The History of The Hobbit (review)

"Alas for the lost lore," Tolkien wrote in his 1936 Beowulf lecture, "the annals and old poets that Virgil knew, and only used in the making of a new thing!" Tolkien may not have been entirely sincere in his lament, for elsewhere he recommends appreciating the work one has rather than demanding to know where it came from. But in any case he himself has been more fortunate than Virgil, and far more fortunate in this respect than the Beowulf-poet. One of the many things these two volumes by John Rateliff do is to lead us into the very engine-room of creation. Some things we can never know, such as how the word "hobbit" came into Tolkien's mind; but against that it can fairly be said that we now know more about the gestation, if not the genesis of The Hobbit, than we do about almost any other work of any period.

The best-known version of "the history of The Hobbit" was, till now, the one given to us in Humphrey Carpenter's biography. In 1977 Carpenter published the famous story of Tolkien "sitting by the window in the study at Northmoor Road," laboriously marking exam papers, finding a blank sheet, and suddenly and impulsively writing on it "In a hole in the ground there lived a hobbit," without at that stage knowing in the slightest what a "hobbit" might be. Carpenter went on to say, quoting Tolkien, that the study was in number 20 Northmoor Road, not number 22, so that it must have been begun in or after the summer of 1930; that Tolkien "wrote the story fluently and with little hesitation" (178); that it was left unfinished apart from some plot notes, and an impromptu conclusion delivered orally to his children; and that it remained so until Tolkien's student Elaine Griffiths borrowed the incomplete manuscript and passed it on to a friend at Allen & Unwin, who urged him to complete it for publication.

Even this account, however, raises some issues. Tolkien's elder sons, John and Michael, retained clear memories of hearing the story told to them in the study at 22 Northmoor Road, i.e. before the summer of 1930. C.S. Lewis saw and read a version—Carpenter says, "lacking only the final chapters"—in, again according to Carpenter, late 1932. And Carpenter says that the version sent to the publishers was a complete typescript done one-handed by the teenage Michael Tolkien (who had cut himself badly on broken glass), which seems a rather casual arrangement for a notoriously finicky author. One cannot blame Carpenter for any defects in the story, for he was relying on the memories of the author and his family, with what looks like a fairly sketchy survey of some of the manuscript material, much of it already delivered to Marquette University. But in these circumstances, one has to look at the documents, and that is what John Rateliff has done, with immense care, thoroughness, and a great deal of illuminating and often amusing commentary.

To begin with, Rateliff is quite sure that the work was "begun in the summer of 1930 and completed in January 1933" (xx). A letter by C. S. Lewis dated 4th February shows that he had read the whole work, and liked it, apart from the ending, about which he was uncertain (one wonders why). As a result of his study of the materials collected at Marquette, Rateliff furthermore divides Tolkien's work on The Hobbit into five "phases." Phase 1 is represented by two texts, a six-page manuscript fragment which Rateliff calls "The Pryftan Fragment," after the name given there for the dragon, and a twelve...