditional story of the Fall of Man. There is no Garden of Eden for humans in The Silmarillion, but when humans do enter Middle-earth from the east all that is known about them to the elves who are imagined as the preservers of these traditions is that something dreadful had happened to them already, a ‘darkness’ which ‘lay upon the hearts of Men’ and which was connected with an unknown expedition of Morgoth: one could believe that Morgoth here is identical with Satan, and his expedition was to lure humanity into their ‘original sin’. The Silmarillion then does not contradict Genesis; but it does offer an alternative view of the origin of sin, in a desire not for the ‘knowledge of good and evil’, but in the desires for creation, mastery, power. ‘A passion for family history’

These desires are then worked out in the long history of elves and humans, which occupies sections two and three of The Silmarillion, as divided above. Most people have found these hard to follow – the most cutting remark made being the apocryphal, ‘a telephone directory in Elvish, yet’ (for which I know no source). There are an awful lot of names in The Silmarillion, it is true, and the genealogies trespass on the short memories of modern literates. Nevertheless, a clear structure for the whole work can be made out if one once masters the central idea of the divisions among the elvish tribes.

The most basic of these is the distinction between the Light-elves, the Calaquendi, who reached Valinor and saw the light of the Two Trees before they were poisoned, and the Dark-elves, the Moriquendi, who refused the journey. (The divisions given here are necessarily something of an approximation. For a full picture, see the information laid out in diagrammatic form in the appendices to The Silmarillion.)

This latter group is largely the same as the speakers of Sindarin, the elvish language which developed by language-change in Beleriand, but not always and not absolutely. One of the three original ambassadors to Valinor was Elwë Singollo (in Quenya, Elu Thingol in Sindarin), who returned to Middle-earth to urge his people to go to Aman, but then remained behind himself, held by his love for Melian the Maia (Maiar are spirits intermediate between the elves and the Valar, and include both Gandalf and the Balrogs). He, accordingly, though king of the Sindarin-speaking Grey-elves, Elves of the Twilight, is not to be counted as a Dark-elf; for he had seen the light once. And yet he is called a Dark-elf, on one occasion. When the Noldor arrive in Middle-earth, Elwë is naturally wary of what may turn into a dispossession, and sends a message warning the sons of Fëanor to stay within the limits he has set. He has been informed of the situation by ‘Ang-rod son of Finarfin’, who is both Feanor’s nephew (on his father’s side) and Elwe’s great-nephew (on his mother’s side). The sons of Fëanor take offence at this, and one of them, Caranthir, calls out:

‘Let not the sons of Finarfin run hither and thither with their tales to this Dark Elf in his cavesl Who made them our spokesmen to deal with him? And though they be come indeed to Beleriand, let them not so swiftly forget that their father is a lord of the Noldor, though their mother be of other kin.’

(my emphases, The Silmarillion, chapter 13)

This is an offensive speech on several levels, and the offence is compounded by containing half-truths. Elwë is not technically speaking a ‘Dark Elf’, for he has been to Valinor and seen the Light of the Trees; on the other hand he is king of the Dark-elves, and he refused to return to Valinor, so there is a basis for Caranthir’s claim of superiority. As for the sons of Finarfin, Caranthir’s sneer is based on the fact that their mother, Elwë’s niece, though a Light-elf, comes from the most junior branch of the Light-elves, the Teleri. But the sneer could easily be turned back on him, for they are also descended, on their grandmother’s side, from the Vanyar, the most senior branch (the Noldor, to which the sons of Fëanor belong, being intermediate). So one of Caranthir’s claims is false in detail but true in general, while the other is true in detail and false in general. This is, then, a subtle and a tense situation, one of many which build up the overall effect of the tragedy of the Noldor. But the subtlety and the tension depend on carrying in one’s head a string of distinctions between elvish groups, and a whole series of pedigrees and family relationships. The audiences of Icelandic sagas could do this, but readers of modern novels are not used to it, and easily miss most of what is intended.

Though Tolkien meant The Silmarillion to fill a gap in English tradition, he does indeed seem to have drawn mostly on Old Norse or Icelandic literature for its main themes. The Silmarils themselves, in my opinion, are an attempt to solve the mysterious riddle of the sampo, an undefined object often referred to in the Finnish Kalevala – Tolkien was fond of Finnish, modelled aspects of Quenya on it, and furthermore admired the Kalevala as a product of exactly the kind of literary rescue-project which he would have liked to see in England, see pp. xv, xxxiv above. However, much of The Silmarillion can be seen as a complex tragedy of mixed blood, of the kind seen in several poems of the Elder Edda. To revert to Dark-elves, the Dark Elf par excellence in The Silmarillion is Eöl, a great smith, a close associate of the dwarves (so answering one of the points bodged by Jacob Grimm, see p. 229), a relative of Elu Thingol and hostile to the incoming Noldor. He captures and marries Aredhel, lost in the woods of Beleriand, and so sets up a sequence of tragedies – all dependent on genealogy. Aredhel is first cousin both to the sons of Finarfin and to the sons of Fëanor. She is also the sister of Turgon who, distrusting the confidence of the sons of Fëanor, has withdrawn into one of the three ‘Hidden Kingdoms’, Gondolin. Why does Aredhel leave Gondolin, and so start the tragic sequence? Her interview with her brother suggests both pride and deliberate deception – one of several snowball-that-starts-an-avalanche scenes in The Silmarillion. Be that as it may, she uses her permission to visit her Telerin/Vanyarin cousins, the sons of Finarfin, to try to reach her Noldorin cousins, the sons of Fëanor: an ominous choice. In any case she fails, is captured by Eöl, and bears him a son, Maeglin. They eventually escape into the country of the sons of Feanor, who capture Eöl as he follows them. What is his relationship to them? He claims kinship, punningly and sarcastically, as Curufin expels him, ‘It is good to find a kinsman thus kindly at need’ – and this is as usual part-true, for he is a cousin by marriage. But Curufin rejects the claim and denies the connection: ‘those who steal the daughters of the Noldor…do not gain kinship with their kin’. This scene is however contrasted with the next one, in which Eöl, still pursuing his wife and son, finds his way to Gondolin and is captured by Aredhel’s brother Turgon. Turgon, by contrast with Curufin, magnanimously grants the connection, immediately greeting him with ‘Welcome, kinsman, for so I hold you’. But Eöl (who still bears the grudge of the dispossessed Dark-elves) rejects the offer, demands his wife and son, and when this is refused kills Aredhel and is himself executed.

Both Eöl and Aredhel have in different ways been studies in ambiguity, but that ambiguity now shifts to their son Maeglin. His closest relative is now his uncle Turgon; his uncle however killed his father; his father however killed his mother; and behind all the ‘howevers’ there is the question of inheritance. Should Maeglin inherit from his uncle? He is less than half Noldorin by blood. In any case, whatever his feelings about his father, he may have inherited from him the Telerin grudge, of dispossession. A solution would be for Maeglin to marry his cousin, Turgon’s daughter and only child, Idril; but first-cousin marriage is forbidden in elvish society (though not in Tolkien’s own society). When Maeglin sees himself displaced once more by the human Tuor, who marries Idril and fathers Eärendil, he turns traitor and betrays Gondolin to Morgoth. Who, then, is responsible for the Fall of Gondolin? Maeglin, for treason? Eöl, for abducting Aredhel? Aredhel, for defying her brother? The sons of Fëanor, for their pride, lack of respect for others, and bad example? The strains of mixed blood? Elvish historians, according to Tolkien, see the core of it in Maeglin’s urge towards a kind of incest, and regard this as ‘an evil fruit of the Kinslaying’, the first Noldorin assault upon the Teleri in Aman: it is a sort of sexual retaliation for ancient violence.

The whole train of story, then, is a sad and complex one, with many mixed motives and scenes of hidden tension. But in order to follow it, it is vital to remember who everyone is, who their relatives are, and what they feel about their relatives. As has been said above, readers of Norse sagas could do it, and according to Tolkien, hobbits could do it too, as one sees from remarks like Gaffer Gamgee’s ‘So Mr Frodo is [Bilbo’s] first and second cousin, once removed either way, as the saying is, if you follow me’. Few do follow him, though; and though the hobbits ‘have a passion for family history’, it is not always shared. The organization of The Silmarillion in that respect makes demands upon its readers which no other modern work has ventured, including (for all its complex structure) The Lord of the Rings. The ‘Human-stories’ of the elves

Another way of penetrating the structure of the third section of The Silmarillion is to observe that it is largely organized round the falls of three different ‘Hidden Kingdoms’: Doriath, Nargothrond, Gondolin. Each is set up by an elf-king, respectively Elu Thingol, Finrod Felagund, Turgon, the latter two motivated by lack of faith in the power of the sons of Fëanor to ward off Morgoth (Thingol made his decision before the return of the Noldor to Middle-earth). Each kingdom prospers for a while, even a long while, till each is located by and willingly or unwillingly acts as host to a mortal, a man, respectively Beren, Túrin, Tuor. These stories of human involvement with the elves were above all the works of Tolkien’s heart. Christopher Tolkien has noted his father’s statement that the tale of Tuor and the fall of Gondolin was the first of the Silmarillion complex to be written, while on sick leave from the army in late 1916 or 1917 (The Book of Lost Tales I, p. 10). It leads furthermore to the story of Eärendil, Tuor’s son, a name which we know had caught Tolkien’s attention even earlier, while he was still a student at Oxford, and which had generated what is possibly the very first work in his whole mythological cycle, the poem of’The Voyage of Earendel’, written in September 1914 (The Book of Lost Tales II, pp. 267-9). Meanwhile the story of Beren and Lúthien remained deeply personal to Tolkien till he died: he had the names ‘Beren’ and ‘Lúthien’ carved on his and his wife’s shared tombstone, a striking identification. These tales do indeed give a clue to the original motive and deepest theme of The Silmarillion, and perhaps of all Tolkien’s work.

In his essay ‘On Fairy-Stories’ (first published in 1947), Tolkien remarked that the ‘oldest and deepest desire’ satisfied by fairy-stories is to tell tales of ‘the Great Escape: the Escape from Death’. He added, with clear self-reference which must in 1947 have seemed merely jocose, ‘The human stories of the elves are doubtless full of the escape from deathlessness’. The only such stories, of course, are those written by Tolkien, and not surprisingly, they do contain both themes. Beren escapes from death – he dies, but is brought back from the dead, alone among men, by the songs of Lúthien which move even Mandos, keeper of the Halls of the Dead, to pity. Lúthien correspondingly escapes from deathlessness, for she, like Arwen in Appendix A of The Lord of the Rings, is allowed to choose death and finally accompany her husband. Eärendil and his wife Elwing in their way also escape from mortality, and reach the Undying Lands to beg for aid to Middle-earth. But again conversely, and on a much larger scale, one should note that nearly all Tolkien’s elvish characters choose death in the long term (though to them death is different from what it is for humans), simply by returning to Middle-earth. Their return does not make them immediately mortal, but it does expose them to the malice of Morgoth and the chances of Middle-earth, which are almost invariably fatal. Why do they do it? Why did Tolkien even imagine such a strange motivation?

There is no difficulty in seeing why Tolkien, from 1916 on, was preoccupied with the theme of death, and escape from it. By the end of World War I, as he said himself, his closest friends were dead. He had been an orphan since his mother died when he was twelve, and had never really known his father, who died when he was four. The theme of escape from death might then naturally seem attractive. More puzzling is the theme of the escape to death, the deep love of the elves for the mortal world, which on the one hand they regard as galadhremmin ennorath, ‘tree-tangled Middle-earth’, and which they regard on the other as a paradise, loss of which is not even fully compensated by immortality, see Haldir’s remark quoted on pp. 205-6 above. One might argue that Tolkien, elaborating his stories of a race choosing the fate of mortality, was trying to persuade himself that mortality had after all some attractions, invisible though those might be to humans who have no other choice. Against that, the whole of The Silmarillion, and especially the ‘human-stories’ embedded in it, is deeply sad, sad beyond The Lord of the Rings (though that is not as pain-free as imperceptive critics have said), certainly sad beyond anything normally tolerated in twentieth-century fiction. The question they ask insistently is ‘why? Why do death and pain and evil come? Why are they necessary?’

Tolkien’s answers to these perhaps unanswerable questions, long-evolving and never in fact completed though they are, can be seen in their most developed form in the tale of Túrin. This survives, like so much of The Silmarillion, in several major (and more minor) forms, which I would pick out as follows:

(1) the tale of ‘Turambar and the Foalóke’, in The Book of Lost Tales II (written by mid-1919)

(2) ‘The Lay of the Children of Húrin’, incomplete, written in alliterative verse in two main versions, in The Lays of Beleriand (written between 1922 and 1925)

(3) ‘Of Túrin Turambar’, chapter 21 of the 1977 Silmarillion (constructed from several sources, but perhaps predominantly work prior to 1937)

(4) the ‘Nam î Hin Húrin’, or ‘Tale of the Children of Húrin’, in Unfinished Tales (much the most expanded version as far as it goes, but fragmentary, written from 1951, see The War of the Jewels).

All four versions differ from each other, but the outline remains surprisingly stable.

Very briefly, the start of the story is the self-sacrificing stand of Túrin’s father Húrin at the Battle of Nirnaeth Arnoediad, which allows Turgon to escape to Gondolin and puts Turgon under deep obligation. Húrin is taken alive by Morgoth and allowed to see the fate that unfolds for his children. Turin’s mother Morwen sends him away for safety, and he is received by Elu Thingol in Doriath. But Túrin, angered by taunts at his mother, kills one of the king’s counsellors (Saeros), flees, and becomes an outlaw. Assisted by the marchwarden of Doriath, Beleg, who has remained his friend, he rises to prominence again, but is captured by the ores, and on being rescued by Beleg, kills him by mistake. He makes his way to Nargothrond (where Finrod is now dead), and under a false name once more becomes prominent; he persuades the elves of Nargothrond to emerge from hiding and take a more aggressive role, while the new king’s daughter, Finduilas, falls in love with him. Meanwhile Morwen, with her daughter Nienor, have eventually fled, and found refuge in Doriath, too late to catch up with Túrin. Túrin’s new aggressive policy however only betrays Nargothrond to Morgoth, and it is destroyed by Glaurung the dragon. Glaurung’s ‘binding spell’ holds Túrin immobile while Finduilas is driven away, and the dragon taunts him with abandoning his mother and sister. Túrin tries to rescue his mother, but arrives at their old home to find she has gone; and while he is doing that, Finduilas is killed by the ores. Meanwhile Morwen, now looking in her turn for her son, meets the dragon, and is lost in the confusion, while Nienor, Túrin’s sister, loses her mind, and runs naked through the forest till she collapses on the grave-mound of Finduilas. Túrin finds her there, and since neither he nor she knows who she is, marries her under the name of Niniel. In a last exploit, he wounds Glaurung mortally, but falls unconscious himself; and when Níniel comes to rescue him, Glaurung restores her memory to her, and she realizes she is pregnant by her brother. She commits suicide; Túrin kills the man who tells him what has happened, Brandir; but when it is confirmed, decides on suicide in his turn. In a last scene he asks his sword (the work of Eöl the Dark-elf) whether it will kill him, and it replies (in a scene which changes little from 1919 to 1951, and which is certainly imitated from the Finnish Kalevala):

‘Yea, I will drink thy blood gladly, that so I may forget the blood of Beleg my master, and the blood of Brandir slain unjustly. I will slay thee swiftly.’

(The Silmarillion, chapter 21)

He kills himself, and the sword breaks as he does it.

All this is seen, furthermore, by Húrin, given the gift of vision by Morgoth, and his embitterment once released plays a part in the later destructions of both Doriath and Gondolin. But what is the root of the tragedy? One answer is obviously that Túrin brings his troubles on himself: again and again he lashes out and kills the wrong person, Saeros, Beleg, Brandir, and others. Another could be that it is just terrible bad luck, if you believe that luck is ever ‘just’ luck: Morwen and Túrin criss-cross while looking for each other, and Nienor just happens to be found on the grave of Finduilas, where Túrin’s guilt and protective urges are at their strongest. Or, of course, it could all be the fault of Morgoth and his servant the dragon Glaurung, who spares Túrin at Nargothrond only for a worse fate. But all three of these could be seen as relatively comfortable explanations. As with the Eöl/Aredhel/Maeglin complex discussed above, Tolkien remained keenly interested in the hidden roots of evil or of disaster, in the way that minor outbreaks of selfishness or carelessness mean more than they seem: snowballs leading to avalanches once more.

These concepts are most developed in the latest version, the ‘Narn i Hîn Húrin’, incomplete though it is. The initial scene is found in all versions. In The Book of Lost Tales II Melko (i.e. Morgoth) curses Úrin (i.e. Húrin), putting ‘a doom of woe and a death of sorrow’ on his family, and granting him ‘a measure of vision’, so he can see what happens to them. The scene is there in ‘The Lay of the Children of Húrin’ (The Lays of Beleriand), where the phrase is ‘a doom of dread, of death and horror’. In the 1977 Silmarillion it has become ‘a doom…of darkness and sorrow’, and here Melkor/Morgoth calls himself ‘Master of the fates of Arda’ (i.e. Middle-earth). In the much-expanded ‘Nam’ version of this scene, Morgoth says further:

‘I am the Elder King: first and mightiest of all the Valar, who was before the world, and made it. The shadow of my purpose lies upon Arda, and all that is in it bends slowly and surely to my will. But upon all whom you love my thought shall weigh as a cloud of Doom, and it shall bring them down into darkness and despair.’

(Unfinished Tales, p. 67)

Húrin denies this, or some of it: ‘Before Arda you were, but others also; and you did not make it’. And even if he was the mightiest of the Valar, Húrin adds, he could not pursue even mortals ‘Beyond the Circles of the World’. Morgoth replies that there is nothing beyond the circles of the world. Húrin’s last words are ‘You lie’, and Morgoth replies ‘You shall see and you shall confess that I do not lie’. The question is, how far is Morgoth lying; and the fear is that some of what he says is true. Perhaps Morgoth really is (and Húrin makes allusive reference to whatever it was that caused the Fall of Man long ago, when Morgoth may have taken the role of Satan) the princeps huius mundi. Tolkien was after all in his own life intimately acquainted with ‘the problem of pain’, as Lewis called it. However, if the world is delivered over to a diabolic power, that power, it seems, must work through human wills, as the ‘Narn’ allusively suggests.

Some responsibility, to begin with, is laid in the ‘Narn’ on Turin’s mother Morwen. She is given very clear advice by her husband before he leaves, ‘Do not be afraid!’ and ‘Do not wait!’ She remembers this, but she ignores it, because ‘she would not yet humble her pride to be an alms-guest’, even of Thingol. She sends her son away instead, but it is a son who remembers the unfortunate words of his father’s crippled servant Sador, that the incomers who take over the country have learned from the ores to hunt their slaves with hounds. The fear that this may happen to his mother is clearly Turin’s major trauma – the image of a naked woman running. At Thingol’s court it is the again unfortunate allusion to this – ‘Do [the women of Hithlum] run like deer clad only in their hair?’ – which triggers Túrin’s first outbreak, first manslaughter, and second exile. The taunt which Glaurung levels at him (in the 1977 Silmarillion) is that he has abandoned his mother and sister, ‘Thou art arrayed as a prince, but they go in rags’, and it is his reaction to that which makes him abandon Finduilas to the fate which he fears for other women. But the fate he fears is exactly what comes about, with Morwen lost in the woods, and Nienor hunted by the ores till she runs naked, ‘as a beast that is hunted to heart-bursting’. It is pity for this, and the identification of all the abused women of his imagination in the one figure, which makes Túrin love Nienor, attempt to protect her by marrying her, and set up the fatal and final incest. All this comes from Morwen’s bad decision to separate from her son, and one of its roots is pride.

It comes also from a series of (as I have called them above) ‘unfortunate’ phrases and allusions – none of them (except perhaps Glaurung’s) intentional. But what is meant by ‘fortune’? Is it the same as ‘fate’? Túrin asks Sador ‘What is Fate?’ as a child, and gets no clear answer. But the implication of the story is that Morgoth was not lying, though he may not have been telling the whole truth, when he called himself ‘Master of the fates of Arda’. He cannot make people do wrong, for that would deny human free will. But he can put words into their mouths, and the responses to those words, in the end all Turin’s fatefully bad decisions, are then their responsibility. Characters in Icelandic sagas occasionally say, of loose or provocative speech, ‘trolls must have plucked at your tongues’, and Tolkien repeats the idea in more dignified form, with Mablung saying for instance to Saeros after his taunting, ‘some shadow of the North [i.e. of Morgoth] has reached out to touch us tonight’. Morgoth’s ‘doom’, then, works by ‘shadow’ and suggestion. But The Lord of the Rings shows how shadow, absence, can paradoxically become a presence. In the ‘Narn’ the double explanations seen in The Lord of the Rings (see pp. 145-6 above) are strongly marked and openly discussed. Sador is lamed ‘by ill-luck or the mishandling of his axe’ (which? Morgoth could have sent the ‘ill-luck’, so that Sador would be there to say his ‘unfortunate’ words); Túrin reaches Doriath ‘by fate and courage’ (but the ‘fate’ may be Morgoth’s, for it would be better if he had died young); Sador tells Turin, like Galadriel talking to Sam Gamgee, ‘a man that flies from his fear may find that he has only taken a short cut to meet it’. Turin takes the nickname Turambar, ‘master of doom’, in defiance of this whole train of thought, asserting his own free will; but the epitaph given him by Nienor is Túrin Turambar turún’ ambart-anen, ‘master of doom by doom mastered’. The whole story suggests once more a deep consideration of the nature of Macbeth, where similarly the words of the witches seem to bring about Macbeth’s fall, but could not operate without Macbeth’s responses to them.

A case could be made for seeing Tolkien’s other major ‘human story’, the tale of Beren and Lúthien, as the philosophical antithesis to Túrin. It is a story of love across the species of elf and human, rather than a tale of incest; it contains the defeat of Morgoth and the recovery of a Silmaril, not the fulfilling of his purpose; it leads on to a more sustained triumph yet, for the couple’s granddaughter is Elwing, the wife of Eärendil, who brings the Valar back to Middle-earth, while Túrin has no descendants; Lúthien masters fate and death in a way that Túrin cannot even aspire to; and the last word of the poetic version of the story sung by Aragorn in The Fellowship of the Ring (1/11) is ‘sorrowless’. However, though the story contains both the Escape from Death (for Beren, sung back from the dead by his wife), and the Escape from Deathlessness (for Lúthien, given permission to join her husband in the end and pass beyond ‘the Circles of the World’), it does not read like a ‘comedy’ in even the Dantesque sense. Work of Tolkien’s heart though it was, and existing in even more versions than Túrin – including Aragorn’s poem and an early version of that poem published separately in 1925 – the impression that it makes in the end remains one of crowding, and perhaps of derivation. It is full of motifs taken from earlier story – werewolves and vampires, the healing herb, the rope of hair let down from a window (as in ‘Rapunzel’), the wizards’ singing-contests (from the Finnish Kalevala). Its core may be Beren’s rash promise to Thingol, ‘when we meet again my hand shall hold a Silmaril’, to be proved true in the letter, if false in the spirit, when Beren shows Thingol the stump of his arm, bitten off at the wrist by Carcharoth the wolf, so that his hand with the Silmaril in it is still in the wolf’s belly. Yet Rash Promises between mortals and the inhabitants of Faerie are an old tradition, as in Sir Orfeo (where a mortal rescues his wife from the underworld through the fairy-king’s promise), and Sir Gawain and the Green Knight (where the mortal makes the promise to receive a return-blow); and the motif of the wolf-bitten wrist is one of the most familiar tales from Snorri Sturluson’s Prose Edda, where it is told of the god Tyr and the monster-wolf Fenrir.

Tolkien was happy, of course, to use old motifs and make them familiar once more; and my earlier criticism (in The Road to Middle-earth) of crowding, of the tale in the 1977 Silmarillion version feeling like a compendium, has been answered by Christopher Tolkien, who points out that that is exactly what the ‘Silmarillion’ at an earlier stage was intended to be: a compendium of legend made at the end of the Third Age of Middle-earth, abbreviating much earlier material but using some passages from it, like the quotation from the ‘Lay of Leithian’ given in The Silmarillion chapter 19 (The Book of Lost Tales II, p. 57). To return to the comments on Lord Macaulay made earlier, Tolkien’s ‘Quenta Silmarillion’ of 1937 could readily be seen as taking the role of Livy’s Histories, so to speak, while The Lays of Beleriand behind it, including ‘The Lay of Leithian’, would represent the ‘lost lore’ and the ‘old poets’ later used ‘in the making of a new thing’. Tolkien’s literary intention is then perfectly clear, and admirably consistent over a long period of time, as are his thematic concentrations on death and immortality, sorrow and consolation. One can see, though, that none of this had any connection at all with any literary mode now familiar, still less commercially viable. Stanley Unwin’s gentle hint that the ‘Silmarillion’ material should be used in the writing of new Hobbits was no doubt well meant, but one cannot imagine how it could ever have been taken. Bilbo can co-exist with Thorin Oakenshield, and Frodo with Strider, but reducing Túrin and Beren to the ‘low mimetic’ mode of the modern novel would present a seemingly unscalable challenge. Angels and the evangelium

At the end of all versions of The Silmarillion comes the tale of Eärendil, another story of escape from Middle-earth, a story which blends the escapes of human and elf, Eärendil being descended from both. In it Tolkien leaves the mode of heroic chronicle and returns to that of mythology. Like Saint Brendan, Eärendil continually explores to the west in search of the Undying Lands. In his absence the remaining sons of Feanor attack his settlement in the hope of winning back the Silmaril inherited by his wife Elwing from her grandmother Lúthien. They fail, and Elwing throws herself and the jewel into the sea. But (and here the mythological strain becomes dominant once more) the Valar turn her into a bird, in which shape, and still carrying the Silmaril, she rejoins her husband far out at sea. The Silmaril takes them through the prohibitions of the Valar, through the Shadowy Seas and past the Enchanted Isles, till Eärendil reaches Aman and walks up from the coast towards Valinor, the Guarded Land. There he is hailed by a messenger as:

‘the looked for that cometh at unawares, the longed for that cometh beyond hope!…bearer of light before the Sun and Moon…star in the darkness, jewel in the sunset, radiant in the morning!’

(The Silmarillion, chapter 24)

The nearest parallel to language of this sort, with its -eth endings and its Biblical phrases, is in the psalm-like announcement of the eagle to Gondor in The Return of the King (see p. 209 above); and like the eagle’s announcement this one carries significant ambiguity. The ambiguity had probably been present from Tolkien’s first explorations.

Tolkien had been struck by the name, or the word Earendel as early as 1914, when he encountered it in an Old English poem, now titled (rather unimaginatively) Christ I. The lines go, Eala earendel…, ‘O Earendel, brightest of angels, sent to men above Middle-earth’, but it is not clear even from context what Earendel means or even whether it is a proper name. Tolkien would however soon have realized, from obvious sources like the standard edition of the poem and Grimm’s Teutonic Mythology, that (as with ‘Light-elves’ and ‘Dark-elves’) the material for an imaginative reconstruction was ready to hand, and one which would once again ‘save the evidence’. In the first place the Old English poem (apart from the word earendel) is a translation of a known Latin antiphon, which begins O Oriens…This antiphon is however, in both Latin and Old English, one of a series representing the cries of the patriarchs and prophets still in Hell, before the coming of Christ to release them, calling out for a Saviour for those who ‘sit in darkness and the shadow of death’, or a prophet who will announce the coming of the Saviour; the O Oriens one is taken to refer to John the Baptist. But whatever earendel is, the image that goes with it is people in sorrow looking up from the darkness and hoping both for rescue and for light. Despite this strongly Christian context, though, Earendel, if it is a name, also has pagan connections. Aurvandil (the Old Norse equivalent of Old English Earendel) is present in Snorri Sturluson’s Prose Edda as a companion of the god Thor. The two went together on an expedition, but as the god waded across the freezing rivers of Élivágar he had to put his weaker companion in a basket. Aurvandil’s toe stuck out, got frostbite, and was broken off and thrown into the sky to become a star. Mythographers have been consistently puzzled by this allusion – one of them suggested that since his wife was called Gróa, ‘to grow’, Aurvandil might represent the seed-corn which in Scandinavia is sometimes sown too early and killed by the frost. However, what Tolkien might have taken from it is, first, confirmation that Earendel/Aurvandil is the name of a star, and second, that it was a sign of hope and good tidings to pagans as well as to Christians.

All this provides a suggestive background for Tolkien’s tale of Eärendil, and a justification for the Biblical language of the herald’s announcement quoted above. In Tolkien the people looking up from darkness and seeing a great light are not the patriarchs and prophets of Old Testament story, but the inhabitants of Middle-earth; and the great light they see is not Christ coming to harrow Hell and release them, but the Silmaril announcing the rescue mission of the Valar. The setting is indeed not Christian but pagan, or at least pre-Christian. However, if pagans knew of Aurvandil, and Aurvandil is the same linguistically as Earendel, and Earendel was early equated with Christ, then could pre-Christians not have had some intuition, some sense of a forerunner of their true and eventual Saviour? In Mere Christianity C.S. Lewis calls these ‘good dreams…queer stories scattered all through the heathen religions’ about a god who brings men ‘new life’. Tolkien would probably not have approved Lewis’s phrasing, nor did he mean in any way to confuse Eärendil or the Valar with Ilúvatar, the Creator. But in a clearly limited and deliberately imperfect way, Tolkien’s Silmarillion closes with an analogue of intercession, forgiveness, and salvation coming necessarily from outside a ruined Middle-earth, just as it began with analogues of the Fall of Angels and the Fall of Man.

To return to the questions implied on p. 239 above, does it not seem presumptuous to repeat Christian myth in a work of human imagining, and what is the point of doing so, with variations? Not surprisingly, the answer seems to be the same as that given already for The Lord of the Rings (see pp. 180-2 above). Tolkien wanted all his life to bring together the Christian religion in which he devoutly believed, and the relics of the pre-Christian beliefs of his ancestors embedded in the literature which he spent his professional life studying. He had no sentimental feelings about paganism or heathenism, which he hated (see the discussion of Denethor above, pp. 176-7); but he was not prepared to write off everything pre-Christian as irrelevant, unlike his countryman Alcuin (see the discussion of Frodo above, pp. 183-5). His re-interpretation of Eärendil is accordingly not presumptuous, but respectful. And as for the point of the variations on Christian myth, one might say this.

Tolkien knew that ‘angel’ meant originally angelos, ‘messenger’. But there could be several kinds of messenger. Gandalf is one – very unlike the traditional image of an angel, with his long beard and short temper, but an ‘angel’ just the same. Eärendil is a second, announcing the coming of the Valar to Morgoth and Middle-earth alike. Galadriel, in a way, could be seen as a third. Of course she is not a Maia, like Gandalf, and she also bears some share of the responsibility for the Fall of the Noldor and the exile from Middle-earth, joining in the rebellion against the Valar because of her yearning ‘to see the wide unguarded lands [of Middle-earth] and to rule there a realm at her own will’ (The Silmarillion chapter 9, but see the alternative versions discussed in Unfinished Tales). If, then, Galadriel were to be remembered at some later date as equivalent in status to Gandalf, and so an ‘angel’, she would have to be a fallen angel; and if fallen angels are the same as devils, then this seems inconceivable. Fallen angels are not however the same as devils in all opinions and all traditions. In some traditions, including early English ones, some of the angels exiled from Heaven with Satan became devils, but others, more undecided or more neutral, became elves. At Judgement Day some of these may regain forgiveness and salvation and return to their old home, as Galadriel does at the very end of The Return of the King. This still does not make Galadriel into an angel, even in the sense of a messenger, in the way that Gandalf is; but one can imagine how a human being, looking back at the events of the Third Age and the First Age ‘from some historical distance’, as suggested above, could be confused, could put together Galadriel the Noldo exiled by the Valar and Gandalf the Maia sent by the Valar (both of them allowed in the end to return) and no longer be able to see much difference.

Tolkien knew further that the Greek word for the New Testament, euangelion, contained within it the -angel- element, and meant ‘the Good Message’, neatly translated into Old English as the gód spell, the ‘good story’, the Gospel. In modern English ‘spell’ no longer means ‘story’, but ‘enchantment’, but Tolkien might have thought that example of semantic change entirely appropriate and perhaps not even accidental. Gospel means Christian message; means good story; means powerful enchantment. Angel means winged creature of Christian myth; means messenger; means elf. Earendel means John the Baptist, the forerunner of Christ the Saviour; means star; means seed – though the ‘seed that does not die and cannot be destroyed’ in the very last sentence of the 1977 Silmarillion is the seed of evil which will ‘bear dark fruit even unto the latest days’, i.e. till now, and beyond now. These complexes of meaning suggest that history, and linguistic change, keep on generating new meanings for words and demanding new versions of story, even when they are the same words and the same story. In that case The Silmarillion, centred as it is on the sins of possession and mastery and the desire to exercise skill whatever the consequences, becomes less a mythology for England and more one for its own time, for the twentieth century: a myth re-told, with proper respect for what in myth is unchanging, because myths always need retelling. Some comparisons

For all that has been said, The Silmarillion can never be anything other than hard to read. And for all Stanley Unwin’s tact, back in 1937, it is unlikely that it would ever have been published in any form, let alone so many of its forms, without the prior success of The Lord of the Rings. It has no hobbits – those essential mediating figures which provide the modern audience with a focus and a point of relationship. It scorns novelistic convention: in the ‘Narn i Hîn Húrin’ Tolkien began to develop the process of detail, of the verisimilitude created by subordinate characters and extraneous dialogue which the modern reader expects, but it did not go very far. Even towards the end of his writing career and after forty or more years of development he was clearly still unsure how to bring in some features, like the ‘Dragon-helm of Dor-lómin’ or the animated ‘Black Sword’ itself. It is not that such motifs could not be used in a modern environment – one can imagine one of Tolkien’s many emulators in the field of fantasy integrating them into a commercially successful fantasy – rather, that Tolkien continued to reach (every bit as much as James Joyce) for something beyond the conventionally publishable.

Like Joyce with Finnegans Wake, he demanded too much for most audiences. Christopher Tolkien has stated that ‘To read The Silmarillion one must place oneself imaginatively at the time of the ending of the Third Age – within Middle-earth, looking back’, (The Book of Lost Tales I, p. 4), which is certainly correct. It parallels Tolkien’s own image of Beowulf as ‘a poem from a pregnant moment of poise, looking back into the pit, by a man learned in old tales who was struggling, as it were, to get a general view of them all’ (Essays); as it does Lord Macaulay’s determined attempt to see Livy or Virgil as men looking back on the old lays or ballads of their own tradition, and seeing in those lays or ballads men who were looking still further back, to ‘the brave days of old’. If Tolkien had a literary model for The Silmarillion, furthermore, it must surely be the Prose Edda of Snorri Sturluson, a compendium of pagan mythological materials made by a man who was himself not at all a pagan, but who did not wish to see the old traditions of the poetry of his native tongue vanish for ever; a man who also continually illustrated his prose synopses with quotations from poems, too often poems which have otherwise vanished. The scholarly attraction of reading a work like that is hard to convey, for one feels at once the interest of the material that is there, regret for all the material that is not there (but which clearly could have been if only the author had thought it worth while), and the constant stimulus to the imagination of the gaps and omissions and lacunae. ‘Heard melodies are sweet’, says the poet, ‘But those unheard are sweeter’, and in cases like Snorri’s one might almost believe it to be literally true. Tolkien corroborated the thought with the remark (in a letter to Christopher in January 1945) that ‘A story must be told or there’ll be no story…yet it is the untold stories that are most moving’ (see Letters, p. 110, and also The Book of Lost Tales I, p. 3). One can agree and sympathize. Just the same, appreciation of this kind is rare, recondite, and hard to develop: an acquired taste par excellence – though one can see it as strikingly exemplified as anywhere in Christopher Tolkien’s own ‘introduction’ to another late synopsis preserving intensely moving and suggestive scraps of older tradition, The Saga of King Heidrek the Wise.

Christopher Tolkien, the editor of all the material discussed here, has also suggested that one’s model for how to respond should be Sam Gamgee’s innocent and naive response to Gimli’s song of Moria in The Fellowship of the Ring, where ‘great names out of the ancient world [Nargothrond and Gondolin] appear utterly remote’: ‘I like that!…I should like to learn it’ (The Book of Lost Tales I, p. 3). This is so. But there is an alternative response to ancient story in The Two Towers, on ‘The Stairs of Cirith Ungol’, when Sam has just given yet one more version of the tale of Beren and Lúthien, and remarked that he and Frodo still appear to be in the same tale: perhaps some hobbit-child in the future will demand the story of ‘Frodo and the Ring’. Yes, says Frodo, and he will demand ‘Samwise the stouthearted’ too: ‘I want to hear more about Sam, dad. Why didn’t they put in more of his talk, dad? It makes me laugh.’ This does not count for much as literary criticism, but it does make a point, which is that The Silmarillion, very much unlike The Lord of the Rings, stays resolutely on the level of ‘high mimesis’ or above, eschewing humour, detail, fine texture. It is able conversely to appeal to qualities virtually ruled out in even the most ambitious commercial fantasies: stoicism, nonchalance, irony, magniloquence. But here it has not been followed, and probably cannot be. The Silmarillion is most likely to be seen – paradoxically, for things were meant to be the other way round – as a further and immensely detailed ‘Appendix’ to The Lord of the Rings. The ‘lost lore that was used in the making of a new thing’ is, in this case alone, no longer lost. It tells a great deal about the ‘making’, but it also returns attention to the ‘new thing’.