Bibliotheca Arnamagnæana - 01.06.1999, Page 170

150
Part One
is still open to other philological interpretations. The expression dvergar
d Qxlom from Rigspula (st. 16) Nerman had first interpreted as fibulas,
brooches used by women for fastening their dresses (Nerman 1931: 29),
but in two later articles he returned to Rigspula, remarking that since
fibulas were worn on the breast, not on the shoulders, the dvergar
should probably be identified with some more obscure objects found in
female graves in Sweden from about 650 to about 1000 (Nerman 1954;
1969: 15). If this interpretation is accepted, and to the best of my know-
ledge no better has been proposed so far - which obviously does not of
necessity imply that it is correct - I think it provides a solid piece of
evidence for dating the text;70 this is not the kind of object which could
be expected to exist in poetic tradition without any reference to real his-
torical conditions. He also pointed to the mention in Rigspula (st. 34) of
silk, which was unknown in Scandinavia before the Viking Age, and
consequently is a safe criterion for post quem dating (Nerman 1969:
14—15); in the light of the recent discussion of this poem a comparable
criterion ante quem obviously would be more interesting, however.
In his book from 1931 Nerman attempted a full survey of archaeolo-
gical arguments for dating the Eddie poems, and some reviewers were
disappointed that in the end the evidence adduced was rather slight (cf.
Kendrick 1932: 11; R. Beck 1935: 238-39). A critical examination has
revealed that extremely few, if any, instances are fully convincing.
Nevertheless, as a means of dating this is in my opinion a very demand-
ing but basically sound method, and further progress may be expected,
not least, perhaps, for a poem like Rigspula with its abundance of cul-
tural detail.71
Nerman himself concluded his last article with the statement that the
Eddie poems can be dated only by a combination of philological and
archaeological methods, philology alone being unable to yield chrono-
logical results.72
70 “Ifrågavarande uttryck i Rigsjsula kan då inte gåma vara aldre an från andra halften av
600-talet. Men det kan icke heller vara yngre an omkring år 1000” (Nerman 1954: 212; cf.
1969: 18).
71 Cf. e.g. Heusler 1906, von See 1957 and 1960, which hardly contain any conclusive
evidence. Joseph Harris finds that Nerman’s argument is particularly persuasive in the
case of Rigspula (Harris 1985: 97).
72 “Det gis ingen annan saker vag till datering av Eddadikterna an genom en kombinerat
arkeologisk-filologisk belysning. Med endast filologiska synpunkter kan man icke nå