Bibliotheca Arnamagnæana - 01.06.1999, Page 265

VIII “VinSandin foma”
245
a) The poems are of West Nordic origin, and were composed after the
shift vr- > r- had taken place in ro g, rægja, but before it had taken
place in vreidr.
b) They are posterior to the general shift vr- > r-, but forms of vreiåi
are used as archaisms.
c) They are of East Nordic origin, from a period after the shift vr- >
r- in word-forms corresponding to rog, rægja. In this case inde-
pendent evidence for East Nordic origin should be looked for.
In Håvamål, if rata and Rati are accepted as genuine vr- words ac-
cording to the etymology, one might think of the period of transition,
if not, the poem (Håvamål I) would seem to antedate the shift.
To sum up: According to this test, Vaflprudnismål, and probably Atla-
kvida and Lokasenna, possibly also Håvamål I, Baldr s draumar and
Sigrdnfumål, have a higher relative age than Grlpisspå, Atlamål, and
probably Helgakvida HjQrvardssonar, Alvissmål and Grogaldr, pos-
sibly also Grlmnismål. The absolute age can only be suggested very
broadly. Allowing for a period of overlap of about a century, one might
suggest a dating of the early group before the year 1000, and the late
group after 900.
In this relatively vague form the result corresponds well to the estab-
lished opinions on the age of the Eddie poems, which is far from surpris-
ing, as the criterion has been used in the discussion of dating for more
than a hundred years. But in so far as this result is accepted, it has the
important implication for the whole debate on dating Eddie poetry that it
demonstrates the ability of a linguistic test to distinguish between poems
of different ages far back in preliterary times, to a period at least two or
three centuries anterior to the period when the poems were written
down.