Íslenzk tunga - 01.01.1961, Blaðsíða 85
ICELANDIC DIALECTOLOGY: METHODS AND RESULTS
83
vs. [lau:qla]; aka ‘to drive’ [a:kha] vs. [aig'a]; vekja ‘to awaken’
[ve:kjha] vs. [veig'ja]; and, finally, tap ‘loss’ [tha:ph] vs. [tha:þ(h)] ;
kot ‘hut’ [kho:th] vs. [kho:$<h)]; sök ‘case’ [sö:kh] vs. [sö:§(h)].
The area of the former (harðmœli) is in the North, from Héraðs-
vötn (in Skagajjörður) in the west to Sandvíkurheiði (north of
Vopnajjörður) in the east. The latter (linmœli) reaches from Beru-
nes (north of Berujjörður) in the East to Hrútafjörður in the western
part of the North. In between are two mixed areas, one in the East,
the other in the western part of the North.
2) Voiced vs. voiceless ð, l, m, n before p, t, k. In the former case,
i. e. when ð, l, m, n are voiced, the following occlusive is aspirated,
but in the latter case it is unaspirated. E. g. maðkur ‘worm’
[maðkhYr] vs. [maþgYr]; stelpa ‘girl’ [scjelpha] vs. [stjelba]; lampi
'lamp’ [lamphi] vs. [lamþi]; vanta ‘to lack’ [vantha] vs. [vantja];
banki ‘bank’ [þauijjkjhi] vs. [þaui]j§ji].30a
The area of the former (the voiced pronunciation) is in the North,
in Eyjajjörður and Þingeyjarsýslur. On both sides there are mixed
areas, Skagafjörður on one side and the norlhern part of the East,
with the Eastern hinterland (Fljótsdalshérað), especially north of
Jö/culsá á Brú, on the other. The rest of the country, the South-East,
South, West, and North-West, has the voiceless sounds.31
3) The distinction between initial hv- and kv- vs. the absence of
this distinction. On the one hand, e. g., hvalir (plur. of hvalur
‘whale’) and kvalir (plur. of kvöl ‘torture’) are distinguished, viz.
as [xwa:lir] or [xa:lir]—-i. e. with an unvoiced velar spirant, round
or unround—vs. [khva:lir], with [khv]-. On the other hand, these
areconfused and pronounced with [khv]-.
The area of the former is in the South-East, from Þjórsá in the
30a In the phonetic transcription, lm n ljl are used, for typographical
reasons, instead of [m n IJ] with a super- or subscript circle, to denote unvoiced
tm n q].
31 About the cluster lt, whose distribution is different from the other clusters
and much more complicated, see Einarsson, “On Some Points ...,” pp. 264—
269.