From 1485 until 1934, Malory's text was known only from the print produced by William Caxton: 'enprynted and fynysshed in thabbey of Westmestre the last day of Iuyl the yere of our Lord MCCCCLXXXV' ('printed and finished at Westminster on the last day of July 1485'). Two versions of this print now survive, and are held at the John Rylands University Library, Manchester, and the Pierpont Morgan Library, New York.
In 1934 Walter Oakeshott, then a librarian at Winchester College, was conducting a survey of the medieval books held in the Masters' Library. He discovered a locked safe, and sought permission to open it:
When at last I approached the safe with the key in my hands, it was with some excitement. I slid aside the metal grille, and was dashed to see at a glance that on the twenty or thirty manuscripts not a single medieval binding remained... Two or three which were not in Latin but were in English caught my eye. One was very fat, some 480 leaves, paper not vellum, the text prose not verse, clearly about King Arthur and his knights, but lacking a beginning or an end...
On careful examination, Oakeshott's discovery proved to be none other than a long-lost Malory manuscript! The 'Winchester Manuscript', as it is now known, was completed by two professional scribes at some point between 1470 and 1483. Interestingly, traces of printer's ink on the paper suggest that it may have been in Caxton's workshop at some point – although linguistic differences mean that it is unlikely to have been the main source-text for the Caxton print. The text was edited by Eugéne Vinaver – who, when the discovery was announced, literally appeared on Oakeshott's doorstep asking for the chance to work on it – and published in 1947.
Until the discovery of the Winchester Manuscript, scholars had not known the extent to which Caxton had altered Malory's text when editing it for print in the 1480s. When the two witnesses were compared, the changes proved to be substantial. Where the Winchester Manuscript divides the text into only eight long 'Books', Caxton had split the work into twenty-one 'Books' with 507 sub-chapters. He had also dramatically reduced the content dealing with Arthur's adventures on the Continent (Book Two in the Winchester Manuscript) and appears to have altered the orthography (spelling, punctuation and 'look' of words) to more closely match his own Kentish dialect (P.J.C. Field, The Life and Times of Sir Thomas Malory, p.3). The discovery also solved some long-standing problems regarding errors and holes in the text: for example, while Caxton's text rather nonsensically gives 'sir Tristram slewe kynge Mark' ('Sir Tristram slew King Mark') (unlikely, since the character of Tristram has in fact already been killed by Mark at this point in the text), the Winchester manuscript gives: 'Sir Bellyngere revenged the dethe of hys fadir sir Alysaundir and Trystram for he slewe kynge Mark' ('Sir Bellinger avenged the death of his father, Sir Alexander, and Tristram, for he slew King Mark').
The question of print vs. manuscript raises some interesting questions for a modern-day reader. To begin with, it remains more difficult to get hold of the Winchester text: the editions edited by Vinaver (1947, 1967, revised by P.J.C. Field in 1990) and Stephen Shepherd (2004) are the only available sources of the Winchester version of Malory's works. E-text editions of the Caxton-based text, many of which are based on print editions which have gone out of copyright, are more easily obtained – but may not give as accurate a picture of the work, given the alterations that were made by Caxton. As the Winchester version has been the preferred text for most scholars working after 1947, reading a Caxton-based version might also lead to disjuncts when attempting to apply critical reading to the works. For more examples, Mark Adderley gives a comparison of a short extract from the Caxton and Winchester witnesses at: http://www.markadderley.net/arthur/malory/malory-text.html
Even the title varies – as it turns out, the heading 'Morte Darthur' is probably Caxton's addition: the Winchester Manuscript does not give a title, and the story itself is concerned with Arthur's life, and that of his knights and the other members of his court, just as much as it is with his death. Vinaver chose to overcome this problem by entitling his edition The Works of Sir Thomas Malory, although Shepherd has returned to the traditional Le Morte Darthur. The differences in layout between the versions (eight books in Winchester vs. twenty-one in Caxton) have also led to questions about how the text had been intended to be read – while the twenty-one book version reads as a complete story, intended to be read along the lines of a modern-day novel, the eight-book version can encourage a more disjointed or 'episodic' reading, which has caused significant critical debate.
- The Malory Project: http://www.maloryproject.com/
- Hilton Kelliher, 'The Early History of the Malory MS', in Toshiyuki Takamiya and Derek Pearsall, ed., Aspects of Malory (1986).
- W.F. Oakeshott, 'The Finding of the Manuscript', in J.A.W. Bennet, ed., Essays on Malory (1963).
- R.M. Lumiansky, Malory's Originality (1964).
- Terence McCarthy, 'The Order of Composition in the Morte Darthur', Yearbook of English Studies 1 (1971): 18-29.
- Ralph Norris, Malory's Library (2008)
- Eugéne Vinaver, Works vol I, 2nd ed (1968).
Composition order and novelistic vs. episodic readings: