Added September 21, 1999. Updated April 25, 2006, 14:18 hours.

The dissertation has a copyright. Please give proper credit when quoting its content.

A SPATIAL ANALYSIS OF MEGALITHIC TOMBS

CONCLUSION

By

Maximilian O. Baldia

1993-April 25, 2006©
All rights reserved

Megalithic tombs research has a long history, inextricably linked with the development of archaeology itself. In the beginning (2400 b.c.-1200 A.D.) interest in megalithic tombs cannot be described as research, for such interest consisted primarily of superstitious awe and religious inspiration. Beyond this, the curiosity of unknown prehistoric individuals can only be imagined.

The earliest speculations and excavations regarding tomb function occurred from 1200-1600, but widespread speculation about giants, devil worship, and sacrifices rendered Marschalk's pioneering approach and reasoned opinion inaudible. Speculation was only gradually replaced by his methods during the subsequent period. However, many terms describing megalithic architecture were first recorded.

In the Antiquarian Period (1600-1800) researchers dispensed of the predominant, speculative Giant Theory through better excavation methods. Antiquarians introduced osteology as a permanent part of tomb research. At the same time they improved descriptions of the tombs and their contents. They also developed a rudimentary chronology and introduced the first monument protection laws in the world. With these achievements, the stage was set for major breakthroughs during the following Classifying Period.

The primary achievements of the Classifying Period (1800-1900) were the extraordinarily vigorous cataloging, classifying, and monument protection efforts. New societies furthered the dissemination of information so fundamental to modern research. These publications were often as thorough as modern ones. Cataloging was one way to counter the rapid destruction of the tombs and preserve information for future analysis. It was realized that the origin and social function of the tombs could only be understood through a thorough knowledge of the monuments in all regions. Yet in spite of the pioneering efforts by researchers, such as Krause and Schoetensack, no nationwide catalog was prepared, either in Germany or elsewhere. Even Lukis and Dryden's Drenthe survey remained largely unpublished. Data collection was not standardized, causing the kind and quality of data to vary widely. The lack of standardization, which frustrated and hampered researchers, such as Sello, remained a serious deficiency. Petersen, Montelius, S. Müller and others attempted to classify the tomb architecture, but the typology remained relatively primitive and repeatedly illogical. Even in Scandinavia agreement on rigorous standards was not achieved. Meanwhile, the wanton destruction of megalithic tombs continued, so that only between 5% to 20% of the known tombs can be located in the 20^th Century. Even the archaeologists themselves were still learning by doing and thus contributed to the destruction of the tombs through primitive excavation methods and overzealous preservation efforts.

On the other hand, the general chronological sequences were identified by tool complexes, particularly stone axes and to some extent ceramics. Many of the better data collection approaches, including mapping, illustrations, sampling strategies through test pits, etc. were hardly surpassed in recent times. In fact, with the exception of ceramic analysis and a more detailed chronology, the tomb researchers of the Classifying Period left little new to be discovered by their successors.

During the Culture History Period (1900-1950) regional differences in the TRB were isolated based on ceramic styles instead of tombs. The classification efforts of the previous period bore fruit as ceramic typologies were refined for specific regions and became the basis for a better understanding of chronological sequences. Isolation of regional differences, combined with a reasonably accurate relative chronology based on settlement rather than tomb excavations, fostered renewed interest in the origin and cultural history of the TRB. The origin of the tombs remained a point of controversy. Structures, other than megalithic tombs, were explored and evidence of a larger ceremonial complex continued to come to light. During the Culture History and subsequent periods the "secondary burials" of the KAK, Single Grave-Corded Ware and Bronze Age cultures found in TRB monuments, such as the Pohlsberg, were viewed as unrelated, intrusive burials of different people with unspecified foreign origins.

The Data Reevaluation Period (1950's-70's) saw refinement of the Scandinavian relative chronology and its application to other regions. Some felt that the chronology only needed minor adjustments and that new tomb excavations and large catalogs would explain the evolution of the tombs. Surprisingly the advent of C¹⁴ dating in the fifties, which revolutionized archaeology in many parts of the world, had little effect on dating tombs, in spite of the seminal article by Bakker, Vogel and Wis'lan'ski (1969). C¹⁴ dates began to hint at the real antiquity of the tombs, but ceramic analysis continued to play the major role in TRB dating. New interregional tomb literature included generalized catalogs with brief descriptions occasionally combined with illustrations. Many were regionally specific works.¹ Among the most ambitious and detailed publications were those of Mecklenburg and Central Germany together with Sprockhoff's catalogs. The resulting data permitted the isolation of regional differences in tomb architecture.

Reexamination of previous results through careful excavation led to the conclusion that some of the earliest mounds contained non- and sub-megalithic chambers as well as graves. T. Madsen observed that such structures often coexisted with megalithic chambers. The neatly standardized nordic ceramic typology ultimately proved inadequate for refined interregional diachronic analysis and new regional styles were isolated. In the process it was suggested that some dolmen were constructed after MN Ia while some passage-graves contained pottery that was earlier than MN Ib.

During the Analytical Period (1970's to present) statistical evaluation showed that passage-graves became successively larger, however, there was no simple, one to one relationship between chamber size and the relative chronology. Difficulties in various regional ceramic typologies were isolated and efforts have been made to overcome incongruencies between the typological method used for ceramics, problems in C¹⁴ dating, and theories on megalithic tomb evolution.

The origin of TRB ceramics now appears to be the result of a broad interregional evolution to which the Ertebølle, Late Rössen (Bischheim) and related pottery contributed. A certain kinship with Michelsberg ceramics is obvious (e.g. Lüning 1976, Zápotocký 1992, Zich 1994). The origin of megalithic tombs and German gallery-graves has been a matter of debate for centuries. The Eastern Tomb Origin Theory is currently not persuasive. The western origin is tainted by the diffusionists' mind-set. It is contradicted by radiocarbon dates and the evolutionary trajectory of French tombs. The Autochthonous Theory is supported by a logical, chronological evolution of tomb types and their spatial distribution leads to the following model:

The origin of TRB monumental burial architecture, including earth graves (Danish: jordgraver), wooden chambers, timber facades with long-barrows, and megalithic tombs is an autochthonous development probably related to an expansive communication and trade network. This network was held together by a common ideology expressed through the building of monumental burial architecture which, as a Europe wide phenomenon, lasted from at least 3600 to about 1600 b.c. (C¹⁴ interaquartile ranges of J. Müller 1987, Fig. 1).

The TRB's megalithic chamber building custom may have started on the Danish islands and on or near the adjacent Western Baltic coast, barely crossing the Elbe, while non- and sub-megalithic tomb construction continued over a broad, diffuse area including Denmark, Poland and the Check Republic. Megalithic tomb construction expanded at different rates into Norway, Netherland, Central Germany and parts of Poland. Even the megalithic mound enclosures may have started in the core area. One must envision a stochastic geographical expansion along hierarchical, dendritic lines of communication, where areas that were strategic nodes of the socio-economic communication network would build monumental architecture at approximately the same time. The local and regional tomb types were derived from local traditions, local religious interpretations and customs tied to an interregional ideology. The tombs thus served both as an integrative mechanism for the larger socio-economic system and the need for regional and local autonomy. This explains the differences in the types and numbers of tombs found in a given locale.

Tombs architecture, and especially megalithic chambers evolve in a fashion best described by Chaos Theory, which "says two things. First, that complex systems ... have an underlying order. Second, the reverse ..." - that simple systems can produce complex behavior (Crichton 1991:75). Thus the discernible variety of tombs, ranging from non- to megalithic tombs and other structures are the result of a complex socio-economic system that started with simple earth-graves, which evolved into relatively elaborate wooden structures. These in turn were gradually replaced by stone structures during the EN C in areas where suitably large timbers became a scarce resource. Thus simple urdolmen gradually evolved into complex passage-graves.

However, even the simple urdolmen may already have been part of complex mortuary practices. The perceived dichotomy in urdolmen form is often seen as evidence that long, narrow urdolmen were designed to contain only one supposedly extended skeleton, while the short, but wide urdolmen are viewed as made to contain single, flexed adult or even child burials, although supporters are often rather vague.² This view is still held,³ but must be reexamined for several reasons. Among them is the fact that at least one small, closed urdolmen showed evidence of multiple, largely disarticulated interments during the EN, while at least one large dolmen contains a single extended skeleton that may or may not belong to the EN. It is, therefore, uncertain whether or not the earliest chambers were designed for a single extended inhumation. Indeed, many chambers are too short (cf. Bakker 1992:96). Thus one or more articulated flexed interments are possible, as are successive depositions of several at least partly disarticulated interments. Regardless of the burial position, the invention of top- and side-entrances suggests they they are the result of multiple interments, although repeated ritual entries after the original interment could have resulted in the same innovation.

In time chamber size increased and multiple side- and capstones came into use. The northern polygonal-dolmen evolved from square-dolmen on Sweden's west coast. A second area of development is in the Danish islands, Jutland and parts of Schleswig-Holstein. Extended-dolmen have a somewhat more southerly distribution that expands beyond the original location of the urdolmen.

Polygonal-dolmen become larger through time and take on the appearance of passage-graves in the north. The larger rectilinear dolmen, which overlap with the polygonal-dolmen, switch from a front-entrance to a corner- and finally a side-entrance, turning into passage-graves. This passage-grave architecture spreads as far west as Netherland and as far east as Pomerania and Kujavia (and beyond?).⁴

The majority of grand-dolmen are located in Mecklenburg. There appear to be no grand-dolmen east of the Oder. However, grand-dolmen are found as far west as the Ems and as far south as the Altmark. They are also found in Denmark, but do not seem to be common in Sweden, although one such small tomb exists on Gotland. The western spread of grand-dolmen must have occurred just before passage-graves became popular, and may have briefly coexisted with them.

The gallery-graves also could have non- and sub-megalithic prototypes, but the evidence is slender, implying that they are related to the grand-dolmen found west of the Elbe in the Altmark and the contact zones along Aller, Weser and Ems. The succession of grand-dolmen by passage-graves in the European Plain appears to have influenced the institution of the side-entrance and the short passage in some, but not all gallery-graves of the adjacent highlands. The immense size of some gallery-graves is attributable to the availability of lighter raw material than glacial erratics. Many of the largest chambers are concentrated near the source of the Ems and Lippe. This implies close interaction between gallery-graves and some of the largest passage-graves along and between the Lippe, Ems, Hase and Hunte river valleys.⁵

The occasional evidence of enclosures found with gallery-graves may also be the result of interaction with the zone of large passage-graves that are often encompassed by Dutch-type enclosures. It can also not be ruled out that gallery-graves form a tenuous (late?) link between the megalithic tombs west of the Elbe and the rare gallery-graves/cists west of the Rhein.

On the whole, the mound and chamber distribution suggests that the number and sometimes the size of tombs peaks in strategic areas along lines of communication. Two primary lines are observable. One is oriented east-west and the other north-south. The density peaks may indicate the vicinity of powerful chiefdoms which by virtue of their location could control trade and communication. Furthermore, the tomb distribution resolves itself into regional groups. These groups are often the result of topographic barriers, which in turn must have led to sub-cultural units and linguistic differences, possibly related to Indo-European.

TRB culture change, especially the evolution of megalithic tombs, seems to have occurred along these major lines of communication, a situation unrecognized by pioneers in spatial analysis, who assumed that the tombs represent central places. Their assumption is challenged by observations that some TRB regions built only non-megalithic burial graves with pottery that is similar to that found in megalithic tombs. For instance a county by county survey of megalithic tomb density in eastern Schleswig-Holstein demonstrates that many well documented TRB communities have constructed no megalithic tombs at all, even though they are close to counties where communities erected many such tombs (Hoika 1986:194 and Fig. 8). In contrast to central places, analysis of the relationship of tombs to soil types, villages and topography are a more promising endeavor in determining factors that influenced the tomb location. However, all of these factors are themselves influenced by what appears to be a close relationship between tombs and prehistoric roads.

There is ample evidence that wheels and wagons were used in the TRB by the end of the EN. In addition, oxen were probably used as draught animals, while some even maintain that horses were ridden. The use of wheeled vehicles necessitates at least minimal building and maintenance of roads. Streets with wooden pavements are documented for a Schussenried village, which is coeval with the early TRB, and wooden tracks used for wagons from the Dümmer area date to 3500 B.C. Such Neolithic tracks are found throughout the TRB culture area and may indicate an organized effort to build and maintain roads. However, most tracks are attributed to the Corded Ware/EGK.

The location and orientation of mounds and chambers is best explained in relation to this prehistoric communication system. The long-mounds follow the orientation of the roads and even the tumuli are aligned along the roads. Except for the earliest urdolmen, chambers front the roads with their entrances and passages facing the road side. Yet within these constraints the entrances and passages appear to show astronomical alignments, although not all passages provide a view of the horizon. Mounds and chambers are placed frequently in such a way that they refrain from obscuring probable sight lines. It may well be that a combination of technical and religious factors combined to create the numerous angled passages that, in part, gave rise to the passage-grave. The orientations of the chambers and mounds themselves do not reflect the same strong orientation patterning.

The tombs are located along these prehistoric roads, almost as if they were road markers.⁶ They concentrate in strategic areas, most noticeably near fords and harbors. They also cluster around complex intersections. There seems to have been competition for the location and use of roads, river crossings and harbors, so that parallel road systems develop. The probability of competition is highlighted by the existence of at least one unfinished chamber and one partly razed mound enclosure.

Occasionally chronological differences are ascertainable. Some roads, or sections of roads, are lined exclusively by urdolmen and extended-dolmen. In specific cases the oldest tombs are closest to the most strategic portion of the road, such as a ford. The later tombs are further away. Sometimes the oldest chambers form the core of a tomb distribution and the later tombs are built at the edges.

In a few instances grand-dolmen occur right next to early passage-graves. In such cases the entrance and passage orientation is virtually identical for both chamber types, while the chamber orientations are more or less at right angles to each other. This probably implies a sequential chronological evolution. It also appears that the longest mounds were often built first, while shorter long-mounds with larger, more complex chambers were built later. One can thus occasionally observe nearly the whole evolution from early dolmen to passage-grave along the same sections of road.

However, in at least one instance a road lined by passage-graves takes a different direction from those lined by early dolmen. This may suggest an interregional chronological shift in interaction and communication that could be due to larger social or political events. Such events may have influenced trade in especially valuable items, such as copper, salt and amber.

On an interregional level, all tomb types, including the eastern "unchambered" long-mounds, long-mounds without megalithic enclosure and even gallery-graves form part of a huge road and communication network. The network crosses mountain passes and connects TRB related cultures, such as the Altheim and Cham Group with the TRB South Group, where megalithic tombs were not fashionable. The gallery-graves may have linked the network with other cultural areas such as the Belgian gallery-grave region and perhaps even the Paris Basin. It is not inconceivable that some of these connections developed towards the end of the MN and LN, when the KAK and the far flung Corded Ware culture predominate in the archaeological record. It is possible that the proliferation of gallery-grave-like chambers with porthole in Sweden is the result of a similar process. The same may hold true for EGK/Corded Ware cists in Denmark and elsewhere.

From a sacred perspective, megalithic tombs lined roads, because they also functioned as shrines. As such, they may be connected to pilgrimages and perhaps even seasonal rituals related to ancestor worship. This is evident in the votive layers found in front of the passages of dolmen and passage-graves from Scandinavia to Netherland. The shrines no doubt tied together diverse socio-economic groups, which lived in major population centers located in different niches of the woodland environment. The existence of a hinterland is supported by the absence of tombs adjacent to areas with large numbers of megalithic tombs. This together with the location of tombs vis-à-vis fortified central villages, implies that the tombs may have marked roads, but they are not centers or territorial markers. As public works, they functioned as an integrative mechanism, which must have fostered at least minimal maintenance of roads, fords, harbors and perhaps even canals.

Figure 19.1 Early Bronze Age Long-mound 9, Fischbeker Heide, Hamburg with megalithic enclosure (after Tromnau 1977:102 Fig. 2).

When megalithic tomb construction ended is difficult to say. Their popularity certainly dwindled before the demise of the TRB. However, subsequent cultures often built their own megalithic structures. For example, in Sweden gallery-graves were built during the LN and Bronze Age. Hyenstrand (1989:12) observed that in the central megalithic areas ... a special relationship with passage-graves may ... represent a continued communal, megalithic tradition. Indeed, one such gallery-grave was built as part of the passage-grave alignment at Falbygden. Made from the same megalithic blocks as the passage-grave of Rössberga, the box shaped gallery-grave is almost directly across the likely Neolithic road form the delicately constructed funnel-shaped façade leading into the Rössberga passage. The modern paved road follows the same direction a short distance behind the passage-graves. Similarly, some of the LN chambers in Central Germany are identified as sub-megalithic cists (cf. Beier 1991a), although some use stones of megalithic dimensions, suggesting traditions that occasionally reach into the Bronze Age. In fact, the Bronze Age people in northern Germany erected what may be the last long-mound with megalithic enclosure (Fig. 19.1). In Denmark several Bronze Age long-mound-like structures exist. Among them is a long-mound, surrounded by a sub-megalithic stone curb. It covers a central Neolithic or Bronze Age dolmen/stone cist and other burial structures at Lykkesholm, Lyngby s., Randers amt. The multiple phase mound at Græse, Frederiksborg amt, originated with an ancient urdolmen in long-mound and in time became a complex mound. In Mecklenburg-Vorpommern the long-mound of Wartin was also rebuilt in the Bronze Age.

Even after the Bronze Age the TRB mounds and chambers continued to play a part in the religious and social organization of northern Europe. The earliest Celtic, Anglo-Saxon and Icelandic literature demonstrates this continuity.

---

FOOTNOTES

¹ Alreus 1966, Chmielewski 1952, Fischer 1956, Hingst 1959, Hinz 1954, Kersten 1951, 1958, Röschmann 1963, Strömberg 1971, etc.

² e.g. Aner 1963, Bakker 1992:95-96 with references, Kælas 1966, Knöll 1976 with references.

³ H. Nelson 1988: 96 states that the majority of (ur)dolmen are large enough to contain an extended burial. Häusler 1994:47 writes: ... the older, smaller dolmen served the deposition of single burials ....

⁴ In Kujavia KAK artifacts are found in the passage-graves east of the Weser. In Mecklenburg-Vorpommern such artifacts have been shown to be secondary interments.

⁵ The Lippe leads to the Rhein, past the megalopolis. The Ems, Hase and Hunte cross the TRB territory and could have been navigated (Bakker 1976:84).

⁶ Near Häggum in the Falbygden area of Sweden, megalithic tombs and later burial mounds and graves seem to have been associated with a road lined by menhirs (Olsbo 1986), reminiscent to the avenues found in Great Britain.

Please send comments or questions to Max Baldia.