Wolfgang Taute’s Excavation at the Open-Air Site Feuersteinacker and the Early Mesolithic in the Western Part of Central Germany

ABSTRACT

Surveys near the village of Stumpertenrod revealed one of the largest Mesolithic assemblages in Germany. As a consequence of agricultural activities, the archaeological layers were partly eroded and the lithic artifact consisted mainly of surface finds. Between 1964 and 1966 Wolfgang Taute – a key-figure for the study of the Mesolithic in Europe – opened a trial-trench. Due to a lack of organic material, the campaign did not lead to the expected outcome and the site slowly fell into oblivion. The following article presents the results of typo-technological analyses of the lithic assemblage discovered in the course of the excavation. It provides new insights into the subsistence strategies of people during the early Holocene and re-integrates this significant site in current frameworks and debates. Furthermore, an interregional comparison of archaeological features and topographic parameters involving modern theoretical and methodological approaches, leads to a better understanding of the Early Mesolithic in the western part of Central Germany.

KEYWORDS:

• Mesolithic
• Central Germany
• lithic technology
• microliths
• lithic sourcing
• research history

Introduction

The following paper is concerned with the lithic artifacts discovered during a trial-excavation at the open-air site Feuersteinacker by Wolfgang Taute between 1964 and 1966. The location revealed one of the largest Mesolithic assemblages in Central Germany, dating to the early part of the Holocene (Preboreal). In the framework of the presented study, it was possible to reconstruct the reduction sequence applied for the manufacture of microliths at the locality and at contemporaneous sites in the wider region. Furthermore, provenance analyses of the lithic raw material led to a better understanding of territories and social relations of hunter-gatherers living at Feuersteinacker. Finally, a contextualization of the archaeological features displayed cultural similarities between the study area and adjacent regions and allowed for a better archaeological definition of the Early Mesolithic in the western part of Central Germany.

Topography

The village of Stumpertenrod (Vogelsbergkreis) is situated in the central part of the Federal State of Hesse, about 65 km northeast of Frankfurt am Main as the crow flies, on the flanks of the Vogelsberg Mountain. With an area of 2500 km², the Vogelsberg Mountain is the largest volcanic structure in Central Europe and the surrounding terrain is strongly influenced by basalt formations, which are the result of several eruptive events during the Miocene (Lotz, 1995, p. 125). The mountain range rises to an elevation of 773 m above sea level and constitutes an important landmark in the region. The open-air site Feuersteinacker is situated on a small hill that is enclosed by two streams, supplying the area with freshwater. The terrain forms a gentle southwest-facing slope at an elevation of 440 m above sea level. Due to wooded hilltops in the north and an elongated ridge in the south and southwest of the site, the area is protected from wind. As the distance to the next watercourse is about 500 m, it was possible to settle at the site without scaring away potential prey that was crossing the alluvial meadows (Krüger & Taute, 1964, p. 21). The soil consists mainly of silt and clay containing medium-sized gravel (Figure 1).

Figure 1. Location of the open-air site Feuersteinacker in Central Germany (Figure: T. Hess).

Research history

The study of the Mesolithic in the Western part of Central Germany began with the occasional discovery of surface finds, including microliths, at various locations within the Federal State of Hesse between 1925 and 1934 (Fiedler, 1994, p. 145). Several larger Mesolithic sites, located in close proximity to outcrops of silicified sandstone were subsequently documented along the course of the river Schwalm (e.g. the open-air site Hattendorf) (Quehl, 1982). Another important location dating to this time period is the site Hombressen in the northern part of Hesse (Fiedler, 1979, 1997). A first overview of the Mesolithic in Hesse was published in 1953 (Uenze, 1953). In the 1970s, Surendra K. Arora worked on Mesolithic sites in Western Germany and the raw material economy of the respective cultural groups (Arora, 1976, 1979). In the eastern part of the study area, Brigitte Pflug (1993) studied Mesolithic settlement patterns in the Rhön Mountains and presented a detailed petrographic description of different raw material types. A reconstruction of the paleoenvironment and the human impact by Early Mesolithic groups in the valley of the river Lahn was conducted between 1996 and 1998 (Urz et al., 2002; Bos & Urz, 2003). In Lower Saxony, just a few kilometers to the north of Hesse, Klaus Grote (1993) has worked on Mesolithic rock shelter sites in the Leine Uplands. Current research is aimed at studying the regional chronologies of the early Holocene to the south (Spies & Fach, 2017) and to the west (Gehlen et al., 2021) of the Federal State of Hesse.

The Mesolithic open-air site Feuersteinacker (German for “flint field”) is named after the fact that until the beginning of the last century, the inhabitants of Stumpertenrod collected prehistoric artifacts made of silica-rich rocks that were visible on the surface, to use them as strike-a-lights (Krüger & Taute, 1964, p. 19). After the first antiquarians became interested in the site, children sold the objects to earn pocket money (personal correspondence with local resident M. Semmler, September 2019). The site first came into focus of archaeological research in 1962, when the local farmer Willi Dietz handed over a collection of several hundred surfaces find to Herbert Krüger, the former director of the Upper Hessian Museum in Gießen (Krüger & Taute, 1964, p. 19). Wolfgang Taute (18.05.1934–29.11.1995) – one of the most influential figures for the study of the Mesolithic in Europe who was then working as an assistant at the University of Tübingen and later became a professor for Prehistoric Archaeology at the University of Cologne (Cziesla, 1996) – soon recognized the significance of the finds. As a consequence, members of the museum organized regular surveys during the following years, in order to document the site and collect further objects. Until today, thousands of lithic artifacts dating to the Early Mesolithic were discovered within an area of about 1 ha. Between 1964 and 1966 Wolfgang Taute and his team conducted an excavation aimed at a better understanding of the stratigraphy of the site and at chronologically classifying the archaeological features. The first article with preliminary results was published in the same year (Krüger & Taute, 1964). Furthermore, drawings of the artifacts were produced. It seems that some of the idealized microlith types that form an integral part of Taute’s scientific work were inspired by the astonishing regularity of the pieces discovered at Feuersteinacker. Due to a lack of radiocarbon dates and a chronological scheme – that was only later developed by Taute (1971) based on his observations at Jägerhaushöhle near Beuron (Baden-Württemberg, Germany) – the excavation did not yield the expected results. However, several amateur archaeologists continued to search the site for surface finds. By far the largest lithic assemblage (n = 8089) was collected by Horst Quehl (Fiedler, 2017). These artifacts were systematically analysed between 2018 and 2019, applying petrographic methods in order to understand the use and origin of different raw materials for the technology of Mesolithic hunter-gatherers (Hess & Riede, 2021). Furthermore, it was possible to emphasize the importance of the locality for the settlement system of people in Central Germany at the beginning of the Holocene. As a consequence, the site – that had fallen into oblivion since its initial publication – became the focus of public interest again and in 2021 the finds were displayed in a special exhibition at the Upper Hessian Museum in Gießen.

Archaeological features and finds

In the framework of an excavation in the 1960s, three different trial-trenches with a width of 1 m were opened and excavated in artificial spits of 15 cm (Krüger & Taute, 1964, p. 24). They had a length of 32 m (trench 1), 5 m (trench 2) and 15 m (trench 3) respectively (see Figure 2). The entire sediment was screened using a screen with a mesh size of 5 mm. As a consequence of plowing and land leveling – which had been practiced at least since the Middle Ages – the sediment was intermixed and consisted of homogenous dark clay and silt. While the density of the surface finds decreased from the northeastern to the southwestern part of the investigated area, the excavation yielded several artifact concentrations (reminding lithic workshops), alternating with zones that were less rich in finds (see Figure 3). Accumulations of round stones that first appeared to be remains of hearths or settlement structures, turned out to be a clearance cairn dating to the 1930s (Krüger & Taute, 1964, p. 25). However, it cannot be excluded that the rocks originally were parts of fireplaces or tents during the Mesolithic. The weathered basalt bedrock (C horizon) was encountered at depths between 45 and 75 cm. Besides several hundred Mesolithic artifacts, a few finds dating to the Neolithic, such as groundstone axes or pottery, were discovered. Additionally, a tanged point made of siliceous shale suggests the presence of hunter-gatherers during the Late Paleolithic. Nevertheless, based on typo-technological aspects and the composition of the raw material it can be assumed that the largest part of the lithic debitage dates to the Early Mesolithic.

Figure 2. Location of the trenches. (Figure: T. Hess after the original drawing by W. Taute).

Figure 3. Internal distribution of artifacts within the trenches. (a) Horizontal distribution. (b) Vertical distribution. (Figure: T. Hess).

Materials and methods

Typo-technological analysis

The dataset consisted of 1192 stratified single finds (≥1 cm) and more than hundred pieces of small debitage, dating to the Mesolithic that were discovered during several field campaigns between 1964 and 1966. They are part of the collection at the Upper Hessian Museum in Gießen. Some of the single finds recovered by Taute and his team were not included in the statistical analysis, as they typologically belong to the Paleolithic. Lithic analyses were performed using a Microsoft Access™ database. The latter is an adaptation of a template originally developed by Auffermann et al. (1990) that has been further developed with a focus on Mesolithic assemblages. In addition, the methodological approach is influenced by research traditions emphasizing the dynamic processes behind the manufacture of stone tools and considering various stages of production and (re-)use (Andrefsky, 2005, 2008; Gosden & Marshall, 1999). The classification scheme for the microliths is based on the work of Taute (1971) and Heinen (2012). Metric analyses were conducted with Mitutoyo™ digital calipers. A blade was defined as being at least twice as long as wide and having a width of ≥1 cm, as opposed to a bladelet, which has a width of <1 cm. In order to determine the exterior platform angles of blanks, a goniometer was used. The weight of the pieces was determined using a precision scale with an accuracy of 0.1 g. In order to identify characteristic fractures (Inizan et al., 1995, pp. 34–37) or use-wear, the artifacts were studied under an optical microscope. The observations were tested and confirmed by experimental work, using the same raw materials that are present within the assemblage.

Provenance analysis of lithic raw materials

Provenance analyses of lithic raw materials were conducted by applying petrographic methods. For this purpose, each artifact was studied under a Zeiss Stemi 2000 optical microscope with a magnification of up to 80x. The texture and the components of the pieces were described in order to understand the formation of the rock. Subsequently, the artifacts were compared to a reference collection that had been established between 2018 and 2019. The latter comprises samples of 25 different raw materials (including thin-sections), and contains the most important rock types used during prehistoric times occurring in an area of about 21.000 km². Using this approach, it was possible to link the raw materials with specific outcrops and reconstruct the mobility patterns of people living at Feuersteinacker. In addition to the assemblage presented in this paper, lithic artifacts from several other Late Paleolithic and Mesolithic sites in Hesse were investigated with a focus on the composition of the raw material. The different rock types occurring in the region and the various petrographic methods that were applied in order to distinguish them have already been described in detail in previous publications (cf. Pflug, 1993; Hess & Riede, 2021, 2022).

Results

Raw material economy

Table 1 summarizes the composition of lithic raw materials discovered during the excavation at Feuersteinacker. With a total number of 366 pieces (30.7%), the largest part of the Mesolithic assemblage consists of chalcedony. This material displays a dull to translucent luster and includes a variety of colors, ranging from brown through red and yellow to gray and white. Outcrops are situated in Homberg near the river Ohm (approximately 20 km to the northwest as the crow flies), Braunfels an der Lahn (about 55 km to the west) and Steinheim near Frankfurt (about 60 km to the southwest) (see Figure 4). All three variants are present within the assemblage. Chalcedony is followed by silicified sandstone from Lenderscheid with 293 pieces (24.6%). The material is particularly fine-grained and homogenous. Typical colors include white, grey and light pink. The mentioned outcrop is located around 40 km to the northeast of the site. Furthermore, a red to orange and slightly more coarse-grained variant of silicified sandstone that occurs near Rörshain, in close proximity to Lenderscheid, is present with a number of 258 artifacts (21.6%). With 123 pieces, locally available siliceous shale accounts for 10.3% of the assemblage. The rounded surface of the rocks points towards river gravels about 30 km to the west of the site. A green variant of silicified sandstone that was ascribed to an outcrop in Wahlen (20 km to the north of the site) was documented with a number of 108 artifacts (9.1%). Additionally, a small amount of lithic artifacts (13 pieces, 1.1%) consists of Cretaceous flint that occurs in Lower Saxony and Thuringia. It was imported over a distance of at least 150 km. Materials occurring in river gravels to the south of the site (such as radiolarian chert, quartz and chert from Flysch formations) are present with nine artifacts. Further eight pieces consist of Jurassic chert deriving from the Swabian or the Franconian Jura. Interestingly, the artifacts show characteristic alterations of colors that are a consequence of heat-treatment. Finally, there are three artifacts that are made of silicified sandstone from Ziegenhain (right next to Rörshain) and one object that consists of silicified sandstone from Hausen. In case of ten artifacts, it was not possible to determine the exact provenance of the raw material, due to their small size or patinations.

Figure 4. Origin of different raw materials in combination with Early Mesolithic sites. (1) Niederweimar. (2) Hattendorf. (3) Lahrbach. (4) Kleinsassen. (5) Rüsselsheim-Waldhaus. (6) Sulzdorf-Hoher Stein (Figure: T. Hess).

Table 1. Importance of different raw material types by number and weight (n = 1192).

Raw material	Number	Percentage	Weight (g)	Percentage
Chalcedony	366	30.7%	401.0	26.7%
Silicified sandstone (Lenderscheid)	293	24.6%	388.9	25.9%
Silicified sandstone, red orange (Rörshain)	258	21.6%	324.6	21.6%
Siliceous shale	123	10.3%	127.0	8.5%
Silicified sandstone, green (Wahlen)	108	9.1%	126.6	8.4%
Cretaceous flint	13	1.1%	12.6	0.8%
Materials from river gravels (radiolarian chert, quartz, etc.)	9	0.8%	5.0	0.3%
Jurassic chert	8	0.7%	5.0	0.3%
Silicified sandstone (Ziegenhain)	3	0.3%	18.4	1.2%
Silicified sandstone (Hausen)	1	0.1%	6.0	0.4%
Not determined	10	0.8%	86.9	5.8%
Total:	1192	100.0%	1502.0	100.0%

There are a few important differences concerning the composition of the assemblage recovered during the excavation in the 1960s compared to the surface finds that were collected in the following decades. Interestingly, the percentage of silicified sandstone from Lenderscheid is considerably higher in case of the surface collection. This seems to be due to the bright and light color of the mentioned raw material that differs from the soil at the site and hence makes it more visible. The same is true for objects made of siliceous shale showing a green or grey color. Materials that occur in very low numbers at the site, e.g. Triassic chert from Muschelkalk formations and Jasper from the Kellerwald region are not present within the assemblage (Figure 5).

Figure 5. Comparison of the proportions of different raw materials between stratified artifacts from the excavation by W. Taute (n = 1192) and the surface finds collected by H. Quehl until 2017 (I = 8089).

Table 2 offers an overview over the raw materials used for the production of microliths. For this artifact class, there was a preference for fine-grained raw materials with a high content of silica. With 23 specimens, they are dominated by chalcedony, followed by siliceous shale and silicified sandstone from Lenderscheid with 15 artifacts each. Further 12 microliths consist of silicified sandstone from Rörshain and six pieces are made of silicified sandstone from Wahlen. Two microliths consist of Cretaceous flint. Additionally, a piece that is made of radiolarian chert and a microlith made of chert from Flysch formations were documented. Finally, there is an artifact that is made of silicified sandstone from Ziegenhain.

Table 2. Raw material used for the production of microliths (n = 76).

Raw material	Number	Percentage
Chalcedony	23	30.3%
Silicified sandstone (Lenderscheid)	15	19.7%
Siliceous shale	15	19.7%
Silicified sandstone, red orange (Rörshain)	12	15.8%
Silicified sandstone, green (Wahlen)	6	7.9%
Cretaceous flint	2	2.6%
Materials from river gravels	2	2.6%
Silicified sandstone (Ziegenhain)	1	1.3%
Total:	76	100.0%

Table 3. Overview over the lithic assemblage (n = 1192).

Artifact class	Number	Percentage
Cores & Debitage	358	30.0%
Cores	12	1.0%
Preparation artifacts	285	23.9%
Shatter debitage	61	5.1%
Modification products	82	6.9%
Burin spalls	19	1.6%
Microburins	63	5.3%
Blanks	627	52.6%
Blades	22	1.8%
Bladelets	108	9.1%
Flakes	437	36.7%
Undetermined blanks	60	5.0%
Modified pieces	125	10.5%
Microliths	76	6.4%
Notched pieces	10	0.8%
Laterally retouched pieces	16	1.3%
Scrapers	13	1.1%
Burins	2	0.2%
Truncations	2	0.2%
Perforators	2	0.2%
Multifunctional tools	4	0.3%
Total	1192	100.0%

Lithic technology

The presence of a relatively high number of preparation artifacts (285 pieces) and shatter debitage (61 pieces) suggest a production of stone tools directly at the site. In case of various types of silicified sandstone, chalcedony, and siliceous shale, all stages of the reduction sequence are represented (c. Hess & Riede, 2021) (Table 3). Among the 12 cores in the assemblage, there are unipolar and bipolar pieces, as well as examples with multidirectional negatives, and cores with two striking platforms and a cross-section that displays a trapezoidal shape. Finally, there are a few nuclei on large flakes with centripetal negatives (see Figure 6).

Figure 6. Different core types. (a, b) Cores with centripetal negatives. (c) Core with two striking platforms. (d, e) Cores with multidirectional negatives. (f) Core with two striking platforms and a trapezoidal cross-section. (Drawings: L. Hilmar).

With 627 objects, more than half of the assemblage consists of blanks. Those are dominated by flakes, which are represented by 437 pieces. Furthermore, 108 bladelets, 22 blades, and 60 undetermined blanks were documented. The main goal of the reduction sequence was the production of blades and bladelets of a certain thickness with a straight longitudinal profile. They are surprisingly regular for an Early Mesolithic assemblage and often seem to be standardized (see Figure 7 and see Table 5). This is particularly true for laminar blanks made of silicified sandstone. This raw material occurs in the form of large blocks within the study area and allowed Mesolithic knappers to produce specimens with two parallel ridges on the dorsal surface and a rectangular cross-section in addition to specimens with a triangular cross-section. The blades and bladelets were subsequently processed to microliths using the microburin technique. This step of the reduction sequence is evident by the presence of 63 microburins and semi-finished products. In case of silicified sandstone, it is possible that laminar blanks were simply snapped to the desired size without creating a notch.

Figure 7. Microliths, microburins, and laminar blanks discovered at the site (Photographs: R. Johansen & T. Hess).

Based on technological aspects, such as the shape of bulbs of percussion and external platform angles, it is possible to state that the blanks were produced using direct percussion with a soft hammerstone (e.g. limestone or sandstone) and in some cases with a hard hammerstone (basalt or jasper) (cf. Pelegrin, 2000). This was also confirmed by experimental work. Compared to other lithic raw materials, it is more difficult to distinguish different knapping techniques applied for the processing of the various types of silicified sandstone within the assemblage. However, it can be assumed that a hard hammerstone was used at least during the initial phase of the reduction sequence as the material is significantly more difficult to process. Bulbs of percussion and eraillure flake scars on the distal end of knapping products show that some cores were placed on an anvil. Throughout the reduction sequence they were continuously rotated. At the same time, there are nuclei on large flakes that were reduced by free-hand percussion. The blanks that were produced like this show significantly steeper external platform angles. Although the use of an antler hammer cannot be completely ruled out (cf. Damlien, 2015), due to a bias concerning the preservation of organic materials, it seems that there was a clear preference for percussion tools made of stone.

The modified pieces are dominated by microliths with 76 pieces (see Table 4 and Figure 7). They include 28 microliths with oblique truncation that often show a basal retouch, as well as 15 crescents, eleven triangular micropoints, and ten isosceles triangles. Furthermore, eight micropoints showing either a completely or a partially retouched edge were documented. Finally, there are four specimens in the assemblage that could not be determined. For the manufacture of microliths, the blanks were laterally retouched prior to dividing them into different pieces. Ideally, Mesolithic knappers were able to produce a microlith with oblique truncation in combination with a crescent or an isosceles triangle from the same blank (see Figure 8). By applying this method, it was possible to create artifacts with a similar size and thickness that served as parts of composite tools (see Table 5). The basal retouch of triangular micropoints is almost exclusively dorsoventral and affects also parts of the surface. It is either straight or slightly convex, which has been interpreted as a chronological marker (Taute, 1971; Heinen, 2012, p. 606). Whereas most microliths were made from blades and bladelets, there are also a few cases in which burin spalls served as blanks. Interestingly, Mesolithic people sometimes reworked particular types of microliths and turned them into different shapes. With 58 pieces, more than 75% of the pieces are complete. This is a relatively high value and seems to be linked with the function of the site. While some of the microliths display characteristic fractures deriving from their function as parts of projectiles, others did not reveal any traces of use. Remains of wood-tar are extremely rare, which might be a consequence of postdepositional processes.

Figure 8. Ideal reduction sequence for the production of crescents, isosceles triangles, and microliths with oblique truncation (Drawings and photographs: T. Hess).

Table 4. Different microliths within the assemblage (n = 76).

Artifacts class	Number	Percentage
Microliths with oblique truncation	28	36.8%
Crescents	15	19.7%
Triangular micropoints	11	14.5%
Isosceles triangles	10	13.2%
Other micropoints	8	10.5%
Not determined	4	5.3%
Total:	76	100.0%

Table 5. Metric attributes of complete microliths (n = 58) and complete laminar blanks (n = 86) (measurements in mm).

Complete microliths	Length	Width	Thickness	Length/Width
Mean:	16.8	8.1	2.9	2.1
Maximum:	24.0	15.1	4.8	3.0
Minimum:	10.2	5.0	1.5	1.3
Standard deviation:	2.9	1.9	0.7	0.5
Complete laminar blanks	Length	Width	Thickness	Length/Width
Mean:	21.6	8.9	3.2	2.4
Maximum:	39.3	19.2	7.9	3.5
Minimum:	10.9	4.4	1.4	2.0
Standard deviation:	5.8	2.4	1.2	0.4

Other tool types occur in relatively low numbers (Table 3). They include laterally retouched pieces with a number of 16 artifacts, 13 scrapers, ten notched pieces, four multifunctional tools, and perforators as well as truncations with a number of two pieces each. Finally, two burins and 19 burin spalls were documented. The difference in numbers implies that the tools were resharpened several times and that people took them along when they moved on to the next site (cf. Gelhausen, 2011).

In addition to artifacts made of silica-rich rocks, there are several elongated groundstone tools among the finds from the excavation at Feuersteinacker and the surface collections. The objects are made of fine-grained sand- and siltstones that are rich in mica. Suitable pebbles were further shaped by grinding. The artifacts show characteristic scars on their distal and lateral parts, which are the result of their function as retouching tools by striking or applying pressure (cf. Taute, 1965; Weiner, 2012). Additionally, it was possible to confirm by experimental work that the objects were used to create a notch as part of the microburin technique (see Figures 9 and 10). Similar to the microliths described above, the tools have measurements that appear to be standardised and could be linked with specific tasks. At the same time, they display an ergonomic design that might be adapted to individual knappers (cf. Taute, 1965). This hypothesis is further supported by the fact that there are examples of decorated retouching tools dating to the same period (Grote, 1979). The mentioned objects find their equivalent within several assemblages of comparable age in the Federal State of Hesse and adjacent regions (see Fiedler, 1994, p. 56; Grote, 1979; Lauerbach et al., 1997). Comparable objects are also known from other Early Mesolithic sites in Central Europe (e.g. Leesch, 2017, pp. 109–121). Additionally, there are stone slabs that served as grinding stones for the production of ocher powder, or as a work surface for other tasks.

Figure 9. Groundstone tools found at Feuersteinacker. (a–c) Hessian State Museum Kassel. (d) Upper Hessian Museum Gießen (Photographs: I. Görner, J. Knossalla, T. Hess).

Figure 10. Reconstructed use of groundstone artifacts found at Feuersteinacker. (a) Retouching of an edge by striking. (b) Creation of a notch as part of the microburin technique. (c) Retouching of an edge by applying pressure. (Figure: T. Hess).

Interestingly, a number of objects made of silicified sandstone show typical features of Middle Paleolithic artifacts, associated with Levallois technique, e.g. chapeau de gendarme platform preparation. Similar pieces were documented in large numbers near outcrops of silicified sandstone in entire Hesse (Bosinski, 1967; Fiedler, 1994). At the same time, final artifacts that indicate a longer stay by Middle Paleolithic groups are almost completely absent at Feuersteinacker. As some of the pieces consist of raw materials that show the exact same properties as in the case of Mesolithic bladelets, it can be assumed that there was a reuse of Middle Paleolithic cores and preparation flakes that were collected near the respective outcrops and transported to the site by Mesolithic groups.

Contextualisation

Typo-technological analyses indicate that the open-air site Feuersteinacker was repeatedly used as a workshop for the serial production of stone tools during the Early Mesolithic. The applied technology and the selection of lithic raw materials allowed the manufacture of standardised stone implements and suggest a certain degree of craft specialization. In addition to microliths, the modified pieces mainly include tools with a scraping or grinding function that were used in the framework of retooling and rehafting.

The spectrum of lithic raw materials is extraordinarily diverse and points to all directions with a clear emphasis on outcrops to the north and northeast of the site. Whereas a large part of the rocks used for the production of stone tools occur within a distance of 60 km around the site, imports of lithic raw materials over distances of more than 150 km suggest a far-reaching contact network. As several rivers rise in the Vogelsberg Mountain and flow radially in all directions, the site Feuersteinacker – that is nowadays situated in a rather remote area – was once located at an important junction (see Figure 4).

The combination of different types of microliths is typical for the Beuronian A (∼9000 BC) which corresponds to the Preboreal. Sites of comparable age are known from Niederweimar in Lahn Valley (Schön, 2016; Urz et al., 2002), 35 km to the northwest of the site, as well as Kleinsassen and Lahrbach in the Rhön mountains, 50 and 60 km to the east as the crow flies (Pflug, 1993, pp. 15–37). Other sites in the Federal State of Hesse that yielded microliths that are typical for an early phase of the Mesolithic are Groß-Gerau and Rüsselsheim-Waldhaus (Fiedler, 1994) about 90 km to the southwest (see Figure 4). From Niederweimar (NW 6) there is a radiocarbon date pointing to an occupation during the late Preboreal (UTC7263: 9580 ± 60 BP, 9130–8850 cal BC, 2σ, calibrated with OxCal 4.4) (Schön, 2016, p. 31). At this site, microliths such as crescents and truncated micropoints, elongated pebbles that served as retouching tool, and stone slabs were discovered. Laminar chert from the Franconian Jura links the assemblage with recently investigated Early Mesolithic sites in the northwestern part of Bavaria. The two regions are connected via the river Main that played an important role as transportation route (cf. Spies, 2020), and perhaps as a unified culturally significant entity beyond functional matters (cf. Hussain & Floss, 2016). Lithic assemblages from northwestern Bavaria display great similarities to finds from Hesse, although they mainly consist of Jurassic chert (cf. Lauerbach et al., 1997). Furthermore, there are noticeable parallels concerning the shape and composition of microliths to sites in Lower Saxony (Fabesch, 1986; Grote, 1993), where flint and siliceous shale were the predominant lithic raw materials. In consequence, the size of the final objects is generally a bit larger than in the south. Compared to the Federal state of Hesse, crescents are less common in neighboring regions.

Discussion

In conclusion, it is important to say that the site Feuersteinacker in Stumpertenrod had the function of a major hub within the settlement system of hunter-gatherers at the beginning of the Holocene. This becomes evident by the composition of the lithic raw material that points to all directions. Instead of processing the rock types near the respective outcrops, they were transported to the site over large distances. Topographic parameters and the vicinity to the sources of several rivers allowed people to maintain social relationships despite of generally smaller territories. It is most likely that the site reflects seasonal gatherings of otherwise dispersed groups. At these occasions, an exchange of material and non-material goods took place. The production of microliths and composite tools allowed completely new ways of sharing and exchanging objects as a materialized form of maintaining social relations (cf. Finlay, 2003). Among many recent hunter-gatherer societies, ritualized exchange-networks served as a form of risk management during phases of limited access to resources (e.g. Wiessner, 1983). Besides the possibility that raw materials were stored at the locality, the site also facilitated contacts between Mesolithic people as a persistent place in the landscape (cf. Barton et al., 1995).

Furthermore, the presented results lead to a better archaeological definition of Early Mesolithic cultures in Hesse. The composition of microliths corresponds to the Beuronian A in Southwestern Germany defined by Wolfgang Taute (1971) based on his observations at Jägerhaushöhle (layer 13). Typical artifact types include micropoints with dorsoventral basal retouch, crescents, microliths with oblique truncation, and isosceles triangles in combination with relatively large macrolithic tools. However, there are a few peculiarities evident in the Mesolithic material from Hesse. While Early Mesolithic assemblages in Souhwestern Germany mainly consist of chert from Jurassic or Triassic (Muschelkalk) formations, as well as radiolarian chert (Holdermann, 2006; Kind, 2009), the spectrum of lithic raw materials in the western part of Central Germany is considerably more diverse and there was an emphasis on silicified sandstone, chalcedony, and siliceous shale (Hess & Riede, 2022). A common feature of the Beuronian is the heat-treatment of lithic raw materials, leading to a typical alteration of colors. The percentage of lithic artifacts showing traces of this process is usually around 65% (Jägerhaushöhle layer 13). In case of assemblages from Northern Bavaria, where Jurassic chert equally accounts for more than half of the raw material, this value rises up to 69% (e.g. at the site Sulzdorf-Hoher Stein near Giebelstadt, n = 749). Whereas imported Jurassic and Triassic chert from Muschelkalk formations within Early Mesolithic assemblages in Hesse was always heat-treated, this phenomenon is almost absent in case of regionally available raw materials (although a number of artifacts made of chalcedony implies the presence of the technological know-how). At the same time, there was a positive selection of particularly bright and colorful raw materials at Feuersteinacker. Ethnographic analogies (e.g. Taçon, 2008) suggest that in addition to functional aspects, the raw materials could have had a symbolic meaning, in a similar way as it has been proposed for the heat-treatment of chert in Southern Germany (Eriksen, 2006; Hess, 2019b). As a consequence of the abundance of specific raw materials, in particular silicified sandstone and flint, microliths are generally larger than in the south of the study area. Laminar blanks have a more regular and standardised appearance compared to sites of a similar age (see Figure 7 and Table 5).

The archaeological record of the Early Mesolithic in the western part of Central Germany is characterized by the occurrence of large open-air sites that yielded thousands of lithic artifacts, made of a variety of different raw materials. Based on an interregional comparison, they can be interpreted as repeatedly visited workshops for the production of stone tools as part of a system of logistical mobility (cf. Binford, 1980; Kelly, 1983). Similar sites are also known from Lower Saxony (Fabesch, 1986) and Northern Bavaria (Lauerbach et al., 1997). A common phenomenon in this context is the presence of elongated pebbles that were used as retouching tools. As those are ergonomically shaped and seem to be adapted to the individual needs of the knappers, it is possible to suggest a certain degree of division of labor and a production of stone tools in a serial way. Lithic assemblages discovered at the mentioned sites display a high proportion of microliths and blanks and a relatively low number of recycled objects.

In general, it is possible to state that larger watercourses and low mountain ranges served as axes of communication. The river systems of Schwalm, Lahn, Ohm, Fulda, Neckar, Main, and Rhine linked different cultural areas and allowed people to integrate essential resources into their annual life cycles (in this context see Floss, 2002; Spies, 2020). Besides outcrops of lithic raw materials, the spatial proximity to salt springs that commonly occur within Hesse could have been a parameter for the selection of sites (cf. Hofbauer, 1992). In a terrain with otherwise rather low elevations, mountain ranges such as the Vogelsberg or the Rhön Mountains formed important landmarks. Additionally, they provided potential camp sites near rock shelters. Large open-air sites are frequently situated on southeast- or southwest-facing gentle slopes in a slightly elevated position near smaller streams that flow into tributary waters of larger rivers. Besides being exposed to sunlight and protected from seasonal floods, the topographic features allowed people to approach their prey without being noticed, as originally suggested by Taute (1971, p. 200).

The landscape during the early Holocene was characterised by large meandering rivers and marshlands in the valley bottoms, while mixed open-forests dominated by hazel were typical for higher zones on alluvial terraces or the plateaus of low mountain ranges (Bos & Urz, 2003). Besides charred shells of hazelnuts (Corylus), remains of water lily (Nymphaea), and guelder rose (Viburnum) were identified at sites dating to the late Preboreal. Palynological data suggest that there was an increased human impact at the beginning of the Boreal, which might have involved intentional clearance by fire-setting (Bos & Urz, 2003, pp. 32–33). This practice is known from an ethnographic context and was on the one hand aimed at creating open spaces that attract herbivores (Mellars, 1976). On the other hand, it was used to maintain paths in a more densely wooded landscape and to contribute to the spread of hazel or oak (Davies et al., 2005; Mason, 2000; Moore, 2003). During the Boreal, a variety of edible plants, such as acorns (Quercus), rose hips (Rosa), elderberries (Sambucus), dewberries (Rubus caesius), crabapples (Malus), dogwood (Cornus sanguinea), and blackthorn (Prunus spinosa), were available. Faunal assemblages typically include red deer (Cervus elaphus), roe deer (Capreolus capreolus), wild boar (Sus scrofa), in combination with aurochs (Bos primigenius), or bison (Bison sp.) (Bos & Urz, 2003, p. 31). Additionally, the proximity to watercourses implies that aquatic resources have contributed to the subsistence of Early Mesolithic groups (cf. Jochim, 2020).

Similar as in case of Early Mesolithic communities in Southwestern Germany that were adapted to karst formations within the Jura mountain range (Hess, 2019b), geomorphological peculiarities within the Central Uplands, such as rivers, extinct volcanoes, and large sandstone formations facilitated migration between different regions. Outcrops of lithic raw materials that formed during the Tertiary (such as silicified sandstone and chalcedony), are also present in parts of North Rhine-Westphalia as well as the Ardennes in Belgium and Luxembourg (Floss, 1994; Leesch, 2017, pp. 59–79). Therefore, subsistence strategies and technological systems based on the mentioned natural parameters could have been easily transferred to other regions. In this sense, it is possible to talk about a form of preadaptation (cf. Burmeister, 2000) in the context of prehistoric mobility patterns based on geomorphological similarities, which are in turn linked with the composition of flora and fauna (see Hess, 2019a). While the predominance of regionally available raw materials in smaller sites suggests that low mountain ranges could be a useful indicator for the definition of Early Mesolithic territories, rocks that were imported over distances of more than 200 km show that the mentioned communities were evidently part of larger networks.

To sum up, it is possible to state that Early Mesolithic (Beuronian A) sites in the western part of Central Germany are characterised by the presence of microliths with oblique truncation, isosceles triangles, and crescents. Another common feature is elongated pebbles that served as retouching tools. Imports of lithic raw materials over distances of more than 150 km suggest a far-reaching contact networks and movements along river systems and low mountain ranges. The most important raw materials in the western part of Central Germany are silicified sandstone, chalcedony, and siliceous shale. Large open-air sites that yielded thousands of stone tools made of imported raw materials, argue for specialized workshops and a serial production of microliths. This implies a certain degree of division of labor and a system of logistical mobility. There are strong similarities concerning the material culture between the study area and sites in Bavaria and Baden-Württemberg. This is also evident by the abundance of artifacts made of non-local Jurassic and Triassic chert, which links the area with the south. At the same time, the presence of Cretaceous flint shows contacts to Mesolithic groups in the north.

The open-air site Feuersteinacker near Stumpertenrod provides an almost unique example of longue durée concerning the (re)use of lithic raw materials over ten thousands of years. As it was an important hub in the settlement system of people during the early Holocene, it forms a milestone for the reconstruction of mobility patterns. Future research could focus on GIS-based models, integrating routes between outcrops of lithic raw materials and known locations, in order to discover new archaeological sites dating to the Mesolithic and to excavate them with modern methods (including micromorphology), in combination with a reconstruction of the paleoenvironment. This prevents archaeological features from being destroyed and leads to a more detailed understanding of the chrono-cultural sequence in the area.

Acknowledgements

I would like to thank Prof. Dr. Felix Riede (University of Aarhus) for his support and the opportunity to work together in the framework of a postdoctoral fellowship. Furthermore, I wish to thank Dr. Katharina Weick-Joch, Linda Heintze, and Linn Merten (Upper Hessian Museum Gießen) for the possibility to analyse the lithic finds from Stumpertenrod and Joachim Knossalla for taking photographs of groundstone tools. Furthermore, I would like to express my gratitude to Dr. Irina Görner (Hessian State Museum in Kassel) for giving me access to the museum’s collection of the Hessian State Museum in Kassel, her scientific advice, and for providing me with additional samples. I would also like to thank Rógvi Johansen and Louise Hilmar (both Moesgaard Museum) for taking photographs and creating drawings of lithic artifacts. Finally, I am grateful for the hospitality of the Department of Archaeology and Heritage Studies of the University of Aarhus and the interest in my research.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available from the corresponding author, [T. H.], upon reasonable request.

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

The project was supported by the Swiss National Science Foundation (SNSF) under Grant No. P2SKP1_184038. This particular study was funded by the Independent Research Fund Denmark (DFF) under Grant No. 6107-00059B as well as the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (Grant No. 817564).

Notes on contributors

Thomas Hess

Thomas Hess has studied Prehistoric Archaeology, Earth Sciences, and Social Anthropology at the Universities of Zurich and Cambridge. In 2016 he completed his PhD at the University of Tübingen on the lithic artifacts from a rock shelter site on the Swabian Jura. Currently he is conducting research on Mesolithic and Late Paleolithic cultures in Central Germany and adjacent regions.

Thomas Hess conceptualized the study, conducted the lithic analyses, created the figures, and wrote the paper.