A Landscape of Ancestors: The Heuneburg Archaeological Project
Preliminary report of the 2002 excavation of a Hallstatt tumulus (Tumulus 18) in the Hohmichele ("Speckhau") mound group, Markung Heiligkreuztal, Gemeinde Altheim, Landkreis Biberach.Submitted to the National Geographic Society by Bettina Arnold, December, 2002
AbstractIn the summer of 2002, a team of American archaeologists from universities in Wisconsin, Minnesota, Illinois, and New Mexico excavated Tumulus 18, one of about 36 mounds belonging to the so-called "Speckhau" group near Hundersingen in southwest Germany (Figure 1). The Speckhau group includes the Hohmichele, which is the second largest Iron Age mound in western Europe. The mound was chosen for excavation in consultation with Professor Dr. H. Reim of the Baden-Württemberg Landesdenkmalamt (State Monuments Office), Tübingen branch. The excavation of Tumulus 18 was part of a larger, long-term research project based at the University of Wisconsin-Milwaukee with the title "A Landscape of Ancestors." The project combines analysis of ancient DNA (aDNA) with conventional mortuary analysis to address questions regarding early Iron Age social organization and its relationship to funerary ritual. In addition, systematic study of the structure and construction of early Iron Age mounds will augment the relatively limited data set currently available for mounds of early Iron Age date in the Heuneburg region. The project was initiated in 1997 and thus far has involved excavation of an early Iron Age tumulus (Tumulus 17 NGS Grant #6470-99 and Grant #6854-00) in the same mound group as Tumulus 18, as well as the genetic analysis of previously excavated human remains from early Iron Age contexts in the region.
Project DescriptionTumulus 18 was approximately 20m in diameter and was preserved to a height of 1.55-1.6 meters. It is located approximately 200m southwest of the Hohmichele and roughly 15 meters east of Tumulus 17 (Figure 1). Its size and its proximity to both the recently excavated Tumulus 17 and the Hohmichele suggested a Hallstatt date, and made it a promising subject for investigation given the questions raised by the excavation of Tumulus 17. Due to past disturbances, including the removal of about one meter of mound fill for the amelioration of local farm fields, and the on-going gradual disintegration of the mound through animal burrowing and tree root incursion, a complete and systematic recovery of the remaining information from the mound was recommended by the Landesdenkmalamt. Even relatively well-preserved mounds in forested areas, most planted with coniferous trees for the past century or more, are at risk. Conifers, including the fast-growing pine, generate acidic soils that have completely dissolved the skeletal remains and eventually metal grave goods in mounds without limestone packing; both Tumulus 17 and Tumulus 18 demonstrated this only too clearly, as skeletal material was completely absent in most cases and represented only by tooth enamel in the case of a few burials. About 15 tumulus cemeteries are assumed to have been associated with the Heuneburg hillfort. Roughly 35 burial mounds are known in the Speckhau mound group today. This probably represents only 10% of the original total. Meanwhile, the chemical changes occurring within the mounds represent an additional, mostly invisible destructive force that will have erased all traces of cultural or physical remains in the remaining mounds in a matter of decades.
The experiences gleaned from excavation of Tumulus 17 led to the development of a slightly different approach to Tumulus 18. While Tumulus 18 also was divided into four quadrants, these were not subdivided into trenches. To avoid drawing the same profiles twice, separated by only 60 cm, as had been done with Tumulus 17, standing profiles were recorded after the excavation of the NW and SE quadrants, with the NS profile and the EW profile offset in the center by one meter. The offset central balk was recorded in the SE quadrant to complete the profiles. The NE and SW quadrants were then taken down to sterile. Based on the recorded stratigraphic sequences of Tumulus 17 and the Hohmichele, 25cm increments (plana) were excavated by quadrant, with cultural layers (strata) and features recorded in profile and plan. Excavation continued at least 20cm into the subsoil below the cultural deposits along each standing profile in order to expose as complete a cross-section as possible. Following consultation with our German colleagues, excavation of burials and features was undertaken by hand, with a small mechanized backhoe used to remove undifferentiated fill and in the final trenching of the standing profiles. Two mechanized wheel barrows were used to expedite the transport of backdirt.
The original diameter of the mound was difficult to determine in spite of a surface survey conducted at the beginning of the field season. As with Tumulus 17 and the Hohmichele (Arnold et al. 1999, 2000; Arnold and Murray 2002; Riek 1962), Tumulus 18 provided no evidence of a demarcation of the mound limits in the form of a trench, stone or post settings and the position of the mound on a slope extending to the southeast caused a certain amount of slumping and distortion of the mound outline in that direction. A logging road running northeast to southwest cut into the southeastern portion of the mound, making its original extent on that side difficult to reconstruct, while a drainage ditch had disturbed the northeast quadrant mound foot. The excavated diameter of the mound based on calculations from the completed profile drawings proved to be approximately 20 meters. Earlier reports had suggested a larger diameter of between 24-29m (Kurz and Schiek n.d.), the range a clear indication of the extent of the slumping. Similarly, while the estimated preserved height was originally reported at 1.4m, the excavated preserved height from the ancient ground surface to the preserved apex of the mound was about 1.55m - 1.6m.
History of InvestigationIn addition to obvious indications of looting and the presence of over 30 trees ranging from several years to a decade in age, about a meter of soil seems to have been removed from the mound mantle at some point in the recent past. Amelioration of fields surrounding the nearby Dollhof farm was recorded in association with the so-called "Small Hohmichele", not far away. In addition, the dimensions of Tumulus 17, which had roughly the same diameter, suggested that Tumulus 18 was originally also higher than it appeared in the summer of 2002. This was confirmed by the discovery of the first burials within two weeks of the start of excavations, very close to the surface. At least three, possibly four looter's trenches and pits could be seen in plan and profiles. Some of these may have been contemporary, with the looters digging in one area and then moving to another part of the mound in the course of the same night or series of days.
In 1893, A. Witscher of the Landeskonservatorium is recorded as having dug into at least two tumuli in the Speckhau group. This included Tumulus 17 and nearby Tumulus 18, but it is no longer possible to differentiate between the two in excavation archives. Some of the trenches visible in Tumulus 18 may be related to Witscher's probing, which the few available documents describe as having produced the remains of a decorated ceramic vessel that has since been lost (Kurz and Schiek n.d.). Tumulus 18 Grave 1 was a disturbance containing large fragments of Alb-Salem tradition burial pottery, with red-slip, graphite, and stamped as well as incised decoration. A jumble of sherds, some quite large, others mere fragments, suggest that the vessels were shattered before being dumped back into the trench, which continued all the way down to the level of the central cremation burial. The fragments of a decorated vessel reported by Witscher could very well have been pieces similar to the Alb-Salem pottery uncovered during the 2002 excavations. Unfortunately, similar fragments, though of a later date, were also found at the base of Tumulus 17, making this apparent clue less useful than it seems at first. However, it seems likely, given Witscher's track record of unsystematic exploration of the mounds in this group, that he did in fact dig into both tumuli.
There was much less evidence for recent animal burrowing in Tumulus 18 than in Tumulus 17, which had been honey-combed by several large and extensive fox or badger burrows, some of which were still inhabited at the start of the 1999 field season. This may have been due to the larger preserved size of Tumulus 17, as well as the stabilizing presence of the roots of the 100+ year old Douglas fir that had been planted on that mound.
Project ResultsIn their dimensions and the absence of a basal demarcation Tumulus 17 and Tumulus 18 were quite similar. However, the structural and mortuary history of Tumulus 18 was less well represented in its stratigraphy than was the case in Tumulus 17, where the individual fill layers and other features were clearly visible in the standing profiles. Tumulus 18 also lacked the viscous, stone-free 0.70-1.00cm thick layer of gray clay that separated the mound core of Tumulus 17 from the outer layers of the mound. A similar feature was observed by Riek (1962) in the Hohmichele, and its absence in Tumulus 18 seems significant.
The more homogeneous fill structure of Tumulus 18 may be in part related to the larger number of burials in the mound, which contained almost four times as many graves as Tumulus 17 (Figure 2). Looting pits and trenches certainly played a role in this mixing process as well. Presumably more frequent interment episodes would have caused disturbance of earlier layers, resulting in some mixing and homogenization of the mound fill. The large boulder that marked one of the burials (Grave 4) may have been placed there in order to make it possible for later graves to be inserted near it without its being disturbed. Some superposition and disturbance did in fact occur, suggesting that the intervals between burials may occasionally have been long enough that any markers (such as wooden poles, documented in the Magdalenenberg with a very similar pattern of concentric rings of interments around a central area) were no longer visible. Grave 13, for example, appears to have resulted in the almost complete clearing of Grave 16, which lay immediately below it, while Grave 11 was placed some 50 cm directly above Grave 17 in the vicinity of the right hip of that burial. Likewise, Grave 3 was placed just above Graves 4 and 5, although the nature of this assemblage of grave goods is problematic (see discussion below).
Mortuary FeaturesFeatures not identifiable as burials
Tumulus 18 contained 12 features that were presumably in some way linked to mortuary activity in the mound, but that were not associated with human remains or assemblages of artifacts that could be interpreted as grave goods. All of these features took the form of variably dense concentrations of charcoal, occasionally apparently the remains of small fires burned in situ when evidence of burned earth (Feature 2; Feature 13; Feature 22; Feature 23; Feature 32) was found in addition to the charcoal, sometimes mingled with burned pottery fragments (Feature 13; Feature 32). Similar small "altars" or ritual depositions of charcoal were reported in Tumulus 17 (Arnold et al. 2000, 2001) as well as in the Hohmichele (Riek 1962; Kurz and Schiek 2002). In that mound, the majority of these deposits seem to have been made on successive mound surfaces, and are typically found at the boundaries of individual strata, possibly to mark the completion of a mound construction episode. There seems also to have been a link between the central cremation burial in Tumulus 17 and some of these charcoal features, as is demonstrated by pottery refits. Whether or not the charcoal deposits in Tumulus 18 had the same characteristics and served the same function(s) as those in Tumulus 17 will have to await future analysis. However, the fact that many of these features could be seen in plan and profile to follow the slope of the mound suggests that they were deposited on the mound surface very much like the corresponding features in Tumulus 17 (Feature 2; Feature 8; Feature 11; Feature 19).
Most of the charcoal features in Tumulus 18 were quite small, ranging between 15 and 30cm; the largest was Feature 13, at over 1.5 meter in cross section. In some cases the charcoal layer was extremely ephemeral, i.e. less than 1cm thick, but in a few instances it was as much as 5cm thick (Feature 22). Feature 13 was one of the densest of these concentrations. This feature was unusual in terms of its size, the fact that it contained a large tree branch that had evidently burned in situ, and its peripheral position near the outer edge of the northwest quadrant. On a number of occasions several discontinuous deposits at the same elevation were documented quite close together; these give the impression of belonging to a dumping episode that may have involved separate basket- or bucket-loads of material (Features 2 and 4; Feature 23). Nine of the twelve features were deposited in the southeast and northwest quadrants, with three recorded in the southwest and none in the northeast quadrant. This was probably not coincidentally the quadrant with the densest concentration of burials, containing seven of the 16 secondary burials surrounding the central cremation.
Nor was this the only evidence that the central burial area had been disturbed. Altogether at least four major looter's trenches were observed in the mound's core. In one case each cut of the looters' shovels had left a row of clear serrated impressions along the edge of the collapsed pyre where they had cut through the charcoal layer into the lighter colored prepared soil platform beneath. Fortunately, while the central burial was more or less totally divested of its ceramic assemblage (there may never have been many metal grave goods in this burial, but any it did contain would have fallen victim to the looting of the grave), the remnants of the pyre itself and its foundation were still intact enough to provide information about the disposal of the individual who was buried in the center of Tumulus 18.
Grave 8 (a concentration of charcoal and cremated bone belonging to the central cremation that appeared at the base of the NW quadrant) and Grave 1 (hereafter referred to as Grave 1/8) consisted of a multi-layered concentration of charcoal of varying thicknesses distributed most densely to the west of the mound's central point, following the slope of the terrain on which the pyre was erected. Presumably the pyre platform was constructed so as to be level when the body was laid on it for burning, but the prepared surface on which it was erected could be clearly seen in profile following the natural slope of the terrain. This suggests that as the pyre collapsed, during and after the burning, the tendency would have been for the remains to slump downhill, leading to the distribution seen in the plan view of the pyre area (Figure 2). As one would expect, the area immediately inside the platform foundation ditches was also the area of greatest charcoal density, while further from the center of the pyre charcoal was both more discontinuous and more thinly distributed. This was also where the densest concentration of cremated bone was documented. Calcination was almost complete, and most of the fragments that were recovered were quite small, supporting the theory that the pyre was a large one that had been constructed of well-cured wood. Reddish and reddish-brown earth, as well as a small number of fire-damaged sandstone cobbles and the disposition of the charcoal remains themselves all point to in situ burning. The individual whose death was the catalyst for the construction of Mound 18 was undoubtedly cremated at the location where the remains of the pyre were uncovered, and those remains constituted the foundation of the mound superstructure.
The pyre platform foundation was first observed as a ditch-like feature marked by iron staining/mineralization along its edges, with light yellow, marly clay fill, running southwest out of the east profile wall under the heavy charcoal deposit and the lighter staining designated as Feature 17. The ditch ultimately proved to extend into the NE 1/4, where an identical ditch was discovered running parallel to the first. The two ditches were roughly 3m apart, and about 3 meters long, oriented with the ditches at the magnetic north and south sides of the rectangle, the "open" ends facing east and west. At their maximum extent they were about 40-45 cm wide, suggesting that they were dug to receive timbers of considerable size that presumably served to anchor the construction of the funeral pyre platform. The early radiocarbon dates from charcoal samples taken from the central pyre area confirm the fact that some large timbers from relatively old trees must have been used in the construction of this pyre. (See detailed discussion of the radiocarbon results below.) Both ditches had rounded ends that were slightly wider than the body of the ditch, suggesting that there may have been uprights at each corner.
Grave 8 was a charcoal and burnt bone concentration in the SE corner of the northwest quadrant. Large pieces of charcoal were recovered on the east side of the concentration, toward the center of the pyre area. The burnt bone was concentrated in the northwest corner; some pieces were quite large, and may be anthropologically analyzable. Other fragments were virtually completely calcined. A looter's trench, clearly visible in outline in the south profile, cut into the mound to just above the feature. Apart from badly burned sherds, the only finds were a fragment of burned bronze and an iron rod/pin fragment, also apparently subjected to extreme heat. The charcoal consists mainly of small fragments, rather than large, discrete pieces. Discontinuous concentrations of silver-grey sand, possibly analogous to what Riek described as Pfohsand in some of the Hohmichele burial contexts (1962), alternated with large compressed, discrete packets of charcoal and ash in this portion of the central cremation.
Grave 2 consisted of a concentration/scatter of burned bone and fragments of red slipped pottery, some quite large, possibly from more than one vessel, in the northwest quadrant. The pottery extended into the east profile, and additional fragments were recovered when the northeast quadrant was excavated. The burned bone was piled in a small concentration, and a number of the pieces were large enough that an anthropological analysis may be possible. There was very little charcoal, however. The vessel fragments in the vicinity of the burned bone appear to be associated, but the scatter pattern of the sherds suggests disturbance through a looting episode or burrow.
Grave 3 appeared as a small, irregular soil stain in the southeast quadrant at an elevation of about 604.20m. The feature contained two Fusszier and three Fussschalen fibulae with crossbow spirals, forms associated with Ha D3, as well as an iron belt hook and belt fitting, more characteristic of early La Tène burials, and a bronze nail-headed pin. No human remains were noted. The main challenge in the interpretation of this assemblage is whether or not it should be considered a grave in the usual sense of the word, i.e. a set of grave goods accompanying a human body. The artifacts recovered from this context were concentrated in an area roughly 50x30cm in size, and there were no indications of a coffin or other wood in the immediate vicinity of the objects, as was the case in most of the graves in this mound. Grave goods in the late Hallstatt period in this region are typically bound to the body of the deceased, that is they are placed where they would have been worn (Oeftiger 1984: 74). In the case of Grave 3, the fibulae should have been placed in the shoulder or head region, whereas the belt hook would usually be in the region of the waist. In fact, all the metal objects in this assemblage are found too close together to represent an in situ placement. At the same time, there was no evidence of a disturbance, and the assemblage could be considered complete (though the absence of either weapons or bracelets, male and female markers respectively, is problematic as well).
One possibility is that this grave is a cenotaph, or a grave without a body, perhaps marking the position within the social unit represented by this mound of an individual related to the person buried in Grave 4 (and possibly also Grave 5), interred no more than 10 cm below Grave 3. The marking of Grave 4 with a large boulder might have had something to do with the eventual placement of another grave so close to it, but this cannot be confirmed.
A similar assemblage, with three Fussschalen fibulae and a fourth fibula too poorly preserved to be identifiable, but more significantly, with two, possibly three bronze nail-headed pins that are identical to the one from Grave 3, was found in Tumulus 6 at the Hallstatt cemetery of Deisslingen, a site northwest of the Heuneburg region that is not far from the Magdalenenberg (Oeftiger 1984:60-63). This burial of a female individual is not only one of the wealthiest recovered from the Deisslingen tumulus cemetery, but also the latest in date. In addition to the fibulae and nail-headed pins that are similar to those found in Tumulus 18 Grave 3, it also contained a belt made of interlinked rods with ring ends, a form of female personal ornament more commonly found in the early La Tène period (Oeftiger 1984:79). Tumulus 18 Grave 3 is one of the latest assemblages recovered from Tumulus 18, and one of several that seem to link the mound stylistically to the Black Forest region to the northwest.
Grave 4 was also discovered in the southeast quadrant literally almost immediately beneath and slightly south of Grave 3 (Figure 3). It was marked by an elongated soil stain, probably the remnants of a wooden coffin or small chamber that was roughly rectilinear but poorly defined. A small boulder at the southeast end of grave was just south of two large bronze Schlangenfibel that were probably at the head or shoulder of what was likely a male inhumation. The grave also contained a bronze belt plate with riveted edges and an iron dagger in a bronze decorated wooden sheath.
Some wood, probably from the coffin or chamber ceiling and/or walls, was preserved in direct contact with the belt plate. The feet were oriented east toward Grave 5, the head was oriented to the west. No human remains were noted. The belt plate and the dagger are similar to those found associated with the male interment in Hohmichele Grave VI, which also contained a female individual, bronze vessels and a four wheeled wagon, among other grave goods. This may have some bearing on the interpretation of Grave 5.
Grave 5 shares a feature number with Grave 4 because of its proximity to that burial. Initially it was unclear whether Grave 5 was a separate burial or an extension of Grave 4. Grave 5 is at the same elevation as Grave 4, and plank stains of a coffin?chamber wall? between the two cannot be associated with one or the other of the burials with certainty. The staining associated with the burial assemblage recovered from Grave 5 in the southeast quadrant extended into the north profile, where the rest of the grave was eventually uncovered in the northeast quadrant, confirming that the individuals in Grave 4 and Grave 5 lay with the lower legs and feet either touching or, more likely, partly parallel to one another. The feet of the individual in Grave 5 were oriented west, toward the feet of Grave 4, and the head was oriented east, toward the head of Grave 16 in the northeast quadrant, which was at the same elevation as both Graves 4 and 5. (See discussion below.)
Grave 5 contained a leather belt with bronze staple? decorations, two bronze bracelets, one disturbed by a rodent burrow, an iron ring? or riveted object and a bronze sheet ring at chest level. The head was decorated with six sheet bronze hoop earrings, three on each side, and tooth enamel was preserved in the immediate proximity of the earrings, which also preserved coffin or chamber wood in the form of a dark soil stain.
Grave 6 was uncovered in the northwest quadrant, and initially was visible as a small rectilinear soil stain measuring about 80x80cm. Wood staining also appeared before any grave goods were excavated along the southern edge of the feature. The grave contained a large bronze belt apparently assembled from a number of different sheet bronze pieces, six sheet bronze hoop earrings and approximately 12 bronze hollow-ball-headed pins decorating a bonnet or other head covering (only six were visible on the left side of the head area exposed by excavation; the right side of the head area, which was removed en bloc, presumably will prove to have been decorated with the same number of pins as the left)(Figure 4). Two simple bronze wire bracelets were on each wrist. Tooth enamel was present in the vicinity of the earrings and hair/bonnet pins. Wood staining from the coffin/chamber was extensive, and made it possible to outline the grave, including an upright plank on the southeast side of the grave, which when excavated proved to contain the right bracelet still in an upright position. The belt area, which had been disturbed along its eastern edge by rodent burrowing activity, was largely covered with wood, presumably from the coffin/chamber. A sample of this was radiocarbon dated (see discussion below). The entire belt area was removed en bloc. The body was oriented with the head to the west and the feet to the east.
Grave 7 was excavated in the southeast quadrant. There was no noticeable grave outline, but dark stains were visible where bronze was present below the surface. A red-slipped Kegelhalsgefäß was uncovered in an upright position close to the east profile wall at the foot end of the grave. It appears to have been placed by the left foot, like the other pottery vessels from undisturbed contexts in this mound. There was some disturbance by rodent burrow(s), that cut through one set of three bronze bracelets and disturbed part of the bronze stapel-decorated belt tassel(?) along the right side of the body. Wood was preserved only in close proximity to the belt(?) and under the right hand set of bracelets. The belt(?) appears to have extended up under the right hand bracelets, but was cut in two by a burrow. Leather(?) seems to have been preserved in the form of a rusty brown organic material. A light gray, sandy soil was observed immediately below all the finds, a feature seen in a number of burials in this mound as well as in the Hohmichele, where Riek refers to it as "Pfohsand" (1962).
Grave 10 lay in both the northwest and the southwest quadrants. It contained a red-slipped Kegelhalsgefäß very much like the one found in Grave 7, as well as an unidentified iron and bronze object that appears to be a horse-bit. If this item is determined to be a horsebit following restoration in Tübingen, it will be the only example of horse trappings in Tumulus 17 and 18, and will make the identification of the grave as male more likely. The vessel probably marks the left foot of the probable inhumation, judging by the dominant pattern of burials with vessels in this mound. No obvious grave outline could be identified. Based on the position of the ceramic vessel, this inhumation was oriented with the feet to the northeast and the head to the southwest.
Grave 11 was excavated in the northeast quadrant, just above Grave 17. No obvious grave outline could be observed, and there was almost no staining in the vicinity of the metal grave goods, suggesting that this burial may have been interred without a coffin. Two large Schlangenfibel lay in what was presumably the chest or shoulder area and one small Schlangenfibel was found in what may have been the waist area, possibly acting as a shroud pin. Shroud pins are known from other contexts, for example in the Deisslingen tumulus cemetery (Oeftiger 1984:77). A small iron fragment was also found in the grave. Based on the position of the two large fibulae, this grave appears to have been oriented with the feet to the north and the head to the south.
Grave 12 was excavated in the southwest quadrant, and again there was no visible grave outline. It contained two small ceramic cups, one inverted; two bronze ankle rings (solid cast), one with traces of organic wrapping (bark?); three bronze fibulae, two Spitzpauken- and one Doppelpaukenfibel, all three late forms (Ha D3); one iron object with a wooden handle(?) in close association with the two better preserved fibulae, possibly an awl; and two unidentified iron fragments (Figure 6). An additional iron fragment was discovered in the area of a likely rodent burrow which seems to have dislodged one of the Spitzpaukenfibel and might account for the position of the two ceramic cups as well. The rodent burrows were not visible as soil discolorations, but could be inferred from the obvious dislocation of some of the grave goods. A corpse shadow and some degraded bone stained black was visible in the immediate vicinity of the ankle bracelets. Staining under the ankle bracelets suggests a coffin or chamber of wood was present. The late fibula types in this grave, as well as the absence of bracelets and the presence of ankle rings, a female marker, distinguish this burial from the rest. The iron objects may be remnants of an iron belt assemblage. On the basis of the fibulae and the possible iron belt assemblage Grave 12 can probably be dated around the same time as Grave 3, possibly a little earlier. The two small (almost miniature) cups are unusual as well. The body was oriented with the feet to the north and the head to the south.
Grave 14 lay in the northeast quadrant, and consisted of a large, fragmentary Kegelhalsgefäß with red slipped rim and neck, that was still standing upright but with half of the body and rim of the vessel displaced by a looter's trench or large animal burrow that was no longer visible as a soil discoloration. Root infiltration between the cracks of the dislocated sherds further supports the disturbance theory. Upon excavation, a second vessel was discovered resting on the base sherds and collapsed body sherds of the first (Figure 8). This was a small graphite slipped drinking cup with an omphalos base (see also Grave 9 for similar vessel set). No grave outline was apparent. Subsequently several large fragments of an iron object (sword? knife?), also obviously disturbed and broken, came to light to the north of the vessels; they may have been part of the possible inhumation(?) represented by this grave.
The ceramic vessel appears to be a Ha C type very similar to vessels found in cremation/inhumation graves with limestone slab deposits at the site of Welschingen near Konstanz on the Bodensee (Hald 2002:40). The vessel type (Alb-Hegau style) in Grave 15 is consistent with an interpretation of this grave as a biritual burial. The following objects were recovered from the probable inhumation grave: 1 iron razor or knife; tooth enamel fragments from a pig/boar mandible (a food offering); a bronze or glass bead and a fragment of bronze wire; two extremely long bronze Schlangenfibel with poppy-head terminals associated with human tooth enamel. The vessel was removed with its contents for careful excavation in the lab. If it contains cremated bone, then the grave should be considered a biritual multiple burial, with both a cremation and an inhumation in the same chamber/burial area.
Grave 16 appears to have been almost completely cleared of grave goods by the excavation of the chamber Grave 13 immediately above it. The only remaining artifacts were an iron object (a handle? a belt hook?) and several fragments of a large red-and-white slipped Kegelhalsgefäß. This is an early Ha D1 vessel type at the Heuneburg, which might explain why at the time of the placing of Grave 13, local knowledge of the earlier grave was no longer exact. On the other hand, it is possible that the disturbance of Grave 16 by Grave 13 was intentional. If that was the case, however, the remaining grave goods in Grave 16 are difficult to explain. The position of the body can no longer be determined with any certainty, but if the vessel was at the feet (like the others in this mound) it seems to have been oriented the same way as Grave 13, with the head to the south and the feet to the north. This is the orientation most commonly observed in Hallstatt inhumations in Baden-Württemberg (Oeftiger 1984: 68).
Grave 18 was also excavated in the northeast quadrant, appearing in the form of a small chamber or roughly rectangular coffin pit 2.9x.90m in size. A rectilinear soil stain, probably from a coffin board, was recorded inside the pit. A fragmentary bronze Fusszierfibula and four small iron fragments were recovered at different elevations; two sherds, possibly intrusive were also recovered. The position of the grave pit, outside the inner ring of burials and the late date of the fibula type suggest that this grave was placed in the mound late in its use life. The position of the grave goods suggests the body was oriented with the head to the southwest and the feet to the northeast.
Grave 19 was recovered in the southwest quadrant. The grave contained one bronze Schlangenfibel and an iron fragment. Tooth enamel was preserved, probably partly due to the fact that the grave had been dug into sterile. A rectilinear chamber outline that was 1.5x.90m in size was clearly visible. The chamber is comparable in its dimensions to that of the child burial in Grave 13, but in this case the surviving grave goods cannot confirm the possibility that this, too, was a child's grave.
Grave 20?/Feature 35 An upright vessel was discovered in the SE 1/4 early in the excavation, and was not given a feature designation because it appeared to be an isolated find. There was no indication of any soil discoloration around it suggesting the presence of a grave. Subsequent discoveries in Tumulus 18 of upright vessels with little or no evidence of a grave outline but with other grave goods indicating that a burial was present make it more likely that this was also a burial, albeit one without additional inorganic grave goods. Soil from inside the vessel, which disintegrated when it was removed, was also collected.
Grave 21?/Feature 20 This 1.4 x 0.70m roughly rectangular pit was initially given a
feature designation rather than a grave number because it did not contain any grave goods. The shape of the pit and the fact that it partly cuts into sterile are features it shares with Grave 18 in the southwest quadrant, and it may very well have been the grave of an individual buried without inorganic grave goods.
DatingRadiocarbon dating of five samples from Tumulus 18 was carried out by Beta Analytic Inc. of Miami, Florida. Charcoal samples recovered through water sieving and sorting from Graves 18 and 1/8, wood from Grave 6 and charcoal from Features 13 and 23 were submitted for dating and yielded the following results (at 2 sigma 95% probability):
There does not seem to be any evidence of recent contamination, but an "old wood effect" may be responsible for the extremely early date of the charcoal from the central cremation grave, which can be presumed on the basis of the foundation ditch diameters (40-45cm) to have held large timbers that were likely old trees when they were felled and were probably used first as structures for the living, followed by a period of long-term storage before they were incorporated into the funeral pyre. This is consistent with a pattern from better preserved contexts like the Magdalenenberg and Tumulus 17 Grave 1, where such curation of building materials has been documented as well. While there may be a functionalist explanation for the reuse of such material in mortuary ritual (large timbers of high quality were likely reserved for the needs of the living), there may also have been a ritual aspect to this curation of construction elements from disassembled structures.
Tumulus 17 has produced evidence for the curation of central funeral pyre remnants, including cremated bone as well as burned and broken pottery, suggesting that depositing material from the founding burial on the mound to mark construction episodes was an important part of the mortuary history of late Hallstatt mounds in this area. The timbers used to build the chambers or pyre platforms may have come from dwellings or structures of some significance to the deceased or his/her kin group - one could imagine, for example, that when the house in which an elite individual was born was abandoned or demolished, portions of it could have been stored to be used in the construction of his/her final "domicile."
This might also have been true for some satellite burials of individuals of a status entitling them to a chamber - Tumulus 17 Grave 1 would be an example of this, as is evidenced by the "old wood effect" documented in that grave (Arnold et al. 2000). Obtaining radiocarbon dates of wood and charcoal material from burial chambers throughout the West Hallstatt zone and comparing the results to the presumed dates of these burials based on the relative dates of their grave good assemblages might be instructive in this regard.
In addition the very early date of the pyre remains from Grave 1/8, the relatively late date of Feature 13 is interesting. This feature was located on the outer edge of the northwest quadrant of Tumulus 18, and it is quite possible that it represents the remains of a large in situ burning event at a late date in the history of the mound, possibly close to the end of its use as a burial monument. Any date later than 450 BC is significant for a burial monument associated with the Heuneburg hillfort, which has long been assumed to lack an early La Tène occupation. In addition, if the mound was founded in the late Ha C period - possibly as early as 700 BC - and Grave 3 and Feature 13 represent the end of the mound's use-life sometime between 450 and 400 BC, then the life span of this burial monument, and the longevity of the social group which it presumably represented, are unparalleled among the Heuneburg mounds.
OrientationThe significance of the cardinal directions to mortuary ritual in the "Speckhau" mound group that was suggested by the orientation of the central chamber/enclosure in Tumulus 17 (Arnold et al. 2000) is further underscored by the orientation of the 3x3m pyre platform in the center of Tumulus 18. The "closed" sides of the pyre platform, delineated by the foundation ditches, face north and south, while the "open" sides face east and west. Clearly all the later burials were placed in the mound with reference to the pyre platform of central Grave 1/8. The orientation of fourteen of the burials can be determined on the basis of the placement of grave goods, which are "bound to the body" during the late Hallstatt period (Oeftiger 1984:74). Most significant is the fact that in spite of the distribution of the burials in a tangential, circular pattern around the central cremation, none of the presumed inhumations are oriented with the head oriented due north.
There also appears to be a rough clustering of burials around the central cremation that corresponds to the cardinal directions. Group 1 (North) consists of Graves 6, 9 and 2 in the north; Group 2 (East) is comprised of Graves 14, 11 and 17; Group 3 (South) is the largest group, with Graves 3, 4, 5, 13, 16, 18, Feature 35 (Grave 20?), Feature 20 (Grave 21?), and possibly Grave 7. Group 4 (West) consists of Graves 10, 19, 12, and 15. The avoidance of north-facing entrances in the late La Tène rectangular enclosures known as Viereckschanzen (including the one only a few meters northeast of the Hohmichele) may be a manifestation of the same fundamental ideological precept.
Two sets of paired burials were interred with their feet toward one another: Graves 6 and 9 in Group 1 and Graves 4 and 5 in Group 3. Grave 9 was deposited at a depth of about 30 cm below Grave 6, but Graves 4 (?) and 5 (?) were at precisely the same elevation, and apparently interred in such a way that their feet must have partly overlapped. Dark staining was observed in the area between these two burials, and it is possible that they were actually interred in the same large coffin or chamber. Unfortunately, in the absence of skeletal material a closer determination of the relationship between these two individuals is not possible.
Superposition of burials occurs in three cases: Graves 13 and 16, in which the later grave partially destroyed the one below it; Graves 11 and 17, where the depth difference is about half a meter; and the placement of Grave 3 (if it is a grave) only 11-20 cm above both Graves 4 and 5. The significance of this superpositioning is based in part on the relative dates of these burials based on their grave goods. Grave 4 (the belt plate and dagger are close matches for a similar set of grave goods in Hohmichele Grave VI, dating this grave to Ha D1) and Grave 5 (the earrings suggest a Ha D1 date consistent with an internment contemporary with Grave 4) are both quite early, while Grave 3 is one of the latest burials in Tumulus 18, probably dating to late Ha D3 or even early LT A. It is possible that there is no connection between these burials, and that the apparent juxtaposition is due to chance. Mitigating against this interpretation is the unworked boulder placed near the head of the individual in Grave 4. Once the burial was covered up, this stone cannot have served as a grave marker, since it would have been out of sight. However, it could have been used to locate Grave 4, and by association Grave 5, by inserting a thin rod-like probe through the intervening layers. Since the evidence suggests that Tumulus 18 was once considerably higher than 1.55 meters, Grave 3 could have required the excavation of as much as a meter of mound mantle in order to be positioned in close proximity to Graves 4 and 5. In any case, the concentration of so many graves so close together in Group 3 (3, 4, 5, 13 and 16) seems more than a coincidence and suggests some sort of a relationship existed between these individuals.
Gender and StatusA demographic analysis of Tumulus 18 produces a female:male ratio of 7:2 if only burials with assemblages that can be assigned to a gender category with a high degree of probability are considered.
Cremation graves: 1/8; 2; possibly 15. No gender or age specific grave goods present.
Six of the seven probably female graves had bracelets on both wrists; the only exception, Grave 12, contained ankle rings, one of the most rigidly "female" categories of personal ornament during the late Hallstatt period. Five of the seven probably female burials contained bronze belt assemblages, although the belt styles varied (no two are exactly alike). The two female burials without belt assemblages - Graves 12 and 13 - were different in other ways. Grave 12 is the latest in date of the female burials based on its fibulae, and several iron fragments found in the disturbed waist area may be part of an iron belt assemblage that would be consistent with the chronologically later position of this grave within the mound. It is possible that belt assemblages were an age-specific component of female costume, and children may not have been buried with belts, which would explain the absence of such an item in Grave 13. This hypothesis remains to be tested, however. It is worth pointing out that the female burial in Grave 4 of Tumulus 17 also had a bronze belt assemblage. On the other hand, bronze belts are clearly not restricted to female burials, as the belt plate in Grave 4 demonstrates.
Four, possibly five, of the seven female burials contained quartz pebbles/cobbles (Graves 7, 9, 12 and 17). The fill of both Tumulus 17 and 18 is surprisingly free of stones, and generally it can be assumed that stones found in a consistent pattern within more than one burial represent intentional placement. The red slipped Kegelhalsgefäß in Grave 7 was placed at the left foot of the female interment, and two quartz pebbles were placed beside it. A quartz pebble and a clump of red ocher were found beside the right wrist of the female interment in Grave 9, and a quartz pebble was placed precisely between the footring decorated ankles of the female interment in Grave 12. The feet of the child burial in Grave 13 were framed by several large stones placed in an arc, but whether this represents the same phenomenon as the placement of quartz stones in the previously discussed burials is unclear. Finally, the female individual in Grave 17 was buried with half of a naturally split rose quartz cobble with pyrite inclusions between the upper thighs. Natural, unmodified or only simply modified objects were frequently included in Hallstatt burials, and seem to have served an apotropaic purpose in much the same way as miniatures, glass beads and pendants of various kinds (Pauli 1975).
Another category of costume that acts as a gender marker is the fibula, specifically the number of these clothing fasteners found in a grave. The late Hallstatt period in Württemberg appears to be characterized by the following pattern: 1-2 fibulae = male grave, 3 or more fibulae = female grave (Burmeister 2000). If this pattern applies in the case of Tumulus 18, then Grave 3 should be considered female, whereas Graves 18, 19, 15 and 11 are probably male (the miniature fibula in Grave 11 could be considered a shroud pin rather than part of the individual's costume, based on its size and position in the grave). At this lower level of probability, factoring in the evidence of the fibulae, the female:male ratio becomes 7:6, considerably closer to a demographically normal burial population.
Compared to the Giessübel-Talhau mounds closer to the Heuneburg and the unlooted inhumations in the Hohmichele - even compared to Tumulus 17 - Tumulus 18 makes a middle-class impression. On the one hand, two individuals in this mound were buried with neckrings, which is a grave good category not documented in the unlooted burials of Tumulus 17. On the other hand, none of the Tumulus 18 graves contained bronze vessels comparable to the cauldron from Tumulus 17 Grave 1, and although the central burials in both mounds were looted, the sheet bronze and iron fragments recovered from the looter's trench in Tumulus 17 suggest it must have contained several high status items, metal drinking vessels probably among them. In the matter of weapons, too, Tumulus 18 with its solitary iron dagger (Grave 4) compares poorly to Tumulus 17, which contained a leather helmet with an iron plume clamp (Grave 1), an iron short sword (Grave 1), an iron dagger (Grave 3), and at least seven iron spear points (two each in Graves 1, 3 and 5, and one in the disturbed Grave 2). There is no evidence of gold or wagon parts in Tumulus 18, items found in more or less fragmentary state in the Hohmichele and the Giessübel-Talhau mounds.
Given the still small number of systematically excavated mounds associated with the Heuneburg, it is too soon to draw any conclusions, but Tumulus 18 seems to have been the oldest and apparently the most populous of the social units represented thus far in the "Speckhau" mound group, and while its longevity is clearly documented, the status of this group appears to have been more stable over time than that of the social unit represented by the Hohmichele, which either was unsuccessful in reproducing itself past 540 BC, or shifted its mortuary activities to the Giessübel-Talhau area after the destruction of the mudbrick wall settlement at the Heuneburg. Both Tumulus 17 and 18 seem to have continued in use past this time, but the social position of their lineage or clan seems to have been unaffected by the "new money" that is reflected in the wealth of the Giessübel-Talhau mounds. Clearly, additional excavation and reanalysis of already excavated material will be required to address these questions.
We would like to thank the National Geographic Society and the Landesdenkmalamt Baden-Württemberg, Außenstelle Tübingen for their generous financial support of this project. We would also like to thank the following individuals for their assistance and encouragement: H. Reim, H. J. Teufel, S. and H. Hagmann, H. and H. Williges, R. Stadler, J. Selbherr and the other citizens of Hundersingen, Mengen, and the Dollhof who made us welcome in their communities during our stay in 2002.
Bibliography2000 B. Arnold, M.L. Murray und S.A. Schneider. Untersuchungen in einem hallstattzeitlichen Grabhügel der Hohmichele-Gruppe im "Speckhau", Markung Heiligkreuztal, Gemeinde Altheim, Landkreis Biberach. Archäologische Ausgrabungen in Baden-Württemberg 1999: 64-67.
2001 B. Arnold, M.L. Murray und S.A. Schneider. Abschließende Untersuchungen in einem hallstattzeitlichen Grabhügel der Hohmichele-Gruppe im "Speckhau", Markung Heiligkreuztal, Gemeinde Altheim, Landkreis Biberach. Archäologische Ausgrabungen in Baden-Württemberg 2000: 67-70.
2002 B. Arnold, M.L. Murray. A landscape of ancestors in southwest Germany. News & Notes. Antiquity 76(292): 321-322.
1981 K. Bittel, W. Kimmig and S. Schiek. Die Kelten in Baden-Württemberg. Stuttgart: Konrad Theiss Verlag.
2000 S. Burmeister. Geschlecht, Alter und Herrschaft in der der Späthallstattzeit Württembergs. Tübinger Schriften zur Ur- und Frühgschichtlichen Archäologie 4. Münster: Waxmann Verlag.
2002 J. Hald. Birituelle Hallstattgräber. Archäologie in Deutschland 4:40.
1997 S. Kurz and S. Schiek (eds), Bestattungsplätze im Umfeld der Heuneburg. n.d. (September 30, 1997 version).
1984 C. Oeftiger. Hallstattzeitliche Grabhügel bei Deisslingen, Kreis Rottweil. Fundberichte aus Baden-Württemberg 9: 41-80.
1975 L. Pauli. Keltischer Volksglaube: Amulette und Sonderbestattungen am Dürrnberg bei Hallein und im eisenzeitlichen Mitteleuropa. München: C.H. Beck Verlag.
1962 G. Riek. Der Hohmichele. Berlin: De Gruyter.
1997 J. Wahl. Die menschlichen Knochenreste aus den Altgrabungen im Hohmichele. In Kurz and Schiek (eds), Bestattungsplätze im Umfeld der Heuneburg. n.d. (September 30, 1997 version).
1979 H. Zürn. Grabhügel bei Böblingen. Fundberichte aus Baden-Württemberg 1979: 54ff.
FiguresFigure 1. Plan of "Speckhau" mound group showing the Hohmichele and Tumuli 17 and 18 (Matthew L. Murray).
Figure 2. Schematic plan of burials and significant features in Tumulus 18 (Seth A. Schneider).
Figure 3. T18 Grave 4. Belt plate and dagger in wooden sheath with bronze decoration.
Figure 4. T18 Grave 6. Earrings and hair/bonnet pins in situ.
Figure 5. T18 Grave 9. Drinking cup inside collapsed ceramic vessel.
Figure 6. T18 Grave 12. Ankle rings in foreground with quartz pebble.
Figure 7. T18 Grave 13. Close-up of neckring and beads. Glass-headed pins mark jet and glass beads.
Figure 8. T18 Grave 14. Drinking cup inside collapsed/disturbed ceramic vessel.
Figure 9. T18 Grave 15. Ceramic vessel in situ with part of stone setting.
Figure 10. T18 Grave 17. Bronze bracelets and belt assemblage in situ.