The Earliest Settlement of Denmark
From PalaeoWiki
Af Jørgen Holm
Oprindelig trykt i Larsson, L. (ed.) 1996: The Earliest Settlement of Scandinavia. Acta Archaeologica Lundensia. Nr. 24.
| Table of contents |
Abstract
The last twenty years have seen major growth in research into the Late Palaeolithic in Denmark, which may well be the most significant area of growth in the archaeology of this country. We can now establish that practically all the cultures/industries associated with the Weichselian Ice Age, which are known from other parts of the North European lowlands, are also represented in Denmark: the Hamburgian Culture, the Federmesser Culture and the Ahrensburgian Culture. In addition, there is the Bromme Culture, which appears to be a predominantly South Scandinavian phenomenon.
Introduction
The study of the earliest Palaeolithic in Denmark has never attracted great favour among the archaeological establishment, and accordingly it has received totally inadequate financial resources. For years on end, Paleolithic and Mesolithic research has only received approx. 10% of the funds that have flowed into archaeological research. In spite of this unfair treatment, which has gradually been supplanted by a little more understanding, a couple of handfuls of enthusiastic professional and amateur archaeologists have successfully achieved sensational results over the last few decades, which certainly represent some of the greatest advances in Danish archaeology.
Fig. The earliest settlements and single finds in Denmark, belonging to the Hamburgian and Federmesser Cultures.
Research history
The fact that Denmark was inhabited during the Palaeolithic had already become clear i 1889, when a reindeer antler striking weapon was found at the foot of a coastal cliff at Nørre Lyngby, on northwest Jutland (Müller 1896). These cliffs were again the focus of attention in 1915, when a tanged point made of flint was found in a layer dating from the Younger Dryas (Jessen & Nordmann 1915). Implements made of organic materials, including reindeer antler striking weapons, which, on the basis of the raw material alone, could be dated to the Late Glacial, were also subsequently found distributed over the landscape (Mathiassen 1938 a; Mathiassen 1938 b Mathiassen 1941).
Bromme Culture
A major research breakthrough occurred in 1944, when the distinguished amateur archaeologist Erik Westerby successfully identified a settlement dating from the Allerød Period at Bromme, near Sorø, on Zealand (Mathiassen 1946; Westerby 1985; Fischer & Nielsen 1987). The so-called “Lyngby Culture” was then superseded by the more reliably authenticated “Bromme Culture”, which is still regarded essentially as a South Scandinavian, and, at a pich, North German phenomen. Research activities then stagnated for a couple of decades. It was not until 1970 that the first Bromme settlements were discovered to the west of the great Belt, on Funen (Andersen 1973; Holm 1972) and on Jutland (Madsen 1983; Nilsson 1989). The quantity of finds on Zealand increased steeply more or less at the same time (Rasmussen 1970; Petersen 1974; Petersen 1994; Fischer 1976; Fischer & Mortensen 1977; Fischer 1978; Fischer 1990 b; Jønsson 1984; Johansson 1996).
Hamburgian Culture
The next distinct breakthrough occurred in 1981, when a test excavation at Jels, on South Jutland, confirmed that a site already discovered by the amateur archaeologist Jørn Fynbo in 1968 belonged to the Hamburgian Culture (Holm & Rieck 1983). Ever since Alfred Rust, working during the 1930s, undertook his epoch-making investigations in the Ahrensburg Valley, northeast of Hamburg, the Hamburgian Culture has, in fact, been high on the list of topics of particular interest to Danish archaeology. The view held for many years, to the effect that the North German settlements were situated quite close to the ice edge, and that most of Denmark was still covered by ice at this time, weakened the belief that it might be possible to find evidence of the culture in this country. Nevertheless, in line with the growing conviction that South Scandinavia (Denmark and southernmost Sweden) was ice-free at a very early stage, even prior to the Bølling Period, there was increasing optimism surrounding the possibility of making Danish finds from the Hamburgian Culture. Preliminary indications had already ben given in 1968, with the publication by C.J. Becker of details of a typical shouldered point from Bjerlev Hede (= heath), between Vejle and Horsens in southeast Jutland (Becker 1970). The author himself, however, pointed out that it could not be excluded that this shouldered point had remained in a wounded animal that had arrived here from the area to the south, and that it was not necessarily evidence of the presence of a “Danish” hunter.
Research into these Late Glacial cultures has accelerated considerably, in particular since the investigation of the site at Jels (1981-83). Exploratory investigations conducted in 1983 at a distance of only 30 m from this site, Jels 1, successfully revealed yet another Hamburgian site, Jels 2, which was excavated in 1984 (Holm & Rieck 1987; Holm & Rieck 1992).
Settlements of the Hamburgian and Federmesser Cultures at Slotseng
The most remarkable example of what can be achieved by a combination of targeted investigation and good luck was discovered in 1985 in the archaeological specimen store at Haderlev Museum! This was another Hamburgian artefact asemblage that had already been brought to the museum by an former employee, Peter Lepik, in 1962. These finds proved to originate from Slotseng, in the valley of the river Nørre Å between the villages of Mølby and Neder Lert only ca. 5.5 km to the southeast of, and linked directly to the Jels sites by the valley. Reconnaissance was carried out in this area in the spring and autumn of every year during the period 1985-89. It was remarkable to note that the Late Palaeolithic flint could be found over an area as large as 75 x 75 m and yet without any distinct concentrations. This was nevertheless taken as an indication that we had to be dealing with several settlements. In order, amongst other things, to shed light on this situation, a trial excavation was made in the area in the autumn of 1989.
By using a large number of test pits, each covering only one quarter of a square metre and arranged systematically at a distance of 5-10 m apart incidentally, a method which had been instrumental in the discovery of the Jels 2 site we were able to isolate a find concentration measuring some 60 x 20 m and extending to a cluster of Late Palaeolithic settlements, at least three in number, which were provisionally designated a, b and c. It is appropriate at this stage to include the comment that the quarter-metre-square method makes a convincing contribution to location and delimiting unspecified concentrations of worked flint, but that it appears less suitable for identifying the particular culture involved. It should be supplemented by the excavation of larger holes at the centre of the concentrations, since this will give a greater chance of recording types of implement that are characteristic of the culture.
Federmesser Culture at Slotseng
One of the sites, known as Slotseng b, was excavated in 1990 by the author of this paper in association with the National Museum. It was a considerable surprise to discover that we might be dealing here with a settlement from the Federmesser Culture. Neither the surface reconnaissance nor the trial excavation had indicated the presence of a Federmesser settlement on this site. A flint concentration measuring ca. 6 x 8 m, forming an approximately star-shaped figure with rounded points, was discovered within an excavated area of 111 square metres. A distinct concentration of burnt flint was discovered at the centre of this figure, which indicates that the hearth was located in this area. By far the major proportion of the implements lay around this imaginary hearth, which again emphasizes that the hearth was the focal point around which the settlement activities were played out. In particular the backed points (=Federmesser) were gathered quite distinctly around and in the actual hearth. In all, just under 300 reliable implements, ca. 100 cores and slightly more than 10,000 flakes and blades were found (Fig. 3).
Fig. Slotseng. Flint implements from the Federmesser settlement b. 1-5 federmesser points, 6-8: scrapers (Wehlen-type) and 9-10: burins. Drawing: J. Holm.
The neighbouring flint concentration, c, was investigated in 1991, and was found to be a settlement from the Hamburgian Culture. It measured ca. 8 x 11 m. Remarkably little burnt flint was found (this is a noticeable feature, that was also observed in conjunction with the Jels sites), and it was accordingly not possible to identify a hearth. The investigation yielded in total ca. 200 regular implements, 70 cores and over 8 000 flakes and blades (Fig. 4). The aforementioned two sites have already been described in greater detail in earlier publications (Holm & Rieck 1992; Holm 1992; Holm 1993).
Fig. Slotseng. Flint implements from the Hamburgian settlement c. 1-4: zinken, 5-8: tanged points (Havelte type), 9-11: scrapers and 12: burin. Drawing: J. Holm.
1993 was the turn of the major find concentration, a, the extent of which had been established in the course of the trial excavation in 1989, and which was situated only ca. 30 m to the east of sites b (Federmesser Culture) and c (Hamburgian Culture). Gradually, as the excavation progressed, it separated into two sites from the Hamburgian Culture and the Federmesser Culture respectively (Fig. 5). The designation a was retained and linked with the Hamburgian settlement, whereas the Federmesser settlement directly to the southeast, and partially overlapping, was given the designation e (d is used for a small concentration of flint flakes, discovered in 1989 a short distance down the slope towards the northwest, and presumed to be material that had been removed from the settlements lying above it through a combination of geological processes and ploughing).
Fig. Slotseng. Distribution of typical flint implements. a: settlement of the Hamburgian Culture and e: settlement of the Federmesser Culture. Computer graphics: J. Holm.
By far the majority of the finds beneath the plough layer was found to be lying on the southern periphery of the large concentration that had been identified after the trial investigation in 1989. Thus, to judge from all the evidence, there appears to have been a displacement to the northwest of the uppermost flint material situated in the plough layer. The terrain actually falls uniformly in this direction. The displacement is presumably attributable to a combination of geological processes soil creep and ploughing.
Hamburgian site a measured ca. 7 x 8 m and was highly irregular. It consisted of 6-7 dense, small concentrations. Both flint waste and implements followed this pattern. Also, strangely few pieces of burnt flint were found on this Hamburgian site.
Federmesser site e appeared almost as an appendix to the Hamburgian site, and lay directly adjacent to its southeastern part. In spite of the fact that we were dealing here with a comparatively distinct concentration, it must be pointed out that a little Federmesser material also occurred in the southern part of the Hamburgian site, and that individual reliable Hamburgian types were present in the highly abundant concentration, measuring only approx. 4 x 5 m, which could be associated with the Federmesser Culture by a good margin. In conjunction with this concentration, archaeological remains were discovered over an area of a couple of square metres from the top of the plough layer and down to a depth of c. 1.5 m. The examination of profiles lends most support to the belief that a large uprooted tree was mainly responsible for the material being carried down to this great depth, even if we are not able to exclude the possibility that there was a small, natural depression here during the Late Glacial or even that the Stone Age people had dug down here. No structural features could be identified at any of the sites.
Fig. Slotseng. Profile section through the Hamburgian settlement c seen from north. In a 0.5 m wide baulk, the featureless, soft, artefact-bearing sand has been removed with a combination of digging and hosing down. Remaining, in highly irregular relief, are remains of the harder fluviatile layer series, disturbed by frost phenomena and bioturbations. Photo: J. Holm.
It goes without saying that the find material from these two slightly overlapping settlements can only be attributed to the respective cultures with considerable uncertainty. This can be done with a degree of equanimity as far as the most reliable types of implement are concerned, whereas the subdivision of cores and flint waste is far more uncertain and, until demonstrated otherwise, can only be based on the distribution pattern formed by the types of implement that characterize the culture. Any future refitting attempts and analyses of the raw material may perhaps resolve this question to some extent. Typical Hamburgian implements included 62 zinken and 61 tanged points of the Havelte type. 77 backed points (Federmesser) and 39 scrapers of the Wehlen type must be attributed to the Federmesse Culture. More uncertainty surrounds the affiliation of a further 50 scrapers and 109 burins.
Four Late Glacial settlements, lying together in pairs, have thus been investigated at Slotseng until now, in such a way that a Federmesser settlement at each location lies quite close to a Hamburgian settlement. This is remarkable. This area during the Late Glacial must have provided particularly favourable conditions, which attracted groups of hunters with centuries in between. It is informative in this context to mention the situation of a distinct forced passage, a fairly large lake and a kettle hole, and the abundant occurrences of unworked flint on the hillside immediately to the east of the settlements. This could indicate common features with regard to hunting strategy and utilization of raw materials. Moreover, the high yet at the same time strongly exposed situation of the settlements is only explicable if the landscape in both cultural periods was still open and afforded a clear view in all directions. This may well be an indication that the Federmesser settlements are not much younger than the Hamburgian settlements, dating from the Older Dryas or the start of the Allerød Period.
All the previously investigated sites at Jels and Slotseng are strongly characterized and disturbed by periglacial processes, bioturbations and ploughing (Fig. 6 & 7). A large proportion of the find material lies in the plough layer, and the rest in apparently random positions beneath it as already mentioned, all the way down to a depth of 1.5 m. Archaeological remains have been found distributed in frost wedges, uprooted trees and on animal routes, and fragments of flint blades and implements, which could be refitted, clearly indicate that some of these archaeological remains had travelled for at least 50 cm in the vertical plane. One feature shared by all these sites is the fact that they lie at a relatively high altitude by Danish conditions and in exposed parts of the landscape, where no new sedimentation has occurred to seal and protect the sites, and from where it is more likely for material to have been transported away from the top of the slope and down into adjacent hollows. As a result, these sites have been severely exposed to erosion and frost phenomena, perhaps in particular during the extremely cold Younger Dryas. Add to this the devastation caused by bioturbations (uprooted trees, tree roots and the activity of small digging animals) and, more recently, ploughing on the sites near the surface. These processes are continuing, as can be appreciated from the fact that quite recent artifacts, such as iron objects and fragments of porcelain, can occasionally be found more than 30 cm beneath the plough layer, without any indication that they were ever buried there.
The situation that we are facing is thus far from an in-situ nature. There is nothing to indicate that the flint artefacts are lying where they were placed or lost at the time of settlement. A similar situation has been established at numerous sites in this and other countries, not only in conjunction with Palaeolithic settlements, but also Mesolithic (Barton 1987; Barton 1992). It is a more difficult matter to evaluate the horizontal movements. We have documentary evidence from Jels 1 that the largest of the uprooted trees had a diameter of 2 m, and we must not lose sight of the fact that the horizontal displacement of the archaeological material occurs to at least half this value, i.e. ca. 1 m. This rapidly became clear to us at Jels, and we then organized our excavation and recording methods accordingly. Thus, by far the major proportion of the Jels sites was subjected to examination in quarter-metre units, a method which was consistently applied in conjunction with the Slotseng sites. From time to time, we excavated in 5-10 cm layers within these units, as an experiment, although we had to conclude, with the major vertical displacements in mind, that the results achieved do not make any meaningful contribution to the following analytical process. My own approach to the usefulness of such finds in conjunction with fine-grained site structural analyses is accordingly highly pessimistic and critical. I have also omitted from the discussion at this point the additional complexity and "interference" that may have resulted from reoccupied settlements on the same site a possibility that should be included to a considerable extent in our desk considerations. My general view is that some of the site analyses that have been presented in the literature take either no account at all, or very little account of the disturbances to which the archaeological remains were subjected after the settlements were abandoned (Blankholm 1991). This is a conspicuous weakness in the method.
The Kettle hole
One member of the team investigating Slotseng in 1991 was the geologist Jette Lorentzen, from Aarhus University, one of whose tasks was to take core samples in the hollow to the northwest of the cluster of settlements. This work was carried out as part of a landscape reconstruction exercise. One of the core samples (No. 7) taken in August 1991 at the foot of the slope ca. 70 m to the northwest of the closest settlements, yielded three bone or antler stumps which originated from the top of an olive-green layer of clay gyttja at a depth of ca. 4.40 m. Further "encircling" core samples indicated that we were concerned here with a small kettle hole measuring scarcely more than ca. 20 x 20 m. A trial trench covering barely 2 m2 was dug around the core location, leading to the finding of a reindeer antler and a further three bone fragments. As the base of the trench was dredged out, a completely undamaged, zinken-like flint implement was found close to the bones and at more or less the same level (Fig. 8). The ingress of ground water prevented excavation to a greater depth, and we had to resort to breaking off and recovering the piece of the reindeer antler projecting from the trench. The remaining bones were left where they were, possibly to await more detailed investigation at some future date (Holm 1992; Holm 1993).
An interim report, drawn up by curator Charlie Christensen, of the Environmental Department of the National Museum (Christensen 1991), states that all the evidence suggests that a complete series of Late Glacial layers and large parts of a Postglacial (Holocene) series are deposited in the kettle hole. Bones and flint were lying at the top of a ca. 60 cm thick layer of clay and calcareous gyttja (mud), which also contained well preserved pollen, the remains of insects and perhaps most interestingly wood. The first evaluation already favoured a dating to the Bølling Period. This has now been fully supported by an AMS dating of the bone/antler fragments obtained from core sample No. 7. The dating was performed by the AMS Laboratory, Institute of Physics and Astronomy, University of Aarhus (AAR-906): 12, 520+/-190 BP.
Fig. Slotseng. The completely fresh, Zinken-like flint implement from the kettle hole. The piece shows wearmarks after work with bone or antler. Measures 7.1 cm.
Converted into calendar years, this dating can be put at 12,750 BC. If this dating is then projected into the timetable and the climatic curve based on the Greenland Summit Ice Core (Johnsen et al. 1992; Dansgaard & Steffensen 1992), it shows that humans were already on the move in Denmark at the start of the Bølling.
Subsequent analyses have revealed that the reindeer antler from the kettle hole bears traces of cutting marks (K. Aaris-Sørensen, pers. comm.), and a microwear analysis of the zinken-like implement has shown that it was used for working bones or antlers (H. Juel Jensen, pers. comm.).
Discussion
As can be appreciated from the above, the last twenty years have seen major growth in research into the Late Palaeolithic in Denmark, which may well be the most significant area of growth in the archaeology of this country. We can now establish that practically all the cultures/industries associated with the Weichselian Ice Age, which are known from other parts of the North European lowlands, are also represented in Denmark: the Hamburgian Culture, the Federmesser Culture and the Ahrensburgian Culture. In addition, there is the Bromme Culture, which appears to be a predominantly South Scandinavian phenomenon.
The Hamburgian Culture is represented by a casual find, the previously mentioned shouldered point of the classical type from Bjerlev Hede (Becker 1970), two sites at Jels (Holm & Rieck 1992), two sites at Slotseng (Holm 1992; Holm 1993), and one comparatively recently found cluster of settlements, totalling four, at Sølbjerg, on southwest Lolland (to the south of Zealand) (Petersen & Johansen 1993; Petersen & Johansen 1994; Petersen & Johansen, this volume). Jels 2 is one of the largest/richest Hamburgian sites ever found, with regard to both the area of the settlement, ca. 11 x 8 m, and the number of implements, ca. 700 (see Bokelmann 1978 for a comparison between German and South Scandinavian settlements from the Hamburgian, Federmesser, Bromme and Ahrensburgian Cultures).
Until 1990, the Federmesser Culture was known only from a series of casual surface finds of Federmesser points (Andersen 1977; Madsen 1982; Fischer 1990a) and scrapers of the Wehlen type, and from a single reliable site at Rundebakke, Knudshoved Odde, on South Zealand (Petersen 1974; Petersen 1994). In addition there is a flint workshop, consisting exclusively of cores and blades/flakes, at Egtved on South Jutland. This has been attributed to the Federmesser Culture solely on the basis of analyses of the underlying flint technology (Fischer 1990a) a method considered slightly doubtful by the author of this article. The question is whether we without any typical implements to guide us today possess sufficient knowledge to enable us to distinguish, for example, between the Federmesser Culture and the Bromme Culture. However, the presence of this Culture in Denmark has now been reliably established with the excavation of the Slotseng b and e sites (Holm 1993).
The Ahrensburgian Culture, of which only glimpses are known until now through more or less uncertain individual finds tanged points, reindeer antler striking weapons and largebarbed harpoons with a spade-shaped base (Taute 1968; Becker 1971; Skaarup 1974; Andersen 1974) is now represented by an entirely reliable excavated site at Sølbjerg, where, as mentioned above, settlements dating from the Hamburgian Culture have also been identified (Petersen & Johansen 1993; Petersen & Johansen 1994).
Slow or rapid expansion?
The first settlement in Denmark can be traced back, via the C14 dating of worked reindeer antlers from the Slotseng kettle hole, all the way to the beginning of the Bølling immediately after the noticeable, and almost rocketlike rise in temperature which marked the end of Dryas 1 that is reflected inter alia in the Greenland ice cores. We may presume that it was not until this time that the climatic conditions and the natural environment became sufficiently favourable to permit people to live so far to The north. Once these living conditions were in place, however, the hunters rapidly arrived in South Scandinavia. The author does not subscribe to the idea of slow, gradual expansion, which did not bring the older Hamburgian Culture any closer than North Germany, and in a younger phase to southernmost Denmark, and the Federmesser Culture any closer than the middle of Jutland, etc. (Fischer 1991). The distance from North Germany to northernmost Denmark could have been covered in two weeks at the most! The actual form of the current find picture can presumably be attributed to the fact that far from every region in Denmark and the rest of South Scandinavia has been investigated with the same intensity. Rapid expansion, e.g. in the form of hunting expeditions, on the extreme periphery of the area which offered adequate living conditions, is not contradicted in any way by the image that we have formed of dynamic and highly mobile groups of hunters including with the support of ethnographic observations. This view is also supported by the absence of finds that pre-date the Dryas 1/Bølling transition in the more southerly areas of the northwest European lowlands (Street et al. 1994).
Migration and communication routes
At least two obvious access routes to South Scandinavia can be identified for the Hamburgian Culture. If we first examine the spread of the culture in North Germany, there is a clear tendency towards a westerly route leading from hill island to hill island. The submerged North Sea mainland is an unknown, but particularly interesting factor in this context. Its interest lies in the fact that it is very easy to imagine that the valley of the river Elbe at the time, and its continuation into the area that is now covered by the sea, was a major life nerve and signpost for the reindeer, and accordingly for the movements of the reindeer hunters. This means that the hunting groups may have entered South Scandinavian territory, including the hill islands of southwest Jutland, from a predominantly western direction.
Another, perhaps even more obvious possibility is a more easterly route - up towards the Jutland Ridge, the watershed, or to put it another way: the Hærvejen (= military road) or the Oksevejen (= cattle road), which, for the last thousand years at least, has been the preferred route between North Germany and the northern part of Jutland. In fact, by following this route, one would encounter a minimum of watercourses and boggy areas. This gives food for thought, although one should resist drawing excessively firm conclusions to the effect that the find locations at both Jels and Slotseng and Bjerlev Hede lie either on or in the vicinity of this route.
Recent major finds of settlements from the Hamburgian Culture and the Ahrensburgian Culture at Sølbjerg on Lolland make it necessary to consider the possibility of a third, even more easterly route, which is described in detail by P. Vang Petersen and L. Johansen (Petersen & Johansen 1993). Whereas the author of the present article prefers to use a broad brush in his description of hypothetical migration routes and communication routes, his two colleagues take the reader through every stick and stone of the landscape from the Ahrensburg valley, across Fehmarn and then on via Sølbjerg, Knudshoved Odde and the Køge bugt (= bay) on Zealand, to Scania. It is naturally always stimulating and refreshing to see already established theoretical models developed further (see, for instance, Holm & Rieck 1992), although this does not provide reasonable grounds, either here or elsewhere, for outlining such a detailed, almost concrete, route. If one really does wish to attempt to define a route, then the starting point must be a fairly large number of reliably dated sites from the same culture, situated at regular intervals on the route whereas Petersen and Johansen justify their route with sites from no fewer than four different Late Glacial cultures: the Hamburgian, Federmesser, Bromme and Ahrensburgian. A further consideration is that, whereas we have some idea of at least certain aspects of the hunting strategy of the Hamburgian and Ahrensburgian Cultures from the finds made in the Ahrensburg valley, we know practically nothing about the situation in respect of the Federmesser and Bromme Cultures. The latter is very often referred to, albeit on a particularly flimsy basis, as a culture based on elk hunting and accordingly with a major reliance on a permanent territory (Mathiassen 1946; Bokelmann 1978).
Situation of the Havelte group
It is appropriate and tempting to consider the reindeer bones and the flint implement found in the kettle hole at Slotseng in relation to the above sites, although the distance between them is no less than ca. 70 m, as already mentioned, and we have no way of being certain that coeval and cultural contact occurred. The flint implement most closely resembles a zinken, although it is not sufficiently typical in nature to permit us to attribute it without further ado to the Hamburgian Culture. If we examine the C14 dating, 12,520+/-190 BP (12,750 BC in calendar years), i.e. the beginning of the Bølling Period, we can confidently rule out the Federmesser Culture, which does not appear before the Bølling - Older Dryas transition. Most of the C14 datings fall in the Allerød, with a few at the start of the Younger Dryas (Bokelmann et al. 1983; Houtsma et al. 1984). On the other hand, this dating fits an early phase of the Hamburgian Culture and aligns quite closely with the datings for Poggenwisch (the Polish Olbrachcice site is slightly older (Burdukiewicz 1986), and Meiendorf and Stellmoor (lower layer) are slightly younger (Fischer & Tauber 1986). (Fig. 9). It is at this point, however, that we encounter the next problem: I would place both Slotseng a and c in particular on the basis of the special design of the projectiles (tanged points, rather than shouldered points) in the Havelte group, which is traditionally perceived as a late phase within the Hamburgian Culture (Bohmers 1947; Tromnau 1975; Tromnau 1981; Stapert 1981; Stapert 1982; Stapert 1992), an assessment which appears to be supported in the TL dating to 12,400+1,600 BP (calendar years) obtained on the basis of burnt flint from Jels 1 (Huxtable & Mejdahl 1992), which is attributed to the Havelte group together with Jels 2. In any case, this dating appears to be too young, and it poses along series of problems; for if we agree with it, then we must simultaneously accept that the youngest phase of the Hamburgian Culture is younger than the Bromme Culture! (Trollesgave ca. 11,100 BP in C14 years, calibrated ca. 11,000 BC (Fischer & Tauber 1986).
A relatively late position for the Havelte group is similarly supported by the C14 datings, of between 11,470 BP and 11,810 BP (Stapert 1992) obtained on the basis of charcoal from the Dutch Oldeholtwolde site, which is attributed to this group by the excavator on the basis of a morphological/technological analysis of the projectiles. However, the view of the author of the present paper is that these projectiles exhibit few similarities with the points from the eponymous locality of Havelte Holtingerzand, and the Dörgener Moor, Ahrenshöft, Jels and Slotseng sites. One should also exercise a degree of care with the dating of charcoal from Late Glacial occupation layers, even if it was found in the vicinity of a hearth. If it were to emerge that, together with the dating of the organic material in the Slotseng kettle hole, we had simultaneously obtained a dating of the Hamburgian sites a and c lying above it, then this would naturally have drastic consequences for our perception of the Havelte group as a late phase within the Hamburgian Culture. It lies rather at the early end of the culture, and we must look for an alternative explanation for the typological differences in the implements which emerge in relation to the Poggenwisch/Meiendorf groups. For example, are these differences of a regional, ethnic or functional nature? This brings me back to the question of shouldered point against tanged point of the Havelte type. Could it be conceivable, for example, that the asymmetrical shouldered points were used predorninantly in conjunction with lightweight throwing spears, whereas the symmetrical Havelte projectiles functioned as arrowheads? Could it be, in a particular set of circumstances, that only a functional difference i.e. not a chronological difference to do with hunting strategy and technique distinguishes the two groups apart? A Danish engineer and amateur archaeologist, Jan Friis-Hansen, arrived at the same theory independently from the author of this paper.
As already stated, however, there need not be any connection between the finds in the kettle hole and the Hamburgian settlements. One possibility which must certainly not be excluded is that a site in its own right, either a settlement or a special site (a kill site, for instance) is situated deeply buried in the immediate vicinity of the kettle hole. Perhaps we have one fragile clue in the form of a flint flake, which was retrieved from a depth of ca. 2.70 m in conjunction with core No. 16 ca. 15 m northeast of the kettle hole.
It is hoped that some of these questions will be resolved in conjunction with future investigations, which will involve further studies of the kettle hole and its immediate surroundings and the taking of core samples in the Late Glacial lacustrine deposits further to the north.
Hamburgian Culture vs. Federmesser and Bromme Cultures
There appears to be broad agreement that the Hamburgian Culture must be regarded as a northern, highly specialized offshoot of the Magdalenian technocomplex. In foreign countries to the south, including the central Rhine area, major organizational differences are found between the Magdalenian and the slightly younger Federmesser groups. During the late phases of the Magdalenian, associated with the Bølling, the subsistence strategy was based on the exploitation of a limited number of animal species (primarily the wild horse and reindeer), whereas the Federmesser people engaged in hunting the full range of mammals that existed during the Allerød. Similarly noticeable differences can be seen in the use of lithic raw materials by the two cultures: whereas the Magdalenian is characterized by high selectivity and economy in the use of the raw material, the Federmesser people, who lived largely in the same geographical area, appear to have had a generous attitude towards the economic use of raw materials, quality and standardization (G. Bosinski, pers. comm.). It is possible that similar differences will emerge once we have carried out a more detailed examination of the Hamburgian Culture on the one hand, and the Federmesser Culture and the Bromme Culture on the other. The best indication of the southern roots of the Hamburgian Culture, in the Magdalenian, is perhaps to be found in the highly selective hunting strategy, which was additionally restricted to reindeer here, and in the economic utilization of the lithic raw materials. In spite of the fact that these people came to South Scandinavia where an abundance of flint of excellent quality is available, they retained their original technology. I also regard the frequent use of combination implements and double implements ("Swiss Army pocket knives") as a contributory factor to this economic approach to the raw material. This is not because it was necessary, but because it was "genetically" programmed into the underlying technology. The Federmesser and Bromme Cultures are to be found in diametrical opposition to this, with their obvious waste and extravagance with the flint. The far-ranging migrations appear to have been superseded by a more territorially reliant lifestyle, which appears to have been interrupted with the onset of the Ahrensburgian Culture during the Younger Dryas, where it is possible to identify a return to a lifestyle and a hunting strategy based on reindeer, which in many respects resembles that of the Hamburgian Culture.
