ArticlePDF Available

A Viking Boat Grave with Amber Gaming Pieces Excavated at Skamby, Ostergotland, Sweden


Abstract and Figures

IN THE SUMMER of 2005 the authors directed the excavation of a flat stone setting with a boat-shaped central depression at Skamby, Kuddby parish, Östergötland, Sweden. The stone setting covered a small and poorly preserved boat inhumation, dated by the artefacts recovered to the early Viking period (9th century ad). This is the first excavation of a boat inhumation in the province of Östergötland. The paper reports on the excavations including the discovery of an exceptional collection of 23 amber gaming pieces, which provide a new perspective on Viking-period gaming. The data from this boat grave are considered in relation to the rest of the Skamby cemetery, which remains to be investigated. Judging from a topographical survey of the ridge surrounding the excavated area, and from metal-detector finds recovered from the surrounding fields, the Skamby cemetery appears to be a high-status burial ground divided into two zones, one comprised of boat inhumation graves, the other of circular stone settings likely to cover cremation graves. The results of the excavation lead to a revised picture of boat burial as an élite mortuary rite in southern Sweden during the late 1st millennium ad.
Content may be subject to copyright.
© Society for Medieval Archaeology 2008 DOI: 10.1179/174581708x335440
Medieval Archaeology, 52, 2008
A Viking Boat Grave with Amber Gaming
Pieces Excavated at Skamby, Östergötland,
IN THE SUMMER of 2005 the authors directed the excavation of a flat stone setting with
a boat-shaped central depression at Skamby, Kuddby parish, Östergötland, Sweden. The stone
setting covered a small and poorly preserved boat inhumation, dated by the artefacts recovered
to the early Viking period (9th century ad). This is the first excavation of a boat inhumation
in the province of Östergötland. The paper reports on the excavations including the discovery
of an exceptional collection of 23 amber gaming pieces, which provide a new perspective on
Viking-period gaming. The data from this boat grave are considered in relation to the rest of
the Skamby cemetery, which remains to be investigated. Judging from a topographical survey
of the ridge surrounding the excavated area, and from metal-detector finds recovered from the
surrounding fields, the Skamby cemetery appears to be a high-status burial ground divided into
two zones, one comprised of boat inhumation graves, the other of circular stone settings likely
to cover cremation graves. The results of the excavation lead to a revised picture of boat
burial as an élite mortuary rite in southern Sweden during the late 1st millennium ad.
The medieval kingdom of Sweden appears to have come into being about
ad 1000 when two groups known as the Svear and the Götar elected a shared
king, Olof Eriksson skotkonungr.
The fertile lands of these two peoples were
separated by the rugged forests of Tiveden and Kolmården and they retained
different laws, customs and administration for much of the Middle Ages.
political focus of the Svear was in Mälardalen (the territories around Lake
Mälaren), while the Götar occupied two discrete geographical districts to the
west (Västergötland) and east (Östergötland) of Lake Vättern (Figs 1 and 2).
During the later 1st millennium ad, the Svear and Götar should probably not
be seen as clearly bounded and internally coherent political units or ethnic
groups. Instead, they were heterogeneous entities forged by political alliances,
socio-economic networks, communication routes and environmental constraints.
In both regions, kingship was a slowly evolving and perhaps only intermittent
Royal Academy of Letters, Box 5622, SE-11486 Stockholm, Sweden.
Department of History and Archaeology, University of Chester, Parkgate Road, Chester CH1 4BJ,
England, UK.
Gahrn 1988, 29.
Christiansen 2002, 96.
70 martin rundkvist and howard williams
institution until after the élite’s conversion to Christianity. Power appears to
have remained in the hands of powerful local magnate farmers during the later
1st millennium ad.
The accounts of the Geatas and Scilfingas of the Anglo-
Saxon poem Beowulf, in which elaborate funerary procedures involving a boat
are portrayed, may dimly reflect the far-ranging contacts and military exploits
of these local élites from as early as the 5th century ad.
However, the written
sources, particularly for the land of the Götar, are so few and terse that the
field of study is just barely proto-historical. Any reconstruction of these societies
must therefore rely mainly on the archaeological record. Understanding the
Vendel- and Viking-period societies of these regions has focused upon questions
Ibid, 96–7, 131–2; see also Sanmark 2004, 84; Svanberg 2003a and b.
Heaney and Donoghue 2002, 96–7; see also Overing and Osborn 1994; Fabech 2001; Näsman 1991.
fig 1
Distribution map of
Viking-period boat-
inhumation graves in
Scandinavia in relation to
the Skamby site. Box
locates Fig 2. Map by
Howard Williams redrawn
and adapted after Müller-
Wille 1995, 100.
71skamby viking boat grave
concerning social structure, political organisation and the process of religious
Of the two regions, Svealand has received most of the archaeological
attention. Boat-grave cemeteries excavated to the north of Lake Mälaren can
be interpreted as evidence of élites in Svealand during the Vendel and Viking
periods (Figs 1 and 2). This region has been the focus of nation-oriented archae-
ology’s discussions of kingdom formation and religious conversion.
Also, this
region has been most intensively studied from abroad in discussing the links to
mortuary practices and socio-political trajectories in north-western Europe and
the British Isles.
The land of the Götar has also received some attention but
fig 2
Principal sites with Viking-period boat-inhumation graves in central Sweden shown in relation to
the Skamby site. Box locates Fig 3. Map by Howard Williams redrawn and adapted after Nylén & Schönbäck
1994, 129.
Eg Graham-Campbell 1994; see Svanberg 2003a, 78–80.
Carver 2005, 303.
72 martin rundkvist and howard williams
this has, for the most part, focused on Västergötland.
In contrast, Östergötland
has seen much less archaeological investigation and certainly fewer attempts to
synthesise and analyse the available data.
A number of factors have contributed to this uneven spread of archaeo-
logical interest. Partly it is due to the historical circumstances in which archae-
ology emerged in Sweden during the 19th and early 20th centuries. The hub of
the modern Swedish state is in Svealand with the capital city Stockholm,
it was in this region that scholars most frequently sought Sweden’s 1st-millen-
nium origins, particularly in Uppland.
Furthermore, the uneven geographical
distribution of Sweden’s university departments of archaeology has meant that
Östergötland in particular (equally far from Stockholm to the north-east and
Göteborg to the west) has seen relatively few research excavations. To some
extent, the work of two high-quality contract archaeology units in Östergötland
has remedied this situation. Many new settlements, burial and ritual sites of
the later 1st millennium ad have been uncovered, adding important new infor-
mation to our understanding of the region’s socio-political structure.
the ample resources devoted to contract archaeology in advance of settlement
expansion and road building around and between the towns and cities of
Östergötland are painstakingly geared towards the conservation of well-preserved
Hence, the situation has persisted that the province’s cemeteries of
possible Vendel- and Viking-period date, known from abundant surviving sur-
face features, have escaped the archaeological attention afforded to comparable
ones in Mälardalen.
The research reported on here is part of a broader archaeological project
aimed at redressing this imbalance. The work will investigate Östergötland’s
late-1st-millennium ad social structure and political geography. This is to take
place through the synthesis of earlier published and unpublished archaeological
data and the execution of carefully focused, research-led excavations at key
archaeological sites in the region. The research at Skamby, focusing on the only
categorical example of a boat-grave cemetery from the region, forms an element
of this broader project.
The cemetery at Skamby in Kuddby parish was recognised and surveyed
in the 1940s. From surface examination alone the site was identified as one of
Sweden’s rare and characteristic boat-inhumation cemeteries. A number of such
cemeteries have been excavated since the late 19th century in the provinces
of Uppland and Västmanland north of Lake Mälaren, yielding spectacular
finds associated with the upper echelons of late-1st-millennium ad society. They
Eg Fabech 2001.
Unless otherwise stated, all places are in modern Sweden.
See Svanberg 2003a, 36–99.
See Kaliff 1998. Eg the ‘ritual house’ at Borg: Nielsen 1997; see also Andrén 2007a.
Rundkvist 2007a.
Rundkvist in prep.
73skamby viking boat grave
include the sites of Vendel, Valsgärde and Tuna in Badelunda.
Solitary and
less-impressive boat inhumations have been recognised at otherwise typical
Viking-period cemeteries in Södermanland, a Svealand province that adjoins
In contrast, archaeologists have recognised only three cemeteries
containing suspected boat inhumations in Östergötland: Norra Berga in Mjölby
parish, Malm in Styrstad parish and Skamby. While presumed to be late-1st-
millennium ad boat graves from their distinctive surface features, the date and
character of these three sites remained uninvestigated prior to 2003.
Of the three sites, the project selected Skamby as a prime target for
investigation. Key questions included:
What was burial preservation like at Skamby?
What was the date of the cemetery?
Did the boat graves at Skamby compare or contrast with the ones in
What was the level of investment, mortuary symbolism and monumentality
of the Skamby boat graves?
These questions promised to demonstrate the importance of field-focused
research to complement the increasingly sophisticated and diverse theoretical
approaches to Viking-period mortuary practices in Scandinavian archaeology.
They also built upon our wider research themes concerning mortuary and
commemorative practices in the later 1st millennium ad.
Skamby is a hamlet on the Vikbolandet peninsula, near the eastern end of
Östergötland’s plains belt (Figs 1–5). Situated close to the Baltic coast, the 1st
millennium ad inhabitants of the peninsula would have enjoyed extensive
communications by river and sea. Hillforts and innumerable cemeteries attest to
the dense occupation of the region in the 1st millennia bc and ad (Figs 3, 4 and
Near Skamby hamlet is a low rocky ridge orientated from north-west to
south-east and surrounded by ploughland (Figs 5, 6–8).
Along the spine of the
ridge are ten large, oval stone settings with diagnostic boat-shaped depressions
at their centres (Figs 6–8). At the ridge’s NW end is a cluster of small round
stone settings of a kind commonly seen at typical cremation cemeteries of
the later 1st millennium ad in Östergötland and found widely in Kuddby and
neighbouring parishes in Vikbolandet. However, with its mixture of boat graves
and circular monuments, the site looks atypical for the region and very much
Sandwall 1980; Lamm and Nordström 1983; Nylén and Schönbäck 1994.
Eg Weiler 1975; Norberg 1998.
Rundkvist 2007b.
Eg Rundkvist 2003; Williams 2006; 2007.
Registered site Raä 158, Kuddby parish.
74 martin rundkvist and howard williams
like the boat-inhumation cemetery at Valsgärde in Uppland did before the first
Under the auspices of Östergötland County Museum, the project involved
seven weeks of fieldwork during the summer of 2005. The excavations revealed
Östergötland’s first boat inhumation and the first set of Viking-period amber
gaming pieces discovered in Sweden for over a century. Amidst evidence of
settlement activity pre-dating the cemetery, the excavations recovered important
copper-alloy casting debris possibly dating from as early as the 2nd century
bc, with parallels from the élite metalworking island-site of Helgö in Lake
fig 3
The regional context of the Skamby site. The cemetery at Malm in Styrstad parish has surface features
suggesting boat graves. The crosses denote historic church sites that show a clear contrast between the
concentration of historic settlement on the fertile plains to the west of Skamby and the more forested and
rocky lands to the east on the Vikbolandet peninsula, as well as to the north and south of the peninsula.
In broad terms this reflects settlement patterns and land-use coeval with the boat graves. Box locates
Fig 4. Map by Howard Williams.
Ljungkvist 2006, 71.
The pre-cemetery settlement evidence is reported in Rundkvist et al 2007.
75skamby viking boat grave
fig 4
The local setting of the Skamby site. While situated within a wide valley inland from the Baltic coast, the
Skamby cemetery was adjacent to a navigable water-course linking it to the sea. In the Viking period, this
stretch of the Slätbacken inlet’s shoreline was 3.5 m higher and hundreds of metres north of the present
one because of post-glacial shore-displacement. Box locates Fig 5. Map by Howard Williams.
76 martin rundkvist and howard williams
A 2003 metal-detector survey of the ploughland around the Skamby ceme-
tery by co-author Rundkvist had given no finds from the earlier Vendel-period
phase of the boat-inhumation custom (late 6th-8th centuries ad). It did,
however, turn up a number of Viking-period finds of the 9th and 10th centuries
ad along the edges of the cemetery. Fragments of bronze jewellery, a bronze
caftan button and a silver-sheet pendant cross indicate ploughed-out graves of
Viking-period date situated around the ridge. The dates of these finds corre-
spond with the heyday of the boat inhumation custom and suggest that the ridge
with its boat graves once formed the core of a somewhat larger Viking-period
burial ground.
The 2005 season’s excavations selected the cemetery’s smallest stone setting
with a boat depression: grave number 15 (Fig 8). The rationale for this choice
fig 5
The immediate environs of the Skamby ridge showing its relationship to principal concentrations of
ancient monuments. These are mostly cemeteries, each having been used during either the first or the
second half of the 1st millennium ad. The hillfort is likely to date from the period ad 150–550. Map by
Howard Williams.
Müller-Wille 1969; 1995.
77skamby viking boat grave
to restrict the amount of archaeological work needed to answer the key
research questions identifi ed (ie level of preservation, date and nature of
the grave) while keeping the disturbance to the cemetery as a whole to a
unlike some other graves at Skamby that showed evidence of previous
disturbance, grave 15 bore no discernible surface traces of looting or
unrecorded antiquarian interventions;
grave 15 was one of only two entirely turf-covered boat graves on the site,
which offered the possibility of better fi nds preservation compared with
other graves, where conditions among the exposed stones in the central
depressions are clearly extremely poor due to the perpetual percolation of
grave 15 lay in the centre of the zone of boat graves and it is carefully
spaced in relation to the graves on either side. This suggested that grave
15 would date to the main period of the cemetery’s use and would not be
a chronological outlier.
Given there was no clear relationship between grave wealth and boat size among
the previously excavated graves at Swedish boat inhumation cemeteries,
fig 6
Aerial view of the Skamby
cemetery from the NNW,
14 August 1991. Photograph by
Jan Norrman.
Lidén et al 2001.
78 martin rundkvist and howard williams
was no reason to believe that the relatively small size of grave 15 at Skamby
would mean that the grave was unusually or poorly equipped.
the monument
The grave’s central depression measured 5.0 by 1.5 m on the surface and
was orientated NESW (ie orientated perpendicular to the NWSE orientation
of the ridge). Beneath the turf and topsoil, an irregular octagonal pavement
of stone blocks measuring 11.5x9 m was uncovered (Fig 9). The stones were
mostly local granite (a mixture of pink, grey and white) with some sandstone.
The monument was mostly composed of a single layer of stones. Most of them
could be carried a short distance by a single adult, although a minority were too
large to carry and had to be rolled by hand. A few of the stones were too large
for even two people to roll and are likely to have been glacial erratics occurring
naturally on the site or moved only a short distance employing draught
An orthostat had been standing at the NW side of the grave-cut, but had
fallen into it, perhaps with the decay of a postulated timber roof or cover to the
fig 7
The Skamby boat-inhumation graves prior to excavation. View from the south-east, March 2003.
Photograph by Martin Rundkvist.
79skamby viking boat grave
grave. Three large stone blocks on the edges of the stone pavement may
also originally have been standing up. The edge stones of the pavement did not
form a distinctive kerb, although larger stones were more common toward the
periphery of the stone setting.
fig 8
A plan of the Skamby cemetery from survey work conducted by Rundkvist and Williams in 2005. Map by
Marcus Andersson.
80 martin rundkvist and howard williams
fig 9
The superstructure of grave 15. Vertical photo collage by the authors.
81skamby viking boat grave
The collapse of a perishable cover over the grave-cut containing the
boat clearly created the central depression. Stones from the superstructure
slumped inwards and filled the grave-cut. They showed no sign of any subse-
quent disturbance, ruling out the possibility of grave-robbing since the Viking
the grave
The upper part of the grave-fill was indistinguishable from a surrounding
culture layer belonging to an earlier settlement. This meant that it was not
possible to document the upper edge of the grave-cut with any certainty. Only
the lower, crudely boat-shaped edges of the grave-cut were visible where it cut
into the natural subsoil. Radiocarbon (14C) dates, from sunken features sealed
by the settlement deposit, place occupation most probably in the 2nd century
bc; these features were not related to the boat-grave (contexts 15 and 20 in
Fig 10 belong to this group but were not dated).
Despite the careful choice of monument for excavation and the observation
that it had not been recently disturbed, the preservation within the grave-cut
proved very poor. Judging from rust stains, preserved clench nails and sections
through the boat-shaped grave-cut, a boat had clearly originally been present.
However, the underlying moraine was clayey and nearly impermeable to water.
Rainwater had repeatedly accumulated and evaporated here, consequently,
no unburnt bone or other organic remains were found. Likewise, the grave
preserved little iron.
The boat had originally measured c 5 m in length and c 1.7 m across at
its broadest point. It was not possible to discern any detailed pattern to the rust
stains and preserved clench nails (Fig 10), in contrast to many other excavated
boat inhumations. However, the number of clench nails in itself clearly suggests
that a clinker-built vessel was placed in the grave.
Little else was discernible of the boat’s characteristics. However, it is worth
noting that the medieval provincial law code of Östergötland, codified in the
late 13th century, mentions a similar boat in the section on land rights.
describes a procedure to determine the border in a body of water between a
private farmstead and the commons of the hundred:
If farmstead and commons meet in water: then take a nine-ell boat and put it with one stem
in the reeds and one out on the deep; and a man shall stand in the rear stem holding a
barge pole in his hand and throw it over his shoulder out into the deep; thus far shall the
rights of the farmstead extend, as he can throw, and the commons are outside.
The ell used in medieval Östergötland measured 53.9 cm. A nine-ell boat
would have measured 4.85 m. This suggests that the grave contained a craft that
may have been widely employed for transport upon the Baltic inlets and rivers
of the Östergötland region. The other nine boats buried at Skamby, however,
Rundkvist et al 2007.
Östgötalagen, Byggningabalken §28:3.
Rundkvist’s translation into English of Holmbäck and Wessén’s 1933 translation into modern Swedish of
the Östergötland law code, 217.
82 martin rundkvist and howard williams
fig 10
Boat-grave burial plan. Drawn by Marcus Andersson, adapted by Howard Williams.
83skamby viking boat grave
appear to have been longer. Estimating from the surface features, the largest
boats may have approached 10 to 12 m in length, commensurate with the
longest boats uncovered at Vendel and Valsgärde in Uppland.
It therefore
appears that burial at Skamby may have employed both smaller river-going and
coastal vessels as well as larger sea-going ones.
grave finds
Just south-west of the mid-ship was a cluster of 23 well-preserved amber
gaming pieces (Fig 11), some located on top of collapsed stones. If this placement
was not the result of Viking-period grave-robbing, then the gaming set was
probably originally placed on top of the grave’s cover or roof rather than with-
in the burial itself, and had fallen with the stones into the burial space when
the roof collapsed. Beneath the gaming piece cluster, a group of iron rivets and
nails lay on the bottom of the cut. They may represent a box or a game board,
although they formed no observable pattern and there was no sign of the
L-shaped mounts typical for Viking-period game boards. Small curved frag-
ments of iron rods here may be from rivets, nails or a simple strap buckle. A
small spherical stone was also found here. It is unlikely to have functioned as a
24th gaming piece, as it is much smaller than the amber pieces and has no flat
face to keep it from rolling off the game board.
Larsson 2007, 46.
Ibid, 71–5.
fig 11
Twenty-three amber gaming pieces. Photograph by Martin Rundkvist.
84 martin rundkvist and howard williams
Other artefacts found in the grave are few and modest, belonging to two
functional spheres: personal items and horse gear. The personal items are a
red glass paste bead and a small slate pendant whetstone. Both lay beneath the
gaming piece cluster at the centre of the boat. They may have accompanied a
body (or bodies) of which nothing remained due to the soil conditions. There
is also part of a small iron knife, found in a superficial part of the grave-fill
mid-ship. This is possibly a residual piece redeposited from the culture layer or
a further artefact placed on the grave-covering.
A highly incomplete set of horse gear lay on the bottom of the cut in its
SW half (Fig 12). There is a very finely wrought hook (from one of the shafts of
a sleigh or small wagon), five frost nails (used to keep the horse from slipping
when ridden or yoked to a sleigh in winter), and two iron rings of identical and
rather small size, one of them with a straight iron bar looped onto it. The rings
look somewhat like pieces of a bridle bit, but are smaller than normal bridle
rings of the time. In the fill of the grave-cut were also numerous small residual
pieces of various materials, principally burnt clay from fired house walls, origi-
nating in the surrounding settlement deposit.
fig 12
Ironwork from horse gear: two rings (possibly from a bridle bit), a hook from the shaft of a sleigh or small
wagon, five frostnails (of which only the top one is completely preserved). Photograph by Martin Rundkvist.
85skamby viking boat grave
date of the grave
The range and character of the artefacts, the boat and monument square-
ly date the grave to the Vendel or Viking period. Only the gaming pieces
hint at a more refined date for the grave within the late 1st millennium ad. The
size and proportions of the Skamby gaming pieces (median diameter 36 mm,
median height/diameter = 0.67) find their closest parallels among bone and
antler gaming pieces from the early Viking period.
Lindquist’s 1978 study of Viking-period gaming pieces treated bone, antler
and glass examples together despite their rather dissimilar designs. This prompt-
ed the misleading statement that Viking-period gaming pieces would generally
have a lesser diameter than Vendel-period ones. This is in fact true only for the
glass pieces that bring down the mean. Instead, Viking-period bone and antler
gaming pieces have about as large a mean diameter as those of the Vendel
period, although they are far more consistent in their proportions. The distin-
guishing feature of Viking-period bone gaming pieces is their greater relative
average height (mean height/width = 0.68 for the Viking period, 0.56 for
the Vendel period, 0.53 for the Migration period, 0.25 for the Roman period).
Scandinavian bone gaming pieces became larger and less squat as the centuries
passed from the Roman period to the Viking period.
The simple domed cross-section of the Skamby gaming pieces, where the
width is greatest at the base, indicates an early date within the Viking period,
c ad 790900. Conversely, the Birka graves, most of which date from the
interval ad 90070, predominantly yielded pieces with narrowed bases, many of
them resembling mushrooms. This applies to the single set of amber pieces from
the Birka graves,
which means that those pieces are not very similar to the
ones from Skamby. In terms of context within a boat grave, there is a close
parallel with the discovery of 23 bone gaming pieces within boat grave 1 at
Gamla Uppsala, dated to c ad 900. The form of these gaming pieces more
closely resembles those from Birka and suggests a later date than that of the
Skamby burial.
We do not know when in the sequence of boat burials at Skamby the inter-
ment uncovered in 2005 occurred. If we assume that only select individuals
in a family or community received boat burial, as has been suggested for the
Mälardalen cemeteries, then perhaps one person per generation received the
rite. If the generation length was c 25 years, it is likely that a cemetery with ten
boat inhumations endured for up to 225 years, from the first burial to the last.
It is therefore likely that boat burial began at Skamby with the beginning of
the Viking period (c ad 800) and ceased with the conversion of Östergötland’s
population to Christianity, possibly from the early 11th century (c ad 1025).
However, a far shorter chronology for the site’s duration is also possible.
Lindquist 1978, 17–21.
Arbman 1940–3, 160–1, Taf 149.
Nordahl 2001, 18.
86 martin rundkvist and howard williams
The superstructure of grave 15 appears to have respected the grave adjoin-
ing it to the north-west, and the graves to the north-west are in a single line on
the highest part of the ridge. These are therefore presumably all earlier. It is
tempting to suggest that graves 10 and 12, placed closest to the circular stone
settings on the top of the ridge, were the earliest, and that boat graves were
sequentially added along the ridge moving downslope from the north-west to the
If this hypothesis is accepted, then the boat graves to the south-east
can be regarded as later in date. Hence, graves 18 and 20 were probably the
last boat graves in the sequence. These graves are the only ones to break with
the linear arrangement charted from graves 10 to 19. They also seem to respect
the locations of graves 17 and 19 and lie in the lowest position on the ridge. The
2005 grave therefore appears to have been near the middle of the site’s boat-
burial sequence. The dating of the gaming pieces appears to agree broadly with
the grave’s spatial location.
age and gender
Given the absence of surviving skeletal material and examples of
multiple interments in boats elsewhere in Scandinavia and beyond, there is a
possibility that the Skamby grave may have contained more than one burial.
However, without clear evidence to the contrary and given the precedent of the
majority of Scandinavian boat graves, the discussion here will assume a single
The provision of a boat might suggest an adult male, although a significant
minority (just under 30%) of Scandinavian boat graves contain female inter-
Rites varied between sites, with boat graves at Tuna in Badelunda
being exclusively female,
while at Vendel and Valsgärde they were exclusively
male. The presence of a boat is therefore suggestive of an adult and possibly a
male, but is by no means conclusive as an indication of the social identity of the
grave’s occupant.
Other sites document horse sacrifice associated with boat inhumations.
However, at Skamby no bones were preserved and the grave-cut was not very
spacious. In Mälardalen, horses were apparently killed at the grave-side and
allowed to collapse into the cut, ending up either in the boat or wedged between
its gunwale and the earthen wall, the cut being locally widened or lengthened
for that purpose. Given the overall association of horse sacrifice with male-
gendered Viking-period graves, the retrieval of artefacts associated with horses
might be seen as a further indication of an adult male burial.
However, the
artefacts actually recovered may equally allude to travel by wagon or sleigh, a
mode of transport associated with female-gendered mortuary symbolism in
Viking-period Scandinavia.
A chronological progression downslope has been identifi ed at Valsgärde: Ljungkvist 2006, 152.
Mülle-Wille 2007, 290; Owen and Dalland 1999; Stalsberg 2001, 379–80.
Andrén 1993, 43–5.
Nylén and Schönbäck 1994. See also Stalsberg 2001, 378.
Larsson 2007, 367–74.
Gräslund 1981, 49.
Andrén 1993, 46.
87skamby viking boat grave
In the graves of late-Roman-, Migration- and Vendel-period Scandinavia,
gaming pieces are mainly associated with the graves of adult males.
With the
advent of the Viking period, gaming pieces increasingly enter the female realm,
becoming gender-neutral in rural Swedish cemeteries.
They are, however, still
associated with adult males in the rich cemeteries of the early town of Birka and
in Norwegian graves.
Therefore, the gaming pieces are once more suggestive,
but by no means conclusive, of an adult male interment. Co-author Rundkvist
has documented male-gender connotations for single beads on late-Viking-
period Gotland.
Small slate whetstones are ubiquitous in late-1st-millennium
ad burials in Sweden, and appear to be gender-neutral.
The monument may also reveal clues as to the identity of the occupant.
Petré has suggested a gendered morphological dichotomy in terms of the central
marker stones on late-1st-millennium ad funerary monuments, echoing sexually
diagnostic features of human anatomy.
Standing stones tend to accompany
male-gender artefact kits and stone spheres female ones. The central orthostat
at Skamby suggests a male gender association.
Combining these strands of evidence, the Skamby 2005 boat grave was
probably the burial of an adult male. Viking-period weaponry, which is
very strongly male-gendered in Scandinavian graves, is exceptionally rare in
Östergötland, and its absence from Skamby therefore does not call for special
explanation. Inevitably, the osteological age and sex of the deceased must
remain unknown in the absence of preserved human bones.
social status and possible early grave robbing
Was the Skamby 2005 burial a wealthy grave, indicating an aristocratic
status for the deceased and the mourners? This question is problematic but the
answer is a cautious ‘yes’.
The poor preservation of organic materials and metalwork means that
many artefacts and materials (including any parts of sacrificed horses, dogs or
other animals) found in boat graves elsewhere were not discernible during the
Skamby excavation. Therefore, the relative poverty of Skamby in comparison
to some of the Mälardalen cemeteries may partly be a matter of differential
preservation rather than differences between burial rites or mortuary
Furthermore, grave robbing is a well-recognised practice among Scandina-
vian boat and ship graves, complicating the assessment of relative burial wealth.
The intact nature of the Skamby stone setting with its distinctive line of stone-
collapse seems to rule out any grave robbing in later decades and centuries,
after the grave-cover had rotted through. However, in some cases, grave robbing
clearly took place in the Viking period itself within months or years of the
M Sandberg 1994; Rundkvist 2003, 36–42, 47–54.
M Sandberg 1994; contra Petré 1993.
Eriksson 1993, 12–13; Solberg 2007, 266–7.
Rundkvist 2003, 61.
Petré 1993.
88 martin rundkvist and howard williams
funeral, when both the boat and its roof were intact. Archaeologists may not see
such robbing during excavation because it may have only necessitated the
removal of a few stones. Disturbance would have been further minimised if those
doing the robbing had knowledge of the funeral and the grave’s arrangement.
In any case, it is possible from the flat stone setting, grave-roof and monolith
that the position of the boat would be clear for all to see. Such hypothetical
robbing may have focused on items of greatest value, such as weapons and
feasting gear.
The Skamby grave with its scattered gaming pieces and the lack of either
brooches or weaponry provides a striking parallel with the c ad 900 boat grave
no 1 at Gamla Uppsala, reported by Else Nordahl.
Here, better preservation
enabled the recovery of the skeleton of an adult male as well as animal bone
including the remains of two dogs and a horse. Grave robbing was evident from
the prone posture and peripheral location of the human skeleton. The post-
robbery scatter of artefacts included 23 gaming pieces (here of bone) and the
presence of token items of horse gear means that, had preservation been poorer,
the grave would have left a nearly identical trace to the remains recovered at
Skamby. Hence, the Skamby grave might be simply a less well-preserved version
of the Gamla Uppsala burial: a case of the Viking-period grave robbing of a
high-status adult male furnished burial in a boat beneath a stone setting.
However, the poor preservation of the Skamby find and the lack of any
evidence for the disturbance of the stone setting (in contrast to Gamla Uppsala)
leave open the possibility that the Skamby grave was indeed unrobbed. More-
over, the mixing of stones and gaming pieces tends to rule against the idea that
their distribution was the result of Viking-period robbing. Whether the grave
itself was robbed or not, the temptation is to interpret the gaming-piece scatter
as evidence of objects placed on top of rather than in the grave and subsequently
falling into it with the collapse of the roof.
Furthermore, it is important to question the expectation that aristocratic
male burials should contain feasting gear and weapons. Östergötland’s 1st millen-
nium ad graves are generally not richly equipped, and Viking period male buri-
als there are particularly austere. Therefore, for Östergötland, what is notable
is the presence of burial riches (namely the gaming pieces), not the absence of
Therefore, whether or not grave robbing took place, the presence of
gaming pieces indicate a high-status funeral. Within graves of late-Roman-
period Scandinavia (ad 150375), gaming pieces strongly correlate with expen-
sive burials.
In the Migration, Vendel and Viking periods (ad 3751100), they
no longer confine themselves to such contexts: the average grave expenditure is
This has to do with the proliferation of bone gaming pieces from the
Nordahl 2001, see also Klevnäs 2007, 29–32.
Klevnäs 2007, 36–9.
M Sandberg 1994.
Ibid; Rundkvist 2003, 43–4, 55–7; contra Solberg 2007, 267.
89skamby viking boat grave
Migration period onward. Gaming pieces of glass or amber generally occur only
in richly furnished burials. The Skamby 2005 boat grave is in fact one of the
poorest-equipped members of this group, while gaming with high-quality pieces
suggests a link with an aristocratic (and even ideals of a heroic) lifestyle.
possibility of grave robbing aside, the scattering of gaming pieces, if interpreted
as evidence of gaming pieces placed on the roof of the grave, may indicate that
they were gifts from mourners rather than traditional ‘grave goods’ situated with
the cadaver.
Provisionally ruling out grave robbing, and given that frost nails survived
at the bottom of the grave where rainwater collected and preservation conditions
were particularly poor, the absence of other metalwork is a clear indication that
the grave did not originally contain any weaponry or other large iron objects
such as stirrups, spurs or kitchen and feasting gear. In this sense, the Skamby
grave is clearly inferior to the richest intact coeval graves found at other
Swedish boat inhumation cemeteries. On the other hand, no other Swedish boat
grave contains a comparable amount of amber, an expensive material imported
from the SE shores of the Baltic or the SW coast of Jutland. The nearest known
site with evidence of contemporaneous amber working, including the making
of gaming pieces, is Birka on an island in Lake Mälaren.
Rich similar craft
finds are known from Ribe on the Jutish amber coast (Denmark), Hedeby in
Schleswig-Holstein (Germany) and Dorestad in Frisia (Netherlands).
Mortuary technologies further bias our assessment of the grave’s relative
wealth. Most Swedish Viking-period graves are cremations. Amber burns easily
leaving a slight residue indistinguishable from that of resinous pine wood. The
cremation of the Skamby 2005 assemblage prior to burial would have entailed
a greater expenditure than the investigated inhumation. But cremation would
have obliterated the amber and, even if a greater number of clench nails had
survived in such a cremation deposit due to fire patination, it would have
appeared to us as a poorly equipped burial. Hence, while undeniably possessing
high-status associations, the unique status of the amber gaming pieces may owe
as much to the choice to inhume the dead at Skamby as to the rarity of the
objects themselves.
Stone superstructures of the form identified at Skamby are unknown from
other Swedish boat inhumation cemeteries and represent quite a respectable
labour expenditure. More important is the size of the monument; here status
was afforded not through the building of a large barrow to serve as a landmark,
but by demarcating an exclusive space through the construction of the stone
setting that framed the grave. The monument is also distinctive for its unusual
oblong and irregularly octagonal shape. Rural Swedish Viking-period cemeteries
are usually dominated by modestly sized mounds and flat round stone settings,
with a few four-sided, boat-shaped and triangular stone settings occurring too.
Jennbert 2006, 136.
Arwidsson 1989b.
Botfeldt and Brinch Madsen 1991; Jensen 1991, 41; Schietzel 1981, 75–6; Kars and Wevers 1983.
Gräslund 1981, 67–70.
90 martin rundkvist and howard williams
Such shapes may have embodied social information about the deceased
and religious symbolism.
While the precise choice of an irregular eight-sided
monument at Skamby remains enigmatic, it might be an adaptation of local
monument forms rather than a radical innovation. While both boat inhumation
and oblong monuments are extremely rare, flat stone settings (although gener-
ally built with smaller-size stones) are the rule in both Mälardalen and Östergöt-
land at the time, regardless of grave contents. This choice of rite places the boats
beneath stone settings somewhere between the established categories of ‘boat
grave’ and ‘stone setting’. Building the monument covered the boat from view,
but only by a modest layer of stones that may have allowed it to be distinguished
from the surface.
Therefore the choice of employing a boat in funerary ritual
at Skamby (itself a considerable investment) was followed up by a distinctive
variant in monument form and hence a means of marking the grave as different
from other contemporary burials.
The Skamby 2005 burial presents an unbalanced inventory, caused either
by the initial choices of the mourners or by early robbing, itself possibly per-
formed by the same mourners. It presents a single very expensive set of objects,
a boat and a distinctive monument combined with relatively unassuming traits.
Overall, the grave represents above-average expenditure in goods and labour,
yet far from top level for its time in southern and central Sweden. For the region
however, the grave is unique in qualitative as well as quantitative terms. For
instance, the distinctive choice of inhumation in a largely cremating region and
the choice of a boat may themselves have constituted a statement of wealth
and status, setting the Skamby graves apart from other contemporary mortuary
practices and therefore likely to have served to mark aristocratic status. Equally,
the location and form of the monument distinguished the social standing of the
deceased and mourners even more than the materials and artefacts invested in
the grave itself. Since it is clear that the boat grave was not the first on the site,
the mourners may have deemed status-by-association with an enduring tradition
on the same location, and the use of an abbreviated symbolic message, to be
sufficient to evoke their social status and that of the deceased.
It is likely that the mortuary symbolism as well as the expenditure and form
of the Skamby grave also served as a social statement. The presence of gaming
pieces may have formed a clear statement of both status and mythological
beliefs. Likewise, Viking-period Scandinavian death rituals seem to have
employed boats and their representations as symbolic allusions to afterlife
journeys, as well as to symbolise heroic worldly ideals.
The presence of horse-
related material culture may allude to the same symbolism of both social status
and perceptions of an afterlife journey. The Skamby grave was therefore
an exclusive statement that defined the transformed identity of the deceased in
relation to a vision of the afterlife as well as aristocratic culture.
Eg ‘tricorns’ (three-cornered stone settings) that may evoke the idea of the world-tree Yggdrasill known from
Norse mythology: Andrén 2007a, 125.
Andrén 1993, 45.
Andrén 1993.
91skamby viking boat grave
With such a poorly preserved grave, conclusions concerning the affinities
of its occupant(s) must remain somewhat vague. As Lidén et al have pointed
the Svealand boat-inhumation custom was in any case not strictly
uniform. There is considerable variability even when comparing coeval graves:
‘The overall impression is that the major common, and maybe the only com-
mon, denominator is the boat’ to which should be added the choice of
the inhumation rite at a time when cremation was the dominant practice.
Nevertheless, the grave excavated at Skamby does stand apart from the ones in
Mälardalen with a number of unique features. At Skamby we see a large, distinc-
tive stone setting with at least one standing stone. Unless robbed, there were
no weapons or feasting gear, the horse gear is minimal, and then somewhat
incongruously there are rare and exclusive amber gaming pieces, possibly placed
over, rather than within, the boat.
regional context
How did the Skamby cemetery fit into its region? The answer is that it did
not. As with other parts of late-1st-millennium ad Scandinavia, Östergötland’s
mortuary practices are varied.
However, most Viking-period graves excavated
over the decades in the environs of Skamby on the Vikbolandet peninsula and
the plains of Östergötland to the west are cremations surmounted by small
mounds or stone settings. As repeatedly stated, Viking-period male burials in
Östergötland are rarely rich in grave goods. Most of the six known weapon
graves of the period are early finds with poor documentation. Some members
of the era’s élite appear to rest under great barrows judging from two that
have been trial-trenched and 14C-dated.
None of these have, however, been
excavated in full, and their burials’ gender affiliation is unknown. An enormous
oval barrow at Aska in Hagebyhöga parish has been trial-trenched and 14C-
dated to cal ad 660880 (1 sigma),
but in this case it is uncertain whether the
monument is a grave barrow or a building platform, and it most likely pre-dates
the Viking period. Certainly, boats may have been more widely employed
in late-1st-millennium ad cremation ceremonies than has previously been
recognised in southern Sweden,
but the deployment of boat inhumation is
exceedingly rare.
Surface features indicate that the Malm (Fig 3) and Norra Berga cemeter-
ies also contain boat graves. However, neither site resembles Skamby. Instead
they have individual boat graves integrated into cemeteries comprised of a range
of local, late-1st-millennium ad grave forms. They also lack evidence of any
linear sequence of boat graves suggesting an enduring and organised tradition.
Lidén et al 2001.
Andrén 2007a, 115; Svanberg 2003b.
Ledberg, St-8656, 1140±80 BP; Sjögestad, Poz-18593, 1210±30 BP.
St-11326, 1270±100 BP; Claréus and Fernholm 1999.
Nicklasson 1999.
92 martin rundkvist and howard williams
Therefore, in its regional context, the Skamby cemetery is unique, being similar
to the Mälardalen cemeteries.
From this evidence, it is possible to suggest that the cemetery constituted a
distinctive and overt symbolic statement through the choice of inhumation, boat
burial and (to some extent) the material wealth and symbolism deployed in
the grave. It had at least one standing stone (possibly originally four including
those along the edges of the stone setting), and it was part of a series of visually
uniform boat burials. Therefore, rather than a one-off occurrence, it was part
of a distinctive mortuary expression by an élite social group perpetuated for
select individuals over decades and perhaps centuries.
inter-regional context
Elsewhere along the coasts of the Baltic, boat graves have been regarded
as evidence of the cultural origins and world-views of Swedish immigrants
practising the rite,
or of shifting political connections between centres (ie
Svealand) and ‘peripheries’.
Archaeologists have adopted a similar perspective
for the Atlantic context, with the similarities between the boat-grave traditions
of western Norway with Scotland and Man perceived in terms of migration and
dynastic links between homelands and newly settled territories.
boat graves are regarded as an element of a pan-Scandinavian (supra-regional)
élite mortuary repertoire without a single point of origin; boat burial, in this
view, transcended local variability in mortuary practices.
For Skamby, these models are difficult to apply fully and adequately. To
regard the Skamby cemetery as serving a settlement of immigrants from, or
subordinates to, a dynasty from Uppland would be to ignore the local adaptation
of the rites in terms of both grave and monument. Moreover, the Vikbolandet
peninsula is strategically positioned between Östergötland’s heartlands to the
west and excellent seaborne communications linking it to all areas of the Baltic,
including Mälardalen to the north and Öland to the south (Figs 1–2).
It was
in this sense a micro-region in its own right. In such a situation, dynastic and
political connections with Svealand are certainly likely, yet there is no evidence
to indicate that these would have been exclusive. Even if they were, boat graves
were not the only choice available to the community burying in the Skamby
cemetery. Hence the choices of the local élite’s mortuary customs need not
be regarded as ideas exported from a single centre (such as Uppland) and
passively adopted by a peripheral region, nor simply as a permutation of an
ubiquitous Scandinavian élite rite. Neither are the grave and the monuments’
unusual designs likely to have been due either to poverty or to ignorance of
customs elsewhere.
Gerds 2006, 157.
Andrén 2007b, 298.
Müller-Wille 2007, 294.
Eg Andrén 1993; 2007a, 123; see also Svanberg 2003a, 142; 2003b, 130–4.
Näsman 1991, 29.
93skamby viking boat grave
Instead, it is likely that the social group interring select individuals at Skam-
by adapted the prestigious boat-grave symbolism of the élites of neighbouring
coastal regions, but did so in their own way and for their own purposes.
mortuary repertoire evoked (but did not copy) aristocratic customs from Svealand
and elsewhere in the Baltic region, and it contrasted with the cremation
rites and mound burials employed within elsewhere in Östergötland. Indeed,
the relatively peripheral locations of the long-lived Vendel and Valsgärde
cemeteries in relation to Lake Mälaren finds a parallel in the peripheral (and yet
simultaneously strategic) location of Skamby on Vikbolandet in the far east of
Östergötland and in proximity to the Baltic coast (Figs 3–5).
The gaming pieces (Fig 11) have so far been described and discussed
in terms of their age, gender and status associations. Here, this striking find is
considered further and its implications for understanding the mortuary practices
at Skamby are addressed.
Gaming pieces occur in Scandinavian graves from the late Roman period
through the Viking period until the abandonment of furnished burial (c ad
1501000). Most were made of bone or antler with occasional glass pieces
occurring in richly equipped graves, for example, the Viking-period ones on
Amber gaming pieces are rare and tend to come from boat and ship
graves (Tab 1; the decontextualised Østby find harks from the vicinity of Kaupang
with its numerous boat inhumations). Their rarity may in part reflect the
low likelihood of their identification in the ubiquitous cremation burials of the
period. Yet the repeated association with wealthy graves suggests that amber
playing pieces denoted the power and connections of the mourners and
The Skamby grave contained 23 amber gaming pieces. Taphonomic factors
complicate a reliable assessment of what number would be normal for use:
cremation shatters many bone or antler gaming pieces and preserves the frag-
ments, while soil chemistry may destroy unburnt bone in inhumations. The
best opportunities to find out how many gaming pieces there should be to a
Viking-period set are offered by inhumation graves with glass gaming pieces or
occasional good bone preservation (Tab 2; gaming pieces are unknown from
the Viking-period graves of Gotland, where calciferous soil provides good
preservation conditions).
The median number of gaming pieces per grave in this sample is 25.
Therefore, there is statistically speaking no reason to suspect that there had
originally been any additional bone or antler gaming pieces in the Skamby
grave, obliterated by the soil conditions. All the Skamby pieces are roughly the
On contemporaneous contrasting élite mortuary practices in central Sweden, see Ljungkvist 2006, 190.
On the ‘liminal’ locations of select 1st millennium ad élite graves on a regional scale, see Carver 1986;
Williams 1999.
Lindqvist 1984.
94 martin rundkvist and howard williams
Table 1
Inv no refers to museum fi nds ID number; SHM = Statens Historiska Museum,
Stockholm; NO = Norway; SE = Sweden
Inv no Location No Date, structure Ref
Oslo 11772 NO, Vestfold, Tjølling
parish, Østby
12 10th-century, no
Petersen 1914, fi g
Oslo 23457 NO, southern part of
1 No documentation Dr Augestad’s
Oslo 24743 NO, Hedmark, Furnes
parish, Alu
2 Grave mound, no
Oslo 35251 NO, Østfold, Trøgstad
parish, Grevegg
1 9th/10th-century
cremation grave
Gaming piece
untouched by fi re
Bergen 5150 NO, Sogn og Fjordane,
Gloppen parish, Hauge
1 9th-century boat
Petersen 1914, fi g
Bergen 4438 NO, Rogaland,
Torvastad parish,
18 10th-century ship
Shetelig 1912, fi g
SHM Bj 524 SE, Uppland, Adelsö
parish, Björkö, gr. 524
15 10th-century
chamber grave
Arbman 1940-3,
160-61, Taf 149
Kalmar County
Museum 904:1-2
SE, Småland, Ljungby
parish, Harby nr 5,
c 13 No documentation Magnusson 2001,
543, 576-9
Pending allocation SE, Östergötland,
Kuddby parish,
Skamby, monument 15,
23 9th-century boat
Rundkvist and
Williams 2006
Table 2
Inv no Site Number of pieces, material
SHM Bj 523 Adelsö parish, Birka, grave 523 20 glass
SHM Bj 524 Adelsö parish, Birka, grave 524 15 amber
SHM Bj 581 Adelsö parish, Birka, grave 581 28 antler
SHM Bj 624 Adelsö parish, Birka, grave 624 27 antler
SHM Bj 644 Adelsö parish, Birka, grave 644 22 glass
SHM Bj 750 Adelsö parish, Birka, grave 750 26 glass
SHM Bj 886 Adelsö parish, Birka, grave 886 25 bone
SHM Bj 986 Adelsö parish, Birka, grave 986 17 antler
SHM 20348 Långtora parish, Långtora 26 (21 glass, 4 bone, 1 soapstone)
same size, shape and colour, except for find no 382 that is visibly larger than
the others. This is a common occurrence in gaming sets of the time: one piece
is often visibly distinct. One of the Harby amber gaming pieces from Småland
is the bust of a bearded man. This observation tallies well with the rather sketchy
information given in the Icelandic literature about a game called hnefatafl, ‘hand
95skamby viking boat grave
Here, the distinctive piece is ‘the king’. In a game of hnefatafl, the
objective seems to have been for the king and his retinue to fight their way out
of a surrounding hostile troop and off the board or to the corner squares.
Gaming sets from 1st millennium ad Scandinavian graves, including
Skamby, are rarely divisible into visibly distinct teams for two players. This
remains an unsolved problem despite long discussion. Petersen suggested that
some of the pieces in the uniform bone or antler sets had originally been paint-
ed, therefore distinguishing them as a team.
But this seems unlikely with amber
playing pieces and impossible with glass ones. Sandberg’s attempt to divide the
gaming pieces from Birka into teams based on small visual differences is not
Instead, it can be suggested that the gaming sets found in graves
were intended for only one player, and that the Viking-period custom was
for each player to bring their own set of pieces to the board. This would add
another aspect of friendly competition to the game playing: first one player
would impress the other with the fineness of his pieces, then the other could try
to revenge herself by winning the actual game.
As we have seen, the evidence from Viking-period inhumation graves indi-
cates that one player would possess c 24 identical pieces plus a king. In 1732,
Carolus Linnaeus documented a game played among the Saami.
They called
it tablut (cf tafl) and played on a 9x9 square board with a king at the centre.
Tablut may hypothetically have preserved some or all features of Viking-period
hnefatafl. In the reconstructed hnefatafl, the defensive player has some multiple of
four identical pieces plus one king, while the attacking player has twice the
number of identical pieces but no king. What multiple of four is used depends
on the size of the board (Tab 3).
If, therefore, the 23 Skamby playing pieces were for use by one player in
a game of (reconstructed) hnefatafl, they could, at a pinch, be used to play
attacker on up to an 11x11 square board, or to play defender on up to a
15x15 board. 17x17 boards were probably rare, as sets of at least 48 identical
gaming pieces are very uncommon. Also, a 17x17 board with 4 cm squares
Table 3
Board size Defender Attacker
7x7 sq (28 cm) 1x4=4+king 2x4=8
9x9 sq (36 cm) 2x4=8+king 4x4=16
11x11 sq (44 cm) 3x4=12+king 6x4=24
13x13 sq (52 cm) 4x4=16+king 8x4=32
15x15 sq (60 cm) 5x4=20+king 10x4=40
17x17 sq (68 cm) 6x4=24+king 12x4=48
Berglund 1971 with references; Reallexikon 1978, ‘Brettspiel’; Parlett 1999 with references.
Petersen 1914, 89; cf Grieg 1947.
C-A Sandberg 1976.
Linnaeus 1995.
96 martin rundkvist and howard williams
(large enough to accommodate the largest Skamby piece) would measure over
68 cm a side including the edges. This is more than any known Viking-period
gaming board. The Gokstad ship burial (c ad 900, Norway) contained a 13x13
gaming board measuring 69 cm on one side including edges.
A late-10th-
century Insular-Scandinavian-style gaming board found at Ballinderry, Ireland,
measures 25x24 cm and has 7x7 peg holes.
The boards in the Birka cham-
ber graves 624 and 886 measured c 3540 and 50 cm a side respectively.
gaming boards are, however, known from earlier in the 1st millennium ad. The
one in the 7th-century Valsgärde 8 boat grave measured 8590 cm a side.
Among the group of 11th-century rune-stones with rich figural scenes from
the Swedish province of Gästrikland, a near-contemporary depiction of people
playing a board game can be seen. The stone from Ockelbo church only survives
in archival documentation since it was destroyed by fire, but depicted two beard-
ed men sitting to either side of a square board, drinking from horns.
The board
has a frame, inside of which are five small squares: one in each corner and one
at its centre, connected by diagonal lines to the corner squares. The Ballinderry
board has a similar arrangement. The Ockelbo board’s side reaches from the
lap to the top of one player’s head, and from the lap to the chin of the other.
It is therefore rather large, assuming that the carver accurately rendered the
board at the same scale as the players.
The implications of this interpretation of the gaming pieces for the sym-
bolic message of the Skamby grave are intriguing if not conclusive. Rather than
mourners placing gaming pieces on a board for playing within the grave, the
deposition of a single gaming set would denote the social class of the deceased
as perceived by the survivors. It may also anticipate future games, perhaps im-
agining a lordly lifestyle of gaming, feasting and fighting in the next world.
Alternatively or additionally, the gaming pieces may relate to the ritualised
associations of gaming with concepts of fate and divining as recently suggested
by Solberg.
In the mortuary context, gaming may have been regarded as
a means of assisting the transformation of the deceased into the afterlife or
ancestral state.
Gaming pieces in Viking-period graves may be yet another piece of
evidence that people regarded mortuary practices at this time as ritualised
transformations, involving the engagement of the living with the dead. They
perceived graves as places of communication with the dead. In this sense, the
placing of one set of gaming pieces with the deceased is a means of social
exchange between the living and the dead focusing on the grave monument.
What could be a more apposite means of communicating with the dead than to
provide the deceased with only one set of gaming pieces? Perhaps the survivors
Nicolaysen 1882, pl VIII:1.
O’Neill Hencken 1933.
Selling 1940.
Arwidsson 1954, 93–4.
Jansson 1981, 194–5, Gs 19.
Hall 2007; Jennbert 2006, 136.
Solberg 2007.
Andrén 2007a, 116–17.
97skamby viking boat grave
were expected to bring their own to the side of the grave, and ancestors in the
afterlife to bring theirs! Certainly, the selection of gaming pieces over other
expensive artefacts indicates the centrality of gaming as a practice and metaphor
of high-status life and death in Viking-period southern Sweden.
Not all Scandinavians practised boat burial at all times in the later 1st mill-
ennium ad. Moreover, they could deploy boats in different ways burnt,
buried, monumentalised in stone settings and depicted upon stones.
In cases
where boats were employed in inhumation practices, they show considerable
variability in terms of the size of the boat, the range and character of grave
furnishings and the nature of the monument raised over the boat. Hence, boat
graves were a variable and strategic mortuary custom that served to commemo-
rate the aristocratic dead and articulate the world-views and social identities of
the survivors.
The evidence presented above shows that boat burial at Skamby involved
elaborate ritual performances and the deployment of material culture linked to
both aristocratic lifestyles and metaphors for afterlife journeying. These included
the provision of a clinker-built boat, horse gear and gaming pieces. A distinctive
stone setting upon a locally prominent ridge, which would have created a
striking presence for the aristocratic dead overlooking the surrounding inhabited
landscape, memorialised these performances (Figs 5 and 7). The excavated
monument at Skamby possibly formed part of a tradition of boat-inhumation
that may have persisted for more than two centuries.
This form of mortuary practice allowed the local aristocracy of the
Vikbolandet peninsula to define their social identities and political allegiances.
Moreover, this was a multi-scalar statement, operating on local, regional and
supra-regional stages. In particular, the amber gaming pieces provide a new
perspective on the significance of gaming in both Viking-period aristocratic
culture and death rituals.
Further research is required at Skamby to ascertain the duration, extent
and overall character of the cemetery. Moreover, the investigation of the
landscape context of the cemetery in relation to other coeval cemeteries and
settlements on the Vikbolandet peninsula remains a priority for future research.
Finally, a fuller exploration is required of the regional context of the cemetery
in relation to other high-status burial sites and central places throughout Östergöt-
Yet the 2005 excavation results serve to illustrate the potential of
research-driven fieldwork targeting late-1st-millennium ad burial sites that are
effectively off-limits to contract archaeology in Sweden.
Eg Gräslund 1981, 69–70; Andrén 1993.
Carver 1995; 2001; 2005, 312–13.
Rundkvist in prep.
98 martin rundkvist and howard williams
Many thanks to the Pettersson family of Skamby for their support and hospitality,
and to the Östergötland County Museum and the Linköping offi ce of the National
Heritage Board for their generous support for the project. The text has benefi ted from
suggestions by Ing-Marie Back Danielsson, Alf Ericsson, Ingrid Gustin, Mark Hall, Alison
Klevnäs, Olwyn Owen, Alex Sanmark, Pierre Petersson, Frans-Arne Stylegar, Elizabeth
Williams and an anonymous referee. A 50-page excavation report is available on-line at
<> [accessed 3 May 2008].
Rundkvist and Williams 2006.
Published sources
Andersson, H, Carelli, P and Ersgård, L
1997, Visions of the Past. Trends and Tradi-
tions in Swedish Medieval Archaeology,
Dept of Archaeology, University of Lund,
Lund Studies in Medieval Archaeology
Andrén, A 1993, ‘Doors to other worlds:
Scandinavian death rituals in Gotlandic
perspectives’, J European Archaeol 1,
Andrén, A 2007a, ‘Behind Heathendom:
archaeological studies of Old Norse reli-
gion’, Scottish Archaeol J 27(2), 10538.
Andrén, A 2007b, ‘A petrifi ed patchwork:
the rune-stone at Karlevi and the early
history of Öland’, in Hårdh, Jennbert
and Olausson 2007, 295300.
Andrén, A, Jennbert, K and Raudvere, C
(eds) 2006, Old Norse Religion in Long-term
Perspectives: Origins, Changes, and Interactions,
Lund: Nordic Academic Press, Vägar till
Midgård 8.
Arbman, H 19403, Birka I: die Gräber:
Untersuchungen und Studien, Stockholm:
Royal Swedish Academy of Letters.
Arrhenius, B (ed) 2001, Kingdoms and Region-
ality: Transactions from the 49th Sachsensym-
posium 1998 in Uppsala, Stockholm:
Archaeological Research Laboratory,
University of Stockholm, Theses and
papers in archaeology B6.
Arwidsson, G 1954, Valsgärde 8, Uppsala:
Museum of Nordic Antiquities, Die
Gräberfunde von Valsgärde 2.
Arwidsson, G (ed.) 1984, Birka II:1: system-
atische Analysen der Gräberfunde, Stockholm:
Royal Swedish Academy of Letters.
Arwidsson, G (ed) 1989a, Birka II:3: system-
atische Analysen der Gräberfunde, Stockholm:
Royal Swedish Academy of Letters.
Arwidsson, G 1989b, ‘Bernstein’, in Arwids-
son 1989a, 535.
Bencard, M, Bender Jørgensen, L and
Brinch Madsen, H (eds) 1991, Ribe Exca-
vations 197076, vol 3, Esjbjerg: Sydjysk
Berglund, L 1971, ‘Hnefatafl en gåta?’,
Saga och sed: Kungl Gustav Adolfs akademiens
årsbok 1970, 8092.
Botfeldt, K and Brinch Madsen, H 1991,
‘The amber’, in Bencard et al 1991,
Carver, M 1986, ‘Sutton Hoo in context’,
Settimane di Studio del central Italian di studi
sull’alto medioevo 32, 77117.
Carver, M 1995, ‘Boat-burial in Britain:
ancient custom or political signal?’,
in Crumlin-Pedersen and Munch Thye
1995, 11124.
Carver, M 2001, ‘Why that? Why there?
Why then? The politics of early medieval
monumentality’, in Hamerow and Mac-
Gregor 2001, 123.
Carver, M 2005, Sutton Hoo: A Seventh-
Century Princely Burial Ground and its Context,
London: British Museum.
Christiansen, E 2002, The Norsemen in the
Viking Age, Oxford: Blackwell.
Claréus, C and Fernholm, R 1999, ‘Aska-
högen: från ättebacke till naturbild-
ning till en av Östergötlands största
gravhögar’, in Andersson et al 1999,
Crumlin-Pedersen, O (ed) 1991, Aspects of
Maritime Scandinavia ad 200-1200, Roskil-
de: Vikingeskibshallen, Proceedings of
the Nordic Seminar on Maritime Aspects
of Archaeology, Roskilde, 13-15 March
Crumlin-Pedersen, O and Munch Thye, B
(eds) 1995, The Ship as Symbol in Prehistoric
99skamby viking boat grave
and Medieval Scandinavia, Copenhagen:
National Museum, Studies in Archaeol-
ogy and History 1.
Fabech, C. 2001, The Spatial Distribution
of Gold Hoards in Southern Scandinavia
and the Geography of Power, KVHAA
Konferenser 51, 189–204.
Gahrn, L 1988, Sveariket i källor och histo-
rieskrivning, Göteborg: Meddelanden från
Historiska institutionen vid Göteborgs
universitet 36.
Gerds, M 2006, ‘Scandinavian burial rites
on the southern Baltic coast’, in Andrén
et al 2006, 15358.
Graham-Campbell, J (ed) 1994, Cultural
Atlas of the Viking World, Oxford: Oxford
University Press.
Gräslund, A-S 1981 Birka IV: The Burial
Customs: A Study of the Graves on Björkö,
Stockholm: Royal Swedish Academy of
Grieg, S 1947, Gjermundbufunnet: en høvdinge-
grav fra 800-årene fra Ringerike, Oslo:
Norske oldfunn 7.
Hall, M 2007, Playtime in Pictland: The
Material Culture of Gaming in Early Medieval
Scotland, Cromarty: Groam House
Hamerow, H and MacGregor, A (eds) 2001,
Image and Power in the Archaeology of Early
Medieval Britain, Oxford: Oxbow.
Hårdh, B, Jennbert, K and Olausson,
D (eds) 2007, On the Road: Studies in
Honour of Lars Larsson, Dept of Archaeology,
University of Lund, Acta Archaeologica
Lundensia, series in quarto 26.
Heaney, S (trans) and Donoghue, D (ed)
2002, Beowulf: A Verse Translation, London:
Holmbäck, Å and Wessén, E 1933, Svenska
landskapslagar: Första serien: Östgötalagen och
Upplandslagen, Stockholm: Geber.
Jansson, S B F 1981, Gästriklands runinskrifter,
Stockholm: Royal Swedish Academy of
Jennbert, K 2006, ‘The heroized dead:
people, animals, and materiality in Scan-
dinavian death rituals, ad 2001000’, in
Andrén et al 2006, 13541.
Jensen, S 1991, Ribes vikinger, Ribe: Den
antikvariske samling.
Kaliff, A 1998, Arkeologi i Östergötland. Scener
ur ett landskaps förhistoria, Uppsala: Dept of
Archeology, University of Uppsala, OPIA
Kars, H and Wevers, J 1983, ‘Early-
medieval Dorestad: an archaeo-
petrological study: part VII: amber’,
Berichten van de Rijksdienst voor het
Oudheidkundig Bodemonderzoek 33, 6181.
Klevnäs, A 2007, ‘Robbing the dead at
Gamla Uppsala, Sweden’, in Stevenson
and White 2007, 2442.
Lamm, J P and Nordström, H-Å (eds) 1983,
Vendel Period Studies, Stockholm: Museum
of National Antiquities.
Larsson, G 2007, Ship and Society: Maritime
Ideology in Late Iron Age Sweden, Uppsala:
Department of Archaeology and Ancient
History, University of Uppsala.
Lidén, K, Isaksson, S and Götherström,
A 2001, ‘Regionality in the boat-grave
cemeteries in the lake Mälaren valley’, in
Arrhenius 2001, 2740.
Lindquist, M 1984, ‘Spielsteine, Würfel und
Spielbretter’, in Arwidsson 1984, 215
Linnaeus, C 1995, The Lapland Journey, ed
and trans Peter Graves, Edinburgh:
Lockharton Press.
Ljungkvist, J 2006, En hair atti rikR: Om
elit, struktur och ekonomi kring Uppsala och
Mälaren under yngre järnålder, Uppsala:
Department of Archaeology and Ancient
History, University of Uppsala.
Magnusson, G (ed) 2001, Möre: historien
om ett småland, Kalmar: Kalmar County
Müller-Wille, M 1969, Bestattung im Boot:
Studien zu einer nordeuropäischen Grabsitte,
Offa 25/26 (1968/69).
Müller-Wille, M 1995, ‘Boat-graves, old
and new views’, in Crumlin-Pedersen
and Munch Thye 1995, 10110.
Müller-Wille, M 2007, ‘Four Viking Age
boat burials in comparison’, in Hårdh et
al 2007, 28794.
Näsman, U 1991, ‘Sea trade during the
Scandinavian Iron Age: its character,
commodities, and routes’, in Crumlin-
Pedersen 1991, 23–40.
Nicklasson, P 1999, ‘Excavating post-
processually a theoretical Utopia or a
practical reality?’, Lund Archaeol Rev 5,
Nicolaysen, N 1882, Langskibet fra Gokstad ved
Sandefjord, Christiania: Cammermeyer.
Nielsen, A-L 1997, ‘Pagan cultic and votive
acts at Borg. Expression of the central
100 martin rundkvist and howard williams
signifi cance of the farmstead in the Late
Iron Age’, in Andersson et al, 37292.
Norberg, L 1998, Ett gravfält från yngre
järnålder i Strängnäs. Särskild undersökning,
Nyköping: Sörmlands museum and Land-
s tinget Sörmland, Arkeologiska medde-
landen 1998:1.
Nordahl, E 2001, Båtgravar i Gamla Uppsala:
Spår av en vikingatida högreståndsmiljö,
Uppsala: Department of Archaeology
and Ancient History, University of
Nylén, E and Schönbäck, B 1994, Tuna i
Badelunda. Guld, kvinnor, båtar. Vols 12,
Västerås kommun.
O’Neill Hencken, H 1933, ‘A gaming board
of the Viking Age’, Acta Archaeologica 4,
Overing, G R and Osborn, M 1994,
Landscape of Desire: Partial Stories of the
Medieval Scandinavian World, Minneapolis:
University of Minnesota Press.
Owen, O and Dalland, M, 1999, Scar: A
Viking Boat Burial on Sanday, Orkney, East
Linton: Tuckwell Press.
Parlett, D 1999, Oxford History of Board Games,
Oxford: Oxford University Press.
Petersen, J 1914, ‘Bretspillet i Norge i
forhistorisk tid’, Oldtiden: tidskrift for norsk
forhistorie 4, 7592.
Petré, B 1993, ‘Male and female fi nds and
symbols in Germanic Iron Age graves’,
Current Swedish Archaeol 1, 14954.
Reallexikon 1978, ‘Brettspiel’, Reallexikon der
germanischen Altertumskunde: 2 völlig neu bearb
und stark erw Aufl , 3, Berlin: de Gruyter.
Rundkvist, M (ed) 1999, Grave Matters:
Eight Studies of Burial Data from the First
Millennium ad from Crimea, Scandinavia and
England, Brit Archaeol Rep Int Ser 781.
Rundkvist, M 2003, Barshalder 2: Studies of
Late Iron Age Gotland, Stockholm: Univer-
sity of Stockholm.
Rundkvist, M 2007a, ‘The fi eld-
archaeological paradox’, in Zivkovic
2007, 24950.
Rundkvist, M 2007b, ‘Early medieval burial
studies in Scandinavia 19942003’, in
Semple and Williams 2007, 4755.
Rundkvist, M in prep, Östergötland’s central
places and political geography
AD 400–1000.
Rundkvist, M, Stilborg, O and Williams H
2007, ‘Copper alloy casting at Skamby in
Kuddby parish, Östergötland’, Fornvännen
102, 279281.
Rundkvist, M and Williams, H 2006, Report
on the Excavation of a Viking Period Boat In-
humation and a Pre-Roman Iron Age Settlement
with Bronze Casting Debris, archive report
available on-line: <http://www.archive.
Skamby2005report.pdf> [last modifi ed
12 February 2007].
Sandwall, A (ed) 1980, Vendeltid, Stockholm:
Museum of National Antiquities.
Sanmark, A 2004, Power and Conversion: A
Comparative Study of Christianization in Scan-
dinavia, Uppsala: University of Uppsala,
Occasional Papers in Archaeology 34.
Schietzel, K 1981, Stand der siedlungsarchäolo-
gischen Forschung in Haithabu: Ergebnisse
und Probleme, Neumünster: Wachholtz,
Berichte über die Ausgrabungen in
Haithabu 16.
Selling, D 1940, ‘Svenska spelbräden från
vikingatid’, Fornvännen 35, 13444.
Semple, S and Williams, H (eds) 2007,
Early Medieval Mortuary Practices. Anglo-
Saxon Studies in Archaeology and History 14,
Oxford: Oxford University School of
Shetelig, H 1912, Vestlandske graver fra jernal-
deren, Bergen: Bergens Museums Skrifter
Solberg, B 2007, ‘Pastimes or serious
business? Norwegian graves with gaming
objects, c. 2001000 ad, in Hårdh et al
2007, 26569.
Stalsberg, A 2001, ‘Scandinavian Viking-
Age boat graves in Old Rus’, Russian
History 14, 359401.
Stevenson, A and White, N C C 2007,
Archaeol Rev Cambridge 22.1: The Materiality
of Burial Practices, Cambridge: Depart-
ment of Archaeology.
Svanberg, F 2003a, Decolonizing the Viking
Age 1, Dept of Archaeology, University
of Lund, Acta Archaeologica Lundensia,
series in octavo 43.
Svanberg, F 2003b, Death Rituals in South-
East Scandinavia ad 8001000: Decolonizing
the Viking Age 2, Dept of Archaeology,
University of Lund, Acta Archaeologica
Lundensia, series in quarto 24.
Weiler, E 1975, ‘Båtgrav i Botkyrka’,
Fornvännen 70, 7781.
Williams, H 1999, ‘Placing the dead: inves-
tigating the location of wealthy barrow
burials in seventh century England’, in
Rundkvist 1999, 5786.
101skamby viking boat grave
Williams, H 2006, Death and Memory in Early
Medieval Britain, Cambridge University
Williams, H 2007, ‘Depicting the dead:
commemoration through cists, cairns
and symbols in early medieval Britain’,
Cambridge Archaeol J 17(2), 14564.
Zivkovic, B (ed) 2007, The Open Laboratory:
the Best Writing on Science Blogs 2006,
Chapel Hill, NC.
Unpublished sources
Eriksson, L 1993, Spelgravar i Mälardalen:
en studie av gravar i Uppland och
Södermanland (unpub MA thesis, Dept
of Archaeology, University of Stock-
Lindquist, M 1978, Sällskapsspel uten-
silier och principer under förhistorisk
tid i Norden (unpub BA thesis, Dept of
Archaeology, University of Stockholm).
Sandberg, C-A 1976, Spel och dobbel i Birka
(unpub BA thesis, Dept of Archaeology,
University of Stockholm).
Sandberg, M 1994, En rik man får spel? En
undersökning av spelutensilier tillförda
Statens Historiska Museum mellan åren
1891 och 1983 (unpub MA thesis, Dept
of Archaeology, University of Uppsala).
grant acknowledgements
The fi eldwork and Martin Rundkvist’s involvement enjoyed fi nancial support
from a number of institutions, notably the Berit Wallenberg foundation, the Helge Ax:
son Johnson foundation, The Royal Academy of Letters, History and Antiquities in
Stockholm, the Anér foundation and the Rausing foundation. A British Academy Small
Research Grant and the Student Fieldwork Fund of the Department of Archaeology at
the University of Exeter supported Howard Williams’ involvement.
Excavation d’une tombe à navire contenant des pièces de jeu en ambre, à
Skamby, Östergötland, Suède par Martin Rundkvist et Howard Williams
Pendant l’été 2005, les auteurs ont dirigé les travaux d’excavation d’un arrangement de
pierres couchées comportant une dépression centrale en forme de bateau à Skamby, dans
la paroisse de Kuddby, en Östergötland (Suède). Ces pierres recouvraient une tombe à
navire de petite taille et en mauvais état, qu’on a pu dater du début de la période viking
siècle) grâce aux artefacts récupérés. C’est la première fois que l’on découvre une tombe
à navire dans la province d’Östergötland. Nous relatons les fouilles, y compris la découverte
d’une collection exceptionnelle de 23 pièces de jeu en ambre, qui offrent une nouvelle per-
spective sur le jeu à l’époque viking. Nous considérons alors les données tirées de cette tombe
à navire par rapport au reste du cimetière de Skamby, qui n’a pas encore été étudié. Si l’on
en juge d’après une étude topographique de la crête entourant la zone mise au jour, et les
découvertes faites au détecteur à métaux dans les champs environnants, le cimetière de
Skamby semble être un lieu de sépultures de haut rang, divisé en deux zones: l’une conten-
ant des tombes à navire, l’autre des cercles de pierres qui recouvrent probablement des
tombes à crémation. Les résultats des fouilles permettent de remettre en question l’idée que
dans le sud de la Suède, à la fi n du premier millénaire, les tombes à navire étaient un rite
funéraire réservé aux élites.
Ein Schiffsgrab aus der Wikingerzeit mit Spielsteinfunden aus Bernstein in
Skamby, Östergötland, Schweden von Martin Rundkvist und Howard Williams
Im Sommer 2005 leiteten die Autoren die Ausgrabung einer fl achen Steinsetzung mit
einer bootsförmigen Vertiefung in der Mitte in Skamby, in der Gemeinde Kuddby in
102 martin rundkvist and howard williams
Östergötland, Schweden. Die Steinsetzung bedeckte ein kleines und schlecht erhaltenes
Schiffsgrab, das anhand der geborgenen Artefakte auf die frühe Wikingerperiode (9.
Jahrhundert n. Chr.) datiert wurde. Dies ist die erste Ausgrabung eines Schiffsgrabs in der
Provinz Östergötland. Der Bericht beschreibt die Ausgrabungen, einschließlich der Entdeck-
ung einer außergewöhnlichen Sammlung von 23 Spielsteinen aus Bernstein, was neues Licht
auf die Spielgewohneiten während der Wikingerperiode wirft. Die aus dem Schiffsgrab
erhaltenen Daten werden dann in Bezug auf den Rest des Friedhofs von Skamby, der noch
erforscht werden muss, mit in Betracht gezogen. Den topographischen Vermessungen der
die Ausgrabungsstätte umgebenden Erhöhung und den Metalldetektorfunden aus den
umliegenden Feldern nach zu urteilen, scheint der Friedhof von Skamby ein in zwei Zonen
unterteiltes Gräberfeld für ranghohe Persönlichkeiten zu sein. Ein Teil umfasst die Schiffs-
gräber, der andere kreisförmige Steinsetzungen, die wahrscheinlich Brandgräber bedecken.
Das Ergebnis der Ausgrabung führt zu einem überarbeiteten Bild des Schiffsgrabs als
Elite-Totenkult im südlichen Schweden während des späten ersten Jahrtausends n. Chr.
Rinvenimento di una barca funeraria vichinga con oggetti da gioco in ambra
in uno scavo a Skamby, nella provincia di Östergötland, Svezia di Martin
Rundkvist e Howard Williams
Nell’estate del 2005 gli autori hanno diretto lo scavo di una pietra posata con una depres-
sione a forma di barca presso Skamby, nella parrocchia di Kuddby nell’Östergötland, in
Svezia. La pietra ricopriva una barca funeraria piccola e mal conservata che, in base ai
manufatti rinvenuti, viene fatta risalire al primo periodo vichingo (IX secolo d.C). Si tratta
del primo scavo di barca funeraria nella provincia di Östergötland. La relazione parla
degli scavi, compresa la scoperta di un’eccezionale collezione di 23 pezzi da gioco in ambra,
che offrono una nuova inquadratura sui giochi del periodo vichingo. Quindi i dati della
barca funeraria sono messi in relazione con il resto del cimitero di Skamby, che rimane da
investigare. A giudicare da una rilevazione topografi ca del rilievo che circonda la zona
degli scavi e da alcuni oggetti rinvenuti con metal detector nei campi circostanti, il cimitero
di Skamby sembra un terreno di sepoltura di alto status, diviso in due zone, una di
inumazioni in barche funerarie, l’altra di pietre circolari che probabilmente ricoprivano
tombe a cremazione. Gli esiti degli scavi creano un quadro riveduto delle barche funerarie
come riti funebri elitari nella Svezia meridionale verso la fi ne del primo millennio d.C.
... Boat burials dating from the 6th to the 10th century AD are found across Scandinavia, ranging from the extravagant ship burials at Oseberg and Gokstad (Brøgger and Shetelig 1971), to cemeteries containing many small boat graves (Arbman 1943, Birkedahl and Johansen 1995, Stylegar 2007, Rundkvist and Williams 2008. Boat burials are also found elsewhere in northern Europe, for example at Sutton Hoo (Carver 2000) and Snape (Filmer-Sankey 1992) in England, Île de Groix in France (Price 1989) and Salme, Estonia (Konsa et al. 2009, Allmäe et al. 2011. ...
... If the grave has not been disturbed, the clinker construction can be seen in the alignment of the nails and rivets, and thus the plan of the boat can be reconstructed (Müller-Wille 1974). Many boat and ship burials from the late Iron Age and Viking Age fall into this category, including Sutton Hoo, Føre, Salme, Borre, Hedeby, Scar and Ardnamurchan (Brøgger and Shetelig 1971, Schanche 1991, Crumlin-Pedersen 1997, Owen and Dalland 1999, Carver 2000, Konsa et al. 2009, Harris et al. 2012, along with all the boat graves at Kaupang (Stylegar 2007), Birka (Arbman 1943) and Skamby (Rundkvist and Williams 2008). ...
While boat and ship graves are known from across northern Europe, and are particularly associated with the Viking Age, only seven examples of such monuments have been excavated in Iceland. Furthermore, no shipwrecks are known dating from this period in Iceland, and examples of boat timbers preserved by waterlogging are very rare. As such, the mineralized wood remains from these burials comprise the vast majority of direct archaeological evidence currently available for the boats used in daily life in early Iceland. This paper uses taxonomic identification of mineralized wood remains from the Icelandic boat graves, along with comparative data from Europe and Scandinavia, to discuss boat construction and repair in early Iceland, including the possibilities of driftwood utilization and the importation of boats from Norway. The economic and social significance of the practice of boat burial in Iceland is also explored, with regard to the importance of boats as transport and the limited availability of wood suitable for boat construction and repair.
... They also feature prominently in highstatus burials that contain martial equipment, including graves from Valsgärde and Birka, Sweden, and the mid to late eighthcentury Salme II ship burial, Estonia (Arbman 1943;Hall 2016;Peets, Allmäe, and Maldre 2010;Peets et al. 2012). While gaming pieces and boards are predominantly found in male graves prior to the Viking Age, the recovery of these objects from later female burials suggests that these may have become gender neutral over time (Rundkvist and Williams 2008). ...
Full-text available
Although the Viking Age (ca. 750–1050 CE) is often characterized as a time of violence, significant questions remain regarding how conflict was conducted during the period. For example, there have been few attempts to understand the cultural norms, attitudes, and practices that drove individuals to participate in warfare. This article reports the results of a study that sought to shed light on this issue by considering the process of enculturation during Viking Age childhood. This was achieved by exploring how the influences of militarism and hegemonic masculinity conditioned those living within Scandinavian societies to participate in conflict from a young age. Through examining the archaeological and literary evidence for childhood pastimes, the study found that everyday aspects of Viking Age society reinforced militaristic, hegemonic hierarchies of masculinity. This can be seen, for example, in the form of toy weapons that were modeled on full-sized, functional weapons; strategic board games that conveyed messages regarding the ideological power of kingship; and physical games that provided opportunities for successful individuals to enhance their social status. The evidence therefore suggests that Viking Age societies perpetuated a series of self-reinforcing cultural norms that encouraged participation in martial activities. ©2019 by The Wenner-Gren Foundation for Anthropological Research. All rights reserved. T.
... There are circumstances when the lithic material of unmodified or minimally modified stone or manuports leads to a direct interpretation like fossilized amber gaming pieces from a Viking boat grave or "unmodified pebble-sized piece of hematite" procured for pigment found at a Caddoan village site (Perttula, 2010, p. 65;Rundkvist & Williams, 2008). Most other situations, as Hurcombe (2007, p. 5) describes in a discussion of manuports, " … requires knowledge of what is naturally present in the landscape contemporary to the site and to know whether any natural transport mechanisms, such as seasonal streams, may have affected the site." ...
The recovery of hundreds of unmodified polished pebbles from excavations at Spirit Eye Cave in West Texas provides an opportunity to analyze an underreported class of lithic artifact. When unmodified pebbles are recovered from large shelter sites in the Texas region, they are sometimes associated with a portable form of pebble art. Because of the small size and recovery of avian faunal elements from the cave, this analysis pursued the possibility of the pebbles representing gastroliths associated with the procurement of avian species. Diagnostic attributes of gastroliths from experimental studies in paleontology provide the criteria for assessment. The results indicate that local formation processes, rather than the procurement of avian species by prehistoric inhabitants, better explain the data set. The study also found more polished chert pebbles than any other lithic material and this is argued to represent human agency; a discussion of the implications for these results provides a range of interpretive possibilities.
... Огляд їх здійснювали Г.Ф. Корзухіна (Корзухина 1963), Х. Віттекер (Whittaker 2006) та М. Рундквіст і Г. Вільямс (Rundkvist, Williams 2008); частково проблема також розглядалася автором цієї статті (Хамайко 2012). ...
Full-text available
The article based on the archival sources re-publicate a lost gaming sets that were found by P.I. Smolichev in 1926 during the excavations at the necropolis in floodplain of Desna river. The archival data on the gaming pieces were studied by the analogies, using chronology of contexts and special features of gaming pieces, that have been found in the 10th century burials, and also of single finds in synchronous cultural layers of the settlements in Kyivan Rus' and Scandinavia areas. The gaming pieces made of glass are concentrated in so called «druzhyna» sites in Kyivan Rus' area among the other import objects dealt with through Scandinavian influence. The glass pieces analogies reveal their belonging to North European cultural area irrespectively of their place of production, which is not finally determined today. One of the 10th century Shestovytsia barrows consisted a cremation located at the firring place with the removing of remains in urn as it was usual for Slavs. Discovery of a glass gaming set in it testifies that the game was adopted by the local population, but with remarcable localization of such finds at the main trade routs.
... These objects are regularly found on excavations and described in reports, but in-depth analyses are few and questions about the raw material for the large group made of unspecified bone and antler have always been secondary to chronology, typology, and social contexts (e.g. Selling, 1940;Lindquist, 1978;Mclees, 1990;Kjer Michaelsen, 1992;Sandberg, 1994;Stauch, 1994;Duczko, 1996;Haahr Kristiansen, 1997;Dahl, 2003;Whittaker, 2006;Kristensen, 2007;Solberg, 2007;Rundkvist & Williams, 2008;Ljungkvist, 2008a;Lund Koksvik, 2010, 2013Skomsvoll, 2012;Caldwell & Hall, 2014;Hall, 2016). The present study is based on a database containing more than 200 entries, of which the largest part (about 150 entries) originates from burials. ...
Full-text available
Discussions of pre-Viking trade and production have for many decades focused on products made of precious metals, glass and, to some degree, iron. This is hardly surprising considering the difficulties in finding and provenancing products made of organic matter. In this article we examine gaming pieces made from bone and antler, which are not unusual in Scandinavian burials in the Vendel and Viking period (c. ad 550–1050). A special emphasis is placed on whalebone pieces that appear to dominate after around ad 550, signalling a large-scale production and exploitation of North Atlantic whale products. In combination with other goods such as bear furs, birds of prey, and an increased iron and tar production, whalebone products are part of an intensified large-scale outland exploitation and indicate strong, pre-urban trading routes across Scandinavia and Europe some 200 years before the Viking period and well before the age of the emporia.
... Огляд їх здійснювали Г.Ф. Корзухіна (Корзухина 1963), Х. Віттекер (Whittaker 2006) та М. Рундквіст і Г. Вільямс (Rundkvist, Williams 2008); частково проблема також розглядалася автором цієї статті (Хамайко 2012). ...
Full-text available
GAMING SETS FROM THE EXCAVATIONS BY P.I. SMOLICHEV AT SHESTOVYTSIA FLOODPLAIN NECROPOLIS The article based on the archival sources re-publicate a lost gaming sets that were found by P. I. Smolichev in 1926 during the excavations at the necropolis in floodplain of Desna river near Shestovytsia. The archival data on the gaming pieces were studied by the analogies, using chronology of contexts and special features of gaming pieces, that have been found in the 10th century burials, and also of single finds in synchronous cultural layers of the settlements in Kyivan Rus' and Scandinavia areas. The gaming pieces made of glass are concentrated in so called «druzhyna» sites in Kyivan Rus' area among the other import objects dealt with through Scandinavian influence. The glass pieces analogies reveal their belonging to North European cultural area irrespectively of their place of production, which is not finally determined today. One of the 10th century Shestovytsia barrows consisted a cremation located at the firring place with the removing of remains in urn as it was usual for Slavs. Discovery of a glass gaming set in it testifies that the game was adopted by the local population, but with remarcable localization of such finds at the main trade routs.
... Вибірку аналогій подібним гральним фігуркам проводили Г.Ф. Корзухіна (Корзухина 1963), Х. Віттекер (Whittaker 2006) та М. Рундквіст і Г. Вільямс (Rundkvist, Williams 2008). Варто акцентувати увагу на повних гральних наборах, що походять з камерних поховань доби вікінгів. ...
Full-text available
GAMING SET FROM FLOODPLAIN CEMETERY MOUND OF SHESTOVYTSYA The paper carried republication of gaming set, which had been finding in 1946 during excavation of mound burial in the floodplain of Desna river near Shestovitsa site by Y.V. Stankiewicz. The set was part of the burial items in 10th century pair chamber, accompanied by his horse, arms and weights. It consisted of 26 glass pieces in two colors, among whom were the pawns and two kings. Collected the available data on the remaining 21 playing figure. The data on the remaining gamingpieces were collected. Searching for similarities and their comparative analysis shows that the glass gaming peices are Mediterranean imports consentrated on the river route "from Varangians to the Greeks." The Scandinavian influence is manifested in the spread of the game at Rus’ lands. Attribution of the game reveals that it belongs to the widely known on the North European lands game «hnefatafl» (king table), which is analogous to the Rus’ version can be "tavleya".
... This may be a matter of recognition: in general disturbance can most readily be observed in burials with better skeletal preservation and larger numbers of artefacts, and boat-graves in particular often provide further corroboration in the form of damage to the timbers of the vessel or disordering of the lines of clinker nails left behind when the planks decay ( Figure 1). It can still be difficult to demonstrate conclusively that a later intrusion has occurred, as opposed to a burial being unexpectedly poorly furnished or having suffered other forms of taphonomic damagee.g. at Skamby (Rundkvist & Williams, 2008), Alsike (Arne, 1934), or grave 76 at Tuna i Badelunda (Nylén & Schönbäck, 1994). However, the frequency of reports of disturbance in boat-graves does suggest that this form of burial was particularly subject to secondary opening. ...
This article examines the wide range of grave disturbance practices seen in Viking-age burials across Scandinavia. It argues that the much-debated reopenings at high-profile sites, notably the Norwegian ‘royal’ mounds, should be seen against a background of widespread and varied evidence for burial reworking in Scandinavia throughout the first-millennium ad and into the Middle Ages. Interventions into Viking-age graves are interpreted as disruptive, intended to derail practices of memory-creation set in motion by funerary displays and monuments. However, the reopening and reworking of burials were also mnemonic citations in their own right, using a recurrent set of practices to make heroic, mythological, and genealogical allusions. The retrieval of portable artefacts was a key element in this repertoire, and in this article I use archaeological and written sources to explore the particular concepts of ownership which enabled certain possessions to work as material citations appropriating attributes of dea...
... The ninth-century Skamby boat burial in Sweden (see Table 1) included 23 wellpreserved amber gaming pieces, possibly laid out on the top of the grave's cover or roof, perhaps as a mourning gift (Rundkvist & Williams, 2008). It has however also been suggested that at Skamby the landscape and ceremonial contexts may refer to a belief that the dead in some way continued to occupy the grave (Williams, 2014;Williams et al., 2010), implying that the deceased used the gaming pieces while they waited for their ship to the afterlife to come in (an anonymous referee reminded me that the provision of entertainment for the deceased is recorded in Ibn Fadlan's description of a Rus' funeral. ...
This contribution explores an aspect of boat burials in the second half of the first millennium ad across Northern Europe, specifically boat burials that included equipment for board games (surviving variously as boards and playing pieces, playing pieces only, or dice and playing pieces). Entangled aspects of identity, gender, cosmogony, performance, and commemoration are considered within a framework of cultural citation and connection between death and play. The crux of this article's citational thrust is the notion of quoting life in the rituals surrounding death. This was done both in the service of the deceased and in the service of those wanting to remember the deceased, the argument distills around the biographical trajectories or the different social and individual uses to which people put ostensibly simple things such as gaming pieces.
Full-text available
This study frames research on board games within a body of anthropological theory and method to examine the long-term social changes that effect play and mechanisms through which play may influence societal change. Drawing from ethnographic literature focusing on the performative nature of games and their effectiveness at providing a method for strengthening social bonds through grounding, I examine changes in the places in which people engaged in play over the course of the Bronze Age on Cyprus (circa 2500–1050 BCE), a period of increasing social complexity. The purpose of this research is to examine how the changes in social boundaries concomitant with emergent complexity were counteracted or strengthened through the use of games as tools of interaction. Bronze Age sites on Cyprus have produced the largest dataset of game boards belonging to any ancient culture. Weight and morphological data were gathered from these artifacts to determine the likelihood of their portability and to identify what type of game was present. The presence of fixed and likely immobile games, as well as the presence of clusters of portable games, was used to identify spaces in which games were played. Counts of other types of artifacts found in the same spaces as games were tabulated, and Correspondence Analysis (CA) was performed in order to determine differences in the types of activities present in the same spaces as play. The results of the CA showed that during the Prehistoric Bronze Age, which has fewer indicators of social complexity, gaming spaces were associated with artifacts related to consumption or specialty, heirloom and imported ceramics, and rarely played in public spaces. During the Protohistoric Bronze Age, when Cyprus was more socially complex, games were more commonly played in public spaces and associated with ii artifacts related to consumption. These changes suggest a changing emphasis through time, where the initiation and strengthening of social bonds through the grounding process afforded by play is more highly valued in small-scale society, whereas the social mobility that is enabled by performance during play is exploited more commonly during periods of complexity.
Finds and symbols specific for male respectively female graves from the Germanic Iron Age have been studied for the Mälar area. A preliminary list with weapons, tools and jewellery is presented.
[FIRST PARAGRAPHS] The hypothesis presented in this paper has already appeared in various fragmentary forms (Carver 1986,1993,1998b), but has not hitherto been drawn together, so it is fitting that I should try to do this in honour of my best teacher and most telling critic. After all, it would not do for Professor Cramp's Festschrift to be too burdened by flattering emulation; it should contain something exasperating as well. So I look forward to her leafing through these pages with growing despair, culminating in the tart response familiar from my carefree student days: 'really Mr Carver, I have no idea what you are talking about', meaning: 'actually, I know perfectly well; you on the other hand ...' In brief, the hypothesis concerns the history of the early medieval period in north-west Europe and our ability to read it from archaeology, or more ' specifically from its major investments such as burial mounds, churches, illuminated manuscripts and sculpture: the word 'monumentality' of my title is intended as shorthand for all these things. Confidence in the idea that monuments had (and have) a meaning beyond some vague celebration of an individual or propitiation of an unseen omnipotence has been growing among prehistorians (e.g. Bradley 1993), and is an accepted feature of the historic period. We know that monuments are more than passive memorials because written commentaries, poetry and inscriptions declare their active purposes for us. Monuments comprise the vocabulary of a political language, fossilized versions of arguments that were continuous and may have related more to what was desired than what had occurred (Carver 1993). At the same time, we need not suppose that the expression is necessarily so subtle, sceptical or to use a fashionable word, ironical, as to lose all hope of making equations between a society and its ideas. That architecture, sculpture, burial mounds and brooches have messages beyond the functional which are dependant on their social, economic and above all their ideological context was never an issue: to understand their real meaning is the goal and the aim of each generation that studies them. It is very likely that the motives I attach to the construction of the monuments to be discussed are equally inadequate characterizations of the profound stresses that motivated and were concealed by their makers. That said, the monuments are what survive, and our story must temporarily keep the candle burning untiI their story can be told.
This paper examines Gotlandic picture stones and their relation to death rituals in Scandinavia from AD 400 to 1100. The stones are shown to mediate, as 'doors', between the living and the dead, and their iconography can be related to a specific form of narrative which may also have been relevant to ship burial on the mainland. The interaction between this tradition of burial and Christianity is discussed.This paper examines Gotlandic picture stones and their relation to death rituals in Scandinavia from AD 400 to 1100. The stones are shown to mediate, as 'doors', between the living and the dead, and their iconography can be related to a specific form of narrative which may also have been relevant to ship burial on the mainland. The interaction between this tradition of burial and Christianity is discussed.Cet article est consacré awe blocs figurés du Gotland et à leurs relations avec les rituels funéraires scandinaves, de 400 à 1100 après J.C. Les blocs sent décrits comme des médiateurs, ou des 'portes' entre le monde des vivants et le domaine de la mort. Leur iconographie peut être mise en rapport avec une forme spécifique de narration, qui peut s'ètre également appliquée aux tombes en forme de bateau du continent. L'interaction entre cette tradition funéraire et le christianisme est discutée.Dieser Artikel untersucht gotländische Bildersteine und deren Bedeutung für Begräbniszeremonien in Skandinavien zwischen 400 AD und 1100 AD. Es wird gezeigt, daß diese Steine als 'Pforten' eine Vermittlerrolle zwischen den Lebenden und den Toten einnahrnen, und daß ihre Ikonographie mit einer spezifischen Überlieferung in Zusammenhang gebracht werden kann, die auch in bezug auf die Bootsbestattungen auf dem Festland von Bedeutung ist. Wechselwirkungen zwischen dieser Bestattungstradition und dem Ouistentum werden diskutiert.