The Road to Middle-Earth: How J. R. R. Tolkien Created a New Mythology

by Tom Shippey

  With rocks, and stones, and trees!

  He should have written ‘stocks’, not ‘rocks’. But he preferred the alliteration on r (and the tautology). Milton meanwhile got the phrase right in his sonnet ‘On the Late Massacre in Piedmont’:

  Avenge, O Lord, Thy slaughtered saints, whose bones

  Lie scattered on the Alpine mountains cold,

  Ev’n them who kept Thy truth so pure of old,

  When all our fathers worshipped stocks and stones …

  However in Tolkien’s view everything else in the poem would be wrong: its vengeful ferocity, its equation of God’s truth with Protestantism, most especially its contempt for ‘our fathers’ before they were converted, for the Anglo-Saxons indeed. Milton knew very little about them, and his contempt was based on ignorance. Yet poetry which uses old phrases is not always bound down to its creator’s intention. Reading that line, and adding to it his memories of Finn and Froda, of Beowulf and Hrothgar and the other pagan heroes from the darkness before the English dawn, Tolkien may have felt that Milton was more accurate than he knew. Perhaps ‘our fathers’ did worship ‘stocks and stones’. But perhaps they were not so very bad in doing so. After all if they had not Christ to worship, there were worse things, many worse things for them to reverence than ‘stocks and stones’, rocks and trees, ‘merry Middle-earth’ itself.

  * Here, as a matter of fact, Tolkien was wrong. ‘Roseberry Topping’ in North Yorkshire preserves beneath pastoral euphemism the Viking name Othinesbeorg, ‘Odin’s mountain’. But Tolkien could have replied that this name had been so sharply changed as to suggest a deliberate de-mythicizing policy in the Middle Ages, which would support his general point.

  CHAPTER 7

  VISIONS AND REVISIONS

  The Shaping of ‘The Silmarillion’

  The Hobbit and The Lord of the Rings are the works which have made Tolkien’s reputation. They were not, however, ‘the work of his heart’, as I have called it elsewhere (Author, ch. 5). This was the immense complex of stories, repeatedly told and retold in quite different forms, which I call ‘the Silmarillion’, but distinguish from The Silmarillion: which is the selection from that immense complex made by Christopher Tolkien and published in 1977, arranged as the latter explains ‘in such a way as seemed to me to produce the most coherent and internally self-consistent narrative’. Tolkien worked on his hobbit-cycle for nearly thirty years, if one accepts that he began composing The Hobbit about 1929 and was still working on the Appendices to The Lord of the Rings in 1955 (see Bibliography, pp. 7, 96). He worked on ‘the Silmarillion’, however, for more than twice as long, for his unpublished ‘Story of Kullervo’, written in 1914, contains the seed of the story of Túrin Turambar, later to be one of the ‘Great Tales’ (Letters, pp. 7, 214–5), while he was still thinking and writing about these texts and their problems in the last year of his life, 1973 (see Peoples, pp. 377–92). These sixty years of development are now set out in Christopher Tolkien’s ‘History of Middle-earth’, most particularly in volumes I–V and IX–XII (VI–VIII being concerned for the most part with the four parts of ‘The History of The Lord of the Rings’).

  The best short account of these sixty years is that of Charles Noad, in his essay ‘On the Construction of “The Silmarillion”’ in Legendarium pp. 31–681, now supplemented and extended by John Garth’s detailed and full-length study of Tolkien’s war years, Tolkien and the Great War2 , works of which I can give only brief summary here. But briefly: the first glimmerings of Tolkien’s new mythology are contained in a series of poems, mostly unpublished at the time and written while Tolkien was an Oxford undergraduate, 1914–1916 (a collection of these poems was rejected by the publishers Sidgwick and Jackson in April 1916). For most of the rest of that year Tolkien was training as an officer or on active service at the Battle of the Somme. Late in October 1916, however, he was returned to hospital in England with ‘trench fever’, and remained a convalescent with recurrent bouts of fever for the next two years. During this time, and for the most part in 1917, Tolkien wrote the material eventually published as The Book of Lost Tales, Parts One and Two, the first two volumes of ‘The History of Middle-earth’. These sixteen chapters contain a great part of the material which was to become The Silmarillion, including (under different titles) the ‘Great Tales’ of Beren and Lúthien, of Túrin Turambar, the Fall of Gondolin and the Tale of Eärendil, as also the elvish tales of ‘The Darkening of Valinor’ and ‘The Coming of the Noldoli’, and the mythological tales of ‘The Music of the Ainur’, ‘The Coming of the Valar’ and ‘The Chaining of Melko’. It is not too much to say that the outline of The Silmarillion was visible by the end of 1917 – or would have been if it had found any readers. For some years thereafter Tolkien was no doubt preoccupied with earning a living, but once he had found stable employment at the University of Leeds and then the University of Oxford, he began to put the tale of Túrin into alliterative verse and the tale of Beren into rhymed verse as ‘The Lay of the Children of Húrin’ and ‘The Lay of Leithian’ respectively, both these now published as The Lays of Beleriand, volume III of ‘The History of Middle-earth’; the first task occupied him approximately 1920–25, the second 1926–31. In 1926, though, Tolkien decided to show some of his poems, including a part of ‘The Lay of the Children of Húrin’ to his former schoolteacher R. W. Reynolds, and wrote a brief epitome of the ‘Lost Tales’ to give him some necessary background. This ‘Earliest “Silmarillion”’ led on to a longer epitome, the ‘Quenta’ or ‘Qenta Noldorinwa … drawn from the Book of Lost Tales’, written in 1930, and to two sets of annals, the ‘Annals of Valinor’ and ‘of Beleriand’ written at roughly the same time. These would lead on in their turn to two later sets of annals and a further expanded epitome, the ‘Quenta Silmarillion’, written between 1930 and 1937. These two bodies of work, from 1926–30 and 1930–37, appear in volumes IV and V of ‘The History of Middle-earth’, respectively The Shaping of Middle-earth and The Lost Road. In 1937, furthermore, Tolkien was asked by Stanley Unwin, publisher of The Hobbit, if he had any other material suitable for publication, and he sent him several manuscripts including ‘The Lay of Leithian’ and ‘The Quenta Silmarillion’. The publisher’s reader, Edward Crankshaw, seems to have been given only the former to read and some pages of the latter as background, and seems also to have been quite baffled by both, and by how they related to each other. Stanley Unwin accordingly gave Tolkien a polite rejection and urged him to start working on a sequel to The Hobbit. Tolkien took the rejection as being a rejection of the ‘Silmarillion’ material – which had in fact hardly been read at all – and started work on the Hobbit-sequel which was to become The Lord of the Rings.

  The ironies of this situation are well set out by Christopher Tolkien in The Lays of Beleriand, pp. 364–7, but one effect was that Tolkien ceased working on anything but the hobbit-cycle (with one significant exception discussed below, pp. 336–45) for more than a decade. Once The Lord of the Rings was effectively completed, however, he turned back with renewed energy to ‘the Silmarillion’, and in the early 1950s wrote two further sets of annals, ‘The Annals of Aman’ and ‘The Grey Annals’, along with a yet longer epitome, ‘The Later “Quenta Silmarillion”’, all these published along with much else in volumes X and XI of ‘The History of Middle-earth’, Morgoth’s Ring and The War of the Jewels. From this body of materials, dating from 1917, from the 1920s, the 1930s, and the 1950s, much of it written over and over again so as to become ‘a chaotic palimpsest, with layer upon layer of correction and wholesale rewriting, of riders and deletions’ (Lost Road, p. 199), Christopher Tolkien was eventually to extract the work published as The Silmarillion in 1977. In Unfinished Tales of Numenor and Middle-earth three years later he went on to give significant expansions of some of the ‘Silmarillion’ material, including the longest and most developed account of the Túrin story, ‘The Narn i Hîn Húrin’ or ‘Tale of the Children of Húrin’, as well as of the Second Age (that of Númenor), and the Th
ird Age, which ran from the first defeat of Sauron by Elendil, Gil-galad and Isildur to the destruction of the Ring and the departure of Frodo over sea.

  These two works, The Silmarillion of 1977 and the Unfinished Tales of 1980, are the subject of the rest of this chapter. It is true, of course, and as should be clear from the paragraphs above, that these are posthumous works which never reached the final shape intended by their author. But in the first place their author never reached a final intention, so his wishes are not being flouted; in the second place, he clearly very deeply wished to see the materials on which he had worked for so long at last published, as his son records (Lost Tales 1, p. 5); and in the third place, The Silmarillion has by now found millions of readers to confirm its existence as a substantial text. Many of those readers have furthermore found it a difficult and challenging text: as I remark below, it ‘could never be anything but hard to read’, so that some account of it is not only called for but likely to be actually useful. Meanwhile, the main reason for deciding to treat it and the Unfinished Tales here, instead of where (some would say) they belong, i.e. before The Hobbit and along with the ‘Philological Inquiries’ of chapter 2, is that that is the way most readers experience them. Probably ninety-nine people out of a hundred come to The Silmarillion and the Unfinished Tales only after reading The Lord of the Rings, while because of the uncompromising nature of the posthumously-published works, it will probably always be hard for most readers to understand them except after reading The Lord of the Rings. In that work Tolkien had set himself to write a romance for an audience brought up on novels. In the others, whether we consider them as earlier or later, we are left with far less guidance. It is accordingly the main aim of this chapter to help people to read The Silmarillion (I use once again the phrase which Tolkien and Gordon used of their edition of Sir Gawain) ‘with an appreciation as far as possible of the sort which its author may be supposed to have desired’. Subsidiary to that, though still important, are the issues of what it has to say and how it came to be: ‘sources’ and ‘designs’ once more, both things Tolkien disliked, but useful if not essential to a proper reading.

  I am grateful, however, for the opportunity to correct much of what I wrote twenty years ago in the light not only of the published material discussed above, but also of Christopher Tolkien’s comments on my initial version, for which see Lost Tales 1, pp. 1–4, 7, and Lost Tales 2, p. 57.

  The Dangers of Going On

  Before beginning any commentary, though, there is one very obvious question to ask, which is why Tolkien never saw The Silmarillion into print himself, and why the Unfinished Tales remained unfinished. There were, after all, nearly eighteen years between the appearance of The Return of the King and Tolkien’s death on 2 September 1973 – as long an interval as that between The Hobbit and The Fellowship of the Ring. During most of that period Tolkien was furthermore relieved of distracting academic duties, while he was not putting his energies into other creative work: almost all the sixteen poems in The Adventures of Tom Bombadil (1962) had seen print before, his contribution to The Road Goes Ever On (1968) consisted mostly of explanation and footnote, and Smith of Wootton Major (1967) is on the same relatively small scale as ‘Leaf by Niggle’. Besides, to repeat the point made above, The Silmarillion was very largely in existence from 1937 on; was also known to be in existence, and very much in demand! Why, then, could Tolkien not finish his legends of the First Age off?

  An answer to this, of a personal kind, has been given by Humphrey Carpenter on pp. 318–20 of his Biography. There was in Tolkien’s later life, he notes, ‘a perpetual discontinuity, a breaking of threads which delayed achievement and frustrated him more and more’. Partly the causes were external – loss of friends, hosts of visitors – but partly temperamental: Tolkien could not ‘discipline himself into adopting regular working methods’ (a fault of which he had been aware since the time of ‘Leaf by Niggle’). The Silmarillion was accordingly held up to a great extent, in Mr Carpenter’s view, by procrastination and bother over inessentials, by crosswords and games of Patience, by drawing heraldic doodles and answering readers’ letters – all compounded, one might add, by the failing energies of age (see Letters, p. 228). This is a convincing picture, and no doubt partly true. Yet it is not a picture of someone taking things easy: rather of continual, if misdirected, intellectual effort. One may remark that it is common experience to find that conscientious people who have a job to do that is too much for them (like writing a book) turn in their uncertainty to doing a succession of easier jobs instead (like answering their mail, drawing up syllabuses, or rationalising office organisation). Something like this seems to have been the case with Tolkien. He may have frittered his time away in constructing etymologies and writing kindly letters to strangers. But these activities occupied him, one may well think, because he could see he had painted himself into a corner: there were purely literary reasons for not finishing The Silmarillion, and these can be deduced not only from that work itself, but from almost the whole of Tolkien’s professional career. For one thing, both The Hobbit and The Lord of the Rings can be seen as primarily works of mediation. In the former Bilbo acts as the link between modern times and the archaic world of dwarves and dragons. In the latter Frodo and his Shire companions play a similar part, though the world they move in has also and in more complex ways been ‘mediated’, turned into a Limbo. Outside these works, though, hobbits are not to be met with, it would be almost impossible for them to exist in the much more rarefied air of the legends of the First Age, and without their existence modern readers lack guidance and a secure point of comparison. The very success of the hobbit-cycle was bound to make a work without hobbits a disappointment, or a puzzle.

  But there may have been a more complex reason for Tolkien’s long hesitation. To go back to ‘Beowulf: the Monsters and the Critics’: if this makes one thing clear it is that the literary quality Tolkien valued above all was the ‘impression of depth … effect of antiquity … illusion of historical truth and perspective’ which he found in Beowulf, in the Aeneid, or for that matter in Macbeth, Sir Orfeo, or the Grimms’ Fairy Tales. In all these works there was a sense that the author knew more than he was telling, that behind his immediate story there was a coherent, consistent, deeply fascinating world about which he had no time (then) to speak. Of course this sense, as Tolkien kept repeating, was largely an illusion, even a provocation to which a wise man should not respond. The ‘heroic lays’ which the Beowulf-poet knew and alluded to sound very fine from his allusions, but if we had them we might discover that the fascination came from his art, not theirs. ‘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, and he meant it. However he also meant everyone to realise that the ‘new thing’ was worth more than the ‘lost lore’.

  The application of this to his own career must (once The Lord of the Rings was published) have seemed all too obvious. One quality which that work has in abundance is the Beowulfian ‘impression of depth’, created just as in the old epic by songs and digressions like Aragorn’s song of Beren and Lúthien, Sam Gamgee’s allusions to the Silmaril and the Iron Crown, Elrond’s account of Celebrimbor, and dozens more. This, however, is a quality of The Lord of the Rings, not of the inset stories. To tell these in their own right and expect them to retain the charm they got from their larger setting would be a terrible error, an error to which Tolkien would be more sensitive than any man alive – though as Christopher Tolkien points out (Lost Tales 1, p. 3) the error would be in the expectation, not in the telling. Tolkien saw the problem and expressed his sense of it in a revealing letter dated 20 September 1963. He had clearly been asked for a sequel to The Lord of the Rings, and replied that he could give ‘another volume (or many) about the same imaginary world’. But he had many other things to do, he feared ‘the presentation will need a lot of work’, and he saw that the legends had to be made consistent with each other and with what he had already
published. He went on:

  I am doubtful myself about the undertaking. Part of the attraction of The L.R. is, I think, due to the glimpses of a large history in the background: an attraction like that of viewing far off an unvisited island, or seeing the towers of a distant city gleaming in a sunlit mist. To go there is to destroy the magic, unless new unattainable vistas are again revealed. (Letters, p. 333)3

  To go there is to destroy the magic. As for the revealing of ‘new unattainable vistas’, the problem there – as Tolkien must have thought many times – was that in The Lord of the Rings Middle-earth was already old, with a vast weight of history behind it. The Silmarillion, though, in its longer form, was bound to begin at the beginning. How could ‘depth’ be created when you had nothing to reach further back to?

  The problem was not absolutely insoluble: Milton, after all, had managed to begin his epic very near the beginnings of time, in Paradise Lost. Furthermore one can perhaps see the solution to which Tolkien, in his philological way, was drawn, namely to present the First Age ‘as a complex of divergent texts interlinked by commentary’ (UT, p. 1), the texts themselves being supposedly written by Men, of different periods, looking back across the ages to vast rumours of whose truth they knew only part, like Sam Gamgee responding to Gimli’s song ‘Of mighty kings in Nargothrond / And Gondolin’ (see once more Lost Tales 1, p. 3). The Silmarillion might then have come to look like (for example) The Saga of King Heidrek the Wise, written late but preserving intensely moving fragments of verse from some much older time now lost; even the editorial matter would then reinforce the effect of age and darkness (a device Tolkien used on a much smaller scale for The Adventures of Tom Bombadil). However that avenue was never explored to its end; and if it had been, one may doubt whether many readers would have grasped the total effect. A Silmarillion on that plan could have ended as merely a pastime for scholars. It is better, no doubt, to see it as it is now, ‘a completed and cohesive entity’ (UT, p. 1 again). But in any case The Lord of the Rings had created other problems for its author besides the issue of ‘depth’: these affected The Silmarillion, but show up more strongly in the Unfinished Tales.