Added
- Under Construction!
- If
figures are not visible, it is a browser problem.
- Footnotes
have been removed due to corruption during conversion.
The
dissertation has a copyright. Please give proper credit when quoting its
content.
A SPATIAL ANALYSIS OF MEGALITHIC TOMBS
6. SUB-
AND NON-MEGALITHIC GRAVES AND THE ORIGIN OF
MEGALITHIC TOMBS.
By
Maximilian O.
All rights reserved
6.1
Introduction
TRB
megalithic tombs, as the most spectacular burial architecture, have captivated
researchers for centuries, but to fathom their origin it is necessary to survey
their relationship to the seemingly unspectacular sub- and non-megalithic
graves. Failure to understand this relationship has led to ironic conclusions.
For example Brøndsted (1957:190), quoted by T. Madsen (1979:315), resolved
that the “dolmen is the visible result
of incoming religious ideas.” Thus, according to T. Madsen: “Brøndsted clearly expressed the generally
held opinion that the coming of megalithic graves meant a decisive change in
the religious life of the early farmers.” Contrary to this still common
view, the seemingly less impressive,
difficult to find and hard to date graves (Fig. 6.1) are an integral part of
the TRB's mortuary complex.
Figure 6.1. Mound LA 14 with urdolmen-size burial pits,
Like many of the TRB's culture traits, the burial practices are
inherited from earlier times, although their connection to the preceding
cultures are, as the TRB's origin itself, hardly understood. Yet it is this
seemingly simple grave architecture which gave rise to megalithic chambers and
mounds. Their architectural forms coexisted with megalithic tombs, frequently
in the same mounds. They even outlasted megalithic architecture, perhaps giving
rise to the KAK and EGK/Corded Ware culture burial architecture, as seems
to be indicated by the stone-packed graves and sub-megalithic MN B/LN cists.
Continuous use of the same burial site is, for example, demonstrated at Bordesholm-Tannenbergkoppel. There a Konens Høj type grave (Fig. 6.2) was found near two megalithic tombs, along with MN B/LN earth-graves, Bronze Age tumuli, and Late Bronze Age and Iron Age urn graves (Hingst 1974:20-22, 67). The same is true for the nearby Flintbek settlement area where the earliest occupation starts during the EN I, most likely sometime between 4400-3860 B.C. (Zich 1992, 1994a, b, 1995a ).
Figure 6.2. Konens Høj type grave with EN C/II funnel beaker, Bordesholm-Tannenbergkoppel, Schleswig-Holstein (Hingst 1974:22 Fig. 4).
|
Table 6.1 Non- and sub-megalithic tombs ¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾ Non-megalithic
graves Unlined
burial pits Coffins
(in pits) Wood-lined
and/or covered burial pits ¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾ Sub-megalithic
graves (small to medium size fieldstones and stone slabs) Stone
pavements usually on level ground Fieldstone-frame
on level ground Stone-lined
and/or covered burial pits Coffins
in stone-lined and/or covered pits or on level ground Cists
of sub-megalithic stone slabs Masonry
chambers with stone or wooden roofs Corbeled
chambers |
|
|
6.2
The
roots of the TRB burial practice and architecture
A bewildering variety of burial architecture occurs in the TRB. It may be categorized into non-megalithic, sub-megalithic, and a megalithic architecture, albeit with considerable overlap, including hybrid tombs. The graves may or may not be covered by a mound (cf. Ebbesen 1994a). Without endeavoring to prescribe a new typology, the architecture is listed under two classes (non-megalithic = earth and wood architecture, sub-megalithic = architecture incoporating), according to increasing architectural complexity in Table 6.1
Several of these forms
arose before the TRB became clearly definable in the archaeological record.
Among these are the true graves or burial pits originally called earth- or
flat-graves to distinguish them form burials within mounds. The true graves
were typical of the Danubian LBK, STK, Rössen, Lengyel, and non-Danubian
Ertebølle cultures.
Haüsler views TRB's flexed burials as an inheritance from the Danubian agriculturists. The reappearance of extended burials on the western fringes of the former Danubian culture area is thus seen as a reversal of the traditional agricultural flexed burial practices (Häusler 1992:278, 282, 285, 291; 1994:37-38). The TRB, which straddles the ancient Danubian domain and that of the Mesolithic, is said to express this hunter-gatherer / agriculturist burial dualism in its practices. Thus the southern Baalberge flexed burials (resting on the right side) first oriented west, later east, are stated to be derived from the Lengyel sphere. The later Bernburg phase and MN V KAK continued this norm.
LBK, STK and Rössen burials usually contained a single, flexed
interment similar to the MN TRB graves of
Figure 6.3.
The silhouette of a skeleton in flexed position in a burial pit Heek, Kr. Borken,
Westphalia (Finke 1980:14 Fig. 7).
The graves of each of the older cultures have a relatively consistent
orientation that shows a gradual change through time. More or less standardized
orientations can also be observed in EN graves. In
The older cultures
occasionally interred two individuals in the same grave in close proximity.
Both interments had the same orientation, but the individuals are placed in
exactly the opposite direction, sometimes even
antipodal, i.e. foot to foot (e.g. Behrens 1973). This position is
foreshadowed in the Swedish Ertebølle cemetery Skateholm II.
Grave 10 has an antipodal-like double grave, although the two skeletons
are more side by side rather than foot to foot, with one male extended, the
other in a sitting position (M. Larsson 1989:214-216, Fig. 3, 5). The
four TRB extended skeletons in the (wooden) plank lined grave at
Bygholm Nørremark, Jylland, were fully antipodal (T. Madsen 1979:309;
1993:99 unnumbered Figure; Rønne 1979:5-6). According to Tempel (1979:114) it
is possible that the 40 cm high “stone-bench” architecture of
Cuxhaven-Gudendorf (Fig. 6.__) also contained such a (foot to foot?) double
burial.
Figure 6.4.
Grave (the 3.85:1.20 m ) with stone benches-like courses of small to medium
size field stones Cuxhaven-Gudendorf (Tempel 1979a:115, Fig. 2). Pottery
not to scale.
The two pots from Cuxhaven-Gudendorf were of slightly different periods
(Fig. 6.4**).
Just like these graves, the burial antipodal position within them
probably was in use throughout a wide area for a long time. Group burials with
flexed antipodal position are shown by Behrens (1973:240 Fig. 102) for the
Salzmünde phase. The KAK flexed antipodal double burial at Zauschwitz (Behrens
1993:241 Fig. 103a) shows that this practice continued right to the end of
the MN A.
Figure 6.5.
Stone framed graves below Sarnowo (Chimielewski 1952:50-51, 56 Fig. 17-18, 23).
Left and center = extented burial with collared flask Mound 1, Grave 1. Right =
Mound 2,
The burial pits (Ebbeson’s Type a) plus graves with a single layered
fieldstone frame (Type b) and the supine burials are thought to be a
continuation of the Mesolithic burial traditions (Ebbesen 1994a:80). They are
demonstrated for Sarnowo (Fig. 6.__). But skeletal remains from EN non- and
sub-megalithic graves are extremely rare in
Figure 6.6 Grave II, Drugsholm, Fårevejle s., Holbæk a., Sjælland (Tilley 1996:81 Fig. 2.5 after P. Nielsen 1981). The ca. 20 year old male was buried with a Type A funnel beaker, a Type I battle-ax, transverse arrowheads, pendants etc. in a simple east-west oriented Oxie Group Type a burial pit (Ebbeson 1994:88) dated ca. 3800/3700 B.C.
There are indeed remarkably similar mortuary practices recorded for the Late Mesolithic and the TRB (cf. Tilley 1996:109-115). These include single and multiple interments in flexed and extended positions. Drugsholm is a good, early example (Fig. 6._). The remains of two “dismembered” individuals at Fakkemose, Langeland, found in a pit below a dolmen in tumulus (Skaarup 1985:207 Fig. 222, 1990:79 Fig. 6) may also imply a close relationship.
But perhaps the most convincing evidence stems from EN
C/II grave from Lohals, associated with a ribbed collared flask (Skaarup
1985:324-325 Fig. 374-375). The position of the extended supine male and the
the somewhat supine female on top of him is nearly identical to the Mesolithic
interment of Skateholm I Grave 14 (eg. Coles and Coles 1989:71 Fig. 50).
Also The TRB tree trunk coffin of Himmelpforten Grave V (Tempel
1979a:111-112), which itself foreshadows a Bronze Age practice (Fig.
2.__*), and similar EN burials in
In the Late Mesolithic
Figure 6.7 Extended supine burial with beads and boar’s tusks in stone-framed grave, Sarnowo Mound 3, Grave 1 (Chmielewski 1952:62 Fig. 29).
At Skateholm two sites have been located
approximately halfway between the passage-graves of Ö. Torp and Snårestad.
Skateholm I had fifty-three graves. One grave had a sitting female with a
baby in her lap, as well as 30 tusks from wild boars. Skateholm II is the older
cemetery. It contained sixteen graves.
In one grave a wood/bark coffin held a man, two axes, four blades, a
bone needle etc. In Grave 15 a young man was interred in a seated position with
three antlers used to prop him up. The above mentioned Grave 14 contained
an old man in extended position with a young woman next to and over him in a
flexed position. Grave 26 contained a tightly
flexed elderly woman. There were six dog burials, including Grave 21 - one
of the most “richly” furnished of all the graves, containing a decorated antler
hammer, dear antlers, and three blades (M. Larsson 1989:219-220, 217
Fig. 7). Similarly, the EN TRB burial at Weißenfels,
Within the context of the two Mesolithic cemeteries there were several
hundred structures, including a ca. 60 m2 feature. A 4:4 m structure within
Skateholm II could be interpreted as a chamber-like structure, mortuary or
cult house (M. Larsson 1989:220). Some graves provide evidence of a wooden
superstructure that was later fired and backfilled. This process was
accompanied with ritual flint
knapping. There is also evidence of ritual feasts and food deposits. Some parts
of individual skeletons were removed before burial. None of these practices
would be out of place in the TRB and to some extent even in the German
Gallery-Grave culture.
Ebbesen (1994a) and Tilley (1996) argue for a continuity of Mesolithic
burial practices in
Figure 6.8: Sarnowo double burial (Chimielewski 1952:55 Fig. 24).
However, entombment in early graves did not always
accommodate merely a single, supine, extended individual. For instance, at
Sarnowo Mound 2 contained a pair of extended, side by side skeletons (Fig.
6._). Mound 8 may represent a related burial practice (Fig. 13.6). There
were three individuals near each other in the Iłowo 1
long-mound (Chimielewski 1952:41 Fig. 11). And while Häusler refers to
the supine entombment as the primary Mesolithic burial practice, he does draw
attention to a number of other, less common customs. At Tangermünde most
burials were extended, but the only double interment included a supine and a
flexed skeleton.
Likewise, at Mesolithic Skateholm a double burial (Skateholm I, Grave 14) consisted of a supine male and a flexed female (Coles and Coles 1989:71 Fig. 50, right). “Supine, sitting, and hocker (flexed) positions” were observed and “many variations occur ..., as for example, in the hocker position, where the range is from weakly-marked positions in which the extremities are only slightly angled to an extremely contorted position, which must mean that the deceased was tightly trussed up and bound hand and foot” (L. Larsson 1989:214). These Mesolithic variations are attributed to age and sex differentiation, but there was also an increased tendency towards flexed burials through time.
Similarly, the Falbygden primeval passage-grave Gökhem 17 contains
perhaps the oldest C14 flexed burial in the the
more northerly parts of TRB (Persson and Sjögren 1994, ibid. in press). The
3:2 m chamber (Kammergrab) of
Niederbösa, Kr. Sonderhausen contained 78 interments, both flexed and
extended. The depositions occurred in two
separate chamber divisions, which were assumed to represent moieties.
One individual showed evidence of a projectile point injury. The chamber
contained wood remnants, Salzmünde and Walternienburg pottery, flint tools,
animal teeth and lower jaws. The tomb was later used for Corded Ware interments
(e.g. Beier 1991a Plate 10,2; Müller-Karpe 1974:961 No. 600).
Haüsler views TRB's flexed burials as an inheritance from the Danubian
agriculturists. The reappearance of extended burials on the western fringes of
the former Danubian culture area is thus seen as a reversal of the traditional
agricultural flexed burial practices (Häusler 1992:278, 282, 285, 291;
1994:37-38). The TRB, which straddles the ancient Danubian domain and that of
the Mesolithic, is said to express this hunter-gatherer / agriculturist burial
dualism in its practices. Thus the southern Baalberge flexed burials (resting
on the right side) first oriented west, later east, are stated to be derived
from the Lengyel sphere. The later Bernburg phase and MN V KAK continued
this norm.
Although several archaeologists (e.g. Ebbesen 1994a, Tilley 1996) argue
for a continuity of Mesolithic burial traditions in the TRB, it is risky to
extend this argument to the supine position which supposedly occurs in the
earliest megalithic chambers. Aside from the fact that such chambers were built
several hundred years after the TRB becomes recognizable, there is an
extraordinary paucity of extended burials in urdolmen (cf. Chapter 7). As a
result Tilly (1996:82-83), using Ebbeson’s (1990:66) phrasing, diplomatically
states that "the body was typically
lain out on its back in supine position and no other has been documented."
Tilley can cite only Ølstykke, Kellerød and Grøfte A and B, out of presumably thousands of Scandinavian dolmen (ibid. 130-133) in support for the continuity of Mesolithic burial practices in megalithic chambers. Worse yet, the evidence largely comes from 19th and early 20th century excavations. In the case of Ølstykke, the only true urdolmen as defined below, a supine skeleton was positioned along one side of the chamber (Fig. 8.__; cf. Midgley 1992:424, 449-450; Persson and Sjögren in press). Grøfte had two cist-like chambers, neither being a true urdolmen, containing at least one double burial, whose limited remains were presumably scattered by animal action, so that there is no direct evidence of their burial position (Ebbesen 1990. Bennike 1990). In addition, assigning the rare urdolmen containig extended skeletons to the TRB is sometimes questionable (Raddatz 1979:130). This is illustrated at Bogø By, Denmark, where an extended burial and an EN C/II pot was found together with later EGK artifacts in an urdolmen without entrance (Fig. 8.3***). In fact, in danish passage-graves the only more or less articulated skeletons are the last interred (Bennike 1990). EN flexed, multiple interments occur at Preußlitz, dated to the Baalberge phase (Behrens 1973:239 Fig. 101c, Preuß 1980:24). The sole human evidence in the above mentioned Westphalian MN flat-grave also reveals a flexed burial. It is, therefore, wiser to view the Mesolithic connection in terms of the continued use of a broad variety of burial practices with regional differences that include flexed and extended burials, of adults, child and new born, animal burials, perforated teeth, boar's tusks, (flat-grave) cemeteries, wooden superstructures (later set on fire or otherwise demolished), feasts, ritual flint knapping, etc. But many of these attributes also occur in the Danubian derived cultures.
Figure 6.9 Reconstructed wooden chambers, Granstedt,
6.3
Sub-
and non-megalithic graves and their relationship with megalithic tombs.
Frequently
earth-graves are found in direct association with megalithic tombs. This was
the case in Issendorf, where at least four graves were located 24 m north
of a megalithic tomb. The graves had the same orientation as the megalithic
tomb. Unfortunately the preservation in the sand dunes was very poor so that
additional graves can only be suspected on the basis of surface finds. The
Issendorf graves showed evidence of chamber-like structures suggesting to
Two wooden chambers
(Fig. 6.5) were reconstructed in their own tumulus at Granstedt (Tempel
1984 and verbal communication May 1994). Offerings of pottery were placed on
top of the wooden roof of each chamber in a fashion documented for flat-graves
(
Non- and sub-megalithic graves cross the line
between flat-graves and megalithic chambers by turning up in mounds that also
contain megalithic chambers, thus implying similar functions. For example, at
Flintbek LA 3 a series of successive earth-graves, including the Konens Høj
type, culminated in the construction of four extended dolmen, all contained in
a rebuilt and expanded long-mound (Fig. 17.21). Stone framed and stone lined,
bathtub-shaped pits occur as single features in their own long-mounds with
megalithic enclosure Likewise stone-framed graves, somtimes apperantly erected
on the ancient soil surface, and stone-lined, bathtub-shaped graves occur in
flat-graves and mounds with rectangular and trapezoidal megalithic enclosures
(Fig. 6._).
On the west side of
the
Figure 6.10.
Stone-lined grave and bathtub-shaped pit in megalithic enclosures (Schuldt
1972:184 Fig. 30).
A pavement also appeared below the tumulus
Horneburg 1, Kr. Stade, near a megalithic tomb. The pavement ended
with an upright stone plate (Tempel 1979a). The plate seems similar to the ones
from the Issendorf flat-graves and a grave in Sievern, Kr. Stade. An
analogous flat-lying stone plate was found in the stone-lined grave within a
small rectangular long-mound with megalithic enclosure at Rothenmoor, Kr.
Sternberg in
A major concentration of such cists is located in Wollschow, Kr. Pasewalk, where roughly 40 cists and urdolmen are located in a group. Schuldt (1975) suspected that most of them belonged to the EN. Two cists were contained in a 20 m long-mound with typical trapezoidal megalithic enclosure. The proximal end harbored a cist constructed of red sandstone slabs 1.9-1.7 m long and 1.5-1.2 m high. Near the distal end was a much smaller cist (0.85:0.5 m and 0.5 m high), covered with a single stone slab. It once contained three children, two were found facing each other in a flexed position, aged six to eight years old.
Figure 6.11. Histogram of 116 TRB tombs and graves containing EN collared flasks, handled jugs and related ceramics.
Figure 6.12. Distribution of earthen long-barrows with earth-graves and timber structures (after Madsen 1979 Fig. 1). The site numbers refer to Madsen's site list and identify tombs mentioned in the quoted text. A = Rectangular and trapezoidal long-mounds and related structures containing earth graves, b = Timber façades, c = Konens Høj type grave, d = Troelstrup type grave.
Figure 6.13.
EN I long-mound with wooden façade and two sub-megalithic cists, Rude,
The Barker stone cist and the two small slab cists were found on top of
the long-mound at Rude, East Jylland, may indicate an evolution from stone slab
cist to megalithic small dolmen. The Rude cists could date to the EN I,
since the mound’s wooden façade dates 2960±90 b.c. / 3715±95 B.C. and
2860±70 b.c. / 3660±40 B.C. (T. Madsen 1980). EN parallels exist
(T. Madsen 1980:98-99). But Rude cists could also belong to a much later
period, because a somewhat similar urdolmen-like chamber (identified as a dyssekammer) with one lengthwise placed
granite erratic block (ca. 2.5:0.75:0.5 m) and two sandstone plates
contained MN IV/V pottery (Skaarup 1985:168-169 No. 209, Fig. 153-156).
The late C14 date from the skeletal remains also imply a later date.
In spite of the difficulties in dating some architectural forms, it is obvious that a variety of architectural experiments, leading to different resolutions in creating the proper (final?) resting place for revered dead, yielded the coexistence of diverse structures. Fig. 6._ compares dolmen and the various non- and sub-megalithic graves, as well as stone-cists, containing EN collared flasks and related EN pottery. The results show that megalithic chambers coexisted with non-megalithic graves at a ratio of 6 to 5 in the EN C/II. On theoretical grounds one would actually expect a larger number of graves, because they are more easily destroyed and less regularly observed than megalithic tombs. This is precisely the precondition one would expect for an autochthonous evolution leading from the common graves to megalithic tombs.
Further more, at the EN C/II (Ebbesen 1994a Phase 2/3) there
is virtually no difference between the function of megalithic and non- or
sub-megalithic architecture. All contained the same burial goods. To name just
one example, the urdolmen in rectangular long-mound from Bogö, Sjælland
contained a collared flask which is virtually identical with one from Triangle,
Gem. Berend, Kr.Schleswig, Schleswig-Holstein (Knöll 1976 No. 137, Plate 4, 1).
It was found in a probable "chamberless" long-mound, illustrating the
connection between megalithic and "non-megalithic" architecture. Thus
one must concur with T. Madsen (1979:315-317) who analyzed the relationship
of non-megalithic structures in Danish long-mounds (Fig. 6.3) and
concluded that “there is no sudden and
total shift from earth graves to megalithic graves.” To the contrary, there
is ample evidence for continuity right into the MN. The Troelstrup type graves,
together with Rustrup (Fig. 6._, 15) of Lindebjerg variant and the
Skibshøj (Fig. 6._, 9) may even be precursors of MN mortuary houses of
Vrue Hede. In fact, at Troelstrup (Fig. 6._, 4) the mortuary structure
may have been built after the megalithic chamber. The successive construction
of graves and megalithic chambers is well demonstrated (Fig. 6._, 2, 4, 9,
20, 21, 29). Both types of burial architecture provided possible access, often
incorporated large stones and wood, and exhibit evidence of fire. Large
quantities of ceramics were found at the façades of EN long-mounds and by the
entrance of MN megalithic chambers. Thus the same rituals were used and the
structures consequently had the same function in society.
In light of the evidence, megalithic tombs are an autochthonous
development whose origins are rooted in the graves of the cultures which
preceded the TRB, but precisely how and why the architectural evolution
resulted in megalithic chambers still needs to be addressed.
Figure 6.14. Figure 6.4 Wood and fieldstone burial chamber, Konens Høj, radiocarbon dated at 2900±100 b.c. or 3645±116 B.C. (Ashbee 1970 Fig. 52 after Stürup 1966).
6.4
Causes
for the advent of megalithic tombs
So,
why did megalithic architecture begin to supersede graves with wooden
superstructures? The answer is that originally the use of megaliths was merely
a matter of expediency. It seems likely that urdolmen became popular due to
socio-economic considerations by the population living in Southern Sweden,
parts of Denmark, Schleswig-Holstein, Mecklenburg, and the adjacent Elbe
region, especially where land is limited by an overabundance of water. In this
region Neolithic agricultural endeavors, grazing practices together with the
building of palisaded villages, wagons with solid wheels, canoes, probable
corduroy roads, etc. gradually exhausted the large trees in the community. Thus
large trees must have become an "expensive" resource in those areas.
In fact T. Madsen observed that EN long-mounds normally used trees up to
one meter in diameter, while the later mortuary houses and palisades of causewayed
enclosures were only 20-40 cm thick (T. Madsen 1991:29-30). He attributes the decline to woodland
management. Among the long-mounds is Rude, where the oldest wooden façade was
dated at 2960±90 b.c./3711±99 B.C., close to the earliest proposed landnam phase dated at
3200 b.c./3800 B.C. (S. Andersen 1988:89 with references).
Indeed, pollen analysis of the mound fill covering the passage-grave of
Klekkenhøj, Møn, implies that trees may have been scarce in the immediate
surroundings of megalithic tombs (S. Andersen 1988:88).
Figure 6.15
The
Since stone-cists, such as those of Rude and Barkær, and rudimentary masonry chambers were in use anyway, and having plenty of erratic blocks at hand, the urdolmen architecture may actually have been more cost effective than wooden chambers, suggesting that the construction of urdolmen could have taken less time, energy and resources than the building of, let us say, a Konens Høj-type timber structure.
The popularity of the urdolmen and subsequent megalithic chambers may, therefore, have been the result of an economic choice when the demand for suitable large timber accelerated, rather than a purely cultural or religious phenomenon. It just so happened that the raw material had an additional benefit. It was durable.
6.5
Chronological
model of mortuary structures leading to the first megalithic chambers
The
precise appearance of the earliest megalithic chambers is not known and
opinions vary (cf. Bakker 1992:68; Hoika 1990b; Nielsen 1984; Persson an
Sjögren 1994 and in press). However, it is generally agreed that non- and
sub-megalithic greves below ground, in tumuli and long-mounds precede
megalithic tombs. Bägerfeldt (1993:63,73) lists the beginning of the so-called
earthen long-barrows at 3900 B.C. As discussed earlier, Midgley (1985) proposed
that they were derived from the Kujavian mounds, but the early Polish dates are
suspect and those tombs are not near the center of the tomb distribution.
Similarly, Midgley (1992) lists the averaged date of 3961±97 B.C. for the
Barkær long-mounds (cf. Liversage 1992:102, T. Madsen
1979:314 Fig. 5a, Midgley 1992:497, Skaarup 1982:44 Fig. 5). The
shell dates merely date occupation activity that predates the tomb construction
by an unknown number of years. The mounds contained Konens Høj type chambers
and a parallel stone-cist. These (closed ?) wooden chambers coexist at least
partially with the first megalithic chambers built in the EN II/C Fuchsberg
phase, which according to P. O. Nielsen (1993:85) ranges from
3500-3300 B.C. Skaarup (1991: 75) and most other Danish archaeologists
date the start of megalithic chambers to ca. 3400 B.C. However, Bägerfeldt
(1993: 66, 73) dates the first dolmen to ca. 3600 B.C. Thus interpretations on
EN tomb dating vary depending on which C14 dates are emphasized in
relation to the pottery. Since the most
reliable dates come from
Lichardus (1976), like Skaarup (1975:205),
suggested that Becker’s type A pots were later than Type B in
Table 2
|
Ebbesen’s 1994 Early
Neolithic chronology and grave classification |
||||||||||||||
|
3300 B.C. |
REGION |
|
SOUTHWEST
BALTIC AREA |
SJÆLLAND EAST GROUP |
SKÅNE / BORNHOLM |
GRAVE TYPE |
||||||||
|
|
Phase 3 |
STYLE |
LOCK HEDE |
FUCHSBERG |
VIRUM |
VIRUM |
a |
b |
c |
d |
e |
f |
g |
h |
|
|
|
Group |
|
|
|
Bellevuegård |
X |
X |
X |
X |
X |
|
X |
X |
|
|
Phase 2 |
STYLE |
VOLLING |
HAVENELEV/SVALEKLINT
|
|
SVENSTRUP/SIRETORP |
a |
b |
c |
d |
e |
f |
g |
h |
|
|
|
Group |
|
Siggeneben
Süd |
|
|
X |
X |
X |
X |
X |
|
X |
|
|
|
Phase 1 |
STYLE |
|
OXIE STYLE |
OXIE STYLE |
OXIE STYLE |
a |
b |
c |
d |
e |
f |
g |
h |
|
|
|
Group |
|
(Rosenhof) |
( |
Oxie group |
X |
|
|
|
|
|
|
|
|
4000 B.C. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Since the Oxie style is viewed as the earliest TRB manifestation in Southern Scandinavia one would expect to first Scandinavian long-mound to occur there. Indeed, the oldest apparently “unchamberd” long-mound is, as expected, reported from the Oxie group area at Kyrkoköpinge, near Trelleborg, Skåne (e.g. Tilley 1996:84-86). Locally known as Jättegraven, Person and Sjögren (1996) date list the date from its wooden façade as 3270±70 b.c. / 4150 (0.86) 3980 B.C. (Fig. 6.__). However, P. O. Nielsen (1993:85) starts the Oxie group in Denmark at 3900 B.C. On the other hand, the earliest C14 date from the Rosenhof phase, which Ebbesen includes under Oxie, is 4360BC (0.78) 4220 B.C. The last is 4150BC (0.82) 3950 B.C. (cf. Midgley 1992:496 No. 24-30). Unfortunately, the Oxie group’s Danish dates are more difficult to assess due to their wide standard deviations and large spread. The oldest date in the Muldbjerg I series is (K-129: 2890±160 b.c. / 3900BC (1.00) 3350 B.C.; Liversage 1992:104 Fig. 84, 14-20; Midgley 1992:497), but the vast majority fall after 3800. Most importantly, the Drugsholm grave, which is supposedly the earliest Type a grave in Ebbesen’s sequence, dates around 3800/3700 B.C. (Fig. 6.__ and 6.__***). So Jättegraven’s date seems perhaps just a little early. Additional dates from the mound would be helpful.
Jättegraven’s nearest Danish neighbor is the earliest Lindebjerg three long-mounds (sb 43) near Kalundborg, Sjælland (Liversage 1981 Plate I). Its earliest pottery belonging to the supposedly later Svaleklint group (Fig. 6.___ ; 6. __, 29).
Near its eastern façade four Type B potts were found (cf. T. Madsen 1979:308). This tomb is dated 3980 (1.00) 3700 B.C. (K-1659b) / 3960 (1.00) 3690 B.C. (K-1659a). The Lindebjerg finds are similar to those from Langeland’s Stengade (“House”) II (Liversage 1981:148-149, Skaarup 1975:187). The latter is thought to be a possible wood enclosed (megadendric”) burial structure without evidence of a mound Liversage 1981:149). Like Lindebjerg, it is the earliest of three nearby “long-mounds,” having two TL dates calibrated as µ3820 and µ4124 B.C. (Skaarup 1975:193 Table 29-30). Ebbesen (1994:77) relates the ceramics of Stengade II to similar finds from Siggeneben Süd (Meurers-Balke 1983), which has a series of dates ranging from 3330±55 b.c. / 4400 (0.97) 4040 (KN-2425) to 2960±60 b.c. / 3820 (0.88) 3620 B. C. (KN-2405), if one excludes the oldest anomalous date (KN-2273). Also related in time is the oldest monument of the Volling Group, Mosegården/Toftum, Jylland, dated 3130±90 b.c. / 3990 (1.00) 3780 B.C. and 2940±90 b.c. / 3790 (0.85) 3620 B.C. (6. __, 20). This 60:16 m mound was surrounded by a post bedding trench, containing split trunks. The trunks had their flat sides set against the barrow (e.g. T. Madsen 1979). All these tombs would appear to belong to Ebbesen’s Phase 2.
Figure 6.16
Oxie group C14 date from the Dragsholm grave, Sjælland, and the
“unchamberd” long-mound Jättegraven, Kyrkoköpinge, Skåne (Midgley 1992:497 No.
65, Persson and Sjögren 1996).
Figure 6.17 Figure 6.5. Long-mound of Lindebjerg, Sjælland dated µ3060±100 b.c. (Midgley 1985 after Liversage 1981). The mound appears to be a conglomerate multiple-phase structure that evolved from the mound-like cover of several different graves. A megalithic enclosure surrounded the later mound and seems to have been constructed in association with a dolmen.
If
one looks beyond Scandinavia, one finds that the single date of 3220±180b.c. from
the Kujavian long-mound group at Wietrzychowice is only minutely younger than
that from Mosegården/Toftum. But its huge standard deviation merely implies
that it could have been built sometime between ca. 4160 (0.92) 3780 B.C. The
tighter date (3140±45 b.c.) from one of the two parallel long-mounds of Beřzno,
Bohemia, would suggest a building period of
3980 (0.44) 3910 B.C. But this date is the oldest in a series
of Baalberge dates and the second mound
yielded an inappropriately younger date. Most likely the tomb was built at the
very end of its more probable range of
3880 (0,56) 3810 B.C. Therefore, one may conclude that the majority of
the earliest datable long-mounds started around 3800 B.C. give or take 50
years (Fig. 6.___1-2). One may further assume that the Rude façade and its
associated B pottery came into being around 3700 B.C., the original
construction being perhaps somwhat earlier and the rebuilding occurring no
later than 3600 B.C. Rude thus may signify the end of the longmound
construction associated with B pottery, i.e. Ebbeson’s (1994) Phase 2.
Theoretically, its stone cists could, therefore, mark a transition to the next
period.
The next and more intensive (?) phase of mound
building has to be associated with Type C pots of the FN II Fuchsberg and
Virum group, which P. O. Nielsen (1993:85) dates to
3500-3300 B.C. This scheme appears to be due to a large number of late
dates from two settlements: “1) Sarup I (five dates), mean value uncal.
2700 bc; cal. 3490-3370 BC. 2) Toftum (eleven dates), mean value uncal. 2650
bc; cal. 3360 BC” (P. O. Nielsen personal communication April
1996). However, in my opinion the date from only two settlements may be sqewd
by the founding date of the village, which will not necessarily coinside with
the earliest appearance of a pottery style. Besides, as Deetz and Dethlefsen
(1966) have shown, Styles overlap each other. And P O. Nielsen
(ibid.) adds that “the transition between
EN I and EN II … should be placed at 3600/3500 BC”. Therefore, the three
dates from Rustrup, Graves 1 and 2, which appear to be earlier than Rude, are
not necessarily out of place (Fig. 6._. 2). One may, however, expect the dates
fall between ca. 3650-3500 B.C. This is strengthened by the dates from Konens Høj, Storgård and Bygholm Nørremark.
The long-mounds of this period often exhibit
transverse oriented hurdle fences as is the case with the double mounds of
Østergård and Barkær, as well as Bygholm Nørremark and Rustrup. They
usually also have a timber façade as described by T. Madsen (1979) and the
earliest graves appear to be small, axially oriented graves. It is not
inconceivable that the long-house at Halle Dölauer Heide was also a burial
mound with façade, transverse hurdle fence, associated burials (Behrens
1973:197 Fig. 78). One should probably
also add Niedüwiedü “House” 2 and Stengade “House” I
to this group. These tombs may contain
burial pits, pavements, stone frames and Konens Høj type structures. The
graves undoubtedly coexisted at least partly with primeval dolmen, but the end
of their construction his highly uncertain. The dating of the
extended dolmen-like cist (Grave 1) in the southern long-mound of
Barkær appears to be related to the adjacent Konens Høj type Grave 2
(Liversage 1992:23-25, Fig. 11-12; T. Madsen 1979:314 Fig. 5a) The chamber
could, by virtue of its central location within the mound, be rather early
according to Liversage. But perhaps the extensive use of wood in the
construction of both mounds exhausted the readyly available supply of large
timbers, leading to the use of stone slabs for the chamber and the construction
of a megalithic enclosure, as evidenced by the remaining corner stones
(Liversage 1992 Plate 1). At Flintbek LA 3 they were certainly
followed by parallel extended dolmen.
In Schleswig-Holstein the Bordesholm-Tannenbergkoppel
Konens Høj type grave (Fig. 6.2) seems to have provided no evidence of a
mound. Those in the nearby Flintbek LA3 long-mound were constructed each in
their own nondescript little tumululus that gradually evolved into a
long-mound. These graves precede the small 2+2 side-stone, axially oriented,
open dolmen in the same mound.
The axially oriented graves are frequently followed
by a later stage of mound construction, which contains larger (often)
Troelstrup (Ebbesen type h) graves, oriented parallel to long-mounds. This is
best illustrated at Storgård Fig. 6._), Flintbek LA 3 (Fig. 17.21)
and Rustrup, but probably also at Østergård (Madsen 1979). The added graves
appear to occur towards the outer end of rectangular mounds, such as
Flintbek LA 3, and at the opposite or distal end of rectangular to
trapezoidal mounds with façades, such as Rustrup, Storgård, and perhaps Niedüwiedü
“House” 2. Among the best preserved Troelstrup graves is Skibshøj (Fig. 6.__).
The Hejring dates would indicate a beginning of ca. 3500 B.C. and an end
at around 3300 B.C. They undoubtedly coexisted with extended- and possibly
the smaller grand-dolmen.
The
observation that there is a chronological succession from axially oriented
Konens Høj type graves which appear to lack an entrance, to somewhat
larger, laterally oriented Troelstrup type wooden chambers with obvious
entrance is highly significant, because the same sequence has traditionally
been stipulated for the evolution of the earliest dolmen (Chapter 7-8). The
Konens-Høj type and similar graves with stone frames, such as Forum, Brøndum
s., and Sallinge graves A-C (Ebbesen 1994:52, 65 Fig. 6, 21-23) are, therefore,
not only the precursors, but also the architectural equivalent to the earliest
dolmen, which also have no entrance and are oriented parallel to the long-mound
axis. The graves coexisted at least partly with these dolmen in time, but not
necessarily in space.
The graves, especially the Konens Høj type occur
in Jylland and adjacent Schleswig-Holstein, whre relatively few dolmen are
known. The majority of dolmen occur in and around the Danish Islands, where
extremely few graves are known and no Konens Høj or Troelstrup type
structures have thus far been found. The change in orientation and an apparent
increase in grave size is also mirrored in the change from primeval dolmen to
extended dolmen. One may therefore hypothesize that the earliest dolmen occur
around 3600 B.C. give or take 50 years and the change to axially oriented,
slightly larger dolmen could have occurred around 3500 B.C.
Figure 6.18. Calibrated C14 dates of long-mounds originally containing only non- and sub-megalithic chambers.
Thus events leading to the practice of monumental tomb architecture must have been in motion by ca. 3900 B.C. Assuming that the TRB in the North Group started during the Rosenhof phase (max. 4274±90 B.C., µ4182.49 B.C.), it took perhaps 300 years to develop a concept of monumental mounds. It took perhaps another 200 or so years for megalithic chambers to become an established architectural form. But the inception of megalithic chambers raises a fundamental question: Why switch from timber to megaliths?
Figure 6.19. Skipshøj long-mound (top) with floor, burial and wooden roof layers (bottom) of the large eastern Troelstrup type chamber. The chambers southwest entrance shows evidence of a wooden post on each side. (T. Madsen 1979a:313 Fig. 4d, 1993:97 unnumbered Fig.)
Figure 6.20. Long-mound with burial pits and lateral construction ditches of Storegård, Denmark.
6.6
Conclusion
TRB megalithic tombs, as the most spectacular burial architecture, have captivated researchers for centuries, but to fathom their origin it is necessary to understand their relationship to the seemingly unspectacular sub- and non-megalithic graves. These burial pits ("earth graves"), stone- and wood-lined chambers and associated burial practices are inherited from earlier times. These forms continued throughout the TRB and outlasted it. Evidence of continuity and discontinuity is often evident in the TRB burial grounds, which often exhibit more or less continued use right into the Bronze Age.
Thus events leading from basic graves to the practice of monumental tomb architecture in form of earthen long-barrows (actually long-mounds with wooden chambers and graves) must have been in motion by ca. 3900 B.C. Assuming that the TRB in the North Group started during the Rosenhof phase (max. 4274±90 B.C.), it took perhaps 300 years to develop a concept of monumental mounds. It took perhaps another 200 or so years for megalithic chambers to become an established architectural form. But the inception of megalithic chambers raises a fundamental question: Why switch from timber to megaliths?
6.7
Related
Links
Central
and North European Neolithic Chronology with summaries of
individual cultures
Neolithic/Copper
Age Link Index: Links to News Bulletins, Articles, Site Reports, Databases, etc.
about the Neolithic/Copper Age in
Please send comments or
questions to Max Baldia.