sign in sign up
<<
Home , Pleistocene, Xiaochangliang

Manuscript Click here to access/download;Manuscript;Ma et al. manuscript_R2_clear.docx

1 Technological adaptations of early at the Lower Pleistocene 1 2 3 2 Nihewan Basin, North China: the case of the bipolar technique 4 1,2,3 1,2 4 1,2,3 1,2 5 3 Dongdong Ma , Shuwen Pei , Ignacio de la Torre , Zhe Xu , Hao Li 6 7 4 8 9 5 1 Key Laboratory of Vertebrate Evolution and Origins, Institute of Vertebrate 10 11 6 Paleontology and Paleoanthropology, Chinese Academy of Sciences, Beijing 12 13 7 100044, China 14 15 8 2 CAS Center for Excellence in Life and Paleoenvironment, Beijing 100044, China 16 17 3 18 9 University of Chinese Academy of Sciences, Beijing 100049, China 19 4 20 10 Department of , Institute of , National Research Council-CSIC, 21 22 11 Albasanz, 26-28, 28037 Madrid, Spain 23 24 12 25 26 13 *Corresponding author (E-mail: [email protected]) 27 28 14 29 30 31 15 ABSTRACT 32 33 16 The Nihewan Basin in North China has proved to be a key area for the study of human 34 35 17 evolution outside of Africa due to its continuous record of hominin occupation since 36 37 18 the . Lower Palaeolithic lithic assemblages at Nihewan are attributed 38 39 19 to the East Asian Mode 1 techno-complex, which is often defined by the widespread 40 41 20 use of freehand knapping techniques. However, our ongoing investigation of several 42 43 44 21 early Pleistocene archaeological sites at Nihewan has revealed a higher prevalence of 45 46 22 bipolar stone artefacts than previously considered, which may have been 47 48 23 underestimated in the past due to the disparity of analytical techniques applied to Early 49 50 24 Stone assemblages and the poor quality of the Nihewan Basin raw materials. This 51 52 25 has constrained the identification of bipolar attributes and their differentiation from 53 54 26 freehand knapping products. This study aims to investigate technological and 55 56 27 economical differences between the two techniques based on experimental results of 57 58 59 28 chert from the Nihewan Basin, creating a referential framework for the study of bipolar 60 1 61 62 63 64 65

29 artefacts that we apply, to the Early Pleistocene assemblages of and 1 2 3 30 Cenjiawan. Our results highlight morphological and technological differences between 4 5 31 bipolar and freehand products, but they also demonstrate that both techniques share 6 7 32 significant similarities in terms of dimensions and productivity. Overall, our results help 8 9 33 to contextualize the technological flexibility of East Asian Mode 1 assemblages in the 10 11 34 Nihewan Basin, where early hominins employed alternative flaking techniques, often 12 13 35 in the same assemblage, to overcome constrains imposed by the poor quality of most 14 15 36 of the raw materials available. 16 17 18 37 19 20 38 Keywords: Nihewan Basin, North China, Early Pleistocene, Bipolar technique, 21 22 39 Experimental Palaeolithic archaeology 23 24 40 25 26 41 1. Introduction 27 28 42 The bipolar technique, in which a core is placed on an anvil stone and struck with 29 30 31 43 a stone percussor, is one of the most important lithic techniques from the earliest 32 33 44 archaeological sites to the ethnographic present (Leakey 1967; Bordes 1968; S Bordaz 34 35 45 1970; Schick and Toth 1993). The occurrence of bipolar knapping in sites 36 37 46 such as Omo (de la Torre 2004; Boisserie et al. 2008), Fejej FJ-1a (Barsky et al. 2011), 38 39 47 FxJj3 (Ludwig and Harris 1998) and others show it was one of the earliest knapping 40 41 48 techniques used to obtain sharp stone tools (Schick et al., 1999; Wynn et al., 2011). 42 43 44 49 Whilst bipolar knapping has often been considered an expedient knapping technique 45 46 50 used to overcome poor flaking quality of raw materials, new evidence indicates that 47 48 51 this technique can be applied to produce even bladelet-like products or Levallois cores 49 50 52 (Tabrett 2017), and archaeologists have long sought to understand the technological 51 52 53 characteristics and variability of human behaviour associated with bipolar knapping 53 54 54 (Teilhard de Chardin and Pei 1932; Hayden 1979; Shott 1989, 1999; de la Torre, 2004; 55 56 55 de la Peña and Wadley 2014; de la Peña 2015; Byrne et al. 2016; Arroyo and de la Torre, 57 58 59 56 2018). 60 2 61 62 63 64 65

57 Bipolar knapping involves three elements, i.e. a hammer, a core and an anvil 1 2 3 58 (Patterson and Sollberger 1976). Flaking experiments have shown that striking force 4 5 59 rebound causes damage to the distal end of the core that is characteristically different 6 7 60 from freehand knapping (Patterson and Sollberger 1976; de Lombera-Hermida et al. 8 9 61 2016), but it is unlikely that all strikes produce cores with identifiable bipolar features 10 11 62 on both proximal and distal ends. Vergès and Ollé (2011) argued that the spatial relation 12 13 63 between the impact point and the counterstrike can vary in the bipolar knapping process, 14 15 64 and typical bipolar fracture depends on the core morphology and delineation of the 16 17 18 65 resting surface of the anvil. Since not every bipolar flake will be initiated with 19 20 66 equivalent forces in the proximal and distal ends, a significant proportion of bipolar 21 22 67 products may possess attributes similar to those obtained during freehand knapping 23 24 68 (Bordes 1968; Barham 1987; Bradbury 2010; Byrne et al. 2016), which will largely 25 26 69 depend on the type and quality of raw materials (Jeske and Lurie 1993). Thus, assessing 27 28 70 the extent of bipolar flaking in archaeological assemblages faces the challenges of both 29 30 31 71 correctly identifying the characteristic attributes of this , and accounting for 32 33 72 the invisibility of this technique among the ‘non-textbook’ products that lack distal 34 35 73 damage. Various experiments (de la Peña and Wadley 2014; de la Peña 2015; Byrne et 36 37 74 al. 2016) have been applied to explore the constrains of bipolar identification in lithic 38 39 75 assemblages. However, theses are often site-specific and therefore fail to establish 40 41 76 universal criteria that can be applied to Palaeolithic sites in other regions such as Asia, 42 43 44 77 or more specifically, the Nihewan Basin, the case study discussed here. 45 46 78 In this paper, we present the results of a of bipolar and freehand knapping 47 48 79 experiments with local chert collected in Nihewan Basin, compare such experimental 49 50 80 results with the Early Pleistocene assemblages of Xiaochangliang and Cenjiawan, and 51 52 81 we discuss the technological and economical differences between the two knapping 53 54 82 techniques in the context of the Palaeolithic in China. Our primary goal is to develop 55 56 83 analytical protocols for the identification of bipolar artifacts in the Early Palaeolithic of 57 58 59 84 the Nihewan Basin, and to establish the proportion of bipolar pieces that can be 60 3 61 62 63 64 65

85 expected in Early assemblages in this area. Additionally, we aim to compare 1 2 3 86 the productivity of freehand and bipolar techniques using Nihewan chert, and thus 4 5 87 contribute new insights into the benefits and constraints involved in the choice of each 6 7 88 technique during the East Asian Early Palaeolithic. 8 9 89 10 11 90 2. Materials and methods 12 13 91 2.1 Geographic and archaeological context 14 15 92 The Nihewan Basin, located in the transition zone between the North China Plain 16 17 18 93 and the Inner Mongolian Plateau (Zhou et al. 1991; Zhu et al. 2001, 2004; Deng et al. 19 20 94 2008), is a key area for the study of human behavioural evolution worldwide. Most of 21 22 95 the Early Pleistocene archaeological sites from have been excavated in this 23 24 96 basin (Wei and Xie 1989; Zhu et al. 2001, 2003, 2004; Wang et al. 2005; Deng et al. 25 26 97 2006, 2007; Xie et al. 2006; Ao et al. 2010, 2013a,b; Liu et al. 2010; Pei et al. 2019; 27 28 98 Yang et al. 2019)(Fig. 1), yielding important materials for the study of human origins 29 30 31 99 and dispersal during “Out of Africa I” (Schick et al. 1991; Zhu et al. 2001, 2004; Gao 32 33 100 et al. 2005; Deng et al. 2008; Braun et al. 2010; Dennell 2010, 2013; Pei et al. 2017). 34 35 101 It is widely accepted that the technological characteristics of the Nihewan Basin 36 37 102 sites fall within the East Asian Mode 1 techno-complex, characterized by core and flake 38 39 103 assemblages (Pei et al., 2017, 2019). Most studies have emphasized the predominance 40 41 104 of freehand knapping of poor- quality chert local outcrops (Schick et al. 1991; Keates 42 43 44 105 2000; Shen and Wei 2004; Gao et al. 2005; Dennell 2008; Liu et al. 2013; Guan et al. 45 46 106 2016; Yang et al. 2016; Pei et al. 2017) (see Table 1). In recent , however, re- 47 48 107 analyses of previously excavated assemblages from Xiaochangliang and Donggutuo 49 50 108 have identified higher frequencies of bipolar artifacts than originally reported (Yang et 51 52 109 al. 2016, 2017); Yang et al. (2016) attributed the underrepresentation of bipolar 53 54 110 products in earlier studies to a lack of systematic definitions of bipolar attributes and 55 56 111 the particularities of the Nihewan raw materials. Chert nodules primarily selected for 57 58 59 112 tool manufacture in the Nihewan Basin contain abundant internal flaws that lead to 60 4 61 62 63 64 65

113 unpredictability during knapping and the shattering of products (Shen and Wei 2004; 1 2 3 114 Liu et al. 2013; Pei et al. 2013, 2019; Guan et al. 2016; Yang et al. 2016), thus 4 5 115 complicating recognition of bipolar artefacts. 6 7 116 In other Nihewan sites (Table 1), a reconsideration of flaking techniques is still 8 9 117 pending and divergent interpretation of the same lithic assemblages exist. This 10 11 118 highlights some of the problems of current research in the region, which are 12 13 119 compounded by a lack of experimental referential frameworks. Using criteria identified 14 15 120 from our experimental replications, we present a re-evaluation of available results 16 17 18 121 published in earlier studies, along with a preliminary re-examination of archaeological 19 20 122 samples from the Cenjiawan site, both of which illustrate how variable proportions of 21 22 123 bipolar products are identified depending on the attributes that are used. 23 24 124 Most studies of sites in the Nihewan Basin regard bipolar knapping as a cultural 25 26 125 marker of these sites but have not considered its broader implications. However, as a 27 28 126 reduction technology, bipolar flaking could be employed by early knappers for a variety 29 30 31 127 of reasons, including ecological or economic constrains, cultural traditions, and/or, 32 33 128 cognitive abilities, among others. Therefore, the traditional typological studies of 34 35 129 bipolar knapping limit our understanding of human behavioural variability and 36 37 130 adaptations in each particular context, and they should be complemented by 38 39 131 technological studies aimed at understanding similarities and differences in the 40 41 132 reduction processes involved in freehand and bipolar flaking. 42 43 44 133 2.2 Raw materials 45 46 134 The central area of the Cenjiawan Platform (Barbour 1924; Barbour et al., 1927) 47 48 135 in the eastern part of the Nihewan Basin contains outcrops of rocks and 49 50 136 volcanic rocks (Fig. 1c). Precambrian rocks (Sinian rock system), were formed 51 52 137 in a lagoon environment where siliceous dolomite was developed. Chert usually 53 54 138 appears within the Sinian rock system in bands and as irregular, dense, and 55 56 139 homogeneous nodules (with a thin outer layer of cemented silica) (Pei and Hou 2002). 57 58 59 140 Due to tectonic movement and Jurassic volcano eruptions, the Sinian rock system 60 5 61 62 63 64 65

141 underwent secondary fracture transformations that created a brecciated structure for 1 2 3 142 chert and siliceous dolomite (Pei and Hou 2002; Pei et al. 2017). The chert is fine- 4 5 143 grained silica-rich microcrystalline or microfibrous, and while it varies greatly in color, 6 7 144 it is often brown, gray, grayish brown, or rusty red (Pei et al. 2017). The chert often 8 9 145 exhibits internal flaws, fractures and a brecciated structure, which decrease its flaking 10 11 146 quality (Guan et al. 2016; Yang et al. 2016, 2017; Pei et al. 2017). Despite these flaws, 12 13 147 chert is generally the most suitable rock for flaking in the area. It was locally abundant 14 15 148 around the archaeological sites across the Cenjiawan Platform from two sources: most 16 17 18 149 derives from Sinian rock outcrops where chert is usually available as blocks that usually 19 20 150 weathered into smaller pieces suitable for human collection: and some was available as 21 22 151 cobbles from the braided channels where Sinian lithologies (in contact with Jurassic 23 24 152 volcanics) are transported by water (Yang et al. 2016; Pei et al. 2017). 25 26 153 The chert blocks and cobbles used in our experiments were collected from the 27 28 154 Cenjiawan Platform fluvio-lacustrine deposits, where most of the Early Pleistocene 29 30 31 155 archaeological sites at Nihewan are clustered. Samples were collected on the basis of 32 33 156 their shape and size suitability for flaking. We selected chert with quadrangular or 34 35 157 tabular morphology for BO1 experiments (Fig. 2a) and rounded blanks for BS 36 37 158 experiments (Fig. 2b), while for the freehand experiments, blanks with an adequate 38 39 159 platform angle were preferred. In addition, one 134×124×62 mm quartzite block with 40 41 160 a flat surface was selected as the anvil for the bipolar experiments. 42 43 44 161 2.3 Bipolar and freehand experiments 45 46 162 Variability of gestures potentially involved in bipolar flaking, in which the striking 47 48 163 angle plays an essential role (Fig. 3), has long been recognized (see review in Byrne et 49 50 164 al. 2016). When cores are made from quadrangular or tabular blanks, flaking can be 51 52 165 initiated from one of the core ends, a flaking motion that was in our experiments named 53 54 166 as “bipolar knapping with orthogonal hammer (BO1)” (Fig. 2a); when blanks had an 55 56 167 elliptical shape, cores were struck on their halving axis and named here as “bipolar 57 58 59 168 splitting (BS)” (Fig. 2b). We chose not to replicate flaking motions in which the core 60 6 61 62 63 64 65

169 rests obliquely on the anvil (BO2) (Fig. 2c), or where the hammer is used with an 1 2 3 170 oblique trajectory (BOB) (Fig. 2d), as in such cases, fracture mechanisms are similar to 4 5 171 those operating in freehand techniques (Duke and Pargeter 2015; Hiscock 2015). 6 7 172 Given that the search for suitable angles is not a priority in bipolar knapping, and 8 9 173 to facilitate the morphological and technological analysis of resulting products, all 10 11 174 bipolar cores in our experiments were reduced using only a single platform. Conversely, 12 13 175 no constraints were imposed in the positioning and rotation of cores during freehand 14 15 176 knapping, which were manipulated freely to find appropriate striking platforms. Two 16 17 18 177 quartzite (93 × 68 × 55 mm and 111 × 83 × 47mm) were used in the 19 20 21 178 experiments, weighing 506g and 717g respectively. One quartzite anvil was used during 22 23 179 the bipolar experiments. 24 25 180 The primary goal in each experiment was to produce as many flakes as possible; 26 27 181 when the core became too small to hold or when no was produced after twenty 28 29 182 consecutive strikes, the experiment was deemed as completed. As discussed above, 30 31 183 Nihewan chert is riddled with internal flaws, which led some cores to split during the 32 33 184 experiments; when a fragment was large enough to be hand-held, it was regarded as a 34 35 36 185 core and knapping continued, while the smaller core fragments were categorized as part 37 38 186 of the resulting debitage. 39 40 187 Two intermediate-skilled knappers (Yingshuai Jin and the first author of this paper) 41 42 188 participated in the experiments. A total of 25 sets, comprising eight BO1, six BS and 43 44 189 eleven freehand cores, was produced (Table 2). Experiments were conducted at the 45 46 190 Institute of Vertebrate Paleontology and Paleoanthropology, Chinese Academy of 47 48 49 191 Sciences (Fig. 3), where the experimental assemblages are currently housed. All 50 51 192 sessions were video recorded, and all materials produced in each experiment were 52 53 193 collected; shatter less than 20mm was not studied for this paper, while all materials at 54 55 194 or above 20 mm were analyzed. 56 57 195 2.4 Measured variables and statistical analysis 58 59 196 In order to compare bipolar and freehand artefacts produced with Nihewan chert, 60 7 61 62 63 64 65

197 we measured the same attributes in all assemblages. Table 3 compiles all the attributes 1 2 3 198 considered in our study, which encompass standard variables widely used in the 4 5 199 analysis of bipolar products (Kobayashi 1975; Byrne et al. 2016), and some less- 6 7 200 commonly used attributes that we hoped could provide further insights in the 8 9 201 differentiation between freehand and bipolar flaking. 10 11 202 Normality tests were conducted to determine adequacy of the sample. Independent 12 13 203 sample T tests and Mann-Whitney U tests were performed to compare metric variables. 14 15 204 Chi-square tests were used to compare categorical variables of bipolar and freehand 16 17 18 205 cores and flakes. Fisher’s Exact tests were used in variables where the sample was less 19 20 206 than five. The Kruskal-Wallis test was used to compare ratios of length to width of BO1, 21 22 207 BS and freehand flakes. All statistics were computed using SPSS. 23 24 208 25 26 209 3. Results 27 28 210 3.1 Techno-typological analysis 29 30 31 211 3.1.1 Types of artefacts 32 33 212 Apart from cores, all detached pieces can be classified into four categories: 34 35 213 complete flake (abbreviated here to flake), flakes with transversal fractures (incomplete 36 37 214 flakes with missing distal ends), Siret flakes (split along the long axis through the bulb 38 39 215 of percussion), and shatter (products with no identifiable attributes). Table 4 shows the 40 41 216 distribution of artefacts produced in bipolar and freehand experiments. Bipolar 42 43 44 217 experiments yielded 123 flakes, 35 transverse fracture flakes, 35 Siret flakes and 105 45 46 218 shatter pieces, while freehand experiments produced 107 flakes, 28 transverse fracture 47 48 219 pieces, 33 Siret fractures, and 111 shatter pieces. The Chi-square test shows no 49 50 220 statistically significant differences in the proportion of products between bipolar and 51 52 221 freehand knapping (2(3)=1.492, p=0.684). 53 54 222 3.1.2 Cores 55 56 223 In the bipolar experiments, BO1 and BS motions show significant differences in 57 58 59 224 the resulting cores. Proximal and distal ends of most BS cores show conspicuous 60 8 61 62 63 64 65

225 battering and crushing, whereas BO1 cores are considerably less damaged. Even more 1 2 3 226 significantly, all BS cores present heavily battered striking platforms, which is always 4 5 227 absent in BO1 cores and becomes the most obvious difference between the two core 6 7 228 types (Fig. 5). 8 9 229 Clear differences are also observed in crushing between bipolar and freehand cores, 10 11 230 which are statistically significant (Fisher’s test: p=0.000). As shown in Table 5, all BO1 12 13 231 and BS experiments present medium or heavy crushing, while no freehand cores show 14 15 232 heavy crushing. 59% of bipolar cores possess bipolar scars (as defined in Figure 4), 16 17 18 233 which are absent in all freehand experiments and constitute the most distinctive 19 20 234 attribute in identifying bipolar cores. 21 22 235 3.1.3 Flakes 23 24 236 A total of 56.1% bipolar and 85% freehand flakes show no dorsal battering or 25 26 237 crushing in the proximal end (Fig. 6). The proportion of slight crushing in bipolar flakes 27 28 238 (30.9%) is larger than in freehand flakes (12.2%). Medium crushing also occurs more 29 30 31 239 frequently in bipolar (7.3%) than in freehand (2.8%) flakes, and heavy crushing is only 32 2 33 240 observed in 5.7% of bipolar flakes (Fig. 7a). The Chi-Square test shows ( (3)=24.284, 34 35 241 p=0.000) statistically significant differences between the two samples, with bipolar 36 37 242 flakes being more likely to present crushing in their proximal ends. 38 39 243 Feathered terminations dominate in both bipolar and freehand flakes (Fig.7b). 40 41 244 Hinged, plunge and axial terminations (i.e., where fracture proceeds directly through 42 43 44 245 the core to its opposite end, and often roughly bisects the core [Cotterell 1987]) all 45 46 246 account for small proportions in the two samples. Flakes with impact points at the 47 48 247 opposite ends are often regarded as typical bipolar flakes (Shott 1989; Bradbury 2010; 49 50 248 Gurtov and Eren 2014). In our experiments, however, only 13.8% percent (n=17) of 51 52 249 bipolar flakes possess this . With the exception of flakes with two opposite 53 54 250 impact points (only present in bipolar knapping), there are no statistically significant 55 56 251 differences in the termination of flakes obtained through the two techniques (Fisher 57 58 59 252 Test p=0.439). 60 9 61 62 63 64 65

253 Bulbs are absent in 74% of bipolar flakes (n=91), and the proportion (68%) is 1 2 2 3 254 similar in freehand flakes (n=73), with the Chi-square test ( (1)=0.928, p=0.335) 4 5 255 confirming that no statistic differences exist between the two techniques. 6 7 256 Regarding flake profile shape (Fig. 7c), straight patterns dominate in both bipolar 8 9 257 (47.5%) and freehand (49.5%) flakes, and convex profiles are at least represented in 10 11 258 both techniques. Again, no statistically significant differences (2(3)=0.662, p=0.882) 12 13 259 are observed for this attribute between bipolar and freehand knapping. In bipolar and 14 15 260 freehand experimental results with quartz collected from , Li (2016) 16 17 18 261 described that the curvature of blanks is generally straight for bipolar flakes (62/74 19 20 262 =83.8%), while freehand flakes have more curved profiles (10/20=50%). De la Peña 21 22 263 (2015) also reported similar experimental results (where the profile of bipolar blanks 23 24 264 tends to be rectilinear) on quartz collected from Johannesburg, Limpopo and Namibia. 25 26 265 The high ratio of punctiform or linear striking platforms (Fig. 7d) is regarded as 27 28 266 an important attribute of bipolar flakes (52%), and in fact this becomes fully dominant 29 30 31 267 in the BS experiments, where such platform types reached 81%. In contrast, only 18.7% 32 33 268 of freehand flakes possess punctiform or linear platforms, and statistically significant 34 35 269 differences exist between both techniques (2(1)=27.437, p=0.000). 36 37 270 3.2 Dimensional analysis 38 39 271 Since size of original cores varies greatly, and in order to ensure that there is no 40 41 272 statistical difference in experimental data between bipolar and freehand knapping 42 43 44 273 products, those with an initial weight between 400~2000g were considered in the 45 46 274 dimensional analysis, which included 8 bipolar and 9 freehand experiments. The mean 47 48 275 weight of bipolar original blanks is 668g, and 876g for those employed in freehand 49 50 276 knapping. A t-test (t(16)=1.594, p=0.081) comparison shows no significant differences 51 52 277 between bipolar and freehand blanks, which produced 82 flakes in the case of bipolar 53 54 278 cores and 94 flakes in freehand experiments. 55 56 279 The mean size and weight of bipolar flakes are 33.6×25.7×9.6 mm and 15g, and 57 58 59 280 37×30×10.3 mm and 21.6g in the case of freehand flakes (Table 6). Results of the 60 10 61 62 63 64 65

281 Mann-Whitney U test indicate that differences in proximal width (U=3400.5, p=0.022) 1 2 3 282 and mesial width (U=3321, p=0.012) of bipolar and freehand flakes are statistically 4 5 283 significant, while other variables such as length (U=3852, p=0.303), distal width 6 7 284 (U=3945, p=0.439), proximal thickness (U=4130.5, p=0.796), mesial thickness 8 9 285 (U=3973.5, p=0.488), distal thickness (U=4051.5, p=0.633) and weight (U=3941, 10 11 286 p=0.433) are similar in both samples. 12 13 287 In order to compare the size of flakes produced in bipolar and freehand 14 15 16 288 experiments, the length×width ratio (U=3516, p=0.05) (Fig. 8a) and length (Fig. 8b) of 17 18 289 flakes were also compared and plotted. The results do not seem to significantly 19 20 21 290 segregate bipolar and freehand flakes, even when the latter are normally larger. 22 23 291 The mesial width/length ratio provides information on flake elongation. In our 24 25 292 experiments, this ratio is 1.43 in BO1 flakes, 1.59 in BS, and 1.3 in freehand knapping. 26 27 293 The Kruskal-Wallis test (H=9.545, p=0.008, df=2) shows there are significant 28 29 294 differences between the three groups, whereas a pairwise comparison indicates there 30 31 295 are no statistically significant differences between BO1 and BS flakes (2(1)=17.478, 32 33 2 34 296 p=0.223) or between BO1 and freehand flakes ( (1)=-18.701, p=0.418). Nonetheless, 35 2 36 297 BS flakes are statistically more elongated than freehand flakes ( (1)=36.178, p=0.008). 37 38 298 3.3 Flake and edge productivity 39 40 299 Bipolar experiments produced an average of 26 debitage pieces (SD=13.2), 33 41 42 300 cutting edges (SD=22.7) and 803.6 mm (SD=510.8) of edge length (Fig.9a), while 43 44 301 freehand experiments yielded an average of 25 debitage pieces (SD=14), 37.4 cutting 45 46 302 edges (SD=22.1) and 1018 mm (SD=685.9) of cutting length. The Mann-Whitney U 47 48 49 303 test shows no statistically significant differences between bipolar and freehand 50 51 304 knapping in either of these variables (number of detached pieces U=39.5, p=0.929; 52 53 305 cutting edges U=30, p=0.351; cutting length U=29, p=0.310). 54 55 306 Only 82.3% of bipolar products presented cutting edges, compared to 91.5% in 56 57 307 the freehand sample (Table 7), with the Chi-square test indicating statistically 58 59 308 significant differences between the two techniques (2(1)=8.528, p=0.004). Focusing 60 11 61 62 63 64 65

309 on the flakes, 89% (n=73 out of 82) of bipolar flakes presented cutting edges, whereas 1 2 3 310 99% (n=93 out of 94) of freehand flakes had cutting edges (Table 7), with the Chi- 4 5 311 square test again indicating statistically significant differences between the two samples 6 2 7 312 ( (1)=8.029, p=0.006). 8 9 313 The mean cutting length of all bipolar products is 37.3 mm (SD=19), while 10 11 314 freehand pieces yield 44.9 mm cutting length in average (SD=28.3), showing 12 13 315 statistically significant differences between the two samples (Mann-Whitney U=16927, 14 15 316 p=0.03). Focusing on the flakes, the mean cutting length in bipolar is 46.14mm 16 17 18 317 (SD=20), considerably less than in freehand flakes (60.13mm; SD=28) (Fig.9b), again 19 20 318 with statistically significant differences between the two techniques (Mann-Whitney 21 22 319 U=2322.5, p=0.000). 23 24 320 Analysis of the videotapes enabled counting the number of -to-core 25 26 321 strikes in each experiment, which can be used as a measure of flaking efficiency and 27 28 322 help to evaluate differences in productivity between bipolar and freehand knapping. 29 30 31 323 The quantitative analysis of strikes per experiment (Table 8) and the Mann-Whitney 32 33 324 test shows there are no statistically significant differences in the production of flakes 34 35 325 (U=21, p=0.145), other detached pieces (U=23.5, p=0.228), number of cutting edges 36 37 326 (U=17.5, p=0.075) and cutting length per strike (U=18, p=0.083) between bipolar and 38 39 327 freehand experiments. The ratio of core weight loss to number of strikes per experiment 40 41 328 indicates that bipolar cores lost 3.8g per strike (SD=2.84), and freehand lost 5 g 42 43 44 329 (SD=3.5), with the Mann-Whitney test showing no statistical differences between the 45 46 330 two techniques (U=25, p=0.290). 47 48 331 49 50 332 4. Discussion 51 52 333 4.1 New perspectives on the identification of bipolar products at the Nihewan 53 54 334 Basin 55 56 335 In our experiments with Nihewan chert, cores are the most conspicuous proxies to 57 58 59 336 distinguish bipolar from freehand flaking; 59% of bipolar cores presented bipolar scars, 60 12 61 62 63 64 65

337 and 61% of such cores showed heavy crushing at the proximal or distal ends. BS cores 1 2 3 338 are more prone to present crushing than BO1 cores. 4 5 339 Identification of bipolar flakes has often been considered substantially more 6 7 340 difficult than in the case of cores. In our experiments, most bipolar flakes possess a 8 9 341 feathered termination, present no bulbs and a straight profile shape, which are all 10 11 342 techno-morphological attributes also typical of freehand flakes. Previous experiments 12 13 343 have also encountered similar issues when attempting to differentiate bipolar from 14 15 344 freehand flakes (e.g., Patterson and Sollberger 1976; Bradbury 2010; Byrne et al. 2016), 16 17 18 345 and often reliable determinations come down to classic bipolar features such as 19 20 346 crushing both on the proximal and distal ends of flakes and crushing of platforms. In 21 22 347 our experiments, only 5.7% of bipolar flakes show heavy crushing on platforms and 23 24 348 13.8% have impact points both in the proximal and distal ends. Based on these two 25 26 349 attributes combined, around 18% of bipolar flakes could be confidently identified for 27 28 350 this technique. Absence of bulbs and straight profiles are common features in our 29 30 31 351 experimental freehand flakes, which we link to the poor quality of the local chert. Thus, 32 33 352 these attributes are not strong indicators of freehand versus bipolar flaking for the raw 34 35 353 materials analysed here, and hence they cannot be used to distinguish knapping 36 37 354 techniques. 38 39 355 Apart from attributes exclusive of bipolar knapping, punctiform or linear striking 40 41 356 platforms are often thought to have a strong relationship with bipolar knapping (Binford 42 43 44 357 and Quimby 1963; Hayden 1979; Barham 1987; Bradbury 2010; Berman et al. 2017). 45 46 358 In our experiments, not all bipolar products presented punctiform or linear platforms, 47 48 359 but 27% of flakes obtained through this technique did show crushing associated to 49 50 360 punctiform or linear platforms, which is in stark contrast to freehand flakes (3%). This 51 52 361 suggests that shape of the is an important discriminatory attribute. 53 54 362 Conversely, attributes such as profile shape, termination, or bulb presence yielded 55 56 363 similar proportions in both bipolar and freehand products with Nihewan chert, which 57 58 59 364 reinforces previous observations on other raw materials (e.g., Byrne et al. 2016). 60 13 61 62 63 64 65

365 Bipolar and freehand knapping produced similar proportions of flakes, transverse 1 2 3 366 and split fragments, and shatter in our experiments. Thus the widespread notion that 4 5 367 bipolar knapping produces higher percentages of debris and is more wasteful than 6 7 368 freehand knapping (Shott 1989; LeBlanc 1992; Jeske and Lurie 1993; Gurtov and Eren 8 9 369 2014) is not supported by our results for Nihewan chert. It is therefore likely that the 10 11 370 high proportion of debris in some Nihewan early Pleistocene sites is associated with 12 13 371 the poor quality of raw materials, rather than to predominance of bipolar knapping in 14 15 372 the assemblages (Yang et al. 2016, 2020; Pei et al. 2019). As discussed above, Nihewan 16 17 18 373 chert often exhibits internal fractures and a brecciated structure that makes this raw 19 20 374 material prone to shattering. 21 22 375 Bipolar flakes are similar to freehand flakes in size, weight, surface area and the 23 24 376 ratio of length to mesial width, whereas distal and mesial width of freehand flakes is 25 26 377 larger than in bipolar flakes, with the former also being more elongated. When 27 28 378 productivity is considered in terms of number of strikes to obtain a flake, no significant 29 30 31 379 differences exist in the number of cutting edges and total cutting length produced in 32 33 380 bipolar and freehand experiments. However, the mean cutting length of debitage and 34 35 381 flakes is longer in freehand than in bipolar products, and freehand pieces possess more 36 37 382 sharp cutting edges. In addition, this study supports results by de Lombera-Hermida et 38 39 383 al. (2016) that higher accuracy can be inferred for freehand knapping than in bipolar. 40 41 384 Our experiments have also shown that, although the overall advantages of 42 43 44 385 productivity of freehand knapping are not entirely evident, this technique is more 45 46 386 effective in producing cutting edges and cutting length. However, several benefits have 47 48 387 been recognized in the use of bipolar knapping as , such as enabling reduction of 49 50 388 round pebbles/cobbles or poor-quality raw materials, increasing productivity in 51 52 389 situations of high scarcity by maximizing reduction (Barham 1987; Knight 1988; Shott 53 54 390 1989; Hiscock 2015), and producing particular tools such as (de la Peña 55 56 391 2013). In the Nihewan Basin, however, freehand techniques were the primary choice of 57 58 59 392 Early Pleistocene hominins, and bipolar knapping may have been used essentially as a 60 14 61 62 63 64 65

393 complementary technique (see Table 1). Given the proximity of Nihewan Early 1 2 3 394 Pleistocene sites to the raw material sources and the large size of some cores in the 4 5 395 assemblages, we conclude that bipolar knapping was not employed as a response to raw 6 7 396 material scarcity or to maximize return by exploiting miniaturized cores. As a working 8 9 397 hypothesis, we propose that the inherent unpredictability of the Nihewan chert may 10 11 398 potentially have incentivized the use of poor-accuracy techniques such as bipolar 12 13 399 flaking, as knappers would be aware that high quality flakes were unattainable due to 14 15 400 the internal flaws of most of the raw materials locally available. Such expedient 16 17 18 401 methods would have complemented freehand knapping techniques, which predominate 19 20 402 at Nihewan but seem to be consistently accompanied by bipolar flaking. 21 22 403 4.2 Tracing bipolar technology in the Nihewan Basin Early Pleistocene sites 23 24 404 Bipolar flaking is yet to be widely recognized in the Nihewan assemblages, yet it 25 26 405 is relevant not only to early human technological choices but also to analytical factors 27 28 406 and the history of research. For instance, the properties of Nihewan raw materials 29 30 31 407 hamper the identification of bipolar products. As we have shown in our experiments, 32 33 408 less than 20% of bipolar flakes show features that would allow them to be recognized 34 35 409 in archaeological contexts. Soft raw materials such as wood can be used as anvils for 36 37 410 bipolar flaking and would produce flakes with undamaged distal edges (Soriano and 38 39 411 Villa 2017). While we have no evidence for wood use in the Early Pleistocene Nihewan 40 41 412 sequence, it is worth highlighting that stone anvils are yet to be reported in any of the 42 43 44 413 assemblages, which makes the presence of bipolar flaking at the sites even less 45 46 414 conspicuous. Together with the potential ‘invisibility’ of bipolar elements, it should be 47 48 415 borne in mind that the sites with higher proportions of bipolar products are those studied 49 50 416 in the last few years, which indicates that an awareness of the potential existence of 51 52 417 bipolar flaking is a recent analytical trend. 53 54 418 An example of this trend is the recent analysis of the Xiaochangliang assemblage, 55 56 419 where chert is the main raw material (96.4%) used in stone knapping. Yang et al. (2016) 57 58 59 420 identified bipolar cores and bipolar splinters. While standards for the identification of 60 15 61 62 63 64 65

421 bipolar cores were not clarified, bipolar splinters were defined as small pieces with 1 2 3 422 crushing on either the proximal platform or the base, and always without evidence of 4 5 423 Hertzian initiation (e.g. bulbs of force, éraillure scars, ripple marks) (Yang et al. 2016). 6 7 424 The criteria used by these authors to identify bipolar products was looser when 8 9 425 considered in the light of our experimental results. Thus, a great number of “chips” and 10 11 426 “chunks” were classified as bipolar products, with 56.92% of artifacts in the assemblage 12 13 427 (including 111 bipolar cores and 328 bipolar splinters) attributed to the bipolar 14 15 428 technique. This is a much higher proportion than in previous studies (Table 1). 16 17 18 429 However, according to our experiments some attributes used by Yang et al (2016), 19 20 430 such as the bulb of force, are also absent in most freehand flakes (68%), and no 21 22 431 significant differences can be discerned between bipolar and freehand flakes. Only 20% 23 24 432 of the bipolar products could be identified confidently in our experiments, and this ratio 25 26 433 would be even lower when considering that in many archaeological assemblages the 27 28 434 use of both freehand and bipolar techniques co-existed. Therefore, in the light of our 29 30 31 435 experiments we would argue that the actual ratio of bipolar products at the 32 33 436 Xiaochangliang site may have been overestimated, and the assemblage needs to be re- 34 35 437 examined using standards set in our experimental replications. 36 37 438 In addition to re-examining recently published case studies in the light of our 38 39 439 experimental results, we also apply these to our own analysis of the archaeological 40 41 440 assemblage of Cenjiawan assemblage. This site has been dated to ca. 1.1 Ma, and is 42 43 44 441 located on the margin of the Cenjiawan platform (Fig. 1), yielding a total of 1996 stone 45 46 442 artefacts (81.5% of them on chert). Although bipolar cores retain more identifiable 47 48 443 attributes, only two bipolar cores (2 out of 79) with bipolar scars and obvious crushing 49 50 444 were identified. In addition, among the 634 flakes in this assemblage, only eight could 51 52 445 be confidently identified as bipolar flakes, when our experimental standards are applied. 53 54 446 Regarding the Cenjiawan flakes regarded as freehand products, 16.7% (88 out of 55 56 447 528) present linear or punctiform platforms, 48.9% show no bulbs, and 28.8% present 57 58 59 448 a straight profile shape. Except for the proportion of linear or punctiform striking 60 16 61 62 63 64 65

449 platforms that is similar to our experiments, the proportions of absent bulbs and straight 1 2 3 450 profiles in flakes interpreted as freehand features are a little lower than in our 4 5 451 experimental results. Therefore, we concluded that attributes such as bulb shape, profile 6 7 452 shape and flake termination types cannot be employed to confidently identify bipolar 8 9 453 knapping products on chert in the Nihewan Basin, especially when both freehand and 10 11 454 bipolar products co-exist in an assemblage. 12 13 455 4.3 Bipolar technology and hominin adaptations in Early Palaeolithic China 14 15 456 Bipolar knapping has long been recognized in the Early Stone Age of China, 16 17 18 457 starting with the excavations at Zhoukoudian (Breuil 1932; Pei 1932, 1933), where this 19 20 458 flaking technique is still its most distinguished feature (Li 2016). However, recent 21 22 459 experimental studies have suggested that the proportion of bipolar products might be 23 24 460 lower than previously estimated, as 75% of debitage may be attributed to bipolar flaking 25 26 461 (Pei and Zhang 1985; Li 2016). 27 28 462 Since the 1970s, bipolar products have been reported in other Early Palaeolithic 29 30 31 463 sites in northern China (e.g., Zhang 1998; Gao 2000), although often in very low 32 33 464 proportions. Some exceptions exist, however, such as Xiaochangliang (Yang 2016), and 34 35 465 Donggutuo (Yang et al. 2017), where the proportion of bipolar products is higher, 36 37 466 although always playing a secondary role to freehand knapping. 38 39 467 In south China, cobbles are commonly used as blanks for stone tools, and freehand 40 41 468 knapping is also the primary knapping technique. Similar to the situation of north China, 42 43 44 469 very few southern China sites have been reported to contain bipolar products and, where 45 46 470 present, the proportion is also significantly low. An interesting pattern, as revealed 47 48 471 recently by Li et al. (2017), is that bipolar knapping was also used to obtain large 49 50 472 flakes at the Danjiangkou Region, as a flexible technological 51 52 473 adaptation to the local poor-quality raw material of quartz phyllite. In addition, a 53 54 474 particular bipolar technique similar to the non-axial bipolar technique described in 55 56 475 western Europe (eg., de Lombera-Hermida et al. 2016), is known in China as Ruileng 57 58 59 476 bipolar knapping (Cao 1978; although see Gao et al. 2008). It involves selection of large 60 17 61 62 63 64 65

477 cobbles with tabular shapes, which are placed on the anvil obliquely (rather than 1 2 3 478 axially), similar to the motions shown in Fig 2 c except that the cores were exclusively 4 5 479 made on cobbles with tabular shapes. Flakes produced via Ruileng bipolar knapping 6 7 480 usually have punctiform or linear platforms, and they lack crushing on the distal ends 8 9 481 (Cao 1978). Ruileng bipolar products are found extensively across southern China, 10 11 482 potentially spanning from the Early to the (Xie, 2017). 12 13 483 To summarize, and following Gao (2000), bipolar knapping in Early Palaeolithic 14 15 484 China is complementary to freehand knapping at most sites, except for Zhoukoudian 16 17 18 485 Locality 1, where bipolar knapping was reported as the main knapping technology to 19 20 486 exploit quartz. Quartz is the most commonly used raw material in bipolar flaking, and 21 22 487 bipolar products are normally small and few are retouched. Nonetheless, it also seems 23 24 488 clear that the relevance of bipolar knapping has been consistently underestimated for 25 26 489 most sites in China, and it is probably not a coincidence that most of the assemblages 27 28 490 with a higher proportion of bipolar knapping derive from recent re-analyses of 29 30 31 491 collections. This hints at our incomplete understanding of bipolar attributes across Early 32 33 492 Palaeolithic sites in China, although new experimental studies have started to consider 34 35 493 the impact of local raw material idiosyncrasies in their identification (but see Lin 1987; 36 37 494 Li 2016; Li et al 2017). 38 39 495 While bipolar knapping has long been regarded from a cultural history perspective 40 41 496 (e.g., Zhang 1998) that would link culture traditions from the Lower to the Upper 42 43 44 497 Palaeolithic in China, emphasis should now be placed on understanding its role in 45 46 498 hominin organization and adaptive strategies. Among these, raw material exploitation 47 48 499 has proved to be a major factor influencing technical choices of makers 49 50 500 (Braun, et al., 2009). Shifting perspectives such as those adopted in this paper are 51 52 501 considering bipolar knapping as an economical, technological and/or adaptive response 53 54 502 by ancient tool-makers (Li 2016), rather than an evolutionary one, although 55 56 503 efforts should also consider its palaeoecological implications. 57 58 59 504 60 18 61 62 63 64 65

505 5. Conclusions 1 2 3 506 This paper has explored the role of bipolar flaking in early hominin technological 4 5 507 organization in North China. We have analyzed the attributes that can used in the 6 7 508 identification of bipolar products in the Nihewan Basin by conducting experiments with 8 9 509 the same chert that was used by Early Pleistocene hominins in the area. Our results 10 11 510 show that cores are the most reliable indicators to identify bipolar knapping, with most 12 13 511 bipolar cores presenting attributes like bipolar scars or heavy crushing of the proximal 14 15 512 or distal ends. These are traits that are significantly different from freehand cores. 16 17 18 513 Proportions of flakes confidently attributable to bipolar knapping are consistently low: 19 20 514 only attributes like heavy crushing on platforms or impact points both at the proximal 21 22 515 and distal ends could confidently be identified as bipolar flakes, while profile shape, 23 24 516 termination, or bulb shape cannot be used to distinguish freehand from bipolar 25 26 517 techniques. 27 28 518 Freehand knapping materials produced more cutting edges and cutting length than 29 30 31 519 bipolar experiments, which may explain the dominance of freehand assemblages in the 32 33 520 Nihewan Basin. Bipolar knapping, however, consistently appears throughout the Early 34 35 521 Pleistocene at Nihewan and had a complementary role, which we propose may be 36 37 522 associated with the poor quality of raw materials generally available in the basin. This 38 39 523 may have restrained the use of more controlled flaking gestures such as those involved 40 41 524 in freehand knapping. 42 43 44 525 Results presented in this paper are intended to explore the specific characteristics 45 46 526 of bipolar knapping on chert from Nihewan Basin. We advocate that similar 47 48 527 experimental work should be done in other areas of China using raw materials specific 49 50 528 to those sites. This will contribute to assessing the technological and economical 51 52 529 differences between bipolar and freehand knapping, and it will create a referential 53 54 530 framework for the study of bipolar artefacts in each particular research area. 55 56 531 The bipolar technique, which is widely spread across China not only in the Early 57 58 59 532 Pleistocene but also across the Middle Pleistocene, may have been associated with 60 19 61 62 63 64 65

533 economic and functional constraints that are still poorly understood, and which deserve 1 2 3 534 further in-depth studies on the significance of technological choices in the adaptations 4 5 535 of Early Palaeolithic hominins in East Asia. 6 7 536 8 9 537 10 11 12 538 Acknowledgements 13 14 539 We thank Yingshuai Jin from Institute of Vertebrate Paleontology and 15 16 540 Paleoanthropology and Xin Ding from University College London for participating in 17 18 541 the experiments. We are grateful to Nicholas Conard and the three anonymous 19 20 21 542 reviewers for their constructive feedback on an earlier version of this manuscript. 22 23 543 24 25 544 26 27 545 Funding information 28 29 546 This research was supported by the Strategic Priority Research Program of 30 31 547 Chinese Academy of Sciences (Gran No. XDB26000000), the National Natural Science 32 33 548 Foundation of China (41872029, 41372032), the Chinese Academy of Sciences 34 35 36 549 President's International Fellowship Initiative (Grant No. 2017VCA0038), an ERC- 37 38 550 Advanced Grant (Horizon 2020, BICAEHFID grant agreement No. 832980), and the 39 40 551 Chinese Academy of Sciences Pioneer Hundred Talents Program. 41 42 552 43 44 553 45 46 47 554 Authors' contributions 48 49 555 S.W.P., D.D.M. and I.T. designed the research. D.D.M., and S.W.P. collected the 50 51 556 data. D.D.M., S.W.P., I.T., Z. X. and H.L. performed the analysis. D.D.M., S.W.P. and 52 53 557 I.T. wrote the paper. 54 55 558 56 57 559 58 59 560 60 20 61 62 63 64 65

561 References 1 2 3 562 Ao H, Deng C, Dekkers MJ, Liu Q, Qin L, Xiao G, Chang H (2010) Astronomical 4 5 563 dating of the Xiantai, Donggutuo and Maliang sites in the Nihewan 6 7 564 Basin (North China) and implications for early in East Asia. 8 9 565 Palaeogeogr Paleoclimatol Palaeocl 297: 129–137 10 11 566 Ao H, Dekkers MJ, An Z, Xiao G, Li Y, Zhao H, Qiang X, Chang H, Chang Q, Wu D 12 13 567 (2013a) Magnetostratigraphic evidence of a Mid- onset of the Nihewan 14 15 568 Formation - implications for early fauna and hominid occupation in the Nihewan 16 17 18 569 Basin, North China. Quat Sci Rev 59: 30–42 19 20 570 Ao H, An Z, Dekkers MJ, Li Y, Xiao G, Zhao H, Qiang X (2013b) Pleistocene 21 22 571 magnetochronology of the fauna and Paleolithic sites in the Nihewan Basin: 23 24 572 Significance for environmental and hominin evolution in North China. Quat 25 26 573 Geochronol 18: 78–92 27 28 574 Arroyo, A., de la Torre, I., (2018) Pounding tools in HWK EE and EF-HR (Olduvai 29 30 31 575 Gorge, Tanzania): percussive activities in the Oldowan e Acheulean transition. J 32 33 576 Hum Evol 120: 402–421 34 35 577 Barbour GB (1924) Preliminary observation in the Kalgan Area. Bull Geol Soc China 36 37 578 3:153–168 38 39 579 Barbour GB, Licent E, Teilhard de Chardin P (1927) Geological study of the deposits 40 41 580 of the Sangkanho Basin. Bull Geol Soc China 5: 263–278 42 43 44 581 Barham LS (1987) The bipolar technology in southern Africa a replication experiment. 45 46 582 S Afr Archaeol Bull 42: 45–50 47 48 583 Barsky D, Chapon-Sao C, Bahain JJ, Beyene Y, Cauche D, Celiberti V, Desclaux E, 49 50 584 Lumley H de, Lumley M-A de, Marchal F, Moullé P-E, Pleurdeau D (2011) The 51 52 585 early Oldowan stone-tool assemblage from Fejej FJ-1, . J Afr Archaeol 9: 53 54 586 207–224 55 56 587 Berman MJ, Sievert AK, Whyte TR (2017) Form and function of bipolar lithic artifacts 57 58 59 588 from the Three Dog Site, San Salvador, Bahamas. Lat Am Antiq 10: 415–432 60 21 61 62 63 64 65

589 Binford LR, Quimby GI (1963) Indian sites and chipped stone materials in the Northern 1 2 3 590 lake MicHigan area. Fieldiana Anthropol 36: 277–307 4 5 591 Boisserie JR, Guy F, Delagnes A, Hlukso LJ, Bibi F, Beyene Y, Guillemot C (2008) 6 7 592 New Palaeoanthropological research in the Plio–Pleistocene Omo Group, Lower 8 9 593 Omo , SNNPR (Southern Nations, Nationalities and People Regions), 10 11 594 Ethiopia. Comptes Rendus Palevol 7:429–439 12 13 595 Bordaz J (1970) Tools of the Old and New Stone Age. Natural History Press, New York 14 15 596 Bordes F (1968) The Old Stone Age. McGraw–Hill, New York 16 17 18 597 Bradbury AP (2010) Bipolar reduction experiments and the examination of Middle 19 20 598 Archaic bipolar in west-central . N Am Archaeol 31: 67–116 21 22 599 Braun DR, Harris JWK, Maina DN (2009) Oldowan raw material procurement and use: 23 24 25 600 evidence from the Koobi For a formation. Archaeometry 51(1):26–42 26 27 601 Braun DR, Norton CJ, Harris JWK (2010) Africa and Asia: comparisons of the earliest 28 29 602 archaeological evidence. In: Norton, C.J., Braun, D.R., (Eds.), Asian 30 31 603 Paleoanthropology: From Africa to China and beyond. Vertebrate Paleobiology 32 33 604 and Paleoanthropology, Springer Science + Business Media B.V. 34 35 36 605 Breuil PH (1932) Le Feu Et L’industrie Lithioue et Osseuse À Choukoutien. Bull Geol 37 38 606 Soc China 11: 147–154 39 40 607 Byrne F, Proffitt T, Arroyo A, de la Torre I (2016) A comparative analysis of bipolar 41 42 608 and freehand experimental knapping products from , Tanzania. 43 44 609 Quat Int 424:58–68 45 46 610 Cao ZT (1978) The Paleolithic site of Xiaohui in Shuicheng, Guizhou Province. 47 48 49 611 Vertebr Palasiatica 16(1): 67–72 (in Chinese) 50 51 612 Chen C, Shen C, chen WY, Tang YJ (1999) 1998 excavation of the Xiaochangliang site 52 53 613 at Yangyuan, . Acta Anthropol Sin 18(3): 225–239 (in Chinese) 54 55 614 Cotterell B, Johan K (1987) Formation of Flakes. Am Antiq 52(4): 675-708 56 57 615 de la Peña A (2013) Bipolar knapping in occupation at El Palomar 58 59 616 rochshelter (yeste, southeastern spain). J Anthropol Res 69:33–64 60 22 61 62 63 64 65

617 de la Peña P, Wadley L (2014) Quartz knapping strategies in the at 1 2 3 618 Sibudu (KwaZulu-Natal, South Africa). PLoS One 9: e101534 4 5 619 de la Peña P (2015) A qualitative guide to recognize bipolar knapping for flint and 6 7 620 quartz. Lithic Technol 40:316–331 8 9 621 de la Torre I (2004) Omo revisited: evaluating the technological skills of Pliocene 10 11 622 hominids. Curr Anthropol 45: 439–465 12 13 623 de Lombera-Hermida A, Rodríguez-Álvarez XP, Peña L, Sala-Ramos R, Despriée J, 14 15 624 Moncel M-H, Gourcimault G, Voinchet P, Falguères C (2016) The lithic 16 17 18 625 assemblage from Pont-de-Lavaud (Indre, France) and the role of the bipolar-on- 19 20 626 anvil technique in the Lower and Early Middle Pleistocene technology. J 21 22 627 Anthropol Archaeol 41: 159–184 23 24 628 Deng CL, Wei Q, Zhu RX, Wang HQ, Zhang R, Ao H, Chang L, Pan YX (2006) 25 26 629 Magnetostratigraphic age of the Xiantai Paleolithic site in the Nihewan Basin and 27 28 630 implications for early human colonization of Northeast Asia. Planet Sc Lett 29 30 31 631 244: 336–348 32 33 632 Deng CL, Xie F, Liu CC, Ao H, Pan YX, Zhu RX (2007) Magnetochronology of the 34 35 633 Feiliang Paleolithic site in the Nihewan Basin and implications for early human 36 37 634 adaptability to high northern latitudes in East Asia. Geophys Res Lett 34(14): 38 39 635 L14301 40 41 636 Deng CL, Zhu RX, Zhang R, Ao H, Pan YX (2008) Timing of the Nihewan formation 42 43 44 637 and faunas. Quat Res 69: 77–90 45 46 638 Dennell R (2008) The Palaeolithic settlement of Asia. Cambridge University Press, 47 48 639 Cambridge 49 50 640 Dennell R (2010) “Out of Africa I”: Current problems and future prospects. In: Fleagle 51 52 641 J.G, et al. (eds.), Out of Africa I: The first hominin colonization of , 53 54 642 Vertebrate Paleobiology and Paleoanthropology. Springer Science+Business 55 56 643 Media B.V. 57 58 59 644 Dennell R (2013) The Nihewan Basin of North China in the Early Pleistocene: 60 23 61 62 63 64 65

645 continuous and flourishing, or discontinuous, infrequent and ephemeral 1 2 3 646 occupation? Quat Int 295: 223–236 4 5 647 Duke H, Pargeter J (2015) simple solutions to complex problems: an 6 7 648 experimental study of skill in bipolar cobble-splitting. Lithic Technol 40:349–365 8 9 649 Feng XW, Hou YM (1998) Huojiadi-a new Paleolithic site discovered in the Nihewan 10 11 650 Basin. Acta Anthropol Sin 17(4): 310–316 (in Chinese) 12 13 651 Gao X (2000) A study of flaking technology at Zhoukoudian locality 15. Acta 14 15 652 Anthropol Sin 19 (3): 199–215 (in Chinese) 16 17 18 653 Gao X, Wei Q, Shen C, Keates S (2005) New light on the earliest hominid occupation 19 20 654 in east Asia. Curr Anthropol 46: 115–120 21 22 655 Gao X, Wei Q, Li GH (2008) A report of the third excavation of the Ranjialukou 23 24 656 Paleolithic site. Acta Anthropol Sin 27(1): 1–12 (in Chinese) 25 26 657 Guan Y, Wang FG, Xie F, Pei SW, Zhou ZY, Gao X (2016) Flint knapping strategies at 27 28 658 Cenjiawan, an Early Paleolithic site in the Nihewan Basin, North China. Quat Int 29 30 31 659 400: 86–92 32 33 660 Gurtov AN, Eren MI (2014) bipolar reduction and hominin selection 34 35 661 of quartz at Olduvai Gorge, Tanzania: What’s the connection? Quat Int 322: 285– 36 37 662 291 38 39 663 Hayden B (1979) Confusion in the bipolar world: bashed pebbles and splintered pieces. 40 41 664 Lithic Technol 9: 2–7 42 43 44 665 Hiscock P (2015) Making it small in the Palaeolithic: bipolar stone-working, miniature 45 46 666 artefacts and models of core recycling. World Archaeol 47: 158–169 47 48 667 Hou YM, Wei Q, Feng XW, lin SL (1999) Re-excavation at Donggutuo in the Nihewan 49 50 668 Basin, North China. Quat Sci 19: 139–147 (in Chinese) 51 52 669 Jeske RJ, Lurie R (1993) The archaeological visibility of bipolar technology: an 53 54 670 example from the Koster site. MidCont J Archaeol 18:131–160 55 56 671 Keates SG (2000) Early and Middle Pleistocene hominid behaviour in northern China. 57 58 59 672 BAR international series, vol 863. British Archaeological Reports, Oxford 60 24 61 62 63 64 65

673 Knight J (1988) Technological analysis of the anvil (bipolar) technique. Lithic 12: 57– 1 2 3 674 87 4 5 675 Kobayashi H (1975) The experimental study of bipolar flakes. In: Swanson EH (eds), 6 7 676 : making and using stone tools. Walter de Gruyter, Berlin 8 9 677 Leakey LSB (1967) Working stone, bone, and woods. In: Singer C, Holmyard EJ, and 10 11 678 Hall AR (eds) A , Vil.1. Clarendon, Oxford 12 13 679 LeBlanc R (1992) Wedges, pieces esquillees, bipolar cores, and other things: an 14 15 680 alternative to Shott’s view of bipolar industries. N Am Archaeol 13: 1–14 16 17 18 681 Li F (2016) An experimental study of bipolar reduction at Zhoukoudian locality 1, 19 20 682 North China. Quat Int 400: 23–29 21 22 683 Li H, Li C, Sherwood NL, Kuman K (2017) Experimental flaking in the Danjiangkou 23 24 684 Reservoir Region (central China): A rare case of bipolar blanks in the Acheulean. 25 26 685 J Archaeol Sci: Reports 13: 26–35 27 28 686 Lin SL (1987) Bipolar technique and bipolar elements: a review on the discoveries 29 30 31 687 aboard. Acta Anthropol Sin 6(4): 352–360 (in Chinese) 32 33 688 Liu P, Deng CL, Li SH, Zhu RX (2010) Magnetostratigraphic dating of the Huojiadi 34 35 689 Paleolithic site in the Nihewan Basin, North China. Palaeogeogr Paleoclimatol 36 37 690 Palaeocl 298: 399–408 38 39 691 Liu Y, Hou YM, Ao H (2013) Analysis of lithic technology of Lower Pleistocene sites 40 41 692 and environmental information in the Nihewan Basin, North China. Quat Int 42 43 44 693 295:215–222 45 46 694 Liu LQ, Wang FG, Yang SX, Yue JP (2018) A preliminary report on the stone artifacts 47 48 695 of 2016 excavation of Maliang Locality 10 in the Nihewan Basin. Acta Anthropol 49 50 696 Sin 37(3): 419–427 (in Chinese) 51 52 697 Ludwig BV, Harris JWK (1998) Towards a technological reassessment of East African 53 54 698 Plio-Pleistocene lithic assemblages. In: Petraglia M, Korisetter R (eds) Early 55 56 699 Human Behavior in the Global Context: the rise and diversity of the Lower 57 58 59 700 Paleolithic period. Routledge, New York 60 25 61 62 63 64 65

701 Patterson LW, Sollberger JB (1976) The myth of bipolar flaking industries. Lithic 1 2 3 702 Technol 5: 40–42 4 5 703 Pei SW (2002) The Paleolithic site at Dachangliang in Nihewan Basin, North China. 6 7 704 Acta Anthropol Sin 21(2): 116–125 (in Chinese) 8 9 705 Pei SW, Hou YM (2002) Preliminary study on raw materials exploitation at Donggutuo 10 11 706 site, Nihewan Basin, North China. Acta Anthropol Sin 21 (suplement): 53–66 (in 12 13 707 Chinese) 14 15 708 Pei SW, Gao X, Wu XZ, Li XL, Bae CJ (2013) Middle to Late Pleistocene hominin 16 17 18 709 occupation in the Three Gorges region, South China. Quat Int 295: 237–252. 19 20 710 Pei SW, Xie F, Deng C, Jia ZX, Wang XM, Guan Y, Li XL, Ma DD, De la Torre I de 21 22 711 (2017) Early Pleistocene archaeological occurrences at the Feiliang site, and the 23 24 712 archaeology of human origins in the Nihewan Basin, North China. PLoS One 12: 25 26 713 e0187251 27 28 714 Pei SW, Deng CL, de la Torre, I. Jia ZX, Ma DD, Li XL, Wang XM (2019) 29 30 31 715 Magnetostratigraphic and archaeological records at the Early Pleistocene site 32 33 716 complex of Madigou (Nihewan Basin): implications for human adaptations in 34 35 717 North China. Paleogeogr Paleoclimatol Paleoecol 530: 176–189 36 37 718 Pei WC (1932) Notice of the discovery of quartz and other stone artifacts in the Lower 38 39 719 Pleistocene hominid-bearing of the Choukoutien Cave deposit. Bull 40 41 720 Geol Soc China 11: 109–146 42 43 44 721 Pei WC (1933) Preliminary note on some incised, cut and broken bones found in 45 46 722 association with Sinanthropus remains and lithic artifacts from Choukoutien. Bull 47 48 723 Geol Soc China 12: 105–112 49 50 724 Pei WC, Zhang SS (1985) A study of the lithic artifacts of Sinanthropus. Palaeontologia 51 52 725 Sinica. Science Press, Beijing. 53 54 726 Peng F (2012) in North China—the for SDG1 and 55 56 727 Xinjiang collections. Dissertation, University of Chinese Academy of Science (in 57 58 59 728 Chinese) 60 26 61 62 63 64 65

729 Schick K, Toth N, Wei Q, Clark JD, Etler D (1991) Archaeological perspectives in the 1 2 3 730 Nihewan Basin, China. J Hum Evol 21:13–26 4 5 731 Schick K, Toth N (1993) Making silent stones speak. Simon and Schuster, New York 6 7 732 Schick K, Toth N, Garufi G, Savage-Rumbaugh ES, Rumbaugh D, Sevcik R, (1999) 8 9 733 Continuing investigations into the stone tool-making and tool-using capabilities of 10 11 734 a (Pan paniscus). J Archaeol Sci 26: 821–832 12 13 735 Shen C, Wei Q (2004) Lithic technological variability of the Middle Pleistocene in the 14 15 736 eastern Nihewan Basin, Northern China. Asian Perspect 43: 281–301 16 17 18 737 Shott MJ (1989) Bipolar industries: ethnographic evidence and archaeological 19 20 738 implications. N Am Archaeol 10(1): 1–24 21 22 739 Shott MJ (1999) On bipolar reduction and splintered pieces. N Am Archaeol 20(3): 23 24 740 217–238 25 26 741 Soriano S, Villa P (2017) Early Levallois and the beginning of the 27 28 742 in central . PloS One 12(10): e0186082 29 30 31 743 Tabrett A (2017) The detachment of Levallois flakes using bipolar percussion at 32 33 744 Howiesons Poort Shelter, South Africa. J Archaeol Sci: Reports 15: 620–629 34 35 745 Teilhard de Chardin P, Pei W (1932) The lithic industry of the Sinanthropus deposits in 36 37 746 Choukoutien. Bull Geol Soc China 11: 315–362 38 39 747 Vergès JM, Ollé A (2011) Technical microwear and residues in identifying bipolar 40 41 42 748 knapping on an anvil: experimental data. J Archaeol Sci 38(5): 1016–1025. 43 44 749 Wang HQ, Deng CL, Zhu RX, Wei Q, Hou YM, Boeda E (2005) Magnetostratigraphic 45 46 750 dating of the Donggutuo and Maliang Paleolithic sites in the Nihewan Basin, North 47 48 49 751 China. Quat Res 64: 1–11 50 51 752 Wang HQ, Deng CL, Zhu RX, Xie F (2006) Paleomagnetic dating of the Cenjiawan 52 53 753 Paleolithic site in the Nihewan Basin, Northern China. Sci China Ser D 49: 295– 54 55 754 303 (in Chinese) 56 57 755 Wei Q, Meng H, Cheng SQ (1985) New Paleolithic site from the Nihewan Beds. Acta 58 59 756 Anthropol Sin 4(3): 223–232 (in Chinese) 60 27 61 62 63 64 65

757 Wei Q, Xie F (1989) Selected treatises on Nihewan. Beijing: Cultural relics Publishing 1 2 3 758 House (in Chinese) 4 5 759 Wei Q (1994) Banshan Paleolithic site from the Lower Pleistocene in the Nihewan 6 7 760 Basin in North China. Acta Anthropol Sin 13(3): 223–238 (in Chinese) 8 9 761 Wei Q (2014) New observations on stone artefacts from the Donggutuo site. Acta 10 11 762 Anthropol Sin 33(3): 223–232 (in Chinese) 12 13 763 Wynn T, Hernández-Aguilar RA, Marchant LF, McGrew WC (2011) ‘‘An ape’s view 14 15 764 of the Oldowan” revisited. Evol Anthropol: Issues News Rev 20 (5): 181–197 16 17 18 765 Xie F, Cheng SQ (1990) Paleoliths excavated in Cenjiawan village, Yangyuan County, 19 20 766 Hebei Province. Acta Anthropol Sin 9(3): 265–272 (in Chinese) 21 22 767 Xie F, Li J (1993) On the cultural relics and archaeological features of the Lower 23 24 768 Paleolithic site at Cenjiawan village. Acta Anthropol Sin 12(3): 224–234 (in 25 26 769 Chinese) 27 28 770 Xie F, Li J, Liu L (2006) Paleolithic archaeology in the Nihewan Basin. Huashan 29 30 31 771 Literature＆Art Press, Shijiahzuang (in Chinese) 32 33 772 Xie GM, Lin Q (2017) A study of Ruilengzaji technique in Guangxi Province and 34 35 36 773 relevant issues. Archaeology and Cultural relics 2017(1):52–61 37 38 774 Yang SX, Hou YM, Yue JP, Petraglia MD, Deng CL, Zhu RX (2016) The lithic 39 40 775 assemblages of Xiaochangliang, Nihewan Basin: implications for early 41 42 776 Pleistocene hominin behaviour in North China. PLoS One 11: e0155793 43 44 777 Yang SX, Petraglia MD, Hou YM, Yue JP, Deng CL, Zhu RX (2017) The lithic 45 46 778 assemblages of Donggutuo, Nihewan basin: knapping skills of early Pleistocene 47 48 49 779 hominins in North China. PLoS One 12: e0185101 50 51 780 Yang SX, Deng CL, Zhu RX, Petraglia MD (2019) The Paleolithic in the Nihewan 52 53 781 Basin, China: Evolutionary history of an Early to Late Pleistocene record in 54 55 782 Eastern Asia. Evol Anthropol https://doi.org/10.1002/evan.21813| 56 57 783 Yang SX, Wang FG, Xie F, Yue JP, Deng CL, Zhu RX, Michael D Petraglia (2020) 58 59 784 Technological innovations at the onset of the Mid-Pleistocene Climate Transition 60 28 61 62 63 64 65

785 in high-latitude East Asia. Natl Sci Rev https://doi.org/10.1093/nsr/nwaa053 1 2 3 786 You YZ, Tang YJ, Li Y (1980) Discovery and significance of the Xiaochangliang site 4 5 787 from the Nihewan formation. Chinese Sci Bull 8: 365–367 (in Chinese) 6 7 788 Zhang SS (1998) The bipolar technique in the Paleolithic industry in China. In: Peking 8 9 789 University (ed) Proceedings of the International Conference on “Chinese 10 11 790 Archaeology Enters the Twenty-first Century”. Science Press, Beijing (in Chinese) 12 13 791 Zhou TR, Li HZ, Liu QS, Li RQ, Sun XP (1991) Study on the Paleogeography 14 15 792 of Nihewan Basin. Science Press, Beijing (in Chinese) 16 17 18 793 Zhu RX, Hoffman KA, Potts R, Deng CL, Pan YX, Guo B, Shi CD, Guo ZT, Yuan BY, 19 20 794 Hou YM, Huang WW (2001) Earliest presence of humans in northeast Asia. 21 22 795 413: 413–417 23 24 796 Zhu RX, An ZS, Potts R, Hoffman KA (2003) Magnetostratigraphic dating of early 25 26 797 humans in China. Earth-Sci Rev 61: 341–359 27 28 798 Zhu RX, Potts R, Xie F, Hoffman KA, Deng CL, Shi CD, Pan YX, Wang HQ, Shi RP, 29 30 31 799 Wang YC, Shi GH, Wu NQ (2004) New evidence on the earliest human presence 32 33 800 at high northern latitudes in northeast Asia. Nature 431: 559–562 34 35 801 36 37 802 38 39 803 40 41 804 42 43 44 805 45 46 806 47 48 807 49 50 808 51 52 809 53 54 810 55 56 811 57 58 59 812 60 29 61 62 63 64 65

813 FIGURE CAPTIONS 1 2 3 814 Fig. 1 Position of Nihewan Basin, showing the location of early Pleistocene sites. a & 4 5 815 b Nihewan basin in North China. c relevant sites in the eastern part of the Nihewan 6 7 816 Basin. The area marked with dark brown color represents the Sinian rock system where 8 9 817 chert was formed; the area outlined with light blue color refers to the Jurassic system 10 11 818 from which volcanic rocks derive. 12 13 819 14 15 820 Fig. 2 Different motions potentially employed in bipolar knapping. a bipolar knapping 16 17 18 821 with orthogonal hammer (BO1). b bipolar splitting (BS). c bipolar knapping with 19 20 822 orthogonal hammer, but obliquely- placed core (BO2). d bipolar knapping with oblique 21 22 823 hammer (BOB). 23 24 824 25 26 825 Fig. 3 a Bipolar core (right) and refitted set after the BO1 experiment (left); b bipolar 27 28 826 core before and after the BS experiment; c bipolar knapping; d freehand knapping. 29 30 31 827 32 33 828 34 35 829 Fig. 4 Relevant attributes in bipolar and freehand experimental sets. a Impact point on 36 37 830 scars. car1: impact point on the striking platform; scar2: distal impact point resulting 38 39 831 from rebound against the anvil; scar3: proximal impact point on a scar that travels 40 41 832 across the core face; scar4: bipolar scar containing 2 bipolar impact points on a scar 42 43 44 833 that travels across the core face. b Flake profile shapes according to ventral faces (right 45 46 834 side) (after Peng 2012). c Crushing intensity on the striking platform of flake or core 47 48 835 (after de la Peña 2015). absent: pieces with no scars shorter than 1 cm. slight: pieces 49 50 836 with 1~2 layers of scars shorter than 1 cm. medium: pieces with 3 or more layers of 51 52 837 scars shorter than 1cm. heavy: pieces with crushed edges and abundant microfractures. 53 54 838 55 56 57 839 Fig. 5 BO1 (a and b) and BS (c and d) cores produced in bipolar experiments. 58 59 840 60 30 61 62 63 64 65

841 Fig. 6 Bipolar flakes produced in the experiments. a-d and g are flakes with punctiform 1 2 3 842 or linear striking platforms and heavy crushing on the platform. e and f are flakes with 4 5 843 impact points in both the proximal and distal ends of ventral surfaces. 6 7 844 8 9 845 Fig. 7 Technological comparisons between bipolar and freehand flakes. a Crushing. b 10 11 846 Termination. c Profile shape. d Platform. 12 13 847 14 15 16 848 Fig. 8 Dimensional comparison of bipolar and freehand flakes. a Length×mesial 17 18 849 width. b Length. 19 20 21 850 22 23 851 Fig. 9 Cutting length of flakes (a) and products excluding flakes (b). 24 25 852 26 27 853 Fig.10 Summary of Paleolithic sites of early and middle Pleistocene in Nihewan 28 29 854 basin. a Complete lithic assemblages; b Bipolar products identified in each site. 30 31 855 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 31 61 62 63 64 65 Ma et al.Tables-R1-Clear

1 Technological adaptations of early humans at the Lower Pleistocene

2 Nihewan Basin, North China: the case of the bipolar technique 3 Dongdong Ma, Shuwen Pei, Ignacio de la Torre, Hao Li

4

5 TABLE CAPTIONS

6 7 Table 1 Age and main technological attributes of early Pleistocene sites in the Nihewan

8 Basin. Dating Raw material1 Archaeological Dating Artefacts (n) Site references (%) references (Ma) Chert Dolomite Total Bipolar Majuangou III 1.66 (Zhu et al. 2004) main 111 0 (Gao et al. 2005) MajuangouII 1.64 (Zhu et al. 2004) main 226 0 (Xie et al. 2006) MajuangouI 1.55 (Zhu et al. 2004) main 215 0 (Xie et al. 2006) Xiaochangliang 1.36 (Zhu et al. 2001) 96.4 1468 439 (Yang et al. 2016) main 901 48 (Chen et al. 1999) 98.2 804 0 (You et al. 1980) Dachangliang 1.36 (Deng et al. 2006) main 33 0 (Pei 2002) Banshan 1.32 (Zhu et al. 2004) 63.2 95 0 (Wei 1994) Feiliang 1.2 (Deng et al. 2007) main 982 0 (Pei et al. 2017) Madigou 1.2 (Pei et al. 2019) 44.7 33.4 1517 213 (Pei et al. 2019) Donggutuo 1.1 (Wang et al. 2005) 96 2315 269 (Yang et al. 2017) 97.85 1667 9 (Wei 2014) main 702 0 (Hou et al. 1999) 85.81 1442 1 (Wei et al. 1985) Cenjiawan 1.1 (Wang et al. 2006) 92.7 1625 0 (Guan et al. 2016) 94.6 891 6 (Xie and Cheng 1990) 89.09 5.56 486 0 (Xie and Li 1993) Huojiadi 1 (Liu et al. 2010) 78.3 6.7 60 1 (Feng and Hou 1998) Maliang 0.8-0.9 (Wang et al. 2005) 54 197 0 (Liu et al. 2018) 9 1Where no quantitative data is available, predominance is noted 10 11 12 13 14 15 16 17

1

18 19 Table 2 Original dimensions (mm) and weight (g) of cores used in the experiments. Experiment Blanks Length Width Thickness Weight BO1 1 Block 107 82 60 536 BO1 2 Block 96 72 53 621 BO1 3 Block 91 84 48 588 BO1 4 Block 107 94 69 1078 BO1 5 Block 97 89 56 782 BO1 6 Cobble 91 85 64 687 BO1 7 Block 81 76 47 585 BO1 8 Block 85 67.4 34.2 276.4 BS 1 Cobble 100 58 44 370 BS 2 Cobble 87 51 49 387 BS 3 Block 105 58 55 467 BS 4 Cobble 88.2 55.4 47.3 296 BS 5 Cobble 69.4 51.3 46.7 243 BS 6 Cobble 80 78 60 382 FH 1 Block 125 124 60 1096.7 FH 2 Block 165 151 54 2011.2 FH 3 Cobble 103 86 62 757.9 FH 4 Block 134 91 53 883 FH 5 Cobble 141 91 76 1221 FH 6 Block 97 69 65 818 FH 7 Block 134 73 59 791 FH 8 Block 106 105 77 1535 FH 9 Cobble 98 88 71 676 FH 10 Block 100 80 55 491 FH 11 Block 94 86 63 718 20 21

22

23

24

25

26

27

28

29

30 2

31

32 Table 3 Technological and morphological variables considered in this study. Category Attribute Variable1 Core Size Length; Width; Thickness Weight (grams) Crushing Absent; Slight; Medium; Heavy (Fig. 4: C) Platform Punctiform or linear; plain Scar Bipolar scar; Scar with one impact point Axial length (mm) Flake Size Length; Width (proximal, medial and distal); thickness (proximal medial and distal Weight (grams) Platform Punctiform; linear; plain Bulb2 Absent; Present Profile3 Concave; Straight; Convex; Irregular (Fig.4: B) Crushing Absent; Slight; Medium; Heavy Cutting edge4 Number of cutting edges; Cutting Length Termination Feather; Axial; Plunge Axial length (mm) Fragments Type Transversal fragment; Siret; Shatter Size Length; Width; Thickness Weight (grams) 33 1 All dimensions in mm. 34 2 Bulbs with a prominent convexity fall into the ‘present’ category, and flakes with 35 inconspicuous/flat bulbs are grouped within the ‘absent’ category. 36 3 The profile shape considered the morphology of the overall ventral face of the flake. 37 4 Cutting edge are considered here as those with a sharp profile and an angle lower than 60°. 38 39 40 41 Table 4 Distribution of debitage categories produced during the experiments. Technique Experiment Products Flake Transverse Split Shatter fracture fracture N N N Weight N Weight N Weight N Weight (g) (g) (g) (g) Bipolar 14 298 123 2078.6 35 347.3 35 281.3 105 961.2

Freehand 11 279 107 2991.1 28 139 33 260 111 1522.2 42 43 44 45

3

46 47 Table 5 Percentages of different crushing stages in cores. Absent Slight Medium Heavy BO1 0 9.1 36.4 54.5 BS 0 0 33.3 66.7 Freehand 25 50 25 0 48 49 Table 6 Dimensions (mm and g) of bipolar and freehand flakes. Bipolar Freehand Min Max Mean St. Dev Min Max Mean St. Dev Length 13 69.2 33.6 13.2 11.1 78.3 37 16.7 Proximal width 4.7 85.1 21.6 11 8.6 81.9 25.3 11.9 Mesial width 10.5 56.3 25.7 10.3 9.8 80.1 30 13.4 Distal width 3.5 59.9 21.2 12 5.5 82.3 22.8 13.2 Proximal 1.7 30.3 9 5.7 2 27.4 9.4 5.9 Thickness Mesial thickness 2.8 32.7 9.6 6.1 1.6 34.2 10.4 6.7 Distal thickness 1.3 34.9 7.3 6.7 1.3 45.4 7 5.8 Weight 1.3 141 15 23.6 0.7 272 21.6 37.8 50 51 Table 7 Flake and length productivity in bipolar and freehand products. Total Pieces with Pieces with Total number Total length cutting no cutting of cutting (mm) of edges edges edges cutting edges Bipolar Flakes 82 73 9 129 3368.2 Other debitage 211 173 38 264 6428.7 Freehand Flakes 94 93 1 181 5591.7 Other debitage 225 206 19 337 9164.6 52 53 54 Table 8 Knapping efficiency based on flaking strikes involved in the production of

55 debitage and cutting edges. Technique Flakes per strike Other debitage Cutting edges per Cutting edge length per strike strikes per strike Mean St. Dev Mean St. Dev Mean St. Dev Mean St. Dev Bipolar 0.08 0.039 0.21 0.086 0.25 0.12 6.15 3.03 Freehand 0.12 0.054 0.3 0.132 0.45 0.211 12.21 6.64 56

4

Fig. 1 Position of Nihewan Basin, showing the location of early Pleistocene sites. a & b Click here to access/download;Figure;Fig.1 R1.png Nihewan basin in North China. c relevant sites in the eastern part of the Nihewan Fig. 2 Different motions potentially employed in bipolar knapping. a bipolar knapping Click here to access/download;Figure;Fig.2.jpg with orthogonal hammer (BO1). b bipolar splitting (BS). c bipolar knapping with Fig. 3 a refitted core and core of a bipolar experiment; b bipolar core before and after Click here to access/download;Figure;Fig.3.jpg the experiment; c bipolar knapping; d freehand knapping. Fig. 4 Relevant attributes in bipolar and freehand experimental sets. a Impact point on Click here to access/download;Figure;Fig.4-R1.png scars. car1: impact point on the striking platform; scar2: distal impact point resulting Fig. 5 BO1 (a and b) and BS (c and d) cores produced in bipolar experiments. Click here to access/download;Figure;Fig.5.jpg Fig. 6 Bipolar flakes produced in the experiments. a-d and g are flakes with punctiform Click here to access/download;Figure;Fig.6.jpg or linear striking platforms and heavy crushing on the platform. e and f are flakes with Fig. 7 Technological comparisons between bipolar and freehand flakes. a Crushing. b Click here to access/download;Figure;Fig. 7-R1.png Termination. c Profile shape. d Platform Fig. 8 Dimensional comparison of bipolar and freehand flakes. a Length×mesial width. Click here to access/download;Figure;Fig.8.jpg b Length. Fig. 9 Cutting length of flakes (a) and products excluding flakes (b). Click here to access/download;Figure;Fig.9.jpg Fig.10 Summary of Paleolithic sites of early and middle Click here to access/download;Figure;Fig.10.jpg Pleistocene in Nihewan basin. a Complete lithic assemblages; b