The t1racing of this long evolution is to me of deep interest, and I hope that it may prove so to others who have a taste for this kind of enquiry: whether the major transformations of plot or cosmological theory, or such a detail as the premonitory appearance of Legolas Greenleaf the keen-sighted in the tale of The Fall of Gondolin. But these old manuscripts are by no means of interest only for the study of origins. Much is to be found there that my father never (so far as one can tell) expressly rejected, and it is to be remembered that ‘The Silmarillion’, from the 1926 ‘Sketch’ onwards, was written as an abridgement or epitome, giving the substance of much longer works (whether existing in fact, or not) in a smaller compass. The highly archaic manner devised for his purpose was no fustian: it had range and great vigour, peculiarly apt to convey the magical and eerie nature of the early Elves, but as readily turned to the sarcastic, sneering Melko or the affairs of Ulmo and Ossл. These last approach at times a comic conception, and are delivered in a rapid and lively language that did not survive in the gravity of my father’s later ‘Silmarillion’ prose (so Ossл ‘fares about in a foam of business’ as he anchors the islands to the sea-bed, the cliffs of Tol Eressлa new-filled with the first sea-birds ‘are full of a chattering and a smell of fish, and great conclaves are held upon its ledges’, and when the Shoreland Elves are at last drawn over the sea to Valinor Ulmo marvellously ‘fares at the rear in his fishy car and trumpets loudly for the discomfiture of Ossл’).
The Lost Tales never reached or even approached a form in which my father could have considered their publication before he abandoned them; they were experimental and provisional, and the tattered notebooks in which they were written were bundled away and left unlooked at as the years passed. To present them in a printed book has raised many thorny editorial problems. In the first place, the manuscripts are intrinsically very difficult: partly because much of the text was written rapidly in pencil and is now in places extremely hard to read, requiring a magnifying glass and much patience, not always rewarded. But also in some of the Tales my father erased the original pencilled text and wrote a revised version over it in ink—and since at this period he used bound notebooks rather than loose sheets, he was liable to find himself short of space: so detached portions of tales were written in the middle of other tales, and in places a fearsome textual jigsaw puzzle was produced.
Secondly, the Lost Tales were not all written progressively one after the other in the sequence of the narrative; and (inevitably) my father began a new arrangement and revision of the Tales while the work was still in progress. The Fall of Gondolin was the first of the tales told to Eriol to be composed, and the Tale of Tinъviel the second, but the events of those tales take place towards the end of the history; on the other hand the extant texts are later revisions. In some cases nothing earlier than the revised form can now be read; in some both forms are extant for all, or a part, of their length; in some there is only a preliminary draft; and in some there is no formed narrative at all, but only notes and projections. After much experimentation I have found that no method of presentation is feasible but to set out the Tales in the sequence of the narrative.
And finally, as the writing of the Tales progressed, relations were changed, new conceptions entered, and the development of the languages pari passu with the narrative led to continual revision of names.
An edition that takes account of such complexities, as this does, rather than attempt to smooth them artificially away, is liable to be an intricate1 and crabbed thing, in which the reader is never left alone for a moment. I have attempted to make the Tales themselves accessible and uncluttered while providing a fairly full account, for those who want it, of the actual textual evidences. To achieve this I have drastically reduced the quantity of annotation to the texts in these ways: the many changes made to names are all recorded, but they are lumped together at the end of each tale, not recorded individually at each occurrence (the places where the names occur can be found from the Index); almost all annotation concerned with content is taken up into, or boiled down into, a commentary or short essay following each tale; and almost all linguistic comment (primarily the etymology of names) is collected in an Appendix on Names at the end of the book, where will be found a great deal of information relating to the earliest stages of the ‘Elvish’ languages. In this way the numbered notes are very largely restricted to variants and divergences found in other texts, and the reader who does not wish to trouble with these can read the Tales knowing that that is almost all that he is missing.
The commentaries are limited in their scope, being mostly concerned to discuss the implications of what is said within the context of the Tales themselves, and to compare them with the published Silmarillion. I have eschewed parallels, sources, influences; and have mostly avoided the complexities of the development between the Lost Tales and the published work (since to indicate these even cursorily would, I think, be distracting), treating the matter in a simplified way, as between two fixed points. I do not suppose for one moment that my analyses will prove either altogether just or altogether accurate, and there must be clues to the solution of puzzling features in the Tales which I have failed to observe. There is also included a short glossary of words occurring in the Tales and poems that are obsolete, archaic, or rare.
The texts are given in a form very close to that of the original manuscripts. Only the most minor and obvious slips have been silently corrected; where sentences fall awkwardly, or where there is a lack of grammatical cohesion, as is sometimes the case in the parts of the Tales that never got beyond a first rapid draft, I have let them stand. I have allowed myself greater freedom in providing punctuation, for my father when writing at speed often punctuated erratically or not at all; and I have gone further than he did in consistency of capitalisation. I have adopted, though hesitantly, a consistent system of accentuation for Elvish names. My father wrote, for instance: Palыrien, Palъrien, Palurien;
Lastly, the division of this edition into two parts is entirely due to the length of the Tales. The edition is conceived as a whole, and I hope that the second part will appear within a year of the first; but each part has its own Index and Appendix on Names. The second part contains what are in many respects the most interesting of the Tales: Tinъviel, Turambar (Tъrin), The Fall of Gondolin, and the Tale of the Nauglafring (the Necklace of the Dwarves); outlines for the Tale of Eдrendel and the conclusion of the work; and
I
THE COTTAGE OF LOST PLAY
On the cover of one of the now very battered ‘High School Exercise Books’ in which some of the Lost Tales were composed my father wrote: The Cottage of Lost Play, which introduceth [the] Book of Lost Tales; and on the cover is also written, in my mother’s hand, her initials, E.M.T., and a date, Feb. 12th 1917. In this book the tale was written out by my mother; and it is a fair copy of a very rough pencilled manuscript of my father’s on loose sheets, which were placed inside the cover. Thus the date of the actual composition of this tale could have been, but probably was not, earlier than the winter of 1916–17. The fair copy follows the original text precisely; some further changes, mostly slight (other than in the matter of names), were then made to the fair copy. The text follows here in its final form.