Archaeologia Islandica - 01.01.2011, Síða 62

ADOLF FRIÐRIKSSON AND ORRI VÉSTEINSSON
grounds, then in comparably liminal
locations, outside homefields (Vésteinsson
2000a, 48-57). Recent archaeological
work has undermined both versions.
Several excavations of very early churches
in direct association with farmhouses, e.g.
at Keldudalur, Neðri Ás, Hofstaðir,
Þórarinsstaðir, Stöng and Hrísbrú, suggest
that this part of the Christian burial
paradigm was established all around the
country ifom the outset. The case of
Hrísbrú is particularly significant as it
constitutes one of the clearest examples of
a saga reference to a church apparently
built some distance away from the
farmhouses being relocated to the
emerging parish centre at Mosfell in the
early 12th century (Vésteinsson 2000a,
49). Excavations at the site have revealed,
not only the church plausibly associated
with the saga reference, but also a very
large hall adjacent to it (Byock et al.
2005). This suggests that references to
relocations of churches are just as likely or
more likely to reflect relocations of
dwellings within the homefield as
indicating any post-conversion changes in
the burial paradigm. As for the suggestion
of Christian burial outside homefields this
remains problematic because no dates are
yet available for such sites and recent
research suggests that they may very well
fit within the pagan burial paradigm; that
the absence of grave-goods and an
east-west alignment are not grounds for
exclusion from the pagan burial
assemblage.
Support for the final scenario, that a
large number of farms lost their
cemeteries, can be surmised from the fact
that churches and churchyards, distinctive
features where they do occur, are not
frequently observed at farms abandoned
in the 11 th-14th centuries. A large number
of such farms has been recorded in the
Northeast, the majority small to middle
sized operations judging from their
home-field sizes, but also a few which
appear quite substantial. In Mývatnssveit,
Reykjahverfi and Kelduhverfi not a single
church-like ruin has been recorded at
more than 100 sites, at least 2/3 of which
are likely to have been inhabited beyond
1100 if the best studied group of sites, that
of Mývatnssveit, reflects a general trend
(Vésteinsson ed. 2011; on Reykjahverfi
see Lárusdóttir 2007; on Kelduhverfi
Ólafsson ed. 2008). Only in
Þegjandadalur, where seven farms were
abandoned before the end of the middle
ages have two church-like structures been
recorded (Lárusdóttir & Hreiðarsdóttir
2011,125). This data can be interpreted in
a variety of ways but at least it is clear that
there were significant numbers of farms in
the early llth century where churches
were not built, and where, as a result,
Christian burial presumably was never
initiated.
We conclude therefore that there must
have been more than 1800 farms where
churches were built in the early llth
century although how many more is
difficult to assess. It is likely that it was
particularly farms in the 13-24 hundred
category which lost their churches in the
course of the 1 lth to 13th centuries and it
may well be that these ran to several
hundreds nationwide. It is however also
clear that many farms, particularly among
the smallest ones, never had a church built
and may therefore never have had their
own Christian cemetery. Although we
have found the evidence for some sort of
intermediate stage or type of burial
wanting we still think that this notion
merits fúrther investigation. The lack of
proof is not the same as refútation and
60