I am afraid all the same that the presentation will need a lot of work, and I work so slowly. The legends have to be worked over (they were written at different times, some many years ago) and made consistent; and they have to be integrated with The L.R.; and they have to be given some progressive shape. No simple device, like a journey and a quest, is available.

I am doubtful myself about the undertaking…

When after his death the question arose of publishing ‘The Silmarillion’ in some form, I attached no importance to this doubt. The effect that ‘the glimpses of a large history in the background’ have in The Lord of the Rings is incontestable and of the utmost importance, but I did not think that the ‘glimpses’ used there with such art should preclude all further knowledge of the ‘large history’.

The literary ‘impression of depth…created by songs and digressions’ cannot be made a criterion by which a work in a wholly different mode is measured: this would be to treat the history of the Elder Days as of value primarily or even solely in the artistic use made of it in The Lord of the Rings. Nor should the device of a backward movement in imagined time to dimly apprehended events, whose attraction lies in their very dimness, be understood mechanically, as if a fuller accou1nt of the mighty kings of Nargothrond and Gondolin would imply a dangerously near approach to the bottom of the well, while an account of the Creation would signify the striking of the bottom and a definitive running-out of ‘depth’—‘nothing to reach further back to’.

This, surely, is not how things work, or at least not how they need work. ‘Depth’ in this sense implies a relation between different temporal layers or levels within the same world. Provided that the reader has a place, a point of vantage, in the imagined time from which to look back, the extreme oldness of the extremely old can be made apparent and made to be felt continuously. And the very fact that The Lord of the Rings establishes such a powerful sense of a real time-structure (far more powerful than can be done by mere chronological assertion, tables of dates) provides this necessary vantage-point. To read The Silmarillion one must place oneself imaginatively at the time of the ending of the Third Age—within Middle-earth, looking back: at the temporal point of Sam Gamgee’s ‘I like that!’—adding, ‘I should like to know more about it’. Moreover the compendious or epitomising form and manner of The Silmarillion, with its suggestion of ages of poetry and ‘lore’ behind it, strongly evokes a sense of ‘untold tales’, even in the telling of them; ‘distance’ is never lost. There is no narrative urgency, the pressure and fear of the immediate and unknown event. We do not actually see the Silmarils as we see the Ring. The maker of ‘The Silmarillion’, as he himself said of the author of Beowulf, ‘was telling of things already old and weighted with regret, and he expended his art in making keen that touch upon the heart which sorrows have that are both poignant and remote’.

As has now been fully recorded, my father greatly desired to publish ‘The Silmarillion’ together with The Lord of the Rings. I say nothing of its practicability at the time, nor do I make any guesses at the subsequent fate of such a much longer combined work, quadrilogy or tetralogy, or at the different courses that my father might then have taken—for the further development of ‘The Silmarillion’ itself, the history of the Elder Days, would have been arrested. But by its posthumous publication nearly a quarter of a century later the natural order of presentation of the whole ‘Matter of Middle-earth’ was inverted; and it is certainly debatable whether it was wise to publish in 1977 a version of the primary ‘legendarium’ standing on its own and claiming, as it were, to be self-explanatory. The published work has no ‘framework’, no suggestion of what it is and how (within the imagined world) it came to be. This I now think to have been an error.

The letter of 1963 quoted above shows my father pondering the mode in which the legends of the Elder Days might be presented. The original mode, that of The Book of Lost Tales, in which a Man, Eriol, comes after a great voyage over the ocean to the island where the Elves dwell and learns their history from their own lips, had (by degrees) fallen away. When my father died in 1973 ‘The Silmarillion’ was in a characteristic state of disarray: the earlier parts much revised or largely rewritten, the concluding parts still as he had left them some twenty years before; but in the latest writing there is no trace or suggestion of any ‘device’ or ‘framework’ in which it was to be set. I think that in the end he concluded that nothing would serve, and no more would be said beyond an explanation of how (within the imagined world) it came to be recorded.

In the original edition of The Lord of the Rings Bilbo gave to Frodo at Rivendell as his parting gift ‘some books of lore that he had made at various times, written in his spidery hand, and labell1ed on their red backs: Translations from the Elvish, by B.B.’ In the second edition (1966) ‘some books’ was changed to ‘three books’, and in the Note on the Shire Records added to the Prologue in that edition my father said that the content of ‘the three large volumes bound in red leather’ was preserved in that copy of the Red Book of Westmarch which was made in Gondor by the King’s Writer Findegil in the year 172 of the Fourth Age; and also that

These three volumes were found to be a work of great skill and learning in which…[Bilbo] had used all the sources available to him in Rivendell, both living and written. But since they were little used by Frodo, being almost entirely concerned with the Elder Days, no more is said of them here.

In The Complete Guide to Middle-earth Robert Foster says: ‘Quenta Silmarillion was no doubt one of Bilbo’s Translations from the Elvish preserved in the Red Book of Westmarch.’ So also I have assumed: the ‘books of lore’ that Bilbo gave to Frodo provided in the end the solution: they were ‘The Silmarillion’. But apart from the evidence cited here, there is, so far as I know, no other statement on this matter anywhere in my father’s writings; and (wrongly, as I think now) I was reluctant to step into the breach and make definite what I only surmised.

The choice before me, in respect of ‘The Silmarillion’, was threefold. I could withhold it indefinitely from publication, on the ground that the work was incomplete and incoherent between its parts. I could accept the nature of the work as it stood, and, to quote my Foreword to the book, ‘attempt to present the diversity of the materials—to show “The Silmarillion” as in truth a continuing and evolving creation extending over more than half a century’ and that, as I have said in Unfinished Tales (p. 1), would have entailed ‘a complex of divergent texts interlinked by commentary’—a far larger undertaking than those words suggest. In the event, I chose the third course, ‘to work out a single text, selecting and arranging in such a way as seemed to me to produce the most coherent and internally self-consistent narrative’. Having come, at length, to that decision, all the editorial labour of myself and of Guy Kay who assisted me was directed to the end that my father had stated in the letter of 1963: ‘The legends have to be worked over…and made consistent; and they have to be integrated with the L.R.’ Since the object was to present ‘The Silmarillion’ as ‘a completed and cohesive entity’ (though that could not in the nature of the case be entirely successful), it followed that there would be in the published book no exposition of the complexities of its history.

Whatever may be thought of this matter, the result, which I by no means foresaw, has been to add a further dimension of obscurity to ‘The Silmarillion’, in that uncertainty about the age of the work, whether it is to be regarded as ‘early’ or ‘late’ or in what proportions, and about the degree of editorial intrusion and manipulation (or even invention), is a stumbling-block and a source of much misapprehension. Professor Randel Helms, in Tolkien and the Silmarils (p. 93), has stated the question thus: