Íslenskt mál og almenn málfræði - 01.01.2021, Blaðsíða 155

Also striking in Figure 4 is the scant number of neoformations found with refer-
ence to loanwords first attested in works composed before the 12th century.14
This might indicate that this strategy was somewhat less productive during that
period, and that other strategies, chiefly calquing, were preferred. However, this
result may also be influenced (and possibly biased) by the limited extent of early
data and by the lack of variety in textual genres (refer to Tarsi 2019b:100–101 and
errata corrige for a deeper discussion of these data).
Finally, two special classes of nouns are worth mentioning, namely geographic
names and Latin integral borrowings, for they constitute special lexical classes.
The creation, and hence the meaning, of geographic names is directly connect-
ed to a given geographic entity. Thus, the process whereby a geographic name is
created is in part distinct from that of nouns or verbs. However, both geographic
names and other lexical categories follow the general principles of phonological
resemblance, especially when borrowed or adapted, and of motivation, in particu-
lar with reference to their creation. Motivation in geographic names is the princi-
ple by which they acquire their designation in relation to meaningful referents in
the language in which they arise. Having said this, it is significant to note that geo-
graphic names appear most often in explicative insertions, whereby the native
term is introduced as an equivalent of the foreign one, and only once the other way
around (cf. § 3.3 above). Only in a handful of occurrences do geographic names
appear in simple alternation or intrastemmatic variation. All the native renderings
of geographic names found in the corpus are neoformations, with the sole excep-
tion of viðsmjörsviðarfjall, a structural calque of Mons Oliveti in Oddur Gott -
skálksson’s translation of the New Testament. As regards the formation of geo-
graphic names, they were found to be the result of four different processes: 1)
determinative compounds (where the noun head is a geographic term, e.g. Blá -
land), 2) native adaptations (usually following a principle of phonological resem-
blance together with semantic motivation, e.g. Jórsalir),15 3) epexegetical com-
Loanwords and native words in Old and Middle Icelandic 155
14 This is an arbitrary chronological division made to ensure that the loanwords were
certainly old in the language. Nothing prevents, however, that loanwords first attested
only after the 12th century to have been originally acquired well before that. Striking cases
are those of skrifa (possibly acquired in North Germanic by the 8th century, Tarsi 2019a)
and paradís (also a very old loanword, cf. Tarsi 2016a:89–90).
15 It is comforting that Myrvoll (2021) recently arrived at the same conclusion as
regards the formation of the toponym Jórsalir. Neither of us knew about each other’s
attempt at etymologizing this name. Mine dates from my first stay at Uppsala University
in 2017–2018. Earlier mentions that I know of the paretymological genesis of Jórsalir are
in von Friesen (1942:280), who identifies jór- as a cognate of OE eofor, OHG ebur ‘wild
boar’ (on this issue see Myrvoll 2021:45–47), and Jansson (1954:43), who only mentions
the paretymological analogical process which produced a toponym with -salir as the head
of the compound. For the sake of completeness, I hereby quote my reasoning from the dis-
sertation (Tarsi 2020a:17), a revised version of which is currently forthcoming: