Studia Islandica - 01.06.1962, Side 194

192
Excursus. (Pp. 52—56). Lastly there is a summary presentation of
a special statistical study, which perfectly supports the conclusions
above. It concerns the verb kveSask, usualiy in preterite: kváSsk (sing.)
or kváöusk (plur.). This (indirect) verb of saying reveals in our main
texts the frequency shown on p. 52. The italicized figures to the right
are as usual adjusted with reference to Egla as the standard of com-
parison: Egla keeps its actual figures from the left to the right co-
lumn; the other sagas get their figures increased or reduced to what
they would have been, if they had been of the same size as Egla.
The list reveals that Egla in such an important and probably often
unconscious detail, a matter of habit, in the individual use of language,
comes much nearer to Snorri than do any of the other sagas. The dif-
ference between Egla on the one hand and Snorri A and Snorri B on
the other is in fact smaller than that between Snorri A and Snorri B.
With regard to the nature of this linguistic detail one might safely
maintain that the statistics practically exclude the possibility that Snorri
could be the author of any other of the five family sagas than —
Egla.
To demonstrate how absolutely unique the low frequency of kveÖask
in Egla and Heimskringla really is, additional saga material has been
consulted, comprising about 400000 words, a volume comparable to
that of the texts in the original research. The corresponding list is
presented on p. 54. First and foremost this completing test reveals that
the more material we include in the statistical comparison — in this
instance practically all the relevant texts — the more emphatically
do Snorri and Egla maintain their common and unique position. The
figure for Egla is 7; the adjusted number for Snorri A and Snorri B
together — 16—j-9 cases on 119 000+109 000 words — is also 7. This
is only a third of the lowest frequency found (in FóstbrœÖra saga)
among the other thirty sagas.