Bibliotheca Arnamagnæana - 01.06.1999, Blaðsíða 248

228
Part Two
In both cases the correlation is positive — as it should be - but far lower
than in skaldic poetry. In the case of Sonderegger the correlation is only
faintly positive,16 and if he is right in his relative datings of Eddie
poems, the question arises whether there is a plausible explanation for
the difference between skaldic poetry and Eddie poetry in this respect. If
such an explanation is found, there is no problem, and there is a possibil-
ity that such an explanation provides a valuable contribution to the study
of Old Norse stylistics. However, in so far as no explanation is found,
the criterion of/um may be used as an argument that Finnur Jonsson’s
dating as a whole is nearer the mark than Sonderegger’s.
Finally, it should be noted that the use of the particle in Eddie poetry
or parts of Eddie poetry might be an aspect of style related to particular
authors or particular subject matters rather than related to chronology.
According to Ingerid Dal, “In Eddie poetry of has mainly become a
stylistic device”, and the particle is above all used “in lively narrative”.17
It may also be tempting to surmise that some poets used it as a deliber-
ate means of archaizing the style.
Different explanations are thus possible, but as a matter of principle
ad hoc solutions should be avoided; arguments of this kind should be
supported by independent evidence. If archaization is intended, there
ought to be other signs of archaization as well. The question would also
have to be raised whether a medieval writer is likely to have possessed
the necessary knowledge of language history to be able to use this
means of archaization consistently. In his treatise Kuhn has pointed to a
number of instances where modem editors have added the particle by
conjecture in contravention to the rules that he was later to find on the
basis of the transmitted texts (Kuhn 1929: 13 n., 24-25). But he men-
tions only one instance where a skald has not followed the rules (of
svangan in Einarr Skulason, cf. Kuhn 1929: 32, 33). The archaization
hypothesis would in my mind have been considerably stronger if there
had been more instances of this kind, also in Eddie poetry.
16 To calculate the correlation according to de Vries’s datings (cf. pp. 183-84 above) is
not very meaningful, since he groups a great number of poems together in the same period.
The result would not be very far from that of Sonderegger, however, and far beneath that
of Finnur Jonsson (for de Vries p is approx. 0.10).
17 “Of ist in der Edda hauptsachlich ein Stilelement geworden; es wird innerhalb der leb-
haften Erzahlung verwendet, um den aktionellen Charakter der Verbalhandlung empha-
tisch hervortreten zu lassen” (Dal 1930a: 9).