Again this week On the Reading Rest I have the material needed to publish an Iron Age house from the mid-first millennium Ölandic ring fort Eketorps borg – the Eketorp ring fort. House 013 was excavated 1968 (0).

Fig 01The Situation. When it was built, House 013 was the only detached building in the ring fort. As a structure the house is not quite symmetrical because corners and short ends have been adjusted to the surrounding buildings in an effort to squeeze-in the house among the existing buildings. If we model what the situation would have been had the houses been symmetrical the adjustments become apparent.

Since the floor covers 51.6 sqm making the house one of the largest buildings in the ring fort, we can conclude that the stone masons intended to build as large a house as possible given its situation in the settlement. Its asymmetry, therefore, has to do with different needs such as the need to pass through the ring street and to drive a cart into the small square created behind House 013 when it was built. In all probability, those who live in the street and around the once much larger square insisted on their right of way when some of them deemed it necessary to build the house. The minimum measure used in the ring fort when planning passages, is a pole of 6 ell or 9 foot, c. 2.82 cm, i.e. the width of the large gates. The whole radial structure of the ring fort is based on dividing peripheries into sections measured in poles (1).

Fig 02

Owing to the building chronology and the chains of abutting and contemporary houses created during the erection of the Northeast quarter we can infer that House 013 was the last to be built.

Fig 03

Fig 05It was erected late in the settlement period when house S had already been rebuilt and enlarged. To begin with it was the smallest house in the ring fort, dominated by a large oven (House S1) when rebiult it was a dwelling house (House S2). House S2 is in all probability the last but one house to be built in the quarter. Logically speaking House U may be as young, but given the flat Southeast corner of the house, the position of its door and the space in front of House S, the planning of House 013, with its withdrawn entrance seems to have been meant to protect an open space in front of House S2 and House R. Likewise the architects took the passage between the existing House U and the southeast corner of House 013 into account.

Fig 04If we ask ourselves who benefits from House 013, the answer is: those who live in House R and House S2. It is conceivable, therefore, that when House R was rebuilt to become a small dwelling house and S2 rebuilt to become a larger one, the storage once a pert of House R was compensated by House 013. In practice rebuilding the houses R and S, as well as adding House 013 to the settlement, may well have happened as one series of events.

Turning the axis of the House 013 more to the South, making the entrance wall parallel to the short end of House T, would have made it possible to build an even larger house facing House T or U. Since this did not happen the overall situation of House 013 indicates that it was meant to function together with House R and House S2.

Indirectly, the reason why House 013 was built had to do with the growing general need for more roof-covered dwelling space. The situation of the house is thus the result of a compromise between these needs and the communal interest in the Northeast quarter to organize daily life around a comparatively large square. The latter interest guided those who built House 011, the first detached house in the ring fort, and later on the situation of House 011 governed that of the houses 010 and 012.

Fig 06Interior constructions. House 013 is divided into two parts by its constructions, primarily because of the floor in its southern part. This floor consisted of a limestone packing and above it a layer of limestone gravel, which created an even, hard and stable floor (construction P).The northern part of the house was meant to have an earthen floor, but the builders found no reason completely to take away the radial house walls of the first ring fort. In some places these remains were left as pavements in the floor, constructions A and B. There are only four interior postholes in two pairs (E/F and G/H) and they are visible because the posts had to be embedded and fixed in those parts of the house where the posts were standing in loose soil rather than directly on the gravel floor or on flat lime stones in earthen the floor. The mid-aisle is wide c. 50 percent of the inner breadth of the house.

The entrance (N) is paved and constructed in the traditional way with and an outer and inner post pair (I/K and L/M). Being the entrance to an outhouse with a storage, the entrance is relative wide and the door blade heavy. Probably it swung on pins in the back of the door fitted into the supporting construction next to the southern entrance post (M).

In addition to these constructions that have to do with the building itself, there are two other constructions, which have to do with furnishing the house. At some stage a small stone foundation was built on top of the gravel floor in the southern part of the house along the wall just south of the entrance (const. O). This kind of foundation, which, except for the one in House 013, belongs to dwelling houses, cosisted of several lime stone courses and could be framed by vertical slabs. They are difficult to interpret, but meant to support a construction that had to be kept well above the floor. A wooden box or crate of some kind – perhaps for storing yarn or grain — has been suggested, but archaeological or ethnographical records have little to support this suggestion. In a settlement with too little roofed dwelling space it stands to reason that whatever the function of this construction, it could be dispensed with if floor space was needed. The second construction consists of two postholes (C and D) which supported a vertical loom. The loom and its workspace is discussed below. It is typical, nevertheless, that the two functional constructions the ’pallet’ O and the loom C/D are usually found in the dwelling houses. If there is but one loom in a house it tends to be situated along the wall and close to the fire place even in contemporary Iron Age farms on Öland and Gotland. If there were more looms in a household they would follow the long walls further away from the hearth (2). Although weaving is closely linked to the main house on a farm, weaving in the Pre Carolingian Iron Age is also a characteristic of pit houses and thus an activity that may well be moved out of the main house into an outhouse.

Stratigraphy. When House 013 was built the ground was leveled. The floor layer on the gravel floor was thin, 5-10 cm. In the northern part the earthen floor could reach c. 15 cm because more finds could be buried in it. The thin floor layers of House 013 were sealed by the gravel and garden slug layer which covered nearly the entire ruined fort, and later on part of the medieval settlement was constructed on a surface that was defined as the top of the first preserved cause of the walls of House 013. The compressed stratigraphy is caused by the thin turf and lack of debris from the ring wall as well as an easy access to building material in the ruined lime stone walls.

Find distributions. During the excavations in 1968 it was not yet possible to spend time pinpointing fragmented bones in the floor layers. In House 013 this was not a great problem because there were precious few fragments. Charcoal pieces and ashes, nevertheless, were mapped because of their diagnostic qualities when trying to define floor levels.

Fig 07

The division of the house into two parts indicated by the constructions is enhanced by the find distribution. On the hard gravel floor there were few finds, oddly enough mostly beads, relatively few pieces of charcoal and few signs of occupation. Since nobody is expected to live on a hard gravel floor this is consistent with this part of the house being a storage. Now and then things were dropped on the floor, but the southwest corner seems to have been more protected, i.e. regularly cover. Contrary to the southern part, the northern is characterized by finds that indicate that somebody lived and worked in the house albeit perhaps not permanently. In the northwestern corner potsherds and resin from a bentwood vessels indicate a living area where people ate.

Fig 08

The area was protected from charcoal, but not from the occasionally broken pot. This pattern suggests that furniture, perhaps a bed to sit on, may have been standing in the northwest corner of the house. The charcoal and the two small ash concentrations enhanced the picture of daily life, but there is no fireplace or hearth and thus the charcoal, the ashes and the potsherds indicate that light, warmth and food were consumed by those who stayed in the house, but the food wasn’t cooked there.

Fig 09The most dominant feature in this part of the house is the 27 loom weight that seems to have fallen to the floor from the post where they were hung up when the fabric was taken down. The remains of two broken loom weights were found in the floor closer to the door. Some of the 27 weights were only partially burned or baked. Nevertheless, if each of the 27 weight were suspended in one or two inches of warp the breadth of the fabric would have been 1.5 or 3 ells, i.e. c. 71 or 142 cm. These breadths are both in accordance with the distance between the postholes C and D. Since they are not part of the roof construction, they are best explained as a way of anchoring the two pots of the vertical loom in the floor. Above the trestles the posts were fitted to a beam that in its turn linked the two trestles anchored on the postholes E/F and G/H. If this beam was attached to the trestles by halving, its diagonal position, indicated by the ‘loom’ holes, would be reasonable inasmuch as the halving would fit the trestles in an oblique angel, thus locking the beam firmly between the trestles. It meets the eye that the loom wasn’t, as usual, set up against the wall. This anomaly indicates that the house walls were lower in House 013 than in the radial houses, i.e. lower than c. 1.7 metres (3).

Fig 12As an additional indication of textile production we must point to the large whetstone with a furrow for producing bone needles buried in the floor next to the loom weights as well a smaller one found in the floor next to it.

Fig 10

Apart from a find distribution that we can understand only as the result of ‘activities’, the western corner of the northern part of House 013 is characterized by textile handicraft and daily life in a small household that doesn’t cook. Owing the organization of the workspace, and to the dismantled loom, the production would seem to vary perhaps also in intensity, maybe following a yearly circle. We may thus conclude that the ‘pallet foundation’, the textile production and one or more of those who worked with textiles could be permanently or seasonally moved out of the dwelling house to live in House 013. This indicates a population pressure demonstrated also in the very building of the House 013, but obviously also a social pressure, which tented to exclude some members of the local or regional community.

The Artefacts. There are very few artefacts in the house floor, and the way they relate to each other is partly odd: there are no beads in the earthen floor where activities were most common, but well on the gravel floor where activities were less frequent. If we judge by the frequency of potsherds there are many iron fragments or few potsherds if we reverse out point of departure. Loom weights and whetstones, moreover, dominate the artefacts. In House 013 as well as in House T wearing beads when going about one’s business would seem to be in vogue c. 600 CE. The lack of beads, needles and spindle-whorls in the northern part of the house, as well as the presence of an arrowhead, may or may not indicate that the weaver was specialized and a man.

Fig 11To sum up: Apart from being an outhouse, House 013 was an even humbler abode than House T, because small fires and broken pots are signs of an insignificant household. The dwelling-and-workshop area is small and for subsistence, whoever lived in the house, must have done so in cooperation with its owners, to whom it was primarily an outhouse. We may think that living in House 013 is something one does only temporarily, but given the general trend in the rebuilding of the houses in the ring fort, changing them from a variety of primary functions to dwelling houses, makes it unlikely that a temporary dwelling would result in a situation such as the one in House 013, if there had not been an urgent need to find someone a place to live. Temporarily setting up a loom against a wall in a dwelling house takes very little room compared to what it took setting it up in House 013. That too suggests that the problem was not the work space, but the space occupied by the worker or workers.



(0) Näsman, Ulf et al. 1976. Ulf Näsman, Kaj Borg & Erik  Wegraeus (eds). Eketorp – fortification and settlement on Öland/Sweden, The Monument. Almqvist & Wiksell.

(1) Näsman, Ulf. 1976. The settlement of Eketorp-II. In: Ulf Näsman, Kaj Borg & Erik Wegraeus (eds). Eketorp – fortification and settlement on Öland/Sweden, The Monument. Almqvist & Wiksell. Stockholm.

(2) Nordström, Karin & Herschend, Frands. 2003. Det ideologiska inslaget i väven. In: Thomas Lindkvist & Janken Myrdal (eds). Tralar. Ofria i agrarsamhallet frab vikingatid till medeltid.. Skrifter om skogs- och lantbrukshisrotia. Vol 17. Nordiska museet. Stockholm. Pp. 50–76.

(3) Edgren, Bengt and Herschend, Frands. 1979. Nya gamla hus. Riksantikvarieambetets rapport 1979:3. Riksantikvarieambetet. Stockholm.


This week On the Reading Rest I have the material needed to publish an Iron Age house from the mid-first millennium Ölandic ring fort Eketorps borg (0)

Översikt Eketorp Hus T

House T, excavated in the early 1970s, is a dwelling house in the North-east block of the ring fort. It is a primary house radiating from the ring wall, i.e. a free-standing building later accompanied by the neighbouring houses S and U. The description of the house will approach it from its entrance, speaking of its outer and inner parts – the ones closest to the ring wall – looking left and right. The house was ruined and partly pulled down already in the Late Iron Age. But not until c. 1200 CE, when the medieval settlement was laid out, was the ground completely leveled and the skins of the walls of House T knocked down to what was by then the surface of the ground. In this process, the Iron Age floor levels were sealed off by debris from the ring wall, new wall foundations supporting wooden sills, and limestone paving.

House T fron the ring street

Interior Constructions. The entrance is simple and there is neither an entrance paving nor post holes to support its construction. The roof-supporting foundations are few and uncertain: two post supports (C and D) and one posthole (A). Probably the lime stone packing below the floor supported most of the roof and the four posts needed to construct the entrance. It seems significant that the only posthole is situated in an earth-filled depression in the limestone packing. The flat hearth for the open fire (E) is located a little behind the centre of the house and next to it farther into the interior there is a cooking pit filled with brittle-burned and fire-cracked stones (G) where the glow could be kept. The outer part of the house is characterized by a hearth refuse (F) consisting of brittle-burned stones, charcoal and some ashes. Adjacent to F and around E and G there are ash concentrations (H, J and K). In the interior, in the right corner, two pieces of a charred plank had been preserved (L).

The limestone packing below the earthen floor is characteristic of the houses in the eastern part of the ring fort, i.e. the part that borders on the intermittent lake east of the ring fort. The ring wall encloses a small section of this lake around a small spring, which was built-in to become the well of the settlement. In order to create dry and drained conditions this area was filled up with limestone – a packing on which the house, its cavity walls with their two limestone skins, stood. Later on more packing within the walls was added as a foundation for the floor constructed just above the packing.

In House T, as in most Iron Age houses there are intended constructions such as the limestone packing, the hearths, the posthole and post supports. The fill that levels out the depression in the packing is also a primary construction, but in addition to proper constructions there are features that were made semi-constructions by daily life. The ash concentrations are a case in point and so is the inner right corner where so little happened that charred planks, eventually concealed in the floor, were not trodden on. Daily life, moreover, results in a number of things that were lost on the floor and preserved as distributions.


Stratigraphy. The house has only one floor – a seemingly continuously growing floor level. The floor lies directly upon the packing or, in the partly filled-in depression in the interior right side of the house, somewhat above it. The floor varies in thickness from c. 5 to c. 15 cm. Stratigraphically the floor is sealed between the packing and the so-called gravel and garden slug layer, which represents the end-phase of the LIA settlements, i.e. the period  when the roof was dismantled and the walls partly pulled down (1).

Several factors have contributed to the characteristics of the floor layer. First there is a tendency, albeit not a very strong one, for the floor to follow the lever of the packing. Secondly, there is a tendency for the floor more markedly to grow in the inner part of the house than in the outer part owing to the way the house was used. Thirdly, the remains of the first ring wall in the outermost part of the house served as a kind of threshold for the inner part of the house – a threshold on which only a very thin occupation layer was deposited. The result of these factors working together is a leveled floor, with a tendency for the central parts being slightly higher than the floor level closest to the long walls.

Floor layer


If we interpolate a corridor of bones and charcoal pieces on a schematic profile in the inner part of the house, covering the depression in the packing, it becomes apparent that the levels where bones have been found are deeper than the levels containing pieces of charcoal. Since it is difficult to preserve bits of charcoal on a floor when people live and walk on it, it is conceivable that the charcoal (pieces with a length above 1 cm) represents quite recent floor levels with few pieces in the deeper and older layers. Bones, which survive much better than charcoal, may be found in deeper layers too, suggesting that the floor layer as we see it, is the result two kinds of behavior: Daily life producing a growing earthen floor tidied-up on a daily basis, and the habitual digging out the floor, i.e. the throwing out the upper floor layers before spreading a new earthen floor. In the process charcoal tends to be come fragmented or dust while some robust objects such as pieces of bone, the odd bead or potsherds are left in or pressed into the deeper otherwise untouched the new layers of soil. In tandem with a wish now and then to clean out the floors there was also a continuous retrieval of things dropped on the floor, such as bones.

Bead stratigraphyOne third of the artefacts found in the house were beads, and if we look at their accumulation in the floor in relation to other finds it is obvious that beads do mainly belong to the upper part of the floor. This indicates that in daily life beads are often found and retrieved before they disappear buried in the floor. Since the floor represents the end phase of the settlement, the distribution indicates that the beads were dropped on the floor shortly before the house was given up. Correspondingly, the representation of potsherds is fairly constant suggesting a propensity in Iron Age man not to retrieve a potsherd when seeing it trodden into the earthen floor.

Find distributions. The stratigraphic analysis suggests that most of the objects found in the floor reflect a relatively short period of usage, which happened to become the end-phase of the occupation. Occupation habits including furnishing and cleaning habits are also reflected in the find distributions.

FInd dist House T

If e.g. we look at the distribution of the complementary categories charcoal bits and bones in relation to hearth, cooking pit, hearth refuse and ash layers, it become apparent that there is a deposition of ashes, charcoal and hearth refuse in the outer part of the house. If checked against the artefact distribution which mirrors daily life around hearth and cooking pit in the inner part of the house, it is obvious that the amount of ashes etc. in the outer part isn’t matched by a sufficient number of artefacts. The amount of preserved pieces of charcoal on the other hand is greater than in the interior. This means that the outer part of the house serves as a (temporary?) dump for waste related to the fire place rather than dwelling. The distribution, moreover, indicates that the lifetime of a hearth before it is maintained is less than that of a floor. The hearth and its immediate surroundings are cleaned up at least once before the general cleansing of the inner part of the house.

The find distribution and the finds reveal the routines of a short period of everyday life as well as boundaries and paths or floor space that look empty either because it were covered and/or little used. Cooking, eating and work related to textiles as well as other kinds of activities on the floor took place in the inner part of the house especially in its right side next to the hearth area. To the left, i.e. north of the hearth area, less has fallen to the floor and its top level is thus a little lower. It stands to reason that in this part of the house some kind of furniture such as beds have to some degree prevented the accumulation of charcoal and artefacts.

The movement pattern in the house is indicated by the empty ‘corridor’ – a path where little is dropped and bits of charcoal not preserved – leading from the entrance into the interior of the house. There is an equally empty space dividing the house into an inner and an outer part and it is tempting to suggest that this space represents a partition wall and a storage area next to it. The absence of finds in the floor could indicate the presence of furniture such as chests.Conclusion House T

The find distribution fits the idea of a house divided into two: an outer entrance room and a passage leading into the interior, a family room or kitchen dwelling informally divided into different areas around the hearths in its centre. Preservation forces us to interpret the find distribution as positive as well as negative indications forming a tentative picture of the way people lived in the house.

The artefacts. Given the relatively large number of beads, the rest of the artefacts indicate a variety of different dwelling-room activities not by frequency but by presence. There is thus one knife, one piece of charred cake (2), one spindle whorl, one loom weight, one whetstone, one comb, one piece of slag and one piece of forging waste. Potsherds are more numerous, but without fittings. The artefacts on this scene in the kitchen dwelling indicate a compact and multi-purposed living. Cooking activities south of the hearths and perhaps spinning and weaving along the south wall are the most visible activities. The most obvious absence is the lack of complete artefacts such as those one would have stored in chests, on a shelf or on top of a wall: no jewelry, no weapons, no gaming pieces, no belt buckles, no cups, no glasses etc. This no doubt has something to do with the way the people moved out of the house, but probably also with the relative humble life lived on a limited space by those occupying House T. We may even suggest that they sat in their beds in the interior, along the northern wall, when they ate,

The size of the dwelling area is c. 25-30 square metres. That equals the size of the kitchen dwelling on a nuclear family farm in South Scandinavia during the Pre Roman Iron Age, i.e. a farm 500 years older than House T. When the floor in House T was lived on there was nevertheless no byre or cowshed belonging to it in the houses surrounding House T and the household wasn’t part of a farm. Compared to a contemporary farmhouse on Öland, the house is just the dwelling part, i.e. half a house. There is, so to speak, no door in the ring wall leading into the byre. Instead of a household on a farm, we should imagine a family living in the ring fort trying to fend for itself perhaps engaged in herding sheep (given the general economy of the settlement), but mostly in keeping warm, cooking food, spinning and weaving, mending a tool and using the outer room as a dump. The people might have had a bench to sit on just outside their door almost in the ring street, but in the summer climbing the ring wall and sitting in the afternoon sun might have been a pleasant alternative. Eventually, the people left with their belongings.

What the excavation has revealed is only a little of what resulted from a short period just before the house was left and pulled-down. In reality the house was used for hundreds of years, but in the end it was but a humble abode in which the last noticeable event before leaving the house might well have been when the string in a necklace snapped and  beads were spread all over the floor.


(0). ‘Eketorps borg’ can be googled. Following the link  and searching for ‘Eketorp’ will lead to some more information.

(1).General descriptions of the three ring fort phases is found in: Borg, K, Näsman, U & Wegraeus, E (red). 1976. Eketorp. The Monument. KVHAA, ALmqvist &Wiksell International, Lund.

(2). Hansson, A.-M., 1997. On plant food in the Scandinavian peninsula in Early Medieval Times. The Archaeological Research Laboratory, Stockholm University. Thesis and Papers in Archaeology B :5 (diss.).