Archaeologia Islandica - 01.01.2006, Qupperneq 38

Steffen Stummann Hansen and John Sheehan
A couple of examples of the sub-circular
type are located within the Þjórsár Valley
(Þjórsárdalur), at the foot of the Hekla
volcano, which is one of the most thor-
oughly investigated archaeological land-
scapes in the Viking world. Two of these
have been archaeologically excavated,
Skeljastaðir and Stöng. The latter site,
which is adjacent to the well-known hall,
featured the remains of a small turf- and
stone-built church with intemal dimen-
sions of 4.8 m x 2.8 m. It was surrounded
by burials and, according to the excavator,
“much indicates that the churchyard had a
circular form or a circular enclosure ...”
(Vilhjálmsson 1996, 133). He conclud-
ed that the church dated to the eleventh
century (Vilhjálmsson 1996, 130-131).
The excavations at Skeljastaðir revealed
an extensive cemetery but, unlike at
Stöng, no structural remains of a church
or enclosure were evident. It is possible,
of course that these were turf-built and
consequently they may have been eroded
away in this harsh environment. The exca-
vator of Skeljastaðir, however, concluded:
“Due to the location and orientation of
some of the marginal graves the church-
yard is likely to have been circular ... .
The distance between the most southerly
grave and the most northerly graves indi-
cated that it had a diameter of approxi-
mately 20 m” (Þorðarson 1943, 134, figs.
181-182). A possible location for the
church was represented by an 8 m x 4 m
area in the centre of the cemetery within
which no burials occurred. Furthermore,
recent excavations on the Icelandic sites
Neðri Ás and Keldudalur, both with cir-
cular enclosures, have established that
these sites pre-date the 1104 Hekla erup-
tion (Vésteinsson 2000, 2005).
The majority of churches in
Norse Greenland are contained within
rectilinear enclosing elements (Roussell
1941; Krogh 1976; Keller 1989; Arneborg
1991). A small number, however, features
sub-circular enclosures. The most impor-
tant of these forms part of the famous
Brattahlið settlement complex (0 29a),
established by Eric the Red in the 980s10.
Excavations were undertaken at this
church site during the 1960s and the
results indicated that it dates to the pri-
mary settlement period of the late tenth
century; it may, in fact, be the church
mentioned in the Saga of the Greenland-
ers as having been built by Eric’s wife,
Þjódhildur (Krogh 1982a; 1982b, 34-50;
Meldgaard 1982. It was mainly built of
turf, measured approximately 4 m x 3 m
intemally, and was oriented east-west
with the entrance at the west. Although no
traces of an enclosure were noted during
the excavations, careful study of the plan
of the associated cemetery, within which
over 150 burials were excavated, indicates
the likelihood of the former existence of
a sub-circular boundary to it (Keller 1989,
187; Fig. 7). This is particularly noticea-
ble on the northern and southern sides of
the cemetery where the orientations of
the graves shift as if to follow or respect
such a boundary. Given that the church
was primarily built of turf, it is likely that
this putative enclosure was also turf-built,
and this may explain why the excavator
did not note it. If it is accepted that the
Brattahlið church dates to the later tenth
century, as has recently, based on circum-
10The conventional way of referring to Norse settlements in Greenland has been by the so-called 0-numbers (for Danish “ost”
= east) for sites in the Eastern Settlement and by the V-numbers (for Danish “vest” = west) for sites in the Westem Settlement
respectively.
36