Archaeologia Islandica - 01.01.2011, Qupperneq 83

FARMSTEAD RELOCATION AT THE END OF THE VIKING AGE.
RESULTS OF THE SKAGAFJÖRÐUR ARCHAEOLOGICAL SETTLEMENT SURVEY
et al. 2001). Changes in the landscape and
farmstead production practices would
likely favor different locations. Other
models of the settlement also suggest
biases against a high degree of stability in
early farm location. Various researchers
have looked to saga accounts of early land
claims, such as the one found in Egil’s
Saga, as a model for farm establishment
(Durrenberger 1991; Herschend 1994;
Smith 1995). There the original settler in
Borgarfjörður, Skallagrím, established
outposts in diverse ecological zones, each
contributing a specialized product to the
farm as a whole. These specialized
outposts are seen as the seeds for later
farms. Others have seen evidence in place
names for the origin of farms in
specialized activity areas (Carter
2010:268; Tetzschner 2002). Whether or
not these are apt models for early land
division and fann establishment in Iceland,
it seems unlikely that a site originally
chosen to contribute a product based on the
exploitation of a specific ecological niche
would be well situated as a standalone
farmstead. It also seemed unlikely that the
distribution of farms resulting from the
initial settlement would transition
smoothly into the more densely settled and
politically hierarchical society of medieval
Iceland. All of this added up, in our minds
at least, to compelling evidence that
hidden under the contemporary landscape,
and inaccessible to traditional survey
methods, there was a largely unknown
Viking Age Iceland.
After years of methodological
innovation and survey we can say that,
while many things were very different in
the Viking Age, for the most part, our
assumption was wrong. Based on the
SASS survey, it appears that the
overwhelming majority of farmsteads are
located where they were originally
established and that somewhere deep
undemeath most contemporary
farms-mounds there is the original Viking
Age longhouse. In fact, in most cases the
date and size of the Viking Age farmstead
is predictive of the value and productivity
of the farmstead in early modem
inventories. In general, the earlier the farm
was established the larger it was in the
Viking Age and the more valuable it was
in the early 18th century (table I).
If the systematic survey in Langholt
has confirmed that, with certain important
caveats, much of the medieval settlement
pattem can be pushed back into the Viking
Age, it has reinforced the idea that the sites
most accessible to archaeologists are
atypical. For some reason these sites were
abandoned or relocated. In cases of
abandonment archaeologists typically
recognize these farms as belonging to a
special class of environmentally vulnerable
farms, primarily in the highlands
(Einarsson 1994; Sveinbjamardóttir 1992;
Vésteinsson 2003). The survey in Langholt
suggests that unlike the highlands, the
abandonment of Viking Age farms in the
lowlands was a rare event. In fact, there is
no evidence that any farms in the survey
area were abandoned during the Viking
Age and that very few farms were ever
permanently abandoned. However, two
major fannsteads relocated toward the end
of the Viking Age. These sites are
important in part because leaving the
Viking Age component of these fanns out
of the settlement sequence would have
81