In studying the history of the Bevis story in medieval and Renaissance England, one is confronted not by a single literary phenomenon but by several. The various redactions of the story in English must be considered not only in relation to the Anglo-Norman Boeve de Haumtone, their putative source, but also in relation to one another; for the differences between them go beyond the lexical and stylistic to manifest varying attitudes towards the story and varying conceptions as to its meaning. While there are too many missing links in the transmission of the Middle English text for it to be possible to trace a straightforward chronological development of the story in English, certain conceptual tendencies and shifts of emphasis can be detected in the English tradition taken as a whole, these being developed in varying degrees in the different surviving texts. My object in this chapter is to characterize the different verse redactions of Sir Bevis of Hampton, both in manuscript and in print, and to demonstrate that they all have something to tell us about the original version of the story in English and about the complexities of its textual transmission.

Most of those who read the Middle English Bevis at all read it in the version represented by the Auchinleck MS (A), since this is the only text that has been published in a readable form, providing as it does the principal base-text for Eugen Kölbing's edition of the romance. Because it is also considerably the earliest of the extant manuscripts and can be shown to have unique points of contact with Boeve, it has acquired a position of presumed textual authority or ‘correctness’ which, however, is not entirely, or not consistently, justified. By indicating A's relationship both to Boeve and to the other Middle English redactions of the romance, and by examining those episodes in which it differs most significantly from other versions of the story written in England, I hope to demonstrate how anomalous it in fact is.

In order to account for the interrelationships between the extant Bevis texts, one must assume at least one intermediary between A and the Middle English original. Not all the features peculiar to A can, therefore, be ascribed to the Auchinleck redactor himself with 100 per cent certainty.