Archaeologia Islandica - 01.01.2004, Qupperneq 74

Orri Vésteinsson
tive and countrywide analysis on a scale
that would be difficult to duplicate else-
where. If this sort of work is to be car-
ried out however the data-set as a whole
must be scrutinized first in order to make
certain that it is really comparable. There
are two sides to this problem. On the one
hand there may be variations in geo-
graphical and temporal coverage. It is
necessary to know whether all the sites
belong to the same period and same part
of the country or whether they are more
evenly spread. On the other the methods
used during excavation and analysis have
a significant impact on the degree to
which the site in question can be com-
pared to others. A site which has been
superfícially dug with limited retrieval of
artefacts and faunal remains can only be
compared to certain elements of a care-
fully excavated site. This is easy to rec-
ognize when say late 19th century exca-
vations are compared to the most recent
ones, but there are a number of hazards
when it comes to excavations from the
second half of the 20th century. These
tend to have a scientific character but
when analysed more closely are often
revealed not to have recorded all the evi-
dence which would be expected today
and to include assumptions we would not
make today without supporting evidence.
While the study of past methodologies
therefore has a practical value for mod-
em researchers who want to be able to
use earlier data in a safe and systematic
manner, such a study is also interesting in
itself, for it can be very revealing about
attitudes and assumptions that have
shaped the archaeological discourse in
the past and continue to influence it to
this day.
The present set of papers is the out-
come of a postgraduate seminar held at
the University of Iceland in 2002-2003.
In the seminar these issues were dis-
cussed in relation to farmstead excava-
tions and these will also be the focus
here. From the early 20th century, exca-
vations of farmsteads have overshad-
owed all other types of archaeological
sites in Iceland (e.g. pagan burials,
churches, assembly sites, ports of trade)
and no other site-category encapsulates
as well the development of field methods
in Icelandic archaeology. In the seminar
each student made a study of a farm site
and its excavation history. Two of these
reports are printed here, each represent-
ing an important stage in the develop-
ment of field methods in the mid 20th
century. The other sites discussed in the
seminar were: Foma-Lá and Sandártunga
(Óðinn Haraldsson), Stöng (Guðrún Alda
Gísladóttir), Þórarinsstaðir (Birna
Lárusdóttir), Sámsstaðir (Anna Rut
Guðmundsdóttir), Herjólfsdalur
(Guðmundur Ólafsson) and Ytre Moa in
Norway (Howell M. Roberts). The
results of these studies have informed the
discussion below.
The background to the seminar is a
growing interest within Icelandic archae-
ology in early excavation methods. This
is to a large degree due to a number of re-
excavations of sites dug in the early 20th
century - Hofstaðir in Mývatnssveit and
Eiríksstaðir in Haukadalur being the
most notable cases. In these re-excava-
tions care has been taken to understand
the progress and rationale of the previous
excavations and it has been emphasised
72