The Lost Road and Other Writings

The Lost Road and Other Writings is not a story but an edition: a gathering of unpublished manuscripts from J.R.R. Tolkien's papers, presented with dates, variant readings and a running editorial commentary by his son. As the fifth instalment of the series it also marks a boundary. The four volumes before it work forward from the earliest phase of the mythology, and this one carries that line as far as the middle and later 1930s, the point at which the writing of The Lord of the Rings interrupted everything else. Readers coming to it in sequence therefore reach a natural pause here rather than a conclusion; the sixth volume opens an entirely different body of drafts. The book is arranged in three parts of very unequal character, followed by appended material, and each text is introduced with an account of the manuscripts on which it rests.

The first part collects the abandoned time-travel narrative that gives the volume its title, together with the associated Númenórean matter; Tolkien himself later recalled, in correspondence, the agreement with C.S. Lewis that had prompted him to attempt such a story at all. The second part assembles the versions of the Silmarillion tradition standing immediately before the long interruption — annals, cosmogonic prose, a treatise on the Elvish tongues, and the prose Quenta in its fullest pre-war state. The third part is a philological document of a wholly different kind, a dictionary of roots and their derivatives across the Elvish languages, which has since become one of the most consulted items in the whole series. Genealogies, a list of names and cartographic material close the volume.

The editorial voice throughout is dry, precise and deliberately unliterary: it dates hands and papers, tracks emendations, and refrains from smoothing the material into a readable whole. Nothing here is offered as settled narrative, and the book assumes a reader who already knows the published Silmarillion and wants to see how its constituent texts stood at a particular moment. Its value in the legendarium is thus documentary rather than canonical — a cross-section of the mythology taken just before its author turned aside to a sequel he had not planned to write.

The manuscripts gathered in this volume belong, with few exceptions, to a single stretch of work in the mid- and later 1930s, so that the book is in effect a dossier of about four years. The Númenórean matter and the modern frame designed to reach it were taken up around the middle of the decade and never carried far: what survives amounts to a pair of opening chapters set in England of the author's own day, an episode on the island itself, and sketches for stages of the journey that were never written. The annals of the Elder Days, the cosmogonic prose and the account of the Elvish tongues were rewritten or newly made in the same span, which allows the historical and the philological strands to be dated against one another with unusual closeness. The prose Quenta was recast from its predecessor of the early 1930s and pushed forward steadily until the end of 1937, when it stopped in mid-course.

The break at the close of 1937 was not a change of mind but an accident of publishing. In the late autumn of that year a portion of the unpublished mythology was submitted to the firm that had brought out The Hobbit, which was looking for a further tale of hobbits rather than for a legendarium; the papers were returned, and within weeks Tolkien had begun the sequel that would hold him for the better part of two decades. The Quenta and the time-travel story were both left where they stood, and neither was ever resumed in the form it then had. Writing to a correspondent nearly thirty years later, Tolkien recalled the abandoned attempt without much regret, noting that the Númenórean material it had been built around outlived the frame and passed into the later work.

Almost fifty years separate the writing from the printing. The book was published in 1987, by George Allen & Unwin in London and by Houghton Mifflin in Boston, as the fifth instalment of a series that had been appearing at roughly annual intervals since 1983; the sixth volume followed in 1988. Preparing it involved rather more than transcription, since several of the texts exist in layered manuscripts and typescripts whose stages had first to be separated before any of them could be presented as a text at all; the editor sets out these findings before each item, and dates them by handwriting, paper and the relation of one document to another. The genealogical tables, the list of names and the second map of the Elder Days were added as an appendix, and the volume carries its own index.

The text has since been reprinted in paperback and kept in print by HarperCollins, but it has not been substantially revised by its editor, and the readings established in 1987 remain the standard ones. The exception is the philological dictionary of Part Three, whose manuscript is the most difficult in the book: a substantial body of corrections and additions to it, drawn from a fresh examination of the papers, was published in the journal Vinyar Tengwar in 2003 and 2004, and serious work on the Elvish roots now proceeds from the 1987 text together with those addenda. The volume has never been translated into German; of the whole series only the two parts of the earliest collection of tales exist in German, so that the English title remains the correct form of citation in German-language reference as well.

Part One falls into three chapters. The first traces how the Númenórean legend arose and how it came to be joined to a scheme of inherited memory; the second gathers the short text known as The Fall of Númenor, given first in a preliminary outline and then in successive versions whose divergences are set out in numbered notes rather than conflated; the third presents what exists of the narrative itself, chapter by chapter, with the fragmentary drafts placed after the finished portions. The modern setting turns on the Errol family — a father, Oswin, his son Alboin, and in the following generation Audoin — whose names carry the Lombard legend the story was built to use, while the Númenórean episode is given to Elendil and his son Herendil. Attached to this part is an alliterative poem on a legendary king who came to his people out of the sea as a child, printed with an account of the place it was to hold in the chapters never written.

Part Two, much the longest, has six chapters. It opens with a discussion of the texts and their relations, which is the place to begin for anyone trying to establish what was written when; there follow the later annals of Valinor and of Beleriand, the cosmogonic prose, the treatise on the Elvish tongues in two versions with a shorter supplementary piece, and finally the prose Quenta. The editor divides the annals and the Quenta into numbered paragraphs of his own devising, and the commentary that follows each text is keyed to those numbers, so that a point of detail can be looked up directly instead of read through. The Quenta breaks off in the course of the tale of Túrin, and the remainder of the story is accounted for by reference to the earlier version printed in the preceding volume of the series, with the later alterations noted separately.

Part Three consists of a single item, arranged alphabetically by root rather than by word. Each entry gives a base in capitals followed by its derivatives grouped under the several languages, abbreviated according to a table set out before the text; doubtful readings are marked as such, and the editor indicates where entries were struck through or added at a later date. Because the material is organised from the roots outward, a reader seeking the derivation of a particular name will generally need the index rather than the alphabetical sequence itself.

The appended matter comprises three items: genealogical tables for the princely houses of the Elves and for the houses of the Edain; an unfinished alphabetical list of names, largely etymological in character, which breaks off before completion; and the second map of the Elder Days, reproduced in sections and redrawn, with a grid of lettered and numbered squares that later volumes continue to use for reference. A full index closes the book and covers all three parts together with the appendix, which makes it the practical entrance to the volume for anyone approaching it with a single name or question in hand.

In the sequence of texts that ends in the published Akallabêth, this volume holds the first station. The Downfall as The Silmarillion tells it — the island granted to the Edain after the wars of the Elder Days, the prohibition against sailing west out of sight of their own shores, the last king's armament against the Undying Lands, the drowning of the land and the removal of the Blessed Realm from the circles of the world, the escape of Elendil and his sons and the founding of the realms in exile — is present in the 1930s papers only in its bones, and a good deal that the later account treats as settled was still being found. Everything the volume contains about Númenor stands well before the fuller Second Age matter of Unfinished Tales: the line of the kings from Elros onward, the tale of the mariner Aldarion and his wife, and the account of the island's shape, harbours and customs all belong to later phases of work and cannot be read back into these drafts. What the volume does establish is the outline of the legend itself, and that outline held: the Númenórean story enters the mythology at this point and never afterwards leaves it.

Toward The Lord of the Rings the relation is one of foundation rather than of anticipation. The Appendices assume as known a drowned western kingdom, a surviving faithful remnant and two successor realms in Middle-earth, and they set that history out in a chronology of the Second Age which nothing in the 1930s work yet resembles in detail; the legend was in place before the sequel began, and was drawn upon rather than devised for it. The philological side of the volume stands in a different relation again. The Elvish tongues as the published Appendices describe them — a high speech brought out of the West and the language of the Grey-elves of Beleriand, with their names and their histories — do not correspond point for point to the scheme worked out before the war, in which the Beleriandic tongue was still held to be the speech of the exiled Noldor. The roots and derivations of the earlier material remain usable, but they are read through a later account of who spoke what, and where.

Where these texts disagree with the published books, the disagreement is generally one of stage rather than of standing. The Silmarillion was assembled from manuscripts of very different dates, its editor's foreword being explicit that the work was broken off in the late 1930s and resumed only long afterwards, so that the Elder Days narrative as printed rests in places on this phase of writing and in others on revisions made twenty years later. The volume therefore lets a reader see which parts of the published compilation were already fixed before the interruption and which were not, a distinction the finished book deliberately does not mark. Tolkien's own remarks on Númenor, made in letters to correspondents who asked how the tale fitted the whole, describe the conception at which he eventually arrived and give no authority to the trial forms; the drafts are evidence of how he got there, not statements of what is so.