Bibliotheca Arnamagnæana - 01.06.1959, Qupperneq 93

79
and occasionally the text is corrected in accordance with a Kms reading12.
Theoretically, a critical text must of course be based on all relevant
versions, but before a foreign translation can be used as Stengel has used the
saga in his edition, it must be established beyond doubt that this version
is independent of other versions, and that it is a correct and painstaking
translation, and this had not been done.
With the almost complete victory of Bédier’s view on the authority of
the Oxford MS over all other versions, interest in the latter, including
Kms, faded for more than forty years. Recently, with the renewed interest
in the development of the Roland theme in the centuries following the
appearance of the oldest version, Runzivals påttr has been the subject of
a book by Professor Paul Aebischer13. He discusses the MSS of the Norse
version, including Fri, which had not been used by Koschwitz, and he
stresses the value of the Swedish version of the frattr (pp. 42-78). Then
follows a translation into French, laisse by laisse (the text is divided into
laisses in accordance with the O version), with notes on the Norse text,
and on omissions, additions, and parallels in other MSS of the Chanson de
Roland than the Oxford MS. In his introduction M. Aebischer points
out the numerous mistakes in the Norse translation, and he repeatedly
affirms that the translator was not well enough acquainted with the
French language to undertake a task of this magnitude. But,
si mal qu’ait été traduit le manuscrit k, force nous est d’admettre que n (i.e. the
Kms), et par conséquent k (i.e. the French source), représente une version de la
Chanson de Roland en partie identique au texte d’Oxford, mais en partie aussi dif-
férente de celle que nous connaissons gråce å ce dernier14.
M. Aebischer’s book is a valuable contribution to Kms studies, not
least because it makes the text of the Runzivals fråttr available in French.
But the translation is, unfortunately, marred by a large number of errors.
The commentary is useful in many respects, but the author has occasionally
been lead astray by his insufficient familiarity with the language of the
saga, and while he is certainly right in principle when he avers that the
Norse translator was not really qualified for his task, he is certainly wrong
in ascribing all the omissions and changes in the saga text to the trans-
r~ E.g. in the additional verses after v. 105, and in the addition after v. 200,
where Stengel has even adopted the inferior reading of a.
13 Rolandiana Borealia (Lausanne 1954), henceforth quoted as Rol. Bor.
14 Rol. Bor. p. 285.