The megalithic construction process and the building of passage graves in Denmark
Abstract
On the basis of observations made since the 1980s during investigations of scheduled megalithic monuments in Denmark in connection with restoration work, an attempt is made to illustrate the process of building a passage grave. It is demonstrated that decisions relating to aspects such as the drainage system, the intermediary layer and the construction where the passage meets the chamber must be taken at a relatively early stage. It was therefore necessary, already at the planning phase, to take account of the final design, form and size of the finished monument. It is also concluded that the most difficult part of the construction process was the placement of the capstones. This step has left traces in the surrounding mound in the form of a horizontal platform at the level of the top of the orthostats or alongside the intermediary layers. It is concluded that the remains of the building site and traces of the process involved in the construction of individual passage graves suggest a planned, well-considered and continuous operation. However, once constructed, the monuments were subsequently maintained, and altered as desired, both by the original builders and by their descendants. Further to these were the alterations and additions of later prehistoric periods.
Keywords: Megalithic construction, architecture, building site, passage grave
Introduction
A large number of megalithic graves were excavated in Denmark in the decades around 1900. Some were also restored to assist preservation and to enable public access. However, public access brought with it an obligation to maintain them in good condition. In the 1980s, a systematic appraisal of the c. 500 scheduled passage graves revealed that a dangerous process of decay was in progress, resulting from variable quality of the restoration work. In an attempt to reverse this decline, an intensive effort was initiated to restore and investigate the most threatened, and the most visited monuments.
This paper builds upon observations made during restoration work on about 70 megalithic monuments to illustrate the processes involved in building a Danish passage grave. These interventions were of varying extent and were located in different places in the monuments, as required. Some of the details observed are seen at one or only at few monuments, whilst others are more general features. The construction could be ambitious as tombs can be large and complex, or relatively simple. However, common to them all is their building material – moraine boulders, i.e., glacial erratics from the Ice Age, because bedrock is not accessible in the Danish landscape (Dehn et al. 2013).
Phases of building
Planning
The nature of the planning process preceding construction of a megalithic monument is unknown, but it is obvious that the type and design of the structure were determined from the outset. Certain monument types are, to some degree, also the result of local traditions. However, generally it must have been decided at the start whether the monument was to be a twin passage grave, with two chambers each with its own passage, or a structure with only a single chamber and passage, in order to obtain the required building materials. Calculation for the passage grave of Maglehøj shows that it took 990 tons of building materials, including 832 tons for the covering mound, to build a medium-sized, singlechambered monument (Holten 2009, fig. 2). The dimensions and form of the capstones, and several of the orthostats, were exploited to the utmost, and complemented each other in a way that must have been worked out in advance. As an example, close fitting to leave minimal space between the capstones is only achievable if the capstones have been chosen according to their length and shape, and their individual and relative positions have been determined in advance. This is clearly illustrated by the Rævehøj passage grave (Fig. 6.1). Only when the exact form and size of the roof – i.e., the capstones – are known, is it possible to design a more detailed ground-plan for the chamber. The precise width of the chamber and the inclination of the orthostats will then be adjusted to fit the chosen capstones. It was also necessary to know in advance whether there were any specific architectural requirements, such as stones needing to be put in a particular place in the construction (Dehn 2009, 21–23). This is the case, for example, in the passage grave of Ørnhøj, where a pair of twin stones – two pieces of the same moraine boulder – apparently had to be used as corner stones, even though they are too tall relative to the rest of the construction (Dehn and Hansen 2000, 220). The passage grave of Kong Svends Høj includes five pairs of twin stones – one pair in the passage and the remainder arranged together at one end of the chamber. As twin stones are always arranged symmetrically relative to each other, the positioning of a large number of twin stones must have been arranged in advance (Dehn et al. 1995, fig. 57).
Fig. 6.1: Cross-section of the passage grave at Rævehøj with elevation plan of the passage. The chamber is 2.3m high and the capstones lie on three intermediary layers above the orthostats. They slope inwards so the walls form a kind of vault. The capstones overlap only a few centimetres with the uppermost intermediary layers, and their length is considerably less than the width of the chamber at floor level. (Drawing: T. Dehn)
The place where the capstones of the chamber and passage meet constitutes a weak point. The transition from passage to chamber can be constructed with or without a keystone. The keystone is the innermost capstone of the passage, on which one or two of the chamber capstones rest. It is therefore part of the passage roof, but also forms part of the chamber wall. The style without a keystone is characteristic of a small geographical region: solutions involving a keystone are much more common (Hansen 1993, 46–50; Westphal 2015). The ground-plan needs to take account of the chosen solution, as the height and form of the cornerstones are crucial with respect to the keystone.
It was probably decided at an early stage whether there would be a closing stone, because this is important relative to the ground-plan and to the entire construction process. The closing stone is a commonly-observed phenomenon. It stands in the chamber wall, but is not part of the load-bearing construction. It is smaller than the other stones making up the wall, and was put in place from outside the chamber at the end of the construction process. This was possible because of a channel running through the partially-built mound: it led into the chamber, providing alternative access to the passage, and was useful for transporting materials in and out of the chamber. It may also have had a ritual significance (see Fig. 1.6; Dehn et al. 2004, 165–167).
When establishing the building site, it can be presumed that the most important building materials, such as capstones, twin stones, and other significant stones in the construction were transported to the site. Setting out the most important elements of the ground-plan would have been done at a very early stage. On the mound base in the twin passage grave at Østrup there was a row at least 3m long of 10–15cm diameter stones running towards a point in the middle of the wall line that separates the two chambers (unpublished investigation 2007). In Kong Svends Høj, just above the base of the mound next to an orthostat, there was a row of fist-sized stones. This was aligned to a passage wall and is therefore interpreted as a sighting line used during the early building phases (Dehn et al. 1995, 28–30). It was presumably important that the orientation and direction of the passage could be marked out precisely by eye. This is apparent from the fact that, in the majority of passage graves, a line continuing from one – or both – of the sides of the passage generally strikes a space between two orthostats in the rear wall of the chamber (Dehn and Hansen 2006b, 53–54). An investigation into the orientation of passages in 105 Danish passage graves proposes three hypotheses relative to the determining factor: the rising sun, the rising full moon and the rising full moon before an eclipse. The conclusion reached in this study is that orientation relative to the moon is most likely (Clausen et al. 2008, 227).
The initial phases of construction
Early work on the construction site involved preparation of the chamber area. Sometimes, as seen in Jordehøj, the topsoil was removed and laid over the intact soil surface directly behind the planned position of the chamber walls, in order to form the base of the mound (Dehn et al. 2000, fig. 3.19). The orthostats were generally erected in individually tailored sockets, such that their final height in the structure must have been fixed at this stage. This shows that key architectural components, for example, stones with special characteristics in terms of form, colour, symmetry or other significant features, were also worked out at this point. In addition, the architectural and technical requirements necessary for the capstones would have been considered, although they were not placed into position until much later.
Fig. 6.2: Cross-section through the lower part of the mound construction at the gable of the Maglehøj passage grave. The chamber lies to the right of the edge of the drawing, and the original soil layer is labelled “9”. The edge of the capstone can be seen over the intermediary layer comprising four slabs, lying on the orthostat. A continuous packing of crushed flint runs from the orthostat foundation ditch, stepwise up through the mound construction to the sloping flagstones of the capstone roof. The construction is not preserved above this level due to an intrusion later in prehistory. (Drawing L. Holten and M. Nissen)
Once the orthostats were erected, the decision was made as to whether the drainage system should include a flint-filled ditch behind the orthostats, as seen in Jordehøj (Dehn et al. 2000, 86–88) and Maglehøj (Figs 6.2 and 6.3; Dehn and Hansen 2007, 18–20). This ditch formed the base for the flint packing that was gradually raised in height, together with the dry-walling and the core of the mound. The upper part of the flint packing was in contact with the flint in the roof construction, and acted to divert water percolating down through the mound.
Fig. 6.3: Close-up of the excavation section in Fig. 6.2. The packing of crushed flint can be followed from the trench in which the orthostats stand, up behind them and then behind the intermediary layer. (Photo T. Dehn)
Dry-walling
The walls and ceiling of the chamber had to be completely watertight so that penetrating water did not wash the mound fill into the chamber, destabilising the construction. The often irregular gaps between the orthostats were therefore sealed with shaped, flat flagstones – usually of sandstone – laid horizontally on top of each other. This is referred to as dry-walling, as today these flagstones appear to have been laid without a binding agent. However, originally clay, chalk mass, or birch bark were all used (Dehn and Hansen 2006a).
As the dry-walling was built, its rear face was secured with a packing of crushed flint that, in turn, was held in position by the mound fill, comprised of stones, soil, or a combination of the two. This progressed in alternating stages: courses of dry-walling, then crushed flint and mound fill, followed by further courses of dry-walling and so on. Dry-walling construction would have required access to both sides. Waste stone chips from shaping of the flagstones, and splinters from crushing of the flint, were seen in the rear of the dry-walling at Maglehøj passage grave (Figs 6.2 and 6.3).
Once the orthostats were in place, the dry-walling was built up level with their top, and the core of the covering mound was raised to hold the flint packing in place. Then the seating for the capstones had to be adjusted. In some areas, the capstones lie at a level corresponding to the top of the tallest orthostat: in others the wall height is increased by adding one to three layers of intermediary stones. The orthostats lean slightly inwards and the stones of the intermediary layer also lie slightly displaced inwards relative to one another, so that a kind of vault is created. The intermediary layer is also secured at the rear with flint packing, and chalk mass or bark fills the horizontal spaces.
Capstones
Details of the chamber ground-plan must have been adjusted around the capstones, and these must hence have been selected and brought to the building site at a very early stage in construction. Observations suggest a number of ways in which these capstones, weighing 10–15 tons, could have been placed on top of the orthostats and the intermediary layer at a height of up to 2.5m above ground level. However, just as there are regional differences in construction, so there may also have been regional differences in building methods. Evidence relating to the handling of capstones is rare but includes traces of temporary platforms, level with the top of the orthostats or the intermediary layer. Flat-topped platforms were formed by the uppermost part of the core of the mound, constructed in parallel with the wall, the dry-walling and the possible intermediary layer in order to hold these elements in place, and were 1.5–2.5m wide. In Jordehøj, it was apparent simply as a thin wash layer in the homogeneous clay fill (Dehn et al. 2000, 90–92), whereas the platform in Birkehøj was consolidated with a very even, cobbled surface made of small stones laid in sand. In the latter case, the platform had been established along parts of both longitudinal sides of the chamber. It is not clear whether the actual capstones lay on this platform during the operation or whether it was just used by the builders during their handling of the stones (Fig. 6.4).
In addition to the two above-mentioned platforms, four more examples are known across Denmark. The first observation of the remains of a platform was at the passage grave at Tustrup (Kjærum 1955, 30) where, at a level matching the tops of the orthostats: “it was possible to follow a pale-coloured stripe of fine sand that was covered with a very thin dark layer, presumably an old humus layer”. At Lundehøj, 5km from Jordehøj, at a corresponding level, there was a 15–20cm thick horizontal layer of hard, chalk-rich clay, (Fig. 6.5; Ørsnes 1957, 229–230, fig. 3). The passage grave at Kragnæs contained “a 3–5 cm thick, greyish-black stripe which sloped upwards towards a point just beneath the top of the orthostats. This stripe consisting of slightly sandy loam can hardly be anything other than the remains of a vegetation layer formed during a first phase of the mound’s construction” (Skaarup 1985, 251). As it appeared to have been trampled by activity, a c. 10cm thick horizontal layer in Kong Svends Høj is also interpreted as the traces of a platform (Dehn et al. 1995, 30, fig. 30).
Fig. 6.4: The passage grave of Birkehøj. Behind the surface of the chamber’s intermediary layer a cobbled platform can be seen which presumably was established in conjunction with laying the capstones in place. The orthostats in the lower part of the picture have partly toppled into the chamber. (Photo T. Dehn)
These observations suggest that once the chamber had been erected and stabilised by the mound core, laying the capstones was the next step. The surface of the mound core bears evidence of this, either as a trampled activity layer, or as a layer reinforced with stones or clay/chalk. Some of the layers are horizontal, others appear to be slightly sloping: two have traces of an associated vegetation horizon.
Fig. 6.5: Cross-section of the chamber in the passage grave of Lundehøj, excavated in 1955 due to its poor state of preservation. The layer labelled “3” is the reinforced platform of clay/chalk, which was established in connection with laying of the capstones. The crosshatching denotes crushed flint, which is seen, for example, in the ditch behind one of the orthostats. (After Ørsnes 1957)
Fig. 6.6: A suggestion as to how the task of laying the capstones on Birkehøj may have been accomplished. The drawing shows a situation where the capstones have been raised and stabilised by the core of the surrounding mound. To the lower left, a channel can be seen running through the mound where the closing stone has not yet been put in position. (Drawing T. Bredsdorff)
These differences could reflect both different methods of laying on the capstones and the fact that the surface of the mound core was not prepared in the same way all the way round the upper edge of the chamber. This could depend on where the capstones had been transported from and how they were to be raised. A ramp and platform need only be constructed on part of the mound core and the chamber. All the above-mentioned observations were made in one section of a ditch within individual structures. It is therefore not possible to conclude more than that ramps are evident in those constructions in those sections. The fact that a ramp cannot be demonstrated in a particular section does not necessarily mean that one was not present in the monument.
In four cases, traces of posts have been demonstrated in the mound structure. The posts must have been removed or sealed by the continuing building process. Three extended down into the ground beneath the mound core, all located very close to the rear of an orthostat (Dehn and Hansen 2007, 25), while the fourth was located in the intermediary layer (Dehn et al. 2004, 166–168). Two of these posts had been rocked back and forth in line with the longitudinal axis of the chamber, (Fig. 6.6), perhaps in connection with post removal. All the postholes were carefully sealed before the building work continued.
Laying the capstones must have been the most complicated part of the building process. Moraine boulders weighing 10–15 tons had to be moved and positioned very precisely at a height of 2.5m above ground level, with limited room for manoeuvre and a relatively loose working surface. Examining the underlay of the capstones, either from inside the chamber or, more rarely, from outside, suggests the surface was adjusted with flat or wedge-shaped stones before the capstones were finally placed in position. Adjustments achieved the correct fall of the flat roof surface in relation to the capstones’ contact points with the orthostats and the dry-walling, and compensated for the natural irregularity of the moraine boulders.
In two cases it appears that control of a capstone was lost as it was being manoeuvred into position: it slipped too far down into the chamber and became lodged. One accident led to the major fracturing of a capstone and a partially destroyed dry-wall: it was then roughly rebuilt from inside the chamber (Dehn 2009, 24). Considering how skilfully capstones were generally handled, it is remarkable that the stones were not brought up into the correct position. However, a capstone positioned lower than the others could have been an intentional architectural feature that, like a lintel, separated one end of the chamber from the rest (Hansen 1993, fig. 75). Only exposure of its upper surface can reveal the reason behind a capstone that hangs further down into the chamber.
Once all the capstones were placed and the gaps sealed, the floor was established by levelling. Clay or stone pavements of flagstones or cobbles could also be added. Birkehøj had a sunken floor covered with a layer of clay, under which was a capillary layer of gravel (Dehn et al. 2004, 164).
The passage-grave mound
When the chamber and the inner part of the passage were secured by the mound core, construction of the outer mound had to be completed. A few chambers are located in rectangular long barrows, but the majority lie encapsulated within round barrows. A round mound could be made of stone, clay, sand, or a combination of these. Various means were applied to keep the chamber and passage sealed and dry. For example, in the mound fill covering both the chamber and the passage, layers of overlapping flagstones or sloping layers of clay were intended to divert water. These layers may be in contact with the flint packing behind the walls and in the drainage ditch.
The question has been asked whether the capstones always were covered by mound fill. It is argued that some of the more simple dolmens were established as “open dolmens” with exposed chambers (Andersen et al. 2014). However, my own investigations found that two of the proposed “open” dolmens were actually covered by a mound (Dehn 2013).
Rows of single closely spaced stones are often seen forming concentric circles around the chamber within the mound. These circles have been observed at the base of the mound or at different levels within it, often behind the walls of the chamber in mounds built of stones and those with a sand fill, but they were not found in Jordehøj, which was built of clay/chalk. However, that monument was found to have two cavities in its mound fill. A plaster cast revealed that a post, almost 10cm in diameter and 64cm in length, stood vertically in the Neolithic topsoil (Fig. 6.7). Excavation suggested that posts such as this extended up to about 1.2m, reaching a level directly below the horizontal platform (Dehn et al. 2000, 88–90). A wooden structure was also found in the ploughed-out passage grave of Vroue Hede III: there were the remains of two concentric rings of burnt wood between the traces of missing kerbstones and the chamber (Jørgensen 1977, 107).
These concentric stone rows may mark the various stages of mound construction, and could have stabilised materials during building work. They could also have controlled the form of the mound during construction, like the Bronze Age wooden wattle structures seen in Lusehøj (Thrane 1984, 93–98). Pottery deposits associated with these stone rows in the Neolithic mounds suggest that different stages in the construction process were marked by particular activities. This was seen at the passage-grave site of Damsbo Mark A1, where there was apparently a spiral rather than concentric circles (Andersen 2011, 152, fig. 5). At Klekkendehøj, a hearth was seen in the stone-built lower part of the mound (Dehn unpublished investigation 2011), and a dolmen at Tårup also had traces of fire in the initial phase of the mound construction (Holst 2006, 12). Were such activities and deposits connected with longer pauses in the building work, or did they represent ritual events during construction?
Fig. 6.7: Vertical cavity at Jordehøj. In the clay-rich mound fill of the passage grave, vertical cavities were apparent in the area beneath the platform. On being filled with plaster of Paris, one of them proved to be the impression of a post 10cm diameter and 64cm long. The post had stood in the topsoil at the base of the mound and observations during the excavation suggest that it reached 1.2m in height. (Photo T. Dehn)
Fig. 6.8: Cross-section of the twin passage grave at Klekkendehøj from 2001, with elevation plan of the northern passage and the wall stones between the northern and southern chambers, a section through the eastern part of the mound, and the section trench in the terrain at the foot of the mound. The pit at the foot of the mound contained a layer of small detached stone fragments, presumably waste from the construction of the chambers and passages. The edge of the pit is overlain by the lowermost of two presumed terrace steps. In a profile section from 1987, only one terrace step could be recognised. (Drawing P. E. Skovgård and J. Westphal)
Kerbstones
Kerbstones were erected in one of the final building phases, often with dry-walling. Kong Svends Høj has a long barrow covering a 12.5m long chamber: the dry-walling between the kerbstones was so high that a channel was necessary between them and the mound core in order to provide rear access during construction (Dehn et al. 1995, 32). At c. 1.5m, the kerbstones in Kong Svends Høj are some of the tallest recorded, and the gable stones are almost 4m high. A similar approach was observed at a 90m long mound at Ibjerg where the kerbstones were erected and stabilised with stones and crushed flint, forming a 1m wide band associated with the actual earthen fill of the mound (Juel et al. in press).
Kerbstone constructions can also be considerably more modest. For example, those around Maglehøj are less than 0.5m in height, with dry-walling between them. Some monuments appear to have only small, widely-spaced kerbstones, with such large gaps between them that dry-walling was apparently not needed. However, bearing the variable preservation in mind, it is not normally possible to understand the method of kerbstone construction without some form of study, usually sectioning the mound in order to clarify the monument’s history before verifying the construction method. Mounds may have been plundered for their kerbstones in recent times, but passage graves were also often expanded and altered in later prehistoric periods, typically in the late Neolithic and Bronze Age, such as at the passage grave at Bigum, which was rebuilt and altered several times. A new lower kerbstone construction was laid outside the original structure from the passage grave mound, and the latter was raised with a layer of horizontal stones (Dehn et al. 2000, fig. 14, 17).
The stone pavements seen in front of the passage entrance or around the entire mound also constitute elements of the kerbstone construction. For example, the passage grave at Sarup Gamle Skole II (Andersen 1997, fig. 127) had traces of a c. 4m wide stone pavement running around the mound at the foot of the kerbstones. At Nissehøj, a semicircular stone pavement sealed an Early Neolithic grave in front of the entrance with an offering area (Holten 2000, 290–291). However, it is not certain that these elements were part of the monument’s construction phase: they could represent later changes.
The latter is not the case for another, rarely-occurring element – terracing. All around Klekkendehøj, there is a 1–1.5m flat horizontal platform at the foot of the mound and c. 1m above ground level. The floors and entrances to the passages are at the same level as this platform entrance. In 1987, a section trench in the south side of the mound revealed it to be an original feature (Dehn et al. 2000, 39–44). A section trench through the west side of the mound in 2001 confirmed that the terracing was the final step in the building process and that there was potentially an additional terrace. The outermost part of the terrace feature lay stratigraphically over the edge of a pit, on the bottom of which were small stones and fragments of larger stones. These, together with other material, are interpreted as waste from the construction of the chamber/passage and the core of the mound (Dehn and Hansen 2007, 30). Within the trench, the pit reached 1.8m in depth and extended 7m in one direction, but its actual size remains unknown. It is interpreted as a source of mound fill material, and was backfilled in connection with extensive levelling-out of the terrain after construction was completed (Figs 6.8 and 6.9).
The sophisticated construction of passage graves lead us to conclude that construction was largely carried out by specialists, perhaps even travelling master craftsmen. One calculation demonstrates that, in the islands of Bogø and Møn (an area of 231 km2), an average of one megalithic monument a year was constructed, and a similar calculation for the Sarup area suggests an average of one newly-constructed megalithic monument every two years (Andersen 1997, 40). According to yet another calculation, on average two passage graves a week were built across the area that is now Denmark (Ebbesen 2011, 239). Regardless of the uncertainty associated with such calculations, these monuments represent the investment of unprecedented efforts by a relatively modest population over the course of a few generations. This must, in turn, have resulted in a great accumulation of knowledge and experience among the people who took part in the building work. The planning and leadership was presumably in the hands of a few individuals, but gathering and transport of the materials, as well as the actual construction work, must have involved a considerable number of people with extensive associated logistical support. As a consequence, in the generations during which this building work influenced daily life, there must (to some extent) have been widespread knowledge and experience of megalithic construction. In addition to building new monuments, it would also have been necessary to carry out regular maintenance, alteration, and adaptation of existing monuments.
Passage graves as architecture
The megalithic construction work described above represents only the technical and tangible part of the process. It is beyond doubt that the building of a structure able to stand for 5,000 years constituted a significant engineering and logistical achievement (Westphal 2015). However, Neolithic monument builders had motives beyond just raising a structure. To a great extent, the driving force was the social, religious, political, and strategic dimensions inherent in both the construction process and the actual end product – a monument that stood as confirmation of agreed alliances. The commitment of resources to build even a single monument suggests that each involved more people than there was a space for in the area around the individual structure. During the construction process, participants became part of the fellowship associated with this, and would later have been able to swap stories about special events that took place during the work, and about details of the construction known only to those who took part.
An interrupted or a continual construction process?
The structure of large megalithic monuments can reveal details of the organisation that lay behind the undertaking. Were there multiple discrete construction periods over several years, or did the work involve a continual sequence, ending with a completed structure? In the investigations described here, no indications of stagnation or breaks in the work have been found. The various components of the monuments combine to form a complex whole. In only two cases are there indications of a vegetation horizon associated with a platform established in connection with the placement of the capstones. Other traces, in the form of wash layers, subsidence or other forms of deterioration within the structure, have not been observed. A break of longer duration would, for example, have left traces in the very homogeneous flint packing behind the orthostats and the dry-walling. On the other hand, activity layers have been observed that reflect trampling during the building work and transport of materials. There are also traces of presumably ritual-based activities such as fires and the deposition of pottery. The vegetation horizons could represent a pause in the building work, but they relate to the placement of the capstone, a particularly crucial part of the process, so the platform may have stood unused while activities proceeded elsewhere. There is nothing to suggest any extended interruptions to the process.
Fig. 6.9. Detail from the section drawing Fig. 6.7 with a presumed material pit which was back-filled during extensive levelling of the terrain. The stones lay c. 0.5m above the base of the pit and at the side closest to the chamber. They presumably represent waste from the building work. (Photo T. Dehn)
It is stated above that the building process illustrated here led to the production of a finished monument. This product is not, however, identical with the final form of the monument as we see it during an archaeological investigation. In addition to the alterations and additions during later periods of prehistory, especially in the late Neolithic and Bronze Age, the builders and their immediate Funnel Beaker culture descendants probably carried out regular maintenance and modifications. Anthropological sources demonstrate that, with time, houses develop in form according to the people who use them and the circumstances surrounding them: they are changed by alterations and additions. “Building, then, is a process that is continually going on, for as long as people dwell in an environment” (Ingold 2000, 188). It seems likely that the same was also true after a megalithic monument had been constructed. The passage grave of Birkehøj demonstrates that both maintenance and alterations were carried out. The unusual architecture in one end of its chamber suggests that an orthostat toppled over after the capstones had been put in place, requiring it to be re-erected and secured as best as possible – although the only access was from inside the chamber. This incident presumably occurred as the toppled orthostat stands at the end of the chamber formed by an earlier small dolmen chamber that was later altered and extended to produce the 11m long passage-grave chamber (Dehn et al. 2004, 169–171).
Fig. 6.10: Plan of the two ploughed-down long dolmens at Ibjerg, excavated between 2007 and 2009. A structure associated with the construction became apparent at the base of the southern mound, in the form of a longitudinal central axis with several sections at each side. (After Juel et al. in press)
Conclusion
This attempt to illustrate megalithic building processes is based exclusively upon the author’s own archaeological investigations and observations of the architecture in about 70 Danish megalithic monuments, especially passage graves, supplemented by examples taken from the literature. It is not certain that the same conditions and situations prevailed in all areas with megalithic monuments, or with respect to earlier monuments such as long barrows and long dolmens. In one of the two, now ploughed-out, parallel long dolmens at Ibjerg, there were suggestions in the mound construction of a cell-like structure that can be interpreted as either a step-like building process or the involvement of several construction teams (Fig. 6.10; Juel et al. in press). A similar phenomenon, seen in the English long barrow chamber at Ascott-under-Wychwood, Oxforshire, where the phases have different dates, is interpreted as follows: “… the dead were incorporated into a construction site and not a finished tomb” (Bailey et al. 2010, 565). The Danish record on which this description of the building process is based gives no grounds for arriving at a similar conclusion. However, the idea of continual construction and rebuilding could be the reason why megalithic monuments most often lie in clusters comprising long barrows, round dolmens and passage graves, as seen for example at Lønt (Gebauer 2015). In the individual monuments, the traces left by the construction site and work on the passage graves reflect a planned, well-considered and continual effort.
Acknowledgements
The foundation for this article was a fruitful collaboration with my colleagues Svend Illum Hansen and Jørgen Westphal, who do not necessarily agree with all my interpretations.
Translation from Danish: David Earle Robinson and Anne Bloch.
Andersen, N. H. 1997. The Sarup Enclosures. The Funnel Beaker Culture of the Sarup site including two causewayed camps compared to the contemporary settlements in the area and other European enclosures. Jutland Archaeological Society Publications 33(1): Moesgaard.
Andersen, N. H. 2011. Causewayed enclosures and megalithic monuments as media for shaping Neolithic identities. In: Furholt, M., Lüth, F. and Müller. J. (eds.) Megaliths and Identities. Early Monuments and Neolithic Societies from the Atlantic to the Baltic. Frühe Monumentalität und soziale Differentierung 1. Dr. Rudolf Habelt GmbH: Bonn, pp. 143–154.
Andersen, N. H., Eriksen, P., Kjærum, P. and Scarre, C. 2014. Stendysser. Arkitektur og Funktion. Jysk Arkæologisk Selskab: Aarhus.
Bailey, D. and McFadyen, L. 2010. Built objects. In: Hicks, D. and Beaudry, M. C. (eds) The Oxford Handbook of Material Culture Studies. Oxford University Press: Oxford, pp. 562–587.
Clausen, C., Einicke, O. and Kjærgaard, P. (2008) The orientation of Danish passage graves. Acta Archaeologica 79, 216–229.
Dehn, T. 2009. The megalithic building site. In: Scarre, C. (ed.) Megalithic Quarrying, Sourcing, Extracting and Manipulating the Stones, BAR S1923, Archaeopress: Oxford, pp. 21–25.
Dehn, T. 2013. “Open dolmens” – a matter of decay? In: Bloo, S. B. C., Bakker, J. A. and Dütting, M. K. (eds) From Funeral Monuments to Household Pottery – Current Advances in Funnel Beaker Culture (TRB/TBK) Research. Proceedings of the Borger Meetings 2009, The Netherlands. BAR S2474, Archaeopress: Oxford, pp. 96–109.
Dehn, T. and Hansen, S. I. 2000. Doubleness in the construction of Danish passage graves. In: Ritchie, A. (ed.) Neolithic Orkney in its European Context. McDonald Institute for Archaeological Research: Cambridge, pp. 215–221.
Dehn, T. and Hansen, S. I. 2006a. Birch bark in Danish passage graves. Journal of Danish Archaeology 14, pp. 23–44.
Dehn, T. and Hansen, S. I. 2006b. Megalithic architecture in Scandinavia. In: Laporte, L., Joussaume, R. and Scarre, C. (eds) Origin and Development of the Megalithic Monuments of Western Europe. Musée de Bougon: Bougon, pp. 39–61.
Dehn, T. and Hansen, S. I. 2007. Examples of megalithic technology and architecture in Denmark. In: Bloemers, J. H. F. (ed.) Tussen D26 en P14: Jan Albert Bakker 65 jaar. Amsterdams Archaeologisch Centrum, Universitet Van Amsterdam: Amsterdam, pp. 17–31.
Dehn, T., Hansen, S. I. and Kaul, F. 1995. Kong Svends Høj. Restaureringer og undersøgelser på Lolland 1991. Stenaldergrave i Danmark 1. Skov- og Naturstyrelsen: København.
Dehn, T., Hansen, S. I. and Kaul, F. 2000. Klekkendehøj og Jordehøj. Restaureringer og undersøgelser 1985–1990. Stenaldergrave i Danmark 2. Skov- og Naturstyrelsen: København.
Dehn, T., Hansen, S. I. and Westphal, J. 2004. Jættestuen Birkehøj. Restaurering af en 5000 år gammel storstensgrav. Nationalmuseets Arbejdsmark 2004, 153–173.
Dehn, T., Hansen, S. I. and Westphal, J. 2013. Restoration of megalithic tombs in Denmark. In: Bergerbrant, S. and Sabatini, S. (eds.) Counterpoint: Essays in Archaeology and Heritage Studies in Honour of Professor Kristian Kristiansen. BAR S2508, Archaeopress: Oxford, pp. 695–702.
Ebbesen, K. 2011. Danmarks Megalitgrave 1(1). Attika: København.
Gebauer, A. B. 2015. Two types of megaliths and one oddball dolmen at Lønt, Denmark. In: Laporte, L. and Scarre, C. (eds) The Megalithic Architectures of Europe. Oxbow Books: Oxford, pp. 137–144.
Hansen, S. I. 1993. Jættestuer i Danmark. Konstruktion og restaurering. Skov- og Naturstyrelsen: København
Holten, L. 2000. Death, danger destruction and unintended megaliths. In: Ritchie, A. (ed.) Neolithic Orkney in its European Context. McDonald Institute for Archaeological Research: Cambridge, pp. 287–297.
Holten, L. 2009. Åbninger til en anden virkelighed. Megalitanlæg som mediatorer mellem her og hisset. In: Schülke, A. (ed.) Plads og rum i tragtbægerkulturen, Det Kongelige Nordiske Oldskriftselskab: København, pp. 159–177.
Ingold, T. 2000. The Perception of the Environment. Essays on Livelihood, Dwelling and Skill. Routledge: London and New York.
Juel, C., Hansen, J. and Andersen, A. B. 2015. Grave i lange baner – et højkompleks fra tragtbægerkulturen ved Ibjerg sydøst for Odense Aarbøger for nordisk Oldkyndighed og Historie.
Jørgensen, E. 1977. Hagebrogård – Vroue – Koldkur. Neolithischer Gräberfelder aus Nordwest-Jütland. Arkæologiske Studier IV. Akademisk Forlag: København.
Kjærum, P. 1955. Tempelhus fra stenalder. Kuml 1955, pp. 7–35.
Ørsnes, M. 1957. Om en jættestues konstruktion og brug. Aarbøger for nordisk Oldkyndighed og Historie 1956, pp. 221–234.
Skaarup, J. 1985. Yngre stenalder på øerne syd for Fyn. Meddelelser fra Langelands Museum. Langelands Museum: Rudkøbing.
Thrane, H. 1984. Lusehøj ved Voldtofte – en sydvestfynsk storhøj fra yngre bronzealder. Fynske Studier XIII. Odense Bys Museer: Odense.
Westphal, J. 2015. In the Eye of the Beholder. Emergence of architectural elements in the visual analysis of Danish megalithic tombs during the last 25 years. In: Laporte, L. and Scarre, C. (eds) The Megalithic Architectures of Europe. Oxbow Books: Oxford, pp. 89–97.