Alliterative Patterning as a Basis for Emendation in Middle English Alliterative Poetry

Alliterative Patterning as a Basis for Emendation in Middle English Alliterative Poetry Hoyt N. Duggan University of Virginia Edito I have taken only the first five instances, and in each instance the superi­ ority of Thornton's reading is evident. At line 262, Offord notes that Ware's reading "can be made to give a feeble sense, but is obviously corrupt" (p.52n.).At line 266 one can compare line 195: "And sayd, 'Sir, bemy soule...'" forcorroborationof Thornton.The loss of men is obvious in line 302. In line 327, Neptolemus is the son of Achilles, and the defectivealliterationin the b verse of Ware reflectsthe loss of thefollowing a verse, which reads in Thornton: "Palamedes, a prise knyghte, and preued in armes." Only the variationbetween thro/y and tnstyly in line 326 involves choosing between the readings primarily on the basis of allit­ erative patterning.The five instances presented here are very much like the remaining 38 instances in that the superiority of the regular reading in T is obvious.26 In that circumstance, 93.2 percent ofthe lines attested by Ware are accounted for as regular. There are forty-one irregular lines in T, but of those only twelve are correctable to aa!ax from W. Three lines, 461, 604, and 626, are not recorded in W and may be Thornton's addition, though there is no compelling reason to think so.Of the lines correctable by reference to W, at least three require some comment: 25 Karl Schumacher, Studien uber den Stabreim in der mittelenglischen Allitera­ tionsdichtung, Bonner Studien zur englischen Philologie,vol. 11 (Bonn: Peter Hanstein, 1914),p. 173,takes the caesura to fall after nede and thus concludes that the a verse is too long. Caesura follows knyghte, and emendation is not required. 26 The remaining 38 instances ofirregular alliteration in Wbut correctable by reference to T appear in lines 349,350,355,360,377,380,383,384,386,395,409,412,424,427,429, 459,464,472,481,483,485,499,502,520,524,535,536,539,545,548,557,603,625, 646,651,653,659,and 661. The 12 lines in T corrected by Ware 319,481,497,499,534, 546,548,558,569,577,588,and 608. 88 ALLITERATIVE PATTERNING 497 T Bot Arthur oure athell kyng, and Wawayne his knyghte (W reads Ewan both here and in the following lines). Offord rejects Gollancz's emendation of T to W's Ewan on the grounds that "there are fairly numerous examples in the poem ofaa/xx alliteration" and speculates that W's Ewan is a scribal correction (pp. 62-63n.497ff.). The speculation is distinctly odd, since the clear evidence ofthe variants is that W is far less careful of alliterative patterning than T by a ratio of almost two to one in such errors. Thus there is little reason to expect him to change his text for the sake ofthe alliteration which he so readily violates elsewhere. And, of course, by assuming that the Thornton manuscript's alliterative patterns were identical to those of the poem, Offord had prejudged the issue without analysis of the relevant available evidence. Neither Gawain nor Ywain in any version ofArthurianlegend does what the poet here says that he does, that is, throw Excalibur in the lake. Instead, as Offord notes, "this rather bald account ofArthur's passing does not suggest that he [the poet] had any source in front of him. It is not intended to be more than a summary, and was probably drawn from memory" (p. 62). The reasonable conclusion is surely that the balance of probability lies with the reading that is regular in alliteration, for there are no grounds at all on which one might take T to be correct. 499 T Than Arthure Sir Wawayne athes by his trouthe W Than Sir Ewan hym hentes by his trowth Both manuscripts are irregular. The a verse in W lacks a stressed stave, and there is no reason to doubt that T's Arthure is correct. Ewan here as in line 497 provides regular alliteration, and the original probably read: "Than Arthure Sir [E]wayne athes by his trouthe." 608 T And graythen...