This page was
last updated on December 1, 2000
[1] This work is dedicated to Torsten Madsen, who introduced me
to the Danish TRB monuments and shared his data freely. Since the typology
presented here is mainly based on the German classification, where there is no
plural ending for dolmen, this practice is continued throughout the text. This
also accentuates the unique meaning of dolmen as used in the TRB culture area
vis-à-vis other regions, where the same term refers to a different
architecture. Note that the number following German tombs are those of
Sprockhoff 1966, 1967, 1975; those of Mecklenburg-Vorpommern are from Schuldt
1972 Table A and/or B, those from Sweden are from Bägerfeldt and Kihlstedt
1985, and those from Denmark are from A. P. Madsen 1868, 1896, 1900 and/or the
Sb. number from various sources, including Hansen 1993.
[2] All calibrated dates are based on equivalent dates from
Midgley 1992, unless otherwise indicated.
[3] Netherland instead of Netherlands is used here in accordance
with Bakker 1979.
[4] Tombs with wooden superstructure include Konens Høj and
Barkær in Denmark (T. Madsen 1979), as well as Bordesholm-Tannenbergskoppel
(Hingst 1975:20-22, Fig. 3 and 4) and Flintbek LA3, Schleswig-Holstein (Zich
1992, 1994a, b, 1995).
[5] However, the skeleton within the cist post dates the TRB.
[6] Bogø By's two dolmen have different dimensions, attributable
to the available raw material or a time lag in construction. Dolmen B also
seems to have had two small capstones and the mound could have been built in
more than one phase.
[7] Hoika (1990b:86) lists GrN-10346:4710±45 B.P. in Footnote 39,
but 2750±45 b.c. instead of 2760±45 b.c. in the text. The date is
identical with Ølstykke's date K-2356: 4710±65 B.P. (2760±65 b.c. calibrated as
3480±131 B.C., e.g. Midgley 1992:497 No. 104).
[8] All dates in brackets are currently preferred Danish dates
according to P.O. Nielsen (personal communiction April 1996).
[9] The literature is not always succinct enough to permit a
rigorous classification, because it often fails to distinguish between
transitional multiside-stone dolmen with a combination of lengthwise and
upright side-stones and extended-dolmen. Difficulties also occur with
classifying polygonal-dolmen, which incorporate small side-stones or use
relatively large but thin capstones, depending on the expediently available
building material. Larger ones are difficult to distinguish from polygonal/oval
passage-graves. Further more, rectilinear chambers in long-mounds are generally
only roughly perpendicular to the mound's axis, while at least three dolmen in
Schleswig-Holstein (Kampen 2, Keitum 5, Utersum 13; Sprockhoff 1966), one odd
chamber on Møn (Stupperup Have, Magleby s., Præstø a., A. P. Madsen 1868:40 No.
63), and one grand-dolmen in Mecklenburg (Gaarzerhof B46, Schuldt 1972:166
Plate 4) are rare exceptions. Such exceptions have led to contradictory
conclusions (Hoika 1990b). But these chambers are eccentric by virtue of their
floor size, position in the long-mound or subterranean location, number of
entrances, the shape and size of the mound's enclosure etc. They do not follow
the norm.
[10] Knöll 1976 considers the chamber an open dolmen that also
contains MN artifacts. Similar artifacts in Germany are assigned to the
Single-grave culture (Laux personal communication May 1994).
[11] Hjortegårdene,
Horns h., Frederiksborg a., North Sjælland (A. P. Madsen 1868:4 No. 9, Kaul
1987:23). The human remains include two crania, four femurs, at least
three tibias, other long-bones, six innominate bones, ribs (some seem partly articulated),
and one (left?) shoulder bone.
[12] Alternatively, several bodies may have been entombed at the
same time, but in a partly disarticulated state, as was the case with the
dismembered inhumations below the dolmen at Fakkemose 266, Langeland (Skaarup
1985:205-210, Fig. 222; 1990:76, Fig. 6; 1994:350-351, Fig. 3). Disarticulation
has also been attributed to animal activity (Gruppe 1984).
[13] Closed parallel urdolmen = 14, probably closed parallel =
3, probably open parallel = 1, open parallel = 2, open perpendicular = 3,
closed perpendicular = 0.
[14] A similar rotation seems to have occurred among
non-megalithic graves as indicated by Flintbek's long-mound LA 3 Grave G (not
definitively attributable to the TRB) and Grave F (Phase 5). Both graves have a
virtually identical orientation as the first megalithic chamber
(Extended-dolmen I, Phase 5; Zich 1992, 1994a, b, 1995). However, the
perpendicular chamber orientation does not always guarantee that the chamber
was entered by a (non-megalithic?) passage from the side of the long-mound
(Hoika 1990b:57, 64).
[15] The date came to my attention after J. Hoika provided a
reprint of his 1994 publication at the 1995 Røsnæs, Kalundborg meeting.
Unfortunately, it is not known whether the chamber had a single axial capstone,
or two perpendicular ones (Bokelmann 1977:92 Fig. 1-2), so it could also be
classified as a slightly earlier transitional dolmen (3500 B.C./2670 b.c.).
[16] cf. Bägerfeldt 1993:78, 84 unnumbered figures for an
independently developed classification that results in a similar typology.
[17] Such polygonal grand-dolmen range from nearly 4 m2
to about 9 m2 in area and straddle the categories of
polygonal-dolmen, grand-dolmen and primeval passage-grave, implying a
continuous evolution into passage-graves.
[18] However, watertight construction is sacrificed with multiple
capstones, forcing additional precautions in form of shingle-like stone slab
roof coverings with clay linings and elaborate drainage systems demonstrated in
passage-graves (e.g. Hansen 1993).
[19] Hansen
1993:107 Fig. 76, p. 108 Fig. 79, p. 110 Fig. 90, p. 134 Fig. 192.
[20] The location suggests that the more easterly parts of
Mecklenburg-Vorpommern, housing three- and even four-yoke grand-dolmen, did not
participate in the original development of passage-graves. The sole
passage-grave on Rügen is a four-yoke Holstein-type chamber, deemed intrusive
by Rennebach (1986).
[21] Cited with kind permission by the authors, the collagen
dates, derived from the same flexed skeleton, are St-11267: 3055±235 b.c. and
Ua-1894: 2750±110 b.c. (Persson and Sjögren 96).
[22] Katbjerg, Onsild h., Maringer ls., Randers a. (A. P. Madsen
1896: 16 No. 28; Kjærum 1969/70). The earliest pottery is MN A Ib.
23] The earliest pottery discovered within the three other
two-capstone chambers on the Hondsrug could be roughly 100 years younger than
that from G3.
[24] The single-capstone passage-graves in order of floor area are
Valla 77/27, Güby 73, Slangerup Sb. 26, Stinsedt-Meckelstedt 1, Brastad 27,
Tegneby 64, Krike Hvalsø Sb. 5, Sieben Steinhäuser 180/D. Note that the
destroyed Meckelstedt 1 chamber has also been classified as a polygonal-dolmen
(cf. Bakker 1992, Laux 1970).
[25] Skärv 81, Falbygden (Rooseveltsson 1992:3 Fig. 2), has an
identical set of three normal capstones plus one undersized capstone as does Sieben
Steinhäuser 807/B, implying a common TRB tradition, but cf. Laux (1984, 1990,
1991) for an alternative interpretation.
[26] A surprising exception is the double passage-grave (ca.
four pairs of side-stones each chamber?) in a long-mound at Kierzkowo,
Bydgoszcz, Kujavia, excavated by Wislanski during 1982-1984 (cf. Bakker 1992:73
with references). The tomb is thought to date to the Globular Amphora culture.
Collagen dates are GrN-15412: 2320±40 b.c. and GrN-15411: 2185±40 b.c.
[27] Ormehøj, Katbjerg, Mariager ls., Onsild h., Randers a.,
East Jylland and Kornerup, Sømme h., Kjøbenhavns a., East Sjælland (A. P.
Madsen 1896:16 No. 27 and 1868:9 No. 20 respectively).
[28] A. P. Madsen's (1868:19 No. 30) illustration of two
immediately adjacent passage-graves (Onshøj, Kjærby, Rørby s. Sb.
12, Arts h., Holbæk a.) may show that the third capstone of the smaller
three-capstone passage-grave straddles the two end-stones of the larger
four-capstone chamber. However, the original 1881 drawing by Løffler makes this
perception less likely, and Rosenberg's 1915 sketches negate it altogether (cf.
Hansen 1986:43-48).
[29] Ubby/Udby, Arts h., Gl. Holbæk a. (A. P. Madsen 1868: 18 No.
29). Although Ubby 29 is a triple passage-grave, its southern part could
either be a double passage-grave or, if not divided in the middle, a 7/8+8 side
stone dual passage-grave.
[30] Bistrup, Fodby
s., O. Flakkebjerg h., Sorø a. (A. P. Madsen 1868: 26 No. 41).
[31] Gundestrup,
Skvæm s., Østerhan h., Hjørring a. (A. P. Madsen 1896: 12 No. 21).
[32] Røddinge, Møn,
Præstø a. (A. P. Madsen 1868: 37 No. 59). It is derived from similarly
shaped simple chambers, such as Sparresminde and Nebbølle, Møn, Præstø a. (A.
P. Madsen 1868:36-37 No. 57-58 respectively). Sneslev Sb. 1, Gl. Sorø a.,
Sjælland (A. P. Madsen 1868: 25 No. 39, Hansen 1993:114 No. 104).
[33] The Swedish gallery-graves are not part of this research,
because they are normally attributed to the Late Neolithic, but they may in
some cases signify a continuation of TRB burial practices (Hyenstrand
1989:12-16).
[34] For additional chambers with steps cf. Bakker 1992. The
oldest steps probably stem from the rectangular long-mound with open
perpendicular urdolmen Ovsted s. (Sb. 16), Voer h., Jylland (Baldia 1995:183
Fig. 9.5).
[35] However, the earliest collagen date from the 8.60 m long
gallery-grave of Niedertiefenbach is KN-2773: 2300±50 b.c. (Beier 1991:183a).
[36] Among these chambers is the Wechte-type grand-dolmen-like
15 m long chamber of Hilter, Kr. Osnabrück, classified as a gallery-grave (Hessisch-Westfälische
Steinkiste, Laux 1991:82-85, Fig. 27). The fact that it contains TRB West Group
Horizon 4 pottery implies a close and late connection between gallery-graves
and the TRB, not to mention the (long lasting?) grand-dolmen tradition
(Fig. 26).
[37] For an alternative view cf. Behrens 1973.