Bibliotheca Arnamagnæana - 01.06.1991, Side 204

190
Siguråsson in the Southern part of Snæfellsnessysla 1713-1714. The
identification and discussion of 41 has clarified important aspects of
the stemmatic relationship between the manuscripts of GS (especially
226 and 229) and has answered questions to problems which Gu6-
mundur f>orlåksson was unable to solve on the basis of the four
manuscripts known to him.
655 is the oldest manuscript and, on the basis of its dating, is prob-
ably not far removed from the original translation, which is generally
thought to have been undertaken in the beginning of the 1260s.
Unfortunately 655 and 238 cover different sections of GS, which makes
it virtually impossible.to come to any conclusion as to their inter-tex-
tual relationship. Textually 655 appears to be superior to 238, and it
preserves comparatively more of the Latin text than 238, but whether
such omission in 238 was made in the original translation or in the
transmission of the text cannot be ascertained. The only manuscript
with which the two fragments can be compared is 226, and this
comparison reveals that GS in 226 is reduced by about one-third in
relation to the fragments. It is impossible to determine in how many
stages of transmission this abridgement took place, but it seems
reasonable to assume that the main abridgement was the work of a
single redactor, and it must clearly have taken place before the date of
226. The fragments 229 and 41 belong to the abridged redaction and
are closely related to 226.3 The significance of 41 is weakened by the
faet that its text is extremely corrupt; nonetheless, reference to the
Latin reveals that in several instances it has a more original rendering
than 226 or preserves material not present in 226. A comparison of 229
with 226 shows that although the text of the two manuscripts is in
many places identical, there is still evidence enough to suggest that 229
cannot be a copy of 226. In a section of GS covered by both 41 and 229
there are instances of an identical or similar reading in 229 and 41 as
opposed to 226. In one instance there is an apparently secondary
reading common to 226 and 229 as opposed to 41: the Latin »corpore
ejusdem« is in 226 and 229 rendered as »f>essu bannsettu (+ sinu 229)
valningar keri«, whereas 41 has »hans kropp«.4 This would suggest a
relationship among the manuscripts of the abridged redaction as illus-
trated in the following diagram:
3 SeeJ6n Helgason, p. 372.
4 Thephrase appeais in theprinted edition p. 1001. The readingin226 and 229 may
be taken from 10020