Sauron Defeated

This is a documentary edition rather than a narrative one. Like the other instalments of the series to which it belongs, it prints unfinished manuscripts and typescripts in the order in which they appear to have been written, with an editorial apparatus that dates them, describes their physical state, records deleted readings and alternative versions, and sets each layer against the one before it. The presiding voice is the editor's, and it is a deliberately restrained one: it explains what a page is, when it was probably made and how it relates to what survives elsewhere, and it declines to smooth the material into a finished story. Readers who open the book expecting a continuous tale will find instead the workings that lie behind one.

The volume is unusual within the series for being visibly made of two unrelated undertakings. Its first part closes the long sequence of drafts for The Lord of the Rings that occupied the three preceding volumes, and with it that editorial project comes to an end; this portion was later reissued on its own under its own title, which indicates how self-contained it is. The remaining two parts turn to a quite different body of work, one belonging to the older matter of Númenor and to Tolkien's experiments with framing that matter for a modern reader. The three parts are unequal in length and unlike one another in kind, and the editor makes no attempt to argue them into a single subject; the binding, not the material, is what holds them together.

The intended reader is one who already knows the published books well enough to notice where a draft departs from them, and who is interested in how the writing moved rather than in what it finally said. Nothing printed here carries the standing of a finished text: these are stages, several of them abandoned, and the editor is consistent in marking them as such. The book's place in the legendarium is therefore an oblique one. It documents the point at which the long labour on the Third Age was set down and the older Númenórean strand taken up again, and it is valuable chiefly as evidence about composition — for chronology, for the history of names and of invented language, and for the shape of intentions that were never carried through.

The papers gathered in the first part of the volume belong to a labour whose outline Tolkien himself set down. In the foreword he wrote for the second edition of The Lord of the Rings he described a work taken up not long after The Hobbit was finished and then advanced only in fits, with intervals in which the story stood still for a year or more; he recalled sending the chapters of the fourth book out in instalments to his son on war service in South Africa, and stated that the tale was not brought to its end until 1949, some five years before the first of the three published volumes appeared in 1954, with the third following in 1955. The drafts printed here thus fall for the most part in the second half of the 1940s, in the final push towards an ending and in the long revision that came after it. Among them are stages of the closing chapters that were rewritten more than once, and a closing scene set some years after the main narrative which Tolkien composed, circulated and eventually withdrew, so that it never reached the published book at all.

The second and third parts of the volume come out of the interruption rather than the labour. Looking back in a letter written in 1964, Tolkien recalled an agreement with C. S. Lewis by which each of them would try to write the kind of story they found too little of, Lewis taking space and himself time, and he described the abandoned time-travel book that resulted on his side, one designed to end with his protagonist present at the drowning of the great island of the west. That first attempt was broken off in the 1930s; the material printed here represents a second and much fuller assault on the same design, made in the middle 1940s while the long tale of the Ring was set aside, and abandoned in its turn. With it belongs a body of linguistic invention of the same years, presented within the fiction as the report of one of the story's characters, which is the earliest sustained account of the language of the drowned island.

The reason these versions waited so long for print is that the matter they treat had already reached readers by other routes. The downfall of the island was told in a compressed narrative form in the collection edited from the papers and published in 1977, and further Second Age material — a geographical account of the island and a tale of one of its early rulers — followed in the volume of unfinished pieces issued in 1980. Both of those books were assembled by Christopher Tolkien, and both drew on writings later than the drafts printed here. What appeared in 1992 was therefore not new subject matter but the earlier and rejected shapes of it, set out with the dates and manuscript descriptions that allow one version to be placed against another.

The volume was published in 1992 as the ninth instalment of a series that began in 1983 and ran to twelve numbered books, the last of them appearing in 1996. The British edition came from HarperCollins, which by then had taken in the firm of George Allen and Unwin that had published Tolkien in his lifetime, and an American edition followed from Houghton Mifflin. In the letter of about 1951 in which he laid his whole invented history before a prospective publisher, Tolkien had pressed for the older matter and the Ring narrative to be brought out together; the series in which this volume stands is a late and very different answer to that wish, printing the drafts rather than a finished text. The first part was afterwards detached and reissued on its own in paperback as The End of the Third Age. No German translation exists: of the whole series only the two earliest volumes have been rendered into German, by Klett-Cotta, so that German-language reference works cite this book under its English title.

The book falls into three numbered parts of markedly unequal size, each carrying its own title, and the ordering within them follows two different principles. Part One is arranged by the finished work: it moves through the last of the six books of the Ring narrative chapter by chapter, in the sequence the published text has, from the resumption of the story of the two hobbits on the eastern side of the mountains, through the passage across the plain and the ruin of the Dark Tower, the judgement and rejoicing on the field before the Gate, the crowning and the marriage at the White City, the long series of leave-takings on the northward road, the reckoning in the Shire and the last ride to the sea. Each chapter is given its own editorial section, so that a reader who knows where an incident stands in the published book can find the drafts for it directly. The abandoned closing scene is placed after the whole sequence, and more than one state of it is printed.

Part Two occupies the centre of the volume and is the longest single stretch of continuous text in it. The material is presented in two divisions, reproducing the fiction's own separation of its record into an earlier and a later series of sittings, each sitting numbered and dated within the invented frame; the editor's commentary follows each division rather than interrupting it, and a further section sets out the places where the earlier drafts of the second division depart substantially from the text as printed. Because Tolkien rewrote the opening several times and shifted the names and identities of the speakers, a good deal of the apparatus here is concerned with sorting one layer of the frame from another, and the reader who wants the narrative rather than the frame will find it concentrated in the second division.

Part Three prints the successive versions of the Númenórean narrative one after another instead of conflating them, so that the same events can be read three and four times over in different shapes, with the editor's discussion set between; a comparison of that matter with the account known from the published collection is made separately. The linguistic report attributed to one of the characters follows at the end of this part, with the editor's notes on the phonology and grammar it sets out. The volume closes with an index of the usual kind for the series, covering names in all three parts together, which is the practical way into a book whose three sections share almost no vocabulary. There are no maps and no genealogical tables; the geography assumed in Part Three is that of the descriptive account printed in the earlier volume of unfinished pieces, and the reader is left to bring it from there.

Nothing printed in this volume has standing as an account of what happened. The matter it preserves reached its settled form elsewhere: for the Third Age in the published tale of the Ring, whose appendices supply the chronology, the descent of the Númenórean and Gondorian kings and the notes on speech and script by which the whole invention is ordinarily read. Those appendices are the fixed points, and the drafts gathered here fall on the far side of them. Where a page in this book differs from the published text — in a name, a sequence of events, a motive — the difference is evidence about how the writing moved and not an alternative reading of the history. The presupposition therefore runs one way only: the volume assumes a reader who already holds the finished books and can lay a draft beside them.

The same holds, with a further complication, for the Second Age. The received account of the island kingdom is the one edited from the later papers and published in 1977, in which the land is granted to the Edain as a reward, a prohibition on sailing westward is laid upon them, that prohibition is broken at the instigation of the captive enemy, and the ruin that follows changes the shape of the world itself. Writing to a prospective publisher, Tolkien himself described the episode in these terms, as a second fall of Men and the moment at which the ways west were closed. The versions printed in the present volume tell the same events but from a different vantage: they are shaped as a tradition carried by Men, in which the parts of the Powers and of the Elves are dimmer, the theology less explicit, and the persons and places named in the tongue of the island rather than in the Elvish forms familiar from the published narrative. That is best read as a difference of transmission rather than a contradiction to be adjudicated.

Two things the book carries that no published work supplies. The first is the speech of the island: the appendices to the Ring narrative name it as the ancestor of the Common Speech and say what became of it, but give nothing of its sounds or its grammar, and the sustained description here has no counterpart in print elsewhere. The second is the device of approach. The published tale presents itself as a translation from a hobbit chronicle, a frame handled lightly and confined to the prologue and the appendices; the material in the middle of this volume represents a quite different and abandoned solution, in which the old matter was to reach a modern reader through the recollections of living men. Tolkien never used that solution, and its absence from everything he published is itself the point of interest.

Finally, the volume is overtaken rather than supplemented by the later Second Age writings. The account of the island's shape and harbours, and the long tale of the mariner king and his wife with its quarrel over the sea, both belong to a period well after these drafts and describe a society the drafts have not yet imagined. The order of composition and the order of publication thus run against each other here, and the reader who comes to this book from the collections of 1977 and 1980 is moving backwards. Its use in the legendarium is accordingly diagnostic: it shows which elements were early and which were late, and it marks the seam at which one great labour was laid down and an older one briefly resumed.