128 F. B. Luqiiiens, 



Miiller's text is practically the Oxford manuscript. Therefore it 

 possesses, of course, marked and almost consistent technical ex- 

 cellence. Let us turn to the examination of Stengel's text.i 



The fact that Stengel's supposed x' contains over six hundred 

 more lines than O, arouses at the outset suspicions of lack of technic. 

 His text tells the same story as is told by O, but uses about five 

 thousand more words in doing so. This smacks of the poetaster.^ 



Our suspicions are confirmed b}^ closer examination. Considered 

 in its entirety, first, Stengel's text shows lack of emphasis, hi O, 

 it will be remembered, emphasis is due to the proportionate lengths, 

 and to the proportionate intensities, of the various natural chapters. 

 But in Stengel's text The Quarrel of Roland and Ganelon, and The 

 Treachery, are awkwardly combined into one chapter ^ ; and The 



vv. 537—549, and the Baligaut Episode. — Even if Stengel's stemma is 

 correct, however, our arguments against it remain justified in spirit, if 

 not in letter ; for tliej^ help to prove that the reconstruction of x'", not 

 that of x', nor even that of x, is the most important task of Roland study. 



^ I feel that it "would be ■wrong not to acknowledge emphatically the 

 extent of my indebtedness, in common with all Roland students, to Professor 

 Stengel. It may be said of his edition that it is at the same time the 

 most useful, and the most harmful, of books dealing with the Roland \ 

 It is invaluable, and always will be, because of the mass of critical 

 material ■which it supplies in convenient form. It is harmful because, 

 in sjHte of its author's own doubts as to its finality (cf. Verhandhmgen 

 dcr fiinfundvierzigsten Vcrsaminhing Detitscher Philologen tind Schiilmdnner in 

 Bremen^ p. 133), it is no"w generally accepted as the authoritative text of 

 the Roland : upon it are based practically all Roland studies of the past 

 decade. And an English translation of it has lately appeared, which 

 w^ill give to general readers a very -wrong idea of the Roland [The Song 

 of Roland neivly translated into English by Jessie Crosland with an Introduction 

 by Professor L. M. Branding London, 1907). 



2 In this connection I cajinot refrain from quoting a passage from an 

 essay in the Fortnightly Revie-iv for November, 1906, bj' C. F. Keary. The 

 unpoeticalness of redundancy does not need confirmation, but the appli- 

 cability of Mr. Keary's remarks to the question in hand is striking. " A 

 redundancy of words is the most unpardonable of all faults in versifi- 

 cation. It is almost always such redundancy that marks off the poet- 

 aster from the poet. We may ez'en say that when there are many words em- 

 ployed — although they cannot fairly be called red^tndant — the Mnse is ift a pe- 

 destrian vein^ The italics are mine. 



'^ For the shift of v. 365 involves the obliteration of the dividing line 

 between the two chapters. Notice also that the correspondence of v. 365 

 and V. 660, pointed out snpra^ p. 117, Note 9, is no longer effective. 



