more complex when one considers the early Viking Age settlements. Several sites have
been interpreted as producing a mixture of indigenous cellular and Scandinavian
longhouse architecture and/or both indigenous and Scandinavian portable artefacts.
The ‘type site’ for this argument is Buckquoy in Orkney, where combs and pins of
indigenous style (i.e. with local pre-Viking Age parallels) were recovered from
rectangular houses under a mid-tenth-century pagan grave (Ritchie 1974 , 1977 ;
Brundle et al. 2003 ; Thäte forthcoming). However, the co-occurrence of indigenous and
Scandinavian material culture (in the form of objects and/or architecture) early in the
Viking Age has also been argued at three other sites in Orkney (the Brough of Birsay,
Pool and Skaill Deerness) and at Old Scatness in Shetland (Curle 1982 ; Hunter et al.
1993 ; Buteux 1997 ; Forster et al. 2004 ; Turner et al. 2005 ). Moving to the Outer
Hebrides, one must note the added complication of continued pottery production
(rather than a switch to soapstone vessels) in the Viking Age (Lane 1990 ; Campbell
2002 ) and some continuity in architectural methods (the building of semi-subterranean
houses revetted into sand) (Sharples and Parker Pearson 1999 ; Owen 2002 ). Further
south in Argyll, one can note the present lack of characteristically Scandinavian settle-
ment of any date and the likelihood of continued occupation of indigenous sites such as
Iona (as a monastery and perhaps royal burial ground) and Dunadd (as an elite centre)
( Jennings 1998 ; Lane and Campbell 2000 ).
The discrepancies evident in most of the categories of evidence surveyed above
are largely matters of chronology and regional diversity. The geographical issue is one of
unsurprising divergence between the Northern Isles, the Western Isles and Argyll. In
terms of chronology, the place-name, tenth- to thirteenth-century settlement and hoard
evidence (and in other ways the genetic data) all post-date the early Viking Age settle-
ment evidence. The same may not apply, however, to the burial evidence, which
raises the last main reason for the existence of widely divergent interpretations of Viking
Age Scotland – poor chronological resolution. Although the pagan graves have now
been carefully studied and relatively tightly dated (Graham-Campbell and Batey 1998 :
152 – 4 ), Viking Age settlements in Scotland can only be assigned the broadest of date
ranges (see Barrett 2003 : 82 – 8 ). We therefore cannot tell whether or not settlements
with ‘mixed’ indigenous and Scandinavian assemblages predate or are contemporary
with Scotland’s pagan graves – or even if the different elements of the ‘mixed’
assemblages were really used at the same time. Moreover, we do not know whether
‘unmixed’ late Pictish settlements preceded sites with Scandinavian material culture or
were contemporary with them. Similar problems exist regarding the question of
whether Christian and pagan practice were contemporary or sequential in the ninth
and tenth centuries – a complex topic which is also clearly important to the question
of Scandinavian impact in Scotland (Morris 1996 b; Dumville 1997 ; Barrett 2002 ;
Crawford 2002 ).
These issues arise for a variety of reasons. Early Viking Age settlements at the Brough
of Birsay and Skaill both lack convincing stratigraphy – due to antiquarian methods in
the former case and the unfortunate premature death of the excavator in the latter case
(Curle 1982 ; Buteux 1997 ). Another important site, Buckquoy, lacks radiocarbon dates
and there is some debate regarding whether the (only partially preserved) rectangular
buildings in its latest phase are actually of Scandinavian style (cf. Graham-Campbell
and Batey 1998 : 160 – 4 ; Brundle et al. 2003 ). These three settlements – plus Pool and
Old Scatness also noted above – are also multi-period, sometimes making it difficult to
–– James H. Barrett––