The Return of the Shadow

The Return of the Shadow is not a story but an edition of the papers behind one. It belongs to the four-volume account of how The Lord of the Rings was written, and it opens that account, carrying the record from the first tentative attempts at a sequel to The Hobbit through the successive rewritings of the early chapters and the journey eastward, breaking off long before the shape of the finished book was settled. Its bulk is therefore made up of drafts: manuscript and typescript texts, most of them abandoned and superseded, printed in the order in which they were composed rather than in the order of the narrative they were groping towards. Around these texts stands an editorial framework of introductions, dating arguments, numbered notes on readings and cancellations, and running commentary that marks where a passage stands in relation to what came before it and what the published book eventually made of it. The reader who comes to it expecting continuous narrative will find instead the same episode attempted three or four times over, each attempt annotated.

The editorial voice is deliberately restrained. It explains the state of the manuscripts, weighs evidence for their sequence, and declines for the most part to judge the drafts as literature or to speculate beyond what the papers support; where the evidence is thin, that is said plainly. This suits the material, because the interest of these texts lies in watching a book find itself — names attaching to the wrong figures, motives assigned and withdrawn, whole roles collapsing into one another. Tolkien himself described the writing as a process of discovery rather than design, in which characters arrived unbidden and had to be accounted for afterwards, and the volume exists to show that process in its documentary form.

Its place in the legendarium is accordingly a subordinate one. Nothing printed here has the standing of the published narrative; these are stages that were passed through and left behind, and their value is evidential — they show what the finished work grew out of, including the gradual pull of the older mythological matter into what had begun as a lighter sequel, a convergence Tolkien later described as the drawing together of two bodies of writing that had long developed apart. The audience is thus a particular one: readers already familiar with the completed book, who want to see its composition rather than to read it again.

The papers gathered in the volume were written over a span that begins in the last weeks of 1937. Allen & Unwin, encouraged by the reception of The Hobbit, pressed for a further book about hobbits, and Tolkien answered in October of that year that he could think of nothing more to say about them, offering instead the mythological matter he had long been at work on. By the middle of December he had changed his mind, or been overtaken by it, and reported to his publisher that a first chapter of a new hobbit story stood written, though he did not yet know what it was about. The correspondence of the following years records progress in irregular bursts separated by long silences, with the manuscript stalling, restarting, and being pushed back through revision rather than forward; as late as the winter of 1942 he was reporting a chapter count and a hope of finishing soon that proved badly optimistic.

Tolkien's own retrospective account of that labour, set down in the foreword he wrote for the second edition of the finished work, describes a book that grew in the writing over some twelve years, interrupted by his professional duties and by the war, halted for long periods together, and brought to an end only in 1949. Publication followed at a further remove: the three volumes appeared in 1954 and 1955, the last delayed by the appendices. The drafts printed in the present volume therefore belong to the earlier part of that span, roughly the first four years of composition, and stop well short of the point at which the story was finished. Dating them is one of the editorial problems the book has to solve, since the papers themselves are only intermittently dated and the letters supply fixed points at wide intervals.

The volume itself was assembled some four decades later, as part of a series that had begun in 1983 with the two volumes of the Lost Tales and had worked forward through the earlier legendarium as far as the writings of the later 1930s. Its subtitle marks the change of subject: it opens a subsidiary sequence within the series devoted to the making of The Lord of the Rings, and is designated the first part of that sequence. Christopher Tolkien explains in the foreword why the manuscripts were treated at this length and in this manner, and sets out the physical circumstances that governed the work — chiefly that the bulk of the relevant papers had passed to Marquette University in the 1950s, while related material remained in England, so that the two collections had to be brought into relation before the drafts could be ordered at all. Assistance in that collation is acknowledged there by name.

The book was published in 1988, in London and in the United States, as the sixth volume of the series. Three further volumes completed the account of the writing of The Lord of the Rings over the next four years, appearing in 1989, 1990 and 1992, so that the four together were issued across a five-year span. The text has not been revised since first publication; later printings, including the paperback and the collected reissues under the HarperCollins imprint, reproduce it substantially as it stood, and the sixth to ninth volumes have since circulated grouped as a single unit. No German translation of this volume exists. Of the whole series only the two Lost Tales volumes have been rendered into German, with the result that the English title remains the correct form of citation in German-language reference as well.

The volume opens with a title-page subtitle that fixes its place in the larger scheme: it is presented as the first part of a history of The Lord of the Rings, and everything that follows is arranged to serve that purpose. After Christopher Tolkien's foreword, which sets out the condition of the papers and the principles governing their presentation, the body of the book falls into three parts named not for stretches of story but for stages of work. Each is a phase of composition, and each begins again at the beginning. The consequence for the reader is unusual and needs stating plainly: moving from the front of the book to the back means passing over the same opening ground three times, in three separate attempts, rather than following one line of narrative from Hobbiton outwards. The chapters are numbered consecutively through all three parts, so that a chapter number identifies a place in the book even though the story doubles back on itself.

The first phase carries the earliest drafting from the opening birthday-party chapter through the road east, the Old Forest, Bree and the flight to the Ford, and ends at Rivendell; it closes not with narrative but with a chapter given over to Tolkien's own memoranda of unresolved questions and proposed alterations, a document that shows the sticking points at which the first attempt stalled. The second phase returns to Hobbiton and rewrites forward again, and it is here that the long expository conversation about the Ring's history takes something like its eventual shape, the material that the published book places in its second chapter. The third phase begins once more from the start and is carried furthest of the three: it works through Bree and Rivendell to the council at which the Ring's fate is decided, then south and into Moria, where the volume breaks off. A reader looking for a particular episode should therefore expect to find it in two or three places, and should consult the part divisions before the chapter titles.

Within a chapter the arrangement is regular. A draft text is given, in whole where it differs substantially from what was printed in 1954 and in summary where it had already come close to it, and is followed by discussion and by numbered notes keyed to points in the text, recording cancelled readings, alternative names and the physical state of the manuscript. Cross-references to the finished work are made by book and chapter, so that any draft can be set beside its published counterpart. An index of names closes the volume and is the quickest route in for a reader pursuing a single figure or place through its changes of name and role. Because the book stops in the middle of the drafting rather than at a natural division, anyone following a thread beyond Moria must continue in the next volume of the sequence, where the same editorial arrangement is kept up.

The volume sits at a hinge between two published books and takes nearly all of its meaning from them. Everything in it presupposes The Hobbit as a finished work already in print: a hobbit had gone east, had come back with a ring picked up in the dark, and that ring had been treated as a convenient trinket rather than as an object of consequence. The sequel that grew out of these papers had to make it carry a weight the earlier book had never given it, and the published result shows the repairs plainly — a prologue that explains how the finding was first reported and why the first report was false, and a later revision of the riddle chapter itself, which brought the older book into agreement with the newer one. The drafts gathered here belong to the years before either repair had been devised.

Set against the finished narrative, the volume carries no authority at all, and this is the point on which it is most often misused. Where a draft and a published chapter diverge, there is nothing to reconcile and no rival tradition to weigh: the printed book is the sole account of what happened, and the papers record only what was tried and set aside. The asymmetry runs deeper than wording. The finished work supplies its own internal frame of reference — a year-by-year chronology of the Third Age, the genealogies and linguistic notes appended to it, the fiction of a manuscript transmitted from hand to hand — and that apparatus belongs to a stage of labour far beyond anything represented here. The drafts cannot be read as competing testimony about Middle-earth; they can only be read as the prehistory of the text that eventually produced such testimony.

The relation to the older mythology is similarly oblique. The background on which the finished book leans — the making of the Rings, the alliance of Elves and Men that threw Sauron down at the end of the Second Age, the drowning of Númenor and the founding of the realms in exile — reached settled form only in writings edited and issued after Tolkien's death, and during the years covered by these drafts much of it was neither fixed on paper nor available to any reader. Two consequences follow. What the posthumous Silmarillion presents as an ordered account stands, in this period, as work still in motion and still capable of going otherwise; and the join between the hobbit story and the greater matter — which Tolkien afterwards explained at length in a separate piece on why the wizard sent dwarves to Bilbo's door — was contrived well after the drafting recorded here had broken off. The volume thus documents the interval before the two strands were fitted to each other, which is its chief value to anyone studying the legendarium entire rather than the completed book alone.