The passage grave at Mysinge , Öland , SE Sweden in a long-term perspective

During the latest part of the early Neolithic period around 3500–3300 BC, megalithic tombs in the form of dolmens and passage graves are built in many different parts of Scandinavia. In Denmark thousands of dolmens and hundreds of passage graves are known, and in northern Germany huge number of tombs are documented as well (Midgley, 1992:418ff). On a Swedish level, dense concentrations of tombs are found along the coasts in Scania and western Sweden, as well as in the area Falbygden in central Västergötland (Srömberg, 1971; Persson & Sjögren, 2001; Sjögren, 2004) (see Fig. 1). Outside these core areas, dolmens and passage graves occur sporadically (e.g. Janzon, 2009; Wallin, 2010) and in large parts of the distribution area of the TRB culture, megalithic tombs have not been built. On the island of Öland, one dolmen and three passage graves are located in the parish of Resmo (Arne, 1909; Papmehl-Dufay, 2006; see below). Only one of these has been excavated, namely the southernmost passage grave at Mysinge, which is the focus of the present paper. Neolithic passage graves are often regarded as the earliest form of collective burial in Sweden (e.g. Ahlström, 2009). The numerous excavated graves typically contain the disarticulated skeletal remains of tens or even more than a hundred individuals (Strömberg, 1971; Persson & Sjögren, 2001). In the Falbygden area, where some 250 passage tombs are concentrated to a restricted area of cambro-silurian sediments, the absolute majority of 14C-dated human bones from passage grave chambers are of middle Neolithic age, and the construction and use of the tombs thus typically connected to the TRB culture (see Midgley, 1992). In the mounds surrounding the tombs, however, secondary burials from the Bronze and Iron Age are extremely common (see Sjögren, 2004:163f). In other cases the The passage grave at Mysinge, Öland, SE Sweden in a long-term perspective


InTRoducTIon
During the latest part of the early Neolithic period around 3500-3300 BC, megalithic tombs in the form of dolmens and passage graves are built in many different parts of Scandinavia.In Denmark thousands of dolmens and hundreds of passage graves are known, and in northern Germany huge number of tombs are documented as well (Midgley, 1992:418ff).On a Swedish level, dense concentrations of tombs are found along the coasts in Scania and western Sweden, as well as in the area Falbygden in central Västergötland (Srömberg, 1971;Persson & Sjögren, 2001;Sjögren, 2004) (see Fig. 1).Outside these core areas, dolmens and passage graves occur sporadically (e.g.Janzon, 2009;Wallin, 2010) and in large parts of the distribution area of the TRB culture, megalithic tombs have not been built.On the island of Öland, one dolmen and three passage graves are located in the parish of Resmo (Arne, 1909;Papmehl-Dufay, 2006; see below).Only one of these has been excavated, namely the southernmost passage grave at Mysinge, which is the focus of the present paper.
Neolithic passage graves are often regarded as the earliest form of collective burial in Sweden (e.g.Ahlström, 2009).The numerous excavated graves typically contain the disarticulated skeletal remains of tens or even more than a hundred individuals (Strömberg, 1971;Persson & Sjögren, 2001).In the Falbygden area, where some 250 passage tombs are concentrated to a restricted area of cambro-silurian sediments, the absolute majority of 14 C-dated human bones from passage grave chambers are of middle Neolithic age, and the construction and use of the tombs thus typically connected to the TRB culture (see Midgley, 1992).In the mounds surrounding the tombs, however, secondary burials from the Bronze and Iron Age are extremely common (see Sjögren, 2004:163f).In other cases the The passage grave at Mysinge, Öland, SE Sweden in a long-term perspective Ludvig Papmehl-dufay seemingly large buried collective in megalithic chambers is an illusion, however, and instead reflects a very extensive period of use, sometimes several millennia.Although activities at the tombs after the TRB early and middle Neolithic by no means are controversial, and often evidenced in the artefact assemblages recovered from the chambers, the use of passage graves in the later Neolithic and the Bronze Age has seldom been discussed.

ÖLAnd In ThE nEoLIThIc
The island of Öland is situated off the Swedish east coast in the Baltic Sea (see Fig. 1).The elongated island measures 130 km in length and c. 20 km in width.Topography is generally flat, and the bedrock consists of Ordovician limestone and, below the western slopes, Cambrian slate.A wealth of prehistoric remains all over the island contributes to create a notion in the landscape of a very present past.The time depth of visible structures in the landscape is considerable, and thus the "presence of the past" goes for past times as well.
The Neolithic coastline is located some 8-12 m above that of today.Stray finds from the period are numerous all over the island, indicating a relatively dense settlement in the central and especially western parts of the island during the early and middle Neolithic (Åberg, 1923).The presence of megalithic tombs and the numerous finds of artefacts associated with the TRB culture, such as thin-butted flint axes, have led early researchers to speculate about an immigrating community of Danish/Scanian TRB farmers to the island in the early Neolithic (Åberg, 1913, 1923;Stenberger, 1948;Hagberg, 1979).
The isolated group of four megalithic tombs in the parish of Resmo is a natural focus for studies of the TRB culture in the area (Arne, 1909).The three passage graves are located a few hundred m apart in the village of Mysinge, and the dolmen is situated c.

ThE MYsIngE PAssAgE gRAVE
The passage grave RAÄ 85 in the parish of Resmo is located in the village Mysinge on southwest Öland (Arne, 1909).Topographically it is situated on the western escarpment at c. 45 m above the present sea level, some 200 m to the east of the steep slopes towards the sea in the west.It has a limited visibility of the sea, and the entrance to the tomb faces away from the Kalmar Straight.In the direction of the tomb entrance towards the southeast the landscape is extremely level, characterized by a thin soil cover on a vast limestone plateau, "The Great Alvar" (Königsson, 1968).
The monument (Fig. 2) consists of a slightly oval mound c. 14.5 × 16 m in size and 1.5 m high, in which a megalithic passage grave is set.The oval chamber, c. 4 m in length and 3 m wide, is built from erratic boulders and oriented in NE-SW, with a c. 6 m long narrow passage running perpendicularly from the chamber towards the SE.Excavations in front of the entrance have revealed a few limestone slabs lying in the length direction of the passage, as well as a small stone-pack-ing in front of the entrance (Arne, 1937;Alexandersson, 2005).The mound is built of pebbles and small stones with larger stones towards the centre, providing an almost cairn-like construction.Only a few stones are visible today in the turf-covered surface.
The passage grave at Mysinge was first excavated in 1908 by archaeologist Ture Jässon Arne (1879Arne ( -1965)), and the results were published the following year (Arne, 1909).The excavation concerned the chamber and the area immediately around the chamber and the passage.The stratigraphy of the chamber consisted of more than 1 m of gravel, sand and soil containing large amounts of unburned human and animal bones as well as a number of artefacts.The bones were mostly found disarticulated, but three human individuals could be identified of which the bones were still in articulated position.In a gap between two of the large roof boulders, fragments of a ceramic vessel were found in the uppermost part of the stratigraphy together with cremated bones, all covered by a limestone slab.Cremated bones were also found just outside the chamber wall to the west.These finds were interpreted as secondary burials from the late Bronze Age or the Iron Age.
The artefacts recovered include a varied set of objects made of flint, bone, antler, amber and pottery.The finds represent a wide chronological sequence, spanning from the TRB culture in the late EN/early MN and up to the Bronze Age.The ceramic urn with cremated bones could possibly be of Iron Age date (Bergenstråhle, 1986:9).The TRB culture is represented in the artefact assemblage mainly through the amber beads, of which at least 17 were collected.Similar finds have been reported from numerous passage graves in various parts of Scandinavia (Ebbesen, 1995;Axelsson & Strinnholm, 2003).A thin-bladed flint axe and a rectangular bone-plate, suggested to be a wrist-guard of Beaker-type, have been interpreted as representing burials of the late MN Battle Axe culture (Malmer, 1962:467).A similar object has been found at Hunneberget, Scania and interpreted as a device connected to textile production (Balic & Knarrström, 2007:90ff).A long flint blade possibly also connects to this sequence.Two bifacial flint artefacts, an arrowhead and a fragment of a flint dagger or sickle, clearly belong to the late Neolithic as do several of the bone implements found.These include three bone pins of a type similar to those found in other late Neolithic burials on the island (Papmehl-Dufay, 2010 b).Finally, two cylindrical bone buttons most probably date to the Bronze Age.In conclusion, the artefact assemblage from the Mysinge passage grave bear witness of a long sequence of human activity at the site, involving the deposition of artefacts and, as we shall see, human and faunal remains in the chamber of the tomb.
The faunal assemblage from Mysinge has not been properly published, but recent analyses indicate the presence of a range of species including dog, fox, hare, cow, sheep/goat, pig and duck (Ahlström, 2009:166;Eriksson et al., 2008, Table 4).Four animal bones from the chamber has been 14 C-dated: a sheep/goat was dated to the late Neolithic, a dog was dated to the transition from Bronze to Iron Age, a cow was dated to the early Iron Age and a sheep/goat was dated to the 16 th -18 th century (Eriksson et al., 2008, table 4).Furthermore, two dogs that display marine dietary signatures have been suggested to be "Pitted Ware dogs", i.e. of Middle Neolithic date, however this has not been confirmed and the dogs remain undated (see Eriksson et al., 2008:540f).
In 1937, Arne returned to Mysinge for a small complementary excavation of the area outside the entrance of the passage grave.Only 4 m 2 were investigated this time and the finds were sparse.Pottery found in front of the entrance was interpreted as being of earlier date.The results from the second excavation at Mysinge were never published.In 2004, an additional area outside the entrance as well as parts of the mound was excavated (Alexandersson, 2005).The trench through the mound revealed a number of construction details previously not known, including a central cairn of large stones.In the uppermost layers towards the centre, just outside the SW end of the chamber, c. 130 fragments of burned bone were found.This find corresponds to burned bones found by Arne in the same area but at a lower level, and probably represents one or several cremations placed in the mound next to the chamber in the late Bronze Age or Iron Age (Arne, 1909:90).
Two trenches were excavated in front of the tomb entrance in 2004 (Fig. 5).In both trenches stone packing appeared, interpreted as an entrance cairn deposited in front of the entrance.In between and mainly underneath the stones, scatters and concentrations of flint, burned bones and pottery were found.The pottery amounts to c. 1.5 kg, while the amounts of flint and burned bones are much more modest.Flakes of flint, quartzite and other materials were found in both trenches and seemingly spread throughout the entire entrance area, whereas pottery and burned bones were restricted to the trench placed right in front of the entrance.The highly fragmented pottery was found in at least three concentrations, possibly corresponding to individual vessels smashed and deposited in front of the tomb.Only a small number of potsherds are decorated, and typologically they can be attributed to the TRB culture and especially its later phase (see Alexandersson, 2005:9).Judging from this, it is clear that the pottery found by Arne in 1937 most probably should be connected to this phase as well.
Apart from the artefacts in front of the entrance, two large sandstone slabs were found and interpreted as facade stones.Both were covered by the stone packing, as were the depositions of pottery, flint and bone.This shows that the entrance cairn was built at a relatively late stage in the history of use of the tomb, possibly in the late Neolithic or the Bronze Age corresponding to observations in Scania and Falbygden (Persson & Sjögren, 2001:214).
In recent years, a research team headed by Professor Kerstin Lidén at Stockholm University has carried out laboratory analyses on the Mysinge osteological assemblage including analyses of stable C and N isotopes as well as mitochondrial DNA (Lidén, 1995(Lidén, , 1996;;Kanstrup, 2004;Eriksson et al., 2008;Linderholm, 2008).In connection with this a series of 34 14 Cdates have been presented (Eriksson et al., 2008; see  Papmehl-Dufay, 2006;Ahlström, 2009).So far, a total of 35 human individuals are included in the analyses, but not all individuals have been checked for all parameters (see Table 1).At the time of excavation in 1908, the number of buried individuals was estimated at 30-40 (Arne, 1909).In recent years, Torbjörn Ahlström has re-evaluated the osteological assemblage from the tomb and presented a minimum number of individuals (MIND) at 56, cremated bones not included (Ahlström, 2009:83).Convincing evidence shows that the bodies have been placed in the chamber in a squatting position, and thus the seemingly chaotic arrangement of bones observed during excavation most probably is a result of later processes (Ahlström, 2009:69).Of the 56 individuals identified, 34 have so far been 14 C-dated of which 12 can be chronologically attributed to the TRB culture, 11 to the late Middle Neolithic Battle Axe culture, one to the Late Neolithic and finally 10 to the early and middle Bronze Age (Table 2).To this should be added the unknown number of cremated individuals, the dates of which are not yet established.The results from the analyses of stable isotopes reveal some interesting patterns (Eriksson et al., 2008).The buried collective at Mysinge display a rather varied dietary signature of a mixture of marine and terrestrial protein sources, but divided up chronologically some general trends emerge (Fig. 6) (see Eriksson et al., 2008:531ff).The TRB individuals (n = 9 in this case) of the early Middle Neolithic all display an isotope signature indicating a mixture of terrestrial and marine protein sources, where hypothetically the terrestrial source derive from cattle or sheep/goat and the marine input consists of fish.The individuals buried in the tomb during the late Middle Neolithic and early Late Neolithic (n = 12 in this case) display a greater variation including completely terrestrial signals as well as samples with a significant marine input.As was the case with the TRB individuals, trophy levels (i.e.δ 15 N values) are indicative of the terrestrial source involved being some kind of herbivore, possibly cattle.For the Bronze Age individuals (n = 10 in this case), there is a much lower degree of variability and a clear dependence on completely terrestrial protein sources from a low trophy level.This signature corresponds well to what would be expected from a traditional agricultural community growing crops and keeping cattle and/or sheep/goat (see Eriksson et al., 2008:537ff).Another highly interesting observation from the analyses of stable isotopes is the notion of intra-individual variability (see Eriksson et al., 2008:533, 536).The dietary life history of an individual can be an indicator of her cultural as well as geographical whereabouts at certain stages during her lifetime (Linderholm, 2008:31).For one of the individuals dated to the early Middle Neolithic, a clear dietary change can be seen during his/her lifetime towards a progressively less marine input and a lower trophy level, indicating an increased dependence on (domesticated?) terrestrial herbivores.Amongst the individuals dated to the late Middle Neolithic, at least four display rather extreme intra-individual changes in diet quite possibly reflecting some form of migration.As for the Bronze Age individuals, the variation in diet is much smaller both between and within individuals, suggesting a rather solid dependence on agricultural sources and a geographically more stable demography.

A BIogRAPhY oF PLAcE
A traditional archaeological interpretation of the Mysinge passage grave would pinpoint that the grave was originally built and used within a TRB context, and that subsequent generations in the late Middle Neolithic and the Bronze Age have used it for secondary burials.However, I think that by regarding the longterm use of the site as "secondary", we are missing the most important notion: every generation has had their specific and unique relation to this site, and the place has a biography of its own in which activities in and around the tomb are deeply embedded.When did the passage tomb become "old"?What was considered a proper attitude towards and use of the monument, and how did this change over time (Kopytoff, 1986:66f)?There is reason to believe that people burying their dead in the tomb 2500 BC did not think of it as a secondary practice, rather it was heavily loaded with symbolic significance and something probably of great importance to the people involved.The use in the later Neolithic and the Bronze Age of this already ancient monument as a burial chamber was no doubt connected to aspects of historical significance and memory (Jones, 2007:21f), but also to reflections Fig. 6. Results from the analyses of 13 C and 15 N isotopes on human samples from Mysinge, plotted according to archaeological phase.Data from Eriksson et al., 2008. in the present and prospects for the future (see Holtorf, 1996:121f).If it was simply a matter of convenience ("oh look, someone already built a tomb here, let's use it!"),we should not expect such clear hiatuses in its utilisation.Also, at least from the available evidence, it seems that the practices involved in interring have remained surprisingly similar throughout a very long period of time.Thus, to date there is nothing to suggest that bodies placed in the chamber in the late Middle Neolithic or the Bronze Age were treated radically different from the TRB ones during the early Middle Neolithic.It should be recalled that the general burial practices in the late Middle Neolithic and the early Bronze Age are quite far from what is seen in the Mysinge passage grave (e.g.Malmer, 1962;Rudebeck & Ödman, 2000).Is it possible that people of the early Bronze Age had some form of knowledge of, and imitated, TRB megalithic burial practices reaching some two millennia back in time?
The reasons for using the tomb in post-TRB periods will never be fully established, but we can assume that they were good and that the practice of returning to the oldest monument in the region for further burials was of great symbolic and social significance.So far, we can only guess what characteristics qualified for burial in the passage grave during the Bronze Age.Was it people of a certain ancestry, a specific talent or profession?Or was it during a specific and socially turbulent phase that ancestry and origin in general became important to emphasise?"In times of trouble <…> people and thus social groups, tend to return to a conservative retrospection and preservation of the old, to seek comfort and legitimacy from an idealized history" (Knutsson & Knutsson, 2004:20) The place biography of the Mysinge tomb is probably intimately connected to several other monuments and archaeological structures in the surroundings.During the Neolithic, the Mysinge area was the locus of at least three passage graves built only a few hundred meters apart from each other and probably within a relatively narrow timeframe (Papmehl-Dufay, 2006).It should be recalled that the distance to the nearest region with a concentration of passage graves is about 150 km to the SW as the bird flies, and the Öland tombs are situated in the eastern periphery of the distribution area for the megalithic TRB culture (see Sjögren, 2004).During the Bronze and Iron Age, the areas immediately to the north and south of the Mysinge tombs were used as necropolis and a large number of grave structures of various size and shape can be seen here still today.That is, the concentration of passage graves is situated between two large Iron Age grave fields, and even though it is not that apparent today it is possible that people of the Iron Age have avoided the old graves in the placing of their cemeteries (see Gillings & Pollard, 1999:186).
Built as a monumental megalithic tomb in the late Early Neolithic (c.3500 BC) and used as a burial chamber repeatedly for more than two millennia, the Mysinge passage grave has gathered a complex and varied life history that continues up to this day.Fragments of this can be reached through the archaeological data and written accounts of antiquarian activities, but one should also be aware of that much of what was once remembered about this site is forever lost.A material biography such as the one suggested here is by necessity partial, not only for taphonomic reasons but also because objects have multiple biographies overlapping and completing each other.A car will have an economic biography in which fluctuations in its monetary value are seen, a technical biography in which its fabrication and history of repair are defined, as well as one or several social biographies related to members or groups in the society/ies in which it functions (Kopytoff, 1986:68).One aspect of the biography of the Mysinge tomb relates to its use as a site for burial, but there is also biographic narratives connected to notions of the site as a material link to a mythical and distant past and, in the case of archaeology, the site as an invaluable source to archaeological knowledge.By appreciating the multiple narratives of place and adopting a biographic approach to sites and monuments, we can reach some of the aspects of past lives that otherwise remain obscure (Kopytoff, 1986:67) Atlikus mitybos analizę, nustatytas jūros ir sausumos baltymų šaltinių mišinys.Chronologiškai padalijus analizės duomenis, išryškėjo tam tikros bendros tendencijos (6 pav.) (Eriksson et al., 2008, p. 531ff).Atlikus piltuvėlinių taurių kultūros individų tyrimus nustatytas jūros ir sausumos baltymų šaltinių mišinys, šie individai tikriausiai mito stambiaisiais raguočiais ar avimis / ožkomis, taip pat žuvimis.Tyrimų rezultatai rodė, kad vidurinio neolito pabaigoje / vėlyvojo neolito pradžioje individų tarpusavio skirtumai buvo didesni.Žalvario amžiuje individų mityba buvo panašesnė ir jų izotopų tyrimai rodo aiškią priklausomybę nuo sausumos baltymų šaltinių, todėl galima agrokultūrinė ekonomika (Eriksson ir et al., 2008, p. 537ff).
2.5 km to the north at Resmo church.For a long time, the megalithic tombs and the stray finds have been the main source of knowledge concerning the TRB culture of Öland.It is only during the last couple of years that several settlement sites have been found and excavated (Papmehl-Dufay, 2009; Alexandersson & Papmehl-Dufay, 2009), allowing the megalithic tombs to be placed within a larger Neolithic social context.It now seems that the TRB culture on the island shared several traits with southwest Scandinavia, such as houses of so-called Mossby type, but also that many traits point towards east Central Sweden in the north (see Alexandersson & Papmehl-Dufay, 2009; Papmehl-Dufay, 2010 a).It is clear that the island has been far from isolated during this time, on the contrary.Possibly the communicative geographic setting

Fig. 2 .
Fig. 2. The passage grave at Mysinge as it appears today.Photo from the SE by the author.