Arda Reconstructed

Douglas Charles Kane. Bethlehem (PA): Lehigh University Press, 2009. ISBN 978-0980149630. Hardcover. 280pp. $65.00.

buy online

Reviewed by Jason Fisher

[This review originally appeared in Mythlore 27.3/4 (#105/106) (2009): 189–95.]

W.H. Auden, Tolkien’s dedicated advocate and one-time student, once wrote that “The words of a dead man / Are modified in the guts of the living” (93). Auden was thinking of Yeats in these lines, not Tolkien, but they describe very well the situation Tolkien’s readers face today: attempting to digest an enormous (and still growing) body of Tolkien’s posthumous writings, with new works continuing to appear nearly forty years after the author’s death. Douglas Kane’s Arda Reconstructed: The Creation of the Published Silmarillion is one particular attempt, in which the author seeks to digest a wide range of these texts, to sample the complex flavors of their interrelationship(s), and to chew on the role Tolkien’s son, Christopher, played in making them more palatable to a larger audience. In Arda Reconstructed, then, Kane explores how Tolkien’s words were “modified” in Christopher’s “guts,” and in the process reveals how they have been “modified” in his own as well.

To put it another way, there are many possible “Silmarillions” — Christopher’s (published in 1977), Charles Noad’s, Douglas Kane’s, yours, mine. Arda Reconstructed seems to be about objective questions — what Christopher cut, what he changed — but at least as much, it is about subjective aspects of the work — why was it cut, what was the effect, how could the book have been different, and so on. In this, Kane acts both as a reader of The Silmarillion, and as a reviewer of Christopher’s efforts to makes sense of his father’s “Silmarillion” papers. But we would do well to remember Tolkien’s admonition (Letters 304): “[a] sharp distinction must be drawn between the tastes of reviewers [...] and of readers!” So warned, this reviewer will proceed with caution.

One marvels at the amount of work Kane has invested in his project and appreciates the rigor with which it is documented. Meticulous as it is, one has the feeling that — like all icebergs of scholarship — only perhaps one-tenth of the author’s labor has actually made it onto the printed page. The only more thorough inspection of Tolkien’s innumerable recensions, drafts, jottings, and marginalia that I can think of was Christopher’s own — which Kane has largely retraced and reconstructed (to the extent possible). Kane’s book is the most extensive and systematic use to which The History of Middle-earth has ever been put. Indeed, one justification for the very existence of that series must have been to make possible studies like Arda Reconstructed. Kane’s book can be used as a roadmap to the vast welter of writings that form the “hypotext” of The Silmarillion, and it has enormous value for this alone, even before one considers the opinions Kane shares and the conclusions he ventures.

Where Kane is at his best is in his relentless ferreting out — so far as he can — of the source(s) of each and every paragraph in the published Silmarillion, from the tangle of source texts underlying it. Kane admits the likelihood that “some of the changes, omissions, and additions that I describe reflect textual material not included (for whatever reasons) in those works [The History of Middle-earth, etc.], or some other source only available to Christopher (including, perhaps, personal conversations that he had with his father)” (25). This is a prudent disclaimer, and it necessarily circumscribes Kane’s analysis. Not to have seen the original manuscripts housed in the Bodleian (and elsewhere, including any that might still be in Christopher’s hands) makes it difficult for Kane to push his argument to a definitive (or authoritative) conclusion, particularly because Christopher “mostly does not show the final step: his actual creation of the published work” (24).

So it must remain educated guesswork on Kane’s part, limited by what has been made public heretofore. Fortunately for Kane, with The Children of Húrin, Christopher has probably now published nearly everything of relevance to Kane’s pursuit. I say “probably” and “nearly,” because I can think of at least one text for which we have still seen none of the intermediate steps, and little discussion of them: Of the Rings of Power and the Third Age. But working within these logical limits, Kane has managed to collate the sources of almost every paragraph of The Silmarillion, a considerable achievement. These sources consist, in the main, of an intricate mélange of various versions of the Quenta, The Annals of Aman, and The Grey Annals, with frequent smaller borrowings from other texts — and in one case, from correspondence by Tolkien. In one extraordinary instance, Kane shows how a single paragraph was constructed from no fewer than six source texts (76). If much of Arda Reconstructed is not revelatory, it is because of the thoroughness of The History of Middle-earth. But what Kane does do is to put the copious source texts into a logical, digestible order; explain how The Silmarillion was assembled from that farrago of sources; summarize and comment on the most important changes, omissions, and (more rarely) inventions Christopher made; and finally, speculate as to some of the reasons and motives for them.

But hold on a moment. To make judgments about changes or reductions in “Tolkien’s vision” for “The Silmarillion” presupposes an understanding of just what Tolkien’s vision was — to the extent this was ever fixed and knowable. It seems that Kane sometimes presumes he understands that vision better than Christopher does. Kane repeatedly claims that “Tolkien clearly intended” this or that (26, 63-4, 75, 84, 93, 98, 106-7, 190, et passim), sometimes on the authority of Tolkien’s own words, but often not. This is a difficult position to defend — first of all, because Christopher has had access to material and personal experience that Kane has not. Kane accuses Christopher of having made presumptions of his own about Tolkien’s intentions for “The Silmarillion” — fair enough — but in his own way, Kane is guilty of this too. To be fair, neither Kane, nor anyone else, can help but make presumptions. To go back to Auden, this is the unavoidable consequence of interpreting the unfinished works of a dead author. Kane laments virtually every omission Christopher made, but surely Tolkien would not have published everything he drafted? At one point, Kane cites Verlyn Flieger, quoting Tolkien’s own words, that “it is the untold stories that are the most moving” (116). Later, Kane points out that Tolkien himself began a sixth version of the Beren and Lúthien story, designed to be a “somewhat more compressed text” than the fifth, “abandoned because it was getting too long” (173). Do not these clues from the author justify some of the omissions by the editor?

In addition to calling for the reinstatement of virtually every omission, Kane argues almost universally that Tolkien’s latest writings should trump all others, but should they? There are well-known counterexamples. Perhaps most famous is the story of the Sun and Moon, where Christopher retained the earlier version despite intimations, and even drafts, of a major rethinking late in Tolkien’s life. Kane — as I think most readers do — praises Christopher’s retention of the earlier telling. Could there have been other situations (other than those where Kane concurs with Christopher’s judgment) where the earlier was the better draft? It is difficult to know where to draw the line.

Kane occasionally compliments Christopher on his solution to a particularly thorny problem (63, 151, 175, 188), but much more often, he is critical of how Christopher “cannibalized” (92) his father’s works to produce The Silmarillion. (I should point out that it was Christopher who first used “cannibalized” to describe his father’s own writing habits, but in Kane’s application of the word to Christopher, it can’t help taking on a more judgmental tone.) Kane often acknowledges that Christopher had few choices for resolving these manifold difficulties, but he takes Christopher to task for not adopting a more inclusive (or “maximalist”, as I have called it) approach. But Christopher indeed considered something like this. “[F]or a time,” Christopher wrote in 1977,

I worked toward a book that would show something of this diversity, this unfinished and many-branched growth. But it became clear to me that the result would be so complex as to require much study for its comprehension; and I feared to crush The Silmarillion beneath the weight of its own history. I set myself, therefore, to work out a single text, by selection and arrangement. (Christopher Tolkien [4])

I think Kane (like Charles Noad, and no doubt others) wishes for something in between the two extremes: something in between the many-branched tree, so over-grown that it collapses under its own weight, as I daresay some would describe The History of Middle-earth; and the tree pruned of many of its youngest branches and loveliest leaves, as Kane might describe Christopher’s Silmarillion. For myself, I would tend to give Christopher the benefit of the doubt on most matters (as all trees need pruning), but Kane is absolutely justified in questioning what seem to be a large number of small and arbitrary omissions.

But Kane’s opinion was shaped in the hindsight of The History of Middle-earth. Had Christopher followed Kane’s more “inclusive” strategy from the beginning, it is entirely possible that The Silmarillion would have been a commercial failure — perhaps such a failure as to have prevented any subsequent material from ever reaching readers. Kane realizes this, but his arguments carry a note of assumed authority which I do not feel is entirely appropriate. Let us not forget that “The Silmarillion,” in much the same raw form as Christopher had to work with, was rejected by two publishers during Tolkien’s lifetime. Tolkien’s reputation by the time of his death was the main reason for its publication in 1977, but Christopher and Guy Gavriel Kay’s work to assemble a single text (though not without mistakes, we know) was a large part of its success. And only a modest success, at that.

And what a task it must have been, to assemble a single text from such raw materials! “Dizzying,” a word Kane uses more than once (63, 117), aptly characterizes the complexity of the Silmarillion text and its relationship to its many precursors. One of the most useful features in Arda Reconstructed is its twenty-five tables, detailing the sources, paragraph by paragraph, from which Christopher assembled the work for publication. In rare cases, particularly in the last chapters of the Quenta Silmarillion, Kane is unable to identify the source of a sentence or paragraph. He concludes either that Christopher consulted a text as yet unpublished, or else that he introduced the material editorially (Christopher has acknowledged several such inventions explicitly). The tables provide an easily navigable guide for intrepid readers who would follow in Kane’s footsteps, as he followed in Christopher’s. Thus, readers may be allowed to judge Kane’s conclusions for themselves.

For his base text, Kane uses the second edition of The Silmarillion (1999), yet he never says anything in his book about the changes from the first to the second edition. A systematic guide to these changes, however few and small they may be, remains a desideratum. Perhaps we may see this as an appendix to a revised edition of Arda Reconstructed. If not, we will have to wait for the revised edition of Wayne Hammond’s Descriptive Bibliography (in preparation). Here is my point: one can, and should, ask — but Kane never does — why, having come to regret some changes and omissions over the course of preparing Unfinished Tales and The History of Middle-earth, Christopher never made such corrections (reversing alterations, reinstating omissions) as could be done without significant modification. If anything, Christopher’s work on The History of Middle-earth would seem to have been the ideal preparation, and justification, for a genuine revised edition of The Silmarillion.

Throughout Arda Reconstructed, Kane takes the most pain to point out when Christopher has omitted passages in the source material from the final published text. Often, the wisdom of such omissions is a matter for legitimate debate, as in Christopher’s choice to remove the device of the narrative frame and its satellite characters, Rúmil, Ælfwine, and Pengoloð (36, et passim). But Kane spotlights many instances where omissions seem to have been both unnecessary and detrimental to the final text. For instance, Christopher omitted a fuller and more powerful account of Melkor’s attack on Formenos, with the much grislier details recounted to Fëanor by his own son (106-7). So too, the actual words of the Oath of Fëanor, of which Kane finds it “remarkable [...] that Christopher chose to leave out this incredibly powerful text [...], and to replace it with the older version in which the oath is simply described in bald terms” (111). I agree. And this might have been an opportune moment for Kane to refer readers to the “Sí Qente Fëanor” text, another extant version of the actual Oath, published in the linguistic journal, Parma Eldalamberon. This is a very early, but equally interesting, prose passage, representing Fëanor’s Oath in Qenya, the earliest form of the High-elven language.

One objection I would raise in Kane’s endless cataloguing of Christopher’s omissions is that he too rarely takes the time to consider why they might have been made. Instead, again and again, he “cannot imagine why” — or some variation thereof — Christopher would have cut whatever it is he cut (72, 90, 91, 96, 109, 140, 161, 166, 179, 212, 213, 235, et passim). Arda Reconstructed would have benefited from Kane’s putting more effort into trying to imagine why — that is, considering what legitimate reasons there could have been for each omission. This would have strengthened those cases where there genuinely does not seem to have been any good reason. There are a few instances where Kane does dig into the matter, attempt to see both sides, and then register his opinion; had this been the rule and not the exception, his conclusions would have carried more persuasive weight than they do.

Two of the most controversial of Christopher’s cuts are the Second Prophecy of Mandos (236-9) and the fuller account of the story of Finwë and Míriel (75-6). The latter, Kane sees as only one example in a larger trend of reducing the roles of female characters in The Silmarillion: “There are at least eight female characters whose role or character could be said to be reduced to a greater or lesser extent by the editorial decisions made [by Christopher]: Uinen, Galadriel, Míriel, Nerdanel, Indis, Ungoliant, Arien, and Nellas (in addition to the removal of the two or three daughters of Finwë and Indis, of Baragund and Belegund’s older sister, Beleth[,] and of Andreth from the Athrabeth)” (252). Of all the changes Christopher made, this is “perhaps [Kane's] biggest complaint” (26). But if the changes and omissions Kane describes do in fact constitute a purposeful reduction, then there is just as much reduction of the male characters (and almost certainly more). For Kane to call attention to only the female characters in this way — and to impute a motive to Christopher to actively reduce their presence in the narrative — strikes me as either disingenuous or careless.

Kane decries also the excision of short passages of description and characterization, and here I tend to agree. Had Christopher retained such passages, this might have helped to mitigate criticism of the spartan nature of The Silmarillion, especially as contrasted with The Lord of the Rings. One early review, for example, held that ”

here [in The Silmarillion] Tolkien cares much more about the meaning and coherence of his myth than he does about these glories of the trilogy: rich characterization, imagistic brilliance, powerfully imagined and detailed sense of place, and thrilling adventure. [...] Numerous characters here have interest, [...] and though each important character has his fascinating quirks, the compression of the narrative and the fierce thematic focus give Tolkien no room to develop and explore those quirks as he does in the trilogy. (Gardner)

Many such descriptions and images as Gardner finds wanting were in fact written by Tolkien, but omitted by Christopher, and the compression Gardner alludes to was largely imposed by Christopher. Such objections raise once more the question of the viability of an uncompressed volume (as well as the question of whether Tolkien ever intended it to be like The Lord of the Rings) — but by now the point may be, as they say, academic.

Kane also highlights a few inventions by Christopher. Two chapters, “Of Aulë and Yavanna” and “Of the Sindar,” were constructed by Christopher out of ancillary material by his father but were never part of any version of the Quenta. More (in)famously, there is the matter of the Nauglamîr (141-2) and “The Ruin of Doriath” (207). Regarding the latter, Kane observes that “Tom Shippey cites [one passage] as an example of Tolkien’s genius for creating compelling images. Yet, as we have seen, Thingol’s death in the dark recesses of Menegroth was completely an invention of the editors” (216). Though Kane is circumspect enough, pointing out Shippey’s “mistake” is perhaps insensitive. Handled just a bit more carelessly, it could have appeared that Kane was making a fool of Shippey, implying gullibility. Alternatively, one might say that if Shippey could be “taken in” (as of course, we all were), it is a sign that Christopher’s invention was of a quality the equal to his father’s. Kane dodges the bullet — just. And let me note here, since I have quoted Kane referring to the “editors” (plural), that throughout Arda Reconstructed, Kane accords Guy Gavriel Kay equal status to Christopher as a co-editor. Though probably no one but Christopher and Kay really know the extent of the latter’s assistance, Kane is clearly overstating his role. It would have been better not to inflate Kay’s contribution without clear cause.

Summing up, I find Arda Reconstructed to be a meticulously researched and valuable new reference work (one of all too few) on The Simarillion. If I have been hard on it, take that for engagement with the book and its author’s ideas, and not as discouragement to potential readers. Moreover, it has the added benefit of approaching the work from the relatively new angle of considering Christopher’s role as a vigorous editor, and Kane is to be congratulated for confronting the matter directly. He presses Christopher hard on many points, even candidly questioning his motives and judgment in a couple of cases (98, 239). He sometimes goes too far, but on balance, I find much of his criticism valid, and most of his questions worth asking. Even when his reach exceeds his grasp, at least he is reaching in interesting new directions. His study also throws a brighter light on just how complex the underlying texts and their interrelationships really are, and how Herculean a task Christopher faced in bringing these inchoate works to a larger audience, both in The Silmarillion and fourteen subsequent books. It is a tight and functional abridgement of much of The History of Middle-earth itself — an abridgement, but not a replacement. Finally, it is a blueprint to another possible “Silmarillion” (one I might actually like to read!) — and a roadmap to further exploration in that mythopoeic space.

Works Consulted

Content copyright 1967- The Mythopoeic Society All rights reserved