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

by Tom Shippey

  Tolkien of course had an answer to the question, a private theory. It had been on his mind since 1926, when in his ‘General Philology’ chapter for The Year’s Work in English Studies vol. 5 he had hinted there might be such a science as Lautphonetik, translatable as ‘a phonology of sounds’. But all phonology is about sounds. Tolkien seems to have meant ‘an aesthetics of sounds’, a science that would explain why certain sounds or combinations of sounds produced different effects from others. Thirty years later he came back to the same idea in his last major learned work, the O’Donnell lecture on ‘English and Welsh’ given in Oxford in 1954 just after The Fellowship of the Ring came out. It is a discursive piece which covers many points, but one of them is a considered though not scientific attempt to say what makes a language beautiful. There is a pleasure, insisted Tolkien, ‘in the phonetic elements of a language and in the style of their patterns’. More pleasure may come from ‘the association of these word-forms with meanings’, but that is a separate stage. Tolkien said that he had only needed to see a vocabulary-list of Gothic for his heart to be taken by storm. The same was true of Finnish; and all along something of the sort had flashed on him at the sight of Welsh names on English coal-trucks, or such simple inscriptions as adeiladwyd 1887 on Welsh chapels.

  What kind of pleasure was this? At the age of 62 Tolkien felt no urge to found a new branch of learning, and fell back on the word ‘style’: the pleasure comes from ‘fitness … to a whole style’, is felt in ‘the reception (or imagination) of a word-form which is felt to have a certain style’. One feature of the Welsh ‘style’ might be ‘the fondness for nasal consonants … and the frequency with which word-patterns are made with the soft and less sonorous w and the voiced spirants f and dd contrasted with the nasals: nant, meddiant, afon, llawenydd, cenfigen, gwanwyn, gwenyn, crafanc, to set down a few at random’ (p. 40). The word and the theory were also in Tolkien’s head when he wrote Appendix F to The Lord of the Rings and declared that he had used names like Bree, Combe, Archet and Chetwood because they contained non-English elements and he needed words to sound ‘queer’, to imitate ‘a style that we should perhaps vaguely feel to be “Celtic”’. This was Tolkien’s major linguistic heresy. He thought that people could feel history in words, could recognise language ‘styles’, could extract sense (of sorts) from sound alone, could moreover make aesthetic judgements based on phonology. He said the sound of ‘cellar door’ was more beautiful than the sound of ‘beautiful’. He clearly believed that untranslated elvish would do a job that English could not.

  Could he have been right? Tolkien’s heresy was against the belief that language is only in a very limited way onomatopoeic, that we just happen from long habit to think ‘pig’ sounds piggish, while Danes (presumably) think pige sounds girlish. It was like him to think, Bombadil-style, that beneath all this there might be a ‘true language’, one ‘isomorphic with reality’, and that in any case there might often be a close connection between thing-signified, person-signifying, and language-signified-in, especially if the person who spoke the language lived on the thing. Legolas puts this view strongly in The Two Towers when he listens to Aragorn singing in Rohirric, a language he does not know, and then remarks (pp. 496–7), ‘That, I guess, is the language of the Rohirrim … for it is like to this land itself, rich and rolling in part, and else hard and stern as the mountains. But I cannot guess what it means, save that it is laden with the sadness of Mortal Men.’ He is right, but his is only one of many correct appraisals in the trilogy. The hobbits hear Gildor and the elves singing, and even the ones who know no Quenya find that ‘the sound blending with the melody seemed to shape itself in their thought into words which they only partly understood’ (p. 78). The dirge of Gléowine for Théoden has the same effect on p. 954. Gandalf uses the Black Speech of Mordor in Rivendell on p. 247 and his voice turns ‘menacing, powerful, harsh as stone’, so that the elves cover their ears and Elrond rebukes him, not for what he says but for the language he says it in. Conversely Merry ‘felt his heart leap’ at the songs of the Muster of Rohan (p. 775); and when Gimli sings of Durin Sam Gamgee – not a learned character – responds simply and directly to the ring of elvish and dwarvish names. ‘“I like that!” said Sam. “I should like to learn it. In Moria, in Khazad-dûm!”’ Obviously his response is a model one.

  One can see from all this why Tolkien made the seemingly wild assertion in 1955, that to him his work was ‘largely an essay in linguistic aesthetic’ (Letters, p. 220). One can also see that he was convinced his heresy had worked, for at the end of his remarks on ‘the Welsh linguistic style’ in ‘English and Welsh’ he brought forward The Lord of the Rings not as fiction but as evidence, declaring: ‘the names of persons and places in this story were mainly composed on patterns deliberately modelled on those of Welsh (closely similar but not identical). This element in the tale has given perhaps more pleasure to more readers than anything else in it.’ ‘Mainly’ is a bit of an exaggeration; the Welsh-modelled names in Middle-earth are only those of Gondor and of Elvish, or more accurately of Sindarin, and these are precisely the most doubtful cases. Many English readers, however, accustomed to the linguistic map of England with its varying Anglo-Saxon, Old Norse and Welsh components, might in all sobriety be able to say ‘Garstang sounds northern’ or ‘Tolpuddle sounds West Country’, and be able to go on from there to cope with the varying styles of the Shire, the Riddermark and the dwarves. There must be much more doubt over how many readers grasp first-hand that the Rivendell song on p. 231 is in Sindarin but Galadriel’s on p. 368 in Quenya, and that these two languages are furthermore related. Still, it would be as wrong to say that readers understand nothing of alien songs as to say they understand everything. As with place-names, landscapes, mythic fragments, ‘feel’ or ‘style’ is enough, however much it escapes a cerebral focus.

  Tolkien’s linguistic map of Middle-earth furthermore shows exceptionally well the relation in his mind between ‘inspiration’ and ‘invention’. One could argue that much of what he decided was forced on him: mere ‘invention’ to get out of difficulty. Thus it was inevitable that the story should be in modern English, and from the start of The Hobbit it was clear that the Bagginses at least were English by temperament and turn of phrase. Now Tolkien knew (none better) that logically this was impossible. He was committed then to a fiction in which ‘hobbitic’ was an analogue of English, was in fact a ‘stylistically’ neutral variant of a Common Speech. What then of dwarvish? Dwarf-names were already there in The Hobbit, and were in Old Norse, a language whose relationship to modern English was to Tolkien all but tangible. The dwarves then must have spoken a language analogous to the Common Speech in exactly the same way as Old Norse is to modern English; and since that was hardly likely in the case of two totally different species (men and hobbits are not really different species, see LOTR p. 2), Tolkien found himself committed also to the notion that the dwarves spoke human languages and used human names for convenience, but had a secret language and secret names of their own, the latter not even to be carved on tombs (a belief which he no doubt enjoyed because of its corroboration in the Grimms’ ‘Rumpelstiltskin’). Having fitted in English and Norse, Old English could not be far behind: hence the Riders with their entirely Old English terminology, their names which are often Old English nouns capitalised (like Théoden King, a phrase of exactly the same type as Bree-hill),13 the sense the characters occasionally indicate that ‘hobbitic’ is a worn-down variant of Rohirric in which words are changed but sound (p. 544) ‘not unfitting’. But by this stage ‘invention’ has stopped and ‘inspiration’ taken over. In the conversation between Pippin, Merry and Théoden outside Isengard Tolkien is no longer trying to explain old inconsistencies from The Hobbit, but writing ever deeper into a world with a life of its own.

  This led him, indeed, into yet further inconsistencies, or rather disingenuousnesses. Tolkien was obliged to pretend to be a ‘translator’. He developed the pose with predictable rigour, feigning not onl
y a text to translate but behind it a whole manuscript tradition, from Bilbo’s diary to the Red Book of Westmarch to the Thain’s Book of Minas Tirith to the copy of the scribe Findegil. As time went on he also felt obliged to stress the autonomy of Middle-earth – the fact that he was only translating analogously, not writing down the names and places as they really had been, etc. Thus of the Riddermark and its relation to Old English he said eventually ‘This linguistic procedure [i.e. translating Rohirric into Old English] does not imply that the Rohirrim closely resembled the ancient English otherwise, in culture or art, in weapons or modes of warfare, except in a general way due to their circumstances …’ (note on p. 1110). But this claim is totally untrue. With one admitted exception, the Riders of Rohan resemble the Anglo-Saxons down to minute details. The fact is that the ancient languages came first.14 Tolkien did not draw them into a fiction he had already written because there they might be useful, though that is what he pretended. He wrote the fiction to present the languages, and he did that because he loved them and thought them intrinsically beautiful. Maps, names and languages came before plot. Elaborating them was in a sense Tolkien’s way of building up enough steam to get rolling; but they had also in a sense provided the motive to want to. They were ‘inspiration’ and ‘invention’ at once, or perhaps more accurately, by turns.

  ‘The Council of Elrond’

  The gist of what has been said in this chapter is that The Lord of the Rings possesses unusual cultural depth. ‘Culture’ is not a word Tolkien used much; it changed meaning sharply during his lifetime, and not in a direction he approved. Still, one can see a deep understanding of its modern meaning of ‘the whole complex of learned behaviour … the material possessions, the language and other symbolism, of some body of people’ in chapter 2 of Book II of The Fellowship of the Ring. This marks a jump-off point for the characters, whose objective is disclosed within it. It was also I suspect a jump-off point for Tolkien, since after that he was no longer writing his way through landscapes he had travelled before. It is therefore perhaps not surprising that as with the house of Beorn in The Hobbit ‘The Council of Elrond’ should provide a sudden introduction to archaic and heroic worlds confronting and overwhelming modern, practical ones. The later work is, however, many degrees more complex than its earlier analogue, being indeed an interweaving of at least six major voices besides minor ones and reported ones; as well as telling a complex tale in complex fashion what all these voices do is present, in our language, a violent ‘culture-clash’.

  This comes out most in the speeches and scripts impacted inside Gandalf’s interrupted monologue of pages 243–58, the fifth and longest from a major speaker (the others coming from Glóin, Elrond, Boromir, Aragorn, Legolas). Within that monologue Gaffer Gamgee functions as a kind of base-line of normality – and, concomitantly, of emptiness. ‘I had words with old Gamgee’, Gandalf reports, ‘Many words and few to the point’:

  ‘“I can’t abide changes,” said he, “not at my time of life, and least of all changes for the worst.” “Changes for the worst,” he repeated many times.

  ‘“Worst is a bad word,” I said to him, “and I hope you do not live to see it.”’

  It is indeed a bad word, especially when all the Gaffer has to complain about is the Sackville-Bagginses; Denethor uses it as well, much later (p. 796), but again with ominous effect. As for ‘abide’, as used by Gaffer Gamgee it has almost no semantic content at all; in context it means ‘bear, tolerate, put up with’, but in that sense is simply untrue. The Gaffer can abide changes; he just has. He means only that he doesn’t like them. But there is a moral for him in the history of the word, which has the frequent early sense of ‘to await the issue of, to wait (stoically) for, to live to see’. In this last sense the Gaffer could ‘abide’ changes, and he does. Right at the end he moralises, stubborn as ever, ‘It’s an ill wind as blows nobody any good, as I always say’ (my italics), ‘And All’s well as ends Better’ (p. 999). At least he has learnt to eschew superlatives. But his language in Gandalf’s monologue conveys an unwelcome reminder of psychological unpreparedness.

  His son Sam re-establishes the hobbits slightly with his terminal comment, ‘A nice pickle we have landed ourselves in, Mr Frodo’, for though he is as obtuse as his father – Sam got himself into trouble, but Frodo did not – this blindness does coexist with a thoughtless courage, a relish for gloom, and a refusal to see Doomsday as more than a ‘pickle’, all adding up to the notorious Anglo-hobbitic inability to know when they’re beaten. However there is another modern voice in Gandalf’s monologue to act as vehicle for cultural contrast: this is Saruman’s. He has hardly been mentioned before, and the question whether he is good or bad is more difficult to decide than with most. But when he is introduced by Gandalf, we know what to think very soon; the message is conveyed by style and lexis. Saruman talks like a politician. ‘We can bide our time’, he says, using a fossilised phrase:

  ‘we can keep our thoughts in our hearts, deploring maybe evils done by the way, but approving the high and ultimate purpose: Knowledge, Rule, Order; all the things that we have so far striven in vain to accomplish, hindered rather than helped by our weak or idle friends. There need not be, there would not be, any real change in our designs, only in our means.’ (p. 253)

  What Saruman says encapsulates many of the things the modern world has learnt to dread most: the ditching of allies, the subordination of means to ends, the ‘conscious acceptance of guilt in the necessary murder’.15 But the way he puts it is significant too. No other character in Middle-earth has Saruman’s trick of balancing phrases against each other so that incompatibles are resolved, and none comes out with words as empty as ‘deploring’, ‘ultimate’, worst of all, ‘real’. What is a ‘real change’? The OED’s three columns of definition offer nothing appropriate; the word has got below dictionary level. As we all know, ‘real’ is now a word like ‘sincere’ or ‘genuine’, a word whose meaning its speaker asks you to take for granted, a politician’s word, an advertiser’s word. ‘Real change’ shows Saruman up with even greater economy than ‘changes for the worst’ does Gaffer Gamgee.

  By contrast with these familiar styles and voices several of the other participants in the Council come over as archaic, blunt, clear-sighted. Gandalf himself uses an older vocabulary than usual, as if to authenticate himself, and Elrond’s speech (pp. 237–9), as is only suitable for one so old, is full of old-fashioned inversions of syntax and words like ‘weregild’, ‘esquire’, ‘shards’. Its burden is to state the Northern ‘theory of courage’, as Tolkien called it in his British Academy lecture, whose central thesis is that even ultimate defeat does not turn right into wrong.* Elrond has seen ‘many defeats, and many fruitless victories’, and in a way he has even given up hope, at least for his adopted people the elves (see p. 262 and further p. 1006); but this does not make him change his mind or look for easy options.

  The heroic note is struck most firmly, however, by the dwarf Glóin, or rather by his report of the dialogue between Sauron’s messenger and that exemplar of stubbornness King Dáin. The messenger offers ‘great reward and lasting friendship’ in return for information about hobbits, or for the Ring. If Dáin refuses, he says:

  ‘ “… things will not seem so well.”

  ‘At that his breath came like the hiss of snakes, and all who stood by shuddered, but Dáin said: “I say neither yea nor nay. I must consider this message and what it means under its fair cloak.”

  ‘“Consider well, but not too long,” said he.

  ‘“The time of my thought is my own to spend,” answered Dáin.

  ‘“For the present,” said he, and rode into the darkness.’

  We get exchanges like this several times in The Lord of the Rings, mostly involving dwarves: Elrond and Gimli swap grim proverbs in the next chapter, Théoden King silences Merry in similarly abrupt style in Book V chapter 2, and Appendix A offers several dwarvish dialogues around the battle of Azanulbizar. Their unifying feature is
delight in the contrast between passionate interior and polite or rational expression; the weakness of the latter is an index of the strength of the former. Thus the messenger’s ‘things will not seem so well’ works as violent threat; ‘not too long’ means ‘extremely rapidly’. In reply Dáin’s ‘fair cloak’ implies ‘foul body’ and the obscure metaphor of spending the ‘time of my thought’ indicates refusal to negotiate under threat. Both participants seek to project a cool, ironic self-control. If Elrond’s recommendation was courage, and Gandalf’s hope, the dwarves’ contribution to the ethical mix of the Council is a kind of unyielding scepticism. This virtue is no longer much practised, swept away by the tide of salesmanship, winning friends and influencing people, the belief that all aggression is dissolved by smiles. We no longer even have a name for it – except perhaps that people who call their tea their ‘baggins’ might recognise it in their approving use of ‘bloody-mindedness’ (not recorded by the OED). Whatever it is, it comes over in Dáin’s speech as a force: words imply ethics, and the ethics of the spokesmen of Middle-earth fit together, beneath surface variation. None of them but Saruman pays any attention to expediency, practicality, Realpolitik, ‘political realism’.

  Any one of the counsellors in this chapter would bear similar analysis. Gandalf’s account of Isildur makes a point through its combination of ancient words and endings (‘glede’, ‘fadeth’, ‘loseth’, etc.) with sudden recall of the words of Bilbo and Gollum. ‘It is precious to me, though I buy it with great pain’; the ‘reality of human nature’ persists. More subtly Aragorn and Boromir strike sparks off each other through their ways of speech as well as their claims, Aragorn’s language deceptively modern, even easy-going on occasion, but with greater range than Boromir’s slightly wooden magniloquence. There is even significance in Aragorn letting his rival have the last word in their debate, with a clause which is perfectly in line with modern speech – ‘we will put it to the test one day’ – but also relates easily to the vaunts of ancient heroes, like Ælfwine’s nú mæg cunnian hwá céne sy in The Battle of Maldon, ‘now who is bold can be put to the test’. Still, the overriding points are these: the ‘information content’ of ‘The Council of Elrond’ is very high, much higher than can be recorded by analyses like this; much of that information is carried by linguistic mode; nevertheless most readers assimilate the greater part of it; in the process they gain an image of the ‘life-styles’ of Middle-earth the solider for its occasional contrasts with modernity. Language variation gives Tolkien a thorough and economical way of dramatising ethical debate.