Long thin blade production and Late Gravettian hunter-gatherer mobility in Eastern Central Europe

Quaternary International 406 (2016) 166e173

Contents lists available at ScienceDirect

Quaternary International journal homepage: www.elsevier.com/locate/quaint

Long thin blade production and Late Gravettian hunter-gatherer mobility in Eastern Central Europe € rgy Lengyel a, *, Wei Chu b Gyo a b

ros, Building B/2, Hungary University of Miskolc, Institute of History, Department of Prehistory and Archaeology, 3515 MiskolceEgyetemva €ln, Germany University of Cologne, Institute of Prehistoric Archaeology, Weyertal 125, 50923 Ko

a r t i c l e i n f o

a b s t r a c t

Article history: Available online 10 February 2016

The regular and systematic production of long blades (>120 mm) that maintain a thickness (<10 mm) of regular blades (<120 mm) is a particular phenomenon of the Upper Palaeolithic (40e10 ka BP) archaeological record of Eastern Central Europe. However, the mechanical underpinnings of manufacturing these long blades are still not fully understood. This paper presents experimental research that used heavy (~800 g) and light (~570 g) antler percussors to test the effect of percussor weight on the manufacture of Upper Palaeolithic type blades. Statistical analyses showed that a heavier percussor effectively increased the ability to achieve a fine thickness for long blades. The results are then compared to other experimental and archaeological assemblages to suggest that the use of heavy percussors may have played a role in lithic economy of Upper Palaeolithic mobile hunter-gatherers. © 2016 Elsevier Ltd and INQUA. All rights reserved.

Keywords: Knapping experiment Long blade technology Antler percussor Upper Palaeolithic

1. Introduction In prehistoric technology, soft hammer percussion techniques were common in the Upper Palaeolithic (UP) (40e10 ka BP) when modern humans used them to produce long, narrow, and flat removals commonly referred to as blades (Bar-Yosef and Kuhn, 1999; Pelegrin, 2011). Variations in blade technologies are most striking through the metric attributes of the products (Nigst, 2012; Zwyns, 2012). In the UP of Eastern Central Europe, hunter-gatherers rarely manufactured blades longer than 120 mm (Hahn, 1977, 1988; Otte, 1981;  ski, 2007; Moreau, 2009; Sobczyk, 1996; Otte et al., 2007; Wilczyn Noiret, 2009; Nigst, 2012). According to the regional archaeological record, a systematic long blade (>120 mm) production first appeared in the Late Gravettian (24e21 ka BP) at Willendorf II Layer 9 (WII9), where a collection of long blades was recovered (Felgenhauer, 1956e1959; Haesaerts et al., 1996). Long blade technology became more prevalent later in the Western European Middle and Upper Magdalenian (15e12 ka BP) where it remained in use until the end of the Palaeolithic (Adouze et al., 1981; Pigeot, 1987; Fouches and San Juan, 1991; Karlin et al., 1993; Valentin,

* Corresponding author. E-mail addresses: [email protected] (G. Lengyel), [email protected] (W. Chu). http://dx.doi.org/10.1016/j.quaint.2016.01.020 1040-6182/© 2016 Elsevier Ltd and INQUA. All rights reserved.

1995; Bodu et al., 1997, 2006; Angevin and Langlais, 2009; Janny, 2010; Caron-Laviolette, 2013). The systematic production of long blades (>120 mm) is uncommon in UP blade assemblages (Hahn, 1977, 1988; Otte, 1981;  ski, 2007; Moreau, 2009; Sobczyk, 1996; Otte et al., 2007; Wilczyn Noiret, 2009; Nigst, 2012), so they do not represent the usual blade type for UP hunter-gatherers. Manufacturing long blades requires high-level technical skills to manufacture the blade as a single unbroken piece (Bodu et al., 1990; Bodu, 1993; Karlin et al., 1993). Long blade technology is thought to be related to highly mobile Western European groups, such as those found during the Magdalenian when lithic tools were commonly transported from one site to another across a broad ecological niche (Angevin and Langlais, 2009). There, long blade production may have played a chief role in the organization of mobile technology (Debout et al., 2012). Compared to regular size blade production, long blade technology requires increased percussive force (Crabtree, 1972; Dibble and Whittaker, 1981; Tixier, 1982). This is achieved either by increased percussor velocity (Dibble and Pelcin, 1995), or by greater percussor weight (Crabtree, 1972; Dibble and Whittaker, 1981; Tixier, 1982; Whittaker, 1994). Controlled experiments (Dibble and Whittaker, 1981) suggest that increased percussive force can simultaneously increase blade thickness. However, a striking feature of the Central European Late Gravettian long blade

G. Lengyel, W. Chu / Quaternary International 406 (2016) 166e173

167

technology was the ability to retain the thickness (<10 mm) of the regular blades (Felgenhauer, 1956e1959). This paper presents the results of a knapping experiment that explores the maintenance of regular blade thickness for long blades using soft organic percussors. The results directly shed light on a technological efficiency that may have aided hunter-gatherers to cope with the logistical challenges of their high degree of mobility.

2. Materials and methods Experiments on lithic tool production follow two different methodologies. One, authentic experiments, use genuine materials and the skills of modern knappers (Crabtree, 1972; Pelegrin, 1991; Whittaker, 1994; Eren et al., 2008, 2014; Eren and Lycett, 2012; Tixier, 2012; Lycett and Eren, 2013; Driscoll and García-Rojas, 2014). Others, controlled experiments, substitute humans for apparatus and chert for modern siliceous materials isolating individual dependent variables while attempting to keep other controlled variables constant (Speth, 1972; Dibble and Whittaker, 1981; Dibble and Pelcin, 1995; Dibble, 1997; Pelcin, 1997a, 1997b, 1997c; Magnani et al., 2014). Here, an authentic experiment was conducted to reproduce UP blade technology, with the aim to accommodate a fuller range of variables that influence knapping that cannot be replicated with machines such as strike accuracy, lithic inhomogeneity, and knapper error (Whittaker, 1994; Inizan et al., 1999; Pelegrin, 2000, 2011). According to the technical skills in the UP (Karlin et al., 1993) the experimenter (G. L.) was a medium level knapper that able to produce 10e20 cm long blades. Because inhomogeneity in siliceous rocks can negatively affect knapping success (Crabtree, 1972; Roche and Tixier, 1982; Andrefsky, 1994; Tixier, 2012; Lengyel, 2013), fine-grained isotropic chert was chosen as the raw material in this experiment. Chert nodules were taken from the Lithic Raw Material €tvo €s Lora nd University of Budapest in Reference Database of Eo Hungary (Mester et al., 2012), which had been collected at the outcrops of Kremenec, Ternopil, Western Ukraine (Mester and , 2013). Farago In the WII9 archaeological long blade collection, most specimens were no longer than 150 mm except for two specimens of 174 mm and 180 mm in length (Felgenhauer, 1956e1959). Thus this experiment aimed to produce a comparable sample of 60e150 mm blades to study changes in blade thickness formation through the transition from regular to long blade, while manipulating percussive force by increasing the weight of percussor and the percussor velocity. Two sets of blades were produced from fourteen cores with two percussors of different weights. Antlers were used as percussors (Fig. 1) as they are widely thought to have been used in UP blade technology (Pelegrin, 2011). The appropriate weight for the percussors was estimated from previous authentic experiments (Tixier, 1982; Eren et al., 2008) as well as from our own experiences. Both suggested that a percussor best able to produce a blade of the length studied here is between 500e900 g. Thus, the light percussor (LP) was a 570 g North American moose antler. To obtain blades >120 mm with the LP, percussive force needed to be increased. This was achieved by increasing percussor velocity, which we were unable to measure as in controlled experiments (Dibble and Pelcin, 1995), and the increment was based on the experimenter's estimation. The heavy percussor (HP), 802 g, was an Eastern European red deer antler that was able to remove blades between 120 and 150 mm without increasing force by accelerating the percussor as it was necessary with LP. The percussors heads were prepared by removing the

Fig. 1. Percussors used in this experiment: (a) light, moose antler; (b) heavy, red deer antler (scale is in cm).

antler burrs and the percussor faces were rasped into a convex surfaces. Knapping was performed seated in a chair. Cores were held in the left hand with the flaking surface in the palm and the left elbow resting on the left thigh. The percussor was held in the right hand and the elbow was supported with the right thigh (Fig. 2). Supporting the elbows with the thigh enabled a constant trajectory to precisely hit the edge of the core's striking platform. This setting provided a comfortable position that frequently resulted in successful blade production.

Fig. 2. The position of knapping holding the heavy percussor.

168

G. Lengyel, W. Chu / Quaternary International 406 (2016) 166e173

Fig. 3. Diagram of a core's striking platform outline (a), a core used for the experiment at the beginning of knapping (b). (Scales are cm. Please note the difference in size between the two views.)

Each core had a single striking platform and a ~200 mm long flaking surface. The outline of the flaking surface, from the view of the striking platform, followed a radial line (Fig. 3), and the flanks of the cores converged. The striking platforms were constantly renewed to keep an optimal angle between the flaking surface and the striking platform. The experiment aimed to keep an acute angle as close to 90
as possible between the flaking surface and the striking platform, because the greater the angle (upper limit at 90
), the longer the removal (Dibble and Whittaker, 1981). Before each removal, the edge of the platform where the strike was to be delivered was isolated, lightly abraded, but remained unfaceted. The edge of the core platform was targeted as the principle of soft percussion technique to increase removal length while simultaneously decreasing platform thickness that was expected to minimize blade thickness (Crabtree, 1972; Dibble and Whittaker, 1981; Dibble, 1997; Pelcin, 1997c). Length (L), maximum thickness (MT), and platform thickness (PLT) were measured for each blade following Inizan et al. (1999) to two digits of precision in millimeters. These measurements fit the most common method of recording the length and thickness of blades in the archaeological literature (Inizan et al., 1999; Andrefsky, 2005). PLT was a dependent variable that we chose to represent striking accuracy on the core edge. The smaller the value of PLT, the closer the impact was to the core's platform edge. We excluded crested blades and overpassed blades from the analyses, the maximum thickness of which was often greater than that of regular blades. Pearson's correlation, tetest, and multiple linear regression models were used to compare the two datasets to predict the effects of the percussor weight, velocity, and accuracy in striking the core platform edge. Additional tests were run by breaking the samples into three ordinal length categories: 60e90 mm, 90e120 mm, and 120e150 mm. This aimed to find the blade length at which the difference in blade thickness began to significantly differ by the use of two different percussors. The division was arbitrary along 30 mm increments and the 120 mm threshold was regarded here as the border between regular and long blades. A one-way ANOVA was then used to compare experimental and archaeological samples.

Data of blade length and thickness from other knapping experiments where the weight and the material of the percussor was known was also available for comparison. Eren et al. (2008) obtained blades using a 611.5 g boxwood percussor. Of these blades, all specimens between 60 and 150 mm in length (N ¼ 161, M ¼ 83.33, SD ¼ 16.93), except crested blades were used for comparison. The blades from the WII9 assemblage were also used to compare the experimental results with archaeological data (data in Felgenhauer, 1956e1959). From here, besides blank specimens, we included edge retouched blades as well, where the length and maximum thickness remained unmodified by retouching. WII9 sample consisted of blades between 78 and 142 mm length (N ¼ 28, M ¼ 109.18, SD ¼ 19.99). Another, Upper Magdalenian long blade collection from Pincevent, France (Bodu, 1993), was also included that consisted of blank specimens between 60 and 150 mm length (N ¼ 41, M ¼ 101.34, SD ¼ 29.62). 3. Results Knapping resulted in 227 complete blades (Fig. 4; Table 1). According to the mean PLT differences between the percussors, t(198.293) ¼ 3.733, p ¼ 0.000 (equal variances not assumed), HP produced significantly thinner platforms, suggesting greater accuracy in striking the core edge with the percussor. L/MT scatterplots showed HP blade MT values were smaller than LP blades, especially above 120 mm length (Fig. 5). A significant negative correlation was found between percussor weight and blade thickness which was also supported by visual observation, r ¼ 0.379, p ¼ 0.000. As a result, the MTs also differed significantly, t(194.649) ¼ 6.088, p ¼ 0.000. PLT also significantly correlated with MT, r ¼ 0.507, p ¼ 0.000. The multiple regression model (R2 ¼ 0.326, F(2, 224) ¼ 54.176, p ¼ 0.000) indicated that PLT significantly predicted MT (b ¼ 0.441, p ¼ 0.000), as did percussor weight inversely (b ¼ 0.271, p ¼ 0.000). Running the tests by the threefold length groups showed a coefficient increase towards the longest specimens, indicating a greater effect on MT for long blades (Tables 2e4).

G. Lengyel, W. Chu / Quaternary International 406 (2016) 166e173 Table 1 Measurements of the experimental blades produced in this experiment. Length [mm]

Percussor

L

MT

PLT

60e90

LP

77.3583 40 7.44237 77.8841 44 8.27792 104.5158 55 8.64956 101.9987 52 8.76322 129.9863 16 6.48205 129.9760 20 7.65709 98.8954 111 20.26904 97.6757 116 20.22700

6.4248 40 2.22718 5.1911 44 1.47095 8.6056 55 1.86216 6.6798 52 1.80235 11.1938 16 2.99716 7.8610 20 1.20367 8.2443 111 2.72942 6.3188 116 1.85691

2.3723 40 1.29021 2.0909 44 0.99891 2.8925 55 1.24312 2.1167 52 0.85369 3.2694 16 1.48434 2.5845 20 0.99012 2.7623 111 1.31745 2.1876 116 0.94420

M N SD M N SD M N SD M N SD M N SD M N SD M N SD M N SD

HP

90e120

LP

HP

120e150

LP

HP

Total

LP

HP

Table 2 Pearson correlations between MT and percussor type, and MT and PLT by the length categories. Length [mm]

MT/Percussor

Comparing our data with other experimental data, according to the ANOVA and the post-hoc tests, Eren et al. (2008) blade thickness values (M ¼ 6.161, SD ¼ 2.66) were similar to HP blades and dissimilar to LP blades (Table 5B). By length categories, however, Eren et al. (2008) blade thickness significantly differed from HP between 120 and 150 mm (Table 6B).

Table 5 A) Oneeway ANOVA of the experimental blades and Eren et al., 2008, Willendorf II/ 9, and Pincevent; B) Tukey postehoc test. A.

Between Groups Within Groups Total

(I) Percussor LP

HP

Eren et al. (2008)

a

60e90 90e120 120e150 a

p

r

p

n

0.316a 0.468a 0.615a

0.003 0.000 0.000

0.441a 0.485a 0.561a

0.000 0.000 0.000

84 107 36

Correlation is significant at the 0.01 level (2etailed).

Sum of squares

df

Mean square

F

p

329.748 2779.007 3108.755

4 449 453

82.437 6.189

13.319

0.000

(J) Percussor

Mean difference (IeJ)

SD

p

0.33033 0.46759 0.52612 0.30692 0.33033 0.46501 0.52384 0.30298 0.30692 0.30298 0.44869 0.50940

0.000 0.005 0.834 0.000 0.000 0.987 0.086 0.986 0.000 0.986 0.907 0.031

B.

MT/PLT

r

169

HP Pincevent Willendorf Eren et al. (2008) LP Pincevent Willendorf Eren et al. (2008) LP HP Pincevent Willendorf

a

1.87400 1.64016a 0.54994 2.03099a 1.87400a 0.23384 1.32406 0.15699 2.03099a 0.15699 0.39083 1.48106a

The mean difference is significant at the 0.05 level.

The WII9 blade thicknesses (M ¼ 7.64, SD ¼ 1.89) did not differ from either the LP or HP (Table 5B), while by length categories, there was a significant difference between LP blades but not in HP between 120 and 150 mm (Table 6B). Pincevent blade thickness (M ¼ 6.64, SD ¼ 3.10) differed from

Table 3 ANOVA in the multiple linear regression model for predicting MT from percussor type and PLT. Length [mm] 60e90

90e120

120e150

Sum of squares Regression Residual Total Correlation Regression Residual Total Correlation Regression Residual Total Correlation

df

60e90

90e120

120e150

(Constant) PLT Percussor (Constant) PLT Percussor (Constant) PLT Percussor

F

p

84.188 234.189 318.377

2 81 83

42.094 2.891

14.559

0.000

152.953 299.105 452.058

2 104 106

76.476 2.876

26.591

0.000

142.255 118.748 261.003

2 33 35

71.128 3.598

19.766

0.000

Table 4 Multiple linear regression model for predicting MT from percussor type and PLT. Length [mm]

Mean square

B

SD

5.810 0.696 1.038 8.081 0.668 1.408 10.898 0.918 2.704

0.737 0.164 0.374 0.765 0.154 0.349 1.473 0.264 0.661

b 0.408 0.266 0.367 0.342 0.425 0.499

t

p

7.880 4.253 2.772 10.562 4.326 4.031 7.401 3.478 4.088

0.000 0.000 0.007 0.000 0.000 0.000 0.000 0.001 0.000

R

R2

0.514

0.264

0.582

0.338

0.738

0.545

the LP but not from the HP (Table 5B). By length categories, the thickness of Pincevent blades did not differ from either HP or LP between 60 and 90 mm (Table 6B). From 90 to 150 mm, a significant difference in MT was found comparing with the LP, while the HP blade thickness did not differ from the Pincevent sample in this length category. Using the sample of Eren et al. (2008) as another experimental set of blades produced with a percussor of comparable weight to LP, Pincevent did not show a significant difference while WII9 did (Table 5B). By length categories, Eren et al. (2008) blades differed only between 120 and 160 mm from the archaeological samples by being thicker (Table 6B). This comparison seems to fit the results of the present experiment, but it should be noted that Eren

170

G. Lengyel, W. Chu / Quaternary International 406 (2016) 166e173

Table 6 A) Oneeway ANOVA of the experimental blades and Eren et al. (2008), Willendorf II/9, and Pincevent by the three length categories. B) Tukey postehoc test. A. Length [mm] 60e90

90e120

120e150

Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total

Sum of squares

df

Mean square

F

p

59.872 892.862 952.734 123.456 839.334 962.790 191.778 349.162 540.940

4 211 215 4 175 179 4 53 57

14.968 4.232

3.537

0.008

30.864 4.796

6.435

0.000

47.945 6.588

7.278

0.000

B. Length [mm]

(I) Percussor

(J) Percussor

Mean difference (IeJ)

SD

p

60e90

LP

HP Pincevent Willendorf Eren et al. (2008) LP Pincevent Willendorf Eren et al. (2008) LP HP Pincevent Willendorf HP Pincevent Willendorf Eren et al. (2008) LP Pincevent Willendorf Eren et al. (2008) LP HP Pincevent Willendorf HP Pincevent Willendorf Eren et al. (2008) LP Pincevent Willendorf Eren et al. (2008) LP HP Pincevent Willendorf

1.23361 0.36593 1.07525 0.95649 1.23361 0.86769 2.30886 0.27712 0.95649 0.27712 0.59057 2.03174 1.92583a 2.33291a 1.10564 1.44105a 1.92583a 0.40708 0.82019 0.48478 1.44105a 0.48478 0.89186 0.33542 3.33275a 3.49375a 3.19375a 1.83375 3.33275a 0.16100 0.13900 5.16650a 1.83375 5.16650a 5.32750a 5.02750a

0.44940 0.59557 0.90058 0.38028 0.44940 0.58744 0.89523 0.36742 0.38028 0.36742 0.53641 0.86260 0.42360 0.72334 0.65558 0.43258 0.42360 0.72681 0.65941 0.43836 0.43258 0.43836 0.73208 0.66521 0.86090 1.03467 1.11142 1.43483 0.86090 0.99408 1.07373 1.40584 1.43483 1.40584 1.51848 1.57178

0.051 0.973 0.755 0.091 0.051 0.579 0.078 0.943 0.091 0.943 0.806 0.132 0.000 0.013 0.445 0.009 0.000 0.981 0.726 0.803 0.009 0.803 0.741 0.987 0.003 0.012 0.044 0.706 0.003 1.000 1.000 0.005 0.706 0.005 0.008 0.019

HP

Eren et al. (2008)

90e120

LP

HP

Eren et al. (2008)

120e150

LP

HP

Eren et al. (2008)

a

The mean difference is significant at the 0.05 level.

et al. (2008) used a percussor of a different material and the experimental settings did not explicitly focus on knapping long blades. Based on the statistical evaluations, it was possible to predict the percussion technique for blades longer than 120 mm by MT. A heavy percussor with great accuracy can be predicted ~55% of the variance within the population of blades 120e150 mm (R2 ¼ 0.545) (Table 3). In the experimental sample, however, there was no long blade (>120 mm) obtained with LP thinner than 7.90 mm (Fig. 5). We assume that the similarity between the MTs of the archaeological samples and the HP detached blades indicates that the successful removal of a long (>120 mm) thin (<7.90 mm) blade was most likely achieved with the use of a percussor weighing ~800 g and striking accurately the very edge of the blade core's striking platform. The results of the statistical tests showed that increased percussion force through greater percussor velocity produced thicker

blades and platforms. The heavy percussor facilitated striking accuracy near the edge of the core platform. Long thin blades were best achieved when the heavy percussor managed to strike closer to the edge of the core's platform. Therefore, the advantage of using a heavy percussor in long thin blade production can be postulated, which in turn relates with percussion accuracy that can be maintained with the lower velocity of the swinging arm. 4. Discussion According to the results, a heavy percussor produced thinner specimens than a lighter one in long blade production. The relationship between the heavy percussor and blade thinness most likely relies on strike accuracy that removes the blade. Achieving this accuracy seemed to also depend on the velocity of the swinging arm that tended to miss the expected point of impact when the swing is accelerated up to a speed under which there is less control

G. Lengyel, W. Chu / Quaternary International 406 (2016) 166e173

171

Fig. 4. Samples of blades from the experiment: (upper row) heavy, and (lower row) light percussor detached specimens (scale is in cm).

over the trajectory of the percussor in hand. Speeding up percussor velocity in turn was necessary for the lighter percussor to remove long blades. Meanwhile, the heavy percussor weight generated sufficient force with a controlled trajectory to obtain the same blade length, but significantly thinner. Missing the intended point of impact with the light percussor usually resulted in a hit along a radial line closer to the center of the core's striking platform instead of the edge. This usually resulted in a thicker blade platform. The experimental long blade sample obtained with the heavy percussor showed a significant similarity in thickness with the Late Gravettian and Magdalenian archaeological assemblages. Assuming that UP hunter-gatherers tended to maximize or at least balance the rate of energy gained over their investment in foraging for food (Kelly, 2013; Bettinger et al., 2015), the technological investment in

Fig. 5. Scatterplot of length (L) and maximum thickness (MT) by percussors in mm.

making thin long blades may have had its own benefits as well. Regarding the Magdalenian, where long blade production was common, long thin blade production has been related to high mobility and shorter occupation periods (Angevin and Langlais, 2009; Debout et al., 2012). Assembling a mobile tool kit requires general rather than specialized tool types (Shott, 1986). While bifacial tools have been seen as elements of multifunctional mobile tool kits (Kelly, 1988), blades can also occupy a similar niche because most UP tool types can be produced from blades (Demars and Laurent, 1992). While small artifacts can be efficient as mobile tools (Kuhn, 1994), large stone tools can also accomplish the same task as they can be re-sharpened and repaired more times, and can circumvent additional labor investment in making hafts and handles that are often required for small tools efficiently (Morrow, 1996). Constant re-sharpening, resulting in intensive lithic reduction, has been shown to be a common option for maintaining tool kits during periods of high mobility (Parry and Kelly, 1987; Hiscock, 1996). Concerning thinness, a theoretical benefit of a thin long blade is that when removals are thinner, a greater the number of artifacts can be produced from an equivalent unit of lithic raw material. This advantage can also play a role in hunter-gatherer technological organization where raw material economy and tool-size are simultaneously important. In hunter-gatherer societies, raw material parsimony may have been strongly related to mobility where ad hoc tasks were frequent while lithic raw materials resources were unpredictable along the landscape (Binford, 1979; Bamforth, 1986). Meanwhile, optimizing transport logistics was likely a crucial concern in dealing with increased mobility (Kuhn, 1994; Brantingham, 2003). In the Eastern Central Europe, there is the potential for UP blot-Augustins, 1999), mobility patterns to be easily detectable (Fe as the tools used in mobile tool kits can easily be identified by the exogenous component of the lithic assemblage since raw materials are highly heterogeneous and regionally specific (Prichystal, 2013). In the Late Gravettian record, lithic raw materials from southern Poland are frequent in Moravia (Czech Republic), Lower Austria, and the Carpathian Basin indicating a mobility range of Late Gravettian hunter-gatherers around the Western Carpathians over blot-Augustins, 1997; ~80,000 square kilometers (Otte, 1981; Fe Kozłowski, 1998, 2000, 2013; Kaminsk a, 2001; Lengyel, 2015). The exogenous raw material component of the assemblages can be

172

G. Lengyel, W. Chu / Quaternary International 406 (2016) 166e173

highly dominated by blades, and the ratio of tools within the exogenous blades is significantly greater than that among blades made from local raw materials (Lengyel, 2015). This suggests that blades were an important part of a mobile tool kit. At the Late Gravettian site of WII9 in Lower Austria, where local raw materials available from the Danube gravel lacked the appropriate size and quality to produce long blades (Nigst, 2012; Brandl et al., 2014), the long thin blades were made of high quality flint most likely originating ~270 km to the north in glacial moraines of Silesia (Otte, blot-Augustins, 1997). Thus, a relationship between hunt1981; Fe er-gatherer mobility and long thin blade technology can be supposed, similar to what has been pointed out for Magdalenian hunters. This begs the question; can the absence of long blade technology be an indicator of lower mobility? Because mobile tool kits can be extensively reduced, it is difficult to estimate the original blank size of mobile tools. Therefore, we cannot link long blade absence to low mobility. The case of WII9 assemblage may be a unique archaeological phenomenon with a good preservation of long blades. 5. Conclusion Long blades are a unique feature of the Eastern Central European Upper Palaeolithic archaeological record. However, there has been little attempt to understand how or why they were manufactured. Here, experimental knapping has shown that a larger soft hammer produced significantly thinner and longer blades than a similar soft hammer percussor by increasing both percussion accuracy and force. When compared to an additional experimental assemblage and two archeological assemblages, they suggest that large percussors may have been a parsimonious technical solution to creating long blades. The experimental tests should be repeated and more experimental comparisons should be conducted that explore a variety of different percussor weights. There are of course numerous factors that interact in complex ways to influence blade length production such as percussor material and morphology, percussor speed and raw material not to mention that highly skilled knappers were likely able to maintain high accuracy even with a lighter percussor in obtaining long thin blades. However so far, our experimental results, when considered within the archaeological context of Eastern Central Europe, suggest that a heavier percussors may have been one way to facilitate the production of long thin blades which in turn may have been part of the answer to technological and logistic demands in organizing a mobile tool kit by Late Gravettian knappers. Acknowledgements The European Social Fund in the framework of Hungarian  TAMOP 4.2.4. A/2e11e1e2012e0001 project (A2eMZPDe13e0181), the Campus Hungary Program (B2/4H/ 17968), and the DFG Funded project SFB 806 Our Way to Europe supported this research. References Andrefsky, W., 1994. The geological occurrence of lithic material and stone tool production strategies. Geoarchaeology 9, 375e391. Andrefsky, W., 2005. Lithics Macroscopic Approaches to Analysis, second ed. Cambridge University Press, New York. ^te sur les «caches» et Angevin, R., Langlais, M., 2009. Où sont les lames? Enque po ^ts» de lames du Magdale nien moyen (15000e13500 BP). In: «de riel au spiriBonnardin, S., Hamon, C., Lauwers, M., Quilliec, B. (Eds.), Du mate alite s arche ologiques et historiques des « De po ^ts » de la pre histoire  tuel. Re a ologie et d'histoire d'Antinos jours. XXIXe rencontres internationales d'arche  bes. Editions APDCA, Antibes, pp. 223e242.

nien du Audouze, F., Cahen, D., Keeley, L.H., Schmider, B., 1981. Le site Magdale  Verberie (Oise). Gallia Pre histoire 24, 99e143. Buisson Campin a Bamforth, D.B., 1986. Technological efficiency and tool curation. American Antiquity 51, 38e50. Bar-Yosef, O., Kuhn, S., 1999. The big deal about blades: laminar technologies and human evolution. American Anthropologist 101, 322e338. Bettinger, R.L., Garvey, R., Tushingham, S., 2015. Hunter-Gatherers: Archaeological and Evolutionary Theory. Interdisciplinary Contributions to Archaeology. Springer, New York. Binford, L.R., 1979. Organization and formation processes: looking at curated technologies. Journal of Anthropological Research 35, 255e273. riel lithique de quelques unite s Bodu, P., 1993. Analyse typoetechnologique du mate nien de Pincevent (SeineeeteMarne) (Ph.D. thesis). In: Applidu site magdale conomiques et sociales, vol. 1e3. University of Paris I, Paris. cations spatiales, e https://tel.archives-ouvertes.fr/tel-00577069. Bodu, P., Karlin, C., Ploux, S., 1990. Who's who? The Magdalenian flintknappers of Pincevent (France). In: Cziesla, E., Eickhoff, S., Arts, N., Winter, D. (Eds.), The Big Puzzle: International Symposium on Refitting Stone Artifacts, Studies in Modern Archaeology 1. Holos, Bonn, pp. 143e163. Bodu, P., Hantaï, A., Valentin, B., 1997. La Long Blade Technology au sud du Bassin couvertes re centes. In: Fagnart, J.P., The venin, A. (Eds.), Le TardiParisien: de s National des Socie te s glaciaire en Europe du nordeouest. Actes du 119e Congre Historiques et Scientifiques, Amiens, 1994, Paris, pp. 211e222.  Pincevent: les Bodu, P., Julien, M., Valentin, B., Debout, G., 2006. Un dernier hiver a niens du niveau IV0 (Pincevent, La Grande Paroisse, SeineeeteMarne). Magdale histoire 48, 1e180. Gallia Pre Brandl, M., Hauzenberger, C., Postl, W., Martinez, M.M., Filzmoser, P., Trnka, G., 2014. Radiolarite studies at Krems-Wachtberg (Lower Austria): northern alpine versus Carpathian lithic resources. Quaternary International 351, 146e162. Brantingham, P.J., 2003. A neutral model of stone raw material procurement. American Antiquity 68, 487e509. CaroneLaviolette, E., 2013. Etiolles, rive gauche La Fontaine au Soulier Etude  d'occupation magdale nienne (MA thesis). Unitechnoespatiale d'une unite  Paris I e Panthe oneSorbonne. Me moire De Master 2, Paris. versite Crabtree, D.E., 1972. An Introduction to Flintworking. Occasional Papers of the Idaho State University Museum 28, Pocatello, Idaho. Debout, G., Olive, M., Bignon, O., Bodu, P., Chehmana, L., Valentin, B., 2012. The Magdalenian in the Paris Basin: new results. Quaternary International 272e273, 176e190. olithique supe rieur Demars, P.eY., Laurent, P., 1992. Types d'outils lithiques du Pale en Europe. Cahiers du Quaternaire No 7. Presses du CNRS, Paris. Dibble, H.L., 1997. Platform variability and flake morphology: a comparison of experimental and archaeological data and implications for interpreting prehistoric lithic technological strategies. Lithic Technology 22, 150e170. Dibble, H.L., Pelcin, A., 1995. The effect of hammer mass and velocity on flake mass. Journal of Archaeological Science 22, 429e439. Dibble, H.L., Whittaker, J.C., 1981. New experimental evidence on the relation between percussion flaking and flake variation. Journal of Archaeological Science 8, 283e296. Driscoll, K., GarcíaeRojas, M., 2014. Their lips are sealed: identifying hard stone, soft stone, and antler hammer direct percussion in Palaeolithic prismatic blade production. Journal of Archaeological Science 47, 134e141. Eren, M.I., Greenspan, A., Sampson, C.G., 2008. Are Upper Paleolithic blade cores more productive than Middle Paleolithic discoidal cores? A replication experiment. Journal of Human Evolution 55, 952e961. Eren, M.I., Lycett, S.J., 2012. Why Levallois? A morphometric comparison of experimental ‘preferential’ Levallois flakes versus debitage flakes. PloS One 7, e29273. Eren, M.I., Roos, C.I., Story, B.A., von Cramon-Taubadel, N., Lycett, S.J., 2014. The role of raw material differences in stone tool shape variation: an experimental assessment. Journal of Archaeological Science 49, 472e487. blot-Augustins, J., 1997. La Circulation des Matieres Premieres au Pale olithique. Fe es Perspectives Comportementales, 75. Tome I-II. In: Synthese des Donne ge. ERAUL, Lie blot-Augustins, J., 1999. La mobilite  des groupes Pale olithiques. Bulletins et Fe moires de la Socie te  d'anthropologie de Paris 11 (3e4), 219e260. Me Felgenhauer, F., 1956e1959. Willendorf in der Wachau. Monographie der €historischen Kommission der Pal€ aolitheFundstellen IeVII. Mitteilungen der Pra € Osterreichischen Akademie der Wissenschaften, Wien, pp. VIIIeIX. ^ts (Leigne -les-Bois, Vienne), atelier de taille Fouches, P., San Juan, C., 1991. Les Gene nien dans son contexte re gional. Revue Arche ologique du Centre de la Magdale France 30, 7e29. Haesaerts, P., Damblon, F., Bachner, M., Trnka, G., 1996. Revised stratigraphy and chronology of the Willendorf II sequence, Lower Austria. Archaeologia Austriaca 80, 25e42. € Hahn, J., 1977. Aurignacien. Das Altere Jungpal€ aolithikum in Mittele Und Osteuropa. €hlau Verlag In: Fundamenta Monographien zur Urgeschichte Reihe A, Band 9. Bo €ln Wien. Ko €sterle-Ho € hle im Achtal bei Blaubeuren I. FundhorHahn, J., 1988. Die Geißenklo €olithikum und im Aurignacien. In: izontbildung und Besiedlung im Mittelpala Forschungen und Berichte zur Vor- und Frühgeschichte in Baden-Württemberg, vol. 26. Kommissionsverlag, Konrad Theiss Verlag, Stuttgart. Hiscock, P., 1996. Mobility and technology in the Kakadu coastal wetlands. IndoPacific Prehistory Association Bulletin 15, 151e157. Inizan, M.L., ReduroneBallinger, M., Roche, H., Tixier, J., 1999. Technology and Terminology of Knapped Stone. CREP, Nanterre, Meudon.

G. Lengyel, W. Chu / Quaternary International 406 (2016) 166e173  Verberie. In: Zubrow, E., Adouze, F., Enloe, J.G. Janny, F., 2010. Technologie lithique a (Eds.), The Magdalenian Household. Unraveling Domesticity. State University of New York, pp. 51e75. €olitikum der Kaminsk a, L., 2001. Die nutzung von steinrohmaterialien im Pala Slowakei. Quart€ ar 51 (52), 81e106. Karlin, C., Ploux, S., Bodu, P., Pigeot, N., 1993. Some socio-economic aspects of the knapping process among groups of hunteregatherers in the Paris Basin area. In: Berthelet, A., Chavaillon, J. (Eds.), The Use of Tools by Human and Nonehuman Primates. Oxford University Press, Oxford, pp. 318e337. Kelly, R.L., 1988. The three sides of a biface. American Antiquity 53, 717e734. Kelly, L.R., 2013. The Lifeways of HuntereGatherers. The Foraging Spectrum. Cambridge University Press, Cambridge. Kozłowski, J.K. (Ed.), 1998. Complex of Upper Palaeolithic Sites Near Moravany, Western Slovakia. Vol. II: MoravanyeLopata II (Excavations 1993e1996).  w. Jagellonian University, Institute of Archaeology, Krako Kozłowski, J.K. (Ed.), 2000. Complex of Upper Palaeolithic Sites Near Moravany, Western Slovakia. Vol. III. Late Gravettian Shouldered Point Horizon Sites in the Moravany e Banka Area. Slovak Academy of Sciences, Jagellonian University, Nitra. Kozłowski, J.K., 2013. Raw materials procurement in the Late Gravettian of the Carpathian Basin. In: Mester, Z. (Ed.), The Lithic Raw Material Sources and Interregional Human Contacts in the Northern Carpathian Regions. Polish  weBudapest, pp. 63e85. Academy of Arts and Sciences, Krako Kuhn, S.L., 1994. A formal approach to the design and assembly of mobile toolkits. American Antiquity 59, 426e442. Lengyel, G., 2013. Knapping experiments on lithic raw materials of the Early Gravettian in Hungary. In: Mester, Z. (Ed.), The Lithic Raw Material Sources and Interregional Human Contacts in the Northern Carpathian Regions. Polish  weBudapest, pp. 39e51. Academy of Arts and Sciences, Krako Lengyel, G., 2015. Lithic raw material procurement at Bodrogkeresztúr-Henye Gravettian site, northeast Hungary. Quaternary International 359e360, 292e303. Lycett, S.J., Eren, M.I., 2013. Levallois economics: an examination of ‘waste’ production in experimentally produced Levallois reduction sequences. Journal of Archaeological Science 40, 2384e2392. Magnani, M., Rezek, Z., Lin, S.C., Chan, A., Dibble, H.L., 2014. Flake variation in relation to the application of force. Journal of Archaeological Science 46, 37e49.  , N., 2013. The lithic raw material sources and interregional human Mester, Z., Farago contacts in the Northern Carpathian regions: report and preliminary results of the field surveys. In: Mester, Z. (Ed.), The Lithic Raw Material Sources and Interregional Human Contacts in the Northern Carpathian Regions. Polish  weBudapest, pp. 23e37. Academy of Arts and Sciences, Krako  , N., Lengyel, G., 2012. The lithic raw material sources and interMester, Z., Farago regional human contacts in the Northern Carpathian regions: a research program. Anthropologie 50, 275e293. € sterle. Das Gravettien der Schwa €bischen Alb im EuroMoreau, L., 2009. Geißenklo €ischen Kontext. Kerns Verlag, Tübingen. pa Morrow, T.A., 1996. Bigger is better: comments on Kuhn's formal approach to mobile tool kits. American Antiquity 61 (3), 581e590. Nigst, P.R., 2012. The Early Upper Palaeolithic of the Middle Danube Region. Leiden University Press, Leiden. olithique Supe rieur de Moldavie: Essai de Synthe se d'une Noiret, P., 2009. Le Pale ge. Evolution Multi-Culturelle, vol. 121. ERAUL, Lie Otte, M., 1981. Le Gravettien en Europe Centrale. De Tempel, Bruges. tude de Otte, M., Noiret, P., Chirica, V., Borziac, I.A., 2007. MitoceMalu Galben: e l'industrie lithique. In: Otte, M., Chirica, V., Haesaerts, P. (Eds.), L'Aurignacien et le Gravettien de MitoceMalu Galben (Moldavie Roumaine), vol. 72. ERAUL, ge, pp. 85e136. Lie

173

Pelcin, A.W., 1997a. The effect of indentor type on flake attributes: evidence from a controlled experiment. Journal of Archaeological Science 24, 613e621. Pelcin, A.W., 1997b. The effect of core surface morphology on flake attributes: evidence from a controlled experiment. Journal of Archaeological Science 24, 749e756. Pelcin, A.W., 1997c. The formation of flakes: the role of platform thickness and exterior platform angle in the production of flake initiations and terminations. Journal of Archaeological Science 24, 1107e1113.  technique expe rimentale des techniques de Pelegrin, J., 1991. Sur une recherche ologie experimentale. Tome 2 e La Terre, l'Os et la debitage laminaire. In: Arche riPierre, la Maison et les Champs. Actes du Colloque International “Expe ologie: bilan et perspectives” 1988. Errance, Paris, mentation en arche pp. 118e128. bitage laminaire au Tardiglaciaire: crite res de Pelegrin, J., 2000. Les techniques de de flexions. In: Valentin, B., Bodu, P., Christensen, M. (Eds.), diagnose et quelques re L'Europe Centrale et Septentrionale au Tardiglaciaire. Confrontation des modles re gionaux de peuplement. Actes de la Tableeronde internationale de e moires du Muse e de Pre histoire Nemours 14e15e16 Mai 1997. Me d'IleedeeFrance n
7, A.P.R.A.I.F., Nemours, pp. 73e86. bitages laminaires du Pale olithique supe rieur. In: Pelegrin, J., 2011. Sur les de histoire. Actes du colloque Delpech, F., Jaubert, J. (Eds.), François Bordes et la pre ditions du CTHS, international François Bordes, Bordeaux, 22e24 avril 2009, e historiques, n
29, Bordeaux, pp. 141e152. coll. Documents pre Parry, W., Kelly, R.L., 1987. Expedient core technology and sedentism. In: Johnson, J.K., Morrow, C.A. (Eds.), The Organization of Core Technology. Westview Press, Colorado, pp. 285e304.   niens d'Etiolles. Pigeot, N., 1987. Magdale Economie de debitage et organisation histoire supple ment 25. CNRS, Paris. sociale. Gallia Pre Prichystal, A., 2013. Lithic Raw Materials in Prehistoric Times of Eastern Central Europe. Masaryk University, Brno. Roche, H., Tixier, J., 1982. Les accidents de taille. In: Cahen, D. (Ed.), Tailler! Pour histoire et Technologie Lithique II, Recent Progress in Microwear Quoi Faire: Pre e Royal d'Afrique Centrale, Studia Praehistorica Belgica, Studies, Tervuren, Muse vol. 2, pp. 65e76. Shott, M.B., 1986. Settlement mobility and technological organization: an ethnographic examination. Journal of Anthropological Research 42, 15e51. Sobczyk, K., 1996. KrakoweSpadzista unit D: excavations 1986e1988. Folia Quaternaria 67, 75e127. Speth, J.D., 1972. Mechanical basis of percussion flaking. American Antiquity 37, 34e60. bitage: osons ne plus affirmer. In: Cahen, D. (Ed.), Tixier, J., 1982. Techniques de de histoire et Technologie Lithique II, Recent Progress Tailler! Pour Quoi Faire: Pre e Royal d'Afrique Centrale, Studia Praein Microwear Studies, Tervuren, Muse historica Belgica, vol. 2, pp. 13e22. oLogique 4. Centre Tixier, J., 2012. A Method for the Study of Stone Tools. Arche ologique Muse e National d'Histoire et d'Art, National de Recherche Arche Luxemburg. Valentin, B., 1995 (PhD thesis). Les groupes humains et leurs traditions au Tardie, glaciaire dans le Bassin parisien. apports de la technologie lithique compare  de Paris I Panthe on, Sorbonne. http://tel.archives-ouvertes. vol. 1e2. Universite fr/tel-00267435. Whittaker, J.C., 1994. Flintknapping: Making and Understanding Stone Tools. University of Texas Press, Austin.  ski, J., 2007. The Gravettian and Epigravettian lithic assemblages from Wilczyn  w-Spadzista BþB1: dynamic approach to the technology. Folia QuaKrako ternaria 77, 37e96. Zwyns, N., 2012. Laminar Technology and the Onset of the Upper Paleolithic in the Altai, Siberia. Leiden University Press, Leiden.