Wynkyn de worde's lost manuscript of the canterbury tales: With new light on HRC MS 46

Previous scholarship has argued that the manuscript (^∗W) used by Wynkyn de Worde for his 1498 edition of the Canterbury Tales (Wn) was derived from the same line of descent as Gg (Cambridge University Library MS Gg.4.27) and Ph¹ (The University of Texas, The Harry Ransom Center MS 46). Most recently, Stephen Partridge, examining the glosses of Wn, has proposed a theory that two types of glossed text, an El-Ll² type and a Gg-Ph¹-Wn type, were derived from a common lost source, and that a further stage of copying of a Gg-Ph¹-Wn type led to Gg and Ph1 with glosses largely omitted and Wn with glosses. My extensive textual analyses of Wn have advanced this theory, and show that Wn/^∗W is textually closer to Ph¹ than Gg and retains more archetypal readings than these two manuscripts. These findings strengthen the possibility that a copying of ^∗W led to the production of Gg and Ph¹, or a manuscript behind them.