These traditional fairy-tales, however, have severe limitations in at least two ways. One is that they are detached from each other. There may be a vague sense that they all take place in something like the same world, a dimly-perceived far past which, as Bilbo says of Gandalf’s stories, is all about ‘dragons and goblins and giants and the rescue of princesses and the unexpected luck of widows’ sons’. But this world is connected to no known history or geography, and furthermore there is no connection between any of the tales themselves. They cannot, then, be developed. They stimulate the imagination, but do not entirely satisfy it – not, at least, in the way that modern readers expect, with a full plot and developed characters and, perhaps most of all, a map.
And there is another problem with fairy-tales which Tolkien sensed very keenly. This is that from their very beginning, from the time, that is, when scholars began to take an interest in them and collect them, they seemed already to be in a sense in ruins. The Grimm brothers, in the nineteenth century, quite certainly had as a main motive for making their collection of Haus-und Kindermärchen the wish to do a kind of literary rescue archaeology. They were convinced that the tales they collected, brief as they were and deep sunk in the social and literary scale, still preserved fractions of some older belief, native to Germany but eventually suppressed by foreign missionaries, foreign literacy, and Christianity. Jacob Grimm, the elder brother, indeed tried to fit the pieces together, or at least collect as many as he could, in his extensive work Deutsche Mythologie, or ‘Teutonic Mythology’. The attempt has since then been generally ignored or derided, but there were some true observations behind it. One was that in some cases, like ‘dwarf (see pp. xiv-xv above), all Germanic languages had preserved the same word; though they had clearly not borrowed it from each other, because the word had always changed as the languages had changed, over the millennia. Accordingly English speakers said ‘dwarf, Germans Zwerg, and Icelanders dvergr. What this seemed to indicate was that the word was very old, much older than the fairy-tales in which it was preserved. But it must have been used in fairy-tales all along. What could those old tales have been like, before the whole mythology had been downgraded to children and their nursemaids?
Strong confirmation of this theory furthermore came from the rediscovery, during the eighteenth and nineteenth centuries, of fragments of the old adult and aristocratic literature of Northern Europe. It should be noted that this had been totally lost and forgotten for many centuries – Shakespeare, for instance (though he clearly knew something about fairy-tales, more than he was prepared to show) can have known nothing about the higher literature that lay behind them. The one surviving copy of the Old English epic Beowulf, with its strong interest in monsters, elves and orcs included, lay as far as we can tell unread and almost unnoticed from the Norman Conquest in 1066 till its eventual publication in Copenhagen in 1815. The Old Norse poems of the Elder Edda likewise lay unknown and for the most part in one manuscript in an Icelandic farmhouse, till they were rediscovered and slowly and patchily republished by scholars including the Grimms. The Middle English poem Sir Gawain and the Green Knight, with its very similar interest in elves and ettins, was hardly known and certainly not a part of university syllabuses till it was edited by Tolkien himself and his junior Leeds colleague E.V. Gordon in 1925. To those, however, who did read these poems and their many badly-preserved analogues, there came a feeling that their authors had indeed known something, something consistent with each other and with the much later fairy-tales of modern times: and that you might just possibly be able to work out what it was. This is the philological activity of ‘reconstruction’, as discussed in the Foreword above, p. xv.
In two ways, then, fairy-tale and its ancestors provoked the imagination, suggested a wider world which they then did not explore. You could work back from the dwarves of ‘Snow White’; or you could work out from the dwarves of Ruodlieb (a poem written in Latin by a German poet of the twelfth century) or of the Elder Edda (a collection of poems written in Old Norse, some of them probably even older than Ruodlieb). This is what Tolkien was doing: as is proved, for instance, by his stubborn insistence on writing, and making the printers print the word ‘dwarves’, even though (as he says in his opening note to The Hobbit) ‘In English the only correct plural of dwarf is dwarfs’. If that is the only correct form, why use an incorrect one? Because the -ves ending is a sign of the word’s antiquity, and so its authenticity. Even in modern English, old words ending in -f make their plural with -ves, as long as they have remained in constant use: so hoof/ hooves, life/lives, sheaf/sheaves, loaf/loaves. Dwarf/dwarves might have developed the same way, but clearly fell out of general use, and so was assimilated (probably by literates, schoolteachers and printers) to the simpler pattern of tiff(s), rebuff(s), and so on. Tolkien meant to turn back this particular clock. The Grimms had done exactly the same with their insistence that the German plural for ‘elf’ ought to be Elben, not Elfen (a form borrowed late on from English, itself by that time historically mistaken). Tolkien furthermore quite clearly had in mind from the start of The Hobbit a poem from the Elder Edda. It gave him all the names of ‘Thorin and Company’.
There are, one might note, surprisingly few names in The Hobbit, certainly by comparison with The Lord of the Rings. Most natural features have names which are just common nouns and adjectives with capital letters, like The Hill, The Water, Dale, the Long Lake, the River Running, the Lonely Mountain, Ravenhill and indeed The Carrock. To Bilbo’s timid question about the meaning of the last Gandalf replies crushingly that ‘[Beorn] called it the Carrock, because carrock is his word for it. He calls things like that carrocks, and this one is the Carrock because it is the only one near his home’. In addition to this we have a few hobbit-names (Baggins and Took, Hobbiton, and the auctioneers Grubb, Grubb, and Burrowes); rather more names from Tolkien’s already developed but here only hinted-at elvish mythology (Elrond, Gondolin, Girion, Bladorthin, Dorwinion, and more doubtfully Orcrist and Glamdring); and a few incidentals (Radagast, Bolg and Azog the goblins, Care and Roäc the ravens, Bard). But when it comes to dwarf-names, Tolkien gives full measure.
He found them in the poem Völuspá, ‘The Sybil’s Vision’, one section of which is called the Dvergatal, ‘the Tally of the Dwarves’. In the original Old Norse, this contains rather more than sixty names, mostly strung together as a simple rhythmic list, repeated in slightly different form in Snorri Sturluson’s thirteenth-century guide to Norse mythology, the Skaldskaparmál, the ‘Treatise on Skald-ship’, or one might say, ‘Art of Poetry’. Part of Snorri’s version goes as follows, and one can see immediately the connection with Tolkien:
Nár, Náinn, Nípingr, Dáinn,
Bífur, Báfur, Bömbur, Nóri,
Órinn, Ónarr, Óinn, Miöðvitnir,
Vigr og Gandálfr, Vindálfr, Porinn,
Fíli, Kíli, Fundinn, Váli,
Þrór, Þróinn, Þettr, Litr, Vitr…
Eight of the thirteen dwarf-names of Tolkien’s Thorin and Company are here, along with the name of Thorin’s relative Dain, his grandfather Thror, and something close to his father Thrain. Four of the other five (Dwalin, Gloin, Dori, Ori) are not far away, as are Durin, in both The Hobbit and Völuspá the dwarves’ legendary ancestor, and Thorin’s nickname Oakenshield, or Eikinskjaldi. Only Balin – a famous name in Arthurian story, though that is perhaps a coincidence – is not in Snorri’s list.
However Tolkien did not just copy the ‘Tally of the Dwarves’,