more concerning either the subject chosen or a suggested date, wrote again to Tolkien on 18 January 1939. Tolkien replied on 1 February that his chosen topic was ‘Fairy-stories’ and suggested 8 March for its delivery.
Tolkien had little time to prepare the lecture before its delivery at St Andrews on 8 March 1939, but already had given thought to the subject. Probably towards the end of 1937 he was invited by the Lovelace Society at Worcester College, *Oxford to read a paper at a meeting of 14 February 1938. In a letter to C.A. Furth of George Allen & Unwin (*Publishers) on 24 July 1938, Tolkien said that he had rewritten *Farmer Giles of Ham the preceding January ‘and read it to the Lovelace Society in lieu of a paper “on” fairy stories’ (Letters, p. 39). It seems likely that he did not give a paper on fairy-stories on that occasion, because he found that he did not have enough time to write one, or to finish writing, and it was easier in the event to revise Farmer Giles of Ham, a version of which was already in hand.
Drafts for the lecture preserved in the Bodleian Library, Oxford (*Libraries and archives) show that Tolkien drew upon resources that became available only after he accepted the invitation, and evidently worked on his text until the eleventh hour. He included several references to The Coloured Lands by *G.K. Chesterton, which was published in November 1938; and it seems likely that a comment in the second version of the lecture was inspired by a letter of 11 February 1939 to Tolkien from C.A. Furth, who found Farmer Giles of Ham hard to categorize for a prospective market. (‘Grown-ups writing fairy-stories for grown-ups’, Tolkien wrote, ‘are not popular with publishers or booksellers. They have got to find a niche. To call their works fairy-tales places them at once as juvenilia; but if a glance at their contents shows that that will not do, then where are you? This is what is called a “marketing problem”’ (Tolkien Papers, Bodleian Library, Oxford).)
Notes and working papers, as well as citations of many tales and authorities within the lecture, indicate that Tolkien did a considerable amount of background reading. It is also clear from his drafts that composition of the lecture did not come smoothly. After writing a first text, Tolkien decided that it needed revision and reorganization and wrote a second version, reusing some of the pages from the first. Both versions are heavily marked with revisions, and neither seems suitable as a reading copy for the actual lecture at St Andrews; it seems, in fact, that the text delivered in 1939 has not survived. One can say for sure, as Tolkien does in an introductory paragraph to On Fairy-Stories in Essays Presented to Charles Williams, only that the 1939 text was ‘abbreviated’ relative to that of 1947. The earlier papers contain nothing about eucatastrophe, or any of the material contained in the published ‘epilogue’. It is certainly significant that a lengthy summary of the lecture in the St Andrews Citizen for 11 March 1939 makes no mention of ‘eucatastrophe’ or any reference to Christianity (‘Andrew Lang’s Unrivalled Fairy Stories: Oxford Professor’s St Andrews Address’, p. 6).
The rough handwriting and frequent emendation of the On Fairy-Stories manuscripts, together with many miscellaneous notes and memoranda of various dates in the Bodleian papers, make it hard to trace the history of writing of the lecture. But much evidence exists, both among and outside of the preserved papers, to show that Tolkien returned to On Fairy-Stories only a few years later, revising and enlarging it, now including the ‘epilogue’. The first reasonably legible and continuous surviving manuscript (though still with many deletions and replacements) cannot be earlier than 1943, since it contains a new reference that a story about the Archbishop of Canterbury slipping on a banana skin might be disbelieved if it was said to have taken place between 1940 and 1943 (1940 and 1945). Drafting for this is on the verso of a sheet referring to cadets whom Tolkien taught at Oxford beginning in spring 1943.
It may be that Tolkien was inspired to look at On Fairy-Stories again after being appointed on 4 December 1942 an examiner of the B.Litt. thesis of *Roger Lancelyn Green, Andrew Lang as a Writer of Fairy Tales and Romances. Tolkien drafted a new introductory paragraph for the lecture on an unused calendar page for 16–22 August 1943, where it is said that only ‘some part’ of On Fairy-Stories ‘was actually delivered [at St Andrews]. Its present form is somewhat enlarged … and … made longer and I hope clearer than the lecture’ (Tolkien Papers, Bodleian Library, Oxford). In a letter to his son Christopher in 1944, Tolkien wrote, after giving an account of a miracle: ‘And all of a sudden I realized what it was: the very thing that I have been trying to write about and explain – in that fairy-story essay that I so much wish you had read that I think I shall send it to you. For it I coined the word “eucatastrophe” …’ (7–8 November 1944, Letters, p. 100).
Circumstantial evidence that Tolkien was working on the lecture, and perhaps reading it to or discussing it with the *Inklings, is contained in a letter by *C.S. Lewis to Gerald Hayes on 12 March 1943, defending his taste for works such the Morte Darthur, The Faerie Queene, Arcadia, the High History of the Holy Grail, and the prose romances of William Morris. ‘But ought we not both to defend our tastes more stoutly?’ he wrote. ‘To all this about being “grown up” may we not answer that the desire to be grown up is itself intrinsically puerile but the love of “fine fabling” is not. These books were written neither by children nor for children. Because they are now out of fashion they have gravitated to the nursery as the old furniture has – the same is true of fairy-tales themselves’ (Collected Letters, vol. 2 (2004), pp. 562–3). There are similar statements by Tolkien in versions of the essay. By 5 August 1943 *Charles Williams had arranged for the newest version to be typed by his friend Margaret Douglas, and on that date she wrote to a friend that she had agreed to type the lengthy essay and that Tolkien’s handwriting was difficult to read.
In late 1944, with the prospect that Williams would soon be leaving Oxford for the London office of Oxford University Press, Tolkien and C.S. Lewis began to organize a Festschrift to honour Williams’ work for Oxford University. When Williams died unexpectedly on 15 May 1945, the Festschrift became a memorial volume. Tolkien’s contribution, On Fairy-Stories, needed only a few emendations to the carbon copy of the typescript made by Margaret Douglas.
Tolkien felt that the ideas he developed in On Fairy-Stories had influenced the writing of *The Lord of the Rings, and said so in letters at least as early to correspondents including Peter Hastings (September 1954) and Dora Marshall (3 March 1955). In a letter to *W.H. Auden on 7 June 1955 he complained that Oxford University Press had ‘most scurvily’ allowed the lecture (in Essays Presented to Charles Williams) to go out of print (Letters, p. 216). But Allen & Unwin were eager to publish it themselves, perhaps as a small book if Tolkien could expand it by about half and remove references that revealed its origin as a lecture. In August 1959 Tolkien signed a contract with Allen & Unwin for publication of an expanded version of On Fairy-Stories, which he hoped to have ready by the end of the year; in the event, the idea lay dormant until 1963, when Tolkien and *Rayner Unwin discussed the possibility of publishing On Fairy-Stories to keep Tolkien’s name in the public eye while he continued to work on *The Silmarillion. At length it was decided to publish the lecture together with the story *Leaf by Niggle in a new volume, *Tree and Leaf.
In 2008 On Fairy-Stories was published by HarperCollins, London, in an ‘expanded edition’ as Tolkien on Fairy-Stories, with commentary and notes by editors Verlyn Flieger and Douglas A. Anderson. The essay proper was reprinted ‘in its final form as edited by Christopher Tolkien and published in The Monsters and the Critics and Other Essays’ (p. 27). With this are partial transcriptions, edited to form ‘a readable text’, of the two manuscript versions we refer to in the present article (numerous extracts also appear in Reader’s Companion); editors’ introductions and annotations; and a comparison of the essay as published in Essays Presented to Charles Williams with that published in Tree and Leaf (similar to the analysis in Descriptive Bibliography, pp. 184, 186–9).
CRITICISM
Published on 4 December 1947,
