Bibliotheca Arnamagnæana - 01.06.1999, Blaðsíða 308

288
Part Two
proved, but his general view on the development in religious sentiments
will be confirmed at least for one part - and an important one - of Old
Norse poetry.
In the following I refer to mythological kennings in court poetry, as de-
fined in Fidjestøl 1982 (cf. nos. 1-95 in the corpus list pp. 170-75). This
list provides a corpus established independently of the problem under ex-
amination; it is a collection of poems covering the greater part of skaldic
poetry, and happens above all to be the part which, due to its relation to
political history, presumably had the best conditions for transmission. It
is also, due to its content and function, the most homogeneous corpus it is
possible to establish for such a long period of time. The corpus has been
criticized for not including poems like Ynglingatal or Haraldskvædi and
others, and although this criticism is not unjustified, the purpose of such
exclusions is to establish as homogeneous a collection of one genre as
possible (cf. Fidjestøl 1982: 179-82), which in the present context is an
obvious advantage. One item has been excluded from the list, however,
namely nr. 72, Øxarflokkr by Einarr Skulason. In the original corpus list
it was included with some reservation (Fidjestøl 1982: 156), but in the
present context it can be seen to be strongly biased, as it more or less
amounts to a skaldic exercise in the use of mythological kennings con-
centrating on the goddess Freyja.
The corpus is preserved in a fragmentary manner, but I do not see any
reason why this faet should affeet the relative amount of mythological
kennings transmitted in it.
In Table 22 I have noted all occurrences of names of divine beings in
either the defining word or basic word of the kennings - in one case in
both.26 Most of them are proper nouns, but appellatives like åss or regin,
as well as kennings for a god have been included. Valkyries are includ-
ed, with the exception of the words hildr and gunnr, which may be ap-
pellatives denoting ‘war’. Giantesses are not included, partly because it
26 Although the list pretends to completeness, I am aware that I have not succeeded totally.
Whereas the method is the same as in Fidjestøl 1993, a new examination of the material has
led to the discovery of as many as 29 new occurrences (out of 212). There is no reason to
believe, however, that these examples are not scattered over the periods in an arbitrary
manner, and that is why the chart presented in Fidjestøl 1993 differs so little in its form from
the present one. As for the total number of kennings within each period, the material has not
been revised after 1992, so there may still be some inexactitudes. - An asterisk means that
the word in question represents a textual emendation, but in every case I refer to my own
evaluation, made prior to and independently of the present study.