Studia Islandica - 01.07.1963, Qupperneq 98

96
thing indicates that the family saga, as the more independent and artis-
tic of the two, would be likely to reveal the author’s individual traits
more distinctly than Knýtlinga, which owes so much to written models.
The adjectives (adverbs) from Laxdœla turned out to be the Í9
ones enumerated on p. 28. Table 4 (p. 29) shows their frequency in
Laxdæla and Knýtlinga as well as in the seven control texts. But to
have the relative frequency, which is in this case the decisive one,
we have to take into account the varying size of the texts. The control
material with its 529000 words has 470 instances of the words in
question. Let us indicate that average frequency with the quotient 1.0
and then calculate the frequency for each separate text in relation to
the control material as a whole. Thus, for example, we obtain the
quotient for Egla (62000 words) in accordance with the following
simple rule-of-three formula:
62 470 • x
--------- = ----------; x = 1.1
62000 529000
The corresponding figures for all the nine included texts are given
on p. 30. As is seen, the frequency in both Laxdœla and Knýtlinga is
three times as high as in the control material, where the seven texts
all keep fairly close to the average.
It is worth noticing that some five of these adjectives, highly char-
acteristic of both Laxdœla and Knýtlinga, belong to a similar sphere
of meaning; they denote something important, honourable, magni-
ficent: mikils verSr ‘distinguished’, stórmannligr ‘magnificent, stately’,
stórœttaSr ‘of good family, of noble parentage’, sœmiligr ‘honourable,
stately’, virSuligr ‘dignified’.
The 16 abstract nouns, which were fixed as comparatively ifrequent
in Laxdæla, are enumerated on p. 31, and their distribution over all
texts engaged in the comparison is shown in Table 5 on pp. 31—32.
The quotients in relation to the average frequency of the control ma-
terial are calculated in the same manner as for the adjective series.
Once more Laxdœla and Knýtlinga tum out to be on a level of their
own: 5.2 and 3.5 respectively.
On p. 32 a survey is presented of the frequency of the 19 adjectives
and 16 nouns taken together; the quotient for each text is given in
the column to the right. For Laxdœla (3.9) and Knýtlinga (3.2) the
figures are respectively nearly three and two and a half times as high
as the highest (Oddr 1.4) in the control material. One could not reason-
ably have wished for a more convincing result. That the quotient for
Laxdæla should be somewhat higher than for Knýtlinga is only what
was to be expected, as the test words were chosen from Laxdœla.