Editiones Arnamagnæanæ. Series A - 01.06.1997, Blaðsíða 42

XL
agrees, the complete disparity between 179 and 181 g in this passage is baf-
fling. It is not credible that both have been copied from a legible, or mainly
legible, S6. The difference between them is so great that in at least one or
other S6 must have been largely given up as the source for this passage.
If it is correct, as put forward in this edition, that we have in A4, A5 and
other manuscripts like them a third witness to the text of S6 independent of
179 and 181 g, then unless something different applies in this passage alone, it
must be said that they confirm that 18Ig has reproduced the text of S6 fairly
well. One mildly opposing point, a partial agreement between 179 and A4-A5
(with 181g diverging), occurs at 1217: ‘og sijner Hlodver kongi brefit’ 179
(A2), ‘Nú síndi Mírmann jall Hlpdver kóngi þetta rit’ A4-A5. But this is negli-
gible, and the conclusion must be that in this passage 179 is the one in which
the innovation has taken place.
lt is possible to compare the passage with texts B, D and F. They have their
own features, but all three agree with 181 g in that they have the king suspi-
cious and attempting to discourage Mírmann from going, and B agrees with it
in several substantial readings. The similarity suggests that there was once an
original text something like this, which descended to S6 and is reproduced in
181 g; then there has been innovation in 179.
No other manuscript except a late copy of 179 has the passage in the form
it takes in 179, so the wording of 179 here (if it is a departure from S6, as sug-
gested) may perhaps be due to the scribe’s composing rather freely where his
exemplar had become more difficult to read than it was when others copied it
before him. The same might be said about two short but notable variants just
before and after the passage; see 1213 and 1221 in the variant apparatus. Use of
another source here seems unlikely, since the space left blank at ll88 was al-
lowed to stand. In either case (free treatment of S6, or use of another source),
his behaviour here was in marked contrast to his choosing to leave a blank
space a little earlier.
The alternative is to believe that 179 (written by the reliable Jón Erlends-
son) was an accurate copy of S6 in this passage and re-assess the relationship
between S6 and 18Ig and between S6 and A4, A5 etc. The agreement of 18lg
and A4-A5 in this passage would then indicate that they are not independently
derived from S6. Since their own variants show that neither can be derived
from the other, they must then be supposed to be derived from a lost common
original which was a copy of S6. Then other readings on which they agree
against 179 in the second part of the saga could be attributed to this supposed
lost common original rather than to S6; indeed one would be almost bound to
suppose that quite a number of them arose there, as otherwise it would have
been an unusually accurate copy of S6.
What counts against this alternative view is that in the first part of the saga,