Bibliotheca Arnamagnæana - 01.06.1999, Síða 185

VI From the tum of the century to Jan de Vries
165
that he never presents his material in an explicitly statistical manner.
Selected examples are described with a lack of critical distance, counter-
examples are ignored, and no effort is made to evaluate the weight of the
statistical evidence - the instances are not even counted. Another reason
is that he arrives at a very sympathetic picture of the religion of the lat-
est decades of Old Norse heathendom, where the divine powers and the
human beings are seen as one great family. Evidently, there may be
some truth in this picture. His examples are real, and even if one may
suspect him of casual overinterpretations, in the main the article makes
in my opinion a valuable contribution to the description of Old Norse
heathendom. What I contest is the chronological consequences he draws
from the comparison of datable skaldic poems and undated Eddie
poetry. Thus I think that de Vries’s judgement was fair when he declared
that nothing can be concluded from the faet that the actual terminology
first occurs in EjoSolfr’s Haustlpng - “in what older sources are we to
search for these words?” he asks.92
Like de Boor, Konstantin Reichardt concentrated on one particular
poem in his sympathetic attempt to overcome the unsatifactory situation
in the study of Eddie chronology by exploring the Eddie vocabulary.93 In
an article on the age of Hymiskvida (Reichardt 1933) he concentrated on
the vocabulary of this poem, in particular on its hapax legomena, of
which, according to Reichardt, it has an unusual high number, 31 ex-
amples in all. Ten of these are kennings; but since the use of kennings
will be discussed in a broader context later, I shall leave out the discus-
sion of the kennings here (cf. chapter XI below). Most important in
Reichardt’s argument is a group of eight Old Norse hapaxes that still ex-
ist in modern Icelandic, according to Sigfus Blondal’s dictionary
(1920-24). In addition to this group Reichardt pointed to twelve words
which are rare in Old Norse poetry but occur in Old Norse prose and in
modem Icelandic (Reichardt 1933: 137-41). On the basis of these find-
ings, Reichardt concluded that the vocabulary of Hymiskvida is compar-
92 “Worter wie regin, hQpt, bgnd sind daher nicht als Ausdruck flir eine erst ziemlich spat
vollzogene Zusammenfassung in eine kollektive Einheit zu betrachten (so de Boor [...]);
aus dem Umstand, daB diese Terminologie erst in Pjobolfs Hausthjng vorkommt, laBt sich
flir eine solche Betrachtung nichts gewinnen; in welchen ålteren Quellen hatten wir diese
Worter wohl zu suchen?” (de Vries 1957: 6).
93 “[...] die ergebnisse sind eher deprimierend als befriedigend, und die zahlreich gebote-
nen begriindungen entbehren fast ausnahmslos des festen bodens” (Reichardt 1933: 147).