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

by Tom Shippey

  APPENDIX A

  TOLKIEN’S SOURCES: THE TRUE TRADITION

  Tolkien himself did not approve of the academic search for ‘sources’. He thought it tended to distract attention from the work of art itself, and to undervalue the artist by the suggestion that he had ‘got it all’ from somewhere else. This appendix accordingly does not attempt to match ‘source’ to ‘passage’ in Tolkien. It does however offer a brief guide to the works which nourished Tolkien’s imagination and to which he returned again and again; since many of them are not well known, this may give many people who have enjoyed Tolkien something else to enjoy. Whether that changes their reading of The Lord of the Rings or The Silmarillion is less important: though in fact comparison with ‘the sources’, in my experience, almost always brings out Tolkien’s extremely keen eye for the vital detail.

  He was also very quick to detect the bogus and the anachronistic, which is why I use the phrase ‘true tradition’. Tolkien was irritated all his life by modern attempts to rewrite or interpret old material, almost all of which he thought led to failures of tone and spirit. Wagner is the most obvious example. People were always connecting The Lord of the Rings with Der Ring des Nibelungen, and Tolkien did not like it. ‘Both rings were round’, he snarled, ‘and there the resemblance ceases’ (Letters, p. 306). This is not entirely true. The motifs of the riddle-contest, the cleansing fire, the broken weapon preserved for an heir, all occur in both works, as of course does the theme of ‘the lord of the Ring as the slave of the Ring’, des Ringes Herr als des Ringes Knecht. But what upset Tolkien was the fact that Wagner was working, at second-hand, from material which he knew at first-hand, primarily the heroic poems of the Elder Edda and the later Middle High German Nibelungenlied. Once again he saw difference where other people saw similarity. Wagner was one of several authors with whom Tolkien had a relationship of intimate dislike: Shakespeare, Spenser, George MacDonald, Hans Christian Andersen. All, he thought, had got something very important not quite right. It is especially necessary, then, for followers of Tolkien to pick out the true from the heretical, and to avoid snatching at surface similarities.

  The single work which influenced Tolkien most was obviously the Old English poem Beowulf, written in Tolkien’s opinion somewhere round the year 700. The best edition of this is by F. Klaeber (Boston: D. C. Heath, 1922, 3rd edn. 1950). There are many translations of it, including the one by J. R. Clark Hall and C. L. Wrenn to which Tolkien wrote the ‘Preface’ in 1940, and Tolkien’s own, as yet unpublished. The reasons for the poem’s appeal to him, however, seem to me to be expressed best in R. W. Chambers, Beowulf: an Introduction (Cambridge: Cambridge University Press, 1921, 3rd edn. with supplement by C. L. Wrenn, 1959). The first two chapters of this show with particular force and charm the way in which history and fairy-tale are in Beowulf intertwined. Other Old English poems which Tolkien used include The Ruin, The Wanderer and The Battle of Maldon, all edited and translated in Richard Hamer’s A Choice of Anglo-Saxon Verse (London: Faber and Faber, 1970), and the ‘Treebeard-style’ gnomic poems Maxims I and II, edited and translated, along with Solomon and Saturn II, in my Poems of Wisdom and Learning in Old English (Cambridge: D. S. Brewer and Totowa, N.J.: Rowman and Littlefield, 1976). Tolkien’s own editions of Exodus and Finn and Hengest (see ‘Abbreviations’ for publishing details) provide much insight into his views on history, heroic continuity, and the relationship between Christian and pagan thought. I have discussed both in my review ‘A Look at Exodus and Finn and Hengest’ in Arda (the journal of the Swedish Tolkien Society), no. 3, 1982–3, pp. 72–80. Very briefly one might say that Tolkien valued Exodus especially as an example of Christian material treated in an old-fashioned or heroic style: his own fiction being a similar mixture but the other way round.

  The poem of Solomon and Saturn just referred to centres on a riddle-contest, a form with two other prominent examples, both in Old Norse. One is Vafðrúðnismál, one of twenty-nine poems in the Elder or Poetic Edda, a collection made in Iceland perhaps about AD 1200. Tolkien knew this collection well, drawing on the poem Völuspá for the names of the dwarves in The Hobbit, on Fáfnismál for the conversation with Smaug, and on Skirnismál for the ‘tribes of orcs’ and the ‘Misty Mountains’. More generally the whole collection gives a sharper edge than Beowulf to the ideal of heroism, and a stronger sense of a tumultuous history filtering down to echo and hearsay. Both points are well brought out in the old, now-superseded edition of the Corpus Poeticum Boreale, by Gudbrand Vigfusson and E York Powell (2 vols., Oxford: Clarendon Press, 1883), as also in Ursula Dronke’s much later partial editions, The Poetic Edda, Volume I: Heroic Poems and Volume II: Mythological Poems (Oxford: Clarendon Press, 1969, 1997). The best translation of the whole of The Poetic Edda remains Norse Poems, by Paul B. Taylor and W. H. Auden (London: Athlone Press, 1981). An earlier version of this last was dedicated to Tolkien, and Auden and Tolkien in the end became friends and correspondents.

  The other major riddle-contest in Old Norse appears in The Saga of King Heidrek the Wise, edited and translated by Christopher Tolkien (London: Nelson, 1960). The relevance of this to all Tolkien’s work, including The Silmarillion, should be obvious; the combination of pride, ferocity and sadness in the older poem of ‘The Battle of the Goths and Huns’ which has found its way into the saga seems to be the note that Tolkien often aimed at, and as often disapproved. Another fornaldarsaga or ‘saga of old times’ of much interest to Tolkien readers is the Völsunga Saga; William Morris’s translation of it in 1870 was reprinted with an introduction by Robert W. Putnam (London and New York: Collier Macmillan, 1962); it has also been edited and translated by R. G. Finch in the same series as Christopher Tolkien’s Heidrek (London: Nelson, 1965), while there is a recent translation by Jesse Byock, The Saga of the Volsungs (Berkeley: University of California Press, 1990). Meanwhile the other great work of Old Norse mythology, later and more ‘novelistic’ in tone than the poems, is the Prose Edda of Snorri Sturluson, written in Iceland between 1225 and 1241. This too is a work of ‘mediation’, like Tolkien’s; Snorri was a Christian trying to preserve pagan material for his countrymen and for the cause of poetry. In several ways, especially its combination of respect for antiquity with a certain detached humour, Snorri prefigures Tolkien. One of the ‘lost’ poems known only by its quotations was a model for ‘Aldarion and Erendis’, see above; another poem added to a manuscript of Snorri’s Edda by some well-wisher is Rígsþula, for the relevance of which see note 13 to chapter 4. There is a good translation of the whole of this work in the Everyman Classics series, Snorri Sturluson, Edda, trans. Anthony Faulkes (London: Dent, 1987).

  It is a jump of many centuries to the great ‘fairy-tale’ collections of the nineteenth, but, as mentioned above, Jacob Grimm at least thought the similarity between German fairy-tale and Scandinavian ‘Edda’ striking enough to prove that both were the debris of a greater unity. Whether this is so or not – the question is discussed intensively in the collection The Shadow-walkers, see note 16 to chapter 2 – the folk-tales of North-West Europe affected Tolkien profoundly. The major collections (from his point of view) certainly included that of the brothers Grimm, printed first in 1812, but expanded, revised and translated ever since: I have used The Complete Grimm’s Fairy Tales, no translator named, published in London by Routledge and Kegan Paul in 1975, but Tolkien no doubt read them in German – he relished the dialect forms of ‘Von dem Machandelboom’, quoting it in the original in ‘On Fairy-Stories’ (Tree, pp. 31–2). Another work he refers to is Popular Tales from the Norse, collected by P. C. Asbjörnsen and J. I. Moe and translated by Sir George Dasent, published first in English in Edinburgh, 1851, but reprinted in London by The Bodley Head, 1969. In the same modern series (1968) is English Fairy Tales by Joseph Jacobs, a reprint from 1890; No. 21, ‘Childe Rowland’ is a ‘Dark Tower’ story, see above. Tolkien also quoted from J. F. Campbell’s Popular Tales of the Western Highlands (4 vols., Paisley and London: Alexander Gardner, 1890–3).

>   Parallel to the fairy-tale tradition collected by the Grimms and others is the ballad tradition, also preserved by collectors of the nineteenth century and containing much similar, and similarly archaic material. The greatest collection of these is certainly F. J. Child’s The English and Scottish Popular Ballads, first published in five volumes by Houghton Mifflin, Boston, 1882–98, and reprinted by Dover Publications, New York, 1965. Particularly vital to this are the philological introductions to each ballad, see especially no. 19, ‘King Orfeo’, no. 60, ‘King Estmere’, and others; while Tolkien also almost certainly read Lowry C. Wimberly’s commentary Folklore in the English and Scottish Ballads (Chicago: University of Chicago Press, 1928). Tolkien probably also knew the Danish collection begun by Svend Grundtvig, Denmarks gamle Folkeviser, out in 12 volumes from 1853 onwards, and partly available to English readers in A Book of Danish Ballads, ed. Axel Olrik, trans. E. M. Smith-Dampier (Princeton: Princeton University Press, 1939). The collection includes several elf-and-mortal or mermaid-and-mortal ballads like Tolkien’s own poems mentioned or reprinted and translated in Appendix B below. The collector’s father, Nicolai Grundtvig, was in my opinion the ‘Beowulfian’ whom Tolkien most respected – he appears in ‘Monsters’ as one of the ‘very old voices’ calling ‘“it is a mythical allegory” … generally shouted down, but not so far out as some of the newer cries’. Grundtvig senior was also remarkable for his efforts to reconcile his studies in pagan antiquity with his position as evangelistic reformer and ‘apostle of the North’, arguing for Óthinn as a ‘forerunner’, Earendel-like, of the Messiah, both ‘sons of the Universal Father’.

  But Tolkien was also interested in later traditions, and even in American traditions: anyone who reads the ‘Introduction’ to English Folk-Songs from the Southern Appalachians, collected by Olive D. Campbell and Cecil I. Sharp (New York and London: G. P. Putnam’s Sons, 1917) will be struck by the strange resemblance of the mountain country of North Carolina before the First World War to ‘the Shire’ as Tolkien described it. Nor is this accident. A piece by Mr Guy Davenport in the New York Times (23 February 1979) records Tolkien grilling an American classmate of his for ‘tales of Kentucky folk … family names like Barefoot and Boffin and Baggins and good country names like that’. Old country names, one might add: in Kentucky and its neighbours, Tolkien obviously thought, there had for a time been a place where English people and English traditions could flourish by themselves free of the chronic imperialism of Latin, Celtic and French. In the same way Fenimore Cooper’s hero Natty Bumppo prides himself on his English ancestry, while Tolkien recorded an early devotion to Red Indians, bows and arrows and forests (‘OFS’ in Tree, p. 41). The journey of the Fellowship from Lórien to Tol Brandir, with its canoes and portages, often recalls The Last of the Mohicans, and as the travellers move from forest to prairie, like the American pioneers, Aragorn and Éomer for a moment preserve faint traces of ‘the Deerslayer’ and the Sioux, see above. The complaint in one of the sillier reviews of The Lord of the Rings, that none of its characters (except Gimli) had ‘an even faintly American temperament’, is as imperceptive as irrelevant. The ‘American temperament’ has roots in many places, but England is not the least among them: caelum non animam mutant qui trans mare currunt.

  The medieval or middle period between the high vernacular culture of North-West Europe and the collecting or ‘reconstructing’ era of Child and the Grimms was in several ways a disappointment to Tolkien, though of course he found much in its more traditional poems such as Pearl, Sir Gawain and Sir Orfeo. His translations of these must be recommended (see ‘Abbreviations’ under SGPO), as also the edition of Sir Gawain by himself and E. V. Gordon (SGGK), and of Pearl by E. V. Gordon alone (Oxford: Clarendon Press, 1953). Tolkien’s assistance to the latter is acknowledged. Tolkien also lived for many years with the Ancrene Riwle, or Ancrene Wisse, and those concerned to seek out an influence on him might read The Ancrene Riwle, translated by Mary Salu, a pupil of his (London: Burns & Oates, 1955). That work was written c. 1225, in Herefordshire. Close by in both place and time was the Brut, an Arthurian Chronicle-epic by one Lazamon. Tolkien certainly valued this as a repository of past tradition, borrowing from it, for instance, Éowyn’s word ‘dwimmerlaik’. At some stage he must also have noted that the stream by which the poet lived – it is a tributary of the Severn – was the River Gladdon. Part of the poem can be found in Selections from Lazamon’s Brut, ed. G. L. Brook with preface by C. S. Lewis (Oxford: Clarendon Press, 1963, 2nd edn. 1983), or in Lazamon’s Arthur: the Arthurian Section of Lazamon’s Brut, eds. W.R.J. Barron and S. C. Weinberg (Harlow: Longman, 1989), while the whole work is translated by Rosamund Allen as Lawman’s Brut (London: J. M. Dent, 1992). I am also persuaded that Tolkien found stimulus in the slightly later legends of St Michael and St Brendan in The Early South English Legendary, edited by C. Horstmann for the Early English Text Society (London: Trübner, 1887).

  Two other clear medieval English influences on Tolkien are Mandeville’s Travels, written about 1375, and available in a modern translation by M. C. Seymour (London: Oxford University Press, 1968); and the Lais of Marie de France, also available in translation by Glyn Burgess and Keith Busby (London: Penguin Classics, 1986). The latter is a clear source for ‘Aotrou and Itroun’, the former perhaps the best guide to Tolkien’s notions of the trees of Sun and Moon, the Paradis terrestre, and the road to it encumbered by enchantments like those of the Dead Marshes. Many phrases from this book seem to have stayed in Tolkien’s mind. One should add that for all their names and preferred languages, both Sir John Mandeville and Marie de France were certainly English by nationality.

  Dealing with Tolkien’s knowledge of other languages could protract this essay interminably, but a source of the highest importance was clearly the Finnish epic Kalevala, which Tolkien knew in the translation of W. F. Kirby (London and New York: Dent and Dutton, 1907). For a more modern and scholarly treatment, see note 15 to chapter 7. I would like to take the opportunity here (prompted by Dr Osmo Pekonen) of correcting the remark made on Author, p. xxxiv, that ‘Lönnrot’s Kalevala is now viewed with suspicion by scholars because … you cannot tell what is by him and what is “authentic”’. Dr Pekonen informs me that Lönnrot kept such careful scholarly records that this distinction can readily be made, and is made throughout in ‘the flagship of Finnish philology’, the 34 volumes of Suomen Kansan Vanhat Runot, ‘The Ancient Runes of the Finnish People’ (1908–1999). Dr Pekonen further cites Lönnrot’s ‘Preface’ to the Kalevala, in which he declares that he felt he had ‘the right to arrange the poems [he had collected] in such a way that they formed the best possible combination’, just as the traditional singers did. This, perhaps, is a right Tolkien would also have wished to claim.

  Also recommendable is the Irish Imram, The Voyage of Bran Son of Febal, ed. Kuno Meyer (2 vols., London: David Nutt, 1895–7). Tolkien’s wanderings in German romance, though probably considerable – see the remarks on Orendel and others above, here and here – are too complex for me to trace. Some guides through the wilderness of heroic legend can be found, however, in the philologists: and when it comes to it these were the men whom Tolkien probably followed with the keenest and most professional interest. Three major works may be cited, though they give the interested reader no more than a taste: Grimm’s Teutonic Mythology, trans. J. S. Stallybrass (4 vols., London: George Bell, 1882–8); R. W. Chambers, Widsith, A Study in Old English Heroic Legend (Cambridge: Cambridge University Press, 1912); and R. M. Wilson, The Lost Literature of Medieval England (London: Methuen, 1952). It should be noted that a vital part of this latter came out as early as 1941, in plenty of time for Tolkien to recall it in The Lord of the Rings, see note 19 to chapter 5.

  The last major ‘old’ source for Tolkien which need be mentioned lies in history and chronicle. Gibbon’s Decline and Fall of the Roman Empire certainly stayed in Tolkien’s mind, though probably in the same compartment as Wagner; ‘Radagaisus’ may be found in its ‘Index’, if not �
��Radagast’, as also ‘Fredegarius’, though not ‘Frodo’. Of the Latin histories which Gibbon used the most interesting for Tolkien were probably Saxo Grammaticus’s History of the Danes, of which Books 1–9 were translated by Oliver Elton, with an introduction by F. York Powell (London: David Nutt, 1894), now superseded by Peter Fisher’s translation in two volumes, edited by Hilda Ellis Davidson (Cambridge: D. S. Brewer, and Totowa, NJ: Rowman and Littlefield, 1979–80); and The Gothic History of Jordanes, translated by C. C. Mierow (Princeton: Princeton University Press, 2nd edn. 1915). One has to add that Mr Mierow’s grasp of Gothic, unlike his Latin, is feeble. The true opinions of Jordanes lie buried in Karl Müllenhoff’s notes to Mommsen’s edition of 1882. A final note on the Germanic tribes as they appealed to Tolkien’s imagination may be found in Sir Charles Oman’s classic, A History of the Art of War in the Middle Ages (London: Methuen, 1898). Its description on pp. 48–51 of the Lombards, that other Germanic ‘horse-folk’ par excellence, strongly recalls the Riders of the Mark.