Bibliotheca Arnamagnæana - 01.06.1960, Side 208

198
as Miss Brown herself has pointed out, “the scribe of 440 has tidied
the text of his original, not only correcting mistakes (which survive
in Br and H), but omitting or simplifying words or phrases which he
has not understood”20. It seems therefore reasonable to suppose
that in many of these cases 440 would have shared the readings of
Br and H had it not been for the activity of its scribe. For instance,
at 36/21 occurs “perhaps the most significant correspondence
between Br and H”21, the misreading ofansu (in Br corrected later
to ofusu) for gfusu; now at this point the scribe of 440 has adopted
åfysi into his text from the marginalia, so that there is no means of
telling what his immediate original had in the body of the text—
but there is no reason at all why it should not have been ofansu, in
which case the correspondence between Br and H loses its signifi-
cance. It therefore seems unsafe to build too much upon this class
of readings.
Although Miss Brown’s arguments are, as we have tried to
show, individually insufficient to support her conclusions, it must
be admitted that cumulatively they have some weight. One might
in faet be inelined to accept her view if no positive evidence to the
contrary could be brought forward. However, we think that such
evidence is to be found in a series of readings in Prestssaga and Gud-
mundar saga dijra, which are difficult to account for except on the
hypothesis that H and 440 had a common original which was not
the same as the original of Br. They are variants or, more signifi-
cantly, misreadings peculiar to H and 440, where Br agrees with,
or is doser to, the other MSS of Gudmundar saga biskups and S tur-
lunga, including in some cases II. Examples from Prestssaga are:
129/19 Årni for Ilroi (Br I 657 204 439, correctly).
132/3 vdru gnnur for mr/rg gnnur (Br), mart annat (I R 657 204).
132/19 rænt for ræntan (Br R 657 204).
133/3 pat vår andadisk Grundar-Ketill omitted by H and 440
only.
137/3 gripit for prifit (Br), hlaupit (I R 657 204); it is possible
that H and 440 have here retained the reading of Sk, but
prifit, the less obvious word, may be the original.
20 Acta 1952, p. 39.
21 Acta 1952, p. 38.