As before, the story continues with nothing more than a paragraph break between what are now chapters IX and X in the middle of manuscript page 129 (Marq.1/1/11:1). On the back of this page is a faint sketch-map showing the forest’s eastern edge, the precursor to the ‘Home Manuscript’ map shown on Plate I [bottom].
They rounded a steep shoulder of land that came down upon their rightTN1 under which the rocky feet [> under the rocky feet of which] the deepest stream flowed lapping and bubbling. Suddenly it fell away. The trees ended. Then Bilbo saw a sight. The land opened wide about them, filled with the waters of the river which broke up and wandered into marshes and pools and isles on either side, though a strong water flowed ever on through the midst. And far away, his dark head in cloud, there loomed the Mountain. Its nearest neighbors to the N.E. and the tumbled land that joined it to them could not be seen.TN2 All alone it rose, and looked across the plain to the forest. The Lonely Mountain. Bilbo had come far and through many weary adventures to see it. And now he did not like the look of it at all!
Listening to the talk of the raftmen he soon realized that dreary as had been their emprisonment, and unpleasant as was their position even now, they were really more fortunate than they guessed.
They would have had small chance of doing more than glimpse that mountain from afar had they gone on and found the way out of the forest unhindered. The lands had changed since the days of the dwarves. Great floods and rains had swollen the waters. The marshes and bogs had spread wider and wider on either side. Paths had vanished, and many a wanderer and a rider too, who had tried to find his way across. Only by the river was there any longer a safe way from the skirts of Mirkwood to the mountain-shadowed plain beyond.TN3 So they went on and on; and always the Mountain seem to threaten them more closely. At last late in the day its shores grew more rocky, the river gathered its wandering waters together; and then turning with a sweep southward it passed [added: through] a wide mouth with stony gates at either side piled with shingles at the feet into the Lake. The Long Lake! It was wide enough indeed, so that the far shore was small and far; but it was so long that its northern end pointing away towards the shoulders of the Mountain could only be guessed. At that end the Running river ran into it, and with the Forest stream filled what must once have been a great deep rocky valley, and then passed out againTN4 southward with a doubled stream and ran away hurriedly to the South.
Not far from where the Forest Stream entered it, there was a strange town. It was not built upon the shore, though there were many huts and buildings there; but right out on the surface of the lake protected from the swirl of the moving river by a bay of rock. Great bridgesTN5 ran out into the water and to where on large piles made of the trunks of forest trees was built a busy wooden town. It was not a town of elves, but of men, who still dared to live under the shadow of the mountain, protected by the water, and the bridge that could be doubly defended or destroyed from enemies and even as they thought from dragons.
They grew rich [> still did well] on the trade that came up the great river from the south and was carted past the falls to their town, though the great days when Dale to the North was thriving and [they were rich >] there were <both> wars and a busy town of boats were now but a legend. The rotting piles of <another> greater <town> could be seen along the shores when there was a drought.
But they remembered little about it; though songs were still sung of the King Under the Mountain Thror and his son Thrain of the race of Durin, and of the coming of the Dragon, and the fall of the Lords of Dale.TN6
Added in the top margin and marked for insertion at this point: ‘Some sang that Thror and Thrain would come back one day and gold would flow in rivers through the northern falls, and all that land would be filled with new song and new laughter. But that was a pleasant fable, which did not much affect their daily business, or their occasional quarrel with wood elves over tolls and such like troubles.’
Boats came out from the town and hailed the raftmen: [and soon the >] Ropes were cast, oars were pulled; and soon they were drawn out of the currents of the <merry> Forest River, and towed away to [the piles of >] round the shoulder of rock to lie ashore by the head [> some way from the head] of the chief bridge to Lake-town. Soon men would come up from the South and take some away, and fill others with stuffs they had brought to be taken back up to the wood elves’ home. In the meanwhile the raftmen went to feast at [> in] Lake Town.
They would have been surprised if they could have seen what happened down by the shore as soon as evening fell. A barrel was opened by Bilbo (and the help of pushes and groans from inside); and out crept a most unhappy dwarf. Wet straw was in his draggled beard; he was so sore and stiff [added: so bruised & battered] he could scarcely stumble through the shallow water to lie groaning on the shore. He had a famished [added: & a savage] look like a dog that has been forgotten in a kennel for a week. It was ThorinTN7 – but you can only have told it by his golden chain, and the colour of a now-tattered sky blue hood with a very tarnished silver tassel. It was some time before he would even be polite to the hobbit.
‘Well are you alive or are you dead?’ said Bilbo quite crossly at last. Perhaps he had rather forgotten that he had had at least one good meal more than the dwarves, and also the use of his legs and arms not to speak of air [> a greater allowance of air]. ‘Are you still in prison or are you free? Have you arrived at last clear of the wood, and reached the Lake or not?’
‘If you want food, and if you want to go on with this silly adventure, which is after all yours first not mine, you had better rub your legs and arms and try and help get the others out, [before >] while there is a chance!’
Thorin of course saw the sense of [> in] this. And after a few more groans he got up and helped the hobbit. They had a time of it in the gathering dark and the cold water finding which were the right barrels. Knocking outside and calling only discovered about six. These they got out. Some had to be helped or carried ashore and laid down helpless;TN8 they were soaked as well as cramped and [?starved >] bruised and hungry [Dori and Nori were not much use yet nor Ori. >] Dwalin and Balin were two of the most unhappy. They were no use just yet. Bifur and Bofur were less knocked about, and drier but they couldn’t be got to help. Fili and Kili, however, – who were young (for dwarves) – and had been packed more neatly with plenty of straw into smaller casks, came out more or less smiling, with only a bruise or two, and a stiffness that soon wore off.
‘I hope I never smell the smell of [butter >] apples again’, said Fili. ‘My tub was full of it. To smell apples [when you can’t move and can’t >] everlastingly when you can scarcely move, and are getting cold and sick with hunger is exasperating. [But >] I could eat anything in the wide world now, for hours on end – but not an apple’. With the help of F. & K. they discovered the others at last and got them out. Poor fat Bombur was asleep or senseless. Dori Nori Ori Oin & Gloin were waterlogged, [added: only] half alive it seemed, and had to be carried and laid helpless on the shore.
‘Well here we are!’ said Thorin ‘And I suppose we ought to thank Mr Baggins. I am sure he expects it. But I wish he could have arranged a more comfortable journey. Still all very much at your service. No doubt we shall feel properly grateful when we are fed and recovered. In the meanwhile, what next?’
‘I suggest lake-town’ said Bilbo. ‘what else is there?’
So Thorin and Fili and Kili went [> left the others and went] with Bilbo to [the] chief Bridge.
There were guards there, but they were [added: not] keeping careful watch; it was so long since there had been much need. Otherwise they would have heard [added: something of] the disembarking of the dwarves. Now their surprise was enormous when Thorin OakenshieldTN9 stepped into the doorway of their hut.
‘Who are you?’ they shouted leaping to their feet.
‘Thorin son of Thrain son of Thror king under the Mountain.’ said he in a great deep voice and he looked it in spite of his torn and bedraggled dress. The gold and silver gleamed on his neck and waist; his eyes were dark and deep. ‘I have come back. I wish to see the master of your town!’
Then there was a tremendous excitement. Some of the more foolish ran out as if they expected to see the mountain turned golden in the night and all the waters of the lake go yellow right away.
The captain of the guard came forward.TN10 ‘And who are these?’ said he pointing to Fili and Kili and Bilbo. ‘The sons of my father’s daughter’s son’TN11 said Thorin ‘Fili and Kili of the race of Durin, and Mr Baggins our guide from the lands of the West.’
‘Lay down your arms’ said the captain.
‘We have none’ said Thorin; and that was true enough. Their knives and the great sword Orcrist Goblin-slasher had been taken from them by the wood-elves. Bilbo had his knife, but he said nothing about that. ‘What need of weapons we are not enemies, who return at last as spake of old. What use against so many. Take us to your master.’
‘He is at feast’ said the captain.
‘All the more reason for taking us to him’ said Fili who was getting impatient at these solemnities. ‘We are wayworn and famished after our long road, and have sick comrades. Now make haste and let’s have no more words, or your master may have something to say to you.’
‘Follow me’ said the captain, and with six men about them he led the way over the bridges to the market place of the town: a wide circle of still [> gentle] water surrounded with the greater homes, and great wide wooden quays with many steps and ladders going down to the surface of the lake.
From one great house there were many lights and a sound of voices. They passed the door and stood blinking in the light looking at long tables filled with folk.
‘I am Thorin son of Thrain son of Thror King Under the Mountain. I return’ said Thorin in a loud voice from the door before the captain could say anything.
All leapt to their feet. The Master of the Town – the mayor perhaps we should call him – sprang from his great chair. But none knew greater surprise than the raftsmen of the elves, sitting at the end of the hall. They recognized Thorin and the two dwarves as the king’s prisoners!
<Pressing> forward to the master they cried – ‘these are prisoners of our king, that have escaped. Wandering and vagabond dwarves that could not give any good account of themselves; sneaking through the woods and pursuing our people.’
‘Is this true?’ asked the master.
‘It is true that we were wrongfully waylaid by the Elf-king and emprisoned without cause, as we journeyed back to our own land’ said Thorin. ‘But locks nor bars may hinder the home coming spoken of old. Nor is this town in the wood-elves’ realm. I speak to the Master of the Town of the men of the Lake, not to the boatmen of the king.’TN12
Then the master paused and looked from one to the other. The Woodelves’ king was grown powerful in those parts; he did not wish for any enmity with him, and he did not trouble much about old songs; but about trade and tolls, cargoes and gold.
Others were of different mind, however, and soon [the dealing of his >] the matter was settled without him. The news spread from the doors of the hall like fire through all the town.TN13 People were shouting within the hall and outside it. The quays were thronged with hurrying feet. Some began snatches of the old songs concerning the return of the King under the Mountain – that it was Thror’s grandson not Thror himself bothered them not at all. Others took up this song and soon it rolled loud and high over the lake.
The King beneath the Mountain[s],
The King of carven stone,
The Lord of silver Fountain[s]
Shall come into his own!
His crown shall be upholden,
His harp shall be restrung,
His halls shall echo golden
To songs of yore re-sung.
The woods shall wave on mountains
The grass beneath the Sun;
His wealth shall flow like [> in] fountains,
[and the >] The rivers golden run.
The rivers run in gladness
The lakes shall shine and burn,
[And >] All sorrow fail and sadness
When the Mountain-kings return.TN14
So they sang – or very like that, only there was a great deal more of it; and much shouting as well as music of harps and fiddles mixed up with it. Such excitement had not been in the town in the memory of the oldest grandfather.
The woodelves themselves began to wonder greatly and even to be afraid. As yet they did not know how Thorin had escaped, and they were begin[ning] to think their King had made a grievous mistake. As for the master he saw there was nothing else for it, but to obey the general clamour for the moment at any rate.
In fact he gave up his chair to Thorin, and Fili and Kili and even Bilbo – whose presence [had >] was quite unexplained.TN15
Very soon the ten other dwarves were bought into the town [with >] amidst scenes of astonishing enthusiasm; and doctored, and fed, and housed, and pampered in the most delightful and satisfactory fashion.
A large house was given up to Thorin & his company; boats and rowers were put at their service, and crowds sat outside and sang songs all day. Some of them were quite new, and spoke confidently of the sudden death of the dragon, and cargoes of rich presents coming down the river to the Lake town. These were inspired largely by the Master & didn’t particularly please the dwarves. But in the meantime they got fit & strong again. Indeed in a week they were more than recovered, fitted out in fine cloth, with well combed beards and proud steps. Thorin looked all he claimed to be and more;TN16 and as he had said the dwarves’ good-feeling towards the little hobbit grew stronger every day. They made a great fuss of him, which was just as well, for he had a shocking cold, and sneezed for <three> days, and couldn’t go out; and his speeches at banquets were limited to ‘Thag you very buch’.
In the meanwhile the wood-elves [were >] had gone back up the river with their cargoes; and there was not a little excitement in the king’s palace. I never heard what happened to the guard and the butler. Nothing was ever said about keys or barrels while the dwarves stayed in Lake-Town, and B. was careful never to become invisible. Still I daresay more was guessed than was known. In any case the king [sent out >] knew the dwarfs [> dwarves]TN17 errand now or thought he did; and he thought also
‘Very well, we’ll see – no treasure will come back through Mirkwood without my having to say in the matter[, and I >]’. He at any rate did not believe in dwarves <illegible> dragons like Smaug, and he strongly suspected (being a wise elf) burglary or something like itTN18 – which shows he was a wise-elf and wiser than the men of the Lake; and yet not as right as we may see. He sent out his spies [as far to >] about the shores of the lake end as far North towards the Mountain as they would go; and waited.
[Thorin > After >] At the end of a week Thorin began to think of departure. While the enthusiasm still lasted in the town was the time to get help. It would not do to let everything cool down with delay. So he spoke to the Master and his councillors, and [spoke >] said that soon he and his company must go on towards the Mountain.
Then for the first time the Master was surprised and a little frightened. I don’t think he ever thought that the dwarves would dare to go [> approach Smaug];TN19 he probably thought they were frauds who would sooner or later be discovered, and turned out. He was wrong. Thorin was really the grandson of the k. under the mountain; and there is no knowing what a dwarf will not dare and do for revenge or the recovery of his own.
At any rate the Master was not sorry to let them go. They were expensive, and their arrival had turned things into a long holiday; business was at a standstill. ‘Let him go and bother Smaug, and see how he likes it’ he thought. ‘Certainly O Thorin Thrain’s son Thror’s son’ was what he said. ‘You must claim your own. The hour is at hand; what help we can offer shall be given’.
So one day – although autumn was already getting on, and winds were cold, and leaves were turn[ing] – three boats left Lake Town, laden with rowers, dwarves, Mr Baggins, and many provisions; horses went round by circuitous paths to meet them at their appointed landing place. The master and his counsellors bade the solemn farewell from the [steps of the >] great steps that went down to the lake. People sang on the quays and out of windows. The white oars dipped and splashed and off they went north up the Lake.
As before, the text continues with no indication in the manuscript of the point where the chapter break was later inserted – in this case, just before the last paragraph on manuscript page 136 (Marq. 1/1/12:1). As was so often the case, the last line was re-written and augmented in the typescript in order to make a more effective break: ‘. . . and off they went north up the lake on the last stage of their long journey. The only person thoroughly unhappy was Bilbo’ (First Typescript, typescript page 109; 1/1/60:7).
TEXT NOTES
1 In the margin alongside this line is the single word ‘left?’ Since this is written in pencil, we know that it comes from the period when Tolkien was preparing the First Typescript, which takes up the proposed correction (as does every subsequent text). Note that the sudden bend to the right of the Forest River, suggesting that it was rounding some obstacle, dates all the way back to the very first Hobbit map incorporated into the Pryftan Fragment (Frontispiece) and can also be seen, though less prominently, in one of the five maps that accompanied Tolkien’s original turnover of the completed text to Allen & Unwin in October 1936 (Plate I [bottom]). On the other hand, this feature has almost disappeared from another map in the same set, the precursor of the Wilderland map (given on Plate II [top]), indicating that Tolkien remained unsure about the course of the Forest River until relatively late in the process.
The final Wilderland map published in the original and all subsequent editions (DAA.[399]) does not entirely agree with the accompanying published text (DAA.241). The easternmost extension of the hills in which the Elvenking’s halls are located does indeed appear on the left bank of the river† but the river does not ‘round’ any ‘steep shoulder of land’ but instead curves gradually to the right as it flows through marshlands.
† These same heights can also be seen rising on the left (north) bank of the Forest River in both of Tolkien’s paintings illustrating Bilbo’s arrival by barrel at the huts of the raft-elves (H-S#122 & 124). The unused coloured pencil sketch (Plate VIII [top]) clearly shows hills on the north bank of the river, while the published version (DAA plate two [bottom]) shows both these hills and the lack of any corresponding heights on the right (south) bank of the river.
2 This mention of these unseen landmarks in the midst of this vivid descriptive passage is remarkable, since our point-of-view character cannot see them and they have not yet appeared on any of the sketch-maps. These low hills or badlands show up most clearly on the early version of the Wilderland map that accompanied the submission of the completed book to Allen & Unwin in October 1936 (Plate I [bottom]), where they do indeed extend north-east from the Lonely Mountain instead of the more westerly orientation they are given in the final Wilderland map. They can also be seen depicted pictorially in the careful sketch of the Lonely Mountain that with the final map of the Long Lake made up another of the five maps accompanying the October 1936 submission (Plate II [top]).
3 In the next stage of the text, the first typescript, this section was greatly expanded to bring in a reference to the Dragon and a reminder of the missing wizard’s mysterious business. Major changes are marked in italics to highlight the degree of expansion.
. . . The talk was all of the trade that came and went on the waterways and the growth of the traffic on the river, as the roads out of the East towards Mirkwood vanished or fell into disuse; and of the bickering of the lakemen and the wood-elves about the upkeep of the forest-river and the care of the banks. Those lands had changed much since the days when dwarves dwelt in the Mountain, days which nearly everybody [> most people now] remembered only as a very shadowy tradition. They had changed even in recent years, and since the last news Gandalf had had of them. Great floods and rains had swollen the waters that flowed East; and there had been an earthquake or two (which some were inclined to attribute to the dragon – alluding to him chiefly with a curse and an ominous nod in the direction of the Mountain). The marshes and bogs had spread wider and wider on either side. Paths had vanished, and many a rider and wanderer too, if they had tried to find the lost ways across. The elf-road through the wood which the dwarves had followed on the advice of Beorn now came to a doubtful and little used end at the eastern edge of the forest; only the river offered any longer a safe way from the skirts of Mirkwood in the North to the mountain-shadowed plains beyond, and the river was guarded by the wood-elves’ king.
So you see Bilbo had come in the end by the only road that was any good. It might have been some comfort to Mr. Baggins shivering on the barrels, if he had known that news of this had reached Gandalf far away and given him great anxiety, and that he was in fact finishing his other business (which does not come into this tale) and getting ready to come in search of Thorin’s company. But he [> Bilbo] did not know it.
All he knew was that the river seemed to go on and on and on for ever, and he was hungry, and had a nasty cold in the nose, and did not like the way the Mountain seemed to frown at him and threaten him as they [> it] drew ever nearer . . .
—First Typescript, typescript pages 103–4 (Marq. 1/1/60:1–2).
For more on the theme of roads falling into disuse and the increasing difficulty of travel, see also the 1960 Hobbit (p. 818); for a lucid description of the theme of depopulation as settled lands turn into desolate wastelands, see Henry Gee ‘The Gates of Minas Tirith’, Chapter 14 in The Science of Middle-earth (page 151).
4 Added: ‘over <illegible> waterfalls’. The single illegible word is not high (the reading of the typescript and published book) but probably noisy.
The reference to the drowned valley that has now become a great lake sounds like an echo of the many drowned lands in Tolkien’s earlier tales, but here no actual tale seems to underlie the reference; like the ruins of the earlier, greater town and the reference to ‘wars’ (plural), it seems to be a deliberate layering of an untold prehistory for artistic effect.
5 Note the plural ‘bridges’ here and elsewhere in this chapter; back in Plot Notes B the reference had been to a single bridge; see p. 364. The plural here persists through both typescripts and was only changed to ‘A great bridge’ in the page proofs (Marq. 1/2/2: page 198). At the same time, the ‘chief bridge’ on p. 436 became the ‘great bridge’ (1/2/2: page 199) and the plural was removed from the description of the bridges being thrown down during Smaug’s attack (cf. p. 548; page proof 1/2/2: page 253). The decision for Lake Town to have only one great bridge seems to have been determined through the two illustrations Tolkien drew of the scene (Plate VIII [bottom] and DAA.244/H-S#127), apparently created over the Christmas 1936 vacation, the second of which he submitted to Allen & Unwin on January 4th, 1937 (the day after his 45th birthday). If so, the changes in page proof (made between February 24th and March 10th 1937) would have been made to bring the text into agreement with the illustration. The colour sketch ‘Death of Smaug’ (see Part Two) also shows the easternmost end of the fallen single Great Bridge and so definitely postdates the submission of the completed manuscript to Allen & Unwin in Oct. 1936; it was probably created between May and August 1937 along with the other colour pieces Tolkien made for the book at Houghton Mifflin’s request (JRRT/Allen & Unwin correspondence, A&U Archives).
6 This marks the first appearance of the names Thror and Thrain. See part iii of the commentary, starting on p. 455.
7 This is the first appearance of the name Thorin used in the text as the chief dwarf’s name, although the change had been anticipated as far back as Plot Notes A (see p. 293). Tolkien’s rejection of ‘Gandalf’ as the name of the chief dwarf no doubt came because, on reflection, it offended Tolkien’s sense of decorum to have a dwarf named ‘elf’ (Gand-alf: ‘wand-elf’). For more on the name ‘Thorin’, see p. 455.
8 Added: where they sat and muttered or moaned.
9 This is the first appearance of Thorin’s cognomen Oakenshield since Plot Notes A, where remarkably enough it had already been linked with Thorin’s name. Like ‘Thorin’ itself, it comes from the list of dwarf-names that appear in both the Völuspá (in the Elder Edda) and the Gylfaginning (‘The Deluding of Gylfi’, in the Prose Edda) as Eikinskjaldi. However, there it is simply another dwarf-name and has no linkage to ‘Thorin’; the two actually occur in different stanzas of Völuspá (Thorin in the third line of stanza 12 and Eikinskialdi in the last line of stanza 13, respectively). See Appendix III.
10 It is possible that this ‘captain of the guard’ is Bard, who later plays such a major role in the dwarves’ fortunes; see p. 553.
11 This is the first mention that Fili and Kili are Thorin’s close kin. Note that they are originally his great-nephews, his sister’s grandsons, whereas in the final book their relationship is one generation closer (his sister’s sons). The original relationship was still in place when the First Typescript was originally typed (cf. typescript page 106; 1/1/60:4) but had already been changed before the Second Typescript (1/1/41:5) was created.
The uncle/nephew bond was extremely important in heroic medieval literature – cf. Roland and Charlemagne, Beowulf and Hygelac, Gawain and Arthur. This motif is more or less entirely absent from the Silmarillion tradition and only enters the legendarium at this point, but later became of great importance: cf. Éomer and Théoden, not to mention Frodo and Bilbo.
12 This was originally followed by a cancelled line that would have marked the beginning of a new paragraph: ‘Yet at least the King of the woods gave us food, and sh[elter]’ – a none-too-subtle hint on the hungry dwarf’s part, in keeping with his earlier verbal sparring before the Elvenking (p. 315).
13 The simile may be significant, since within a few chapters the town will indeed burn, as had already been foreseen in Plot Notes B (although in the Plot Notes the dragon did not succeed in burning it to the waterline; see p. 364). Note also the line, ominous in retrospect, in the town-folk’s song: ‘The lakes shall shine and burn’.
14 An earlier draft of this poem can be found on the back of the next manuscript page; after this version was superseded Tolkien struck it through with a cancellation line, then turned the paper upside down and over to use the reversed back as a fresh sheet (manuscript page 135; Marq. 1/1/11:8). A large Roman numeral II appears at the top of this page, drawn directly over the first three cancelled lines:
II
When the king beneath the Mountains comes.
the Lord beneath the Hills
the lord of golden Fountains
This is followed by the draft:
The king beneath the Mountains
The king of carven stone
the lord of golden [> silver] Fortress
shall come into his own
[The >] His crown shall be uplifted
his harp shall be restrung
His halls shall be relighted
his praises shall be resung
His wealth shall flow like water
his gifts like light of sun
The river run in gladness
And the grass <stands> under sun [> beneath the sun]
He <sic> crown shall be upholden
his harp shall be restrung
his halls shall echo golden
to song[s] of yore resung.
[The >] His wealth shall flow like fountains
[The > like >] The rivers golden run
the grass [< woods] shall wave [> wax] on mountain
and the grass beneath the sun
The rivers run in gladness
[The > and men >]
the lake be filled with gems [> shall shine and burn]
[And men know no more >]
And sorrow fail and sadness
When the Mountain-kings return
The second draft fair copy incorporated into the main manuscript (see pp. 439–40) required only a very few minor changes (mostly typographical), made between the manuscript and typescript stages, to achieve the text of the published version (DAA.251). The most significant change comes in the final line, which shifts from the plural (‘When the Mountain-kings return’) to the singular (‘At the mountain king’s return’): compare Ms. 1/1/11:6 (manuscript page 134) with Ts. 1/1/60:5–6 (First Typescript, pages 107–8). One change marked in the manuscript which was not taken up into the typescript version is a pencilled change from ‘kings’ to ‘lords’ in the last line.
Curiously enough, between the first and second drafts the word lake (i.e., the Long Lake) in the third line from the end was changed to lakes (plural), a reading which has persisted ever since. While certainly justifiable from the point of view of poetic license, given the dominance of plurals in the closing stanza (it is Mountain-kings, not merely the King under the Mountain, whose return they praise and prophesy), it is nevertheless striking for Tolkien to move from the precise and accurate to the general and ‘poetic’.
15 This passage was revised and expanded to read
In fact he gave up his chair to Thorin, and Fili and Kili sat beside him and even Bilbo was given a place at the high table – no explanation of where he came in (although no songs had alluded to him even in the obscurest way) [beyond >] was asked for in the general bustle.
The cancelled ‘beyond’ suggests some such cover story as that which Thorin had offered the captain of the guard – that is, ‘Mr. Baggins, our guide from the lands of the West’.
According to the much later chronology of The Lord of the Rings, this welcoming feast took place on September 22nd, but clearly no such specific time-scheme was present in the original conception; see ‘Timeline and Chronology’ in The 1960 Hobbit, especially pp. 823 & 832.
16 This sentence was recast between the manuscript and First Typescript to the reading in the published book: ‘Thorin looked and walked as if his kingdom was already regained and Smaug chopped up into little pieces.’
17 The apostrophe is missing in the original, but presumably Tolkien shifted from the singular (dwarf’s) – that is, Thorin’s – to the plural (dwarves’) here rather than offered up an alternative plural (dwarfs’).
18 The illegible word here might begin with a k and end in an -ly (or just possibly -ing), but it is certainly not ‘killing’. The parenthetical ‘(being a wise elf)’ was cancelled, presumably before the section following the dash was added.
19 This passage was cancelled and replaced by the following (written in the margin and marked for insertion at this point): ‘I think that like the king he never believed . . .’
The vivid description of Lake Town that dominates this chapter is another example of Tolkien drawing upon his knowledge of history and prehistory as inspiration for his creative work. As Christina Scull was the first to point out,1 archeological fact often underlies Tolkien’s fiction – an aspect of his writing that is not surprising, given his avowed ‘passion . . . for heroic legend on the brink of fairy-tale and history, of which there is far too little in the world’ (letter to Waldman, 1951; Letters p. 144), as well as his emphatic preference for ‘history, true or feigned’ as a mode of writing (LotR.11, Foreword to the Second Edition). An interest in legends on the edge of recorded history naturally implies knowledge and interest in both sides of that borderland: a solid familiarity with early recorded history and a matching interest in unrecorded prehistory as well. By the same token, anyone who like Tolkien sets out to write ‘feigned history’ must be well acquainted with the real thing if his pseudohistory is to be plausible and persuasive. A good example can be found in the frame story for The Book of Lost Tales, the Eriol legend, set in the murky period just before the Jutes (closely followed by the Angles and the Saxons) invaded Britain and turned it into England, a period Tolkien revealed extensive historical knowledge of in his posthumously published lectures on the Freswæl, or ‘Fight at Finnesburg’ (Finn and Hengest: The Fragment and the Episode, ed. Alan Bliss [1982]). Furthermore, Tolkien’s outline for the unwritten chapters of The Lost Road (HME V.77–8; see also JRRT to Christopher Bretherton, 16th July 1964, Letters p. 347) shows his intention to have that work, starting in the familiar present and ending in the wholly invented mythic world of lost Númenor, bridge the gap between the two through episodes set first in poorly-recorded historical periods (ninth-century England during the collapse of the Angles’ kingdoms under Viking assault,2 Lombardic Italy of the mid-sixth century AD), then in eras known only through legends (Norse lands in the time of Scyld Sceafing, Ireland during the legendary days of the Tuatha dé Danaan), then periods known from archeology but for which all legends and stories have been lost (the Ice Age, the era of the Paleolithic cave-paintings), and finally beyond, into his own imagined prehistory (Beleriand and finally Númenor).3
A closer look at the evidence shows that Tolkien was very well versed indeed in prehistory. We have already seen how he modeled Medwed’s hall on modern archeological reconstructions of a Norse mead-hall (see commentary following Chapter VII, p. 261). Similarly, in his 1932 Father Christmas letter he drew on Paleolithic cave art, such as that found in the caves of Altamira, Spain [17,000 BC] (the similar caves at Lascaux and Chauvet not yet having been discovered). So too with Lake Town, which is closely modeled on the Neolithic lake-dwellings or pfahlbauten (‘pile structures’) first discovered in Switzerland in 1854 on the shores of Lake Zurich.4 That Tolkien is drawing directly on accounts of the Swiss discovery, probably the classic Die Keltische Pfahlbauten in den Schweizerseen by Dr. Ferdinand Keller (literally ‘The Celtic Pile-structures in the Swiss Lakes’ [1854], translated into English by John Edward Lee as The Lake Dwellings of Switzerland and Other Parts of Europe [1866]), is shown by his reference (p. 436) to how
The rotting piles of another greater town could be seen along the shores when there was a drought.
This reference to another, greater, Lake Town (what archeologists excavating the site would no doubt call Lake Town I, to distinguish it from the later Lake Town II visited by Thorin & Company and the still later Lake Town III that replaces it described in the final chapter) is unusual, because we learn nothing else about it; its destruction seems to belong to the distant past, long before Smaug’s advent. It is striking, therefore, that the first prehistorical lake-dwelling was discovered because a dry winter lowered the water level of Lake Zurich, exposing the ancient wooden piles that had once supported the settlement; Tolkien’s description seems to be a direct echo of the archeological discovery of some seventy-five years before. Scull (‘The Influence of Archaeology and History’ page 41; see Note 1), Anderson (DAA.245), and Artist & Illustrator (H-S#125) all reproduce nineteenth- and early twentieth-century images from various archeological texts showing artists’ conceptions of what such Neolithic villages might have looked like; another, found in Bryony & John Coles’ Sweet Track to Glastonbury, is an early twentieth-century depiction of the lake-dwellings closest to Tolkien’s home, those at Meare and Glastonbury (some seventy-five miles from Oxford).5 None of these images corresponds exactly to Lake Town as Tolkien describes it in his text or depicts it in his three drawings of the site (‘Esgaroth’ [Plate VIII, bottom], ‘Lake Town’ [DAA.244/H-S#127], and ‘Death of Smaug’ [in plate section two], all of which clearly date from the time after he had already submitted the book to Allen & Unwin for publication (the first two to December 1936 and the third probably between May and August 1937), since they show only one bridge between Lake Town and the shore (see Text Note 5 above). The closest is that appearing in Robert Munro’s Les Stations Lacustres d’Europe aux Ages de la Pierre et du Bronze [1908] (‘The Lake Stations of Europe during the Stone and Bronze Ages’), said to have been based on an earlier drawing by A. de Mortillet6 (DAA.245 [top], H-S#125). Given the exactness with which Tolkien based some of his drawings upon pre-existing sources – e.g., ‘The Trolls’ on Jennie Harbour’s ‘Hansel and Grethel Sat Down by the Fire’, ‘Bilbo Woke Up with the Early Sun in His Eyes’ on Archibald Thorburn’s chromolithograph of a golden eagle, and ‘Firelight in Beorn’s house’, the original conception of Medwed’s dwelling, on the picture of Hrolf Kraki’s meadhall that had appeared in his friend E. V. Gordon’s An Introduction to Old Norse – it is probable that he had a more direct source for his illustrations of Lake Town that has not yet been discovered.
As might be expected, the sequence of three drawings shows some variation as Tolkien refined his image of the city over the water: the hut beside the head of the bridge in the text wherein Bilbo, Thorin, Fili, and Kili meet the story’s first humans to appear on-stage (see part ii of the commentary, below) becomes a gate-house attached to the bridge in the first drawing (‘Esgaroth’), through which one must pass to enter the city. By contrast, a gap intrudes between gate-house and bridge in the second drawing (‘Lake Town’), offering no explanation of how visitors climb up onto the elevated bridge. This same image also includes an archway in Lake Town’s southern side, allowing access to the water-market at the city’s center described in the text. The most dramatic of the three, ‘Death of Smaug’, depicts the burning city, its bridge already cast down (only the easternmost link can be seen, on the left). It also substitutes long rowhouses for the grander individual buildings shown along Lake Town’s western edge in the two earlier drawings. Similar but smaller buildings had appeared in Lake Town’s southeast quadrant in the second drawing (‘Lake Town’) and, less elongated, in the first (‘Esgaroth’), but we can tell we are not simply looking at the lake-dwelling from a different angle by the position of the moon (just past new) in the upper left (i.e., the northwest) and a dim glimpse of the Lonely Mountain to the north, just left of the center of the picture.7
While Tolkien’s debt to Keller et al. seems clear, it is interesting to note that by the time Tolkien was writing The Hobbit Keller’s theory was under attack by a new generation of archeologists. It is now generally believed that the majority of ‘lake-dwellings’ were not actually out on platforms above the water but built on marshy ground along the shore or on low islands or peninsulas surrounded by marsh or bog; they were more wetland settlements than lake-dwellings per se, and this is definitely the case with the ‘lake villages’ of Somerset at Glastonbury [discovered 1892] and Meare [1895], the latter of which was actually still under excavation throughout the time Tolkien was working on The Hobbit.8 However, we will badly misunderstand any influence on Tolkien from contemporary science and scholarship unless we look not at modern ideas and interpretations regarding a given field but instead at the scholarly consensus of Tolkien’s day. For example, Keller asserted that many of the lake-dwellings seem to have been destroyed by fire (cf. Keller pages 8, 28–9, 33, 43–5, &c.), laying far less stress on the fact that, as modern archeology notes, many were simply abandoned. Tolkien, unrestrained by the demands of historical probability, took the lake-dwellings discovered by Keller, de Mortillet, Bulleid, and their peers and incorporated them into his fiction in their classic, raised-platform-on-pilings-above-the-water form. In the process, he provided a mythic explanation of why Neolithic folk sought such protection and undertook the enormous labor required to construct a lake-town: in a world inhabited by predatory dragons, it would be worth almost any pains to carve out a home in an environment dragons would instinctively avoid. Similarly, he picked up on the theme of destruction by fire, which has inspired many a speculation about warfare and pillage among the historians,9 and gave it an epic interpretation of destruction by dragon in a holocaust worthy of the Beowulf-poet. Just as significantly, he departed from the archeological record when it suited his purpose – for example, while many lake-dwellings were destroyed and rebuilt several times, they were usually rebuilt on the same spot (the lake-dwelling at Robenhausen on Lake Pfaffikon just north of Lake Zurich, discovered in 1858 and described by Keller on pages 37–58, fits Tolkien’s pattern particularly well, since it was built three times, burned down twice, and finally abandoned). By contrast, our Lake Town in The Hobbit shifted its site each time it was rebuilt, and the ultimate fate of its third incarnation is unknown, being beyond the scope of our story.
It is wholly remarkable that, in current editions of The Hobbit, incorporating Tolkien’s second and third edition changes (as well as others made subsequently since Tolkien’s lifetime), we come two-thirds of the way through the story before we meet our first humans, these being the guardsmen in the hut before Lake-town – and that when at last we do so little is made of the fact. The event is somewhat obscured by the description of Gandalf the wizard as ‘an old man’ (DAA.32), or Beorn as ‘a huge man’ (DAA.167), or of the raft-elves as ‘raftmen’ (cf. DAA.240, 241, 250), whereas in fact none of these are truly human (Gandalf the Grey being one of the Istari, an incarnate Maia or angel, Beorn a werebear, and the ‘raftmen’ more properly ‘raft-elves’ – cf. ‘the raftsmen of the elves’, p. 439). The mention of the Men of Dale in Chapter I and the Men of the Long Lake on Thror’s Map, and of the ‘woodmen’ in Chapter VI (see especially DAA.147–8) and on the Wilderland Map, has introduced the idea of off-stage humans from very early on, although the woodmen do not actually appear in the story until very near the end, at the Yuletide celebrations in Beorn’s halls on Bilbo’s return journey (see p. 682 and DAA.353). In addition, modern readers of the book as often as not come to it armed with knowledge from a prior reading of The Lord of the Rings and thus know that Bilbo would have passed through Bree on his journey east and encountered humans there, a fact Tolkien made explicit in the timeline he drew up to accompany the 1960 Hobbit (see ‘Timelines & Itinerary’, pp. 816, 818, 828, 834). Matters would of course have been quite different for a reader of the original manuscript, who would have no reason not to accept the text’s description of Bladorthin as ‘a little old man’ quite literally and at any rate knew Bilbo and company had passed through lands inhabited by ‘men or hobbits, or elves’ in the early stages of their journey (contrast p. 90 with DAA.65). Hence the moment is momentous only because changes made in the course of later editions removed previous encounters with humans, and would itself have been superseded had Tolkien’s proposed 1960 revision been carried to completion and seen print.
The woodmen, from what little we are told of them, are clearly very much in the vein of William Morris’s Goths as described in works such as The House of the Wolfings [1888] and The Roots of the Mountains [1889], which pit Germanic tribesmen against expansionist Romans and marauding Huns, respectively. Compare Tolkien’s account of the villages of the ‘brave woodmen and their wives and children’
In spite of the dangers of this far land bold men had lately been pushing up into it from the south again, and cutting down trees, and building themselves places to live in among the more pleasant woods farther down in the valleys away from the shadows of the hills, and along the river-shores. There were many of them and they were brave and well-armed (p. 205)
with Morris’s Men of the Mark, who live in
a dwelling of men beside a great wood . . . this great clearing in the woodland was not a matter of haphazard: though the river had driven a road whereby men might fare on either side of its hurrying stream. It was men who had made that isle in the woodland.
. . . [T]hey had no lack of wares of iron and steel, whether they were tools of handicraft or weapons for hunting and for war. It was the men of the Folk, who coming adown by the river-side had made that clearing . . . they came adown the river . . . till they had a mind to abide; and there as it fell they stayed their travel, and . . . fought with the wood and its wild things, that they might make to themselves a dwelling-place on the face of the earth.
So they cut down the trees, and burned their stumps that the grass might grow sweet for their kine and sheep and horse; and they diked the river where need was all through the plain . . . and they made them boats to ferry them over . . . and [the river] became their friend, and they . . . called it . . . the Mirkwood-water . . .
In such wise that Folk had made an island amidst of the Mirkwood, and established a home there, and upheld it with manifold toil too long to tell of.
—The House of the Wolfings, pages 1–2.10
In a sense, if Morris is describing the moment when prehistory (the Germanic-Roman strife from the point of view of the tribesmen resisting Roman encroachment) meets history (this war being known to history only from the other point of view through Roman writers such as Caesar and Tacitus), then Tolkien is offering a prequel to that transitional moment, when mythic monsters rather than ambitious empires were the greatest threat to existence; not the time when the Men of the Mark were defending their land against invaders but the earlier time when they themselves were first coming into that part of the world. Perhaps significantly, the Wolfings keep hanging from the roof of their hall a great work of art, the origins of which they have forgotten:
a wondrous lamp fashioned of glass . . . clear green like an emerald, and all done with figures and knots in gold . . . and a warrior slaying a dragon . . . an ancient and holy thing
—The House of the Wolfings, page 6.
This is probably a passing reference by Morris to the story of Sigurd dragon-slayer, which he had translated only the previous year, but within Tolkien’s larger reconstructed prehistory it could just as easily be seen as preserving, like the Franks Casket, a fragment of the story of Bilbo dragon-slayer long after the details of what actually happened have been forgotten. Similarly, Tolkien may have decided to invent a different history for one of the Wolfings’ neighbouring tribes, the Bearings, which is simply the modern English cognate of Beornings, the name bestowed upon those woodmen who later choose to take Beorn/Medwed as their leader; instead of merely being their totem-animal it would thus have been the name of the tribe’s original leader, Beorn/Bear, who himself had long since been forgotten.
That the woodmen were an archetype that strongly appealed to Tolkien can be shown by the presence in the Silmarillion tradition during the First Age of the People of Haleth, a woodland folk who later become the Men of Brethil, the Second Kindred of the Elf-friends. Also very similar are the Northmen of eastern Rhovanion (Wilderland), whom Tolkien created to serve as the common ancestors for the Lake-men, the Men of Dale, the woodmen, and the Men of Rohan; they figure in the history of Gondor told in Appendix A of The Lord of the Rings and in ‘Cirion and Eorl’, especially part (i) ‘The Northmen and the Wainriders’ in Unfinished Tales (cf. UT.289–90 & 295, 297–8). Significantly, the woodmen’s culture and way of life seem to have changed little in the seventeen hundred years that separates them from the Northmen of Vidugavia’s day (a name which, as Christopher Tolkien points out, is itself Gothic for ‘Wood-dweller’, Widu-gauja; UT.311).
It is rather surprising, then, that the first humans we see close-up in the story are quite different. Rather than an Iron Age culture like the men on the other side of the forest (whom Tolkien makes their kinsmen when he comes to write Appendix A of LotR), the Men of Lake Town are urban, even urbane, with a culture right out of the High Middle Ages.11 Lake Town is a free city (at least until the re-establishment of the Kingdom of Dale at the end of the story), belonging to no nation and owing allegiance to no king. It is also Tolkien’s only oligarchy, ruled by a Master of the Town ‘elected from the old and wise’ (p. 551; cf. p. 639 & DAA.309) rather than a noble lord.12 We are told about the Lake-men’s concern with commerce, and how the Master has ‘a good head for business – especially his own business’ (p. 550) and a mind devoted to ‘trade and tolls, cargoes and gold’ (p. 439; the typescript rather tartly adds ‘to which habit he owed his position’ – cf. DAA.250). This, along with two references to ‘the Master and his councillors’ (p. 441), suggests that Lake Town is probably dominated by Merchant Guilds, the guildmasters of whom would choose one among their number to serve as Master of the Town (‘the mayor perhaps we should call him’). As supporting evidence for this, note the Lake-men’s disparagement of ‘old men and money-counters’ (p. 551) after the Master’s poor showing during the attack by Smaug in contrast to Bard’s heroism, and their cry in the typescript of ‘Up the Bowman, and down with Moneybags’ (typescript page 138; 1/1/64:4).
The Master of Lake Town is one of Tolkien’s most interesting minor characters in his own right: an essentially unsympathetic figure who knows so little about his own town’s history that he doubts there ever was a King under the Mountain yet who nonetheless helps our heroes a great deal when they need it most, giving them food and clothing and shelter and sanctuary when they most need it. Indeed, he treats them with a generosity that borders on extravagance, feasting them at banquets and clothing them in fine cloth (‘of their proper colours’ adds the First Typescript), granting them their own large house to stay in and even refusing the wood-elves’ implied request for extradition despite the elvenking’s being his primary trading partner. Yet in all this he is simply cynically going with the tide of public opinion; privately ‘he believed they were frauds who would sooner or later be discovered and turned out’. Later he does them an ill turn just as crucial as the help he had given them earlier, when he defuses his people’s criticism of his behavior during the attack on Lake Town by Smaug by shifting their anger onto the absent dwarves, stirring the lake-men up by reminding them that Thorin & Company must have disturbed the dragon and are thus responsible for his attack on the city (an accusation that is, of course, quite true, although inadvertent on their part) – though typically he frames the accusation in terms of profit and loss, payment and recompense. He thereby helps set in motion the conflict that soon results in the Battle of Five Armies. Yet he is not without skills, as the narrator himself notes (p. 639): it is he who plans out the new Lake Town that rises from the ashes of the old (pp. 552 & 641), and does it so well that the new is fairer than the old (DAA.313).
A wily politician (the only one in Tolkien’s work),13 the Master is sophisticated, subtle, and just a touch corrupt, and his advent on the scene is a harbinger of the ambivalence that is so much a feature of the final chapters, culminating in the tangle of rights and wrongs over ownership of the dragon-hoard. In fact, as Douglas Anderson points out, he is highly reminiscent of that touchstone of bureaucratic greed and double-dealing, the Mayor of Hamelin in Robert Browning’s ‘The Pied Piper of Hamelin’ [1842].14 Like the Mayor of Hamelin (who is similarly unnamed, identified only by title), the Master of Lake Town makes whatever bargain suits his goals at the time, and abandons it without conscience when circumstances change. It is possible that in this wholly unflattering depiction of a town official Tolkien may also owe something to the ‘Town & Gown’ rivalry that had divided Oxford since the thirteenth century; riots between students (‘Gown’) and shopkeepers (‘Town’) persisted until as late as the mid-nineteenth century,15 and the Miller in Farmer Giles of Ham shows that Tolkien was quite willing to use medieval stereotypes when they would yield comic effect. Of course, if we are seeking for applicability it is only fair to point out that while Master of the Town (to give him his full title; cf. p. 439) is a title for what we would now call a mayor, more familiar today in its Middle English form burgomaster (literally ‘town master’), the title Master is also used for the head of several Oxford colleges, including Pembroke, Tolkien’s college at the time he was writing The Hobbit. Anyone who has witnessed academic politics can testify that here is a masterly portrait of a certain type of head of college, or department, or school, pleasant but not trustworthy, accommodating but not sincere.
In the end, though, what is important is not what the Master may or may not symbolize but his role within the story; to first help our heroes and then to greatly complicate their existence. He can fill both roles equally well because the keystone to his personality is that he is motivated entirely by self-interest, and it is a fitting though cruel fate that he winds up starving to death, entirely dependent upon his own too-inadequate resources and abilities, when he seizes for his own what should have been shared among his fellows.
(iii)
Thorin, son of Thrain, son of Thror
With this chapter, the sequence of names familiar from the published books finally makes its appearance, although as we have seen at least one of these changes (Gandalf > Thorin) was mooted as far back as Plot Notes A (see p. 293), written before the Mirkwood chapter had been tackled. And as we shall see, elements of this genealogy remained in flux even while the book was at the printers, with some associated issues not being finally resolved until many years later. Nonetheless, ‘Thorin’ henceforth replaces ‘Gandalf’ as the chief dwarf’s name, and the father and grandfather of Bilbo’s employer finally receive names.16
Like Thorin itself, both the names Thrain and Thror come from the same list of dwarf-names, known as the Dvergatal (‘Dwarf-tally’), that provided the names of all the other dwarves who accompany Bilbo (with the sole exception of Balin); see Appendix III.17 This list appears both in Völuspá (‘The Sayings of the Sybil’ [circa 1000 AD]), the first poem in the collection variously known as the Elder Edda or Poetic Edda, in what is generally considered to be an interpolation to the original poem,18 as well as in Gylfaginning, ‘The Deluding of Gylfi’ in Snorri Sturluson’s Prose Edda [1223].
Jean Young, in her translation of the Prose Edda, glosses ‘Thorin’ as ‘Bold One’ (page 41) but does not explain the meaning of the other two dwarf-names. Ursula Dronke, in her edition of Völuspá, translates the three names as ‘Darer’ (þorinn), ‘Yearner’ (þráinn), and ‘Thrive’ (þrór). Of the other dwarf-names associated with this family, ‘Oakenshield’ (Eikinskjaldi/Eikinskialdi) appears in both lists (indeed, it is repeated twice within the Völuspá itself, in stanza 13 line 8 and again in stanza 16 line 2; see Text Note 9), as does Gandalf (Ganndálf/Gandálfr). Young glosses the latter as ‘Sorcerer-elf’ but Dronke, rather surprisingly, prefers ‘Sprite Elf’; the usual translation is ‘Wand-elf’, although Shippey prefers ‘Staff-elf’ (The Road to Middle-Earth, rev. edition [1992], page 88).
From this point onward, references to Thrain, Thorin’s father, and to Thror, his grandfather and last King under the Mountain, appear frequently, as Thorin asserts his claim to the kingship and treasure and returns to the home of his youth.19 It is now easy to overlook, but important to note, that in the original first edition of the book no mention of either Thror’s or Thrain’s name appeared in the text before the scene at Lake Town (see Note 16), with one notable exception which Tolkien had inserted into the text of Chapter VII when creating the typescript. This sole earlier occurrence is important, because remarkably enough it gave a reversed line of descent – i.e., Thorin son of Thror son of Thrain – and because this error was preserved through both typescripts and into the printer’s proofs. Even more remarkably, when correcting those proofs Tolkien initially decided to change every other occurrence of the names to agree with this exception – that is, he adopted the reverse genealogy and decided to make Thror the father and Thrain the grandfather and Last King throughout – instead of simply altering this single anomalous case to match the rest. And, as Christopher Tolkien points out in his discussion of the two competing genealogies in The Treason of Isengard (HME VII.159–60), it is this reversed line of descent (Thorin–Thror–Thrain) which appears on the map with the moon-runes published in the book.
Since the resulting confusion persisted into the early stages of The Lord of the Rings20 and required Tolkien to make adjustments and additions to The Hobbit as late as 1966 to fully resolve (and which he even then did not perfectly achieve), it seems worthwhile to go into the matter in some detail here to understand how two separate and competing genealogies – what we may call the ‘text tradition’ (father Thrain, grandfather Thror) and the reversed genealogy of the ‘map tradition’ (father Thror, grandfather Thrain) – arose, and how Tolkien ultimately solved the problem that traces of these competing traditions left in the story. The best way to do so seems to be to briefly rehearse the various stages by which Tolkien fixed upon the names of these two characters:
• First Phase: Gandalf is the chief dwarf; his father is unnamed; his grandfather is briefly named ‘Fimbulfambi’, then left unnamed. Bladorthin is the wizard (Pryftan Fragment, Bladorthin Typescript).
• Second Phase: Gandalf is the chief dwarf; his father and grandfather are unnamed; Bladorthin is the wizard; Medwed the werebear. (Second Phase manuscript from middle of Chapter I through what is now Chapter IX). In Plot Notes A, written during a brief pause between the composition of Chapters VII and VIII, Tolkien proposes changing several names (Gandalf > Thorin, Bladorthin > Gandalf, Medwed > Beorn) but does not carry out the changes.
• Second Phase, continued: Chief dwarf’s name changes to Thorin (between Chapters IX and X). His father and grandfather are given the names Thrain and Thror, respectively (Chapter X). These names are used consistently throughout the rest of the Second Phase manuscript (through the scene on Ravenhill in Chapter XV).
• First Typescript: After breaking off the manuscript at the end of the scene with Roäc, Tolkien returns to the beginning of the story and creates the First Typescript. The name changes proposed more than a year before in Plot Notes A are now carried out: Thorin (Oakenshield) is the chief dwarf throughout, Beorn the werebear, and Gandalf the wizard. Thorin’s father and grandfather are unnamed anywhere before Chapter VII, where their names are accidentally reversed when Medwed’s simple remark ‘if it is true that you are respectable dwarves & not friends of Goblins’ (p. 234) is changed to Beorn’s ‘if it is true that you are Thorin (son of Thror, son of Thrain, I believe), and that your companion is respectable . . .’ (1/1/57:5). The typescript transposes two chapters (so that Ms. Chapters XIII and XIV become Ts. Chapters XIV and XIII, the latter represented by a thirteen-page ‘fair copy’) and breaks off shortly before the manuscript it is replacing did (in the middle of Ts. Chapter XIV rather than early in Ms. Chapter XV), but the majority of the references to Thrain and Thror in Chapters X through XV that had appeared in the Second Phase manuscript carry over unchanged, word-for-word, into the typescript (see Note 19), with Thrain the father and Thror the grandfather.
• Third Phase: Tolkien completes the book after a pause of about a year, writing a short, forty-five page manuscript conclusion from the point where the First Typescript broke off. The Thorin–Thrain–Thror line of descent is still in place. The Gem of Girion now becomes the Arkenstone, and in four places (once in a pasteover insertion into the typescript of Chapter XII,21 twice in Chapter XVI, and once in Chapter XVII) it is called ‘the Arkenstone of Thrain’. In one of these points in the new manuscript, and twice in the typescript that ultimately replaced it (the latter dating from autumn 1936), Thrain is explicitly Thorin’s father rather than his grandfather:
‘What of the Arkenstone of Thrain?’ said [Bard] . . .
‘That stone was my father’s, and is mine’ [Thorin] said. ‘. . . how came you by the heirloom of my house’? (Chapter XVII: new Ms. page 21 & 1st Ts. 1/1/67:1)
and again
. . . and now Thorin spoke of the Arkenstone of Thrain, and bade them eagerly to look for it in every corner.
‘For the Arkenstone of my father,’ he said, ‘is worth more than a river of gold in itself, and to me it is beyond price . . .’ (Chapter XVI: 1st Ts. 1/1/66:1)22
Significantly, Tolkien marked this passage in both the First and Second typescripts, changing ‘father’ to ‘fathers’ (i.e., ‘For the Arkenstone of my fathers, 1/1/66:1 & 1/1/47:1), but for some reason he rescinded this correction in the page proofs – see the first set of proofs (1/2/2: page [273]), where Tolkien altered fathers to father, and the second set of proofs (1/2/3: page [273]), where he pencilled in an ‘s’ after father but then erased it. Had he not done so, and had the reference in the following chapter been brought into line, the entire problem need never have arisen.
• Finally, at some point Tolkien makes ‘Thror’s Map’ [Plate I; to distinguish this from the final version appearing in the published book, I will refer to this version as ‘Thror’s Map I’ (TM.1)], based on the original ‘Fimbulfambi’s Map’ that ended ‘The Pryftan Fragment’ [Frontispiece] (or on a now-lost intermediary map that closely resembled it; see p. 23). This earlier version of the map, which accompanied the submission of the completed book to Allen & Unwin on 3rd October 1936, is neatly labelled in the lower left corner: ‘Thror’s Map · Copied by B.Baggins · For moon-runes hold up to a light’. Beneath the Lonely Mountain in the center is the label ‘Here of old was the land of Thrain King under the Mountain’. In the final version of Thror’s Map [‘Thror’s Map II’ (TM.II); DAA.97] published in the book, made in between 10th December 1936 and 4th January 1937, the label in the lower left has been simplified to ‘Thror’s Map’ and the text beneath the mountain now reads ‘Here of old was Thrain King under the Mountain’. This is the clearest expression of the reverse genealogy: Thrain is the grandfather who was the last King Under the Mountain and Thror is the son who years later made the map before setting out on his own final quest.
Unfortunately, we do not know exactly when TM.I was made. It seems probable that it was during the creation of the First Typescript [e.g., sometime in 1932] – that is, after the emergence of the names Thror and Thrain during the writing of Chapter X and long enough afterwards that Tolkien had forgotten the original sequence, just as he did when typing the typescript version of Chapter VII. However, it might have been slightly later, during the composition of the final chapters [e.g., December 1932–January 1933], when the references to ‘the Arkenstone of Thrain’ seem to indicate that Tolkien had become either confused or ambivalent about the correct sequence. It could even be as late as the late summer and early fall of 1936, when Tolkien finally extended the typescript all the way to the end of the book (see Christopher Tolkien’s letter to Father Christmas, December 1937, cited in the Foreword to the 50th anniversary edition, page vii).
In summary, then, the preponderance of evidence from what we may call the text tradition is overwhelmingly in favor of the original genealogy: Thorin, son of Thrain, son of Thror. However, in several places (Beorn’s reference to Thorin’s ancestry, at least some of the references to the Arkenstone of Thrain, and most importantly on the Map) Tolkien either explicitly or implicitly uses the reverse genealogy instead. Whether this was entirely the result of confusion on his part or deliberate choice is impossible to say, but the evidence suggests the former; it seems unlikely, for example, that Tolkien would deliberately revert to the rejected reverse genealogy when drafting the Council of Elrond scenes in ‘The New Hobbit’ (see Note 20) when he had already committed to the other line of descent in print just two years before (cf. HME VI.403). We have here, therefore, a rare case of Tolkien’s losing track of some detail in the course of revising the book; he managed at the last minute to bring the text into accord (with the exception of two of the references to the Arkenstone) but failed to get the troublesome map to agree with it.
Not until years later did he return to the matter, when as part of the 1947 revisions that became the Second Edition of The Hobbit he resorted to what Taum Santoski called the typically Tolkienesque solution of leaving both contradictory pieces of information in place and adding a third element that took both into account and resolved their apparent contradiction: the invention of Thrain the Old. For a detailed discussion of this new character, and the reasons for his insertion into the story, see The Fourth Phase: The 1947 Hobbit, pp. 780, 788, & 791.