Chanson dc Rolajid. 131 



Noble vassal vos i solt horn clamer. 

 Ki la t'enveie ja n'iert de nos amez, 

 Li quens Rollanz nel se doust penser ; 

 Car estraiz estes de si grant parented, 

 Que par Charlon n'iert guariz ne tensez, 

 Ne seit ocis o a honte livrez." 



Analysis is unnecessary. So is further quotation, for whoever desires 

 may make any number of similar comparisons for himself ^ 



The device which we have called the short-strophe group- is 

 not handled skillfully in Stengel's text. In those passages which 

 describe the quarrel and the treachery it is conspicuous by its 

 absence.^ 



The device of ' laisses similaires ' is handled very unskillfully by 

 Stengel's poetaster. There are nineteen instances of the device in 

 his Roland, and seven of them are imperfect.'* Examine, for example 

 Stengel's ' laisses similaires ' in which Oliver begs Roland to summon 

 aid.'' It will be remembered that they are technicall}' perfect in O.^ 

 But here the contrary is true. In the first place, there are five 

 strophes, obviously too man}' : a poet would need to be very skillful 

 to repeat himself effectively five times." In the second place, the 

 last two are exceedingly trite. Finally, in the last three there is 

 no increment : the striking exclamation — that Roland will not wind 

 his horn for Paynims — occurs, to be sure, in the third strophe ; but 

 it has already occurred in the first, and occurs again in the fourth. 



^ Following are the worst strophes in the first 660 lines : ix, xxii, xxiii, 

 XXV, xxvi, xxvii, xxviii, xxxi, xxxiv, xxxix, xliii, xlvii, xlviii, 1, lii. It 

 is needless to cite further. 



- Snpra^ p. 122. 



' It cannot be answered that, owing to the greater length of Stengel's 

 text, its short strophes are necessarily longer than those of O. The 

 average length of Stengel's strophes — fifteen lines — is practically equi- 

 valent to that of O's sti'ophes. 



* The imperfect instances are: 520-562 d; 563- 595 a; 1049-1092; 

 1396-1437 dd; 1482o-1482zz ; 1610- 1627 p ; 1796-1850x. There are be- 

 sides three passages Avhicli perhaps are 'laisses similaires' (1737—17521; 

 1851— 1868n; 2146-21631); this loose use of the device is significant of 

 the supposed author's lack of technical skill. 



5 1049-1092. 



6 Cf. S7,pra, p. 122-123. 



'' O has only one instance of four ' laisses similaires ' (1796—1850), and 

 none of five. 



