Ox Bones and Silver Ladles: The Construction of the Ainulindalë by Dawn Walls-Thumma

Posted on 10 April 2021; updated on 2 May 2021

| | |

This paper was presented at the Tolkien at UVM conference held virtually on 10 April 2021.


Eight years ago, when I first started researching Tolkien's creation myth the Ainulindalë, most of the discussion I encountered, in both scholarly and fan spaces, could be summed up under the heading: Where exactly does the Ainulindalë come from? In other words, what were the myths it was "based on" or that Tolkien drew from as his "sources"? These questions presupposed that there was a source, and indeed, I read articles from scholars investigating roots in the Biblical, Classical, and Northern traditions. In fact, at my very first Tolkien conference, where I argued that there was no single source for the Ainulindalë, I ended up presenting opposite a scholar claiming its source was in Genesis. (I think it may have been his first Tolkien conference too. We pointedly and awkwardly avoided directing any too-hard questions at each other, and our audience mercifully responded in kind.)

Yet when I say that "there was no single source for the Ainulindalë," I don't mean that it doesn't have strong and undeniably mythic roots. In fact, I'd go so far as to say that those roots give the Ainulindalë a lot of its power. Rather, what I mean is that Tolkien held no single myth as a model for the Ainulindalë. And that, too, is part of its power: It is not a subsidiary of another culture's myths, it's not a derivative, but rather it draws from the same deeper mythic source as our real-world creation myths do, giving the Ainulindalë a resonance and meaning apart from that imparted by its connection to another culture's mythology.

And this meaning imparted by the Ainulindalë, independent of its association with a source myth, is important. As a creation myth, the Ainulindalë forms the foundation upon which the rest of the legendarium is constructed. Barbara Sproul calls creation myths "self-fulfilling prophecies" because "they create facts out of the values they propound."1 In the case of the legendarium—while undoubtedly influenced by Northern, Classical, and Christian traditions—those values are uniquely its own, beginning with its creation myth, the Ainulindalë.

To begin, it's worth considering what Tolkien himself said on the subject. Tolkien’s own sense of the sources of his legendarium are complicated. Even as he lamented in a letter to a Mr. Wrigley that “the search for the sources of The Lord of the Rings is going to occupy academics for a generation or two,” his essay “On Fairy-stories” introduces such concepts as the “Cauldron of Story” and “Tree of Tales” that presuppose a common origin for all stories—and in fact assign some of a story’s power to its intertextuality.2 Likewise, his own legendarium is clearly intertextual at points. But again, he is drawing from a more nebulous store of "mythic elements"—archetypes, if you will—and not from a single source. Tolkien himself phrased it best in his 1951 letter to the publisher Milton Waldman: "These tales are 'new', they are not directly derived from other myths and legends, but they must inevitably contain a large measure of ancient wide-spread motives or elements."3 Returning again to the letter to Mr. Wrigley, Tolkien illuminates his method in selecting these elements, noting that "the particular use in a particular situation"—not the element's association with a specific myth—is what imparts meaning to that element's use in the legendarium.4

It is with these "ancient wide-spread motives or elements" where my interest lies, specifically in considering how Tolkien selected and assembled "the particular use" of elements to simultaneously root the Ainulindalë in the broader world mythic tradition while allowing it to carry its own meaning and values independent of other myths. In particular, I'm going to consider two specific mythic elements: the deus faber or god-as-artisan motif and how that motif is used in world mythology, because the theme today is Tolkien and the Classics, the demiurgic motif with a specific focus on Plato's Timaeus.

In 1963, the religion scholar Charles Long classified creation myths as five types: creation from nothing, creation from chaos, world parent, emergence, and earth diver.5 Tolkien's Ainulindalë, which was first set to paper sometime between 1918 and 1920 as The Music of the Ainur, is perhaps the most enduring text in the legendarium, changing relatively little as the decades brought significant revisions to all other areas of his work. To use Long's system, the Ainulindalë is a creation-from-nothing myth: an important concordance with the Hebraic scriptures, since Tolkien's Christian God and Arda's Eru Ilúvatar are the same being. The ability to create from nothing is a distinguishing characteristic of God (or Eru, if you will) that becomes essential later in distinguishing between Tolkien's concepts of creation and subcreation.

The Ainulindalë contains an essential element that distinguishes it from the Hebraic scriptures: the Ainur, created from the thought of Eru within the universe. Here is where the Ainulindalë begins to align more with the elements presented in Plato's Timaeus. Timaeus presents the creation of the universe as demiurgic: carried out by divine beings themselves created by a monotheistic deity. Much as the Ainur are entrusted to make "a new World made visible before them," the demiurgic gods in Timaeus construct the various beings of the world. This places much of the universe at a one-step remove from the ultimate creator god, which may seem on the surface to lessen the power of that god. However, as Tolkien himself notes in his essay "On Fairy-Stories," the effect is to attach to the creative, artistic impulse the mark of divinity. Because God creates, in other words, so do His creations, a phenomenon that Tolkien terms subcreation, summed up in "On Fairy-Stories" as, "we make still by the law in which we're made."6

Building on Long's classification system of creation myths, David Adams Leeming in 2010 further classified world creation myths by motif. The motif described above, at the heart of the Ainulindalë, he terms the deus faber motif, or god-as-artisan. In this motif, the creator god is equivalent to a human artisan, and the creation of the universe, world, or humankind parallels a specific art or craft, such as pottery or tentmaking. Leeming suggests that the deus faber motif adds significance to human creativity by its association with the divine. "The deus faber creation," he writes, "is a celebration of human ingenuity and a justification for what we do. And it turns what we do into something mystical and magical. It makes our crafts microcosmic representations of creation itself.7

Interestingly, the deus faber motif is almost nonexistent in Western culture, though it occurs in myths from the Americas, East Asia, Africa, and Oceania. For Western examples, Leeming cites only a passage from the Book of Job where God describes laying the foundations of the earth, making him a builder in Leeming's interpretation—a bit of a stretch in mine—and a Romanian myth where creation is achieved through weaving.8 Western myths tend to rely on the creation-from-nothing and creation-from-chaos types—with the exception of the Norse, who in their very Norse way, chop up a giant and use his dismembered parts to make the world. This makes Tolkien's own use of the deus faber motif interesting because it holds no parallels with the myths fans and scholars typically like to credit as sources for the legendarium, and the Ainulindalë more particularly.

Another intriguing deviation is the type of artistry represented in deus faber myths. Almost always, they involve some form of physical craft: pottery, weaving, or smithcraft, for example. But Tolkien's use of art as the vehicle by which the universe is created comes at least initially through the mode of music and song, a mode of creation almost entirely absent from world mythology. There is creation-from-word, a motif associated with the creation-from-nothing type, but that is not exactly what happens in the Ainulindalë either, at least where the Ainur are concerned, leaving the Ainulindalë as a bit of both motifs but wholly neither.

Returning to Tolkien's assertion that "the particular use in a particular situation" of an element or motif is the most interesting consideration, it's worthwhile to consider his particular use of these particular motifs. As noted earlier, they root the Ainulindalë within a broader mythic tradition. Thus, the Ainulindalë feels genuine as a creation myth, which is why I suspect so many Tolkien fans feel like it must "come from" somewhere. However, Tolkien's small deviations from that tradition—namely the use of music as the mode of demiurgic creation—are significant to the meaning of the Ainulindalë, both as a standalone text and as the mythic foundation of the legendarium.

Music and song are significant as modes of creation for two reasons. First, they tie into how the Northern cultures Tolkien loved traditionally represented and made sense of their own history and reality. The Anglo-Saxons, the Norse, the Finns—all of these cultures, in ancient and medieval times, relied on poetry and music as a mode within the oral tradition by which the past was remembered and interpreted for a modern audience. Especially considering Tolkien's original aim to represent a pre-Christian mythology for England—an objective that, while ostensibly abandoned, remained vestigial at points within the legendarium—the reliance of these cultures on music and song makes his choice of song as the creative mode significant. If, in his original imaginings, the Ainur of the 1918-1920 Music of the Ainur were at the deepest origins of English myth, the sacred nature of song as a mode of literally creating reality carries forward and imbues the figurative creation of reality by historical figures such as the scops and skalds of Germanic tradition with the coloring of the divine. This aligns with the power that these figures held within Germanic society: far from mere entertainers, as we tend to think of singers today, but trusted with the power to make sense of the world through their selection and presentation of historical details—the influencers of both kings and their followers.9 Perhaps somewhat romantically, one can also imagine Tolkien first encountering the echoes of these songs as written down by Christian scribes and the imaginative universes they kindled for him, in turn represented in a legendarium itself established via a creation myth using song: his concept, described in "On Fairy-Stories," of the Secondary World that becomes akin to reality while one is inside of it.

Secondly, the use of music and song, as Leeming notes of the deus faber motif, associates those arts with the divine. In the context of the Ainulindalë, I'd extend this to narrative. While the Ainulindalë is described as "like unto harps and lutes, and pipes and trumpets, and viols and organs, and like unto countless choirs singing with words," invoking both instrumental music and vocal song, the introduction of conflict via Melkor and the resolution of that conflict as part of the song itself suggests a narrative structure.10 This elevates Tolkien's own work as a poet and storyteller to an impulse derived from his Creator. As Leeming notes, deus faber myths offer "justification for what we do": We are not simply fiddling about with clay or paint or music but imitating the same divine mode used by God Himself. Given Tolkien's frequent self-deprecation of his creative work, his choice of song as the artistic mode for creation always feels more than a little self-validating to me: He is, after all, only "making still by the law by which he's made."

Within the deus faber motif, however, Tolkien also uses demiurgic creation—or subcreation, in his own terminology—similar to that used by Plato in the Socratic dialogue Timaeus. Timaeus is, to put it bluntly, a weird text, involving an idiosyncratic and suspect mix of cosmogony, the four elements, physiology, triangles, taxonomy, casual sexism, medicine, more triangles, astronomy, and theology. The translator of the version I used called Timaeus "obscure and repulsive" to modern readers, which this reader found a bit strong but not entirely inaccurate.11 Certainly, there are sections—mostly those where the word "triangle" is used more than three times in a paragraph—that I have struggled unsuccessfully to attend to despite reading Timaeus more times than I can recall over the years. I will focus on just a small part of Timaeus, which concerns the creation of the universe and world. No triangles, promise.

First, the dialogic structure is very similar to Tolkien's earliest work on his creation myth, The Music of the Ainur. There are multiple speakers summoning a distant past that has faded into obscurity for most of its youthful and ignorant audience and which is being rekindled via the oral tradition. Timaeus describes several world-destroying deluges, similar to the multiple cataclysms in The Silmarillion: the overthrow of the Lamps and the inundation of Almaren, the War of Wrath and the inundation of Beleriand, and the destruction of Númenor and the inundation of, well, Númenor. Speaking of which, the origins of the Númenor story—the famous if roundabout result of a creative challenge between Tolkien and C.S. Lewis—can be found in Timaeus as well, where Plato first presents the story of the drowned, overproud isle of Atlantis. Atlantis, of course, was the prototype of Númenor, one of the few instances of an unequivocal source for part of the legendarium. John Cox notes as well that Tolkien's frequent use of series of objects or people, diminishing in each successive iteration, owes to the same pattern seen in Timaeus, where distance from the original, eternal Creator produces increased corruption.12

Demiurgic creation is one such succession seen in The Silmarillion: the eternal, omnipotent creator Eru Ilúvatar; the immortal but highly fallible subcreators, the Ainur; and the attempts the Ainur make at creating life, described as imitations in the case of the Dwarves, perversions in the case of Melkor's fabrications, and subcreative footnotes in the case of Yavanna's olvar and kelvar.

There are other cosmogonical parallels as well between Timaeus and the Ainulindalë. Both are monotheistic, involving an eternal creator who sets the universal stage for the work of his demiurges: the Greek gods in Plato and the Ainur in Tolkien. The creator in both texts vacates the world, leaving his demiurges in charge, in contrast to the monotheistic (and highly, wrathfully involved) creator god in the Hebraic tradition. In Timaeus, the creator says to his demiurges, "I will myself sow the seed, and having made a beginning, I will hand the work over to you."13 In The Silmarillion, the absence of Eru Ilúvatar is made apparent when Manwë Súlimo must twice pray for Eru's intercession, to grant Lúthien Tinúviel mortality and to unmake the world when Ar-Pharazôn breaks the Ban of the Valar; otherwise, it is understood, the governance of Arda is left to the Ainur who accepted a place within it.

In both texts, the demiurges operate within a structure imposed upon them by the creator. Timaeus describes creation as based "after an unchangeable pattern,"14 which recalls the structure of the Music of Ainur—and Melkor's introduction of evil in his deviation from it—followed by Eru's words to Melkor that "no theme may be played that hath not its uttermost source in me, nor can any alter the music in my despite,"15 a concept that, though phrased less eloquently, extends back to the earliest draft of The Music of the Ainur. Timaeus doesn't introduce a conflict among the demiurges as Tolkien does between Melkor and the other Ainur; however, the creator does state that "only an evil being would wish to undo that which is harmonious and happy,"16 which seems to prefigure exactly Melkor in The Silmarillion. Demiurgic creation is the most distinct element of Plato's cosmogony as presented in Timaeus, and many of its features are present in the actors and process of creation in the Ainulindalë.

As with the deus faber motif, however, I am less eager to declare the Ainulindalë as derived from Timaeus than to consider why the demiurgic element of Timaeus may have appealed to Tolkien and met his needs as he [sub]created his own cosmogonic myth. After all, the use of the deus faber motif doesn't require demiurges. Tolkien could have done away with the Ainur and had Eru Ilúvatar step out from behind the conductor's podium and sing the universe into existence himself. John Cox proposes that the appeal of demiurgic creation lies in the "buffer" it provides between the ultimate and eternal Divine and the grosser, more worldly elements of creation, but as noted earlier, if anything, assigning the full creative act to Eru Ilúvatar makes the composition of poems, songs, and stories more closely affiliated with the divine.17 I would suggest instead that the demiurgic structure possibly appealed to Tolkien for the simple reason that it allowed him to construct a monotheistic Secondary World—essential given his Christian beliefs and the early association of the legendarium with English mythology—that also included the kind of paganistic pantheon so essential, and delightful, in the myth cycles Tolkien most enjoyed.

There is one essential deviation in the demiurgic structure of the Ainulindalë when compared to Timaeus that shows that Tolkien was not engaging in rote imitation so much as selecting and manipulating elements that advanced his purpose. In Timaeus, as noted earlier, the demiurges are the immortal product of the creator god. However, the other organisms of the world—including humankind—are made by the demiurges, not by the creator. The creator justifies this on the basis that "if they were created by me and received life at my hands, they would be on an equality with the gods."18

It is precisely this equality that, in subverting Timaeus here, Tolkien strives to achieve. Both Elves and Mortal Men are pointedly not created by the Ainur but by Eru Ilúvatar himself; not only that, but they are so firmly in Eru's province that much about them and their fate remains unknown to the Ainur. This change is essential, for Elves and Mortal Men must be the Children of Eru in Tolkien's cosmogony. As noted earlier, since Tolkien imagined Arda as a mythology of our own world, certain theological elements had to remain consistent with his beliefs as a Christian, with the primacy of a monotheistic creator god and humankind as His creations without intermediaries.

In "On Fairy-Stories," Tolkien describes what scholars today would call intertextuality, only he calls it "the Soup" or, more eloquently, "the Caldron of Story." This metaphor, which he carries over several pages, reveals much of what he believes about the use of mythic and other story elements within his own work. He cautions that "We must be satisfied with the soup that is set before us, and not desire to see the bones of the ox out of which it has been boiled." But, earlier, he asserts that "the Cooks do not dip in the ladle quite blindly,"19 which implies that Cooks—artists, musicians, storytellers—must by necessity see the bones of the ox. They are, presumably, pardoned for their perverse curiosity in service of their craft.

I hope Tolkien will forgive me for my scholarly interest in ox-bones as well—and the myriad other ingredients mystical and strange within the Caldron of Story that is the realm of creation myth. I have shown how, in his construction of the Ainulindalë, Tolkien dipped the ladle with discerning intent, beginning with a monotheistic creator required by his own Christian faith, overlaid with a pantheon of demiurgic subcreators in the Platonic tradition, and all folded into a deus faber motif as seen in myths around the world. For all of these, Tolkien modified—seasoned, if I may extend his metaphor even further—these ingredients to suit his specific purpose.

Here, I think it's worth returning to Barbara Sproul, whom I quoted at the beginning for her description of creation myths as "self-fulfilling prophecies" from which the details of a myth cycle arise. The Ainulindalë, with Tolkien's careful selection of mythic elements, forms the foundation of the legendarium, with its primacy given to artistic creativity and especially song: where Light derived from Eru himself is sung to life in the form of the Two Trees, where the greatest villain Melkor is subdued not with a sword but a song, where the humble Ringbearers come to understand their purpose not in the role of heroes but as a place within a story, within a song.

Works Cited

  1. Barbara C. Sproul, Primal Myths: Creating the World (New York: Harper & Row, 1979), 3
  2. The Letters of J.R.R. Tolkien, "337 From a letter to 'Mr Wrigley' 25 May 1972."
  3. The Letters of J.R.R. Tolkien, "131 To Milton Waldman."
  4. The Letters of J.R.R. Tolkien, "337 From a letter to 'Mr Wrigley' 25 May 1972."
  5. Charles H. Long, Alpha: The Myths of Creation (New York: George Braziller, 1963).
  6. The Tolkien Reader, "On Fairy-Stories."
  7. David Adams Leeming, Creation Myths of the World: An Encyclopedia, 2nd ed. (Santa Barbara, CA: ABC-CLIO, 2010), 321.
  8. Leeming, Creation Myths, 319.
  9. Morton W. Bloomfield and Charles W. Dunn, The Role of the Poet in Early Societies (Cambridge: D.S. Brewer, 1989).
  10. The Silmarillion, Ainulindalë.
  11. Benjamin Jowett, "Introduction and Analysis" to Timaeus.
  12. John Cox, "Tolkien's Platonic Fantasy." SEVEN: An Anglo-American Literary Review 5 (1983): 59.
  13. Plato, Timaeus, trans. Benjamin Jowett.
  14. Ibid.
  15. The Silmarillion, Ainulindalë.
  16. Plato, Timaeus.
  17. Cox, "Tolkien's Platonic Fantasy," 58.
  18. Plato, Timaeus.
  19. The Tolkien Reader, "On Fairy-Stories."

About Dawn Walls-Thumma

Dawn is the founder and owner of the SWG. Like many Tolkien fans, Dawn became interested in Middle-earth thanks to Jackson's Lord of the Rings films, but her heart was quickly and entirely won over by The Silmarillion. In addition to being an unrepentant fanfiction author, Dawn is an independent scholar in Tolkien and fan studies (and Tolkien fan studies!), specializing in pseudohistorical devices in the legendarium and the history and culture of the Tolkien fanfiction fandom. Her scholarly work has been published in the Journal of Tolkien Research, Transformative Works and Cultures, Mythprint, and in the books Not the Fellowship! Dragons Welcome and Fandom: The Next Generation. Dawn lives on a homestead in Vermont's beautiful Northeast Kingdom with her husband and entirely too many animals.


What a fabulous article, Dawn! I wished I had read it when I was in college, still writing my dissertation about Greek mythology and its parallels with Tolkien. I had no notion, then, of the deus faber motif - but it would have improved my research a lot lol.

I especially liked re-reading this passage: "'We must be satisfied with the soup that is set before us, and not desire to see the bones of the ox out of which it has been boiled.' But, earlier, he asserts that 'the Cooks do not dip in the ladle quite blindly,' which implies that Cooks—artists, musicians, storytellers—must by necessity see the bones of the ox. They are, presumably, pardoned for their perverse curiosity in service of their craft."

Although the reader won't want to see the bones of the ox, we, as storytellers, are obliged to show them the reminiscent flavors, something that will still remind them that the Soup was made of more than just the bones. I really like to think this was Tolkien's intent through and through :)

I hope it is OK to comment a little on the background to Tolkien's use of Timaeus by way of a note on your translator...

Benjamin Jowett was one of the most eminent of Victorians and the most important figure in the history of the universities in that era. Back in his day, Oxford and Cambridge were only for Anglican men. Jowett was a leading light among the so-called 'liberal Anglicans' who opened up England's ancient universities to women, dissenting Protestants, Jews, and Catholics - without their efforts Tolkien would never have taken a degree let alone become a professor at Oxford. In addition, Jowett made his college, Balliol, the intellectual powerhouse of late-Victorian Oxford. His translations of Plato were for a long while definitive, but more importantly they helped redefine what a liberal education at Oxford was all about. 

So when an Oxford man even in the 1930s turns to Plato, he is invoking the spirit and the project of Jowett. What Tolkien was doing with Plato should be read, in part, against the background of how Jowett's translations of Plato were woven into the fabric of various Oxford approaches to life, the universe, and everything. In this case, what Jowett did in the first instance was simply establish the canon, so that Republic, Timaeus, and Critias were noted to follow dramatically on three consecutive days, thus bringing into view a single rounded project in which Plato first outlined his ideal political state, then justified it, first by showing how it was conceived according to the same principles by which the natural world was given form, and then by showing those same principles in operation in history. From this trilogy the conclusion was: (a) the Republic must be read seriously, and as a foundation even for modern political thought (which becomes strong at Baliol), (b) the Timaeus is in the first instance an early attempt to do 'science' (hence all the triangles) that soon came to look monotheistic, and (c) the Critias is an attempt to do philosophy of history, with the unfinished nature of the Critias pointing to the fact that Plato did not get to grips with History. This last conclusion seems the least important, but it was what opened the door to the massive influence of Hegel in late-Victorian Oxford and a profound obsession with the philosophy of history.

When Tolkien touches on the Atlantis story, and indicates that he is so doing (the primary subject of the unfinished Critias but told also in shorter form as a prelude to Timaeus' cosmological monologue), he is - in this context - touching on issues still associated in Oxford with questions pertaining to the philosophy of history. (He seems here quite close to his Pembroke colleague R. Collingwood). And the fact that he so completely reworks Plato's myth of Atlantis - as it were extracting it from the ancient Greek canon and fitting it out for a revamped northern canon - suggests that he is in some sense confident that he, in contrast to Plato, is now making the myth work (an audacious idea!), that with the help of the northern tradition he has fitted the myth rightly into an idea of History.

On the Timaeus and Tolkien's creation myth, however, it seems to me more tricky. Jowett's Oxford did not hold that Plato's cosmology was wrong in the same way as his philosophy of history was wrong. The latter was an invitation to Hegel and speculations about the meaning of History. But Plato's cosmology was not only scientifically antiquated beyond any hope of redemption, the Timaeus has a peculiar relationship to monotheist thought, and much of what can be found in it (rightly or wrongly) went into the making of Christian theology. None of this undermines the idea that the Ainulindale draws on the Timaeus. But it does complicate the attempt to picture what Tolkien might have thought he was doing in drawing on the Timaeus.

All of which is a very long way of getting to a question. Is there any indication of a change in the creation myth before and after Numenor first appears (in 1936)? Because if Tolkien returns to the Timaeus in 1936 and so systematically reworks Plato's story of how the world was made round, one might expect the work to have generated some related re-conception of the music and the original making.

 

Hmmm. Probably my previous comment appears 'scholarly' or 'critical' in a bad way, and there would be some reality in that appearance because the basic idea that the Timaeus is a source of the Ainulindalë caught me by surprise, and intersects curiously with my own research on the Atlantis story of the Timaeus. As may be seen, I have no good reason for denying this idea. What is more, I feel compelled to take it very seriously, because if it is the case then I need to look at the Numenor reworking of the Timaeus differently - as a return as much as a new direction, which latter is what I have always presumed. 

So the above comment has its source ultimately in my own surprise and confusion. Returning some days later, I still do not see clearly, but wish to reiterate that, in Tolkien's reading of Plato in 1936, what is key is the initial distinction made by Timaeus between what does not change and can be known, and is the object of philosophical knowledge, and what changes and so cannot be known, philosophically speaking. This division in effect denies the possibility of natural science as a road to knowledge but opens the door to natural science as myth, defined philosophically as the most rational account that can be given of that which is inherently irrational. Tolkien's reworking of the two myths - Atlantis and the making of a round world - seems to me bound up with a reworking of Plato's notion of myth, which becomes in his hands something quite different (a tale that relates our historical world to an immortal realm, not the eternal realm of the philosopher and the theologian), and I have taken it that this notion of Myth is new to Tolkien in 1936 (and the other side of an equally new notion of History). But again, it is more than likely that Tolkien was in 1936 returning to the source, so to speak, and that I have simply failed to notice this.

On Jowett and Liberal Anglicanism I'd just like to add, in this postscript to a comment, that I feel defensive for these dead men of the Victorian establishment. They made possible the modern English system of higher education (it was on their watch that most of the provincial universities of the country were founded), but when that university system came into its own in the twentieth century it was embarrassed about its own clerical past and tried to hide and ended up rediscovering many of the older concerns only by way of reading Continental philosophy in the second part of the century. 

PPS. Here is a quotation from W.P. Ker, the great critic of the 1936 lecture on Beowulf, who first formulated the 'theory' of northern courage. He is here discussing Hegel and 'The Philosophy of History' in 1925. I copy out this quotation not to prove or disprove anything but because Ker was a product of Jowett's Baliol and so gives an insight into the Oxford tradition on the Timaeus out of which Tolkien steps.

The Philosophy of History has the same place in his [Hegel's] works as the Critias of Plato, after the Republic and Timaeus. The Theory of Justice and the Theory of Nature are followed by the story which shows the adventures of Justice in space and time. There are many things besides philosophy in Plato's story of Atlantis, and the same sort of freedom may be allowed to Hegel.

 

the basic idea that the Timaeus is a source of the Ainulindalë caught me by surprise

It might be a quibble but I don't think Timaeus is a "source" for the Ainulindale. My very first conference presentation, now ten years ago, was on creation mythology and made the case that Tolkien didn't use any existing creation myths as sources for his Ainulindale. I take him at his word that his stories are new but drawing from elements and themes found in other myths/stories. My view here has not changed substantially, though I've shifted my interest from cosmogony to focus elsewhere. (This paper resurrected this old area of interest of mine for Tolkien at UVM because the theme that year was "Tolkien and the Classics," which is a bit out of my wheelhouse ... but creation myths I can do!) Tolkien's myths/stories succeed, in my view, because they are not imitative but deftly weave in enough mythical elements to feel authentic in a way that an imitative work would not, nor an "original" story/myth.

So do I think there are elements that are in common between Timaeus and Ainulindale that Tolkien used because they serve his larger purpose in the Ainulindale? Absolutely. The idea of subcreation, the deus faber motif, and the "corruption" of creativity as it moves away from Eru are all important concepts in Tolkien that, to my reading (and that of others who have viewed the Ainulindale through a Platonic lens), are also found in Timaeus. But do I think that Timaeus is "a source" for the Ainulindale, implying that Tolkien read it and intentionally decided to select these elements for use in his own cosmogonical myth? Absolutely not. Maybe he did! I certainly can't say that he didn't. But with my own origins in the humanities in creative writing, I also know that fiction writers are magpies, and our brains are constantly littered with details and ideas from everywhere that sometimes end up deployed in our own creative work. Tl;dr: I don't take correspondences or connections between texts as an implication that one is the source for another. It is just as likely that the ideas of demiurges and creators-as-artists were selected by Tolkien because they work in his legendarium without any conscious recognition of a source, even as we know he would have been familiar with Timaeus.

This division in effect denies the possibility of natural science as a road to knowledge but opens the door to natural science as myth, defined philosophically as the most rational account that can be given of that which is inherently irrational. Tolkien's reworking of the two myths - Atlantis and the making of a round world - seems to me bound up with a reworking of Plato's notion of myth

I think this could be a fascinating line of inquiry! I can't comment further on specifics, but it seems like there might be some intriguing connections here.

All of which is a very long way of getting to a question. Is there any indication of a change in the creation myth before and after Numenor first appears (in 1936)?

Without pulling out the multiple texts that make up the Ainulindale again, my answer would be no. I did a line-by-line comparison across drafts many, many years ago now, and what has stuck with me, over years of working with the posthumous "Silmarillion" texts, is the consistency across many drafts, which is not typical of Tolkien's work. But, given that the Ainulindale is a creation myth, this is not unexpected either, if one accepts that the creation myth is the foundation of the myth cycle beyond it, making it hard to change once those other myths are in place. In the particular case of the Ainulindale, subcreation and the importance of language/song are concepts that occur over and over again in the legendarium, so it makes sense (to me) that these prominent elements would need to remain in the creation myth, and Ainulindale serves nicely in that regard.

Well, thank you for the explanations. I do apologize for jumping in with a sledge-hammer. The fact is that I have to rethink my own framework in light of what you say here, which caused a minor freakout on my part. But now the dust has settled i see clearly that I must consider to some degree how Tolkien might have encountered the Timaeus as an undergraduate and how that might manifest itself in the early creation myth. And that does require some getting one's head around Oxford Classic around 1910, which is an unwelcome additional headache.

On the changes, it feels a great folly for me to debate such matters here, and I fear that I might have the wrong end of some sticks, but what about this (from my own notes of years back)?

1. The Quenta Silmarillion of the late 1930s:

§1 In the beginning the All-father, who in Elvish tongue is named Ilúvatar, made the Ainur of his thought; and they made music before him. Of this music the World was made; for Ilúvatar gave it being, and set it amid the Void, and he set the secret fire to burn at the heart of the World; and he showed the World to the Ainur. And many of the mightiest of them became enamoured of its beauty, and desired to enter into it; and they put on the raiment of the World, and descended into it, and they are in it. (Lost Road 224)

And from 1946-1948, the revised Ainulindalë (Morgoth's Ring 3).

But when they were come into the Void, Ilúvatar said to them: “Behold your Music!” And he showed to them a vision, giving to them sight where before was only hearing; and they saw a new World made visible before them… and it seemed to them that it lived and grew.... Therefore Ilúvatar called to them and said: “I know the desire of your minds that what ye have seen should verily be, not only in your thought, but even as ye yourselves are, and yet other. Therefore I say: Let these things Be! And I will send forth the flame imperishable into the Void, and it shall be at the heart of the World, and the World shall Be; and those of you that will may go down into it.” And suddenly the Ainur saw afar off a light, as it were a cloud with a living heart of flame; and they knew that this was no vision only, but that Ilúvatar had made a new thing…. But this condition Ilúvatar made, or it is the necessity of their love, that their [i.e. the Valar’s] power should henceforth be contained and bounded in the World, and be within it for ever, so that they are its life and it is theirs. (MR 11-14; emphases added).

Where the first Phase has the making of the music and then the Ainur shown the world, now they are shown but a vision. Ilúvatar then speaks and sends the sacred fire into the void, and the world is actually created. In addition, the life of the world is now said to be bound up with the Valar who enter into it. 

I had assumed that this revision marks the entrance of the Platonizing of the Ainulindalë, but (again) without thinking too much about the original version.