ABSTRACT
The Canterbury Tales as we know it is a fifteenth-century book. When Chaucer died in October 1400 he had abandoned the work. In the Retracciouns, appended to his translation of the penitential summae which became the Parson's Tale, he listed the 'tales of Caunterbury thilke that sownen into synne' among his published works which he wished to revoke as 'enditynges of worldly vanitees'. The manuscripts of the extant 24 tales show that the planned 120 tales envisaged in the General Prologue were not written. Clearly while the work was in progress, no final collection of the tales was feasible, and after the Retracciouns none was intended. The extant manuscripts show further that Chaucer did not write sequentially according to a determined pattern of narration; that he pressed into his narrative some work that had been written much earlier for different purposes, like the Monk's Tale; and that some of the Canterbury Tales, nevertheless, were combined into larger units to have a dramatic impact on the whole work, like the tales of Knight, Miller, Reeve, and Cook. Thus the reference to 'the tales of Caunterbury' in the Retracciouns must be to those single tales or groups of tales which were already known (presumably in manuscript) to 'hem that herkne this litel tretys or rede', among them perhaps Sir Peter Buxton, steward to Henry, earl of Derby (later Henry IV), and Thomas Hoccleve, a clerk of the Privy Seal.
