Lithic Production Strategies at the Early Pleistocene Site of Bizat Ruhama, Israel 9781407313313, 9781407342948
Bizat Ruhama is an Early Pleistocene site located on the fringe of the Negev Desert, Israel, in the southern coastal pla
176 49 37MB
English Pages [175] Year 2014
Front Cover
Title Page
Copyright
Table of Contents
List of Tables
List of Figures
List of Plates
CHAPTER 1: INTRODUCTION
CHAPTER 2: OUTLINE OF THE RESEARCH
CHAPTER 3: THE CONTEXT OF THE SITE AND THE LITHIC ASSEMBLAGES
CHAPTER 4: LITHIC RAW MATERIALS
CHAPTER 5: FLAKE PRODUCTION
CHAPTER 6: SECONDARY KNAPPED FLAKES
CHAPTER 7: EXPERIMENTAL KNAPPING
CHAPTER 8: DISCUSSION
CHAPTER 9: CONCLUDING REMARKS
LIST OF REFERENCES
APPENDIX
Recommend Papers
- Author / Uploaded
- Yossi Zaidner
File loading please wait...
Citation preview
BAR S2685 2014 ZAIDNER LITHIC PRODUCTION STRATEGIES AT BIZAT RUHAMA
B A R
2685 Zaidner cover.indd 1
Lithic Production Strategies at the Early Pleistocene Site of Bizat Ruhama, Israel Yossi Zaidner
BAR International Series 2685 2014 05/12/2014 09:21:41
Lithic Production Strategies at the Early Pleistocene Site of Bizat Ruhama, Israel Yossi Zaidner
BAR International Series 2685 2014
ISBN 9781407313313 paperback ISBN 9781407342948 e-format DOI https://doi.org/10.30861/9781407313313 A catalogue record for this book is available from the British Library
BAR
PUBLISHING
TABLE OF CONTENTS CHAPTER 1: INTRODUCTION
1
APPROACHES AND QUESTIONS IN THE EARLY PLEISTOCENE LITHIC STUDIES: A SHORT REVIEW
3
Oldowan – the “sharp-edged” technology
3
The chaîne opératoire approach
4
CHAPTER 2: OUTLINE OF THE RESEARCH
6
RESEARCH OBJECTIVES AND QUESTIONS
6
STUDIED ASSEMBLAGES AND SITES
6
Bizat Ruhama
6
Supplementary material
7
Nahal Hesi
7
Evron Quarry
7
METHODOLOGY FOR THE STUDY OF THE LITHIC TECHNOLOGICAL STRATEGIES Study of raw material procurement strategies
7 7
Lithology
8
Maximum length
8
Width
8
Thickness
8
Shape
8
Observations on raw materials at the site
9
Study of raw material exploitation strategies
9
Study of lithic production strategies
9
Knapping modes
9
Knapping techniques
9
Debitage methods
9
Elements in the chaîne opératoire
10
Recording of the artifacts
10
Experimental knapping
11
Statistical analysis
11
CHAPTER 3: THE CONTEXT OF THE SITE AND THE LITHIC ASSEMBLAGES LOCATION AND GEOLOGY
12 12
Location
12
Present and past climate and environment
12
Plio-Pleistocene geology
14
Pleshet Formation
15
Ahuzam Formation
16
The Quaternary
16
THE SITE
16
The stratigraphic sequence
17
Chronology
19
The 1996 excavations
19
The 2004–5 survey and excavations
22
Area A and Trench 5 (BRAT5)
22 iii
Bizat Ruhama Trench 1 (BRT1)
28
Bizat Ruhama Trench 2 (BRT2)
28
Bizat Ruhama Trench 3 (BRT3)
28
Faunal remains from BRAT5
28
SITE FORMATION
29
Site size and size of individual concentrations
30
The depositional settings
30
Stratum 5
30
Stratum 4
30
Stratum 3
31
Preservation of finds and its implications for site formation
31
Horizontal distribution
32
Summary: site formation history
34
CHAPTER 4: LITHIC RAW MATERIALS
35
LITHIC RAW MATERIALS IN THE BIZAT RUHAMA AREA Random sampling and characterization of raw material sources
35 35
Lithological composition of the sampled exposures
35
Shape of the pebbles in the sampled exposures
36
Length of the pebbles in the sampled exposures
36
Selective sampling
36
Sample A
39
Sample B
39
Summary: raw material accessibility and variability
39
LITHIC RAW MATERIALS AT THE SITE
39
CHAPTER 5: FLAKE PRODUCTION
42
FLAKED PIECES
42
Fractured flat pebbles
42
Fractured pebbles and pebbles with 1–3 removals
43
Cores
43
Orthogonal cores with series of unidirectional removals (N=30)
43
Multifacial multidirectional cores (N=10; Plates 4:2; 6:3; 7:8, 9)
45
Preferential surface cores (N=4; Plates 5:2; 8:1)
45
Bipolar cores (N=19; Plates9, 10)
45
Exhausted cores (N=48; Plate 6:4–6; 7:1–7, 10)
46
Flaked pieces: an overview
46
DETACHED PIECES
48
Complete flakes
50
Length and width
50
Thickness and width/thickness ratio
50
Butt
51
Dorsal face features
54
Ventral face features
54
Distal end features: signs of bipolar technique
54
Lateral edge curvature
55 iv
Siret accidental breaks
55
Longitudinal breaks
55
Angular fragments
55
Detached pieces: an overview
55
CHAPTER 6: SECONDARY KNAPPED FLAKES
57
FLAKED FLAKES
57
ANVIL FLAKES
58
MODIFIED FLAKES
60
Morphology of scars
61
Edge modification type
61
Edge angle
61
Number and position of modified edges
61
The assemblages
61
SECOND-GENERATION FLAKES
64
CHAPTER 7: EXPERIMENTAL KNAPPING
67
EXPERIMENTS IN FLAKE PRODUCTION
67
Description of experimental DP assemblages
70
Characteristics of experimentally produced flakes
70
Composition of the assemblages
72
Free-hand versus bipolar reduction: what differences should we expect in the DP assemblages?
73
Summary
74
EXPERIMENTS IN BREAKAGE OF FLAT PEBBLES
74
ANVIL-SUPPORTED KNAPPING OF FLAKES
75
The experiment
75
Description of the experimental assemblage
77
Secondary knapped flakes
77
Second-generation flakes
79
Archaeological vs. experimental data
79
CHAPTER 8: DISCUSSION
115
PATTERNS OF INTRA-SITE VARIABILITY
115
RAW MATERIAL EXPLOITATION STRATEGIES
116
Raw material transport
117
Selectivity in raw material acquisition and exploitation
118
The influence of pebble size on patterns of raw material selection
119
The influence of raw material quality on selection patterns
119
Raw material at the Acheulian site of Nahal Hesi: a case study in raw material acquisition and exploitation in the Bizat Ruhama area
120
The complexity of raw material exploitation strategies at Bizat Ruhama
122
KNAPPING STRATEGIES AT BIZAT RUHAMA
122
Debitage techniques
122
Pebble reduction strategies
124
The organization of debitage and knapping skills Secondary flake knapping
126 127
v
Selection of flakes for further knapping
127
Techniques and goals of secondary flake knapping
128
LITHIC PRODUCTION STRATEGIES AT BIZAT RUHAMA AND THE EARLY PLEISTOCENE ARCHAEOLOGICAL RECORD
130
CHAPTER 9: CONCLUDING REMARKS
135
LIST OF REFERENCES
137
APPENDIX
152
APPENDIX 1. DESCRIPTION OF RAW MATERIAL OUTCROPS SAMPLED DURING THE SURVEY
152
APPENDIX 2. RECORDS ON ARTIFACTS
153
APPENDIX 3. DESCRIPTION OF THE KNAPPING EXPERIMENTS
156
vi
LIST OF TABLES Table 2.1. Observations made on material before, during and after the experimental knapping.
11
Table 3.1. Size of the excavated areas, density of the finds and microstratigraphy.
20
Table 3.2. Surface condition of the artifacts.
21
Table 3.3. Composition of the lithic assemblages.
21
Table 4.1. Rock type frequencies among the pebbles in the sampled exposures.
36
Table 4.2. Description of the pebbles collected during selective sampling (Sample A).
39
Table 4.3. Shape of the pebbles in the Bizat Ruhama assemblages.
40
Table 4.4. Chert type frequencies in three archaeological assemblages.
41
Table 5.1. Breakdown of FPs in Bizat Ruhama assemblages.
42
Table 5.2. Breakdown of FPs by raw material type.
43
Table 5.3. Descriptive statistics of FP assemblages.
44
Table 5.4. Debitage systems in different Bizat Ruhama assemblages.
47
Table 5.5. Incipient cones on core surfaces.
48
Table 5.6. Signs of anvil impact on the distal ends of the debitage surface of different core types.
49
Table 5.7. Breakdown of detached piece assemblages.
49
Table 5.8. Raw material frequencies in BR1996 and BRAT5.
50
Table 5.9. Descriptive statistics of DP assemblages.
51
Table 5.10. Crosstabulation of bulb type and signs of impact on the distal end of complete flakes in the BRAT5 and BR1996 assemblages.
56
Table 5.11. Crosstabulation of butt type and signs of impact on the distal end of complete flakes in the BRAT5 and BR1996 assemblages.
56
Table 6.1. The secondary knapped flake assemblages.
57
Table 6.2. Breakdown of secondary knapped flake assemblages by raw material type.
58
Table 6.3. Crosstabulation of raw material type and exploited face of flaked flakes.
59
Table 6.4. Dorsal surface features of anvil flakes: types of impact marks.
59
Table 6.5. Features identified at the intersection between ventral and broken/lateral surfaces of anvil flakes.
59
Table 6.6. Shape of scars at the intersection between the ventral and broken/lateral surfaces of anvil flakes.
59
Table 6.7. Crosstabulation of dorsal and ventral surface features of anvil flakes.
60
Table 6.8. Broken surface features of anvil flakes.
60
Table 6.9. Breakdown of modified flake assemblages by techno-morphological type.
63
Table 6.10. Breakdown of modified flake assemblages by raw material type in BRAT5 and BR1996.
63
Table 6.11. Breakdown of modified flake assemblages by DP category.
64
Table 6.12. One-way ANOVA test for different techno-morphological types of modified flake assemblages.
64
Table 6.13. Impact marks on dorsal ridges of modified flakes.
65
Table 6.14. Kolmogorov-Smirnov test of significance for SGFs and complete flakes.
66
Table 7.1. Descriptive statistics and composition of the experimental assemblages.
68
Table 7.2. Distal end features of flakes produced during bipolar technique experiments.
72
Table 7.3. Frequency of accidents in free-hand technique experiments by raw material type.
72
Table 7.4. Crosstabulation of types of bulb of percussion and distal end features of complete flakes in bipolar technique assemblages.
73
Table 7.5. Descriptive statistics of experiments in breakage of flat pebbles.
74
Table 7.6. Anvil-supported knapping of flakes: length and width of flakes selected for anvil knapping.
75
Table 7.7. Anvil-supported knapping of flakes: composition of the experimental assemblages.
77
Table 7.8. The experimental assemblage: signs of anvil impact.
78
Table 7.9. The experimental assemblage: signs of hammerstone impact.
78
vii
Table 7.10. Descriptive statistics of experimentally produced SGFs.
79
Table 8.1. Cortical cover of complete flakes in the Bizat Ruhama assemblages compared to the experimental assemblages.
118
Table 8.2. Weight of the lithic assemblages and mean weight of the pebbles collected during the raw material survey (g). 119 Table 8.3. Rock types at Nahal Hesi and Bizat Ruhama.
120
Table 8.4. Debitage systems at Bizat Ruhama.
125
Table 8.5. Average length of complete flakes from selected Early Pleistocene assemblages in the Levant and PlioPleistocene assemblages from Eastern and Central Africa.
viii
132
LIST OF FIGURES Figure 2.1. The three main types of chert in the Bizat Ruhama area.
8
Figure 2.2. The shape of the pebbles. “Flat discoid” refers to small, flat, rounded pebbles usually associated with marine beach environments.
8
Figure 3.1. Location map.
12
Figure 3.2. a) Undulating loess hills in the Bizat Ruhama area, view from the site to the east; b) Exposures of Eocene chalk in the Nahal Shiqma riverbeds, view from Tel Nagila to the north; c) Exposures of Pleshet Formation beachrock overlying Eocene chalk in Nahal Sad, view to the north; d) The Bizat Ruhama badland field, view from the site to the south.
13
Figure 3.3. Map of the study area on the northeastern margin of the Negev Coastal Plain. S1–S7 refer to conglomerate exposures sampled during the raw material study. Figure 3.4. Stratigraphic chart of the southern part of the coastal plain (after Gvirtzman and Buchbinder 1969).
14 15
Figure 3.5. Composite stratigraphic section and microstratigraphy of the excavated areas at Bizat Ruhama. The composite stratigraphic chart is based on study of the Bizat Ruhama type-section (Strata 1–5; Laukhin et al. 2001, Mallol et al. 2011, Ronen et al. 1998) and on Bar-Yosef 1964. Figure 3.6. Bizat Ruhama, the site and excavation areas, view to the southeast.
17 18
Figure 3.7. The southern channel with BRAT5 and the 17-meter-high slope (the type-section of the site), view to the northeast.
18
Figure 3.8. Plan of the Bizat Ruhama site
20
Figure 3.9. BR1996, horizontal (a) and vertical (b) distribution of the finds; larger black dots represent bones. The thin curved line marks the edge of the erosion slope.
22
Figure 3.10. BRAT5, view to the southeast.
23
Figure 3.11. BRAT5, vertical distribution of bones and artifacts.
24
Figure 3.12. BRAT5, horizontal distribution of artifacts and bones. .
24
Figure 3.13. BRAT5, densities of bones and artifacts.
24
Figure 3.14. BRAT5, horizontal distribution of artifacts according to size-groups.
25
Figure 3.15. BRAT5, distribution of bones identified to species or size-class.
25
Figure 3.16. BRAT5, spatial distribution of artifacts by group within the lithic assemblage. The bipolar group consists of bipolar cores and flakes.
26
Figure 3.17. BRAT5, spatial distribution of artifacts by raw material group.
27
Figure 3.18. Length histogram of the artifacts from all excavated areas.
33
Figure 4.1. Shape frequencies of the pebbles in the sampled exposures.
37
Figure 4.2. Conglomerate exposures sampled during the raw material study.
37
Figure 4.3. Length distribution of the pebbles in the sampled exposures.
38
Figure 4.4. Length distribution for different rock types among the pebbles in the sampled exposures.
38
Figure 4.5. Chert pebbles from the Bizat Ruhama area. 1) Mishash (brecciated) chert; 2) Eocene chert.
40
Figure 4.6. Length of the pebbles (mm) in the Bizat Ruhama assemblages.
40
Figure 5.1. Length/thickness scattergram of FP types.
43
Figure 5.2. Schematic chart of the orthogonal unidirectional debitage system. a) Cortex removal; b) Sequence of unidirectional removals; c) and d) Two additional series of unidirectional removals struck from different striking platforms and on different debitage surfaces.
44
Figure 5.3. Schematic chart of the preferential surface debitage system. a) Cortex removal; b) Sequence of unidirectional unifacial removals; c) Butt rectification; d) Further exploitation of the debitage surface.
45
Figure 5.4. Box plot graph presenting angles between striking platform and debitage surface by debitage method.
45
Figure 5.5. Amount of cortex on cores and exhausted cores.
46
ix
Figure 5.6. Debitage systems by raw material type.
47
Figure 5.7. Technological and morphological characteristics of flakes in Bizat Ruhama assemblages.
53
Figure 6.1. Length scatterplot of complete flakes, flaked flakes, anvil flakes and modified flakes.
58
Figure 6.2. Characteristics of edges of modified flakes. a) Edge modification type; b) Location of modified edges; c) Angle of edge by modification type; d) Edge modification type of pointed pieces. Figure 6.3. Combined length histogram of complete flakes and SGFs in Bizat Ruhama assemblages.
62 65
Figure 6.4. SGFs and complete flakes in Bizat Ruhama assemblages. a) Length; b) Thickness; c) Width/length ratio; d) Thickness/width ratio.
66
Figure 7.1. Two possible ways of fracturing pebble. a) Pebbles of more discoidal shape often provide an angle to initiate knapping by free-hand removal of the end of the pebble; the fracture surface will be at an acute angle to the cortical surface of the pebble. b) Pebbles of spherical shape can be fractured by the bipolar technique; in this case the fracture surface produced is usually at an obtuse angle to the cortical surface of the pebble.
69
Figure 7.2. Use of the bipolar technique during the experimental study. a) Knapping of the pebble; b) Knapping of the flake.
69
Figure 7.3. Characteristics of flakes produced during the knapping experiments. a) Butt types; b) Types of bulb of percussion. BT – bipolar technique; HH – free-hand hard hammer technique; Anvil butt – plain butt formed by anvil impact.
71
Figure 7.4. Schematic representation of the experiment in anvil-supported knapping of flakes.
76
Figure 7.5. Schematic representation of the sequence of anvil-supported knapping of the flake.
76
Figure 7.6. Angle of the edge of different types of products of anvil knapping of flakes.
78
Figure 8.1. Chert types used at the site compared to sampled exposures of the Pleshet Formation.
117
Figure 8.2. Length of cores and complete flakes by raw material type in the Bizat Ruhama assemblages.
117
Figure 8.3. Length of pebbles at the site compared to pebbles from the Pleshet Formation.
118
Figure 8.4. Length of artifacts by raw material type at Bizat Ruhama and Nahal Hesi.
121
Figure 8.5. Rock types at Nahal Hesi by major lithic group.
121
Figure 8.6. Composition of the Bizat Ruhama archaeological and experimental assemblages (BT = bipolar technique; FH = free-hand hard hammer technique). Figure 8.7. Distal end features of the flakes in the Bizat Ruhama archaeological and experimental assemblages.
123 123
Figure 8.8. Shapes of bulb of percussion of the flakes in the Bizat Ruhama archaeological and experimental assemblages.
123
Figure 8.9. Butt types of the flakes from the Bizat Ruhama archaeological and experimental assemblages.
124
Figure 8.10. Flake selection for secondary knapping: composition of DPs and secondary knapped flake assemblages by a) Type of DP; b) Type of raw material.
127
Figure 8.11. Thickness histograms for complete and secondary knapped flakes.
128
Figure 8.12. Scatterplot of average thickness and edge angle of secondary knapped flakes, SGFs and complete flakes.
129
Figure 8.13. Length values of the longest axis of the scars on flaked flakes, exhausted cores, Clactonian notches and cores.
130
Figure 8.14. The lithic production scheme at Bizat Ruhama.
131
x
LIST OF PLATES Plate 1. Bizat Ruhama archaeological assemblages. Large brecciated chert core weighing 3,8 kg found in BRT6.
81
Plate 2. Bizat Ruhama archaeological assemblages and experimental assemblages. A. Flat discoid pebbles broken on an anvil during the experiment: pebble fragments: B. Broken flat discoid pebbles with signs of impact on broken surface.
82
Plate 3. Bizat Ruhama archaeological assemblages. Orthogonal multifacial unidirectional core with subspheroid shape: 1; Fractured pebbles with signs of opposite impact: 2,3.
83
Plate 4. Bizat Ruhama archaeological assemblages. Orthogonal unifacial unidirectional cores: 1,3,5; multifacial multidirectional core: 2; orthogonal multifacial unidirectional core with subspheroid shape: 4; flat discoid pebble: 6. Cores 3,5 exhibit signs of anvil impact on the distal end of the debitage surfaces.
84
Plate 5. Bizat Ruhama archaeological assemblages. Orthogonal multifacial unidirectional cores with polyhedral/ subspheroid shapes: 1,4; preferential surface core: 2; orthogonal multifacial unidirectional core: 3.
85
Plate 6. Bizat Ruhama archaeological assemblages. Chopper on flat discoid pebble: 1; multifacial multidirectional cores: 2,3; exhausted cores: 4,5,6.
86
Plate 7. Bizat Ruhama archaeological assemblages. Exhausted cores with removals of thin flakes on the last stage of the reduction: 1-7,10; multifacial multidirectional cores 8,9.
87
Plate 8. Bizat Ruhama archaeological assemblages. Preferential surface core: 1; Flaked flakes: 2-5.
88
Plate 9. Bizat Ruhama archaeological assemblages. Bipolar cores.
89
Plate 10. Bizat Ruhama archaeological assemblages. Bipolar cores.
90
Plate 11. Bizat Ruhama archaeological assemblages. Bipolar flakes.
91
Plate 12. Bizat Ruhama archaeological assemblages. Bipolar flakes.
92
Plate 13. Bizat Ruhama archaeological assemblages. Bipolar flakes.
93
Plate 14. Bizat Ruhama archaeological assemblages and experimental assemblages. Shape of the ventral surfaces of bipolar flakes.
94
Plate 15. Bizat Ruhama archaeological assemblages and experimental assemblages. Distal end features of the bipolar flakes in archaeological assemblages (1-9) and experimental assemblages 10-18. Plate 16. Bizat Ruhama archaeological assemblages. Flaked flakes: 1-8.
95 96
Plate 17. Bizat Ruhama archaeological assemblages. Anvil flakes and modified flakes with signs of impact on the dorsal ridges. Pointed piece 1; Anvil flakes 2, 5-7; Clactonian notch 3; Irregular and step-like trimming 4. Arrows point at signs of impact on the dorsal face of the artifacts. Plate 18. Bizat Ruhama archaeological assemblages. Dorsal and ventral features of anvil flakes and modified flakes.
97 98
Plate 19. Bizat Ruhama archaeological assemblages. Clactonian notches: 1,8,12,13,15; Pointed flakes: 2,7; Flakes with marginal scars: 3,4,11; Flakes with scaled trimming: 5,9,10,14; Flake with irregular trimming: 6. Artifacts 2,5,7,8 exhibit signs of impact on the dorsal ridges.
99
Plate 20. Bizat Ruhama archaeological assemblages. Flakes with irregular and step-like trimming: 1,3,4,5,6,9,12,15, 23; Pointed flake: 2,8,10,11,13,14,16,20,21,24; Clactonian notches: 7,18,19; Flakes with scaled trimming: 17,22. Artifacts 1,3,8,9,20,23 exhibit signs of impact on dorsal edges.
100
Plate 21. Bizat Ruhama archaeological assemblages. Pointed flakes: 1,2,4,7,8,9,13,14,16 (pointed flakes 4,9,13,14,16 are shaped by Clactonian notch and additional scars); Clactonian notches: 6,12; Flakes with step-like trimming: 10; Flakes with scaled trimming: 11,15,18. Artifacts 4,5,6,8,9,13 exhibit signs of impact on dorsal edges.
101
Plate 22. Bizat Ruhama archaeological assemblages. Second generation flakes: 1-12. Artifacts 2,3,7,8,9 exhibit signs of impact on the distal end of the ventral surface.
102
Plate 23. Bizat Ruhama archaeological assemblages. Second generation flakes: 1-5; artifact 3 exhibits signs of impact on the distal end of the ventral surface.
103
xi
Plate 24. Bizat Ruhama archaeological assemblages and experimental assemblages. Second generation flakes.
104
Plate 25. Bizat Ruhama archaeological assemblages. Signs of impact on the distal ends of the ventral surfaces of the second generation flakes.
105
Plate 26. Experimental assemblages. Bipolar cores: 1-3; Simultaneous Siret and distal break produced during freehand hard hammer technique flaking.
106
Plate 27. Experimental assemblages. Bipolar flakes.
107
Plate 28. Experimental assemblages. Clactonian notch (1) and pointed flake (2) produced during anvil-supported knapping of the flakes conjoined with SGF that were detached during the knapping. Plate 29. Experimental assemblages. Modified flakes produced during the anvil-supported knapping of the flakes.
108 109
Plate 30. Experimental assemblages. Anvil flakes and modified flakes produced during the anvil-supported knapping of the flakes. Clactonian notch: 1; flakes with irregular trimming 2, 7-9; flakes with scaled trimming 3-6; pointed piece 10. Artifacts 2, 7-10 exhibit signs of crushing on the dorsal ridges (marked by arrow). Plate 31. Experimental assemblages. Dorsal and ventral features of anvil flakes and modified flakes.
110 111
Plate 32. Experimental assemblages. Second generation flakes produced during anvil-supported knapping of the flakes: 1-12.
112
Plate 33. Nahal Hesi archaeological assemblages. Cores on small local Eocene pebbles: 1-4; small unretouched flakes: 5-9; large scrapers made on raw materials that were not found in the area during the raw material survey.
113
Plate 34. Experimental assemblages. Handaxe made on limestone pebble and chopper made on brecciated chert pebble. 114
xii
CHAPTER 1: INTRODUCTION Bizat Ruhama is an Early Pleistocene site located on the fringe of the Negev Desert, Israel, in the southern coastal plain of the southern Levant. This book presents the results of recent excavations carried out at the site and technological analysis of its lithic industry.
The Acheulian presence in the Levant is well documented from as early as 1.6–1.4 Ma at ‘Ubeidiya (Bar-Yosef and Goren-Inbar 1993, Martínez-Navarro et al. 2012, Tchernov 1987), ca. 1 Ma at Evron Quarry and Latamne (Bar-Yosef 1998, Ron et al. 2003, Tchernov et al. 1994) and 0.8–0.7 Ma at Gesher Benot Ya‘aqov (Goren-Inbar et al. 2000). Late Acheulian sites are also well known from both open-air and cave contexts in Israel (Chazan and Kolska Horwitz 2007, Garrod and Bate 1937, Gilead and Ronen 1977, Goren-Inbar 1985, Jelinek 1977, 1982, Marder et al. 1999, 2011, Neuville 1951, Ronen et al. 1972, Stekelis and Gilead 1966), Lebanon (Copeland 1983), Syria (Jagher 2011, Muhesen and Jagher 2011) and Jordan (Copeland 1991, Copeland and Hours 1989, al-Nahar and Clark 2009). Non-handaxe industries are rare. Some assemblages composed of cores and flakes were found on the coastal terraces of Lebanon (Hours 1981). A number of non-handaxe localities have been identified near Kefar Menachem, Israel (Barzilai et al. 2006, Gilead and Israel 1975); one of these localities is currently dated to the Middle Pleistocene and is interpreted as a variation within the Acheulian technocomplex (Malinsky-Buller, 2013, pers. comm.). Additional non-handaxe Acheulian occurrences at Layer G in Tabun Cave, Layers E–G in Umm Qatafa, the base of Layer C in Bezez Cave and Stratum IV in the Yabrud rockshelter, were identified on the basis of absence of bifacial tools as Tayacian (Copeland 1983, Garrod and Bate 1937, Goren-Inbar 1995, Neuville 1951). However, apart from the absence of handaxes, no similarities were found between these industries and that of Bizat Ruhama (Bar-Yosef 1998, Ronen 1979). Hence, the Bizat Ruhama industry has generally been discussed in the context of the core-and-flake industries of the Levant (Bar-Yosef 1994, 1998) or even assigned to a separate cultural unit called the Nagilan, after the nearby Tel Nagila (Ronen 1979).
Since its discovery, Bizat Ruhama has attracted attention from researchers of the Levantine Paleolithic because of its distinctive industry. The first scholars to study the lithic artifacts from the site were participants in the archaeological survey of Nahal Shiqma (Lamdan et al. 1977). They described the industry as being composed of “unique tools, which are very difficult to classify by conventional typological means” (Lamdan et al. 1977: 55). The age of the site was unknown, but on the basis of its stratigraphic position and the simplicity of its tool forms it was assigned to the Lower Paleolithic (Lamdan et al. 1977, Ronen 1979). The pilot excavations at Bizat Ruhama were conducted in 1996 by A. Ronen and J.-M. Burdukiewicz (Ronen et al. 1998). The results of these excavations suggested that Bizat Ruhama is a single-horizon site characterized by a core-andflake industry (Ronen et al. 1998, Zaidner 2003a, Zaidner et al. 2003). The characteristics of the lithic assemblages can be summarized as follows: 1. Absence of handaxes and handaxe preparation flakes. 2. Flake-oriented, rather than core-tool-oriented, technology. 3. Small-sized flakes. 4. High frequency (ca. 40% of the entire assemblage) of notches, denticulates, small awls and small retouched flakes. Among the tools thought to be retouched, many do not exceed 15 mm in length, raising questions about the production techniques, tool use and technological affiliation of the Bizat Ruhama hominins.
In other studies, the small size of the artifacts was viewed as the main characteristic distinguishing Bizat Ruhama from other Lower Paleolithic assemblages in the Levant and beyond and was used to propose cultural/technical affinities with other industries (Burdukiewicz and Ronen 2003a, Derevianko 2009). The industry was often called “microlithic”, not only to emphasize the small size of the
The Bizat Ruhama industry has no clear parallels among known Lower Paleolithic sites in the southern Levant, which usually show Acheulian affinities (e.g. Bar-Yosef 1998, BarYosef and Belmaker 2010, Goren-Inbar 1995, Ronen 1979). 1
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
artifacts but also to link it to the “microlithic complex” of the Lower Paleolithic (see papers in Burdukiewicz and Ronen 2003b, Derevianko 2009, Derevianko et al. 2000). The “microlithic” industries, first described at Vértesszölös (Hungary), Bilzingsleben (Germany) and some other central and eastern European sites, are characterized by the absence of handaxes, the small size of artifacts and the high frequency of small retouched tools and are dated to 400–200 Ka (Burdukiewicz et al. 1994, Burdukiewicz 2003, Kretzoi and Vértes 1965, Kretzoi and Dobosi 1990, Mania 1990, Mania and Weber 1986, Svoboda 1987, Valoch 1989).
The evidence from Dmanisi indicates that the earliest hominins possessing Oldowan-like core-and-flake technologies dispersed from Africa as early as the beginning of the Early Pleistocene (de Lumley et al. 2005, Ferring et al. 2008, Mgeladze et al. 2011), preceding the first Acheulian assemblages in Africa (Asfaw et al. 1992, Beyene et al. 2013, Roche et al. 2003, Semaw et al. 2008). This sheds new light on the earliest hominin presence in Eurasia, since the evidence for pre-Acheulian dispersals out of Africa is now compelling. The Eurasian Early Pleistocene record thus contains evidence for the presence of two African entities, the Oldowan and the Acheulian technocomplexes. The chronological context and absence of Acheulian tools suggest that Bizat Ruhama may belong to one of the Oldowan out-of-Africa sorties.
The claim that the “microlithic complex” represents a distinct technological unit was brought into question by finds from Isernia La Pineta, a Middle Pleistocene site in central Italy. At the site, small notches, denticulates and awls were interpreted as by-products of small flake production rather than intentionally produced tool forms. It was suggested that this specific technology was an opportunistic response to constraints imposed by the availability of raw material (Crovetto et al. 1994, Longo et al. 1997, Peretto 1994).
It is from this perspective that I launched a new field project at the site in 2004. The excavations (2004–5) had three major goals: 1) to reconstruct the paleoenvironmental context of the site; 2) to provide large lithic assemblages for detailed technological and behavioral studies; and 3) to verify the primary context of the lithic and faunal assemblages. The results of the new excavations suggest that Bizat Ruhama is a site complex containing a number of roughly contemporaneous occupations.
Recent dating efforts have led to a breakthrough in the understanding of the Bizat Ruhama site. It is now soundly dated to the Matuyama reversed polarity chron (1.96–0.78 Ma) on basis of paleomagnetic evidence (Dassa 2002, Laukhin et al. 2001, Ron and Gvirtzman 2001, Ronen et al. 1998; see also below) and according to the faunal remains should probably be dated to ca. 1.6–1.2 Ma (Martínez-Navarro et al. 2012). According to this evidence, Bizat Ruhama is an Early Pleistocene site representing one of the earliest records of hominin presence outside Africa. At present, the Early Pleistocene archaeological record of the Levant and Europe is extremely fragmentary. Large assemblages of Early Pleistocene artifacts in primary depositional context have been reported only from Dmanisi (Georgia), identified as an Oldowan site, and ‘Ubeidiya (Israel), assigned to the Early Acheulian (Bar-Yosef and Goren-Inbar 1993, BarYosef and Tchernov 1972, de Lumley et al. 2005, Gabunia and Vekua 1995, Gabunia et al. 2000, Mgeladze et al. 2011). In other supposed Early Pleistocene occurrences, either the context of the artifacts is questionable or the assemblages are small and excavated areas are restricted in size (Arzarello et al. 2006, Barsky et al. 2010, Carbonell et al. 1999, 2008, Chauhan 2009, Dennell 2009 and references therein, Despriée et al. 2010, Duval et al. 2012, Martínez et al. 2010, Muttoni et al. 2013, Oms et al. 2000, Ronen 1991a, Santonja and Villa 2006, Toro-Moyano et al. 2011). At present, Bizat Ruhama is one of very few large Early Pleistocene coreand-flake assemblages discovered in primary context in Europe or southwestern Asia. The scarcity of other evidence makes the prospect of studying the technological behavior, paleoecology and cognitive and motor skills of the Bizat Ruhama hominins paramount for understanding the earliest hominin adaptations in Eurasia.
This book focuses on the analysis of the lithic assemblages from different occupation areas. Chapter 2 presents the methodology for the study of lithic technology. Following this, the geological, sedimentological and archaeological context of the site are presented in order to establish the general paleoenvironmental framework of the hominin occupation at the site and to demonstrate that the studied lithic assemblages are in primary context (Chapter 3). In the subsequent chapters different aspects of the lithic production system are presented and discussed. In the approach adopted here, the lithic production is seen as “a result of a long series of technical, economic, social, and even symbolic options, the combination of which can be expressed in terms of ‘strategies’ ” (Perlès 1992: 225). The existence of options implies that the hominins must have made choices. The aim of this study is to comprehend the circumstances leading to the particular choices that were made by the Bizat Ruhama hominins. The lithic production is presented in sequential form, from raw material acquisition (Chapter 4) to the stages of lithic production (Chapters 5–6). Chapter 7 presents the results of knapping experiments conducted in order to gain understanding of some aspects of the lithic technology and to assist in assessing the level of technological competence possessed by the Bizat Ruhama hominins. The experimental studies led to a breakthrough in the understanding of some of the technological features of the assemblages and resulted in a new interpretation of flakes that were previously thought to be intentionally retouched. The lithic production strategies, together with some other behavioral aspects of the site, are 2
Chapter 1: Introduction
then reconstructed (Chapter 8). The same chapter discusses the knapping skills of the tool-makers, the degree of conceptual complexity displayed and the place of the Bizat Ruhama industry within the Plio-Pleistocene archaeological record. The conclusions are presented in Chapter 9.
and the Acheulian (Leakey 1971, 1975). In Leakey’s view, the Developed Oldowan differed from the Acheulian in the relative frequencies of handaxes and core-tool types and the level of standardization of light-duty tools. In recent years the number of studied Plio-Pleistocene sites has increased significantly (see Schick and Toth 2006 for an overview). The age of the earliest of them goes back to 2.6–2.3 Ma (Kimbel et al. 1996, Roche et al. 1999, Semaw et al. 1997). The Oldowan now covers a much wider time span and exhibits much higher variability (Barsky 2009, Braun et al. 2009, Delagnes and Roche 2005, de la Torre et al. 2003, Goldman-Neuman and Hovers 2009, Hovers 2009a, Isaac 1997b, Kibunjia 1994, Piperno et al. 2009, Roche et al. 1999, Sahnouni 2006, Schick and Toth 2006, Semaw et al. 1997, Semaw 2000, Texier 1995). Although in general, Leakey’s division of Early Pleistocene industries into Oldowan and Acheulian is still accepted, the internal division of the Oldowan into substages is highly debated. On the basis of technological similarities between them, especially the ability to produce large flakes (unrecorded in the Oldowan), most scholars today view the Developed Oldowan and Acheulian as part of a single technocomplex (e.g., de la Torre and Mora 2005, Gowlett 1988, Semaw et al. 2008, 2009, Stiles 1979).
APPROACHES AND QUESTIONS IN THE EARLY PLEISTOCENE LITHIC STUDIES: A SHORT REVIEW Africa, the homeland of the first human-made stone tools, is the region where most of the Early Pleistocene sites are located and where the questions and methodology for the study of the earliest industries developed. Starting from the early 1930s, East Africa became the focus of intensive research on early human origins and evolution. The essence of the first lithic studies was to present the entire lithic assemblage according to distinct typological groups, providing counts, frequencies and some statistical analyses (Balout 1967, Clark 1969a, Clark et al. 1966, Kleindienst 1961, 1962, Leakey 1967, 1971, 1975). These first studies aimed at providing cultural and chronological frameworks for the African Early Stone Age and at understanding the variability within and across the earliest African assemblages (Kleindienst 1962, Leakey 1967).
Some authors argue that Oldowan industries that predate Olduvai Bed I (ca. 1.8 Ma) display characteristics differing from those of later Oldowan industries. These earliest known African assemblages at Gona EG10, EG12, Lokalalei 1, Lokalalei 2C and Fejej FJ-1a, as well as some levels at Dmanisi, lack evidence for systematic bifacial or multifacial flaking, lack spheroids, and are characterized by low percentages of pebbles with intentionally modified edges (choppers or chopping tools) and intentionally retouched flakes (Barsky 2009, Carbonell et al. 2009, de Lumley et al. 2005).
Mary Leakey’s study of the Olduvai Gorge assemblages (1971) became the point of reference for the Oldowan and Acheulian technocomplexes across Africa and beyond. Leakey focused on large core-tools as a means of tackling the question of the spatial and temporal variability that she observed throughout the Olduvai Gorge assemblages. Within the framework of the cultural paradigm, she divided the Olduvai assemblages into three cultural phyla: the Oldowan, the Developed Oldowan and the Acheulian. The Oldowan industrial complex was identified in Bed I of the Olduvai Gorge. The Oldowan assemblages include a number of categories identified by Leakey as core-tools (choppers and chopping tools, polyhedrons, discoids, subspheroids, spheroids, heavy-duty scrapers and broken pebbles), as light-duty tools (scrapers and burins made on flakes), as manuports (hammerstones, anvils and blocks of raw material) and as waste (light-duty flakes and fragments). Although Leakey noted that some of the core-tools could merely be cores for flake production, she also reported that many of them show battering and signs of use-wear (Leakey 1971). In general, she gave greater weight to the core-tools, considering them as deliberately designed for their use and as a manifestation of cultural affinities. Handaxes were not found in Oldowan assemblages, and only infrequent and poor standardized retouched flakes were reported. In Olduvai Bed II, Leakey advocated the coexistence of a number of industrial complexes: the Developed Oldowan (A, B and C)
The occurrence of these features in the sites dated to later than 1.8 Ma is argued to be evidence for more elaborate knapping with higher investment in the shaping of the working edges (Barsky 2009, Carbonell et al. 2009, de Lumley et al. 2005). Another important difference between sites predating 1.8 Ma and later assemblages is the distance over which raw materials were transported from the sources. In sites that predate 1.8 Ma the raw material sources are usually located less than 1 km away from the site (Féblot-Augustins 1997), while in later sites these distances are considerably greater (Blumenschine et al. 2003, Hay 1976, Plummer 2004).
Oldowan – the “sharp-edged” technology Another approach to the study of the Early Pleistocene assemblages was developed by Isaac during his work at 3
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Koobi Fora Formation, Lake Turkana (Isaac 1997b). Isaac viewed the variability of form of the Early Stone Age artifacts as a stochastic outcome of simple subsistence patterns (Isaac 1972, 1977). Thus, instead of studying the variability of form, he focused on the behavioral and formational aspects of the Early Pleistocene archaeological record. The underlying principle of this approach was “seeking recurrent predictable relationships between the circumstances (e.g. geography, habitat, food refuse, etc.) of a site and the characteristics of the artifact assemblage that it contains” (Isaac 1997a: 8). The main aim of this approach to the lithic artifacts was to evaluate their role in hominin adaptations to the environment, a question that was often overlooked by the cultural approach.
pounded pieces (PPs) – pieces modified or shaped by pounding and battering (Isaac 1986, Isaac and Harris 1997). Although this system of classification was considered hierarchical and the group of FPs was further subdivided into typological categories similar to those applied by Leakey, many of the analyses and discussions were based on these three generic groups (Isaac and Harris 1997). The research of this new school dealt first with the formational history of the assemblages. If the composition of the assemblages was determined to be anthropogenic, they were studied in accordance with their environmental background. The main questions that governed the study of the stone tools were environmental and economic in character: 1) Does the composition of the lithic assemblages change in relation to different environmental settings and to the distance from the raw material source? 2) Are the lithics associated with bones? 3) How far and in what quantities were the raw materials transported from their sources? 4) Were the stone tools knapped at the sites or transported in finished form from the raw material sources? (Isaac 1997b, Isaac and Harris 1978, Kroll and Isaac 1984, Ludwig and Harris 1998, Potts 1988, 1991, Schick 1986, Stern 1993, Stiles 1998, Toth 1982, 1985, 1987).
The new approach was put into practice in the archaeology project conducted at the Plio-Pleistocene Koobi Fora Formation on the eastern side of Lake Turkana (Isaac 1997b). Actualistic studies have played a major role in this new research direction. The experiments in stone tool production were conducted by Toth (1982, 1987, 1997). Based on his experimental work, Toth argued that the variability of form among the Early Pleistocene core-tools is a by-product of flake production. During the experimental production of flakes, Toth replicated many of the Oldowan core-tools without premeditation by using suitable knapping angles and choosing the “path of least resistance” (Toth 1982, 1987). One of the major inferences of Toth’s study was that unretouched flakes, rather than choppers, discoids and other core-tool forms, were the desired product. To underline the importance of sharp flakes, some of them were shown to have been used for butchery and wood-working (Keeley and Toth 1981).
The chaîne opératoire approach Current research on the Late Pliocene and Early Pleistocene industries focuses on the complexity of the core reduction technology and the acquisition and exploitation patterns of the raw material. Recent Oldowan studies concentrate on the level of technological competence of the Oldowan hominins, their manual dexterity and the predetermination and foresight reflected in the Oldowan technology (Barsky 2009, Braun et al. 2009, Carbonell et al. 2009, de la Torre et al. 2003, Delagnes and Roche 2005, de la Torre and Mora 2005, Goldman-Neuman and Hovers 2009, Harmand 2009, Hovers 2009a, Piperno et al. 2009, Roche et al. 1999, Schick and Toth 2006, Semaw 2000, Semaw et al. 1997, Stout et al. 2005, Texier 1995, Toth et al. 2006).
Isaac and Harris seem to agree with this inference in their account of the Koobi Fora lithic assemblages (Isaac and Harris 1997), where they also formulate their understanding of the Plio-Pleistocene technology: “…in the simplest possible material culture system, stones would have been broken in ways that would have produced reasonable quantities of convenient sharp edges with the least input of effort and of skill” (Isaac and Harris 1997: 263).
This emphasis on the complexity of the technological process rose from two major developments, one in the field of Plio-Pleistocene archaeology and the other in the general tendencies of prehistoric research. First, the age of the some of the recently discovered Oldowan sites goes back to 2.5 Ma, raising the question of whether in such early sites the hominins had already mastered stone knapping. Second, there was a general shift in prehistoric archaeology toward the behavioral and technical aspects of lithic production and the introduction of the chaîne opératoire approach.
It has been suggested that this sharp-edged technology was associated with increasing reliance on meat resources and was characteristic of hominins but not of apes (Isaac and Crader 1981). According to this new concept of the earliest industries, the form of the core-tools was of less importance, since they were not intentional tools but by-products of flake manufacture. Hence, the new methodology sought to diminish the significance of the variability of form that was previously thought to be a manifestation of cultural or functional differences. Isaac and Harris grouped all the artifacts into three technologically defined groups: flaked pieces (FPs) – pieces from which flakes were detached; detached pieces (DPs) – flakes and flake fragments; and
The chaîne opératoire approach is grounded in the perception of technique and technology in French ethnography (Lemonnier 1992). Technique, according to Maus (1936, 4
Chapter 1: Introduction
images of shapes which already exist in his mind (concepts, templates). These range from the selection of the rough block and its positioning, through performing and shaping, all the way to finishing… At each of these stages, the real situation is compared with the underlying mental image of the ideal next stage. Various modes of action which could be used in the next step of the process are visualized in order to correct and/or progress the work. The knapper then decides which mode of action is both desirable and feasible.”
as cited in Lemonnier 1992), is an action (purposeful body movement) that is effective (planned to give a result) and traditional (learned from others). Each technical process involves: 1) objects – tools employed and pieces of material transformed, 2) gestures – the range of body movement performed, and 3) technical knowledge achieved through learning from others or personal experience (Lemonnier 1992, Pelegrin et al. 1988). These three principles of the technique were developed by Leroi-Gourhan (1964) and Lemonnier (1976) into the concept of chaîne opératoire – a trajectory of technical actions leading to a goal and having a beginning (raw material acquisition) and an end (the final product). In the chaîne opératoire view, lithic production is a succession of gestures and technical actions carried out on a block of raw material in which each successive action is conditioned by a previous one.
Thus, the execution of the schéma opératoire is a dynamic process that is based on knowledge and memory, i.e. what is the shape of the needed product and what should be the sequence of gestures to accomplish it, and on know-how, i.e. the experience and skills of the knapper that allow him to solve the problems arising during deviations from an ideal schéma opératoire (Karlin and Julien 1994, Pelegrin et al. 1988, Pelegrin 1990).
The chaîne opératoire approach was used for the study of the prehistoric lithic technology in the works of French archaeologists during the 1980s (e.g. Boëda 1986, Geneste 1985, Pelegrin et al. 1988). In even the simplest lithic production system, the chaîne opératoire includes three stages: 1) raw material procurement, 2) tool production and 3) tool use. In more advanced systems, the second and third stages can be divided into several substages, i.e. stages in core preparation and blank production, blank selection, tool resharpening and reuse (Inizan et al. 1999, Karlin and Julien 1994, Perlès 1992).
The chaîne opératoire approach has recently been applied to the study of Plio-Pleistocene lithic assemblages in Africa and beyond (Barsky 2009, Delagnes and Roche 2005, de la Torre et al. 2003, de la Torre and Mora 2005, de Lumley et al. 2005, Harmand 2009, Roche and Texier 1996, Roche et al. 1999, Texier and Roche 1995). Many of these studies show that the knapping skills and manual dexterity of Oldowan hominins were more developed than was previously thought (Delagnes and Roche 2005, de la Torre et al. 2003, Roche et al. 1999, Texier 1995). According to these studies, the Oldowan hominins were well aware of the fracture mechanics of different rock types, were able to organize and predetermine the debitage in such a way that long sequence of flakes could be detached from core surfaces, and had a level of know-how that allowed them to maintain knapping even if the debitage surfaces were damaged by knapping accidents (Braun et al. 2009, Delagnes and Roche 2005, de la Torre et al. 2003, Harmand 2009, Hovers 2009a, Stout et al. 2005).
Each chaîne opératoire is an implementation of the goal of the knapping, which is based on the knapper’s needs and his technical knowledge and experience. These needs and knowledge are the foundations of the conceptual schéma opératoire – a succession of mental images or conceptual templates leading to the knapping goal (Pelegrin 1990). As explained by Pelegrin (1990: 303): “The stone tool maker is guided by a series of mental
5
CHAPTER 2: OUTLINE OF THE RESEARCH RESEARCH OBJECTIVES AND QUESTIONS
1.
How do the assemblages recovered vary among different areas of the site? What do these differences tell us about the hominin occupation of the site?
2.
Is the composition of the archaeological assemblages what one would expect if knapping was conducted in situ and, if not, does this indicate hominin transportation of prepared artifacts from/to the site or post-depositional disturbance?
3.
Are the raw materials local or were they transported over long distances?
At present, the general cultural and technical affinities of Bizat Ruhama are yet to be established. The goals of the knapping are unclear. The technological competence of the Bizat Ruhama hominins, their knapping skills and the choices that led to the production of particular types of artifacts are the questions that need to be answered to allow for comprehensive understanding of the industry.
4.
Do the raw material acquisition and exploitation patterns indicate selectivity? If so, what are the reasons for selective use of the raw materials (e.g. knapping qualities, size, shape)?
5.
Which knapping techniques and methods were used at the site?
The first goal of this study is to present a detailed description of the industry using a methodology and terminology resembling those used in other recent Early Pleistocene lithic studies. This will provide a means of comparing the debitage techniques and methods and the morphologies of the artifacts with the same aspects of other Early Pleistocene sites and of assessing the place of the Bizat Ruhama industry within the archaeological framework of the Early Pleistocene.
6.
What was the aim/s of the knapping?
7.
How developed are the conceptual schémas opératoires of the lithic production? What levels of preplanning and foresight do they demonstrate?
8.
What is the place of Bizat Ruhama industry among the Early Pleistocene industries?
This study aims at gaining an understanding of the lithic technology of the Early Pleistocene hominins on the threshold of Eurasia. In the approach adopted here, lithic production is seen as “a result of a long series of technical, economic, social, and even symbolic options, the combination of which can be expressed in terms of ‘strategies’” (Perlès 1992: 225). The existence of options implies that hominins must have made choices. The aim of this study is to understand the causes and circumstances that led to the particular choices made by the hominins of Bizat Ruhama.
The reconstruction of the goals of the lithic production and sheding a light on technological and adaptive skills of the Early Pleistocene hominins, their technological competence and manual dexterities are the second major goal of this study. Given that Bizat Ruhama preserves some of the earliest evidence for human presence outside Africa, the study of these behavioral aspects of the industry is important for comprehension of the earliest hominin adaptations in Eurasia and of the technological skills of the earliest Eurasians.
STUDIED ASSEMBLAGES AND SITES
Bizat Ruhama The lithic industry of Bizat Ruhama is composed of five excavated assemblages and a group of artifacts collected during surveys and surface collections. The excavated assemblages contain 1958 artifacts. Two main excavated areas, BRAT5 and BR1996, yielded 1694 artifacts, while the rest were unearthed in three trenches: BRT1, BRT2 and BRT3. In addition, over the years from the first research
To achieve these primary objectives, a number of questions are addressed: 6
Chapter 2: Outline of the research
at the site in the 1970s to the most recent surveys, a large assemblage of ca. 700 artifacts was collected from surface exposures of the archaeological layer. This assemblage was incorporated into the technological and metrical study of the industry. The data are all arranged in tables, in which survey/surface assemblages are presented separately. In cases where assemblages from excavated trenches are small (BRT1, BRT3 and in some cases BRT2 as well), they are presented together with the survey/surface material.
to discarding the used tool. To be able to understand these choices, we first need to analyze the lithic assemblages in a manner that brings to light the decision-making of hominins at each stage of this long technological sequence. The approach chosen for this study is one that attempts to reveal the intentions, choices and skills of the tool-makers through studies of the chaîne opératoire and conceptual schéma opératoire of tool production (Karlin and Julien 1994, Karlin et al. 1993, Pelegrin 1990, 1993, Perlès 1992, 1993, Pigeot 1990).
Supplementary material
In the study of lithic assemblages with the chaîne opératoire approach, each lithic artifact is seen as a link in a successive technical process. The place of each link can be found with the help of refitting or “mental refitting” – reading the traces of previous technical acts on the dorsal face and butt of the artifact (Boëda 1986, Delagnes and Roche 2005, Inizan et al. 1999, Pelegrin 1985, 1986). Once the place of each artifact in the sequence is defined, the whole technical process can be reconstructed and the aims of the knapping, the decisions taken during the knapping sequence and the skills of the knapper can be comprehended.
Lithic assemblages from the Middle Pleistocene site in Nahal Hesi and the Early Pleistocene site of Evron Quarry were studied to help in answering specific questions raised by the Bizat Ruhama industry. Nahal Hesi Nahal Hesi is a Middle Pleistocene Acheulian site situated less than 4 km north of Bizat Ruhama. It was excavated by David Gilead in the early 1970s. The excavations yielded faunal and lithic assemblages. The fauna were recently published by Yeshurun et al. (2011). According to observations made on fauna and lithics, Nahal Hesi is a site in primary context (Yeshurun et al. 2011). The lithic assemblage is composed of 557 artifacts, all of which were analyzed during this study. The Nahal Hesi lithic assemblage was incorporated into this study to assist in understanding the pattern of raw material exploitation at Bizat Ruhama.
For Bizat Ruhama the chaîne opératoire was reconstructed by mental refitting with the help of knapping experiments. The goal of the knapping experiments was to reproduce elements of the chaîne opératoire whose genesis was not clearly understood. However, while focusing on the reconstruction of reduction sequences, I also attempted to combine the chaîne opératoire approach with the environmental approach adopted by Isaac’s team. Thus, the paleoenvironmental context, association between bones and stone tools, and densities of artifacts were taken into consideration while reconstructing the lithic production system at the site.
Evron Quarry Evron Quarry is an Early Acheulian site located in the Galilee Coastal Plain. The site was excavated by A. Ronen and M. Prauznitz in 1976–1977 and by A. Ronen in 1985 (Ronen 1991b). It is considered to be one of the earliest sites in Israel (Bar-Yosef 1998) and is dated to the end of the Early Pleistocene on the basis of faunal and paleomagnetic studies (Ron et al. 2003, Tchernov et al. 1994). Evron Quarry is a site in primary context that yielded faunal and lithic assemblages (Ronen 1991b, Tchernov et al. 1994). The present study includes metrical data from an assemblage excavated at Evron Quarry during 1977. The assemblage was excavated from an area of ca. 30 m2 and is composed of 376 artifacts.
Study of raw material procurement strategies The question of raw material acquisition was addressed through a combined study of the stone sources in the Bizat Ruhama area, the types of stone used at the site and the knapping qualities of local stone. The raw materials of the nearby Middle Pleistocene Acheulian site of Nahal Hesi were studied as well. Nahal Hesi is Hesihlocated 3.8 km north of Bizat Ruhama, well within the estimated home range of Early Pleistocene hominins (Anton et al. 2002, Dennell 2007, 2009, Marwick 2003). Comparison of the lithic assemblages of Nahal Hesi with those of Bizat Ruhama thus provides a rare opportunity to examine the differences between raw material exploitation patterns of Early and Middle Pleistocene hominins in the same region.
METHODOLOGY FOR THE STUDY OF THE LITHIC TECHNOLOGICAL STRATEGIES Technological strategies are recurrent choices made by hominins from a number of existing options (Perlès 1992). These choices were made at different stages of lithic production, from obtaining a block of raw material
A survey of raw material sources was conducted in the area within a radius of 5 km from the site and each of the sources 7
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
located was randomly sampled. Since the only available raw materials in the area are pebbles found in conglomerate exposures, the sampling was carried out by methods used in geomorphological studies. An area of 1 m2 randomly chosen in each exposure was sprayed with color (Kellerhals and Bray 1971). All colored pebbles were collected and subjected to macroscopic study, including the following observations:
matrix are coarse-grained and similar in color (both in hues of gray); 2) chert in different hues of gray within the same pebble; 3) homogeneous chert in different hues of gray. Since they are similar in color, grain size and homogeneity, all three types are presented here together. The gray chert probably also originates from the Mishash Formation and may in fact be a more homogeneous type of brecciated chert.
Lithology
4.
Lithological identification was made with the naked eye, but in some cases hydrochloric acid was used to distinguish carbonate rocks from chert. The sampled exposures contained only sedimentary carbonate (limestone) and siliceous rocks (different types of chert). The collected pebbles were divided into five types of rock:
Eocene chert: a grayish-brown, brown and dark brown, relatively fine-grained and homogeneous chert with small white patches and spots. This is a typical chert of the Adulam Formation found in the Shephela region, similar to the Adulam chert found in northern Israel (e.g., Ramot Menashe; Lengyel 2005).
5.
Translucent chert: a fine-grained, translucent, glossy chert in gray and brown hues. This is the most homogeneous and fine-grained material in the area. Its probable origin is the Mishash Formation.
1.
Hard white/gray limestone.
2.
Mishash or brecciated chert (named after Kolodny 1967, Kolodny et al. 1965). Brecciated chert is unhomogeneous and unevenly textured and is characterized by a variety of colors (Figure 2.1). This is a typical chert of the Upper Cretaceous Mishash Formation known from the Judean Mountains and the Negev (Kolodny 1967, Soudry et al. 1985). The thickness of chert layers in the formation sometimes reaches several meters. Brecciated chert has been described as being composed of a “matrix” and of “fragments” (Kolodny 1967). The “fragments” can be arranged within the matrix in several shapes that can all occur in the same layer. Both the fragments and the matrix are composed of micro- and crypto-quartz, but they differ markedly in color. The transition between the matrix and the fragments can be sharp or gradual.
3.
Maximum length The longest possible distance between two points on the pebble’s surface. Width Measured at the widest part of the pebble, perpendicular to the axis of maximum length. Thickness Measured at the thickest part of the pebble, perpendicular to the width axis. Shape
Gray chert: a grayish, coarse-grained, homogenous chert. Three types can be distinguished within this group: 1) chert with occasional large gray patches similar to the “fragments” in the brecciated chert, but more homogeneous; the patches and surrounding
The shape was recorded according to the qualitative method developed by Goldman-Neuman and Hovers (2009), with some necessary adjustments (Figure 2.2).
Figure 2.2. The shape of the pebbles. “Flat discoid” refers to small, flat, rounded pebbles usually associated with marine beach environments.
Figure 2.1. The three main types of chert in the Bizat Ruhama area. 8
Chapter 2: Outline of the research
Observations on raw materials at the site 1.
The lithology of each artifact in the lithic assemblages was determined.
2.
Unknapped pebbles found at the site were subjected to the same set of measurements as pebbles collected from the conglomerate exposures.
the piece of raw material. The technique involves knapping tools and gestures, e.g. the type of hammer used, the mode of the force applied, body position and movement (Inizan et al. 1999, Pelegrin 1985, 1986). The identification of the technique is based on the form of the butt and features of the ventral face and distal end of the flakes. The latter are relevant only if an anvil was used during knapping. At Bizat Ruhama, evidence for use of two techniques was found: Bipolar technique. The bipolar technique consists of placing the core on an anvil and striking it along a roughly perpendicular plane. Force is delivered to the core from two opposing points of impact: where the hammer meets the core and where the core rests on the anvil. The anvil allows the knapper to strike with greater force than if the core were held free-hand.
Study of raw material exploitation strategies The initial stage in any chaîne opératoire study, whether it involves actual or mental refitting, is division of the artifacts into groups of raw material that are then studied separately. Eventually, the groups are compared in order to evaluate differences in techniques and methods of reduction, intensity of reduction, blank selection, etc. During the presentation of the lithic industries in the following chapters, many of the observations are presented separately for different raw material types in order to allow for comparison.
The bipolar technique may result in characteristic signs on cores as well as on DPs. The technique does not usually produce a conchoidal fracture (Andrefsky 2005, Cotterel and Kamminga 1987). Thus, the products of the bipolar technique often show no signs of bulbs of percussion on flakes or negative bulbs of percussion on cores. Bipolar cores and flakes may also exhibit signs of force applied to opposite ends and the flakes may have crushed butts (Ahler 1989, Barham 1987, Diez-Martín et al. 2009, Flenniken 1981, Jeske and Lurie 1993, Koboyashi 1975). During the knapping experiments presented in Chapter 7, an attempt was made to identify diagnostic signs of the bipolar technique on the types of raw material used at Bizat Ruhama.
Study of lithic production strategies Knapping modes Knapping applies to any type of action intentionally aimed at fracturing raw material (Inizan et al. 1999). Knapping is a general term that does not refer to aims, methods or techniques and encompasses debitage, shaping and retouching. In this work the term “knapping” is used either in its general meaning or when the exact aim of the removals is not certain. For example, it is unclear whether Clactonian notching was aimed at the manufacturing of Clactonian notches or the Clactonian waste flake was the desired end product. In the first case it would be referred to as retouching and in the second it would be called debitage.
Free-hand hard hammer technique. This technique consists of striking the hand-held core with a hard stone hammer (in the case of Bizat Ruhama, a hard limestone or chert pebble). The technique is usually associated with a conchoidal fracture. Flakes produced by the free-hand hard hammer technique are characterized by the presence of impact points, incipient cones of percussion, pronounced bulbs of percussion, easily identifiable ripples and thick butts (Andrefsky 2005, Inizan et al. 1999).
Debitage/flaking are used as synonyms to describe “the intentional knapping of blocks of raw material, in order to obtain products that will either be subsequently shaped or retouched, or directly used without further modification” (Inizan et al. 1999: 138).
Debitage methods According to Inizan et al. (1999), the debitage method refers to a carefully thought-out sequence of interrelated actions. The debitage method can be simple or organized. If the debitage shows no sign of predetermination and foresight, it can be called simple debitage (Delagnes and Roche 2005, Inizan et al. 1999). For example, if the knapper uses only the appropriate angles of the piece of raw material and after each removal seeks a new appropriate angle, the debitage is executed without predetermination and foresight and thus can be called simple. If, on the other hand, the debitage is executed according to constant rules resulting in a series of recurrent products, it can be called organized debitage (Delagnes and Roche 2005, Inizan et al. 1999). Organized
Shaping is the sculpturing of a single tool from a chunk of raw material (e.g. production of handaxes or choppers) (Inizan et al. 1999). Retouching is the process of intentionally transforming a blank into a desired tool by a single removal or a series of small removals (Inizan et al. 1999). Knapping techniques The knapping technique is the physical action of transforming 9
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
debitage can be carried out by a number of different, more or less elaborate methods. Middle Paleolithic and later methods of debitage are the most complex, including multiple stages of preparation. They entail an elaborate conceptual schéma opératoire and often produce a number of characteristic preparation products, which are very helpful in reconstructing the chaîne opératoire. In earlier periods the debitage methods are much simpler, but debitage that follows constant rules and shows a certain level of predetermination has been identified in even the earliest Pliocene sites (Delagnes and Roche 2005). Elements in the chaîne opératoire The elements in the chaîne opératoire were identified according to the specific technological features of the artifacts. The elements are listed below according to their place within the chaîne opératoire. Most of the elements are identified on the basis of features previously described in the literature; these are, for example, cores, flakes, siret breaks, Clactonian notch waste flakes, etc. However, some elements are described here for the first time. These are fractured flat pebbles, anvil flakes and second-generation flakes. All three were identified on the basis of the data obtained during the knapping experiments. In total, 18 elements were identified. The identified elements can be grouped into four groups along the reduction sequence: flaked pieces, detached pieces, secondary knapped flakes and second-generation flakes.
Fractured flat pebbles – flat discoid pebbles broken on an anvil.
2.
Fractured pebbles – in most cases, pebbles were fractured along their transversal axis.
3.
Pebbles with 1–3 removals.
4.
Cores – pebbles from which more than three flakes were removed.
5.
Bipolar cores – cores that show clear evidence of splitting on an anvil, namely signs of opposite impacts and absence of clear bulb negatives.
6.
Exhausted cores – small, heavily reduced cores lacking identifiable striking platforms and debitage surfaces.
3.
Siret accidental breaks – products of siret accident (Inizan et al. 1999), longitudinal breakage of the flake starting from the butt. Some of the siret breaks are also broken at the distal end.
4.
Longitudinal breaks – similar to siret breaks, but the proximal end is either broken or unidentifiable.
5.
Other fragments – distal and medial fragments that show clearly recognizable ventral faces.
6.
Angular fragments – waste and flake fragments that exhibit a single flat (ventral?) surface without apparent orientation.
1.
Flaked flakes (after Ashton 1992, 2007) – “flakes that have had further flakes removed from lateral, proximal or distal edges and from both the ventral and dorsal. There are characteristically between one and four removals on a single piece, but sometimes several more” (Ashton 2007: 1).
2.
Anvil flakes – flakes showing signs of impact at the intersection of ventral/lateral and dorsal/lateral surfaces that, as demonstrated by knapping experiments conducted during this study, were produced during knapping of flakes on an anvil.
3.
Possible anvil flakes – broken flakes that show indefinite signs of breakage on an anvil.
4.
Modified flakes – DPs that show signs of further modification on one or several edges. The modification consists of scars of different morphology, size and regularity. Some of the scars closely resemble retouch, but could also have been produced during breakage of the flakes on an anvil.
Second-generation flakes (SGFs) flakes that were detached from flakes. SGFs are thin, short, broad flakes that, on the basis of knapping experiments conducted during this study, were identified as products of secondary flake knapping. The group includes Clactonian notch waste flakes.
Detached pieces (DPs) are artifacts that were detached from the cores and pebbles and exhibit identifiable ventral faces. They include the following elements: 1.
Proximal fragments – flakes with butt intact.
Secondary knapped flakes (SKFs) flakes that were knapped, broken on an anvil or modified after their detachment from FPs:
Flaked pieces (FPs) are chunks of raw material from which flakes were removed (after Isaac 1986). These include: 1.
2.
Recording of the artifacts Each of the four groups of elements was subjected to the same set of technological and metrical observations. In addition, individual elements were subjected to specific sets of observations. Since many of the categories applied
Complete flakes – flakes with butt, distal and lateral edges intact. 10
Chapter 2: Outline of the research
in this study are widely known and are frequently used in other technological studies, they are not described in detail. There are, however, a number of features whose description required the introduction of new categories. These are set out in detail in Appendix 2. The complete set of recorded observations is also presented in Appendix 2.
bipolar flaking of pebbles, anvil breakage of flat “beachlike” pebbles and anvil knapping/breakage of flakes. The recording system includes observations made on material before, during and after the knapping (Table 2.1).
Experimental knapping
Statistical tests of significance were frequently employed to examine the differences and similarities between the frequencies of different raw material types, elements in the chaîne opératoire, and archaeological and experimental assemblages. Statistical test were not used as a means of distinguishing between the artifact groups, but rather were used as supporting evidence for inferences made by other methods. The statistical analyses are presented at the 95% confidence level.
Statistical analysis
Experiments in stone knapping were conducted to assist in the understanding of site formation processes, human activities at the site and reconstruction of the chaîne opératoire. Knapping experiments were conducted over three years starting from 2004. The majority were conducted by myself, while Goyrgy Lengiel took part in the experiments in the free-hand hard hammer technique and the bipolar technique. The experiments included free-hand hard hammer and
Flakes
Flat “beach-like” pebbles
Pebbles
Stage of the experiment
• Raw material • Size
• Raw material • Size • Roundedness
• Raw material • Size and weight • Roundedness
Pre-knapping observations
• Number of blows • Number of products produced by each blow • Recording the shape of the flake after each blow • The direction of the blow on each product
• Number of blows • Number of products produced by each blow • Direction of blow on each product
• • • •
Flaking technique Method of reduction Number of blows Number of products produced by each blow • Description of each knapping accident • Direction of blow on each artifact produced by the bipolar technique
Observations made during knapping
• Division of the assemblage according to the system used during the study of the archaeological material • Counting of artifacts 0.3–1 cm long • Study of complete flakes, applying the same list of technological attributes as that applied during the study of the archaeological material • Description of anvil and hammerstone impact marks on flakes and cores
Post-knapping observations
• Division of the assemblage • Description and according to the system counting of knapping used during the study of the products archaeological material • Description of anvil • Counting of artifacts 0.3–1 impact signs cm long • Description of broken • Description of anvil and surface features hammerstone impact marks
Table 2.1. Observations made on material before, during and after the experimental knapping.
11
CHAPTER 3: THE CONTEXT OF THE SITE AND THE LITHIC ASSEMBLAGES LOCATION AND GEOLOGY
undulating topography (Figure 3.2a). Low sand and loess hills (160–190 m a.s.l.) descend gently to the east and north toward Nahal Shiqma, the largest stream in the region (Figure 3.3). East of Nahal Shiqma and the Beer-Sheva–Qiriat-Gat road, the sand is replaced by chalk of the Lower Shephela hills. Numerous small streams, tributaries of Nahal Shiqma, drain the area. The largest are Nahal Sad, Nahal Ruhama and Nahal Hazav (Figure 3.3). Nahal Shiqma, whose source is in the southern Judean Mountains, is one of the main streams of the coastal plain’s drainage system. According to Nir (1970), the current course of Nahal Shiqma took its form during the Pliocene/Early Pleistocene invasion of sand dunes that blocked the path to the west. This caused the stream to flow northward before turning west toward the Mediterranean in the eastern part of the study area. The Quaternary deposits in the eastern part of the study area are thin and Pliocene and Eocene deposits are often exposed (Figure 3.2b, c; BarYosef 1964, Gvirtzman and Buchbinder 1969, Lamdan et al. 1977, Nir and Bar-Yosef 1976, Sneh and Buchbinder 1984, Sneh et al. 1998). The loess hills are disturbed by erosion that creates typical badland fields intersected by erosional gullies and depressions. Bizat Ruhama is located in one of these fields (Figure 3.2.d).
Location Bizat Ruhama is located on the fringe of the Negev Coastal Plain, 25 km east of the present Mediterranean shoreline. The Negev Coastal Plain is 27 km wide in the area of Bizat Ruhama and the site is located on its northeastern margin, between the Negev Desert on the south, the Shephela and the Judean Mountains on the east and the Mediterranean coast on the west (Figure 3.1). The area is characterized by a low
Present and past climate and environment At present the region is characterized by hot, dry summers (mean August temperature 32ºC) and warm winters (mean January temperature 17ºC). The average annual rainfall is 300–400 mm (Figure 3.1). Bizat Ruhama is located on the boundary between the Mediterranean and semi-arid climatic belts in the Irano-Turanian phytogeographical region, just 30 km north of the Saharo-Arabian phytogeographical region that is characterized by an average annual rainfall of less than 200 mm. The Irano-Turanian vegetation in Israel forms a narrow strip (20–30 km wide) separating the Mediterranean and Saharo-Arabian phytogeographical regions. At present most of the area is cultivated and the natural vegetation has largely been destroyed. Originally, the region was probably characterized by an open environment with occasional
Figure 3.1. Location map. 12
Chapter 3: The context of the site and the lithic assemblages
Figure 3.2. a) Undulating loess hills in the Bizat Ruhama area, view from the site to the east; b) Exposures of Eocene chalk in the Nahal Shiqma riverbeds, view from Tel Nagila to the north; c) Exposures of Pleshet Formation beachrock overlying Eocene chalk in Nahal Sad, view to the north; d) The Bizat Ruhama badland field, view from the site to the south.
shrubs and trees. The geographical position of the area on the desert fringe apparently makes the region sensitive to environmental changes deriving from fluctuations of climatic belts during the Quaternary (Horowitz 1979, Magaritz 1986, Magaritz and Goodfriend 1987, Vaks et al. 2006, 2007).
during the Quaternary (Vaks et al. 2006, 2007). Hence, the development of the hamra paleosol in the Bizat Ruhama area may also point to somewhat more humid conditions (Brunnacker et al. 1982, Ronen 1975). At present, hamra soils are found in Israel only in areas with an annual rainfall higher than 450 mm and are usually associated with Mediterranean forest (Dan et al. 1975, Ronen 1975).
At present there is little evidence for the past environment of the Negev Coastal Plain. Beside the stratigraphic sequence of Bizat Ruhama, the only Plio-Pleistocene sequence studied in the area is the Nahal Besor section near Tel-Sharuhen (Menashe 2003). Lengthy erosional phases are evidenced during the Early Pleistocene in the Tel-Sharuhen section (Menashe 2003) and during the Middle Pleistocene in the Bizat Ruhama area (Dassa 2002, Ron and Gvirtzman 2001). The erosional gaps have been linked to humid environmental conditions (Menashe 2003). More humid conditions than those of the present have also been suggested by Dassa (2002) during the formation of the Early Pleistocene hamra in the Bizat Ruhama area and by Rosen (1986) during the formation of the Middle/early Upper Pleistocene terraces of Nahal Shiqma. The data from the northeastern Negev as well point to several fluctuations in the amount of rainfall
In contrast, the data from faunal assemblages from Bizat Ruhama and nearby Middle and Upper Pleistocene sites point to a relatively stable semi-arid environment like that of the present (Yeshurun et al. 2011). The archeofauna are represented by equid, antelope and bovine remains, with no typical woodland species (e.g. cervids and suids) or amphibian or riparian animal species (e.g. hippo or turtle). Overall, the environmental picture stemming from the archaeozoological studies indicates an open environment with sporadic water sources and trees, much like this semiarid region today. The discrepancy between faunal and geological evidence may stem from different resolutions of these paleoenvironmental proxies. Moreover, it is important to note that archaeofaunal assemblages represent only points 13
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Figure 3.3. Map of the study area on the northeastern margin of the Negev Coastal Plain. S1–S7 refer to conglomerate exposures sampled during the raw material study.
in time, not a continuous sequence. Thus, it is possible that they represent only the periods of climate amelioration that allowed hominin settlement in the Negev Coastal Plain, whereas periods of aridification left no archaeological occurrences or associated faunal remains.
and Buchbinder 1984, Zilberman 1984). There is general agreement on the stratigraphy and nomenclature of the pre-Quaternary beds of the region. The stratigraphic chart presented in Gvirtzman and Buchbinder (1969; see Figure 3.4) summarizes the data on the pre-Quaternary stratigraphy of the Negev Coastal Plain.
Plio-Pleistocene geology
A detailed lithostratigraphic study of the Bizat Ruhama area was conducted by Bar-Yosef (1964). The following discussion of the lithostratigraphy of Bizat Ruhama area is largely based on his work. The base of the geological sequence in the area is a Cretaceous limestone of the Judea Group. Because of tectonic uplift during the Cretaceous,
The geology of the Negev Coastal Plain has been discussed by many authors (Gvirtzman 1970, 1990; Gvirtzman and Buchbinder 1969, Horowitz 1974, 1979, Issar 1961, Itzhaki 1961, Nir 1989, Nir and Bar-Yosef 1976, Sneh 14
Chapter 3: The context of the site and the lithic assemblages
Figure 3.4. Stratigraphic chart of the southern part of the coastal plain (after Gvirtzman and Buchbinder 1969).
the bedrock tilts sharply to the west. From that period onward the area was continuously influenced by sea-level fluctuations and tectonic uplift. The sedimentological cycles are linked to marine transgressions. Chalks, marls and sandstones of the Hashephela and Saqiye Groups were deposited during Tertiary transgressions. Phases of marine regression were usually accompanied by erosion, with the last major erosion phase documented in the area taking place after the deposition of the Eocene chalk of the BetGuvrin Formation. In the study area the next documented sedimentary cycle is of Pliocene age, depositing the Pleshet and Ahuzam Formations.
1979, Sneh and Buchbinder 1984, Zilberman 1984, 1986; see, however, Nir 1989 for a different view). According to Bar-Yosef (1964), both formations are well preserved in boreholes as well as on the surface of the studied area. They will be discussed in detail in the present study, firstly because they create a base for the Quaternary sediments in the region and secondly because of their importance as a major raw material source of the Bizat Ruhama hominins. Pleshet Formation The outcrops of the Pleshet Formation are widely distributed in the foothills of the Judean Mountains, in the coastal plain and in the northwestern Negev. The Pleshet Formation represents the littoral facies of the Pliocene transgressions and includes sandstones, conglomerates cemented by calcareous sand, beachrock and, to a lesser extent, uncemented sands and marls. In the subsurface, the Pleshet
The Pleshet and Ahuzam Formations were first discussed by Issar (1961). The use of both names for describing Pliocene marine littoral sediments (Pleshet Formation) and fluvial sediments (Ahuzam Formation) is generally accepted (Gvirtzman 1990, Gvirtzman and Buchbinder 1969, Horowitz 15
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Formation constitutes an almost continuous horizon between Nahal Shiqma and the western border of the studied area. It tilts 1–2º toward the west and unconformably overlies the Bet-Guvrin Formation. Outcrops of the Pleshet Formation are not found on the easternmost margin of the studied area, east of the Beer-Sheva–Qiriat-Gat road. In the study area the Pleshet Formation is 15–20 m thick and is rich in abraded marine skeletal fragments that occasionally appear in dense horizons and in detritus. Its lithological characteristics in the vicinity of Bizat Ruhama indicate coastal or shallow sea environments (Bar-Yosef 1964).
base of the resulting dunes, sandstone (kurkar) or red sandy loam (hamra) often formed. Sandstones appear in almost all boreholes around Bizat Ruhama. Their thickness reaches 30 m and they directly overlie the Pleshet or Ahuzam Formation (Figure 3.5). The Quaternary sandstone is exposed in only a few locations west of Bizat Ruhama, where it is known as the Hirbet Harev Kurkar Ridge (Bar-Yosef 1964, Horowitz 1979, Issar 1961, Nir 1989, Nir and Bar-Yosef 1976). The sandstone is overlain by a 2–4 m thick unit of sandy loam (hamra). The Bizat Ruhama archaeological remains are located on the surface of the hamra. The Nahal Shiqma riverbed, on the eastern margin of the study area, comprises the easternmost frontier of the invasion of Quaternary sands in Israel. The sands here reach their highest elevation in the Israeli coastal plain (ca. 180 m a.s.l.).
Ahuzam Formation The Ahuzam Formation is distributed along the foothills of the Judean Mountains, in the Higher and Lower Shephela. Only a few outcrops are known in the Negev Coastal Plain and their identification is arguable (Sneh and Buhbinder 1984). Ahuzam is a fluvial formation deposited along rivers that followed the regressing Pliocene sea (Bar-Yosef 1964, Gvirtzman and Buchbinder 1969, Sneh and Buchbinder 1984, Zilberman 1984, 1986).
The desert’s influence on the region is evident in the deposition of Peri-Saharan loess from the Middle Pleistocene to the present. The thickness of the loess deposits in the Negev reaches 12–15 m (Amit et al. 2011, Bruins and Yaalon 1979, Crouvi et al. 2010, Yaalon and Dan 1974), but it is lower in the Bizat Ruhama area, located close to the northern boundary of loess deposition in Israel. The average accumulation rate of loess in the Negev during the second half of the Upper Pleistocene and the Holocene is 0.1 mm per year (Bruins and Yaalon 1979, Gerson and Amit 1987, Yaalon and Ganor 1975). The loess provides the upper sedimentological cover of the study area, but in the site’s vicinity it has largely been removed by the erosion that created the badlands and exposed the underlying hamra.
In the eastern part of the studied area, a few exposures along the Nahal Shiqma and Nahal Adorayim riverbeds have been identified by some researchers as belonging to the Ahuzam Formation (Bar-Yosef 1964; Gvirtzman and Buchbinder 1969). Sneh and Buchbinder (1984) challenged this view, arguing for a later age of these conglomerates. The Ahuzam Formation was found in the subsurface in most of the studied boreholes (Bar-Yosef 1964, Gvirtzman and Buchbinder 1969, Sneh and Buchbinder 1984). The average thickness of the Ahuzam conglomerates in the Bizat Ruhama area is 10 m and they are composed of chert, limestone and chalk pebbles in a chalky and sandy matrix. In a few locations, outcrops of the Pleshet and Ahuzam Formations show lateral transition or interfingering. Bar-Yosef (1964: 20) describes one of these exposures near Tel Keshet. These outcrops mark the location of the fluctuating Pliocene shoreline, indicating that during the late Pliocene the shoreline was very close to Bizat Ruhama.
THE SITE
The Quaternary
The archaeological site of Bizat Ruhama was discovered by Yehuda Bach of Kibbutz Ruhama in the 1960s. After the archaeological survey conducted by Lamdan et al. (1977) in the Nahal Shiqma area, the lithic material collected on the surface of the site was published for the first time and the name Bizat Ruhama (Ruhama Swamp in Hebrew) was introduced on the basis of the assumption that the dark clay of Stratum 3 was a swamp or lake deposit (Lamdan 1977: 55).
During the late Pliocene and the Early and Middle Pleistocene, the Negev Coastal Plain was influenced by sea level fluctuations, the proximity of the desert and pedogenetic processes. The major parent material of the rocks and soils that build the Quaternary sedimentological cover of the coastal plain is quartzitic sand swept from the Nile Delta along the eastern Mediterranean shoreline. The sands were transported eastward from the coast during numerous marine transgressions. The strong western winds characteristic of the coastal plain drew the sands further inland. On the
In 1996 the site was excavated by a joint Israeli-Polish team headed by Avraham Ronen and Jan-Michal Burdukiewicz (Burdukiewicz and Ronen 2000, Ronen et al. 1998). During the project the general stratigraphy and chronology of the site were established (Laukhin et al. 2001, Ronen et al. 1998). The lithic material was studied by Zaidner (Zaidner 2003a, b, Zaidner et al. 2003). In addition, an area of ca. 80 km² at a radius of ca. 5 km around the site was surveyed in search of potential sources of raw material (Zaidner 2003b, Zaidner et al. 2003).
16
Chapter 3: The context of the site and the lithic assemblages
Figure 3.5. Composite stratigraphic section and microstratigraphy of the excavated areas at Bizat Ruhama. The composite stratigraphic chart is based on study of the Bizat Ruhama type-section (Strata 1–5; Laukhin et al. 2001, Mallol et al. 2011, Ronen et al. 1998) and on Bar-Yosef 1964.
The stratigraphic sequence
et al. 2001), Ronen and Burdukiewicz in 1996 (Ronen et al. 1998), and Zaidner and Mallol in 2004–5 (Mallol et al. 2011, Zaidner et al. 2010). Pedological and paleomagnetic studies of an additional outcrop (henceforth Ruhama 2), located 700 m southeast of Bizat Ruhama on the other edge of the badland field, was conducted by Ron, Gvirtzman and Wieder (Ron and Gvirtzman 2001, Wieder et al. 2008). The correlation of these studies is incorporated in the following presentation of the stratigraphy.
Bizat Ruhama is located in a badland field, 3 km long and 0.7 km wide, that is intersected by erosional gullies, channels and depressions. The archaeological layer was discovered at the bottom of two erosional channels on the southwestern edge of the field (Figure 3.6). On the western slope of the northern channel the erosion of the loess exposed a sequence of deposits approximately 20 m thick (Figure 3.7). The stratigraphy was studied by Laukhin in 1995 (Laukhin 17
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Figure 3.6. Bizat Ruhama, the site and excavation areas, view to the southeast.
Figure 3.7. The southern channel with BRAT5 and the 17-meter-high slope (the type-section of the site), view to the northeast.
The western slope of the northern channel, 17 m high, constitutes the type-section of Bizat Ruhama badland field and is composed of five strata (Figure 3.5), from base to top: •
et al. 2008). The stratum is placed in the Matuyama reverse polarity chron (1.96–0.78 Ma). The contact with Stratum 4 is locally sharp or diffuse.
Stratum 5: unknown depth. Reddish-orange sandy clays (hamra), massive and sticky. Ferruginous crust at the top. Carbonate nodules at 25–30 cm depth, becoming columns at 1.5–2 m depth. A few artifacts and bones occur in the upper 5 cm of the stratum. A similar layer was identified in the Ruhama 2 section (Dassa 2002, Wieder et al. 2008). This layer was formed on terrestrial sand dunes that had undergone pedogenesis (Weider
•
18
Stratum 4: 0.2–1 m. Gray-yellowish gray sand and clay with horizontal scatters of fossil plant fragments. Manganese impregnation toward the bottom. The bulk of the archaeological remains are located in the lower part of this layer. The thickness of this layer as documented in seven different sections ranges between 0.2 and 0.5 m, but it reaches a thickness of 1 m in one location reported by Laukhin et al. (2001). This layer
Chapter 3: The context of the site and the lithic assemblages
was not found in the Ruhama 2 section. According to Laukhin et al. (2001), Stratum 4 was deposited on the margin of a small lake. However, no evidence supporting the existence of a lake was found during the 2004–5 seasons. The stratum is placed in the Matuyama reverse polarity chron (1.96–0.78 Ma). The contact with Stratum 3 is locally sharp (and microlaminated in some areas) or diffuse. •
•
•
Stratum 2, Stratum 3 and Stratum 4. The total thickness of the sampled section is 4 m and the results show reversed polarity for most of the sequence except for the inconclusive upper 0.5 m (Laukhin et al. 2001). Both studies thus place the archaeological horizon of Bizat Ruhama in the Matuyama reversed polarity chron (1.96–0.78 Ma). Stratum 3, overlying the archaeological Stratum 4, and the lower part of Stratum 2 were also accumulated during the Matuyama chron. Sediments corresponding to the Bruhnes normal polarity chron were not found in the studied section.
Stratum 3: 1–3 m. Grayish black clay, massive, prismatic and greasy. Manganese and iron impregnation. This layer was excavated and studied in the type-section and in seven additional locations. This unit was not found in the Ruhama 2 section. The stratum is placed in the Matuyama reverse polarity chron (1.96–0.78 Ma). The contact with Stratum 2 is diffuse.
A paleomagnetic study conducted by Ron (Dassa 2002, Ron and Gvirtzman 2001) on the Ruhama 2 section shows similar results. The two lower paleosols in their study, correlated with Stratum 5 of the Bizat Ruhama section, show reversed polarity. The transition to normal polarity was detected in the third paleosol, probably corresponding to the lower/ middle part of Stratum 2 in the Bizat Ruhama type-section.
Stratum 2: 11–12 m. Brown silty clays, massive and prismatic with slickensides. Manganese impregnation and carbonate horizons (with nodules and concretions). A similar unit in the Ruhama 2 section was classified as “Brown Grumosol” and divided into four paleosol units (Dassa 2002). The basal 2–3 m of the unit are placed in the Matuyama reverse polarity chron (1.96–0.78 Ma). The contact with Stratum 1 is locally sharp or diffuse.
The faunal evidence from the site supports an Early Pleistocene age (Yeshurun et al. 2011). The equid found at Bizat Ruhama is identified as Equus cf. tabeti, resembling the equids from ‘Ubeidiya, Latamne, Aïn Hanech and Gesher Benot Ya‘aqov and generally dissimilar to the species closer to extant equids that are known from the Middle Pleistocene onward (Eisenmann 2006). The antelope identified as Pontoceros ambiguus or Spirocerus sp. is present at ‘Ubeidiya (Martínez-Navarro et al. 2012) and Dmanisi (Buhksianidze 2005) but absent from Gesher Benot Ya‘aqov (Martínez-Navarro and Rabinovich 2011). It has recently been suggested that the fauna of Bizat Ruhama belongs to the same faunal unit as ‘Ubeidiya and predates the Jaramillo normal event (Martínez-Navarro et al. 2012). The site is thus likely to be roughly contemporaneous with ‘Ubeidya, which is currently dated to ca. 1.6–1.2 Ma (Martínez-Navarro et al. 2012).
Stratum 1: 01.–2 m. Pale yellow, loess-like sandy clays, highly porous and carbonaceous, with local horizontal bedding. A similar unit in Ruhama 2 section was classified as “Loessial Arid Brown Soil” by Dassa (2002) and Ron and Gvirtzman (2001). Unknown age.
Chronology Estimation of the age of the site is based on paleomagnetic studies and bio- and geostratigraphic considerations. The site is located on the edge of the southern coastal plain of Israel at the easternmost boundary of the invasion of coastal sediments. The hamra directly underlying the artifacts is located 28 km off the current coastline at 160 m a. s.l, and thus must be connected to one of the earliest Pleistocene marine ingressions.
The 1996 excavations During the 1996 excavation season at Bizat Ruhama, an area of 11 m2 was excavated (henceforth BR1996; Figure 3.8; Table 3.1). In the excavations, each artifact and bone was plotted by three coordinates using a Total Station. Excavated sediments from 4 m2 were sieved through 1 mm mesh. The stratigraphic section is composed of Strata 3–5 of the typesection (Figure 3.5), from top to bottom:
The stratigraphic sequence of Bizat Ruhama was subjected to paleomagnetic studies in 1995 and 1996. In 1995, samples collected at 5 cm intervals encompassed the lower 15 cm of Stratum 3, the artifact-bearing deposit of Stratum 4 (20 cm thick in the sampled area) and the upper 50 cm of Stratum 5. The samples were taken from the location that was later excavated during the 1996 excavations. The total thickness of the sampled section is 80 cm and the results show reversed polarity for the entire sequence (Ronen et al. 1998). In 1996 a new paleomagnetic study was begun with the aim of extending the sampled sequence. In total, 189 samples were taken, encompassing the lower part of 19
•
Stratum 3: Grayish black clay, 30–50 cm in thickness. The contact with Stratum 4 is sharp and finely laminated, with alternating sand and clay laminae. A few artifacts were found.
•
Stratum 4: Gray, yellowish-gray sand with numerous small manganese concretions 20 cm in thickness. Most of the archaeological remains were found in this unit,
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Area
Size of the Finds area (square meters)
Density of The top of the Microstratigraphy artifacts per hamra (Stratum square meter 5) (below datum)
BRAT5
25
Lithics 701 Bones ~1000
28
~4.55 m
Stratum 4 is 0.3-0.5 m thick. Contact with grayish black clay (Stratum 3) and with hamra (Stratum 5) is diffuse.
BR1996* 11
Lithics 993 Bones ~50
90.2
~ 5.13 m
Stratum 4 is 0.2 m thick. Contact with Stratum 5 is sharp. Contact with Stratum 3 is finely laminated with alternating sand and clay laminae-
BRT1
2
Lithics 103 51.5 Bones – a few small splinters
~4.95 m
Stratum 4 is 0.3 m thick. Gray-yellowish gray sand is gradually getting partly-colored with greenish-gray and purple-red patches at the bottom. Contact with Stratum 3 and with Stratum 5 is diffuse.
BRT2
4
Lithics 149 37.3 Bones – a few small splinters
~5.4 m
Stratum 4 is highly disturbed by clay and yellow sand lenses and pockets. The contact with Stratum 5 is sharp and erosional.
BRT3
1
Lithics 28 Bones 20
~4.15 m
Stratum 4 is 0.25 m thick. Contact with Stratum 3 and with Stratum 5 is diffuse.
28
Table 3.1. Size of the excavated areas, density of the finds and microstratigraphy.
Figure 3.8. Plan of the Bizat Ruhama site (modified after Ronen et al. 1998). BR1996 – area excavated in 1996. BRAT5, BRT1, BRT2, BRT3, BRT4, BRT6 – areas and trenches excavated in 2004–5. V – Sampled locations with in situ artifacts or bones. X – Sampled locations without artifacts or bones. Thick curved lines mark the contour of the erosional channels along which the archaeological layer is exposed. The black line stretching from NW to SE marks the eastern border of the archaeological occurrence. 20
Chapter 3: The context of the site and the lithic assemblages
the majority embedded in the lower 10 cm of the layer at the contact with Stratum 5. The contact with Stratum 5 is sharp. •
(Table 3.2). Of the artifacts in the BR1996 assemblage, 5% had undergone chemical alteration. The degree of patina development was recorded only for artifacts made on brecciated chert (N=192), since difficulty was encountered in distinguishing patina on artifacts made on other types of raw material. The lithic assemblage includes a high number of secondary knapped flakes (Table 3.3). The number of second-generation flakes is low even in the squares where sediments were sieved. Similarly, the number of small chips and fragments smaller than 1 cm is low (N=138). The excavation yielded 50 bones, including a few horse teeth and a number of large- and medium-sized long bone fragments of ungulates.
Stratum 5: Hamra, hard with black coating at the top, depth unknown. The upper 10 cm of the stratum were excavated. Only a few artifacts were found (Ronen et al. 1998). Beyond the upper 10 cm the layer is sterile.
The southeastern corner of the excavated area is partially disturbed by an erosional channel and deposition of reworked clays. The excavation yielded 993 lithic artifacts with an average density of ca. 90 items per square meter. The lithic material shows minor signs of post-depositional abrasion
BRAT5
BR1996
BRT1
BRT2
BRT3
N=689
N=950
N=41
N=138
N=28
Fresh
95%
83%
88%
81%
96%
Slightly abraded
4%
15%
12%
17%
4%
Abraded
1%
2%
0
2%
N=203
N=192
Unpatinated
25%
2%
Slightly patinated
70%
18%
Patinated
5%
80%
Preservation
Patination
Table 3.2. Surface condition of the artifacts.
Pebbles Hammerstones
BRAT5 17 2.40% 2 0.30%
BR1996 23 2.3 1 0.10%
Anvils Flaked Pieces Detached Pieces Secondary Knapped Flakes Second Generation Flakes Chunks Total
38 5.40% 297 42.40% 205 29.20% 137 19.50% 5 0.70% 701
69 6.90% 399 40.20% 431 43.45 20 2% 50 5% 993
BRT1 5 11.10%
BRT2 8 5.40%
BRT3 1 3.60%
1 2.20% 6 13.30% 21 46.70% 9 20% 1 2.20% 2 4.40% 45
17 11.40% 60 40.30% 58 38.90% 3 2% 3 2% 149
1 3.60% 14 50% 5 17.90% 7 25%
28
Table 3.3. Composition of the lithic assemblages.
21
BRT4
BRT6
1 1
1
1
Total 54 2.80% 3 0.20% 1 0.10% 132 6.90% 792 41.30% 708 36.90% 168 8.80% 60 3.10% 1918
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
The vertical distribution of the artifacts and bones in BR1996 is presented in Figure 3.9b. The artifacts appear in a distinct, dense 3–5 cm thick horizon in the southern squares and are more scattered in the northern squares. The southward slope of the archaeological horizon in the northern part of the excavated area most probably reflects the paleotopography of the hamra ground surface. The horizontal distribution of the finds is presented in Figure 3.9a. The lithic artifacts are distributed over the entire excavated area without any apparent clusters. In contrast, most of the bones are concentrated in the southwestern corner of the excavated area.
bearing archaeological remains were found in two small present-day erosional channels (henceforth the northern and southern channels), tributaries of a larger channel (henceforth the eastern channel) running on the east (Figure 3.8). The distance between the northern and the southern channels is 50–120 m. The area excavated in 1996 was located in the northern channel. During the survey, the slopes of the northern and southern channels were cleaned and sampled in sixteen locations, while six additional locations were cleaned in the eastern channel. The stratigraphy of each location was recorded and archaeological remains, when present, were collected. The results of the cleaning show that the archaeological Stratum 4 is present on both sides of the northern channel and in the western part of the southern channel. In the eastern channel, grayish black clay (Stratum 3 of the type-section, see above) and the artifactbearing Stratum 4 are absent from all six studied locations. The section there includes hamra (Stratum 5) overlain by brown clay (Stratum 2 of the type-section, see above). In order to determine the spatial extent of the archaeological horizon, two trenches (BRT4 and BRT6; Figure 3.8) were dug using a backhoe at locations where the archaeological material was not exposed on the surface. In BRT4, excavated in the western part of the site, the artifact-bearing Stratum 4, containing a few bone splinters and a single flake, was detected beneath 1.5 m of clayey sediments, suggesting that the horizon extends to the west, where it is buried under the thick sequence of clays of Strata 1–3. The archaeological layer was not detected in BRT6, excavated in the eastern part of the southern channel. There, hamra is unconformably overlain by laminated alluvial deposits. At the contact between the hamra and the overlying alluvial deposits, a large core weighing 3.8 kg with a few small removals was found (Plate 1), suggesting the former presence there of an archaeological occupation that had been washed away.
Figure 3.9. BR1996, horizontal (a) and vertical (b) distribution of the finds; larger black dots represent bones. The thin curved line marks the edge of the erosion slope.
Area A and Trench 5 (BRAT5) The 2004–5 survey and excavations
During the survey a high density of well-preserved artifacts and bones was recorded in the southern channel. Subsequently, a trench measuring 1x5 m (Trench 5) was excavated in 2004 on its south-facing slope. The artifacts and bones were found in a thin, dense horizon. During the 2005 field season an area (Area A) of ca. 20 m2 was opened to the west of the squares excavated in Trench 5 (Figures 3.8, 3.10; Table 3.1). In the excavations each artifact and bone was recorded by three coordinates using a Total Station, with the exception of deep soundings in squares DN/116 and DP/111, in which the location of the artifacts was not recorded. In squares DR–DS/111 only the upper part of the sandy layer was excavated and only a few artifacts and bones were unearthed. About half of the excavated sediments were wet sieved through 1 mm mesh. The stratigraphy recorded at the northern wall of Area A includes following four units (from top to bottom) (Figures 3.5, 3.10):
A new project was launched at the site in 2004 with the aims of: 1. Determining the spatial extent of the archaeological occupation. 2. Locating and excavating areas with good bone preservation. 3. Recording possible spatial differences in the density, technology, typology and morphology of the lithic artifacts. The fieldwork included survey, geological trenching and archaeological excavations. During the survey, outcrops 22
Chapter 3: The context of the site and the lithic assemblages
Figure 3.10. BRAT5, view to the southeast.
1.
Reworked clays and sands. The contact with the underlying layer is finely laminated. This layer is missing from the type-section.
2.
Stratum 3: Grayish black clay with slickensides and greasy appearance, 1.5–2 m thick. Toward the middle of the unit there is a well-developed calcic horizon with elongated rhyzoconcretions. The contact with the underlying Stratum 4 is diffuse.
3.
Stratum 4: Grayish/yellowish gray sands with numerous small manganese concretions, 40–50 cm thick. Virtually all the archaeological remains were found in this unit. The contact with the underlying Stratum 5 is diffuse.
4.
Stratum 5: Hamra paleosol, depth unknown.
bones is presented in Figures 3.11, 3.12 and 3.13. Eighty artifacts and a few hundred bone splinters that were retrieved during the sieving are not included in the distribution maps. The artifacts and bones appear in a horizon 10–15 cm thick in the lower part of Stratum 4, on top of the hamra. A few artifacts were found in the top few centimeters of the hamra. The vertical distribution of the finds points at a slight undulation of the archaeological layer (Figure 3.11), which is in agreement with the top of the hamra surface as it was recorded during the excavations. The highest point at which the hamra occurred during the excavation is between squares DS/115 and DT/115 (between Easting 4988 and 4990, Figure 3.11). The horizontal distribution of lithic artifacts and bones generally overlaps (Figure 3.12). Both artifacts and bones are scattered over the excavated surface and do not form distinctive clusters. The only distinctive feature is a narrow strip (ca. 50 cm wide) nearly devoid of artifacts and bones in squares DS–DT/115 (Figure 3.12). Since this low-density strip occurs at the highest point of the underlying hamra, it is possibly linked to post-depositional movement of the artifacts to lower locations. In an attempt to gain additional insights into the spatial arrangement of the site, density distribution maps were generated (Figure 3.13). The maps show that squares in the middle of the excavated area are somewhat denser in artifacts and bones, with the densest areas for both roughly overlapping. There is an additional cluster of artifacts in the northeastern corner of the excavated area.
The excavation yielded an assemblage composed of ca. 700 lithic artifacts with an average density of 28 pieces per square meter (Table 3.1). The artifacts are fresh and only slightly patinated (Table 3.2). The degree of patina development was recorded only for artifacts on Mishash Formation chert. DPs compose the main group of the BRAT5 assemblage, followed by secondary knapped flakes and second-generation flakes (Table 3.3). Despite the fact that about half of the excavated sediment was wet sieved through 1 mm mesh, the number of retrieved chips and fragments smaller than 1 cm is relatively low (N=261). The faunal assemblage of BRAT5 includes 591 bones recorded by three coordinates and a few hundred bones retrieved from squares where three-dimensional recording was not employed and from sieving.
Since the spatial distribution of the finds is significant in assessing the agencies responsible for site formation and may
The horizontal and vertical distribution of lithic artifacts and 23
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Figure 3.11. BRAT5, vertical distribution of bones and artifacts.
Figure 3.12. BRAT5, horizontal distribution of artifacts and bones. The thin curved line marks the edge of the erosion slope.
Figure 3.13. BRAT5, densities of bones and artifacts. The thin curved line marks the edge of the erosion slope.
contribute to the understanding of early hominin behavior patterns, the spatial arrangement was examined from four additional perspectives. Firstly, water disturbance can create a pattern in which artifacts are spatially arranged according to their size or weight (e.g. Lenoble 2005, Schick 1986). The artifacts were grouped into size categories to test whether a size-sorting effect is evident within the excavated area. The results show no size-sorting within BRAT5 (Figure 3.14).
Comparisons with other areas will be presented below. Secondly, as with the artifacts, the spatial distribution of bones and the taphonomy of the faunal assemblage can attest to the agencies responsible for site formation. The bones identifiable to a species or body size class were plotted according to skeletal parts. Although no clear anatomical articulations were noted during fieldwork, the distribution 24
Chapter 3: The context of the site and the lithic assemblages
of several fragments hinted at the possibility of some articulated elements having disintegrated in situ. Examples are two concentrations of antelope teeth seemingly from the same jaw that were found isolated but near one another and a concentration of antelope rib fragments, including one conjoin of two fragments with ancient fractures found 80 cm from one another (Figure 3.15). Two additional conjoins, one consisting of two tibia fragments of a mediumsized ungulate and the other of two calcaneus fragments of an antelope, were found. The conjoins show that bone processing and post-depositional decay took place in situ and indicate the integrity of the assemblage. Thirdly, hominin activities may create concentrations of functionally related artifacts, such as pieces with modified edges, pieces with use-wear signs or sharp flakes. The distribution by group of the artifacts that compose the lithic assemblage is presented in Figure 3.16. Overall, artifacts
Figure 3.14. BRAT5, horizontal distribution of artifacts according to size-groups.
Figure 3.15. BRAT5, distribution of bones identified to species or size-class. The black rectangles mark conjoins or bones of the same species forming a concentration that probably belongs to the same specimen.
25
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Figure 3.16. BRAT5, spatial distribution of artifacts by group within the lithic assemblage. The bipolar group consists of bipolar cores and flakes. The circle marks the concentration of FPs.
from different groups are distributed over the entire area and do not form clusters or concentrations that can be clearly interpreted as distinct activity areas. The only exception is a group of five FPs on the boundary between squares DR/114– 115 (Figure 3.16).
belong to the same type of raw material. Six types of chert were identified in BRAT51. The distribution maps for raw materials demonstrate that artifacts are scattered over the 1
Fourthly, knapping forms concentrations of artifacts that 26
In addition to Mishash and translucent chert, four groups of Eocene chert were distinguished on the basis of differences in hue and inclusions.
Chapter 3: The context of the site and the lithic assemblages
entire excavated area and, with only one exception, do not form concentrations in which only one type of raw material prevails (Figure 3.17). The exception here is a group of DPs on Mishash chert in the northeastern corner of the excavated area, which otherwise contains only one Eocene chert artifact. Attempts to refit the Mishash chert artifacts within the group were unsuccessful.
The distribution maps of the two major types of raw material in BRAT5, Mishash chert and Eocene chert group 1, show a scatter of detached and flaked pieces that may reflect the presence of a knapping area. The raw material groups that contain smaller numbers of artifacts show distribution patterns that are not characteristic of knapping areas. The distribution patterns of Eocene chert group 4 and translucent
Figure 3.17. BRAT5, spatial distribution of artifacts by raw material group. Eocene 1–4 – different types of Eocene chert. Large black dots – FPs; small black dots – DPs, secondary knapped flakes and second-generation flakes. 27
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
chert demonstrate that FPs are grouped together and isolated from DPs. This kind of spatial arrangement does not correspond with a knapping spot (e.g. Newcomer 1980, Schick 1986, Toth 1982). Similarly, the isolated position of the sole FP on Eocene chert group 3 is not typical of a knapping area.
DPs and secondary knapped flakes are the largest groups in the assemblage (Table 3.3). Only a few small bone splinters were found in BRT2. Bizat Ruhama Trench 3 (BRT3) A trench measuring 1x7 m was opened on the southern slope of the northern channel, at the easternmost exposure of the artifact-bearing deposit. The archaeological layer was reached and excavated in only 1 m2. The material was collected in quarter-meter squares and the sediments were wet sieved through 1 mm mesh. Twenty-eight artifacts, 96% of which are fresh, and 20 bones were unearthed (Tables 3.1, 3.2, 3.3). The section is composed of Strata 3–5 of the typesection (Figure 3.5), from top to bottom:
Bizat Ruhama Trench 1 (BRT1) A small trench measuring 2 m2 was excavated 5 m to the west of BR1996 (Figure 3.8). The excavation yielded 103 lithic artifacts and a few bone splinters (Tables 3.1, 3.2, 3.3). The section of BRT1 is composed of Strata 3–5 of the typesection (Figure 3.5), from top to bottom: 1. Stratum 3: Grayish black clay, 1.5 m thick. The contact with the underlying Stratum 4 is diffuse.
1. Stratum 3: Grayish black clay, 0.2–0.5 m thick. The contact with the underlying Stratum 4 is diffuse.
2. Stratum 4: Yellowish gray sands. The sand is partly colored with greenish-gray and purple-red patches toward the bottom. The archaeological finds are concentrated in the lower 15–20 cm of the stratum. The contact with the underlying Stratum 5 is diffuse.
2. Stratum 4: Yellowish gray sands, 0.25 m thick. The archaeological finds are concentrated in the lower part of the stratum. The contact with the underlying Stratum 5 is diffuse.
3. Stratum 5: Hamra. A few artifacts were embedded in the upper few centimeters of the stratum.
3. Stratum 5: Hamra paleosol, depth unknown. One conjoin of two flake fragments found in different quarter-meter squares, the possible distance between them being 50–150 cm, was made. The flake was probably broken on an anvil.
Bizat Ruhama Trench 2 (BRT2) A trench 4 m long and 1.5 m wide was excavated on the southern slope of the northern channel, approximately 10 m southeast of BR1996 (Figure 3.8; Table 3.1). The material was collected in quarter-meter squares and the sediments were wet sieved through 1 mm mesh. The section is composed of three units corresponding to Strata 3–5 of the type-section (Figure 3.5), from top to bottom:
Faunal remains from BRAT5 The study of the faunal remains was conducted by Yeshurun et al. (2011). Only faunal remains from BRAT5 were included in the study, since other areas did not provide sufficient faunal samples. The bone assemblage retrieved in BRAT5 is heavily fragmented, but the preservation of cortical surfaces is good. Complete bone elements are nearly absent and the assemblage is essentially composed of isolated teeth and limb bone shaft fragments (Yeshurun et al. 2011). The faunal assemblage is composed entirely of ungulates. Small game and carnivores were not found. The ungulate assemblage includes 141 bones identified to species or size class. The assemblage is taxonomically dominated by Equus cf. tabeti, followed by a mediumsized spiral-horned antelope (Antelopini gen. et sp. indet., probably Pontoceros ambiguus or Spirocerus sp.) and some bovines (probably Bison sp.) and gazelle (Gazella sp.; for a detailed description of the assemblage see Yeshurun et al. 2011). The bone surfaces bear some evidence of hominin modifications. One definite cutmark was observed, along with percussion marks including pits, microstriations and conchoidal notches on five specimens, all of medium-sized ungulates (11% of the relevant NSP; 25% of MNE). In total,
1. Stratum 3: Grayish black clay, disturbed by patches of sandy brown clays and yellowish sands, 1.5–2 m thick. The contact with Stratum 5 is sharp. 2. Stratum 4: Yellowish-gray sands with numerous lenses and pockets of grayish and brown clay and yellow sand. The artifacts appear in the lower 5 cm of Stratum 5. The contact with Stratum 5 is sharp. 3. Stratum 5: Hamra, depth unknown. The artifact-bearing horizon in BRT2 is disturbed by intrusion of clay patches from the overlying Stratum 3. The lithic assemblage is composed of 149 artifacts. The artifacts lay above the interface between sand and hamra, some directly on the top of the latter (Figure 3.5). The density of lithic artifacts in BRT2 is 37 pieces per square meter (Table 3.1). Most of the artifacts are fresh (Table 3.2). 28
Chapter 3: The context of the site and the lithic assemblages
at least a quarter of the limb bones were cracked open for the extraction of marrow. Almost half of the limb-bone shaft fragments from all size classes display “green” (fresh) fractures. Moreover, nearly all shafts retained less than half of their original circumference. The last two observations strengthen the notion that bone marrow was routinely exploited by the Bizat Ruhama hominins.
hominin involvement) and for the evidence for extraction of marrow, usually left untouched after carnivore consumption (Yeshurun et al. 2011).
SITE FORMATION The excavations at Bizat Ruhama unearthed a site complex that yielded a number of lithic and bone assemblages. The site was systematically sampled during the survey and the archaeological remains were found to occur over an extensive area in a distinct horizon, 10–15 cm thick. A few localities in different outcrops were methodically excavated. From the stratigraphic position and the composition of the assemblages it is clear that all the excavated areas belong to a single site complex. The information collected allows us to approach some broader questions concerning the hominin occupation of the site:
The vast majority of the bones and teeth at Bizat Ruhama are coated with black manganese but otherwise display fair preservation. The incidence of bleaching, weathering, cortical exfoliation and abrasion of bone edges is low (Yeshurun et al. 2011), indicating relatively speedy burial of faunal remains in a favorable sedimentological environment. The weathering in particular is surprisingly low for an open-air site. The scarcity of rounded edges, cracking and exfoliation presents additional evidence of the minor role of water or other geological agents in the deposition and destruction of the assemblage. Root (biochemical) marks and trampling striations appear on about one third of the specimens (Yeshurun et al. 2011). The latter may be induced by sediment compaction or by hominin and animal trampling. According to experimental studies conducted in East Africa, unburied bones will totally disintegrate in the course of several years (Behrensmeyer 1978). The good preservation of bone cortical surfaces at Bizat Ruhama thus suggests that they were exposed for a short period of time. The dominant skeletal parts within each size class at Bizat Ruhama are teeth (all isolated), followed by limb-bone shaft fragments. The survival of bone parts correlates significantly with their mineral density. Overall, the densest elements in the skeleton (teeth) are the best represented. The next densest parts, limb bone shafts, are the next best represented, and almost no elements with low density values have survived. While limb-bone ends and skull pieces are nearly absent, denser parts of these elements (long-bone shafts, teeth and petrosum) do exist in the assemblage, indicating that more porous parts of the skeleton were indeed present at the site but were differentially preserved and subsequently obliterated as a result of destruction processes. For instance, dense limbbone shafts outnumber porous limb-bone ends for all limb bones in the assemblage, sometimes in the proportion of 3:1, despite the fact that they accumulated at the site as complete bones before being fractured by hominins or carnivores.
1.
Are the lithic assemblages under study in primary anthropogenic context?
2.
Is the co-occurrence of the lithic artifacts and bones related principally to human activity or to natural causes?
3.
How large is the Bizat Ruhama archaeological site?
4.
Does the site constitute a continuous scatter of artifacts or distinct concentrations?
5.
If distinct concentrations do exist, how large are they and how clear are their boundaries?
6.
Does the site represent a single occupation episode, a palimpsest or a number of distinct episodes?
7.
What does the spatial distribution of artifacts and bones reveal about site formation and hominin behavior?
The following discussion benefits from actualistic and archaeological studies conducted in East Africa. Sites in Olduvai Gorge and Koobi Fora were at the focus of multidisciplinary archaeological research for many years. In Koobi Fora, site formation processes, land use patterns, inter- and intra-site variability and spatial organization were studied in a comprehensive experimental, archaeological and environmental approach. Experimental studies in site formation, tool production and artifact and bone preservation were used as analogies on the basis of which the archaeological interpretations were developed (Behrensmeyer 1978, Bunn 1981, 1983, 1986, Isaac 1997b, Isaac and Harris 1978 and 1997, Kroll and Isaac, 1984, Schick 1986, 1987, Toth 1982, 1987, 1997). The inferences made in East Africa can be used in the interpretation of the results of the excavations at Bizat Ruhama, since the site has some important similarities to
Although there is some evidence for carnivore involvement in the form of several gnawed and tooth-scored bones, the absence of large carnivore remains is noteworthy. On the whole, the evidence points to an accumulation resulting largely from anthropogenic activities representing exploitation of carcass parts of large ungulates. The ungulate remains may have been acquired by hunting. An alternative likely scenario is the acquisition of ungulate carcass parts by scavenging from carnivore kills, accounting for the gnaw marks (as a result of carnivore defleshing before 29
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Early Pleistocene East African sites: 1.
Bizat Ruhama is an open-air site with a single thin archaeological horizon, similar to most of the Koobi Fora and Olduvai sites.
2.
The archaeological remains derive from a sedimentological setting resembling those of some of the Koobi Fora and Olduvai sites.
of Stratum 5. The sand of Stratum 4 shows some features that are indicative of low-energy water deposition; for example, in the BR1996 section (Figure 3.5) part of Stratum 4 exhibits fine laminations. It was not possible to establish in the field whether such processes are related to the archaeological horizon or to a subsequent, unrelated depositional event. The vertical plots of BRAT5 and BR1996 confirm the field observations. The lithic artifacts and bones form a zone 10–15 cm thick with the densest cluster on, or immediately above, the contact between hamra and clayey sand. The vertical distribution of the finds follows the undulating paleotopography of the hamra surface (Figures 3.9, 3.11). At a larger scale, the undulation is clearly evident when the elevations of the top of the hamra in different areas within the Bizat Ruhama site complex are compared (Table 3.1).
Site size and size of individual concentrations The size of the site is reconstructed using the results of the survey and the sampling of the clayey sand layer that bears archaeological remains. Stratum 5 is exposed at the bottom of the sequence across the entire badland field, while Stratum 4 is a restricted phenomenon, found only in the site area and in one additional location ca. 250 m south of the site that was discovered in 2012. At the site, the clayey sand was found over an area of several thousand square meters, most of which is still covered by the clays of Strata 2 and 3. For instance, the approximately 50 m separating between BRAT5 and BR1996 are still covered by 2–3 m of clay. Whether the area between them was continuously occupied will be resolved only when additional excavations are conducted. The results of the survey indicate that artifacts occur not over the entire exposure of Stratum 4 but rather in a few spots along the exposure (see Figure 3.8). Consequently, at this stage it seems reasonable to view the site as reflecting repeated occupations in different locations.
Micromorphological and granulometric studies conducted by Mallol et al. (2011) provide additional insights into the depositional and post-depositional processes associated with Strata 3, 4 and 5 and the archaeological assemblages. Stratum 5 Stratum 5 exhibits micromorphological features that conform with previous descriptions of hamra paleosols (Dan et al. 1969, Yaalon and Dan 1967), i.e., a dune-like aeolian deposit made up of massive subrounded to rounded quartz sand, with traces of bioturbation and clay illuviation related to its permeable nature. Leaching of clay is evidenced by the presence of clay coatings and intercalations. The results also show features of surface water saturation. Dune soils, which are easily and recurrently saturated by water, are often hydromorphic. Extensive iron-manganese segregation, responsible for the characteristic reddish color of hamra soils, is ubiquitous at the site, and indicates pseudogley or surface gley (Stoops and Eswaran 1985). Pseudogley forms in soils in seasonal climates that entail significant percolation of water during the wet season and desiccation, including dehydration of iron oxides and consequent reddening, during the dry season (PiPujol and Buurman 1998, Muckenhausen 1963). The microfissured pattern of the groundmass and frequently cracked edges of large pores (indicative of recurrent surface desiccation) and the few iron-manganese concretions from the Stratum 4–5 interface (indicative of reduction from waterlogging, iron movement and reoxidation) are further evidence of a semiarid environment.
The excavation did not reach the limits of the concentrations of lithic and faunal remains in the excavated areas. Thus, it is hard to determine the size of the individual concentrations. In BRAT5 it is likely that the concentration is far more extensive than the excavated area, because artifacts and bones occur in similar densities in square DN/116 (Figure 3.12) and on the opposite slope of the channel (5 m to the south). It is also possible that BR1996, BRT1 and BRT2 are all parts of a single concentration, since they are located only a few meters apart from each other (Figure 3.8).
The depositional settings During excavation of the different areas it became evident that the archaeological remains are embedded in the lowermost few centimeters of sand of Stratum 4. The transition between Strata 5 and 4 is variably sharp and diffuse in different areas but could be clearly recognized in the field. The change in texture and color of the sediment is quite obvious and was fairly easy to trace during excavation. As a result, it was possible to expose the top of Stratum 5 extensively. It was clear during the field work that the finds were associated with a paleosurface represented by the top
Stratum 4 Stratum 4 is massive and lithologically similar to Stratum 5, but with higher proportions of coarse sand and very little clay. It displays microscopic features of incipient pedogenesis similar to those observed in the hamra unit and hence possibly represents the continuation of the same 30
Chapter 3: The context of the site and the lithic assemblages
kind of sedimentary environment. The close lithological similarity between Strata 4 and 5 suggests that the former derives from the latter. Considering that Stratum 5 represents a poorly vegetated semi-arid landscape, denudation of this layer in the dry seasons could have led to the accumulation of sand in the most depressed areas.
unclear. Sedimentary environments of these types yield delicate structures that are highly prone to destruction by bioturbation, which incidentally has been documented in this layer. Different kinds of low-energy deposition in BRAT5 and BR1996 are suggested by the different microstructures observed in micromorphological samples. BRAT5 shows stronger bioturbation, pseudogley and more abundant ferruginous rootlets, whereas the sediment of BR1996 exhibits well-preserved fine laminations of silt and clay. This points to water deposition involving slightly higher energy in BR1996 and stagnation in BRAT5.
Although both layers are bioturbated, none of the postdepositional features (such as clay infillings or pedotubules) of Stratum 4 penetrate into Stratum 5, indicating that the soil material at the contact between the two layers was not strongly disturbed and could be in a primary position. Occasional rounded aggregates of sandy clay were identified at the base of Stratum 4 and at its contact with Stratum 5. These particles represent relicts of an episode of sedimentary stasis during the early formation stages of Stratum 4 and indicate that the top of Stratum 5 represents a buried surface.
The archaeological remains are depositionally related to sediments of Stratum 4. No reworked microscopic elements from Stratum 3 were found within Stratum 4, which would be the case if the lithic and bone remains derived from it. Moreover, none of the post-depositional features (such as clay infillings or pedotubules) of Stratum 4 cross down to Stratum 5, indicating that the soil material at the contact between the two layers was not strongly disturbed and could be in a primary position. The lithic artifacts and bone remains are associated with the basal Stratum 4 sediment or the Strata 5–4 interface, and were possibly deposited concomitantly with the accumulation of the Stratum 4 sands (Mallol et al. 2011).
The predominance of sand, the aeolian sediment and the lack of visible truncations or coarse-grained lag material (indicative of a sandy surface in the process of deflation) indicates that the inter-dune depression was dominated by deposition rather than deflation. Given their fresh preservation states, the archaeological remains cannot be considered possible lag material. Nevertheless, some degree of seasonal deflation and local redistribution of surface sand could also have been partly responsible for the buildup of Stratum 4. So far we can only infer such a process, as sedimentary structures indicative of wind transport in depressed areas, such as low angle cross-bedding (Kocurek and Dott 1981, Pye 2009), have not been documented. Structures of these types are highly susceptible to bioturbation and are not likely to survive over time.
Stratum 3 Stratum 3 represents a different style of sedimentation and pedogenesis from those of the underlying units. It is much more clayey, finer-grained, and exhibits microstructures that resemble vertisols, such as very little porosity, strongly granostriated b-fabrics and prismatic microstructures. Its dense fabric entails poor drainage, and hence water saturation leading to gley (a bleaching of the groundmass as a result of intense waterlogging), accumulation of organic matter, strong fissuring and cracking of the soil structure upon drying. Such features, together with the low chroma of these sediments, indicate hydromorphism of the kind that results from a semi-permanently waterlogged setting (Stoops and Eswaran 1985).
The presence of bioturbation, plant root structures and pseudogley in Stratum 4 indicates that, similarly to the setting during the formation of Stratum 5, the depression was seasonally damp. We can infer the existence of microtopography from the differences in depth of the top of Stratum 5 across the different exposures, not only in BRAT5 and BR1996 but in the other trenches as well (Table 3.1). This is an additional indicator of a seasonally damp inter-dune surface, as it reflects the irregular distribution of deflation and deposition. Rainfall in such a setting results in low-energy surface wash and rainsplash leading to the formation of colluvial mantles (Pye 2009: 294). The fine laminations locally preserved in BR1996 attest to the removal of finds and their redeposition nearby. The remains of hominin activity were plausibly buried by such lowenergy mechanisms.
Preservation of finds and its implications for site formation The ample literature that deals with the preservation of lithic artifacts in open-air environments links the post-depositional alteration of their surfaces and edges with two major mechanisms: fluvial transportation and trampling (Flenniken and Haggarty 1979, Gifford-Gonzalez et al. 1985, Harding et al. 1987, McBrearty et al. 1998, Nielsen 1991, Petraglia and Nash 1987, Petraglia and Potts 1994, Pryor 1988, Schick 1986, Shackley 1974, 1978, Shea and Klenck 1993, Villa
In sum, Stratum 4 likely formed on a depositional interdune depression in a semi-arid climate, possibly drier than during the formation of Stratum 5. The evidence points toward low-energy modes of deposition by wind and water, although the precise depositional mechanisms remain 31
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
and Courtin 1983). Although experimental studies show that the effects of trampling on the edges of stone artifacts should not be underestimated, there are no clear criteria to distinguish between edge damage caused by fluvial transport and trampling (Shea 1999). Abrasion of artifact surfaces, on the other hand, is soundly correlated with the sedimentary setting and degree of fluvial disturbance in open-air sites (e.g. Bar-Yosef and Goren-Inbar 1993, Harding et al. 1987, Isaac 1997b, Petraglia and Nash 1987, Petraglia and Potts 1994, Schick 1986, 1992, Shackley 1974).
assists in understanding the formation of archaeological sites. Patination results from exposure to light combined with the composition of surrounding matrix and may develop in a short period of up to a few months (BarYosef 1993, Nadel and Gordon 1993). The degree of patina development at Bizat Ruhama was recorded only for artifacts made on Mishash Formation chert from BR1996 (N=192) and BRAT5 (N=203), since it proved difficult to distinguish patina on artifacts made on other types of raw material. The observations were made by the naked eye. Striking differences in the degree of patination were observed between BR1996 and BRAT52. In BR1996 most of the artifacts are patinated and only a few show no patina that can be distinguished by the naked eye (Table 3.2), while in BRAT5 most artifacts are only slightly patinated and 25% show no patina at all. Since the sedimentological settings in both areas are similar, these differences may provide evidence for a longer exposure in BR1996.
The present description of artifact surface abrasion at Bizat Ruhama employs a threefold division adopted from BarYosef and Goren-Inbar (1993) and Shea (1999). The state of preservation of the artifacts was defined as fresh, slightly abraded or abraded. It should be noted that there are no heavily rolled artifacts in any of the studied assemblages of Bizat Ruhama. As a rule, the material shows only minor signs of macroscopic post-depositional abrasion. The frequency of abraded artifacts ranges between 0.2% and 2% in the excavated areas. It can be concluded that the mechanisms leading to abrasion of chert artifacts were minimal at Bizat Ruhama and virtually absent in BRAT5 and BRT3.
Horizontal distribution The spatial distribution of archaeological remains in early hominin open-air sites may be the outcome of a number of sometimes interrelated processes that include hominin activities (food processing, knapping etc.), animal activities and post-depositional biogenic, geogenic and anthropogenic agencies. The depositional issues associated with anthropogenic and biogenic activities have been addressed in faunal and geoarchaeological studies (see above). The effects of flowing water may also be evident in the horizontal distribution of the lithic artifacts. As demonstrated by experimental studies (Schick 1986), water flows tend to:
Archaeological and experimental studies in open-air settings show that heavily rounded and abraded stone artifacts either occur in coarse-grained fluvial contexts or indicate long transportation from their original location by highvelocity and prolonged flows (e.g. Bar-Yosef and GorenInbar 1993, Harding et al. 1987, Isaac 1997b, Petraglia and Nash 1987, Petraglia and Potts 1994, Schick 1986). Since numerous factors influence the degree to which a stone tool is abraded (texture and hardness of the material, sedimentary context, water velocity, duration of the flow event etc.), it remains unclear how much post-depositional disturbance is needed to develop abrasion that can be distinguished macroscopically. According to a four-level division of sites disturbed by water flow proposed by Isaac et al. (1997), artifact surface abrasion is evident only at level 4, i.e. on artifacts transported tens or hundreds of meters from their original location by strong fluvial events. At levels 2 and 3 small-scale events will winnow small-sized artifacts and rearrange them within the living area, but will not leave signs of abrasion on artifact surfaces (Isaac 1997a, Schick 1986). The experimental study conducted by Harding et al. (1987) shows that in the case of handaxes transported in a river bed, abrasion becomes visible after 150 m. To sum up, it seems that while the absence of abraded artifacts suggest that the site was not subjected to large-scale disturbance events, small-scale events would not have left a mark on artifact surfaces. Following these criteria, it might be suggested that the artifacts of Bizat Ruhama were not subjected to any large-scale water disturbance.
1. Create clusters of artifacts and bones. 2. Create imbrications. 3. Arrange artifacts with the longest axis along or transverse to the water current. 4. Create the effect of size sorting. During strong fluvial events, size sorting is usually accompanied by local change in the composition of the lithic assemblage, since flaked pieces tend to be larger, thicker and heavier and have a different volumetric structure from DPs. In small-scale events only small-sized flakes and bones will be winnowed. When interpreting the horizontal distribution of the Bizat Ruhama lithics and bones, one should bear in mind that the largest area excavated so far is 25 m2. This is a small area that may be only a sample of a much larger concentration 2
The degree of patina development is another aspect that 32
It should be emphasized that heavily patinated artifacts were found only on the surface during the survey
Chapter 3: The context of the site and the lithic assemblages
Bizat Ruhama (261 in BRAT5, collected from 6 m2; 138 in BR1996, collected from four squares), despite sieving with 1 mm mesh. If at least some knapping did take place at the site, as suggested by the composition of the assemblages, then on the basis of knapping experiments higher numbers of pieces smaller than 1 cm would be expected (e.g. Delagnes et al. 2006, Schick et al. 1991; Appendix 3C).
of hominin refuse. Nevertheless, some inferences can be made. The horizontal distribution patterns at Bizat Ruhama are characterized by unclustered, scattered distribution of artifacts and bones over the entire excavated area in BRAT5 and BR1996. The finds do not form clusters or imbrications characteristic of water currents. The orientation of remains at Bizat Ruhama does not add significant information, since artifacts and bones only rarely have a pronounced longitudinal axis.
Both the deficit in chips and the variation in frequencies of second-generation flakes may be connected to postdepositional winnowing of small particles during the burial of the site. Stratum 4 contains higher proportions of coarse sand and very little clay in comparison with Stratum 5. This suggests that the fine fraction was either deposited elsewhere or post-depositionally winnowed. Micromorphological evidence of seasonal waterlogging (indicated by the presence of pseudogley) suggests that the removal of sandsized material from the surface may be a response to a complex process of combined seasonal deflation and lowenergy deposition. These processes may have caused the displacement of small, light artifacts as well. In BR1996, the scarcity of second-generation flakes fits with evidence for higher-energy deposition observed during the field work and in micromorphological slides (see above).
The distribution of the artifacts by length group is presented in Figure 3.18. The distribution is unimodal, showing a peak in the 20–24.9 mm group in all the assemblages. Thus, the assemblages do not show the degree of size sorting expected in large-scale water-disturbed sites. Yet, there are noticeable variations between the areas in the frequency of artifacts that are 10–19.9 mm long. The size class of 10–19.9 mm is composed largely of thin, light second-generation flakes, which are considerably more abundant in BRAT5 and BRT3 (Table 3.3). The differences in frequency of second-generation flakes do not seem to a result of hominin activities, since the Clactonian notches and secondary knapped flakes from which they were produced were found in all excavated areas, suggesting a post-depositional origin for the phenomenon.
The distributions of artifacts and bones at Bizat Ruhama clearly overlap. During the survey, bones were found only in association with lithics. In BRAT5 the horizontal distribution
Also noteworthy are the low numbers of chips and fragments (smaller than 1 cm) in all the excavated areas at
Figure 3.18. Length histogram of the artifacts from all excavated areas. 33
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
patterns of bones and lithics overlap as well. In BR1996 only a few bones were found, mostly concentrated in one square. The distribution of the bones suggests in situ butchery and marrow extraction (see above). The distribution of lithic artifacts shows no distinct activity areas and does not display the pattern characteristic of knapping areas (Figures 3.16, 3.17; see also above). Systematic refitting attempts have not yet been conducted for the BRAT5 and BR1996 areas. Hopefully, in the future they will provide additional insight into the spatial organization of the site and its formation processes. Refitting is feasible, as demonstrated by a conjoin of two flake fragments found in BRT3.
bones suggests that hominins occupied the site during the gap in sedimentation that occurred after the formation of the hamra. This gap is observable in micromorphological slides in the form of clayey/sandy aggregates at the base of Stratum 4. The fact that artifacts and bones formed a zone 10–15 cm thick probably stems from vertical displacement of the finds after primary deposition. Experimental and archaeological work in Koobi Fora shows that “where the east African PlioPleistocene open-air sites formed on loose sandy or silty substrates, post-depositional processes could have readily dispersed an archaeological zone through a thickness of at least 10–15 cm of sediments” (Kroll and Isaac 1984:12). The vertical post-deposition movement seems to have occurred even when the objects largely remained in place horizontally. Among the factors responsible for the vertical movement of the archaeological remains, the most significant are mechanical processes involving wetting and drying of sediment, biogenic agencies and human trampling (Isaac 1997b, Kroll 1997, Stockton 1973, Villa and Courtin 1983).
To sum up, the horizontal distribution of the artifacts and bones is not indicative of large-scale disturbance events. The remains are largely in primary deposition, though the deficit in small chips and variations in the frequency of secondgeneration flakes probably result from post-depositional winnowing. The artifacts and bones at Bizat Ruhama were found in what seems to be hominin-induced association. First, the distribution of bones and artifacts clearly overlaps; there are no areas among the exposures of the archaeological horizon in which bones were found without artifacts. Second, the distribution of bones points to in situ processing and post-depositional decay. The distribution of lithic artifacts is not typical of knapping or other activity areas and cannot be interpreted in terms of hominin use of space.
The faunal remains show signs of butchery and marrow extraction and their association with the lithics is clearly hominin-induced. The good preservation of bones and low degree of patination of the artifacts in BRAT5 attest to fast burial of the site. During this burial by sands accumulating in the depression, some small particles were winnowed. Larger artifacts may possibly have been rearranged within activity areas as well, thereby masking the original patterns of hominin use of space. It has been shown by experimental studies that small-scale fluvial events can relocate a single artifact within the activity area without changing the composition of the lithic assemblages to a significant degree (Schick 1986). The unpatterned distribution of artifacts in BRAT5 may result from such a rearrangement or alternatively reflect a lack of spatial patterning on the part of the Bizat Ruhama hominins. Micromorphological and zooarchaeological evidence indicate only minor post-burial disturbance. BRAT5 and BRT3 show better preservation of original site features than other areas, likely because of faster burial. Overall, it seems that among the factors responsible for site formation at Bizat Ruhama, hominins rather than animals or geogenic agencies played the major role. The results show that Bizat Ruhama is a well-preserved site in primary context and that the lithic and faunal assemblages were accumulated by hominins and are largely undisturbed.
Summary: site formation history The Bizat Ruhama site represents hominin occupation in an inter-dune depression, an area confined by dunes that underwent relatively continuous sedimentation and soil formation during the Early Pleistocene. Stratum 5 constitutes a hamra soil that formed in a semi-arid climate. Under the same environmental conditions, the sandy layer above (Stratum 4) represents the input of locally reworked sand and soil aggregates from the hamra topsoil, plausibly through wind and overland flow. Increasing water saturation and sedimentary accretion in wet conditions in the Bizat Ruhama depression (Strata 2 and 3) postdate the hominin occupations and are unrelated to them. At present, the evidence suggests that the archaeological remains accumulated on the ground surface of the hamra during several occupation episodes that took place over a relatively short time-span. The position of the lithics and
34
CHAPTER 4: LITHIC RAW MATERIALS LITHIC RAW MATERIALS IN THE BIZAT RUHAMA AREA
in the area, seven conglomerate exposures mapped during the 2000–2001 survey (Zaidner 2003a, b) were randomly sampled. The outcrops were identified on the basis of the geological study of the area (Bar-Yosef 1964), the geological map (Sneh et al. 1998), the description of the Pleshet and Ahuzam Formations (Issar 1961, Zilberman 1984, 1986) and field observations made with the help of Noam Greenbaum (Department of Geography, University of Haifa).
Bearing in mind the changes in landscape that have taken place since the occupation, it is impossible to identify the specific exposures used by the Bizat Ruhama hominins to obtain raw materials. It is possible, however, to trace the local availability of raw materials by examining the shape, size and lithology of the present-day surroundings. Subsequent comparison between the raw materials in the area and those used at the site may reveal patterns of raw material acquisition, selection and transport.
The five exposures of Pleshet conglomerate nearest to the site were sampled (S1, S2, S4, S5 and S7 in Figure 3.3), along with the only two exposures of Ahuzam conglomerate available in the area (S3 and S6 in Figure 3.3). In order to assess the internal variability within a single exposure, Samples 2 and 4 were collected about 100 m apart from a single extensive exposure of the Pleshet Formation at Nahal Sad. The description of the sampled locations is presented in Appendix 1 and the methodology of sampling in Chapter 2.
In the vicinity of the site, the only hard stone raw material is found in the Pliocene conglomerates of the Pleshet Formation and, to a lesser extent, the Ahuzam Formation. Conglomerates constitute an almost continuous horizon in the subsurface of the eastern Negev Coastal Plain, but at present they outcrop only in riverbeds of semi-perennial streams (see Chapter 3, above). Conglomerates are more extensively exposed east of the Nahal Shiqma channel, beyond the boundary of the Quaternary sands invasion.
Lithological composition of the sampled exposures The study pointed to pronounced differences in the lithological composition of the Pleshet and Ahuzam Formations (Table 4.1). The main difference lies in the considerably higher frequency of limestone within the Ahuzam conglomerates. The low frequency of limestone in the Pleshet conglomerates has been previously reported and is explained by the intensive abrasion undergone by pebbles in the coastal environment (e.g. Issar 1961; Buchbinder pers. com. 2003, Greenbaum pers. com. 2008).
In the Bizat Ruhama area today, Pleshet conglomerates outcrop east of the site along the channels of several small streams and the riverbeds of Nahal Shiqma and Nahal Sad (Bar-Yosef 1964, Sneh et al. 1998, Zaidner 2003a, b), the nearest exposures being 2.5 km from the site (Figure 3.3). Ahuzam conglomerates are much more frequent in the Higher Shephela1 region and are only seldom exposed in the site’s vicinity. Only two outcrops of Ahuzam conglomerate were found in the study area, one to the east of the Nahal Shiqma and another close to the recent course of Nahal Adorayim.
Within the exposures of the Pleshet Formation the differences in lithological composition are less pronounced, but still significant. Only the two samples taken from the same exposure in Nahal Sad (Samples 2 and 4) show insignificant differences in lithological composition (X2 = 8.746; df = 4; p = 0.07). Limestone is the most abundant stone type in the Pleshet conglomerates except for the two samples from Nahal Sad, in which brecciated chert is slightly more frequent (Table 4.1). All of the chert types occur in all of the samples except for Sample 7, in which gray chert is missing. The frequencies of chert types vary, with brecciated chert being the most abundant.
Random sampling and characterization of raw material sources To achieve a comprehensive view of raw material availability 1
Sneh and Buchbinder (1984) argue that the outcrops identified as belonging to the Ahuzam Formation in the eastern coastal plain and Lower Shephela may represent later, reworked Ahuzam pebbles.
35
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Table 4.1. Rock type frequencies among the pebbles in the sampled exposures.
Shape of the pebbles in the sampled exposures
Length of the pebbles in the sampled exposures
In geomorphology, pebble shape is one of the most important tools for determining the depositional environment of the conglomerates and differentiating conglomerate exposures from one another. A number of quantitative methods have been developed to measure roundedness, flatness and sphericity of pebbles (e.g. Dobkins and Folk 1970 and references therein). Unfortunately, these methods are inapplicable to the pebbles found in the archaeological assemblages of Bizat Ruhama. First, the small number of complete pebbles at the site does not permit the statistical testing on which quantitative differentiation can be based. Second, the pebbles found at the site were picked up by hominins who may have looked for pebbles of specific shape, size and lithology, and therefore do not represent random sampling of the exposures. Thus, to create comparable datasets for both archaeological sites and conglomerates a simpler qualitative approach was used, similar to the one applied by Goldman-Neuman and Hovers (Goldman 2004, Goldman-Neuman and Hovers 2009; see Chapter 2 above).
The pebbles from both the Pleshet and Ahuzam Formations rarely exceed 150 mm in length (Figure 4.3), with pebbles larger than 150 mm being relatively frequent only in Sample 2. Sample 2 can clearly be divided into two size groups (Figures 4.2.a, 4.3): one group is composed of large subdiscoid pebbles, while the other is composed of small, flat discoid pebbles. Thus, Samples 2 and 4 indicate that length variability within a single exposure can be significant. The length of the pebbles varies significantly between different raw material groups (Figure 4.4). Limestone and brecciated chert pebbles occur in a wide range of lengths and they are the largest in the area. Virtually all translucent and Eocene chert pebbles are smaller than 100 mm and most of them are smaller than 50 mm. The size differences reflect the initial size of the raw material at the primary sources; Eocene pebbles derive from small nodules of the Adulam Formation, while brecciated pebbles come from the thick chert layers of the Mishash Formation. These clear-cut differences in the size of different raw material types are mirrored in the use of raw material in the Lower Paleolithic sites of the southern coastal plain of Israel. Large bifacial tools were often made on brecciated chert, whereas flake tools were made on Eocene or translucent chert (Chazan 2000, Lamdan et al. 1977, Marder et al. 1999, 2006, Ohel 1976; for Nahal Hesi, see below).
The shapes of the pebbles show clear differences between the Ahuzam and Pleshet conglomerates (Figure 4.1). While in the Ahuzam Formation (Samples 3 and 6) angular pebbles are prevalent, in the Pleshet Formation (Samples 1–2, 4–5, 7) they are virtually absent. All the pebbles collected from the Pleshet conglomerates have highly rounded and smooth edges, and some of them are very flat (Figures 4.1, 4.2). Pebbles of the type classified as flat discoid include very rounded, thin pebbles that are characteristic of beach environments. Flat pebbles were not recorded in the Ahuzam Formation. Among samples of the Pleshet conglomerate exposures, considerable differences were recorded in the frequencies of flat discoid and subdiscoid versus subspheroid and spheroid pebbles.
Selective sampling The results of the random conglomerate sampling give a generally reliable picture of raw material availability in the area, but they do not reflect human choices in the selection of raw materials. To examine the possible impact of human 36
Chapter 4: Lithic raw materials
Figure 4.1. Shape frequencies of the pebbles in the sampled exposures.
Figure 4.2. Conglomerate exposures sampled during the raw material study. a) Sample 2, Pleshet Formation at Nahal Sad (note that pebbles appear in two size groups); b) Sample 4, Pleshet Formation at Nahal Sad, well-sorted flat pebbles; c) Sample 5, Pleshet Formation at Nahal Shiqma near Tel Nagila, mostly well-rounded spheroid and some flat pebbles; d) Sample 3, Ahuzam Formation near Nahal Pura, mostly angular pebbles. 37
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Figure 4.3. Length distribution of the pebbles in the sampled exposures.
Figure 4.4. Length distribution for different rock types among the pebbles in the sampled exposures. The T-test indicates that the differences between the length of pebbles of brecciated chert (average length 80.4 mm) and Eocene chert (average length 39.2 mm) are significant (t = 7.041; df = 206; p = 0.00). 38
Chapter 4: Lithic raw materials
selection patterns, we resampled one of the exposures of the Pleshet conglomerate (Sample 1). The resampling was designed to imitate hominin raw material acquisition patterns, on the assumption that hominins may have looked for large, angular pebbles that provided better angles for the initiation of knapping.
of suitable raw material was small chert nodules of the Adulam Formation outcropping 10–15 km east of the site (Sneh et al. 1998; Buchbinder pers. com. 2003), the hominins had to rely on secondary sources. Pleshet Formation conglomerates are the main source of raw material in the area. Pleshet outcrops are extensively exposed and easily accessible within 2.5 km east of the site. Isolated exposures of the Ahuzam Formation were found only at the eastern edge of the studied area, ca. 5 km from the site. The post-occupation changes in the landscape make it difficult to evaluate whether the outcrops available today were accessible during the occupation of the site. Nonetheless, since the Bizat Ruhama site is mantled by loess and clays accumulated during the Middle/Upper Pleistocene (Chapter 3), it is likely that conglomerates were even more extensively exposed at the time of hominin occupation and may have occurred closer to the site than at present.
Sample A Sample A was collected by two persons whose objective was to locate pebbles larger than 20 cm in a time of 30 minutes. In the random sampling of the exposure (Sample 1), pebbles larger than 150 mm were not found. During the selective sampling, nine pebbles between 200 and 300 mm long were collected, indicating that with investment of time and energy large pebbles could be found in the area. In other parameters such as shape and lithology, the results of the selective sampling confirm the conclusions of the random sampling. All the pebbles collected in Sample A are of brecciated chert or limestone and subdiscoid or subspheroid in shape (Table 4.2).
The results of the raw material sampling show clear differences between conglomerates of the Ahuzam and Pleshet Formations, despite the same primary source of the rocks in both conglomerates (the Adulam and Mishash Formations). The differences lie in both the lithological composition and the shape of the pebbles and are linked to the different deposition environments of the Ahuzam and Pleshet Formations. In metrical attributes, pebbles from the random sampling of both formations show the same range of length values, with a maximum length between 150 and 200 mm. The results of the selective sampling of one of the exposures of Pleshet conglomerates indicate that with investment of time and energy, larger pebbles could be found as well. In lithology, shape and correlation between lithology and size, the results of both the random and the selective sampling show similar results. Chert pebbles occur in all sampled exposures of the Pleshet Formation but are rare in the Ahuzam Formation. The spheroid and discoid pebbles of Eocene and brecciated chert are the predominant types of siliceous rocks in the Bizat Ruhama area (Figure 4.5). Notably, Eocene and translucent chert pebbles never exceed 100 mm in length, while brecciated chert and limestone pebbles occur in all size ranges up to 300 mm. Angular pebbles are very rare in the area.
Table 4.2. Description of the pebbles collected during selective sampling (Sample A).
Sample B Virtually all the pebbles collected from the Pleshet conglomerates during the random sampling are highly rounded and do not provide good angles for the initiation of knapping. The goal of Sample B was to collect angular or highly angular pebbles, i.e. to search for pebbles providing angles that facilitate initial removals. In 30 minutes only three angular pebbles were collected, all relatively small (40–70 mm long). The results confirm the conclusions of the random sampling that angular pebbles are extremely rare in the Pleshet Formation in the Bizat Ruhama area.
LITHIC RAW MATERIALS AT THE SITE The evidence from all the excavated areas in the Bizat Ruhama site complex indicates that pebbles served as the primary, and most probably the only, raw material for the lithic production. Of 67 collected cores, 58 are clearly made on pebbles. For the remaining six cores the shape of the raw material nodule could not be identified, since they have no cortex. In addition, the lithic assemblages of the site include 26 pebbles. Moreover, the shape of the remaining cortex
Summary: raw material accessibility and variability Bizat Ruhama is located in an area lacking hard rocks in primary geological position. Since the closest primary source 39
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Table 4.3. Shape of the pebbles in the Bizat Ruhama assemblages. Mora 2005, Harmand 2009); for instance, three subspheroid pebbles exhibit concentrations of percussion marks typical of hammerstones. Nevertheless, it is clear from the shape of the cortical surfaces on the cores and flakes that flat/ spheroid rounded pebbles were used for knapping as well. The length of complete pebbles at the site does not exceed 85 mm (Figure 4.6), the only exception being a pebble of subdiscoid shape found in BRT1 that is 182 mm long.
Figure 4.5. Chert pebbles from the Bizat Ruhama area. 1) Mishash (brecciated) chert; 2) Eocene chert. on the flakes indicates that they too were detached from pebbles.
Except for the limestone core/chopper found in BR1996, the only raw material used at the site was chert. The lithological composition of the three largest assemblages excavated at the site show the prevalence of Eocene chert, followed by brecciated and translucent chert (Table 4.4). BR1996 and BRT2 contain a number of desilicifed pieces categorized as undefined chert in Table 4.4. The differences in frequencies of the three main types of raw materials in the different assemblages are insignificant (Table 4.4).
Table 4.3 demonstrates that the pebbles found at the site are spheroid or discoid in form. They are very rounded and smooth. The spheroid/discoid shapes of the pebbles would normally be considered unsuitable for knapping, as none of them provides good angles for initiation of the knapping sequence. It is likely that some of the pebbles at the site were brought in to serve as hammerstones or pounding tools, as at other Early Pleistocene sites (de la Torre and
Figure 4.6. Length of the pebbles (mm) in the Bizat Ruhama assemblages. 40
Chapter 4: Lithic raw materials
Table 4.4. Chert type frequencies in three archaeological assemblages. The X2 test indicates that the differences between the frequencies of brecciated and Eocene chert in BRAT5 and BR1996 are insignificant (X2 = 8.972; df = 4; p = 0.06).
41
CHAPTER 5: FLAKE PRODUCTION FLAKED PIECES
Fractured flat pebbles
The term “flaked pieces” (FPs) was introduced by Isaac (1986) to include all pieces from which flakes were removed, including retouched pieces. It is much less commonly used nowadays, since most authors prefer to distinguish between cores and retouched flakes (e.g. Delagnes and Roche 2005, de la Torre et al. 2003, Roche et al. 1999, Sahnouni 2006, Shea and Bar-Yosef 1998). At Bizat Ruhama, due to the simplicity of the core forms, the large number of broken pebbles and the high frequency of exhausted cores (Table 5.1), the term was retained to include all forms from which flakes were detached. Knapped and modified flakes represent different stages of the chaîne opératoire and hence were excluded from the FP group. The FP assemblages of Bizat Ruhama were further divided according to the shape of the raw material and the knapping techniques and methods into five subgroups: flat fractured pebbles, fractured pebbles and pebbles with a few removals, cores, bipolar cores and exhausted cores (Table 5.1). In total, FPs constitute 8.7% of the Bizat Ruhama lithic assemblages.
This group is composed of transversally fractured flat discoid pebbles. The selected pebbles are flat, with an average thickness of 11.1 mm. Pebble fragments exhibit two opposite cortical surfaces and a fracture surface that in most cases is flat (57.5%) but sometimes exhibits conchoidal fracture shock waves (12.5%). On other fragments the fracture surface could not be identified because of further knapping. The angle between the cortical and fracture surfaces is 75– 90º. The group includes 23 flat pebble fragments from the main excavated areas and 17 fragments from trenches and surface collections (Table 5.1). Thirteen of the fragments show signs of opposite impacts on the fracture surface (Plate 2B:1, 2). The knapping experiments (see Chapter 7) showed that these characteristics are common when the flat pebbles are broken using an anvil as a support. The majority of the flat pebble fragments are of Eocene chert, followed by translucent and brecciated chert (Table 5.2). Most of them were found in BR1996.
Table 5.1. Breakdown of FPs in Bizat Ruhama assemblages. The X2 test indicates that the differences between the BRAT5 and BR1996 assemblages are significant (X2 = 15.075; df = 4; p = 0.005). 42
Chapter 5: Flake production
Table 5.2. Breakdown of FPs by raw material type. The X2 test indicates that the differences between brecciated and Eocene chert are significant (X2 = 32.357; df = 4; p= 0.00).
Some of the pebble fragments show signs of secondary treatment: 15 of them (37%) were further used as cores from which a few small and thin flakes were removed (Plates 2B:3; 4:6).
based on repeated patterns identified in core management, on approaches to the core volume and on the technique applied.
Fractured pebbles and pebbles with 1–3 removals
This group is composed of cores with striking platforms and debitage surfaces intersecting at an angle that is close to or steeper than 90º (Figure 5.2). They exhibit a series of 2–5 unidirectional removals struck from a prepared striking platform. The striking platform was prepared by a single blow that removed the cortical surface of the pebble or by splitting of the pebble into two halves (Figure 5.2a). The group of orthogonal cores can be further subdivided into unifacial unidirectional cores (cores with a number of removals struck from a single direction, Figure 5.2b) and multifacial unidirectional cores (cores with a series of unidirectional removals struck from different striking platforms and debitage surfaces, Figure 5.2c, d). According to the shape of
Orthogonal cores with series of unidirectional removals (N=30)
This group is composed of three fractured pebbles and six pebbles that were abandoned after a few removals. Although the group is very small, it sheds some light on the initial stage of the knapping sequence, which is rarely visible on more intensively reduced cores. The three fractured pebbles preserve signs of opposite impacts on the edges of the fractured surface (Plate 3:2, 3). They apparently each constitute one half of a pebble fractured on an anvil, which could be later used as a core. All three pebbles are spheroid or subspheroid in shape. The other six pieces are pebbles with 1–3 small removals, lacking signs of the bipolar technique.
Cores The assemblages of Bizat Ruhama include 43 cores from the main excavated areas and 24 cores from the trenches and surface collection. According to the remains of cortex, 92% of the cores were made on pebbles (the remaining 8% have no cortex). Figure 5.1 and Table 5.3 show that most of the cores exceed 30 mm in maximum length and that they are significantly larger than other FP types in the assemblage. Most of the cores were made on brecciated chert, although Eocene chert is more frequent among small and shattered bipolar cores (Table 5.2). The cores can be divided into four technological groups, each representing a different reduction system. The division is
Figure 5.1. Length/thickness scattergram of FP types. 43
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Orthogonal cores Preferential surface cores Multidirectional cores Cores Bipolar cores Indeterminate Total Exhausted cores Flat pebbles
N Mean Std. D. N Mean Std. D. N Mean Std. D. N Mean Std. D. N Mean Std. D. N Mean Std. D. N Mean Std. D. N Mean Std. D.
L 30 39.20 8.452 4 49.07 15.330 10 42.38 11.730 19 25.44 5.229 4 33.22 5.448 67 36.00 11.108 40 24.48 4.769 25 28.41 6.778
W 30 32.46 9.037 4 45.66 17.840 10 35.59 12.132 19 22.00 4.937 4 27.36 5.927 67 30.45 10.939 40 17.08 4.055
T 30 26.99 8.993 4 26.90 8.567 10 26.82 8.838 19 15.80 5.129 4 19.42 2.588 67 23.33 9.107 40 17.27 4.302 25 11.14 2.697
SN 30 11.37 3.783 4 6.00 2.160 10 7.70 2.541 4 10.00 3.830 4 8.25 4.924 52 9.90 3.932 40 2.39 1.201
SL 30 23.06 7.591 4 28.47 12.919 9 24.24 6.500 3 24.23 6.735 3 16.00 4.583 39 23.44 7.854 40 14.79 4.621
SW 30 19.46 5.277 4 24.83 12.425 9 24.07 7.999 3 16.48 4.733 3 12.67 2.887 39 20.32 7.270 40 14.51 3.315
H 23 2.46 1.266 4 1.25 0.500 6 1.50 1.049 17 1.00 1.173 2 1.00 0.000 52 1.55 1.253
IC 13 1.54 2.665 3 1.33 2.309 6 1.00 1.095 17 0.76 1.300 2 0.50 0.707 41 1.07 1.836
Table 5.3. Descriptive statistics of FP assemblages. L = length; W = width; T = thickness; SN = number of scars; SL = length of largest complete scar; SW = width of largest complete scar; H = number of hinged scars; IC = number of incipient cones on core surfaces.
the cores and remnants of the cortex, orthogonal cores were made on spheroid and subspheroid pebbles. Unifacial unidirectional cores (N=9; Plate 4:1, 3, 5). These are orthogonal cores from which a series of 2–5 flakes was detached from a single dominant striking platform. The debitage surface was extended to the lateral parts and exploited ca. 50–100% of the circumference of the pebble (Plate 4:1, 3, 5). Consequently, some of these cores have a semi-prismatic cross-section. The blows were placed far from the edge of the core in order to remove thick flakes. This often resulted in an angular, jagged shape of the striking platform. The removals are elongated and usually cover the entire length of the debitage surfaces. When the striking platform was abandoned, some of the cores were used for opportunistic removal of a few flakes from other directions and on other surfaces. Multifacial unidirectional cores (N=21; Plates 4:4; 5:1, 3, 4). Some unifacial unidirectional cores were reduced further on other debitage surfaces. After a number of unidirectional removals, the striking platform was abandoned, the core was rotated and a new platform was generated. The unidirectional reduction was terminated either because the deep hinged scars prevented further reduction or because the angle between the striking and debitage surfaces became too
Figure 5.2. Schematic chart of the orthogonal unidirectional debitage system. a) Cortex removal; b) Sequence of unidirectional removals; c) and d) Two additional series of unidirectional removals struck from different striking platforms and on different debitage surfaces.
44
Chapter 5: Flake production
obtuse. The previous debitage surface sometimes served as a new platform, while in other cases a new striking platform was prepared by a single removal. A new series of removals was then produced. Orthogonal cores are often heavily reduced, making it uncertain how many series of removals were produced during the lifetime of the core. On most of them two series are visible, while one core exhibits signs of four such series (Plate 5:1). As in the case of unifacial unidirectional debitage, some opportunistic individual removals from different directions ended the core reduction sequence. Because of the original spherical shape of the raw material, the obtuse angles between the surfaces and the intensive reduction, multidirectional orthogonal cores are parallelepiped/cubic or subrounded in shape. Three of them can be classified as subspheroids (Plates 3:1; 4:4; 5:1).
Figure 5.3. Schematic chart of the preferential surface debitage system. a) Cortex removal; b) Sequence of unidirectional unifacial removals; c) Butt rectification; d) Further exploitation of the debitage surface.
Multifacial multidirectional cores (N=10; Plates 4:2; 6:3; 7:8, 9)
exploitation of the debitage surface (Figure 5.3c), probably to correct the angle between the surfaces. The cores exhibit several unidirectional removals that cover the entire debitage surface. The reduction sequence usually ended with deep hinged scars that prevented further exploitation (Plate 8:1). Unlike orthogonal cores, the intersection between the striking platform and debitage surface in preferential surface cores is at a relatively acute angle (Figure 5.4). One of the cores shows four incipient cones 1 cm from the edge of the striking platform, indicating unsuccessful attempts to remove thick flakes (Plate 8:1).
A group of cores were reduced on several exploitation surfaces without any apparent organization in the reduction process. The cores were rotated after almost every removal in a search for an appropriate angle. On some of these intensively reduced cores only the negative of the bulb of percussion of the last detached flake is visible, the others having been removed by successive blows. In many cases each of the exploitation surfaces is covered by the scar of a single removal. Multidirectional cores that were intensively reduced display angular and polyhedral shapes (Plates 6:3; 7:8, 9).
Bipolar cores (N=19; Plates9, 10) Nineteen cores in the Bizat Ruhama assemblage were identified as bipolar. Bipolar cores exhibit crushed and
Preferential surface cores (N=4; Plates 5:2; 8:1) These are cores exhibiting a flat debitage surface with a few unidirectional removals. The preferential surface cores display the highest level of planning among the cores of the Bizat Ruhama assemblages. They are the only cores that show some level of management prior to production of flakes and their reduction follows simple but unvarying rules. Preferential surface cores are made on subdiscoid pebbles and have two hierarchical surfaces. The narrow part of the pebble served as a striking platform and the flat part as a debitage surface (Figure 5.3). The debitage surface is opposite to the unknapped cortical surface of the pebble. The striking platform, which consists of a fracture plane or was initially prepared by one large removal, was later rectified. The rectification consists of the detachment of two or three flakes from the striking platform prior to the
Figure 5.4. Box plot graph presenting angles between striking platform and debitage surface by debitage method. 45
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
angular debitage surfaces, shattered striking platforms and evidence for distal impact. The distal features include scars on the distal edge of the debitage surfaces that are morphologically similar to scars on pièces esquilles (N=2; Plate 10:4), signs of impact, crushing and shattering (N=9; Plates 9:1, 3; 10:1, 2) and signs of shattering with wedgeshaped crack lines (N=8; Plates 9:2, 4; 10:3). Of these cores, 42% exhibit a cortical surface opposed to a single debitage surface, indicating that the core was constantly struck in the same direction. Other cores exhibit one or several additional debitage surfaces, showing that they were twisted and turned. It is possible that some of these cores were first reduced by other methods and were shattered against an anvil only during the last stage of reduction. Among bipolar cores, 55% are relatively thin and have a wedge-shaped profile similar to that previously reported for bipolar core assemblages of later periods (Plates 9:1, 2; 10:1; Binford and Quimby 1963, Jeske and Lurie 1993, Shott 1989, White 1968 and references therein), while others are more polyhedral in shape.
evident that the characteristic products of flaked fragments are thin flakes with sharp edges, which are considerably smaller than the flakes struck from the cores (Table 5.3).
Flaked pieces: an overview The study of knapping methods shows that in basic technological terms there are no substantial differences between the various assemblages of the site. All five technological subgroups identified in the FP assemblage occur in both main excavated areas and along the exposures of the archaeological layer. Similarly, each of the four core reduction systems was identified in both excavated areas, with insignificant differences in frequencies between the groups (Table 5.4). This indicates that the FP assemblages from all excavated areas belong to a single industry. Orthogonal and bipolar systems of debitage dominate the assemblages, while the preferential surface cores that represent the highest level of planning in the core assemblage of Bizat Ruhama are poorly represented.
Exhausted cores (N=48; Plate 6:4–6; 7:1–7, 10) Many cores in the Bizat Ruhama assemblage were knapped to the point where their knapping methods and techniques could not be identified. These are small and heavily reduced pieces, which lack identifiable striking platforms and debitage surfaces. Exhausted cores are the smallest artifacts in the FP assemblages (Table 5.3). In comparison to other FP classes, little cortex remains on their surfaces (Figure 5.5). Interestingly, the exhausted core group is composed almost entirely of Eocene and translucent chert artifacts, in contrast to other FP classes, where brecciated chert is dominant (Table 5.2). These differences may suggest a more intensive reduction of Eocene and translucent chert. The other major factor that should be taken into consideration is the small size of Eocene and translucent chert pebbles.
Some inter-area differences were recorded in the amount of FPs in the assemblages and in the frequency of subgroups forming the FP assemblages. FPs are more abundant in BR1996 than in BRAT5 (Table 5.1). The BR1996 assemblage is dominated by exhausted cores and exhibits a low frequency of cores. Conversely, the BRAT5 assemblage contains only a few exhausted cores and is dominated by other FP classes. The knapping strategies at Bizat Ruhama were strongly influenced by raw material availability. In particular, the shape and the size of the pebbles were likely the prime factor determining which knapping strategy would be used. Figure 5.6 shows a clear correlation between the type of raw material and the reduction method that was selected for core reduction. Brecciated chert pebbles were mostly knapped by the orthogonal system, while Eocene chert is more frequent among bipolar and multidirectional cores. It seems, however, that the selection of debitage method is linked to the size and the shape of the available pebbles rather than their lithology. In the Bizat Ruhama area the pebbles of brecciated chert are
The exhausted cores frequently bear a few scars of small, thin flakes removed in the last stage of reduction. Some of them exhibit up to five scars of such flakes, struck opportunistically from different surfaces and directions (Table 5.3; Plates 6:5; 7:1–7, 10). The scars show signs of conchoidal fracture with Hertzian cones and ripples. From the size of the scars it is
Figure 5.5. Amount of cortex on cores and exhausted cores. 46
Chapter 5: Flake production
Table 5.4. Debitage systems in different Bizat Ruhama assemblages. The X2 test indicates that the differences between the BRAT5 and BR1996 assemblages are insignificant (X2 = 6.962; df = 5; p= 0.22).
Figure 5.6. Debitage systems by raw material type.
considerably larger than other chert pebbles. The brecciated chert cores (mean length 39.9, s.d. 11.14) are significantly larger than the Eocene chert cores (mean length 31.7, s.d. 9.21; t = 2.954, df = 59, p = 0.01). The size of the brecciated chert pebbles allowed for debitage methods requiring more complex organization of the core volume (i.e. orthogonal and preferential surface debitage). Eocene and translucent chert, on the other hand, are uncommon among the formal cores.
The small size of the pebbles of both chert types demanded good knapping and motor skills in order to organize the debitage. Therefore, the size of the pebbles was probably the main reason for using debitage methods that did not require any preparation or organization of the core volume, namely multidirectional and bipolar reduction. Many of the pebbles of both chert types were probably simply smashed between the hammerstone and anvil, producing flakes, bipolar cores 47
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
and exhausted cores. The clear dominance of Eocene and translucent chert among the exhausted cores is probably connected to the small size of the pebbles as well (Table 5.2).
or until deep hinged scars prevented further reduction. Most of the cores in the Bizat Ruhama assemblages have one or more hinged scars (Table 5.3). The cores were often struck several times after the angle between the surfaces became too obtuse, resulting in a number of incipient cones on the core surfaces (Table 5.5). On some of the cores there are several incipient cones, sometimes very far from the edge of the striking platform. The incipient cones occur either on the striking platforms (marking failed attempts to remove thick flakes) or at the center of the core surfaces. The latter probably indicate attempts to break the core in order to maximize the raw material at any cost, even if success was clearly uncertain.
The selection of the debitage method was also influenced by the shape of the pebble. Flat discoid pebbles were broken on an anvil. The resulting fragments were sometimes further flaked in order to produce thin flakes. The orthogonal debitage system was employed only when the pebbles were spheroid or subspheroid in shape. The spherical shape made it difficult to create acute angles, thus entailing knapping at an obtuse angle. Preferential surface cores were made on relatively flat subdiscoid pebbles that allowed an acute knapping angle to be created in a much easier way.
The reduction strategies at Bizat Ruhama are characterized by frequent use of an anvil. In addition to bipolar debitage, the anvil was occasionally employed during other reduction sequences. Many of the cores reduced by orthogonal debitage methods bear signs of anvil impact such as crushing and wedge-shaped crack lines on the distal end of the debitage surface (Table 5.6). These cores were not identified as bipolar, since in terms of core management and organization of the core volume they are similar to orthogonal cores that do not bear signs of anvil impact. The anvil was probably used as a support, because of the small size and spherical shape of the pebbles.
None of the cores in the Bizat Ruhama assemblage has a cortical striking platform. It is likely that the rounded shape of the pebbles required the formation of a flat surface before initiation of the knapping sequence. This was achieved either by the removal of a cortical flake or by splitting the pebble into two halves. Some of the cores still preserve the bulb from the initial splitting of the pebble. The raw materials at Bizat Ruhama were reduced intensively and aggressively, with an emphasis on maximizing the exploitation of the raw material. Cores were rarely abandoned before they were totally exhausted or the debitage surfaces were destroyed. Although the pebbles used for knapping were small, most of the cores have only a small amount of cortex and a large number of scars in proportion to their size (Figure 5.5; Table 5.3). Many cores were not abandoned until the angle between the surfaces reached 90º or more,
DETACHED PIECES Detached pieces (DPs) include all artifacts that were detached from cores and pebbles and exhibit identifiable ventral faces. DPs at Bizat Ruhama are severely fragmented.
Table 5.5. Incipient cones on core surfaces. 48
Chapter 5: Flake production
Table 5.6. Signs of anvil impact on the distal ends of the debitage surface of different core types.
The DP assemblage includes complete flakes and broken flakes that are further subdivided according to the break type into four groups (Table 5.7). In addition, a category of angular fragments was distinguished. Angular fragments include knapping waste and DP fragments that show no orientation. Core fragments are excluded from the angular fragment group.
The scarcity of complete flakes makes technological and metrical studies of the assemblages difficult, since only a few of the technological observations related to core reduction can be made on broken pieces. Observations of value for technological reconstruction of the core reduction sequence, such as direction of scars on the dorsal face, number of scars, amount of cortex, shape of the butt and bulb, and metrical attributes cannot be made on broken flakes. For that reason, the attribute analysis presented here is as a rule based solely on the complete flakes. For some observations, proximal and siret breaks were used as well.
The number of complete flakes is low in all of the studied assemblages (Table 5.7). The differences in composition of the DP assemblages from BRAT5 and BR1996 are statistically insignificant (Table 5.7). The most important difference is in frequencies of complete flakes. The differences in composition of raw material types are insignificant for Eocene and brecciated chert in BRAT5 and BR1996 (Table 5.8).
Nevertheless, broken DPs do provide valuable insights into the type of knapping techniques used and into the methods of post-detachment treatment. On the assemblage level, the frequencies of broken flakes and types of breaks often serve
Table 5.7. Breakdown of detached piece assemblages. The X2 test indicates that the differences between the BRAT5 and BR1996 assemblages are insignificant (X2 = 12.166; df = 5; p = 0.33). 49
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Table 5.8. Raw material frequencies in BR1996 and BRAT5. The X2 test indicates that the differences between brecciated and Eocene chert are insignificant (X2 = 9/251; df = 5; p = 0.99).
as indicators of the knapping technique used, or even as a means of reconstruction of the site’s function (Ahler 1989, Amick and Mauldin 1997, Andrefsky 2005, Diez-Martín et al. 2009, Jeske and Lurie 1993, Kuijt et al. 1995, Sullivan and Rosen 1985).
differences between width values Z = 1.991; p = 0.001). These differences are especially noticeable among flakes larger than 30 mm. This group includes 29 brecciated chert flakes and only eight Eocene chert flakes. A similar tendency was recorded during the study of the core assemblage, and it clearly reflects the initial size of the raw materials. Since the mean length of Eocene chert pebbles in the Bizat Ruhama area is 39 mm and the largest Eocene chert pebble found during the raw material survey is 90 mm long, Eocene chert flakes are expected to be very small. The size dichotomy of the artifacts and its clear correlation with the availability of raw materials is significant in light of previously suggested cultural interpretations of the Bizat Ruhama industry and will be further discussed below.
Complete flakes According to the knapping strategies reconstructed on the basis of the analysis of FPs, the DP assemblage is expected to be dominated by flakes with unipolar or unipolar and side scar patterns on the dorsal face, plain butts, abundant signs of bipolar reduction and significantly larger sizes of brecciated chert flakes in comparison to other raw material types. As shown during the study of the FP assemblages, most of the cores that still retain technological information resulted from the knapping of brecciated chert pebbles. In contrast, however, Eocene chert flakes constitute the largest group in DP assemblages (Table 5.8). This discrepancy is probably connected to the size of Eocene chert pebbles and the intensity of their reduction, which resulted in small numbers of formal core on the one hand, and large numbers of detached flakes on the other.
Thickness and width/thickness ratio The thickness of flakes often furnishes more information about technological behavior than the length. While the length is constrained by the size of the raw material, the thickness is likely to mirror the technological and motor skills of the knappers and plays an important role in the economic and functional aspects of the industry. Thin flakes usually provide sharper cutting edges and, although the length of the cutting edge is also important, ethnographic examples indicate that the selection of flakes for use is often based on the morphology and angle of the edge rather than its length (Hayden 1977, White 1968, White et al. 1977). At Bizat Ruhama complete flakes are thick, especially with respect to their width (Table 5.9). Brecciated chert flakes are slightly thinner than Eocene chert flakes. In general, the relative thickness values of Bizat Ruhama flakes are in the range of those in the ‘Ubeidiya assemblages (Bar-Yosef and Goren-Inbar 1993).
Length and width Complete flakes in the Bizat Ruhama assemblages are small. The distribution of length values is unimodal, with the vast majority of flakes being 15–25 mm long. Brecciated chert flakes are significantly longer and wider than Eocene and translucent chert flakes (Table 5.9; the Kolmogorov-Smirnov test shows significant differences between length values of Eocene and brecciated chert: Z = 2.04; p = 0.000, significant 50
Chapter 5: Flake production
N Brecciated chert Mean Std. D. N Mean Eocene chert Complete Std. D. N flake Translucent Mean chert Std. D. N Mean Total Std. D. N Proximal Mean fragment Std. D. N Other Mean fragment Std. D. N Angular Mean fragment Std. D. N Flaked flake Mean Std. D. N Anvil flake Mean Std. D. N Pointed piece Mean Std. D. N Scaled trimming Mean Std. D. N Modified Clactonian notch Mean flake Std. D. Step-like/ N Mean irregular Std. D. trimming N Total Mean Std. D. N SGF Mean Std. D.
ML 97 28.79 9.467 139 24.26 6.477 21 24.85 6.993 257 26.10 8.103 185 22.12 6.231 194 19.67 8.945 326 19.03 5.276 58 35.3 11.48 177 22.12 5.391 131 24.70 5.857 38 29.08 9.296 204 24.98 6.443 198 24.58 6.194 571 25.03 6.498 167 14.84 3.310
L 97 26.39 8.725 139 20.3 7.058 21 19.84 8.095 257 22.68 8.332
W 97 21.49 9.048 139 17.6 4.998 21 19.56 3.943 257 19.23 7.081
T 97 8.63 3.296 139 8.26 3.315 21 7.02 2.073 257 8.34 3.265
12 33.43 16.48
58 23.96 7.481
11 22.66 7.211 6 26.74 14.996 26 22.84 6.617 26 23.41 7.573 69 23.27 7.859 151 12.87 2.759
101 19.77 5.121 23 20.54 5.785 150 19.36 5.530 145 18.30 5.545 419 19.16 5.477 151 14.07 3.784
58 15.42 4.097 177 10.62 3.021 125 10.91 3.001 33 10.69 3.505 186 10.75 3.292 181 10.38 3.763 525 10.66 3.408 154 4.14 1.254
W/T 97 2.7 1.104 139 2.39 0.988 21 2.96 0.96 257 2.54 1.045
W/L
BL 97 12.13 5.882 139 11.83 5.188 21 9.42 7.156 257 257 1.13 11.87 0.402 5.473
BT 97 5.08 2.971 139 6.07 3.169 21 4.31 1.967 257 5.67 3.1
LS
WS
58 58 17.51 16.66 6.521 4.517
37 19.47 6.019 19 17.46 5.214 90 16.44 7.175 97 18.80 7.183 243 18.04 6.876 151 149 144 3.53 1.22 11.04 1.038 0.389 4.954
37 6.80 3.115 19 6.59 2.231 90 5.89 2.446 97 7.10 1.628 243 6.51 2.332 145 3.85 1.771
204 204 9.63 15.053 2.865 4.234
Table 5.9. Descriptive statistics of DP assemblages. ML = maximum length; L = length; W = width; T = thickness; W/T = width/thickness ratio; W/L = width/length ratio; BL = length of butt; BT = thickness of butt; LS = length of scars on flaked flakes and Clactonian notches; WS = width of scars on flaked flakes and Clactonian notches.
Butt
remnants of ridges caused by constant rotation and switching between debitage and striking surfaces in multidirectional cores. Figure 5.7c presents butt types of complete flakes and proximal fragments on brecciated and Eocene chert. Although both assemblages are clearly dominated by plain butts, there are some differences that according to the X2 test
Most of the cores at Bizat Ruhama have a flat striking platform prepared by a single removal. Consequently, it was anticipated that plain butts would dominate the DP assemblage. Some butts were expected to exhibit the 51
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
52
53
The X2 test indicates that the differences between brecciated and Eocene chert are significant in scar pattern (X2 = 11.503; df = 5; p = 0.04) and butt type (X2 = 11.203; df = 3; p = 0.01); insignificant in amount of cortex (X2 = 10.503; df = 5; p = 0.06), bulb shape (X2 = 1.716; df = 4; p = 0.79) and distal end features (X2 = 6.457; df = 5; p = 0.26).
Figure 5.7. Technological and morphological characteristics of flakes in Bizat Ruhama assemblages: a. Pattern of scars on dorsal faces of complete flakes on brecciated chert and Eocene chert; b. Amount of cortex on dorsal faces of complete flakes on brecciated chert and Eocene chert; c. Butt types of complete flakes and proximal fragments on brecciated chert and Eocene chert; d. Bulb shapes of complete and proximal flakes; e. Distal end features of complete flakes on brecciated chert and Eocene chert. “Combinations of 1/2/3” refers to different combinations of scars on ventral face, crushing and crack lines; f. Lateral edge curvature of complete flakes on brecciated chert and Eocene chert
Chapter 5: Flake production
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
are statistically significant (X2 = 11.203; df = 3; p = 0.01). The major difference is the higher number of pieces with cortical butts among the brecciated chert flakes (14.5%) in contrast to only a few among the Eocene chert flakes (2.6%). Given that brecciated chert flakes with cortical butts also have a high amount of cortex on their dorsal faces, they clearly represent initial stages of core reduction. Only a few butts have been identified as dihedral. It is likely that the ridges on the dihedral butt surfaces are not part of the intentional preparation but rather residues of core swapping during multidirectional knapping. Shattered butts show evidence of crushing caused by a strong hammer blow and, as shown by the knapping experiments, they were probably produced by the bipolar technique (see below). The differences in the thickness and width of the butts on Eocene and brecciated chert flakes are negligible (Table 5.9).
notable. The complete Eocene chert flakes attest to intensive reduction. Despite the small size of the pebbles, most Eocene chert flakes preserve only a little cortex on the dorsal face. First flakes (entame) were not found among Eocene chert flakes and only a very few of flakes have more than 75% cortex. Ventral face features The shape and flatness of the bulb of percussion and the prominence of compression rings on the ventral surfaces vary according to the knapping technique involved (Andrefski 2005, Cotterell and Kamminga 1987, Inizan et al. 1999). Flat, diffuse or sheared bulbs of percussion, bulbs of percussion with ridges and flat ventral surfaces are often associated with the use of the bipolar technique (Ahler 1989, Barham 1987, Diez-Martín et al. 2009, Kobayashi 1975). Conversely, some of the experimental studies indicate that, along with the aforementioned characteristics, many of the bipolar flakes exhibit the same characteristics as flakes produced by the free-hand hard hammer, i.e. prominent bulbs of percussion, compression rings and similar breakage patterns (Amick and Mauldin 1997, Jeske and Lurie 1993, Kuijt et al. 1995).
Dorsal face features Figure 5.7a, b presents the information recorded on the dorsal faces of complete flakes in the Bizat Ruhama assemblages. The differences between brecciated and Eocene chert are prominent only for the pattern of scars on the dorsal face (Figure 5.7a): the unipolar scar pattern is more frequent among brecciated chert flakes, while the multidirectional pattern is more abundant among Eocene chert flakes. A similar but far stronger tendency was recorded for the cores as well (see above, Figure 5.6). Flakes with multidirectional scar patterns were probably produced by the multidirectional debitage method. Flakes with unipolar scar patterns could be produced either by unidirectional orthogonal or by preferential surface debitage methods. It is difficult to differentiate between two methods, especially given the small size of the assemblage of complete flakes. Flakes struck from preferential surface cores are expected to be flat, to have relatively acute angles between the dorsal surface and the butt (angle de chasse) and to display evidence of more thorough butt preparation. Only two flakes in the Bizat Ruhama assemblage show such characteristics. Some of the flakes with unipolar scar patterns exhibit the shape of naturally backed knifes (Bordes 1961), with cortex preserved on the left or right back of the flake (Plate 11:12– 14). These flakes were probably struck from unidirectional orthogonal cores. Flakes displaying other scar patterns on the dorsal surface are few, making it even more difficult to assign them to debitage systems identified in the core assemblage.
Among complete flakes in the Bizat Ruhama assemblage, 80% exhibit compression rings that are more or less pronounced. Most of the proximal and other fragments also have compression rings. The only DPs that exhibit flat surfaces are angular fragments and some longitudinal fragments. Most of the complete flakes have prominent bulbs (Figure 5.7d). Flat bulbs are second in frequency. Bulbs with ridges and crushed bulbs are also relatively abundant among both chert types. These aspects will be discussed further below in connection with the experimental studies. Distal end features: signs of bipolar technique The distal end is generally defined as the extremity of the flake that is opposite the butt, i.e. opposite the point of applied force. Given that during bipolar reduction force is often applied from two opposite sides, both proximal and distal ends of the flake may exhibit signs of impact. The identification of impact marks on the distal end of the flakes is based on the results of the knapping experiments presented in Chapter 7. The impact marks include small flat scars on the distal end of the ventral surface (Plates 11:2, 4, 5, 7, 12, 13; 12:1, 3, 5; 13:2–4, 6; 14:18; 15:6), small wedge-shaped crack lines developing from the distal end along the axis of the flake (Plates 11:3, 6, 9–11; 12:2, 4; 13:5; 14:10, 17; 15:1, 3–5, 7, 8), opposite bulbs of percussion (Plates 11:8; 13:1; 14:14) and marks of violent crushing (Plates 11:1, 14; 13:2, 6; 14:15, 16; 15:2, 9). According to the knapping experiments, all four of these features are diagnostic for the bipolar knapping technique (see below). More than 40% of the complete flakes in the Bizat Ruhama assemblages exhibit
Dorsal face characteristics point to a slightly lower intensity of reduction of the brecciated chert. Brecciated chert flakes more frequently have a cortical scar pattern (14.5% against 5.1% of complete Eocene chert flakes). They also exhibit a greater amount of cortex (Figure 5.7b) and a lower number of scars (brecciated: mean 2.82, std. d. 1.657; Eocene: mean 3.09, std. d. 1.726). Given the significantly larger size of the brecciated chert flakes, the difference is even more 54
Chapter 5: Flake production
signs of impact on their distal ends. Crack lines, scars on the ventral face and crushing marks occur in similar frequencies in both the Eocene and brecciated chert flake assemblages (Figure 5.7e).
the definition was later extended to include flake fragments and shatter from bipolar reduction (Diez-Martín et al. 2009, Hovers 2003, Isaac and Harris 1997, Semaw 2006). At Bizat Ruhama only fragments of DPs were included in this category, while fragments of cores were included in the group of exhausted cores.
Lateral edge curvature The curvature of the lateral edges was recorded in an attempt to discern repetitive patterns in the form of the flakes. The shape is in general unstandardized, with most of the flakes exhibiting irregular lateral edges. However, a group of flakes, especially in the brecciated chert assemblage, exhibits a more standardized pattern with two parallel edges (Figure 5.7f). Most of these flakes with parallel edges exhibit a unidirectional pattern of scars, indicating that they were produced by unidirectional orthogonal or preferential surface debitage methods.
The category includes fragments of DPs whose flaking axis cannot be defined. Angular fragments have no butt or bulb of percussion, but usually exhibit one flat (ventral?) surface without apparent orientation. The opposite surface often preserves scars of previous removals. Angular fragments are among the smallest in the DP assemblages (Table 5.9).
Detached pieces: an overview The DPs of the Bizat Ruhama assemblages are divided into seven groups according to the type and characteristics of the breakage. The Bizat Ruhama assemblages contain only 257 complete flakes. The breakage and the secondary knapping of the flakes hinders the reading of the scar pattern on the dorsal surface and the identification of butt and bulb features, and masks the methods by which the flakes were produced. Nonetheless, it is clear that the flakes of Bizat Ruhama are part of the same debitage system as the cores. The flakes show unidirectional or multidirectional scar patterns and ample evidence for use of the bipolar technique. No influx of flakes produced by debitage methods other than those defined in the core assemblages was discerned during the study. DP assemblages display a high intensity of reduction. They are very small and exhibit a low amount of cortex and a high number of scars. DPs from different excavated areas show a close similarity in technology, morphology, size and raw materials used, indicating that they all belong to the same industry.
Siret accidental breaks Siret accidental breaks (Inizan et al. 1999) are knapping accidents in which flakes are snapped into two along their knapping axis. Siret breaks occur when two perpendicular flaking planes develop from the point of impact and they often preserve one half of the bulb of percussion. Although siret accidental breaks are commonly associated with the hard hammer technique (Inizan et al. 1999), during the experimental study (see below) they were often produced by the bipolar technique as well. At Bizat Ruhama, siret accidents are more common in translucent and brecciated chert flakes (Table 5.8).
Longitudinal breaks Longitudinal breaks are knapping accidents in which flakes are broken along their flaking axis. They are produced by a similar mechanism to that resulting in siret accidental breaks, but do not preserve remains of the bulb of percussion or the butt. In the experimental assemblages (see below) they are more frequent in bipolar technique assemblages. Longitudinal breaks occur in all excavated assemblages and are more frequent among brecciated chert DPs (Tables 5.7, 5.8).
The assemblages show clear-cut differences between Eocene and brecciated chert flakes. Brecciated chert flakes are significantly larger and most show unipolar scar patterns. They also preserve more cortex and have fewer scars on the dorsal surface. Eocene chert flakes are smaller, more frequently have a multidirectional pattern of scars, preserve less cortex and have a higher number of scars on the dorsal surface. A similar dichotomy was recorded in FP assemblages as well.
Angular fragments
Many of the complete flakes exhibit possible traits of the bipolar technique. These traits include the direct evidence of hammer or anvil impact on the distal end as well as indirect evidence like ventral surface shape (i.e. the shape and size of the bulb of percussion), butt type and flake breakage pattern. To determine whether there is a correlation between possible indicators of the bipolar technique, the shape of the bulb of percussion and the type of the butt were
Angular fragments are the largest group of DPs at Bizat Ruhama, forming 12–15% of the DP assemblages (Table 5.7). Only in BRAT5 are angular fragments slightly less frequent than complete flakes. Angular fragments were originally interpreted as core fragments or exhausted coretools (Bar-Yosef and Goren-Inbar 1993, Leakey 1971), but 55
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
bulb are insignificant; X2 = 2.266; df = 1; p = 0.13). Thus, neither prominent nor flat bulb of percussion seems to be diagnostic of the type of technique used. Among the butt types, shattered butts show the best correlation with signs of distal impact (Table 5.11), while most flakes with a dihedral butt show no signs of impact on their distal end.
intersected with signs of distal impact. Among the bulbs of percussion, only the crushed bulb and to a lesser extent the flat bulb with ridges exhibit good correlation with distal impact marks (Table 5.10). Conversely, 65% of flakes with prominent bulb and 53% with flat bulb do not display signs of distal impact (differences between prominent and flat
Table 5.10. Crosstabulation of bulb type and signs of impact on the distal end of complete flakes in the BRAT5 and BR1996 assemblages.
Table 5.11. Crosstabulation of butt type and signs of impact on the distal end of complete flakes in the BRAT5 and BR1996 assemblages.
56
CHAPTER 6: SECONDARY KNAPPED FLAKES The lithic assemblages of Bizat Ruhama show marked evidence for post-detachment treatment of DPs. As shown above, complete flakes form only 10% of the Bizat Ruhama assemblage. Although many flakes were broken during core reduction, it is clear that a substantial number of flakes was broken or modified after their detachment from the cores. These flakes have been grouped under the generic name of “secondary knapped flakes”. Secondary knapped flakes constitute about half of the DP assemblage. They are further subdivided into four groups identified on the basis of their knapping methods and techniques and the morphology of the resulting edges.
more appropriate to describe the knapped flakes at Bizat Ruhama than the more frequently used “cores-on-flakes”, which is is often applied to the more elaborate production of cores on flakes in Levantine late Acheulian and Mousterian contexts (Barkai et al. 2010, Chazan and Kolska Horwitz 2007, Goren-Inbar 1988, 1990, Hovers 2007, 2009b, Munday 1976, Solecki and Solecki 1970). At Bizat Ruhama, on the other hand, flaked flakes show no preparation but were rather opportunistically used for one or several removals. Most of the flaked flakes were found in BR1996 (Table 6.1). About half of the assemblage is made on brecciated chert (Table 6.2). Flakes selected for further knapping exhibit similar technological characteristics to the rest of the flake assemblage. They were made by either the unipolar or the multidirectional method of reduction. Six of the selected flakes exhibit crushing and small removals on the distal end of the ventral surface, indicating that they were removed from bipolar cores. Flaked flakes are significantly larger than complete flakes (Table 5.9; Kolmogorov-Smirnov test for length values: Z = 3.603, p = 0.000). Many of the largest flakes at Bizat Ruhama were further flaked and the smallest flakes that were knapped are around 20 mm in maximum length (Figure 6.1). In addition, flaked flakes are significantly
FLAKED FLAKES The term “flaked flakes” was used by Ashton (1992, 2007, Ashton et al. 1991) in the British Lower Paleolithic context to describe flakes that were further knapped. “Flaked flakes … consist of flakes that have had further flakes removed from lateral, proximal or distal edges and from both the ventral and dorsal. There are characteristically between one and four removals on a single piece, but sometimes several more” (Ashton 2007: 1). The term “flaked flakes” seems
Table 6.1. The secondary knapped flake assemblages. 57
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Table 6.2. Breakdown of secondary knapped flake assemblages by raw material type.
products detached from the flaked flakes were thin sharp flakes, generally shorter than 20 mm and frequently wider than they were long (Table 5.9). These flakes occur among small, thin debitage in BRAT5 and in small numbers in other assemblages.
ANVIL FLAKES Broken flakes in the Bizat Ruhama assemblages often show marks of impact on their dorsal faces. In most cases these are accompanied by marks of opposite impacts at the intersection of the ventral and broken surface/s or ventral and lateral surfaces of the flake. In the literature similar marks have been linked to flake breakage with the support of an anvil (Bergman et al. 1987, Corvetto et al. 1994, Longo et al. 1997, Owen 1982, Peretto 1994). At Bizat Ruhama 177 flakes with such marks were identified, 117 of which derive from BRAT5 and BR1996 (Table 6.1). Anvil flakes constitute 5–8% of the lithic assemblages from the different excavated areas. About half of them are made on Eocene chert (Table 6.2). As in the case of flaked flakes, the flakes selected for anvil knapping were produced by the same methods as the rest of the flake assemblage. Nevertheless, it seems that the largest and thickest pieces were chosen (Figure 6.1; Table 5.9); the anvil flakes show maximum length values close to those of complete flakes and initially were probably among the largest in the assemblage. Anvil flakes are also significantly thicker than complete flakes (Table 5.9; t = -7.155; df = 398; p = 0.00).
Figure 6.1. Length scatterplot of complete flakes, flaked flakes, anvil flakes and modified flakes. thicker than unmodified flakes (Figure 6.1; Table 5.9; Z = 5.154, p = 0.000). The flaked flakes were not knapped by any consistent method, and no rules governing the placement of the removals on the flake edges and surfaces were observed. The scars occur randomly on the lateral, distal or proximal edges and on the dorsal or ventral surfaces of the flakes (Table 6.3; Plates 8:2– 5; 16:1–8). Usually 1–5 thin flakes were removed (Table 5.9). Only two flaked flakes show a sequence of two parallel removals struck from the same direction. The scars display large and pronounced negatives of the bulb of percussion and show no evidence for anvil support. The blows were usually placed close to the edge of the flake, resulting in shallow scars and indicating an advanced level of skill and precision.
The flakes were usually broken at the point of maximum thickness. The impact marks are evident on the dorsal ridges, at the point of intersection between the dorsal and broken/ lateral surfaces. The most frequent impact marks on the dorsal surfaces are points of percussion, cones and negatives of the cones of percussion (Plate 17:5-7) and crushing on the dorsal edges (Table 6.4; Plates 17:2, 3; 18:1, 3, 6, 8, 9). Often either the cone of percussion or its negative occurs in combination with crushing. Other features are much less
According to the shape of the scars, the characteristic 58
Chapter 6: Secondary knapped flakes
Table 6.3. Crosstabulation of raw material type and exploited face of flaked flakes. and cones of percussion are exceptionally rare. Instead, small isolated scars or marks of crushing on the broken/ lateral surface of the flakes are the most frequent features. The morphology of the scars is presented in Table 6.6. Among impact marks directed from the ventral surface, 68% are step-like fractures or crushing on the edges of the artifacts (Plates 17:2, 4; 18:1, 3, 5, 6). Step-like scars are always wider than they are long, terminate in a step fracture and do not exhibit a clear negative of a cone of percussion. They are common when the angle between the ventral and broken/lateral surfaces is steep. The other marks of impact on the ventral faces include incipient cones of percussion, which often occur in combination with the scars on broken/ lateral surfaces, and wedge-shaped crack lines (Plate 18:3, 5). Both occur on the dorsal face as well, but are more frequent on the ventral surface.
Table 6.4. Dorsal surface features of anvil flakes: types of impact marks.
In most cases impact marks on the ventral surface occur together with dorsal features. There are a number of possible combinations (Table 6.7). When a clear point of percussion or a negative cone of percussion occurs on the dorsal surface, it is usually associated with opposing scars on the broken/ lateral surfaces. When the impact marks on the dorsal face are more violent (e.g. dorsal face crushing or combination of crushing with point and cone of percussion), they are accompanied by more pronounced marks on the ventral surface (e.g. wedge-shaped crack lines, incipient cones of percussion or both).
Table 6.5. Features identified at the intersection between ventral and broken/lateral surfaces of anvil flakes.
common. Wedge-shaped crack lines that develop from the point of impact toward the center of the broken surface generally occur in combination with other features (Plates 17:1; 18:4, 5, 8). Some marks of anvil breakage were identified on the ventral surfaces of the flakes as well. These are distinctly different from marks occurring on the dorsal face (Table 6.5). Points
Table 6.6. Shape of scars at the intersection between the ventral and broken/lateral surfaces of anvil flakes. 59
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Table 6.7. Crosstabulation of dorsal and ventral surface features of anvil flakes. The characteristics of the impact on different flake surfaces indicate that the flakes were likely placed with their ventral surface resting on an anvil. The impact marks at the intersection of the ventral and lateral surfaces are not localized but occur at different spots, pointing to a large area of contact with the anvil. The impact marks on the dorsal faces are localized and pronounced, indicating a small area of contact with a hard object, as would be expected in the case of a hammerstone blow (see Chapter 7). The broken surfaces of anvil flakes are either flat or exhibit compression rings (Table 6.8). In addition, they may display marks of crushing and shattering.
knapped artifacts seems to be quite compelling, a group of broken flakes was classified as “possible anvil flakes” (Table 6.1). Like anvil flakes, these flakes were broken at the point of maximum thickness and exhibit marks of crushing or step-like fractures developing from the ventral surface. Unlike anvil flakes, they do not bear impact marks on dorsal surfaces and therefore in most cases cannot be unambiguously linked to the breakage of the flake.
MODIFIED FLAKES The third group of secondary knapped flakes is composed of artifacts with scars of varying shape, size and regularity. The scars are similar to those found on the edges of anvil flakes. The difference is that they occur in a sequence rather than being isolated, as in the case of anvil flakes. Conventionally, many of the artifacts from this group would be assigned to the retouched tools category. At Bizat Ruhama, however, their assignment to this category is uncertain because of the roughness and low degree of standardization of the scars, and because the bipolar technique and anvil-supported knapping of flakes, which were intensively used at the site during different stages of the reduction sequence, often result in similar scars on the flake edges (Bergman et al. 1987, Corvetto et al. 1994, Longo et al. 1997, Peretto 1994).
In addition to anvil flakes, whose identification as anvil-
Table 6.8. Broken surface features of anvil flakes. 60
Chapter 6: Secondary knapped flakes
Edge angle
Therefore, like the scars on the anvil flakes, they may be not the result of intentional retouch aimed at the production of specific tool forms but rather a by-product of bipolar knapping of the flakes. This chapter aims at presenting techno-morphological features of these flakes as well as describing the characteristics of the scars on their edges. The aims, methods and techniques of their production will be discussed in Chapter 8, after the presentation of the results of the knapping experiments.
One of the most distinctive features of the modified flakes in the Bizat Ruhama assemblages is steep angle of the edges. Modified flakes were divided into four groups according to the edge angle (see Appendix 2 for description of categories). Flakes with flat scars (defined as lower than 45°) are very rare, constituting only 8% of the assemblage. Flakes with semiabrupt and abrupt modified edges dominate the assemblages at 81%. The remaining 11% show angles of 90° or more and were defined as cross-abrupt.
Morphology of scars
The steepness of the modified edge of the flakes is probably the outcome of the blank selection pattern and modification technique. Flakes selected for secondary modification are thick (Table 5.9), being significantly thicker than unmodified flakes (Kolmogorov-Smirnov Z = 5.193; p = 0.000). Thin, sharp flakes were rarely modified.
Modified flakes at Bizat Ruhama show scars of varying morphology and size. The scars were divided into three groups according to their morphology: conchoidal fracture scars, scars with step-like fractures and marginal scars. Conchoidal fracture scars exhibit negatives of the bulb of percussion and in some cases even ripples (Plates 17:1, 3; 18:7). This group also includes Clactonian notches, artifacts with one large conchoidal scar on the dorsal face that creates a concave edge (Bordes 1961) (Plates 17:3; 18:4, 9). The second group consists of step-fracture scars like those described for anvil flakes (Plates 17:4; 18:1, 2, 8). The third group is composed of small, short scars occurring on the margins of the flakes (Plates 19:2, 3, 11; 20:5; 21:17). Because of their small size, it was hard to assign these to either of the previous categories. Some of these scars could be merely edge damage or signs of use-wear.
The edge angle shows good correlation with the edge modification type. Cross-abrupt modification was not identified among artifacts with the scaled type of modification but is very common when the edge is formed by step-like fracture scars (Figure 6.2c). Conversely, flat scars occur only on artifacts modified with scaled or irregular scars and are not found on artifacts with step-like fracture scars. Edge angle thus clearly affects the morphology of the scars. The steeper the edge, the higher the chance that the removal will terminate as a hinge and that scars with step-like terminations will be produced.
Edge modification type
Number and position of modified edges
The three scar types identified above are found in the Bizat Ruhama assemblages in different combinations on the edges of modified flakes (Figure 6.2a). Six types of edge modification were identified. The categories are described in Appendix 2. In most cases the modified edges are rough and unstandardized. Only a small portion of the assemblage exhibits regular scaled retouch formed by a number of sequential conchoidal scars (Plates 19:5, 9, 10, 14; 20:17, 22; 21:11, 15, 18). The Clactonian notch is the most common type of edge modification (Plates 17:3; 18:4, 9; 20:7, 18, 19; 21:6, 12). Clactonian notches sometimes occur in combination with other types of removals – conchoidal, step-like or marginal (Plates 20:1, 18, 19; 21:4, 9, 14, 16). In some cases two Clactonian notches were identified on opposite edges of an artifact (Plates 19:8, 15; 20:16; 21:16).
The modified flake assemblage includes a large group of artifacts with two or more modified edges (Figure 6.2b). In 16% of the cases two edges converge or create a pointed extremity. In an additional 14%, scars occur either on two lateral edges or on lateral and distal edges. Among the artifacts with one retouched edge, the right edge was modified more frequently than others.
The assemblages Modified flakes can into four techno-morphological groups (Table 6.9). Except for the pointed pieces, other groups in the assemblages were defined by the type of edge modification. The scaled modification group contains artifacts that exhibit one edge with a number of sequential conchoidal scars. This group includes the most regular and intensively worked pieces in the modified flake assemblage. The modified edge is usually wavy, convex or denticulate in shape (Plates 19:5, 9, 10, 14; 21:18). No artifacts with two edges shaped by scaled retouch were identified, apart from several pieces
The largest group in the assemblage exhibits a sequence of irregular scars on one or several edges formed by combinations of conchoidal, step-like and marginal scars (Plates 19:6; 20:1, 3–6, 9, 12, 15, 23; 21:10).
61
Figure 6.2. Characteristics of edges of modified flakes. a) Edge modification type; b) Location of modified edges; c) Angle of edge by modification type; d) Edge modification type of pointed pieces.
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
62
Chapter 6: Secondary knapped flakes
Table 6.9. Breakdown of modified flake assemblages by techno-morphological type. The X2 test indicates that the differences between the BRAT5 and BR1996 assemblages are insignificant (X2 = 6.892; df = 3; p = 0.075). with two convergent edges that were included in the pointed item group.
of different shapes with various degrees of standardization. The most standardized are reminiscent of awls, becs shaped by two opposite Clactonian notches or Tayacian points (Plates 19:2, 7; 20:2, 8, 10, 11, 13, 14, 16, 20, 21, 24; 21:1, 2, 4, 7–9, 13, 14, 16), while the majority are unstandardized pieces with thick points and abrupt/cross-abrupt irregular edges shaped by scars of various forms.
The Clactonian notch group is composed of flakes with a single Clactonian notch as well as pieces with a Clactonian notch and step-like or irregular scars on other edges. Clactonian notches are characterized by wide, short scars (Table 5.9). They also frequently occur on the edges of pointed pieces (Plates 20:16; 21:4, 9, 13, 14, 16). The third techno-morphological group is composed of flakes with step-like scars, which often occur in sequence with conchoidal fracture scars (in the latter case identified as irregular modification; Plates 20: 3–6, 9, 12, 15, 23; 21:10). Very few pieces exhibit a sequence of more than 3–4 scars of similar shape, depth and invasiveness.
Raw materials occur in significantly different frequencies among different types of modified flakes (Table 6.10). Flakes with step-like scars are more numerous among Eocene and translucent chert artifacts. Clactonian notches are considerably more frequent among modified flakes on brecciated chert, but are commonly made on the other raw materials as well. Most of the pieces with scaled modification are made on Eocene chert.
The group of pointed pieces includes artifacts with two convergent edges shaped by combinations of different edge modification types (Figure 6.2d). The group contains pieces
The types of DPs selected for modification are presented in Table 6.11. The data show considerable differences in
Table 6.10. Breakdown of modified flake assemblages by raw material type in BRAT5 and BR1996. The X2 test indicates that the differences between the three chert types are significant (X2 = 13.675; df = 6; p = 0.03). 63
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Table 6.11. Breakdown of modified flake assemblages by DP category.
composition of the modified flake and DP assemblages. The absence of longitudinal and Siret fragments and low frequencies of angular fragments among modified flakes indicate that selection of DPs for modification followed consistent rules. Frequently, the flakes that were selected for modification preserve butts. The number of identified proximal fragments is especially high (Table 6.11). The abundance of proximal fragments is probably a result of removal of the distal ends of complete flakes during the post-detachment modification. Thus, the post-detachment treatment is likely one of the reasons for the low number of complete flakes at the site.
modification show slightly higher length and significantly higher thickness values than the complete flakes in the assemblages. Marks of dorsal impact similar to those found on the anvil flakes occur on 20% of the modified flakes (Table 6.13). The impact marks occur on all four techno-morphological types. In some cases the impact marks on the dorsal face occur precisely opposite the negative of the removal that created a Clactonian notch or close to the pointed extremity of a pointed piece (Plates 17:1-4; 18:1, 2, 4, 6, 9; 19:2, 5, 7, 8; 20:1, 3, 8, 9, 20, 23; 21:4, 5, 6, 8, 9, 13). This phenomenon is clearly significant for understanding of the manufacturing techniques and the aims of post-detachment treatment of the flakes. As noted above, similar signs were found on anvil flakes and were interpreted as hammerstone impacts that occurred during bipolar reduction.
Metrical characteristics of modified flakes are presented in Figure 6.1 and Tables 5.9, 6.12. In general, modified flakes are homogeneous in size. Overall, the flakes selected for
Overall, the modified flake assemblage is heterogeneous in the shape of the scars and the types of modified edges. With the exception of the flakes with scaled modification, the groups are very similar in most of the recorded characteristics. Flakes with scaled scars, apart from having the most standardized modified edges, are slightly different in length, blank selection and frequency of signs of bipolar technique.
SECOND-GENERATION FLAKES Secondary knapped flakes constitute a substantial part of the industry and are expected to have produced a large amount of second-generation flakes (SGFs) and waste. Given that these flakes were produced in the second stage of the reduction sequence, they are categorized as second-generation flakes (Plates 22, 23). The identification of SGFs is based on the morpho-technological features of the flakes studied in the archaeological assemblages, on the scars on the secondary
Table 6.12. One-way ANOVA test for different technomorphological types of modified flake assemblages. 64
Chapter 6: Secondary knapped flakes
SGFs clearly differ from complete flakes in length, thickness, width/thickness and width/ length ratios, shape, size and shape of butt, type of bulb and breakage patterns. Figures 6.3–6.4 and Table 5.9 present the metrical characteristics of SGFs and complete flakes. The data show that the two assemblages differ significantly in all metrical characteristics except for the length of the butt (Table 6.14). The length histogram of the combined complete flake and SGF assemblage in Figure 6.6 shows two major peaks of distribution, around 9–16 mm and 18–26 mm. These two peaks correspond respectively with SGFs, which never exceed 20 cm in length, and complete flakes (Figure 6.4a). SGFs are very Table 6.13. Impact marks on dorsal ridges of modified flakes. thin, especially in comparison to complete flakes (Figures 6.4b, d). One of the typical features of knapped flakes from which they were apparently produced, SGFs, which is clearly associated with their thinness, is the and on comparison to the flakes replicated during the sharp edge around most of their circumference. The edge experimental knapping. Thus, small, thin, short and wide angles are usually less than 55º (mean 37.37, std. d. 7.046). flakes with large, often wing-shaped butts were included in SGFs also differ significantly from complete flakes in width/ SGFs’ group. Still, the borderline between flakes produced length ratio (Figure 6.4c). during core reduction and secondary flake reduction is not clear-cut, and it is possible that some flakes categorized as Most of the SGFs are complete (87%) and exhibit a SGFs were manufactured during core reduction and vice pronounced bulb of percussion (79%). The bulb of percussion versa. occupies a large part of the ventral surface (Plate 24b).
Figure 6.3. Combined length histogram of complete flakes and SGFs in Bizat Ruhama assemblages. 65
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
In contrast to the flakes produced during core reduction, which have irregular or parallel edge curvature, the SGFs display rounded (65%) or divergent (26%) curvature of the lateral edges. The butts are large and have a wing-like or lens-like shape (Plates 22, 23, 24b). Overall, the SGFs are very consistent in most of their recorded characteristics and are the most standardized flake type in the Bizat Ruhama industry.
on the distal end of the ventral face opposite the bulb, indicating that many of them were produced during flake knapping on an anvil (Plates 22:2, 3, 7, 8, 9; 23:1; 25). SGFs were found in large numbers only in BRAT5. Because of their small size, they were rarely collected during the survey. They are scarce in BR1996, probably due to varying degree of post-depositional winnowing in different areas of the site (see Chapter 3).
Within the SGF assemblage, 11% exhibit marks of impact
Figure 6.4. SGFs and complete flakes in Bizat Ruhama assemblages. a) Length; b) Thickness; c) Width/length ratio; d) Thickness/width ratio.
Table 6.14. Kolmogorov-Smirnov test of significance for SGFs and complete flakes (including complete flakes from the flaked flake and modified flake assemblages).
66
CHAPTER 7: EXPERIMENTAL KNAPPING Knapping experiments were found to be essential for an understanding of the behavioral and post-depositional processes that resulted in the formation of the Bizat Ruhama archaeological record and of the technological behavior employed in the making of lithic artifacts. The composition and technological characteristics of the Bizat Ruhama assemblage have no parallel in the Levantine Paleolithic. For example, the use of an anvil for stone knapping has not been reported from any other Levantine Paleolithic site, and some of the techno-morphological groups of the lithic assemblage have not been identified in other Lower Paleolithic sites in the Levant and beyond. Consequently, an understanding of the technological behavior of the Bizat Ruhama hominins could not be achieved solely on the basis of existing knowledge and a lithic replication study was seen as essential. The main questions of the study were: 1.
Is the composition of the archaeological assemblages as expected if the knapping was conducted in situ? If not, is it indicative of post-depositional disturbance or hominin transportation of prepared artifacts from/to the site?
2.
Is the small number of small debris at the site as expected if the knapping took place on the spot or was the site disturbed by post-depositional winnowing?
3.
What are the flaking qualities of the raw materials used? Can the variation in the quality, size and shape of the raw materials explain the differences in exploitation of various chert types?
4.
It seems clear that the bipolar technique was used at the site along with the free-hand hard hammer technique. Is it possible to differentiate between the products of both techniques on the raw material used at Bizat Ruhama? Could marks identified as signs of the bipolar technique on the edges of cores and flakes have been produced by the free-hand hard hammer technique?
5.
Were the impact marks and scars identified on the edges of anvil flakes and modified flakes produced by anvil impact? If so, how and why were the flakes broken on
an anvil? Which additional products are expected to be produced during the breakage? 6.
Why were small and flat “beach-like” pebbles broken on an anvil? Which additional products are expected to be produced during the breakage?
The study included experimental knapping of pebbles by both the bipolar and the free-hand hard hammer techniques, anvil breakage of the flat “beach-like” pebbles and knapping/ breakage of the flakes on an anvil (see Appendix 3 for a detailed description of the experiments).
EXPERIMENTS IN FLAKE PRODUCTION The pebbles for the experimental knapping were collected during the raw material survey of Pleshet conglomerate exposures (Samples 2, 4, 5 and 7). In total, the experiments involved the knapping of 87 pebbles. Observations presented in Table 2.1 were applied during the knapping of 30 brecciated and 26 Eocene chert pebbles. Among these, 11 Eocene and 13 brecciated chert pebbles were knapped by the bipolar technique and the remaining 32 pebbles were knapped by the free-hand hard hammer technique. In total, four experimental assemblages were created and studied (Table 7.1). The cores were knapped by debitage methods identified during the study of the archaeological assemblage, namely the orthogonal method with a series of unidirectional removals, the multidirectional method and the unidirectional preferential surface method. The selection of the knapping technique followed the fracture specifics of each knapped pebble. For each pebble, fracture was first attempted by the free-hand technique. If it proved impossible to initiate the fracture, the pebble was then flaked with the support of an anvil. The experiments demonstrate a clear link between the shape and size of the pebble and its suitability for free-hand knapping. The free-hand hard hammer technique was found to be efficient only when the angle between the striking platform and debitage surface was maintained at 1cm) Sub-total
29 32 5 3 6 6 81 8 9 42 80 134 273 354
Eocene chert % within a group 35.8 39.5 6.2 3.7 6.9 7.9 100.0 2.9 3.3 15.4 29.3 49.1 100.0
% of total 8.2 9.0 1.4 0.8 1.1 2.3 22.9 2.3 2.5 11.9 22.6 37.9 77.1 100.0
N
28 18 9 0 3 3 61 4 4 66 58 120 248 313
Brecciated chert % % within a of total group 45.9 8.9 29.5 5.8 14.8 2.9 0.0 0.0 4.9 1.0 4.9 1.0 100.0 19.5 1.6 1.3 1.6 1.3 26.6 21.1 23.4 18.5 48.4 38.3 100.0 79.2 100.0
Table 7.7. Anvil-supported knapping of flakes: composition of the experimental assemblages. 77
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Broken flakes 1. Incipient cones 2. Broken surface scars and crushing 3. Isolated scars on lateral surface 4. Wedge-shaped crack lines 5. Point of percussion and crushing 6. Combinations of 1-4 7. No signs 8. Clactonian notches 9. Step-like fracture scars 10. Conchoidal scars 11. Combinations 8-10
2 13 1 5 6 22 58
1.90% 12.10% 0.90% 4.70% 5.60% 20.60% 54.20%
Clactonian notches
Flakes with retouch-like scars
14
Pointed pieces
1 9 3 6
2
Table 7.8. The experimental assemblage: signs of anvil impact.
Broken flakes 1. Point and/or cone of percussion 2. Negative cone of percussion 3. Incipient cones 4. Dorsal face crushing 5. Wedge-shaped crack lines 6. Combination of 1/2 and 4 7. Combination of 1/2+4+5 8. Combination of 4 and 5 9. No signs Total
Clactonian notches
13
12.1%
1
7.1%
19 0 7 7 2 59 107
17.8% 0.0% 6.5% 6.5% 1.9% 55.1% 100.0%
2
14.3% 0.0% 14.3% 0.0% 0.0% 64.3% 100.0%
2
9 14
Flakes with retouch-like scars 1 5.6% 1 5.6% 1 5.6% 4 22.2% 0.0% 0.0% 2 11.1% 1 5.6% 8 44.4% 18 100.0%
Pointed pieces 1
2 3
33.3%
16 1 1 25 0 9 9 3 66.7% 78 100.0% 142
Total 11.3% 0.7% 0.7% 17.6% 0.0% 6.3% 6.3% 2.1% 54.9% 100.0%
Table 7.9. The experimental assemblage: signs of hammerstone impact. the scars exhibit step-like terminations. An additional group of artifacts was identified as “possible anvil flakes”, following the terminology of the study of the archaeological assemblages. These are broken flakes that show no marks of hammerstone impact on their dorsal surfaces. Flakes that bear sequences of step-like scars (Table 7.7) are indistinguishable from the modified flakes in the archaeological assemblages that were identified as flakes with step-like modification. In many cases they could be mistakenly identified as bearing intentional abrupt retouch (Plates 29:3, 5, 10; 30:2, 8, 9; 31:2, 6, 7). The flakes with step-like scars are similar in general morphology and the angle between the surfaces to the anvil flakes (Figure 7.6). Clactonian notches produced during the experiment are indistinguishable from archaeological samples (Plates 28:1, 2; 29:1, 7, 9; 30:1). Only four pieces with a sequence of conchoidal scars resembling scaled retouch were produced during the experiment (Plates 29:6; 30:3–6).
Figure 7.6. Angle of the edge of different types of products of anvil knapping of flakes. 78
Chapter 7: Experimental knapping
Pointed pieces were produced unintentionally, without any attempt to control the form of the knapped flakes. Pointed forms were often created by a combination of “retouch” and Clactonian notch (Plate 28:2, 29:2, 3, 5, 6, 9, 10). Morphologically, pieces with conchoidal fracture scars (including Clactonian notches) differ from anvil flakes and flakes with step-like fracture scars; their proximal and distal edges are usually intact, and there are substantially lower angles between the ventral and lateral surfaces (Figure 7.6). Marks of hammer impact on the dorsal face are evident on 54% of the modified flakes and 35.7% of the Clactonian notches (Table 7.9; Plates 29:1, 5, 8, 10; 30:2, 7-10; 31:2, 4, 6, 7). These impact marks usually occur on the ridges at the contact between the dorsal and lateral planes of the flakes. The marks are similar to those identified on archaeological artifacts. Crushing is the most frequent mark (37.5%), followed by point and cone of percussion (25%) and combinations of crushing and cone of percussion (18.8%), crushing and wedge-shaped crack lines (12.5%) and wedgeshaped crack lines and cone of percussion (6.3%).
are long and thick, in many cases being the widest and the thickest part of the flake. The butts are large because the contact between the anvil and the flake was not confined to one spot, as is the case during hard hammer percussion, but occupied a larger area (Figure 7.4b). Another factor that seemingly facilitates the removal of flakes with thick butts is the acute angle between the ventral and lateral edges of the flakes (Figures 7.4, 7.5). The butts are often wing-shaped because of the small chips that were detached from the flake edge during the detachment of SGFs. The crack that formed the SGF developed parallel to the edge of the knapped flake and terminated at its dorsal surface (Figure 7.5b). Hence, SGFs are quite even in thickness and as a rule exhibit feather or axial terminations. Hinge terminations are very rare. The SGFs are thin and usually have sharp lateral and distal edges and one blunt edge formed by the butt. Among the SGFs, 26% exhibit contact features at the point of hammer impact opposite the butt. The impact marks include marks of percussion, wedge-shaped crack lines and crushing. The marks occur almost exclusively at the point of detachment of the SGF at the intersection of the lateral and distal surfaces. Incipient cones are virtually absent from the dorsal surfaces, but often occur on the ventral surfaces due to contact with the anvil. Over 74% of SGFs show no marks whatsoever of bipolar loading.
Second-generation flakes The experimental assemblages contain flakes that in the archaeological assemblages were identified as SGFs (Plates 24, 32). Second-generation flakes constitute 77.1 % of Eocene and 77.9% of brecciated chert experimental assemblages. The SGFs were detached by a Hertzian initiation generated by anvil impact. They show a prominent bulb of percussion, conchoidal fracture ripples, a distinguishable butt and crushing on the edge of the butt where it meets the dorsal face. SGFs differ from other DPs in their high standardization of form and size, and in the shape of the butt and the bulb of percussion. This standardization, however, does not signify preplanning or any special skills on the part of the knappers. The process of knapping itself is rather simple and not well controlled, and it is probably the shape, size and edge angle of the knapped flakes that make the final product look uniform. The SGFs are short (often wider than they are long) because their length is limited by the thickness of the knapped flakes (Table 7.10). The butts
Archaeological vs. experimental data Clactonian notches, irregularly “retouched” flakes, pointed pieces and broken flakes in the archaeological and experimental assemblages are similar in general morphology and in the shape of the scars along their edges (Tables 6.5, 6.6, 7.8; Plates 18-30). The main difference between the archaeological and experimental assemblages is in the higher frequencies of broken flakes in the latter (Tables 6.2, 7.7). There are two possible explanations for this difference. Firstly, the knapping experiments produced broken flakes without signs of dorsal impact (Table 7.9). In archaeological sites similar breakage can result from knapping accidents,
Table 7.10. Descriptive statistics of experimentally produced SGFs. 79
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
use or depositional processes, and consequently in the archaeological assemblages only flakes showing signs of dorsal impact were classified as secondary knapped flakes. Secondly, during the experiments most of the flakes were knapped until they broke, while a shorter knapping sequence would have resulted in substantially higher frequencies of Clactonian notches and flakes with “retouch-like” scars.
Second-generation flakes that were produced during the experiments are similar in size and general morphology, as well as in the shape of the butt, to second-generation flakes that were identified in the archaeological assemblages (Tables 5.9, 7.10; Plates 27-30). These flakes occurred in high frequencies in one of the excavated areas of the site (BRAT5, 19.5%), but were rare in another (BR1996, 2%), possibly due to varying degrees of winnowing during the burial of the different areas of the site (see Chapter 3). Most of the second-generation flakes from the archaeological assemblages (87%) are complete, and 79% of these exhibit pronounced bulbs of percussion. The butts are often large and have a wing-like or lens-like shape (Plates 22-23). Second-generation flakes are thin and sharp, with an angle of 25–55º between the lateral and distal edges (mean 37.37, std. d. 7.046). In the archaeological assemblages, 11% of the second-generation flakes exhibit marks of distal impact similar to the impact that resulted from hammerstone blows during the experiments (Plate 25).
The marks of hammerstone impact observed on the experimentally produced flakes are similar to the dorsal impact marks seen on the archaeological artifacts (Tables 6.4, 6.13, 7.9; Plates 18, 31). The same types of impact marks occur in both assemblages, with some differences in the frequencies of combinations of different marks. Slight variations in manufacturing techniques, e.g. changing the weight of the hammerstone or the force of the blow, may be responsible for this discrepancy (see also Bergman et al. 1987).
80
Chapter 7: Experimental knapping
PLATES
Plate 1. Bizat Ruhama archaeological assemblages. Large brecciated chert core weighing 3,8 kg found in BRT6. 81
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Plate 2. Bizat Ruhama archaeological assemblages and experimental assemblages. A. Flat discoid pebbles broken on an anvil during the experiment: pebble fragments: 1; small thin breaks 2; signs of opposite impact on the broken surfaces of the pebbles: 3. B. Broken flat discoid pebbles with signs of impact on broken surface: 1,2; broken flat discoid pebble with few additional small and thin removals: 3. 82
Chapter 7: Experimental knapping
Plate 3. Bizat Ruhama archaeological assemblages. Orthogonal multifacial unidirectional core with subspheroid shape: 1; Fractured pebbles with signs of opposite impact: 2,3. 83
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Plate 4. Bizat Ruhama archaeological assemblages. Orthogonal unifacial unidirectional cores: 1,3,5; multifacial multidirectional core: 2; orthogonal multifacial unidirectional core with subspheroid shape: 4; flat discoid pebble: 6. Cores 3,5 exhibit signs of anvil impact on the distal end of the debitage surfaces. 84
Chapter 7: Experimental knapping
Plate 5. Bizat Ruhama archaeological assemblages. Orthogonal multifacial unidirectional cores with polyhedral/ subspheroid shapes: 1,4; preferential surface core: 2; orthogonal multifacial unidirectional core: 3. 85
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Plate 6. Bizat Ruhama archaeological assemblages. Chopper on flat discoid pebble: 1; multifacial multidirectional cores: 2,3; exhausted cores: 4,5,6. 86
Chapter 7: Experimental knapping
Plate 7. Bizat Ruhama archaeological assemblages. Exhausted cores with removals of thin flakes on the last stage of the reduction: 1-7,10; multifacial multidirectional cores 8,9. 87
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Plate 8. Bizat Ruhama archaeological assemblages. Preferential surface core: 1 (arrow points at incipient cone on the core striking platform); Flaked flakes: 2-5. 88
Chapter 7: Experimental knapping
Plate 9. Bizat Ruhama archaeological assemblages. Bipolar cores. 89
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Plate 10. Bizat Ruhama archaeological assemblages. Bipolar cores. 90
Chapter 7: Experimental knapping
Plate 11. Bizat Ruhama archaeological assemblages. Bipolar flakes 1-14. Distal end features: combination of scars, crushing and wedge-shaped crack lines: 1; sequence of scars: 2,5,13;14; Combination of crushing and wedge-shaped crack lines: 3; scar: 4,7,12; wedge-shaped crack line: 6,9,10,11; Opposite bulbs: 8. 91
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Plate 12. Bizat Ruhama archaeological assemblages. Bipolar flakes 1-5, distal end features: combination of crushing and scar: 1; wedge-shaped crack lines: 2,4; sequence of scars: 3,5. 92
Chapter 7: Experimental knapping
Plate 13. Bizat Ruhama archaeological assemblages. Bipolar flakes. Distal end features: opposite bulbs: 1; sequence of scars on the distal end of the ventral surface: 3,4,6; crushing and scars: 2; wedge-shaped crack line: 5. Bulb: prominent 1,4,6; with pronounced cone and ripples 2; flat: 3; crushed with ridge 5. 93
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Plate 14. Bizat Ruhama archaeological assemblages and experimental assemblages. Shape of the ventral surfaces of bipolar flakes. Archaeological assemblage 10-19; experimental assemblage 1-9. Bulb: prominent: 1,2,10,14,17,19; flat: 6-8,15,16; crushed bulbs and bulbs with ridges: 3,5,9,11,12,13,18; pronounced cone and ripples: 4. Distal end features: opposite bulbs: 1,14; scars: 2,6,8,9,18,19; wedge-shaped crack lines: 10,12, 17; crushing: 11,15; combination of scars, crushing and crack-lines: 16. 94
Chapter 7: Experimental knapping
Plate 15. Bizat Ruhama archaeological assemblages and experimental assemblages. Distal end features of the bipolar flakes in archaeological assemblages (1-9) and experimental assemblages 10-18. Wedge-shaped crack lines: 1,3,4,5,7,8,13,14,17; crushing: 2,12; scar: 6,10,11,15; combinations of different features: 9,16,18. 95
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Plate 16. Bizat Ruhama archaeological assemblages. Flaked flakes: 1-8. Arrows mark post-detachment removals. 96
Chapter 7: Experimental knapping
Plate 17. Bizat Ruhama archaeological assemblages. Anvil flakes and modified flakes with signs of impact on the dorsal ridges. Pointed piece 1; Anvil flakes 2, 5-7; Clactonian notch 3; Irregular and step-like trimming 4. Arrows point at signs of impact on the dorsal face of the artifacts. 97
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Plate 18. Bizat Ruhama archaeological assemblages. Dorsal and ventral features of anvil flakes and modified flakes. Dorsal features: crushing: 1,2,3,6,7,9; wedge-shaped crack line: 4; combinations of features: 5,8. Morphology of the scars on the edges: step-like scar: 1,2; wedge-shaped crack lines and crushing: 3,5; Clactonian notch: 6,9; conchoidal fracture scar: 7; combinations of different scars: 6,8. 98
Chapter 7: Experimental knapping
Plate 19. Bizat Ruhama archaeological assemblages. Clactonian notches: 1,8,12,13,15; Pointed flakes: 2,7; Flakes with marginal scars: 3,4,11; Flakes with scaled trimming: 5,9,10,14; Flake with irregular trimming: 6. Artifacts 2,5,7,8 exhibit signs of impact on the dorsal ridges. 99
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Plate 20. Bizat Ruhama archaeological assemblages. Flakes with irregular and step-like trimming: 1,3,4,5,6,9,12,15, 23; Pointed flake: 2,8,10,11,13,14,16,20,21,24; Clactonian notches: 7,18,19; Flakes with scaled trimming: 17,22. Artifacts 1,3,8,9,20,23 exhibit signs of impact on dorsal edges. 100
Chapter 7: Experimental knapping
Plate 21. Bizat Ruhama archaeological assemblages. Pointed flakes: 1,2,4,7,8,9,13,14,16 (pointed flakes 4,9,13,14,16 are shaped by Clactonian notch and additional scars); Clactonian notches: 6,12; Flakes with step-like trimming: 10; Flakes with scaled trimming: 11,15,18. Artifacts 4,5,6,8,9,13 exhibit signs of impact on dorsal edges.
101
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Plate 22. Bizat Ruhama archaeological assemblages. Second generation flakes: 1-12. Artifacts 2,3,7,8,9 exhibit signs of impact on the distal end of the ventral surface. 102
Chapter 7: Experimental knapping
Plate 23. Bizat Ruhama archaeological assemblages. Second generation flakes: 1-5; artifact 3 exhibits signs of impact on the distal end of the ventral surface. 103
Plate 24. Bizat Ruhama archaeological assemblages and experimental assemblages. Second generation flakes. Note the large butts of the specimens. The butts of experimental specimens were produced by an anvil impact.
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
104
Chapter 7: Experimental knapping
Plate 25. Bizat Ruhama archaeological assemblages. Signs of impact on the distal ends of the ventral surfaces of the second generation flakes. 105
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Plate 26. Experimental assemblages. Bipolar cores: 1-3; Simultaneous Siret and distal break produced during free-hand hard hammer technique flaking. 106
Chapter 7: Experimental knapping
Plate 27. Experimental assemblages. Bipolar flakes. Distal end features: Wedge-shaped crack line: 1,2; crushing and scars: 3,5; scars: 4,6. Bulb: crushed: 1,3,5; prominent: 2,4; flat: 6. 107
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Plate 28. Experimental assemblages. Clactonian notch (1) and pointed flake (2) produced during anvil-supported knapping of the flakes conjoined with SGF that were detached during the knapping. The flakes were placed with the ventral surface on an anvil and struck on the dorsal surface with brecciated chert pebble hammerstone. 108
Chapter 7: Experimental knapping
Plate 29. Experimental assemblages. Modified flakes produced during the anvil-supported knapping of the flakes. The “retouch” was accidentally produced by an anvil impact. Pieces 1,2,3,4,5,8,10 exhibit signs of the hammerstone impact on the dorsal faces. Typology: 2,3,5,6,9,10 – Pointed flakes; 1,7 – Clactonian notches; 8 – Flake with scalar trimming. Morphology of the scars: 1,7 – Clactonian notches; 2,4,9 – Clactonian notches and scars; 3,10 – Step-like scars; 5,6,8 – Conchoidal fracture scars. 109
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Plate 30. Experimental assemblages. Anvil flakes and modified flakes produced during the anvil-supported knapping of the flakes. Clactonian notch: 1; flakes with irregular trimming 2, 7-9; flakes with scaled trimming 3-6; pointed piece 10. Artifacts 2, 7-10 exhibit signs of crushing on the dorsal ridges (marked by arrow). 110
Chapter 7: Experimental knapping
Plate 31. Experimental assemblages. Dorsal and ventral features of anvil flakes and modified flakes. Dorsal features: crushing: 1-4,7,9; wedge-shaped crack lines and crushing: 5,8; scars: 6. Morphology of the scars on the edges: step-like scar: 2,4,9; wedge-shaped crack lines and crushing: 1,3,5; conchoidal fracture scar: 7,8; combinations of different scars: 6. 111
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Plate 32. Experimental assemblages. Second generation flakes produced during anvil-supported knapping of the flakes: 1-12. The butt was always produced by an anvil impact. Replicas 3,4,9,12 exhibit signs of impact on the distal end of the ventral surfaces. 112
Chapter 7: Experimental knapping
Plate 33. Nahal Hesi archaeological assemblages. Cores on small local Eocene pebbles: 1-4; small unretouched flakes: 5-9; large scrapers made on raw materials that were not found in the area during the raw material survey. 113
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Plate 34. Experimental assemblages. Handaxe made on limestone pebble and chopper made on brecciated chert pebble. 114
CHAPTER 8: DISCUSSION PATTERNS OF INTRA-SITE VARIABILITY
Comparison to the data from Olduvai, Koobi Fora and ‘Ubeidiya shows that in terms of density Bizat Ruhama is in the range of other Early Pleistocene sites. Early Pleistocene hominins rarely created the very dense, multi-level sites that we see in later periods. In open-air Plio-Pleistocene sites, the density rarely exceeds 30–40 pieces per horizon up to 50 cm thick (Bar-Yosef and Goren-Inbar 1993, Kroll and Isaac 1984, Isaac 1997b, Leakey 1971, Schick 1987). Only in a few exceptional cases like FxJj 20AB, FxJj 20M and FxJj i8NS in Koobi Fora (Isaac 1997b), Lokalalei 2C in West Turkana (Delagnes and Roche 2005) and A.L. 894 in Hadar (Goldman-Neuman and Hovers 2009), does the density exceed 100 stone artifacts per square meter.
The archaeological remains at Bizat Ruhama are not evenly distributed over the site but rather appear in a number of concentrations. From their stratigraphic positions, it is clear that these concentrations all belong to the same sedimentological episode. The two largest excavated areas, BRAT5 and BR1996, located approximately 50 m apart, sample the densest concentrations at the site. According to the geoarchaeological data presented in Chapter 3, both concentrations were accumulated largely by anthropogenic agencies and have been only slightly disturbed by postdepositional processes. Hence, the composition of the lithic assemblages reflects hominin activities and permits the study of patterns of intra-site variability.
Interestingly, the density of bones also varies drastically between the excavated areas: in BRAT5 the bones outnumber the stone artifacts, while only a few dozen bones were recovered from BR1996. The presence/absence of bones, together with artifact density, have long been central issue in discussions of hominin transport behavior, their use of the environment, the context and structure of the sites and the social structure of early hominin groups (Binford 1981, 1987, Schick 1987, Isaac 1978, 1986, Isaac and Harris 1978, Potts 1984, 1988, Stern 1993). More research is needed to evaluate whether the concentration of artifacts with scarce bones in BR1996 reflects different hominin activities or varying preservation of bones across the site. Longer subaerial exposure of bones in BR1996 could have caused their disintegration in a few years (e.g. Behrensmeyer 1978). The strikingly higher patination rate of artifacts and higher frequency of abraded pieces in BR1996 does indeed speak in favor of longer subaerial exposure.
The study indicates that in most technological and metrical characteristics the lithic artifacts from all of the excavated areas and the survey are strikingly similar. The resemblances in the type, size and shape of raw materials, debitage methods and techniques, together with the morphology and size of the artifacts, unequivocally demonstrate that all the excavated assemblages are part of the same industry and thus should be treated as a single technological entity. The differences that do exist between the two main excavated areas are of behavioral and economic, rather than technological, character. The first major difference is in the density and weight distributions of the lithic artifacts. The density of artifacts per square meter is 90 in BR1996 and 28 in BRAT5. When expressed in weight, these differences are also very striking: 5170 g from 11 m2 in BR1996 versus 3551 g from 25 m2 in BRAT5. In view of the evidence from the site formation studies, these differences should be accredited to variation in hominin activity rather than to postdepositional disturbances. Consequently, we must conclude that larger amounts of stone were transported to BR1996 than to BRAT5. One cannot determine with confidence why the transportation rates in these two concentrations are so different, although hominin activities, group size, duration of occupation or repeated occupation of the same location are among the possible reasons. Further excavation may help to clarify this issue.
Finally, the rates of core reduction and intensity of flake modification seem to be higher in BR1996 than in BRAT5. The frequencies of cores and complete flakes are higher in BRAT5 than in BR1996, while the frequencies of categories that involve more intensive flaking and modification (exhausted cores, anvil flakes and modified flakes) are higher in BR1996. The technological features of complete flakes also point to higher intensity of reduction in BR1996. For example, complete flakes in this area exhibit higher numbers of scars and lower amounts of cortex on the dorsal faces. 115
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
To conclude, the most important insight from the intra-site comparison is that all the excavated assemblages belong to a single technological entity. Some interesting variations in the hominin occupation at the site were recorded, the most notable being the higher rate of transportation of raw material and greater intensity of knapping in BR1996.
1997, Goldman-Neuman and Hovers 2009, Harmand 2009, Stout et al. 2005, Zaidner 2003a). In such cases only the general source area can be identified. Another difficulty arises from the fact that quantifying the composition of raw material involves comparison between complete nodules and knapped material. Here, several factors can cause under/overrepresentation of certain raw material types in the archaeological context. Among these are the fragility of the raw material, the size of the nodules, the use of different reduction strategies and different intensities of knapping. All of these factors need to be considered when claiming that hominins employed intentional selection of raw material.
RAW MATERIAL EXPLOITATION STRATEGIES In addition to being the initial step in every chaîne opératoire and thus constituting an essential part of the reconstruction of hominin technical systems, patterns of raw material exploitation are important for the study of hominin economic adaptations, land use and cognitive abilities. For Plio-Pleistocene sites, the potential value of reconstructing raw material exploitation strategies for the understanding of hominin cognitive and adaptive abilities has long been recognized. For instance, transportation of raw materials over long distances is seen as a major behavioral aspect distinguishing Early Pleistocene hominins from apes (e.g. Brown and Hovers 2009, Marwick 2003, Schick 1987 and references therein). The distance over which raw materials were transported seems to have changed considerably during the Plio-Pleistocene. In sites dated to earlier than 2 Ma, the raw material sources were usually less than 1 km away (Féblot-Augustins 1997), while in later sites raw materials were transported from sources up to 13 km away. This difference marks an important shift in hominin behavior toward higher investment in raw material transport and is one of the most prominent characteristics distinguishing between the lithic production systems of Oldowan sites predating and postdating 2 Ma (Goldman-Neuman and Hovers 2009, 2012 and references therein).
In the exploitation stage, selectivity may be expressed in the use of different raw materials to produce different tool types or the application of different knapping methods to different raw materials (Bar-Yosef and Goren-Inbar 1993, Belfer-Cohen and Goren-Inbar 1994, Braun et al. 2009, Piperno et al. 2009, Stiles 1991). The study of selectivity in exploitation deals only with human impact on the raw materials and is independent of problems arising from identification, sampling and classification of raw material sources. At Bizat Ruhama the study of transport and selectivity in raw material usage is closely tied to the question of raw material size. This is because the absence of Acheulian bifaces and the small size of the artifacts have often been viewed as a consequence of raw material constraints. Hence, a major goal of the present study was to determine whether the available raw material nodules are suitable for the production of Acheulian bifaces. In order to reconstruct the raw material exploitation strategies at Bizat Ruhama, the data collected during the study of archaeological material and raw material sources were intersected with the knapping qualities of local stone and with the data from the nearby Middle Pleistocene Acheulian site of Nahal Hesi. The questions that I attempted to answer during the present study were:
Another aspect of raw material exploitation strategies that has received attention in recent studies is selectivity. Some studies show that as early as the Pliocene, hominins were acquainted with the physical properties of different rock types and tended to pick fine-grained, even-textured rocks for knapping (Goldman-Neuman and Hovers 2009, 2012, Harmand 2009, Hovers and Brown 2009, Piperno et al. 2009, Semaw 2006, Semaw et al. 2003, Stout et al. 2005). Selective use of raw materials may be evident during two stages of the technical system: acquisition and exploitation. In the acquisition stage, selectivity may be apparent in a composition of raw materials at the site that differs significantly from that of raw material sources in the area (Harmand 2009, Goldman-Neuman and Hovers 2009, Stout et al. 2005, Zaidner 2003a). The challenge faced by archaeologists conducting such studies stems from the fact that in most cases the exact location of the sources from which the hominins obtained raw materials remains unknown. This is especially true for Plio-Pleistocene sites, where in most cases secondary raw material sources, namely pebbles from fluvial conglomerates, were used (Féblot-Augustins 116
1.
Did the Bizat Ruhama hominins apply intentional selection during raw material acquisition?
2.
Were the raw materials transported far from their sources?
3.
Are the available raw materials suitable for production of Acheulian bifaces?
4.
Is the small size of the artifacts at Bizat Ruhama an outcome of the size and shape of the available raw materials?
5.
Were the different types of raw materials knapped by different methods and techniques and used for
Chapter 8: Discussion
production of different tool types?
Raw material transport
pebbles all derive from the same source area (GoldmanNeuman and Hovers 2009, Stout et al. 2005). At present, the closest exposures of the Pleshet Formation are located at a distance of 2.5 km from the site, but it is likely that during the Early Pleistocene the Pleshet Formation was more extensively exposed in the area of the site (see Chapter 4). Pebbles from the Ahuzam Formation were not used at the site, probably because of their scarcity in the vicinity.
Bearing in mind the absence of primary raw material sources in the Bizat Ruhama area, it is not surprising that pebbles were the only raw materials used at the site. In view of the shape of the pebbles at the site, the presence of flat discoid pebbles and the roundedness and smoothness of the cortical surfaces of the artifacts, Pleshet conglomerates were probably the only source of raw materials used at the site. Their homogeneity in shape and size suggests that the
Comparison of the lithological composition at BR1996, BRAT5 and BRT2 with the sampled conglomerate exposures indicates that all the raw material types used at the site were available in the vicinity (Figure 8.1). No exotic rock types or evidence for the use of raw materials other than pebbles were found. The flakes at the site are smaller than the cores (Figure 8.2), indicating that there was no influx of large flakes knapped elsewhere. The size of the pebbles unearthed
6.
Do the raw material acquisition, transportation and exploitation strategies at Bizat Ruhama indicate preplanning and foresight?
Figure 8.1. Chert types used at the site compared to sampled exposures of the Pleshet Formation.
Figure 8.2. Length of cores and complete flakes by raw material type in the Bizat Ruhama assemblages. 117
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
during the excavations is in clear agreement with the size of the locally available pebbles, pointing to use of the local sources of raw material (Figure 8.3). Small, flat “beachlike” pebbles were routinely used at the site, usually simply broken on an anvil. It is highly unlikely that the hominins invested effort in transporting such poor raw material from a distance; rather, the raw material source was close, perhaps closer than today.
assemblages the frequency of first flakes is 5–12%, in the sites they are either absent or extremely rare. Toth suggested that cortical flakes are missing because the initial stages of core reduction were executed elsewhere and the cores were brought to the sites only after this procedure. At Bizat Ruhama, however, other options should be considered as well. Unlike the Koobi Fora assemblages, the flake assemblage of Bizat Ruhama is strongly biased by postdetachment treatment. Since it was the largest flakes that were chosen for further knapping, many of these came from the initial stages of core reduction (see for example Plates 19:5, 9, 13, 14; 20:12, 14, 16, 19; 21:2, 10, 17, 18). Thus, many of the fully cortical flakes were further knapped and this, rather than the absence of the initial reduction stages, could be the main reason for their underrepresentation in the archaeological context.
Selectivity in raw material acquisition and exploitation At Bizat Ruhama selectivity in acquisition is clearly visible in the absence of limestone, despite its dominance in the surrounding conglomerates. Although the Early Pleistocene hominins were acquainted with the superior knapping qualities of chert (Hay 1976, Stiles 1998), the virtual absence of limestone at Bizat Ruhama is noteworthy, given that it was used in other Early/early Middle Pleistocene sites in the Levant (Bar-Yosef and Goren-Inbar 1993, Belfer-Cohen and Goren-Inbar 1994, Clark 1969b). Moreover, in the Bizat Ruhama area limestone was used during the Middle Pleistocene (Ohel 1976; see also below).
Figure 8.3. Length of pebbles at the site compared to pebbles from the Pleshet Formation.
Another question connected to the transportation of raw materials is whether the entire knapping sequence took place at the site. The composition of the assemblages in both BRAT5 and BR1996 exhibits no evidence that elements of the chaîne opératoire are missing. Both assemblages contain cores, cortical flakes, flakes, secondary knapped flakes and SGFs. The presence of a few complete and tested pebbles indicates that some of them were brought to be knapped at the site. Nevertheless, the number of first flakes and flakes with 100% of cortical cover produced in the knapping experiments is higher than those recorded for BR1996 and BRAT5 (Table 8.1). A similar tendency was noted for the Koobi Fora assemblages during the experimental study conducted by N. Toth (1987). While in his experimental
1-25% 26-50% 51-75% 76-99% 100% First flake (entame) Total
61 14 4 4 4 2 89
Another characteristic of the raw material composition pointing to intentional selection is the prevalence of Eocene chert, which is at least twice as frequent at the site as Eocene chert pebbles in the sampled exposures (Figure 8.1). All the knapped material at the site was weighed; weight was considered to be a more reliable measure of the amount of the raw material brought to the site, since it is not affected by the fragility of raw materials or different intensities of knapping. When compared to the weight of the pebbles collected during the survey, the weight of the artifacts indicates that the number of Eocene chert pebbles knapped by Bizat Ruhama hominins was substantially higher than
BRAT5
68.5% 15.7% 4.5% 4.5% 4.5% 2.2% 100.0%
55 17 13 7 1 0 93
BR1996 59.1% 18.3% 14.0% 7.5% 1.1% 0.0% 100.0%
Experiments 126 38.8% 54 16.7% 45 13.9% 48 14.7% 34 10.6% 17 5.3% 325 100.0%
Table 8.1. Cortical cover of complete flakes in the Bizat Ruhama assemblages compared to the experimental assemblages. 118
Chapter 8: Discussion
that of brecciated chert pebbles, while translucent chert seems to be negligible (Table 8.2). Three average pebbles would be enough to produce the total number of brecciated chert artifacts in BRAT5, while ten average pebbles would be required to produce the total number of Eocene chert artifacts in the same area. Given its scarcity in the nearby conglomerates, it is clear that the hominins invested extra effort to obtain Eocene chert pebbles.
Eocene chert Brecciated chert Translucent chert Total
BRAT5 1585 1761 205 4555
BR1996 2260 2340 570 6659
during reduction of large pebbles. When knapping small pebbles, organized debitage methods could not be applied because of the smaller volume of the cores and they were usually knapped by the bipolar technique. The influence of raw material quality on selection patterns Together with shape and size, quality is the third vital characteristic influencing raw material selection patterns. Quality seems to have been an important factor in selection of raw material as early as the Pliocene (Harmand 2009, Stout et al. 2005). Given that Eocene chert shows no advantages in size and shape in comparison to brecciated chert, I hypothesize that the preferential use of small Eocene chert pebbles at Bizat Ruhama may stem from its superior knapping quality. During the knapping experiments three questions were asked:
Survey 144 611 159
Table 8.2. Weight of the lithic assemblages and mean weight of the pebbles collected during the raw material survey (g).
Eocene chert pebbles seem to have been more intensively reduced. The number of exhausted Eocene chert cores is far higher than for other types of raw materials. The higher number of scars on the cores and the lower amount of cortex on the complete flakes also point to more intensive reduction of Eocene chert. In addition, Eocene chert flakes were clearly the preferred blanks for the secondary knapping of flakes.
1.
How suitable is the raw material for a series of controlled removals in which the configuration of the core remains as planned by the knapper?
2.
How easy is it to initiate the fracture in the raw material?
3.
How fragile is the raw material?
To answer the first question, the number and types of accidents that led to the abandonment of the core debitage surface during free-hand hard hammer reduction were recorded. In all brecciated chert pebbles the debitage surface was destroyed at least once (in six cases the debitage surface was destroyed three times). In most cases the breakage occurred because of fissures in the raw material. Often a few large flakes were produced by one blow that fractured the pebble at the contact between the chert fragments and the matrix and destroyed the debitage surface. In contrast, while knapping Eocene chert the debitage surface was abandoned because of the loss of an appropriate angle after a series of removals (seven cases) or because of hinge scars that prevented further reduction (two cases). The results indicate that in terms of controlling the configuration of the core, Eocene chert is the preferable raw material.
The influence of pebble size on patterns of raw material selection Although some large pebbles were knapped at the site, on the whole the archaeological assemblages show no evidence for the selection of large pebbles (Figure 8.3). On the contrary, some very flat and small pebbles were knapped. The artifacts at Bizat Ruhama are generally small (Figure 8.2). The largest artifact is a broken brecciated chert core with six small removals found in BRT6, 22 cm long and weighing 3.8 kg (Plate 1), followed by a core fragment 98 mm long from BR1996. Other artifacts at the site, however, are smaller than 70 mm. The small size of the artifacts largely results from the use of Eocene chert pebbles, 40 mm long on average and rarely exceeding 50 mm in maximum length (see Chapter 4). The influence of raw material size on the size of the artifacts is clearly visible while comparing the length values of complete flakes on Eocene and brecciated chert (average length brecciated chert 26.4 mm, Eocene chert 20.3 mm; Kolmogorov-Smirnov test: Z = 2.04; p = 0.000).
During the experiment we noticed that it is much easier to initiate the fracture in Eocene chert pebbles. To knap brecciated chert pebbles, on the other hand, a great deal of power needs to be applied, especially when the lack of suitable edges prevents the initiation of the fracture. In ten cases we did not succeed in initiating a fracture of a spheroid brecciated chert pebble even when using the bipolar technique, whereas this never occurred during the knapping of Eocene chert pebbles. These differences are probably due to the much finer texture of Eocene chert, which may have played an important role in the raw material acquisition patterns and in the selection of knapping methods and techniques at the site.
The size of the pebbles was probably one of the most important features in the selection of core reduction methods and techniques. More organized debitage methods, such as the orthogonal method with a series of unipolar removals or the preferential surface method, were mostly employed 119
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
The fragility of the raw material was assessed by intersecting the total number of removals with the total number of flakes and fragments and with the number and type of knapping accidents for each pebble. In the free-hand hard hammer technique experiments, the number of complete flakes was higher among brecciated chert DPs and the number of accidents was lower (see Tables 7.1, 7.3). The knapping of Eocene chert pebbles produced more flakes and fragments per removal than the knapping of brecciated chert pebbles (Appendix 3C; the ratio of total number of pieces to total number of removals is 1.94 for Eocene chert and 1.61 for brecciated chert). The apparent greater fragility of Eocene chert may be accounted for by its fine-grained texture, but may also be due to the small size and mass of the Eocene chert pebbles. When knapping larger brecciated chert pebbles by the free-hand hammer technique, the blow was often positioned far from the edge of the striking platform, making it possible to produce thick flakes. In contrast, when using the same technique on small Eocene chert pebbles, the blow had to be placed closer to the overhang, producing flakes that were thinner and more likely to break. Thus, it seems that the greater fragility and small size of the pebbles are the factors responsible for the larger number of breaks and accidents among Eocene chert flakes in the free-hand hard hammer experiment.
found at Nahal Hesi occur in local conglomerates. However, the lithological compositions of the two assemblages are markedly different (Table 8.3). At Nahal Hesi limestone was the most frequent type of rock used, followed by Eocene, translucent and brecciated chert, while at Bizat Ruhama limestone is virtually absent. Another important difference is in the size of the raw materials used, as reflected in the size of the artifacts (Figure 8.4). While the size of the raw materials was not a prime concern of the Bizat Ruhama hominins, at Nahal Hesi larger pebbles were used and some of the Eocene and translucent chert artifacts were made on nodules of raw material that are larger than those available in the area. It seems that the Nahal Hesi hominins invested special effort in searching for and procuring large nodules of raw material.
Limestone Eocene chert Brecciated chert Translucent Gray chert Undefined Total
In the bipolar technique experiments, it was found that Eocene chert fractures much more easily and it was possible to remove a large number of flakes and fragments. Brecciated chert was found to be much harder to fracture; many blows were unsuccessful and the knapping resulted in a much higher number of breaks (Table 7.1).
Nahal Hesi 147 26.4% 139 25.0% 81 14.5% 112 20.1% 40 7.2% 38 6.8% 557
Bizat Ruhama 1 0.1% 959 52.2% 460 25.1% 255 13.9% 53 2.9% 108 5.9% 1836
Table 8.3. Rock types at Nahal Hesi and Bizat Ruhama.
The pattern of raw material exploitation at Nahal Hesi clearly shows that different raw materials were selected for the production of different tools. Limestone was used only for production of handaxes and choppers, while Eocene and translucent chert were used for production of flake tools (Figure 8.5). The use of limestone is a rare phenomenon in the Levant, occurring occasionally only at the Early Pleistocene and the beginning of the Middle Pleistocene sites of ‘Ubeidiya (Bar-Yosef and Goren-Inbar 1993), Latamne (Clark 1969b) and Gesher Benot Ya‘aqov (GorenInbar et al. 2000, Belfer-Cohen and Goren-Inbar 1994). In sites that are geographically and chronologically close to Nahal Hesi, limestone is virtually absent. It was rarely used in the southern Levant at sites postdating Gesher Benot Ya‘aqov, i.e. for almost all of the Middle Pleistocene, when chert was the sole raw material in use. The use of limestone at Nahal Hesi is likely a response to the absence of local chert suitable for the production of Acheulian bifaces and core-tools; in the absence of other alternatives, limestone was probably an ad hoc choice.
The results of the knapping experiments show that the knapping quality of raw materials used at Bizat Ruhama is determined by the homogeneity and grain size of the chert type and by the size of the pebbles. While the unhomogeneous and coarse-grained brecciated chert pebbles are hard to knap, Eocene chert is a much friendlier material. It fractures easily and the configuration of cores can be controlled without difficulty. Its superior knapping quality may explain why Eocene chert was used despite the small size of the pebbles.
Raw material at the Acheulian site of Nahal Hesi: a case study in raw material acquisition and exploitation in the Bizat Ruhama area The site of Nahal Hesi is located on the bank of the Nahal Shiqma stream, 3.8 km northeast of Bizat Ruhama (Figure 3.3). The site is dated to the Middle Pleistocene in accordance with the characteristics of the lithic assemblage and the fauna (Yeshurun et al. 2011). Situated only a few kilometers from Bizat Ruhama, Nahal Hesi provides a case study of the use of raw materials during the later Lower Paleolithic in the area. As at Bizat Ruhama, all the lithological types
Eocene and translucent chert artifacts at Nahal Hesi can be divided into two groups, each representing different raw material acquisition strategies, reduction sequences and transportation patterns. The first group consists of cores made on small pebbles and their products – small, thin, 120
Chapter 8: Discussion
Figure 8.4. Length of artifacts by raw material type at Bizat Ruhama and Nahal Hesi.
Figure 8.5. Rock types at Nahal Hesi by major lithic group.
sharp flakes (Plate 33:1–9). This group is clearly made on local raw materials and, judging by the presence of cores and cortical flakes, was probably knapped on the spot. These pieces were not retouched, but many of them show edge damage possibly caused by use. The second group includes
large, uncortical, carefully retouched scrapers (Plate 33:10– 13). The technological and metrical characteristics of the artifacts in this group indicate an exotic origin. Most of them show complex patterns of scars on the dorsal face that indicate different knapping sequences from those readable 121
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
on the cores and the debitage. The scrapers seem to have been produced by different reduction sequences from those practiced at the site and are made on raw materials that are larger than those available in the area. Thus, it is likely that they were produced elsewhere and introduced to the site as blanks or prepared tools.
Bizat Ruhama. Both raw materials were readily used by the Acheulians at Nahal Hesi and at other Acheulian sites in the Bizat Ruhama area (Ohel 1976), mostly for the production of handaxes and choppers. During current knapping experiments we also produced some handaxes from limestone and brecciated chert pebbles collected during the raw material survey. This indicates that both raw materials are suitable for the shaping of core-tools (Plate 34). Thus, a lack of suitable raw materials cannot be the reason for the absence of bifacial tools at Bizat Ruhama. Rather, it seems that the technological repertoire of the Bizat Ruhama hominins did not include bifacial knapping.
Overall, the exploitation of raw materials at Nahal Hesi points to a complex behavior that is distinguished by opportunistic production of handaxes and small thin flakes from locally available raw materials and by careful production and transportation of large flake-scrapers. This pattern is clearly influenced by constraints imposed by the availability of raw material. The most striking features of the assemblage are the use of limestone, a raw material that was usually overlooked by Acheulians, and the transportation of prepared tools over distances.
Comparison with the patterns of raw material use at Nahal Hesi indicates that the two sites had very different habits of raw material acquisition and exploitation. The differences are reflected in longer transportation distances, the influx of large flakes made elsewhere and a significantly more complex pattern of selectivity in the exploitation of raw materials at Nahal Hesi. These differences point to the higher mobility, more complex mental templates and more elaborate technology practiced during the Middle Pleistocene by hominins in the Bizat Ruhama area.
The complexity of raw material exploitation strategies at Bizat Ruhama The technological behavior reflected in the raw material exploitation strategies at Bizat Ruhama is not complex and no evidence for preplanning and foresight was found. The assemblages are characterized by an investment of extra effort in the acquisition and exploitation of Eocene chert and by a total avoidance of limestone. Knapping experiments showed that the probable reasons for preferential use of Eocene chert were its superior fracture mechanics and its higher efficiency in the production of flakes. During knapping of Eocene chert pebbles, initiation of the fracture and control of the configuration of the core are substantially easier. The experiments demonstrated that the knapping of Eocene chert produces larger amounts of flakes and fragments.
KNAPPING STRATEGIES AT BIZAT RUHAMA From the description of the Bizat Ruhama industry presented in Chapters 5 and 6, it is clear that lithic production at the site was geared toward the production of flakes and thus should be identified as debitage sensu Inizan et al. (1999). The evidence for shaping at the site is limited to a single chopper found in BR1996. The morphological and technological characteristics of all other knapped pebbles indicate that they served as cores for flake production. Flakes were thus the goal of pebble reduction. The flakes were often further knapped and modified, although the aim of this secondary modification is not always as clear as the goal of pebble reduction. In this chapter the data gathered during the raw material study, the knapping experiments and the study of the lithic assemblages will be intersected in order to identify the goals of lithic production at Bizat Ruhama and to reconstruct the paths taken by the knappers in order to reach these goals.
The simple patterns of selectivity in raw material acquisition recorded at Bizat Ruhama are documented at the sites of Gona and Lokalalei (earlier than 2 Ma), where specific raw materials were preferred, possibly because of their superior knapping quality (Harmand 2009, Stout et al. 2005). Moreover, at 2–1.5 Ma hominins showed a developed pattern of preplanning, reflected in the transportation of raw materials over long distances (Blumenschine et al. 2003, Hay 1976, Plummer 2004). Such a level of preplanning is not evident in the Bizat Ruhama assemblage. The site’s inhabitants probably used the closest available outcrops and employed a minimal selection based on the fracture mechanics of the local rocks.
Debitage techniques The identification of the knapping techniques used at Bizat Ruhama is based on the study of the archaeological material and its comparison to the products of knapping experiments. Two knapping techniques, the free-hand hard hammer and bipolar techniques, could possibly have been used at Bizat Ruhama. In the literature, the former is usually identified by large and prominent bulbs of percussion, pronounced ripples and thick butts, while the latter is characterized by
The size of the raw materials played only a minor role in the selection patterns of the Bizat Ruhama hominins. Although the evidence indicates that large pebbles were available and knapped, most of the artifacts at the site are still smaller than 3 cm. Limestone and brecciated chert pebbles, the largest in the area, were not among the preferred raw materials at 122
Chapter 8: Discussion
flat ventral faces, shattered butts and signs of impact on the distal ends of the flakes and cores. Many of the characteristic signs of the bipolar technique were found among the cores and debitage in the Bizat Ruhama assemblages. However, recent experimental studies have showed that some of the characteristics of the bipolar technique can be produced by free-hand techniques and vice versa (e.g. Jeske and Lurie 1993, Kiujt et al. 1995). One of the major aims of the experimental knapping conducted during this study was to identify the characteristic signs of the bipolar technique on the raw materials used at Bizat Ruhama, and to assess how central this technique was in the Bizat Ruhama lithic production system. Comparison between the experimental and archaeological data shows that the archaeological assemblages at Bizat Ruhama are very similar to those produced by the bipolar technique and differ from free-hand hard hammer assemblages (Figures 8.6–8.9). The most striking differences between the composition of the experimental bipolar and free-hand hard hammer assemblages are the lower frequency of complete flakes and higher frequency of angular fragments in the former (Figure 8.6). The composition of the archaeological assemblage is very similar to the bipolar experimental assemblage. It should be noted, however, that the number of complete flakes in the archaeological assemblage might have been higher if many of them had not been selected for post-detachment flaking. In contrast, the number of angular fragments in the archaeological assemblages is probably real and unbiased by post-detachment treatment.
Figure 8.7. Distal end features of the flakes in the Bizat Ruhama archaeological and experimental assemblages.
The most diagnostic features of the bipolar technique visible on flakes are signs of distal impact on the ventral
Figure 8.8. Shapes of bulb of percussion of the flakes in the Bizat Ruhama archaeological and experimental assemblages.
surfaces. In the experimental bipolar assemblage, 35% of the flakes exhibit scars, crack lines or crushing on the distal end of the ventral surface, while no such signs were found among flakes produced free-hand. In the archaeological assemblages, similar signs were identified on 40% of the flakes (Figure 8.7), indicating that the bipolar technique was intensively used at Bizat Ruhama. Other signs that seem to be diagnostic of the bipolar technique are crushed bulbs and
Figure 8.6. Composition of the Bizat Ruhama archaeological and experimental assemblages (BT = bipolar technique; FH = free-hand hard hammer technique). 123
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
was most probably the dominant technique employed at the site. The reasons for using the bipolar technique probably lie in the shape, size and knapping qualities of the local raw material and will be discussed below.
Pebble reduction strategies The raw materials used at Bizat Ruhama consist of small rounded pebbles that do not offer good knapping angles. This is probably the main reason for the absence of cores with a cortical striking platform and for the rarity of flakes with cortical butts within the assemblages. In order to start the sequence of the pebble’s reduction, it is necessary first to split the pebble into two halves or to remove an opening flake. After this, the fracture plane created by the opening blow can be used as a striking platform. Most of the cores in the Bizat Ruhama assemblages were intensively knapped and they usually preserve less than 50% of their cortex. Consequently, it is often impossible to reconstruct the shape and size of the pebbles on which the cores were made and the initial stages of core reduction. This is especially true for the cores knapped by the multidirectional and bipolar methods, which often have no cortex at all. When the cortical surfaces of the cores allow the shape of the pebble to be identified, they indicate that pebbles of spheroid/subspheroid, subdiscoid and flat discoid shapes were knapped. These are the three dominant shapes among the unknapped pebbles in the excavated assemblages and in the sampled raw material sources as well.
Figure 8.9. Butt types of the flakes from the Bizat Ruhama archaeological and experimental assemblages.
bulbs with pronounced cones of percussion (Figure 8.8). While these signs are absent from the free-hand products, they occur in low frequencies in both the experimental bipolar and the archaeological assemblages. Butts seem to be much less diagnostic of the type of technique used (Figure 8.9). The only substantial difference between the bipolar and free-hand assemblages is the frequency of shattered butts. These are more frequent in the experimental bipolar and archaeological assemblages than in the experimental free-hand assemblages.
Five debitage systems can be distinguished at Bizat Ruhama (Table 8.4). Flat, discoid “beach-like” pebbles were transversally fractured on an anvil for a purpose that is not entirely clear. Flat pebbles were used in later periods as well, for example in the Pontinian Mousterian of Italy (Kuhn 1995: 97) or Late Stone Age in South Africa (Van Riet Lowe 1946). In these cases the pebbles were split along their longitudinal axis. This method enabled use of the volume of the pebble in a way that created large flakes with long working edges. The method used at Bizat Ruhama seems to be less efficient. According to the results of the knapping experiments, the transversal breakage of the pebble employed at Bizat Ruhama resulted in two groups of products. The first group includes thick pebble fragments with two opposite cortical surfaces. The second group contains thin flakes that are 1–2 cm in maximum length. The pebble fragments were sometimes used as cores from which a few additional small flakes were removed. Small, thin flakes often provide sharp edges and could be the goal of the knapping. Whatever was the goal of the knapping, the use of such poor material may point to either serious limitations in the availability of raw material or the presence of a source of flat pebbles in close proximity to the site during its occupation. In any case, the use of flat discoid pebbles indicates that the Bizat Ruhama hominins invested little effort in searching for raw materials
One of the most interesting features of the experimental bipolar assemblage is the high number of flakes that show no evidence whatsoever of the bipolar technique. More than 50% of the flakes produced by the bipolar technique show no signs of distal impact and have plain butts and prominent bulbs. These flakes are indistinguishable from the products of the free-hand technique. In contrast, the experimental free-hand technique assemblage only rarely shows features associated with the bipolar technique and is clearly dominated by plain butts and prominent bulbs. In light of this, it is clear that many flakes that show apparent features of the free-hand technique can be produced by the bipolar technique, but not the other way round. Consequently, the 40% of flakes showing diagnostic bipolar features in the archaeological assemblages should be considered a minimum frequency. In fact, the frequency of flakes that show no identifiable bipolar features in the archaeological assemblages is similar to the frequency of such flakes in the experimental bipolar assemblage (Figures 8.6–8.9). Thus, it is clear that the bipolar technique was commonly used at Bizat Ruhama and 124
Chapter 8: Discussion
Debitage system Raw material
Bipolar technique Free-hand hard hammer technique Initiation stage
Production stage
Preparation stage
Flat discoid Multidirectional Bipolar Orthogonal debitage Preferential surface pebbles multifacial debitage debitage breakage debitage Flat discoid ? Small, mostly Spheroid/ subspheroid, mostly Discoidal pebbles pebbles Eocene chert brecciated chert pebbles pebbles + ? + + _
+
_
?
_
?
?
Pebble is fractured by bipolar technique in order to create flat surface that will serve as striking platform in the production stage
Pebbles are placed on an anvil and smashed
Opportunistic removal of flakes from different directions and debitage surfaces
Pebbles are smashed by bipolar technique
Inappropriate angles, hinged scars or small size prevent further reduction
Small size prevents further reduction
Production stage
Abandonment stage
+
Entame flake is removed by free-hand hard hammer technique in order to create flat surface that will serve as striking platform on the production stage Series of recurrent One or several recurrent unidirectional removals unidirectional removals struck from the same debitage struck from the same surface. debitage surface. Rotation of the core. In some cases a new platform was generated. In other cases previous debitage surface was used as striking platform Additional series of recurrent unidirectional removals struck from the same debitage surface. Hinged scars or inappropriate angles prevent further reduction. In some cases flake/s were opportunistically removed from different surfaces
Rectification of the striking platform by small removals in order to maintain the knapping angle One or several recurrent unidirectional removals struck from the same debitage surface Hinged scars prevent further exploitation
Table 8.4. Debitage systems at Bizat Ruhama.
but rather used everything that was at hand.
very hard to knap by the free-hand technique. During the experiments, most orthogonal cores could be knapped only by the bipolar technique. A similar link between orthogonal knapping and the bipolar technique has been observed in other bipolar technique experiments (Mourre 1996, 2004). Similarly, the archaeological assemblages show a clear link between orthogonal debitage and the use of an anvil. Among orthogonal cores, 43% exhibit signs of anvil impact on opposed sides of the debitage surface (e.g. Plate 4:3, 5). Moreover, the tested and broken pebbles of spheroid shape found in the archaeological assemblages show opposed signs of impact, indicating that they were split by the bipolar technique. Thus, orthogonal debitage and the bipolar technique are most likely interrelated features connected to
The orthogonal debitage method, with a series of unidirectional removals, was employed during the flaking of spheroid and subspheroid pebbles. The choice of this knapping method was largely predetermined by the rounded shape of the pebble. The knapping experiments demonstrated a clear link between the shape of the pebble, the angle between the striking and debitage surfaces and the need for use of an anvil. The experimental data indicate that it is virtually impossible to fracture spheroid and subspheroid pebbles without the support of an anvil. Even after initial splitting by the bipolar technique, the angle between the fracture and the cortical surfaces often remained at around 90º and it was still 125
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
the rounded, subspherical shape of the pebbles.
translucent chert pebbles are usually too small to be knapped by organized debitage methods. The majority of the cores were knapped by the bipolar technique, resulting in shattered bipolar cores and exhausted cores in FP assemblages and angular fragments in DP assemblages.
During orthogonal debitage, the anvil was used in a calculated way that enabled control of the core’s configuration for a series of 2–5 unidirectional removals. Elongated flakes with cortical backs struck from the lateral edges of the pebbles correspond to this stage of the debitage (Plates 11:12–14; 16:7). The flakes removed from orthogonal cores often cover the entire debitage surface. The striking platform was usually abandoned because of hinge fractures or the loss of appropriate angles. Either the core was then abandoned, or it was rotated and a new series of removals was struck from new platforms. A few orthogonal cores were rotated three or four times. The final products of this multiple rotation are cores of polyhedral and subspheroid shape (Plates 3:1; 4:4; 5:1, 4). At the abandonment stage, many of the orthogonal cores show isolated removals from different directions and on different surfaces. These removals suggest that the knapper sought to maximize the use of the raw material and to remove as many flakes as possible.
The organization of debitage and knapping skills The five debitage systems identified at Bizat Ruhama differ in level of predetermination and organization of the reduction sequence. In general, the debitage methods at the site are simple. The breakage of flat discoidal pebbles and both multidirectional multifacial and bipolar debitage methods are simple, opportunistic reduction sequences that show no evidence of organization or preparation and thus can be classified as simple debitage. These reduction sequences usually display no signs of preparation or rectification of the cores (Table 8.4). The orthogonal and preferential surface debitage methods, on the other hand, show a basic level of organization. Firstly, planning and foresight are discernible in the matching of the debitage method to the shape of the raw material; in particular, the selection of discoidal pebbles for preferential surface debitage was deliberate and probably preplanned. Next, both methods exhibit a clear conceptual division of the core into two surfaces: striking platform and debitage surface. The debitage surfaces demonstrate a number of recurrent unidirectional removals struck from a single striking platform. This in itself implies organization, maintaining of constant technical rules and a detailed knowledge of the consequences of the blows inflicted on the core (Delagnes and Roche 2005, de la Torre et al. 2003, Texier 1995).
Only four knapped subdiscoidal pebbles were found at the site, all of them knapped by the preferential surface method. This method exploits the pebble’s shape to create an acute angle between striking and debitage surfaces, thereby facilitating the use of the free-hand hard hammer technique. The angle was sometimes maintained by rectification of the striking platform. There is no evidence for use of the bipolar technique during preferential surface debitage. Preferential surface cores were abandoned because of deep hinges that prevented further exploitation of the debitage surface. These are the largest and the least intensively exploited cores in the Bizat Ruhama assemblages. Flakes struck from preferential surface cores are expected to be flat, to have relatively acute angles between the dorsal surface and the butt and to display evidence for more thorough butt preparation. Such flakes are rare in the Bizat Ruhama assemblages.
The striking platforms of orthogonal and preferential surface cores show signs of simple preparation (Table 8.4). The initial preparation of the striking platform involves the removing of the cortical surface of the pebble in order to create an angle that is suitable for further knapping. Both debitage methods thus include at least two sequential stages of knapping, the first aimed at preparing the core for the second. In some cases, the striking platform was rectified during preferential surface debitage in order to maintain the knapping angle and knapping was then continued on the preferential surface. Thus, both debitage methods reflect knapping skills above a simple “process of trial and error” (Wynn 1981, 1985) and demonstrate a basic level of preplanning and even an anticipation of the forthcoming, as in the case of the rectification of preferential surface cores.
Multifacial multidirectional cores usually show very little cortical cover, making it difficult to reconstruct the shape of the pebbles on which they were made and the initial stages of reduction. Some of the multidirectional cores in the first stages of the pebble reduction could have been reduced by other debitage methods. Multidirectional cores show some evidence for unsystematic use of an anvil as a support, but in most cases they were probably knapped by the freehand hard hammer technique. The cores were intensively exploited from different directions and on different surfaces. Thick flakes with a multidirectional scar pattern, irregular edges and thick irregular cross-sections were produced during multidirectional debitage.
In contrast to some of the Plio-Pleistocene sites in Africa (A.L. 894, Hovers 2009a; Localalei 2c, Delagnes and Roche 2005), at Bizat Ruhama the cores only rarely show evidence for attempts to exploit debitage surfaces that were covered by a large hinged scar. In most cases either such cores were abandoned or a new debitage surface, and often
Most of the cores discussed above are relatively large with a mean length of 40.2 mm, much larger than the rest of the artifacts in the Bizat Ruhama assemblages. Not surprisingly, they are mainly made on brecciated chert; Eocene and 126
Chapter 8: Discussion
a new striking platform, were generated. The experimental knapping indicated that a debitage surface with hinged scars could only rarely be further exploited. Due to the small size of the cores and the right angle between the striking platform and debitage surfaces, attempts at further exploitation resulted either in short step and hinged scars or, if the blow was placed very far from the core’s edge, in breakage of the core into several fragments. The knappers of Bizat Ruhama seem to have been well aware of these features of the local raw material. For example, on some of the orthogonal cores the debitage surfaces were changed two or three times because hinged scars prevented further exploitation (Plate 5:1). On the other hand, 36% of the cores show one or more incipient cones formed by unsuccessful knapping attempts. By and large, these marks could be connected to the use of the bipolar technique. Some of them are probably signs of an attempt to break cores that had lost knapping angles (Plate 5:4). The abundance of these marks probably shows that knappers had difficulty in handling small pebbles.
extremely high frequency of flakes with signs of postdetachment treatment. About half of the total amount of DPs produced during core reduction were knapped, intentionally broken, notched or modified. The magnitude of this phenomenon in the Bizat Ruhama assemblages clearly indicates that one of the major goals of core reduction was to produce blanks for further knapping. Since the secondary knapped flakes are broken and intensively knapped, it is sometimes difficult to identify the debitage methods by which they were produced. Nonetheless, it is clear that they are part of the same reduction sequences as the unmodified flakes. Secondary knapped flakes were made on flakes with a unidirectional and multidirectional scar pattern. As indicated by the presence of both cortical and uncortical flakes, flakes from all stages of core reduction were used. Nonetheless, some evidence indicates that flakes were not randomly selected for further knapping. There are three important differences between the DP and secondary flake assemblages that point to intentional selection. The first significant difference is in the type of raw material selected (Figure 8.10b; X2 = 47.406; df = 2; p = 0.000): flakes on Eocene and translucent chert were more frequently used as blanks for further modification. The other significant
Secondary flake knapping Selection of flakes for further knapping The Bizat Ruhama assemblages are characterized by an
Figure 8.10. Flake selection for secondary knapping: composition of DPs and secondary knapped flake assemblages by a) Type of DP; b) Type of raw material. 127
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
difference is in the type of DPs selected (Figure 8.10a; X2=408.239; df = 5; p = 0.000): the number of angular fragments among secondary knapped flakes is extremely low, and longitudinal and siret breaks are virtually absent. In contrast, the number of flakes with butts still intact (complete flakes and proximal fragments) is higher than among unmodified flakes. Especially noticeable is the abundance of proximal fragments, which is probably the result of postdetachment modification of complete flakes from which the distal ends were removed by flaking and breakage.
the secondary knapped flakes.
The most important criterion, however, is size. The flakes selected for knapping and breakage are the longest and thickest in the assemblages. This is especially true for the flaked flakes, which are far thicker and longer than the complete flakes. It is also true for anvil flakes and modified flakes, both shaped on the thickest flakes in the assemblages. Figure 8.11 shows that thin flakes were not used for further knapping. Flakes that are thinner than 7 mm contribute approximately 45% to the complete flake assemblages, whereas their frequency ranges between 2% and 15% among
Four groups of secondary knapped flakes and their products were identified in the archaeological assemblages: flaked flakes, anvil flakes, modified flakes and SGFs. The first group includes flakes that were used as cores for the removal of one or several small, thin flakes. The question of how the other three groups were created was addressed during the experimental knapping. It should be noted that anvil flakes, modified flakes and SGFs were not produced during the experiments in bipolar and free-hand reduction of pebbles. Conversely, during the experiments in anvil-supported
In conclusion, flakes selected for further knapping belong to different debitage systems, all recorded in the core assemblages, and to different stages of the reduction sequence. The criteria applied to the selection of the flakes relate mostly to their size: the largest and, especially the thickest, complete flakes or fragments were preferred. Techniques and goals of secondary flake knapping
Figure 8.11. Thickness histograms for complete and secondary knapped flakes. The line marks the 7 mm border, below which only very few secondary knapped flakes are found. 128
Chapter 8: Discussion
knapping of the flakes they were all routinely produced, supporting the hypothesis that at Bizat Ruhama flakes were knapped on an anvil. The products of the experimental knapping show good correlation with archaeological secondary knapped flakes in general morphology, the types of impact marks on dorsal and ventral faces, and the shape of the scars on the edges (see Plates 17-21; 28-31).
thick flakes were intentionally selected as blanks. Second-generation flakes are the thinnest and sharpest of the products of the Bizat Ruhama industry (Figure 8.12), but still have sturdy lateral and distal edges and a blunt edge formed by the butt, which allows a firm grip. In contrast, the anvil flakes and most of the modified pieces are small, thick and carinated and do not seem to be functional. Although pieces with steep edges can be used for scraping or woodworking (e.g. Coutts 1977, Gould et al. 1971, Toth 1982), they need to be large enough to allow a firm grip and still have a sufficiently long working edge. At Bizat Ruhama, however, the secondary knapped flakes are exceptionally small. Complete flakes are situated at the middle of the scale in Figure 8.12, still providing good working edges.
The experiments demonstrate that the entire spectrum of edge modification morphologies identified at the site could have been produced by anvil impact during bipolar knapping of the flakes. During knapping, small chips and microflakes that resemble what Newcomer (1976) identified as “spontaneous retouch” were removed simultaneously with larger flakes that often resemble Clactonian notches. Large and small scars often occur in succession, creating a denticulate or almost rectilinear edge that could be mistakenly identified as intentional retouch. Similar results have been obtained in other knapping experiments in which flakes were knapped by the bipolar technique (Crovetto et al. 1994, Vergès and Ollé 2011). In the experimental assemblages of Bizat Ruhama, spontaneous removals from the edges of modified flakes were generated by anvil impact due to pressure of the edge of the flake against an anvil. The number and type of spontaneous scars depended on the angle of the flake edge, the size of the contact area between the flake edge and the anvil, the strength of the blow and the type of raw material used. When the sequence of spontaneous scars occurs on the edges of experimental flakes, they are usually characterized by irregularity in scar form, size and invasiveness. A similar irregularity characterizes the edges of the modified flakes in the archaeological assemblages as well. Thus, given the results of the recent study, previous descriptions of the industry as containing high numbers of retouched flakes (Ronen et al. 1998, Zaidner et al. 2003) seem to be questionable. It should not be concluded that the Bizat Ruhama hominins applied no intentional retouch at all, since some of the more standardized pieces without signs of dorsal impact could be the result of intentional retouch. Be that as it may, it is clear that the role of intentional retouch in the Bizat Ruhama lithic production system was not as important as was suggested in previous studies.
At Bizat Ruhama the production of small flakes was not restricted to a single operational scheme; small flakes were also produced during pebble reduction and freehand secondary knapping of flakes. According to the size of the scars, all of these operational sequences yielded flakes of similar size (Figure 8.13). The breakage of flat discoid pebbles could also be part of the same lithic production strategy, since it resulted in many thin, sharp fragments. Combining all of the available evidence, I suggest that Clactonian notches and anvil flakes fall within the same group of technological debris as exhausted cores and flaked flakes, while SGFs are the desired product of the knapping of flakes.
During the experiments, each knapping sequence on average produced 4.16 thin flakes and fragments that were 1–2 cm long. These second-generation flakes are the only standardized and systematic outcome of secondary flake knapping and hence are the probable objective of the production. Judging from the experiments, the main reason for consistent detachment of uniform flakes was the use of thick steep-edged flakes as blanks for further knapping. The striking difference in the archaeological assemblages between the thickness values of flakes used as blanks for knapping and those of unmodified flakes indicates that the
Figure 8.12. Scatterplot of average thickness and edge angle of secondary knapped flakes, SGFs and complete flakes.
129
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Figure 8.13. Length values of the longest axis of the scars on flaked flakes, exhausted cores, Clactonian notches and cores.
LITHIC PRODUCTION STRATEGIES AT BIZAT RUHAMA AND THE EARLY PLEISTOCENE ARCHAEOLOGICAL RECORD The lithic production system at Bizat Ruhama is geared toward flake production and is built from three successive stages (Figure 8.14): 1.
Acquisition of raw materials.
2.
Production of flakes.
3.
Further knapping of flakes.
These three stages are connected by a relatively long succession of gestures, technical actions and decisions. The first stage is characterized by simple patterns of selection and transportation of raw materials from the closest available sources. The raw materials available in the area of Bizat Ruhama undoubtedly imposed difficulties on the knappers. However, the occupants of the site did not attempt to overcome these difficulties by transporting materials over long distances, as did their counterparts in some of the Oldowan sites in Africa (Blumenschine et al. 2003, Hay 1976, Plummer 2004) or the later Acheulian hominins of Nahal Hesi. On the contrary, they invested little effort in searching for raw materials but rather used everything that was at hand.
1.
Selection of different debitage methods for pebbles of different shape.
2.
Series of unidirectional removals of elongated flakes that cover most of the core’s debitage surface.
3.
Simple preparation and rectification of the core’s striking platform.
4.
Habitual use of the bipolar technique.
5.
Intensive core reduction.
The methods of core reduction practiced at Bizat Ruhama are known from different Oldowan sites in Africa, although not all of them occur in every assemblage. Unidirectional reduction methods are known from practically every Oldowan assemblage, from the earliest hominin sites at Gona in Ethiopia and Lokalalei 2C in Kenya to some of the late sites of the Developed Oldowan in the Olduvai Gorge and sites belonging to the Karari industry in Koobi Fora (Braun et al. 2008, Delagnes and Roche 2005, de la Torre and Mora 2005, Harris and Isaac 1997, Leakey 1971, Roche et al. 1999, Semaw 2006, Semaw et al. 2003). At 1.8–1.5 Ma, multifacial, unidirectional and multidirectional knapping appeared across Oldowan assemblages in East Africa and led to the first occurrence of cores with polyhedral and subspheroid shapes (e.g. Braun et al. 2009, de la Torre et al. 2003, Isaac 1997a, Leakey 1971, Sahnouni et al. 2002), similar to those resulting from orthogonal multifacial debitage at Bizat Ruhama. The bipolar technique, so intensively used at Bizat Ruhama, is known from many Oldowan sites, some of which predate 2 Ma (Barsky 2009, de la Torre 2004, Diez-Martín et al. 2010, 2011, Harris et al. 1987, Ludwig and Harris 1998, Merrick and Merick 1976). Finally, simple rectification of the striking platform, as seen on preferential surface cores at Bizat Ruhama, is known at 2.34 Ma from Lokalalei 2C1 (Delagnes and Roche 2005). The lithic industry of Bizat Ruhama shows no evidence for the production of large flakes that marks the emergence of the Acheulian in Africa (Beyene et al. 2013, de la Torre and Mora 2005, Semaw et al. 2009). Other advanced technological features, such as shaping, standardization of tool shape and bifacial and discoidal methods of core reduction, are lacking as well. In fact, alternating knapping either for preparation of the working edge or as part of chopper-like or discoidal flaking is not documented at the site, and there is no evidence for the advanced platform preparation seen in some of the Oldowan and Early Acheulian assemblages at Olduvai Beds I and II, Peninj and Melka Kunture (de la Torre and Mora 2005, 2009, Piperno et al. 2009).
The pebbles brought to the site were knapped by a number of methods depending on their shape. The scarcity of goodquality raw materials resulted in intensive knapping of the raw materials that were brought to the site. The following points underscore the most important aspects of this debitage stage:
1
130
Delagnes and Roche (2005) argue that the aim of the rectification at Lokalalei 2C was to remove irregularities on the natural surface of the striking platform, while at Bizat Ruhama the aim was to rectify the angle between the surfaces.
Chapter 8: Discussion
Figure 8.14. The lithic production scheme at Bizat Ruhama.
Two further components of the reduction system at Bizat Ruhama have not yet been reported from other Oldowan sites: systematic secondary knapping of flakes and the matching of a specific anvil technique to flake knapping. The Bizat Ruhama assemblage shows clear evidence for selection of thick flakes for both free-hand and anvilsupported knapping. The key aspects of this debitage stage can be summarized by the following: 1.
Intentional selection of thick flakes.
2.
Habitual use of the bipolar knapping technique.
3.
Production of thin, sharp flakes smaller than 2 cm.
of anvil impact, some of the thickest and longest of the selected flakes were knapped by the free-hand hard hammer technique. The vast majority of the selected flakes, however, were knapped on an anvil. While the technology of core reduction at Bizat Ruhama resembles the Oldowan technology employed throughout Africa, the secondary modification of flakes, so widely used at Bizat Ruhama, is rare in the Oldowan. At Oldowan sites, flakes were usually obtained directly from nodules of raw material (Barsky et al. 2011, Braun et al. 2008, DiezMartín et al. 2009, 2010, Hovers 2009a, Isaac 1997b, Isaac and Harris 1997, Kuman and Field 2009, Leakey 1971, Roche 2005, Semaw et al. 2003, Toth 1985), whereas flakes that were used as cores for removal of small flakes are extremely rare (Barsky et al. 2010, Delagnes and Roche
According to the location of the scars and absence of signs 131
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
2005, de la Torre 2004, de la Torre and Mora 2005, Hovers 2009a, Leakey 1971). Retouched artifacts are also rare in the Oldowan. Only a few roughly retouched flakes were reported in varying frequencies in most localities at Olduvai (de la Torre and Mora 2005, Leakey 1971) and Koobi Fora (Isaac and Harris 1997) and at Peninj (de la Torre et al. 2003), Ain Hanech (Sahnouni 2006), Melka Kunture (Chavaillon and Berthelet 2004, Chavaillon and Chavaillon 2004, Piperno et al. 2009), Atapuerca TD6 (Carbonell et al. 1999) and Dmanisi (de Lumley et al. 2005). In recent studies of the assemblages from Olduvai Beds I and II by de la Torre and Mora (2005), some of the items previously identified as retouched tools have been reclassified as natural pieces. In sites predating 2 Ma, worked or retouched flakes are extremely rare (Delagnes and Roche 2005, de la Torre 2004, Hovers 2009a, Kibunjia 1994, Semaw 2006).
Site FtJi1 (122) Senga 5 (355) Bizat Ruhama (Eocene chert – 86) Omo 123 (110) Omo 57 (45) ‹Ubeidiya III 37 (120) Koobi For a FxJj 50 (111) Ubeidiya K 26 (351) Koobi For a FxJj 20M (245) Evron Quarry (171) Ubeidiya K 28 (473) Bizat Ruhama (Brecciated chert – 80) Ain Hanech (194) Lokalalei 2C (500) Peninj ST Site Complex (~200) Gona OGS 7(76)
The use of an anvil in the Early Pleistocene sites was an important part of an adaptation system for exploiting lithic resources as well as for obtaining food. At Bizat Ruhama, for example, the anvil was used not only on all stages of lithic production, but probably also during the breakage of bones for marrow extraction, which was one of the major food processing activities at the site (Yeshurun et al. 2011). The anvil was used in other African and Levantine Lower Paleolithic sites for similar and other purposes (Goren-Inbar et al. 2002 and references therein).
Mean length Evidence of whole for bipolar flakes technique 16 + 19.85 + 20.3
+
20.8 22.2 22.54 23 24.85 25 25.2 25.81
+ +
26.4
+
+
37.9 38 40.49 44
Table 8.5. Average length of complete flakes from selected Early Pleistocene assemblages in the Levant and PlioPleistocene assemblages from Eastern and Central Africa. The number in parentheses is the total number of measured artifacts. The data for ‘Ubeidiya, Koobi Fora, Omo, Senga 5, FtJi1, Lokalalei 2C, Peninj ST, Ain Nanech B and Gona OGS 7 are from Bar-Yosef and Goren-Inbar 1993, Isaac and Harris 1997, de la Torre 2004, Harris et al. 1987, Merrick and Merrick 1976, Delagnes and Roche 2005, de la Torre et al. 2003, Sahnouni 2006, Semaw et al. 2003, respectively. The data for Senga 5 and FtJi1 include complete flakes and broken fragments.
The use of an anvil in lithic production is well documented for many Early Pleistocene sites (Barsky 2009, Barsky and de Lumley 2010, Barsky et al. 2011, de la Torre 2004, Diez-Martín et al. 2009, 2010, 2011, Gao 2000, Harris et al. 1987, Howell et al. 1987, Kuman and Field 2009, Leakey 1971, Ludwig and Harris 1998, Merrick and Merrick 1976, Pei 1938). Many of these sites (Sterkfontein, Fejej FJ-1a, Omo sites, Shungura Formation sites and Senga 5) exhibit a raw material acquisition pattern similar to that of Bizat Ruhama, in which the closest available outcrops containing small pebbles were exploited (Barsky 2009, de la Torre 2004, de la Torre and Mora 2005, Harris et al. 1987, Howell et al. 1987, Kuman and Field 2009, Ludwig and Harris 1998, Merrick and Merrick 1976). The reason for use of the bipolar technique as suggested in most of these studies is similar to that proposed for Bizat Ruhama, namely raw material constraints, particularly the use of small pebbles. Similarly to Bizat Ruhama, applying the bipolar technique to small pebbles resulted in small-sized artifacts. The metrical data presented in Table 8.5 indicate that in Early Pleistocene sites production of small artifacts is linked to the use of the bipolar technique. Whenever larger nodules of raw materials with good knapping angles were available, the bipolar technique was rarely used and larger artifacts were produced. The small artifacts of Bizat Ruhama, thought to be a unique feature, actually share a range of length values with those of many other Early Pleistocene sites in Africa and beyond, especially those in which the use of the bipolar
technique is documented. Nonetheless, although small pebbles were often used at other Early Pleistocene sites, systematic secondary knapping of flakes and use of the bipolar technique for the knapping of flakes are not documented. This raises the question of whether this particular lithic production system was first used by the Bizat Ruhama hominins. While systematic use of this system seems to be unique to the site, evidence for use of similar techniques of flake knapping may possibly be found among Early Pleistocene retouched artifacts. Many of the retouched flakes at Early Pleistocene sites were formed by large Clactonian notches or irregular abrupt retouch that often created a pointed extremity (see Bar-Yosef and GorenInbar 1993: figures 8; 10:3–4; 14:1–2; 24:1–2; 25:1–2; de la Torre and Mora 2005: figure 7.26; Isaac and Harris 1997: appendices 6AA, 6DD, 6HH; Piperno et al. 2009: figure 10.3:7–9; Sahnouni et al. 2002: figures 5:5–8; 6:7–8; Semaw et al. 2003: figure 4:1). At the site of Gombore 1 in Melka Kunture, Clactonian notches were frequently made on flakes and nodules of raw material (Chavaillon and Chavaillon 2004). In a recent study of the assemblages from 132
Chapter 8: Discussion
Olduvai Beds I and II (de la Torre and Mora 2005), most of the retouched pieces and awls were reclassified as simple flake fragments and chunks, and Semaw et al. (2009) have also suggested that most of the awls were unintentionally produced. At the Early Pleistocene site of Vallparadis, denticulate edges were interpreted as by-products of knapping on an anvil (Martínez et al. 2010). Some of the cases noted above may have resulted from sporadic use of the method identified at Bizat Ruhama.
products. The resulting products include pieces that resemble denticulates and Clactonian notches, as well as small, thin flakes and fragments. On the basis of experimental knapping and use-wear analysis, it was suggested that small, thin flakes and fragments were the desired end product of this knapping sequence and that the entire lithic production system reflects an adaptation to unfavorable raw material conditions in the Isernia La Pineta area (Crovetto et al. 1994, Longo et al. 1997, Peretto 1994).
Only in the Middle Pleistocene did the secondary knapping of flakes become more frequent. At some late Lower and Middle Paleolithic sites, flakes were used as cores for the removal of small flakes (Ashton 1992, 2007, Barkai et al. 2010, Dibble and McPherron 2006, 2007, Goren-Inbar 1988, Hovers 2007, Malinsky-Buller et al. 2011). In most of these cases the use of the bipolar technique is not documented. Intentional breakage of the flakes was sporadically practiced in late Acheulian and later sites in Europe and the Levant (Bergman et al. 1987, Bordes 1953, Mace 1959, Newcomer 1972, Shifroni and Ronen 2000). In some cases it was suggested that the flakes were broken on an anvil (Bergman et al. 1987, Bordes 1953). The intentional breakage of the flakes was often associated with tool production and resharpening, especially with burin manufacture (Bergman et al. 1987, Burdukiewicz 1981, Delarue and Vignard 1958, Mace 1959, Newcomer 1972). Some fragments were used without any additional modification (Bergman et al. 1987). Bergman et al. (1987) linked the intensive use of flake breakage at Hengistbury Head in southern England to the large distances from raw material sources and suggested that intentional breakage represents attempts to economize on raw materials.
The production of small flakes in the Early and Middle Pleistocene was thus a widespread phenomenon not restricted to any particular region or culture. At Bizat Ruhama the small size of the artifacts and the use of anvilsupported knapping techniques are interrelated features that reflect an adaptation to the form, size and knapping quality of the raw material. Going back to the example of the Middle Pleistocene site of Nahal Hesi, the fact that the Bizat Ruhama hominins used strictly local sources and invested little in transport of raw material is especially striking. The differences between the two sites point to higher mobility and more complex mental templates at Nahal Hesi and may lie in culturally imprinted habits of the hominins of the site. It is likely that cultural determinism at Nahal Hesi is the reason for the differences observed in patterns of raw material exploitation and lithic production strategies. The Acheulians of Nahal Hesi responded to constraints of raw material availability by transportation of prepared artifacts and by exploiting usually overlooked lithic resources, but still produced the same culturally determined forms. The best example of such a response is handaxes made on limestone. In contrast, the decisions that governed the choices of the Bizat Ruhama hominins do not have a clear, culturally restricted expression, visible to archaeologists. The technology at the site was most probably determined by functional needs and raw material constraints. Instead of looking, as the Acheulians did, for material to realize their mental template, the Bizat Ruhama hominins adapted their technology to the locally available raw materials.
Thus, the secondary knapping of flakes in most Middle Pleistocene and later cases reflects recycling and tool-kit maintenance (e.g. Ashton 2007, Bergman et al. 1987, Hovers 2007). For Bizat Ruhama it is unclear whether the reuse of flakes was part of the technological organization or the use of flakes as blanks for further reduction was a stage in the primary conceptual framework of the lithic production. The intensity of the secondary knapping seems to speak in favor of the latter.
The intensive and elaborate secondary knapping of the flakes at Bizat Ruhama reflects a longer production sequence and thus entails a more complex schéma opératoire than at other Oldowan sites, where flakes were only casually modified. Still, it seems that the lithic strategies at Bizat Ruhama reflect least-effort solutions (e.g. Braun 2012, Isaac and Harris 1997, Toth 1982) aimed at producing sharp flakes, and they seem to fit well within the now generally accepted perception that sharp edges were the major goal of Early Pleistocene lithic production systems (e.g. Braun and Hovers 2009, Braun et al. 2008, Delagnes and Roche 2005, Harmand 2009, Isaac 1986, 1997a, Isaac and Harris 1997, Roche 2005, Toth 1982, 1985, 1987, 1997). This inference from Early Pleistocene lithic assemblages is supported by use-wear analysis and ethnographic studies showing that flakes with sharp edges, regardless of their size, were frequently utilized by hunter-
Probably the closest parallel to Bizat Ruhama is the Middle Pleistocene site of Isernia La Pineta in Italy, where small flakes and fragments were produced during bipolar knapping of tabular flint nodules (Crovetto et al. 1994, Longo et al. 1997, Mussi 1995, Peretto 1994, Vergès and Ollé 2011). The raw materials in the Isernia La Pineta area are characterized by their unhomogeneous texture and often break during knapping. According to lithic replication experiments, this problem was overcome by using the bipolar technique (Crovetto et al. 1994, Peretto 1994). The nodules were placed with their longest and widest face on an anvil and knapped with little effort to control the shape of the knapping 133
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
gatherers in wide geographical and chronological ranges (Barkai et al. 2010, Hayden 1977, Keeley and Toth 1981, Longo et al. 1997, Peretto 1994, Strathern 1969, White 1967, White et al. 1977). In some cases, it was suggested that thin flakes smaller than 2 cm were systematically used (Barkai
et al. 2010, Dibble and McPherron 2006, Longo et al. 1997, Peretto 1994). Some of these studies propose that small, thin flakes were hand-held and used to cut meat (Barkai et al. 2010, Longo et al. 1997, Peretto 1994).
134
CHAPTER 9: CONCLUDING REMARKS The evidence that has accumulated during the last decade indicates that the earliest out-of-Africa dispersals were made by hominins possessing Oldowan-like technologies at ca. 1.8–1.7 Ma (de Lumley et al. 2005, Ferring et al. 2008, Zhu et al. 2004), preceding the first Acheulian assemblages in Africa (Asfaw et al. 1992, Roche et al. 2003, Semaw et al. 2008). Although the timing of the earliest dispersals and the technological affinities of the first Eurasians seem now to be established, the Eurasian Early Pleistocene record still suffers from a lack of sites in primary archaeological context. Consequently, many of the technological, behavioral, cognitive and paleoecological aspects of the earliest hominin occupations of Eurasia remain poorly understood. The results of the recent excavations at Bizat Ruhama provide relatively large lithic assemblages associated with faunal remains in primary archaeological context and thus allow study of some of the still unanswered questions of Eurasian Early Pleistocene archaeology. This book has focused on the presentation and interpretation of the lithic material, whereas parallel studies have presented other aspects of the site, such as paleoenvironments, paleoecology and subsistence (Mallol et al. 2011, Yeshurun et al. 2011).
scars on the edges belong to this debitage stage and are in fact “cores” from which thin flakes were detached. Thus, they should be classified as technological debris. The lithic strategies at Bizat Ruhama seem to be a response to raw material constraints. Experimental knapping indicated that the production of small-sized artifacts and the use of anvil-supported knapping techniques at the site comprise an adaptation to the shape, size and knapping quality of the raw material. Although the local pebbles undoubtedly imposed difficulties on the knappers, they did not attempt to overcome this obstacle by transporting raw materials from distances, as at the Acheulian site of Nahal Hesi. The absence of evidence for raw material transport and the selection of the best raw materials in terms of fracture mechanics despite their small size suggest that the Bizat Ruhama hominins applied “leasteffort” strategies (Isaac and Harris 1997). The use of an anvil at all stages of the debitage system and the post-detachment knapping of the flakes probably represented an adaptation to the raw materials in the site’s area. It is notable that, in addition to a skillful adaptation to the raw materials, the Bizat Ruhama hominins exploited an ecological niche different from those usually associated with Plio-Pleistocene sites. The hominins inhabited an interdune depression lacking evidence for a river or lake in the immediate surroundings during the occupation (Mallol et al. 2011). The faunal assemblage contains no woodland or riparian species and points to a homogeneous semiarid environment (Yeshurun et al. 2011). In other PlioPleistocene sites in the Levant and Africa, the preferred settings were on lake margins or riverbanks and the fauna are indicative of mosaic environments of woodlands, open areas and water bodies (e.g. Bar-Yosef 2006, Bar-Yosef and Goren-Inbar 1993, Bar-Yosef and Tchernov 1972, Belmaker 2006, 2009, Chavaillon and Berthelet 2004, Ditchfield et al. 1999, Feibel 2001, 2004, Feibel et al. 1989, 1991, GorenInbar et al. 2000, Guerin et al. 1993, Haas 1966, Horowitz 1996, Isaac 1997b, Leakey 1971, Leakey and Leakey 1978, Martínez-Navarro 2004, Plummer et al. 2009, Raynal et al. 2004, Sikes 1994, Tchernov et al. 1994). Bizat Ruhama thus demonstrates that, when equipped with a simple tool-kit, Early Pleistocene hominins were capable of adapting to a
One of the major objectives of this book was to define the goal of the lithic production system and to reconstruct the strategies that were applied to achieve this goal. Another objective was to assess the place of the Bizat Ruhama industry among Early Pleistocene taxonomic units. Based on the experimental knapping and the study of the archaeological material, I suggest that the lithic production system at Bizat Ruhama was directed toward production of sharp-edged flakes. Thin, sharp flakes were either produced directly during pebble reduction or involved a more complicated chaîne opératoire that included two additional stages: the deliberate selection of thick flakes and free-hand or anvil knapping of the selected flakes. According to the signs on their edges, about half of the flakes in the Bizat Ruhama assemblage were subjected to post-detachment knapping. The magnitude of this phenomenon leaves no doubt that the further knapping of the flakes was deliberate and that it played a fundamental role in the lithic production system. According to the available evidence, the Clactonian notches and most of the flakes that exhibit irregular “retouch-like” 135
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
broad spectrum of environments.
Early Pleistocene and present in the African continent, the Levant and possibly southern Europe and China. In this broad perspective, Bizat Ruhama seems to fit well within the Oldowan technocomplex. The Bizat Ruhama hominins did not employ bifacial shaping, possessed an Oldowanlike core reduction technology and exhibited knapping skills comparable to those of Oldowan hominins in Africa. Nonetheless, Bizat Ruhama differs from other Oldowan sites in the complexity of the lithic production system. The schéma opératoire reflected in the Bizat Ruhama lithic industry includes a stage of flake modification, rarely evident in other Plio-Pleistocene sites. The more elaborate schéma opératoire at Bizat Ruhama, however, does not necessarily imply that the mental templates of Bizat Ruhama hominins were more developed than those of their African counterparts. The knapping goals, as far as can be judged from the available data, were similar.
When attempting to assess the technological affinities of the Bizat Ruhama hominins it should first be noted that there have been many developments in Early Pleistocene research during the four decades since the first definitions of the Oldowan culture by Leakey (1967, 1971) or Mode 1 industries by Clark (1970). One of the most important of these developments is the increasing number of sites, which in turn has resulted in greater variability in the lithic assemblages. Another important change is in the methods of study and the classification of the industries. If Leakey’s typological approach were applied to Bizat Ruhama, it would fit the definition of neither the Oldowan nor the Acheulian. Leakey’s approach is probably still valuable for description and presentation of Olduvai Gorge assemblages, but it is not always applicable on a wider front, given the variability of the raw materials used and the fact that core-tools were cores for production of flakes rather than deliberately configured tools.
Finally, this study advocates the need for caution when identifying intentional retouch and challenges the common belief that Clactonian notches and pointed artifacts were intentionally designed tools. The current study supports previous suggestions that Clactonian notches may be sources of small flakes rather than purposefully shaped tools (Crovetto et al. 1994, Dibble and McPherron 2006, Peretto 1994). This is especially true for the Early Pleistocene sites in which the bipolar technique was frequently used.
More recent studies have broadened our understanding of the Oldowan and its technological, geographical and chronological boundaries have been extended by research in the last few decades (e.g. papers in Hovers and Braun 2009). The Oldowan is now considered to be a widespread technological phenomenon preceding the emergence of Acheulian technology, dated to the late Pliocene and
136
LIST OF REFERENCES Ahler S.A., 1989. Experimental knapping with KRF and mid-continental cherts: overview and applications. In: Amick D.S., Maudlin R.P. (Eds.), Experiments in Lithic Technology. BAR International Series 528, Oxford, pp. 199–234.
Approaches to Stone Tool Analysis. Cambridge Scholars Publishing, Newcastle, pp. 1–16.
al-Nahar M., Clark G. A., 2009. Lower Paleolithic in Jordan. Jordan Journal for History and Archaeology 3: 173–214.
Balout L., 1967. Procédés d’analyse et questions de terminologie dans l’étude des ensembles industriels du Paléolithique inférieur en Afrique du Nord. In: Bishop, W.W., Clark J.D. (Eds.), Background to Evolution in Africa. Chicago University Press, Chicago, pp. 707–736.
Ashton N.M., Dean P.D., Mcnabb J., 1991. Flaked flakes: what, where, when and why? Lithics 12: 1–11.
Andrefski W., 2005. Lithics: Macroscopic Approaches to Analysis. Cambridge Manuals in Archaeology, Cambridge University Press, Cambridge.
Barham L. S., 1987. The bipolar technique in Southern Africa: a replicative experiment. South African Archaeological Bulletin 42: 45–50.
Amick D.S., Mauldin R.P., 1997. Effects of raw materials on flake breakage patterns. Lithic Technology 22: 18–32. Amit R., Enzel Y., Crouvi O., Simhai O., Matmon A., Porat N., McDonald E., Gillespie A.R., 2011. The role of the Nile in initiating a massive dust influx to the Negev late in the middle Pleistocene. Geological Society of American Bulletin 123: 873–889.
Barkai R., Lemorini C., Gopher A., 2010. Palaeolithic cutlery 400 000–200 000 years ago: tiny meat-cutting tools from Qesem Cave, Israel. Antiquity 84: 325. Barzilai O., Malinsky-Buller A., Ackermann O., 2006. Kefar Menachem West: a Lower Palaeolithic site in the Shephela, Israel. Journal of the Israel Prehistoric Society 36: 7–38.
Anton S.C., Leonard W.R., Robertson M.L., 2002. An ecomorphological model of the initial hominid dispersal from Africa. Journal of Human Evolution 43: 773–785.
Bar-Yosef O., 1993. Site formation processes from the Levantine viewpoint. In Goldberg P., Nash D.T., Petraglia M.D. (Eds.), Formation Processes in Archaeological Context. Prehistory Press, Madison, Wisconsin, pp. 11–31.
Arzarello A., Marcolini F., Pavia G., Pavia M., Petronio C., Petrucci M., Rook L., Sardella S., 2006. Evidence of earliest human occupation in Europe: the 1 site of Pirro Nord (Southern Italy). Naturwissenschaften 93: 107–112.
Bar-Yosef O., 1994. The Lower Paleolithic of the Near East. Journal of World Prehistory 8: 211–265.
Asfaw B., Beyene Y., Suwa G., Walter R.C., White T.D., Wolde-Gabriel G., Yemane T., 1992. The earliest Acheulean from Konso-Gardula. Nature 360: 732–735.
Bar-Yosef O., 1998. Early colonization and cultural continuities in the Lower Paleolithic of Western Asia. In: Petraglia M.D., Korisettar R. (Eds.), Early Human Behavior in Global Context. Routledge, London and New York, pp. 221–280.
Ashton N.M., 1992. The High Lodge flint industries. In Ashton N.M., Cook J., Lewis S.G., Rose J. (Eds.), Excavations at High Lodge, G. de G. Sieveking 1962–1968, J. Cook 1988. British Museum Press, London, pp. 124–163.
Bar-Yosef O., 2006. The known and unknown about the Acheulian. In Goren-Inbar N., Sharon G., (Eds.), Axe Age, Acheulian Toolmaking from Quarry to Discard. Equinox, London, pp. 479–494.
Ashton N.M., 2007. Flakes, cores, flexibility and obsession: situational behaviour in the British Lower Palaeolithic. In: McPheron S. (Ed.), Tools versus Cores: Alternative 137
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Bar-Yosef O., Tchernov E., 1972. On the Palaeo-ecological History of the Site of ‘Ubeidiya, Jerusalem: Israel Academy of Sciences.
and Interpretations. Springer, New-York, pp. 211–227. Bergman C.A., Barton R.N.E., Collcutt S.N., Morris G., 1987. Intentional breakage in a Late Upper Palaeolithic assemblage from southern England. In: Sieveking, G. de G., Newcomer M.H. (Eds.), The Human Uses of Flint and Chert, Proceedings of the 4th International Flint Symposium. Cambridge University Press, Cambridge, pp. 21–32.
Bar-Yosef O., Goren-Inbar N., 1993. The Lithic Assemblages of ‘Ubeidiya. Qedem 34. Institute of Archaeology, Hebrew University of Jerusalem. Bar-Yosef O., Belmaker M., 2010. Early and Middle Pleistocene faunal and hominin dispersals through Southwestern Asia. Quaternary Science Reviews 30: 1318– 1337.
Beyene Y., Katoh S., WoldeGabriel G., Hart W.K., Uto K., Sudo M., Kondo M., Hyodo M., Renne P.R., Suwa G., Asfaw B., 2013. The characteristics and chronology of the earliest Acheulian at Konso, Ethiopia. PNAS 110: 1584–1591.
Bar-Yosef Y., 1964. Geology of Ahuzam – Nir’am Area. TAHAL, Tel-Aviv (in Hebrew).
Binford L.R., 1981. Bones, Ancient Men and Modern Myths. Academic Press, New York.
Barsky D., 2009. An overview of some African and Eurasian Oldowan sites: evaluation of hominin cognitive levels, technological advancement and adaptive skills. In: Hovers E., Braun D.R. (Eds.), Interdisciplinary Approaches to the Oldowan. Springer Press, Dordrecht, pp. 39–47.
Binford L., 1987. Searching for camps and missing the evidence? Another look at the Lower Paleolithic. In: Soffer O. (Ed.), The Pleistocene Old World. Plenum Press, New York, pp. 17–31.
Barsky D., de Lumley H., 2010. Early European Mode 2 and the stone industry from the Caune de l’Arago’s archeostratigraphical levels ‘‘P’’. Quaternary International 223–224: 71–86.
Binford L.R., Quimby G.L., 1963. Indian sites and chipped stone materials in the Northern Lake Michigan area. Fieldiana – Anthropology 36, 12: 277–307. Blumenschine R.J., Peters C., Madao F.T., Clarke R.J., Deino A.L., Hay R.L., Swisher C.C., Stanistreet I.G., Ashley G.M., McHenry L.J., Sikes N.E., van der Merwe N.J., Tactikos J.C., Cushing A.E., Deocampo D.M., Njau J.K., Ebert J.I., 2003. Late Pliocene Homo and hominid land use from Western Olduvai Gorge, Tanzania. Science 299: 1217–1221.
Barsky D., Celiberti V., Cauche D., Grégoire S., Lebégue F., de Lumley H., Toro-Moyano I., 2010. Raw material discernment and technological aspects of the Barranco León and Fuente Nueva 3 stone assemblages (Orce southern Spain). Quaternary International 223–224: 201–219. Barsky D., Chapon-Sao C., Bahain J.J., Beyene Y., Cauche D., Celiberti V., Desclaux E., de Lumley H., de Lumley M.A., Marchal F., Moullé P.E., Pleurdeau D., 2011. The Early Oldowan stone-tool assemblage from Fejej FJ-1a, Ethiopia. Journal of African Archaeology 9: 207–224.
Boëda E., 1986. Approche technologique du concept Levallois et évaluation de son champ d’application à travers trois gisements saaliens et weischeliens de la France septentrionale. Ph.D. thesis, Université de Paris X. Bordes F., 1953. Notules de typologie paléolithique: i. Outils moustériens à fracture volontaire. Bulletin de la Société Préhistorique Française L: 224–226.
Behrensmeyer A.K., 1978. Taphonomic and ecological information from bone weathering. Paleobiology 4: 150– 162.
Bordes F., 1961. Typologie du Paléolithique Ancien et Moyen. Bordeaux: Institut de Préhistoire de l’Université de Bordeaux.
Belfer-Cohen A., Goren-Inbar N., 1994. Cognition and communication in the Levantine Lower Paleolithic. World Archaeology 26 : 144–157. Belmaker M., 2006. Community Structure through Time: ‘Ubeidiya, a Lower Pleistocene Site as a Case Study. Unpublished Ph.D. dissertation, Hebrew University of Jerusalem.
Braun D., 2012. What does Oldowan technology represent in terms of hominin behavior? In: Dominguez-Rodrigo M. (Ed.), Stone Tools and Fossil Bones: Debates in the Archaeology of Human Origins. Cambridge University Press, Cambridge, pp. 222–244.
Belmaker M., 2009. Hominin adaptability and patterns of faunal turnover in the Early to Middle Pleistocene transition in the Levant. In: Camps M., Chauhan P. (Eds.), Sourcebook of Paleolithic Transitions: Methods, Theories
Braun D.R., Hovers E., 2009. Introduction: current issues in Oldowan research. In: Hovers E., Braun D.R. (Eds.), Interdisciplinary Approaches to the Oldowan. Vertebrate Paleobiology and Paleoanthropology Series. Springer, 138
List of references
Dordrecht, pp. 1–14.
Burdukiewicz J. M., Ronen A., 2003a. Research problems of the Lower and Middle Paleolithic small tool assemblages. In: Burdukiewicz J.M., Ronen A. (Eds.), Lower Palaeolithic Small Tools in Europe and the Levant. BAR International Series 1115, Oxford, pp. 235–238.
Braun D.R., Rogers M.J., Harris J.W.K., Walker S.J., 2008. Landscape-scale variation in hominin tool use: evidence from the Developed Oldowan. Journal of Human Evolution 55: 1053–1063.
Burdukiewicz J.M., Ronen A., 2003b. Lower Palaeolithic small tools in Europe and the Levant. BAR International Series 1115, Oxford.
Braun D.R., Plummer T.W., Ditchfield P.W., Bishop L.C., Ferraro J.V., 2009. Oldowan technology and raw material variability at Kanjera South. In: Hovers E., Braun D.R. (Eds.), Interdisciplinary Approaches to the Oldowan. Springer, Dordrecht, pp. 99–110.
Carbonell E., García Antón M.D., Mallol C., Mosquera M., Ollé A., Rodríguez Álvarez X.-P., Sahnouini M., Sala R., Vergés J.M., 1999. The TD6 level lithic industry from Gran Dolina, Atapuerca (Burgos, Spain): production and use. Journal of Human Evolution 37: 653–693.
Bruins H.J., Yaalon D.H., 1979. Stratigraphy of the Netivot section in the desert loess of the Negev (Israel). Acta Geologica Academiae Scientarium Hungaricae 22: 161–169.
Carbonell E., Bermúdez de Castro J.M., Parés J.M., Pérez González A., Cuenca-Bescós G., Ollé A., Mosquera M., Huguet R., van der Made J., Rosas A., Sala R., Vallverdú J., García N., Granger D.E., Martinón-Torres M., Rodríguez X.P., Stock G.M., Vergès J.M., Allué E., Burjachs F., Cáceres I., Canals A., Benito A., Díez C., Lozano M., Mateos A., Navazo M., Rodríguez J., Rosell J., Arsuaga J.L., 2008. The first hominin of Europe. Nature 452: 465–469.
Brunnacker K., Ronen A., Tillmanns W., 1982. Die jungpleistozanen Aolianit in der sudlichen Kustenzone von Israel; Ein Beitrag zur zeitlichraumlichen Klimaentwicklung. Eiszeitalter u. Gegenwart 32, 23-48. Buhksianidze M., 2005. The fossil Bovidae of Dmanisi. P.hD. Dissertation, University of Ferrara. Bunn H.T., 1981. Archaeological evidence for meat-eating by Plio-Pleistocene hominids from Koobi Fora and Olduvai Gorge. Nature 291: 574-577.
Carbonell E., Sala R., Barsky D., Celiberti V., 2009. From homogeneity to multiplicity: a new approach to the study of archaic stone tools. In: Hovers E., Braun D.R. (Eds.), Interdisciplinary Approaches to the Oldowan. Springer Press, Dordrecht, pp. 25–37.
Bunn H.T., 1983. Evidence on the diet and subsistence patterns of Plio-Pleistocene hominids at Koobi Fora, Kenya, and Olduvai Gorge, Tanzania. In: Clutton-Brock J., Grigson C. (Eds.), Animals and Archaeology 1: Hunters and Their Prey. British Archaeological Report, Oxford, pp. 21-30.
Chavaillon J., Berthelet A., 2004. The archaeological sites of Melka Kunture. In: Chavaillon J., Piperno M. (Eds.), Studies on the Early Paleolithic site of Melka Kunture, Ethiopia, Origines, Istituto Italiano di Preistoria e Protostoria, pp. 25–80.
Bunn H.T., 1986. Patterns of skeletal representation and hominid subsistence activities at Olduvai Gorge, Tanzania, and Koobi Fora, Kenya. Journal of Human Evolution 15: 673–690.
Chavaillon J., Chavaillon N., 2004. The site of Gombore I. Comments and conclusions on the lithic assemblage. In: Chavaillon J., Piperno M. (Eds.), Studies on the Early Paleolithic site of Melka Kunture, Ethiopia. Origines, Istituto Italiano di Preistoria e Protostoria, pp. 437–448.
Burdukiewicz J.M., 1981. The flint technology of the Hamburgian Culture, Olbrachcice, Poland. Third International Symposium on Flint, Maastricht, pp. 67–70.
Chazan M., 2000. Typological analysis of the Lower Paleolithic site of Holon, Israel. Journal of the Israel Prehistoric Society 30: 7–33.
Burdukiewicz J.M., 2003. Lower Paleolithic sites with small artifacts in Poland. In: Burdukiewicz J.M., Ronen A. (Eds.), Lower Palaeolithic Small Tools in Europe and the Levant. BAR International Series 1115, Oxford, pp. 65–93. Burdukiewicz J.M., Śnieszko Z., Winnicki J., 1994. A Lower Palaeolithic settlement at Trzebnica (S.W. Poland). Ethnographisch-Archäologische Zeitschrift 34: 27–40. Burdukiewicz J.M., Ronen A., 2000. Ruhama in the Northern Negev Desert. A new microlithic site of the Lower Palaeolithic in Israel. Praehistoria Thuringica 5: 32–46. 139
Chazan M., Kolska Horwitz L., 2007. Holon: A Lower Paleolithic Site in Israel. American School of Prehistoric Research 50. Peabody Museum of Archaeology and Ethnology, Harvard University, Cambridge. Chauhan P.R., 2009. Early Homo occupations near the “Gate of Tears”: examining the paleoanthropological records of Djibouti and Yemen. In: Hovers, E., Braun, D.R. (Eds.), Interdisciplinary Approaches to the Oldowan. Springer,
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Dordrecht, pp. 49–59.
dunes of the Sharon. Geoderma 2: 95–120.
Clark G., 1970. Aspects of Prehistory. University of California Press, Berkeley.
Dan Y., Raz Z., Yaalon D.H., Koyumdjisky H., 1975. Soil Map of Israel. Ministry of Agriculture, Israel.
Clark J.D., 1969a. The Kalambo Falls Prehistoric Site, Volume I. Cambridge University Press, Cambridge.
Dassa M., 2002. Paleosols of Southern Coastal Plain: Climatic and Environmental Changes of Late Quaternary. Unpublished MA Thesis, Bar-Ilan University (in Hebrew).
Clark J.D., 1969b. The Middle Acheulian occupation site at Latamne, northern Syria. Quaternaria 10: 1–73.
Delagnes A., Roche H., 2005. Late Pliocene hominid knapping skills: the case of Lokalalei 2C, West Turkana, Kenya. Journal of Human Evolution 48: 435–472.
Clark J.D., Cole G.H., Isaac G.L., Kleindienst M.R., 1966. Precision and definition in African Archaeology. South African Archaeological Bulletin XXI: 114–121.
Delagnes A., Lenoble A., Harmand S., Brugal J.-P., Prat S., Tiercelin J.-J., Roche H., 2006. Interpreting pachyderm single carcass sites in the African Lower and Early Middle Pleistocene record: a multidisciplinary approach to the site of Nadung’a 4 (Kenya). Journal of Anthropological Archaeology 25: 448–465.
Copeland L., 1983. The Paleolithic stone industries. In D. Roe (ed.), Adlun in the Stone Age. The excavations of D. A. E. Garrod in the Lebanon 1958-1963. BAR International Series 159, Oxford, pp. 89–365. Copeland L., 1991. The Late Acheulian knapping – floor at C – spring, Azraq Oasis, Jordan. Levant 23: 1–6.
Delarue R., Vignard E., 1958. Intention et fractures moustériennes sectionnant des racloirs. Bulletin de la Société Préhistorique Française LI: pp. 29–31.
Copeland L., Hours F., 1989. The Hammer on the Rock: Studies in the Early Palaeolithic of Azraq, Jordan. BAR International Series 540ii, Oxford.
de la Torre I., 2004. Omo revisited: evaluating the technological skills of Pliocene hominids. Current Anthropology 45: 439–465.
Cotterell B., Kamminga J., 1979. The mechanics of flaking. In: Hayden B. (Ed.), Lithic Use-wear Analysis. Academic Press, New York, pp. 97–112.
de la Torre I., Mora R., Domínguez-Rodrigo M., de Luque L., Alcala L., 2003. The Oldowan industry of Peninj and its bearing on the reconstruction of the technological skills of Lower Pleistocene hominids. Journal of Human Evolution 44: 203–224.
Cotterell B., Kamminga J., 1987. The formation of flakes. American Antiquity 52(4): 675–708. Coutts P.J.F., 1977. Green timber and Polynesian adzes and axes: an experimental approach. In: Wright R.V.S. (Ed.), Stone Tools as Cultural Markers: Change, Evolution and Complexity. Australian Institute of Aboriginal Studies, Canberra, pp. 67–82.
de la Torre I., Mora R., 2005. Technological Strategies in the Lower Pleistocene at Olduvai Beds I & II. ERAUL 112, Liège. de la Torre I., Mora R., 2009. The technology of the ST site complex. In: Domínguez-Rodrigo M., Alcalá L., Luque L. (Eds.), Peninj: A Research Project on Human Origins (1995– 2005). Oxbow Books, Oxford and Oakville, pp. 145–189.
Crabtree D.E., 1972. An Introduction to Flintworking. Occasional Papers of the Idaho State University Museum no. 28, Pocatello.
de Lumley H., Nioradzé M., Barsky D., Cauche D., Celiberti V., Nioradzé G., Notter O., Zvania D., Lordkipanidze D., 2005. Les industries lithiques préoldowayennes du début du Pléistocène inférieur du site de Dmanissi en Géorgie. L’Anthropologie 109: 1–182.
Crouvi O., Amit R., Enzel Y., Gillespie A. R., 2010. The role of active sand seas in the formation of desert loess. Quaternary Science Reviews 29: 2087–2098. Crovetto C., Ferrari M., Peretto C., Longo L., Vianello F., 1994. The carinated denticulates from the Paleolithic site of Isernia La Pineta (Molise, Central Italy): tools or flaking waste? The results of the 1993 lithic experiments. Human Evolution 9: 175–207.
Dennel R., 2007. “Resource-rich, stone-poor”: early hominin land use in large river systems of northern India and Pakistan. In: Petraglia M.D., Allchin B. (Eds.), The Evolution and History of Human Populations in South Asia. Springer Press, Dordrecht, pp. 41–68.
Dan Y., Yaalon D.H., Koyumdjisky H., 1969. Catenary soil relationships in Israel, 1. The Netanya catena on coastal
Dennell R., 2009. The Paleolithic Settlement of Asia. 140
List of references
Cambridge University Press, Cambridge.
Palmqvist P., 2012. On the limits of using combined U-series/ ESR method to date fossil teeth from two Early Pleistocene archaeological sites of the Orce area (Guadix-Baza basin, Spain). Quaternary Research 77: 482–491.
Derevianko A.P., 2009. The earliest human migrations in Eurasia in the Early Paleolithic. Institute of Archaeology and Ethnography Press, Novosibirsk.
Eisenmann V., 2006. Pliocene and Pleistocene equids: paleontology versus molecular biology. In: Kahlke R.-D., Maul L.C., Mazza P. (Eds.), Late Neogene and Quaternary Biodiversity and Evolution: Regional Developments and Interregional Correlations. Proceedings volume of the 18th International Senckenberg Conference (VI International Palaeontological Colloquium in Weimar), 25th–20th April 2004. Courier Forschungsinstitut Senckenberg (CFS), 256, pp. 71–89.
Derevianko A.P., Petrin V.T., Taimagambetov J.K., 2000. The phenomenon of microindustrial complexes in Eurasia. Archaeology, Ethnology & Anthropology of Eurasia 4: 2–18. Despriée J., Voinchet P., Tissoux H., Moncel M.-H., Arzarello M., Robin S., Bahain J.-J., Falguères C., Courcimault G., Dépont J., Gageonnet R., Marquer L., Messager E., Abdessadok S., Puaud S., 2010. Lower and Middle Pleistocene human settlements in the Middle Loire River Basin, Centre Region, France. Quaternary International 223–224: 345–359.
Féblot-Augustins J., 1997. La circulation des matières premières au Paléolithique. ERAUL 75. Université de Liège, Liège.
Dibble H.L., McPherron S., 2006. The missing Mousterian. Current Anthropology 47: 777–803.
Feibel C.S., 2001. Archaeological sediments in lake margin environments. In: Stein J., Farrand W.R. (Eds.), Sediments in Archaeological Context. University of Utah Press, Salt Lake City, pp. 127–148.
Dibble H.L., McPherrron S., 2007. Truncated-faceted pieces: hafting modification, retouch, or cores? In: McPherron S. (Ed.), Cores or Tools? Alternative Approaches to Stone Tool Analysis. Cambridge Scholars Publishing, Newcastle, pp. 75–90.
Feibel C.S., 2004. Quaternary lake margins of the Levant rift valley. In: Goren-Inbar N., Speth J.D. (Eds.), Human Paleoecology in the Levantine Corridor. Oxbow Books, Oxford, pp. 21–36.
Diez-Martín F., Sánchez P., Dominguez-Rodrigo M., Mabulla A., Barba R., 2009. Were Olduvai hominins making butchering tools or battering tools? Analysis of a recently excavated lithic assemblage from BK (Bed II, Olduvai Gorge, Tanzania). Journal of Anthropological Archaeology 28: 274–289.
Feibel C.S., Brown F.H., McDougall I., 1989. Stratigraphic context of fossil hominids from the Omo group deposits: northern Turkana Basin, Kenya and Ethiopia. American Journal of Physical Anthropology 78: 595–622. Feibel C.S., Harris J.M., Brown F.H., 1991. Paleoenvironmental context for the Late Neogene of the Turkana Basin. In: Harris J.M. (Ed.), Koobi Fora Research Project, vol. 3, The Fossil Ungulates: Geology, Fossil Artiodactyls, and Palaeoenvironments, Clarendon Press, Oxford, pp. 321–370.
Diez-Martín F., Sánchez P., Dominguez-Rodrigo M., Mabulla A., Bunn H., Ashley G., Barba R., Baquedano E., 2010. New insights into hominin lithic activities at FLK North Bed I, Olduvai Gorge, Tanzania. Quaternary Research 74: 376–387. Diez-Martín F., Yustos P.S., Domínguez-Rodrigo M., Prendergast M.E., 2011. An experimental study of bipolar and freehand knapping of Naibor Soit quartz from Olduvai Gorge (Tanzania). American Antiquity 76: 690–708.
Ferring R., Lordkipanidze D., Berna F., Oms O., 2008. Geology and formation processes at Dmanisi in the Georgian Caucasus. Abstracts of the 73rd Annual Meeting of the Society of American Archaeology, Vancouver, British Columbia, p. 195.
Ditchfield P., Hicks J., Plummer T., Bishop L.C., Potts R., 1999. Current research on the Late Pliocene and Pleistocene deposits north of Homa Mountain, southwestern Kenya. Journal of Human Evolution 36: 123–150.
Flenniken J.J., 1981. Replicative Systems Analysis: A Model Applied to the Vein Quartz Artifacts from the Hoko River Site. Washington State University Laboratory of Anthropology Reports of Investigations No. 59, Pullman.
Dobkins J.E., Folk R.L., 1970. Shape development on TahitiNui, Journal of Sedimentary Petrology 40: 1167–1203.
Flenniken J.J., Haggarty J.C., 1979. Trampling as an agency in the formation of edge damage: an experiment in lithic technology. Northwest Anthropological Research Notes 13: 208–214.
Duval M., Falguères C., Bahain J.-J., Grün R., Shao Q., Aubert M., Dolo J.-M., Agustí J., Martínez-Navarro B., 141
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Gabunia L., Vekua A., 1995. A Plio-Pleistocene hominid from Dmanisi, East Georgia, Caucasus. Nature 373: 509– 512.
Goren-Inbar N., 1985. The lithic assemblage of the Berekhat Ram Acheulian site, Golan Heights. Paléorient 11: 7–28. Goren-Inbar N., 1988. Too small to be true? Reevaluation of cores on flakes in Levantine Mousterian assemblages. Lithic Technology 17: 37–44.
Gabunia L., Vekua A., Lordkipanidze D., Swisher C.C., Ferring R., Justus A., Nioradze M., Tvalchrelidze M., Antón S.C., Bosinski G., Jöris O., de Lumley M.-A., Majsuradze G., Mouskhelishvili A., 2000. Earliest Pleistocene hominid cranial remains from Dmanisi, Republic of Georgia: taxonomy, geological setting, and age. Science 288: 1019– 1025.
Goren-Inbar N., 1990. The lithic assemblages. In: GorenInbar N. (Ed.), Quneitra: A Mousterian Site on the Golan Heights. Qedem 31. Institute of Archaeology, Hebrew University of Jerusalem, pp. 61–167.
Gao X., 2000. Core reduction at Zhoukoudian locality 15. Archaeology, Ethnology & Anthropology of Eurasia 3: 2–12.
Goren-Inbar N., 1995. The Lower Paleolithic of Israel. In: Levy I. (Ed.), The Archaeology of Society in the Holy Land. Leicester University Press, London, pp. 93–109.
Garrod D.A.E., Bate D.M., 1937. The Stone Age of Mount Carmel. Clarendon Press, Oxford.
Goren-Inbar N., Feibel C.S., Verosub K.L., Melamed Y., Kislev M.E., Tchernov E., Saragusti I., 2000. Pleistocene milestones on the out-of-Africa corridor at Gesher Benot Ya‘aqov, Israel. Science 289 : 944–974.
Geneste J.-M., 1985. Analyse lithique d’industries moustériennes du Perigord: une approche technologique des comportements des groupes humains au Paléolithique moyen. Thèse doctorat, Université de Bordeaux.
Goren-Inbar N., Sharon G., Melamed Y., Kislev M., 2002. Nuts, nut cracking, and pitted stones at Gesher Benot Ya‘aqov, Israel. PNAS 99: 2455–2460.
Gerson R., Amit R., 1987. Rates and modes of dust accretion and deposition in an arid region – the Negev, Israel. In: Frostick L.K., Reid I. (Eds.), Desert Sediments: Ancient and Modern. The Geological Society of London, Special Publication no. 35, pp. 157–169.
Gould R.A., Koster D.A., Sontz A.H.L., 1971. The lithic assemblage of the western desert of Australia. American Antiquity 36: 149–69. Gowlett J.A.J., 1988. A case of Developed Oldowan in the Acheulean? World Archaeology 20, 1: 13–26.
Gifford-Gonzalez D.P., Damrosch D.B., Damrosch D.R., Pryor J., Thunen R.L., 1985. The third dimension in site structure: an experiment in trampling and vertical dispersal. American Antiquity 50: 803–818.
Gvirtzman G., 1970. The Saqiye group (Late Eocene to Early Pleistocene) in the coastal plain and Hashefela regions, Israel. Unpublished PhD. dissertation, Department of Geology, Hebrew University of Jerusalem (in Hebrew).
Gilead D., Israel, M., 1975. An Early Palaeolithic site at Kefar Menahem. Tel Aviv 2: 1–12.
Gvirtzman G., 1990. The geology and geomorphology of the Sharon and its Mediterranean shelf. In: Dagni A., Grossman D., Shmueli A. (Eds.), The Sharon between the Yarkon and the Carmel. Tel Aviv University, Tel Aviv, pp. 19–60 (in Hebrew).
Gilead D., Ronen A., 1977. The Acheulean industries of Evron, Western Galilee Coastal Plain, Israel. Eretz Israel 13: 56–86. Goldman T., 2004. Raw material selection patterns in A.L., 894, Hadar, Ethiopia. Unpublished M.A. thesis, Hebrew University of Jerusalem.
Gvirtzman G., Buchbinder B., 1969. Outcrops of Neogene Formation in the Central and Southern Coastal Plain, Hashphela and Be’er Sheva Regions. Israel Geological Survey Bulletin 50, Jerusalem.
Goldman-Neuman T., Hovers E., 2009. Methodological issues in the study of Oldowan raw material selectivity: insights from A. L. 894 (Hadar, Ethiopia). In: Hovers E., Braun D.R. (Eds.), Interdisciplinary Approaches to the Oldowan. Springer, Dordrecht, pp. 71–84.
Haas G., 1966. On the Vertebrate Fauna of the Lower Pleistocene Site of ‘Ubeidiya. Israel Academy of Sciences and Humanities, Jerusalem.
Goldman-Neuman T., Hovers E., 2012. Raw material selectivity in Late Pliocene Oldowan sites in the Makaamitalu Basin, Hadar, Ethiopia. Journal of Human Evolution 62: 353–66.
Harding P., Gibbard P.L., Lewin J., Macklin M.G., Moss E.H., 1987. The transport and abrasion of flint handaxes in a gravel-bed river. In: Sieveking G. de G., Newcomer M.H. (Eds.), The Human Use of Flint and Chert. Cambridge 142
List of references
Hovers E., 2009b. The Lithic Assemblages of Qafzeh Cave. Oxford University Press.
University Press, Cambridge, pp. 115–126. Harmand S., 2009. Variability of raw material selectivity and technoeconomic behavior in the early Oldowan: evidence from the Late Pliocene sites of Lokalalei. In: Hovers E., Braun D.R. (Eds.), Interdisciplinary Approaches to the Oldowan. Springer, Dordrecth, pp. 85–97.
Hovers E., Braun D.R., 2009. Interdisciplinary Approaches to the Oldowan. Vertebrate Paleobiology and Paleoanthropology Series. Springer, Dordrecht. Hours F., 1981. Le Paléolithique inférieur de la Syrie et du Liban. Le point de la question en 1980. In Sanlaville, P., and Cauvin, J. (Eds.), Préhistoire du Levant. Maison de l’Orient, Lyon, pp. 165–184.
Harris J.W.K., Isaac G.L., 1997. Sites in the Upper KBS, Okote, and Chari Members. In: Isaac G.L. (Ed.), Koobi Fora Research Project, vol. 5: Plio-Pleistocene Archaeology. Clarendon Press, Oxford, pp. 115–220.
Howell F.C., Haesaerts P., de Heinzelin J., 1987. Depositional environments, archaeological occurrences and hominids from Member E and F of the Shungura Formation (Omo Basin, Ethiopia). Journal of Human Evolution 16: 665–700.
Harris J.W.K., Williamson P.G., Verniers J., Tappen M.J., Stewart K., Helgren D., de Heinzelin J., Boaz N.T., Bellomo R.V., 1987. Late Pliocene hominid occupation of the Senga 5A site, Zaire. Journal of Human Evolution 16: 701–728.
Inizan M.-L., Reduron-Ballinger M., Roche H., Tixier J., 1999. Technology and Terminology of Knapped Stone. Préhistoire de la Pierre Taillée 5. Cercle de Recherches et d’Etudes Préhistoriques, Paris.
Hay R.L., 1976. Geology of the Olduvai Gorge. University of California Press, Berkeley. Hayden B., 1977. Stone tool functions in the Western Desert. In: Wright R.V.S. (Ed.), Stone tools as cultural markers: change, evolution, complexity. Australian Institute of Aboriginal Studies, Canberra, pp. 178–188.
Isaac G.L., 1972. Early phases of human behaviour: models in Lower Palaeolithic archaeology. In: Clark D.L. (Ed.), Models in Archaeology. Methuen, London, pp. 167–199.
Horowitz A., 1974. The Late Cenozoic Stratigraphy and Paleogeography of Israel. Institute of Archaeology, Tel Aviv University.
Isaac G.L., 1977. Olorgesailie. The University of Chicago Press, Chicago and London. Isaac G.L., 1978. The food-sharing behavior of protohuman hominins. Scientific American 238: 90–106.
Horowitz A., 1979. The Quaternary of Israel. Academic Press, New York.
Isaac G.L., 1986. Foundation stones: early artefacts as indicators of activities and abilities. In: Bailey G.N., Callow P. (Eds.), Stone Age Prehistory. Cambridge University Press, Cambridge, pp. 221–241.
Horowitz A., 1996. Review of Lower Paleolithic site locations in Israel, possibly controlled by deposition and erosion processes. Israel Journal Earth Science 45: 137–145. Hovers E., 2003. Treading carefully: site formation processes and Pliocene lithic technology. In: Martínez-Moreno J., Torcal R.M., de la Torre I. (Eds.), Oldowan: Rather More than Smashing Stones. An Introduction to “The Technology of First Humans” Workshop. Universitat Autonoma de Barcelona, Barcelona, pp. 145–163.
Isaac G.L., 1997a. Introduction. In: Isaac G.L. (Ed.), Koobi Fora Research Project, Volume 5: Plio-Pleistocene Archaeology. Clarendon Press, Oxford, pp. 1–11.
Hovers E., 2007. The many faces of cores-on-flakes: a perspective from the Levantine Mousterian. In: McPherron S.P. (Ed.), Cores or Tools? Alternative Approaches to Stone Tool Analysis. Cambridge Scholars Press, Cambridge, pp. 42–74.
Isaac G.L., Harris J.W.K., 1978. Archaeology. In Leakey M.G., Leakey R.E. (Eds.), Koobi Fora Research Project, Volume 1: The Fossil Hominids and Introduction to their Context, 1968–1974. Clarendon Press, Oxford, pp. 64–85.
Isaac G.L., 1997b. Koobi Fora Research Project, Volume 5: Plio-Pleistocene Archaeology. Clarendon Press, Oxford.
Isaac G.L., Crader D.C., 1981. To what extent were early hominids carnivorous? An archaeological perspective. In: Harding S.O., Teleki G. (Eds.), Omnivorous Primates. Columbia University Press, New York, pp. 37–103.
Hovers E., 2009a. Learning from mistakes: flaking accidents and knapping skills in the assemblage of A.L. 894 (Hadar, Ethiopia). In: Schick K., Toth N. (Eds.), The Cutting Edge: New Approaches to the Archaeology of Human Origins. Stone Age Institute Publication Series 1, Gosport and Bloomongton, pp. 137–148.
Isaac G.L., Harris J.W.K., 1997. The stone artefact assemblages: a comparative study. In: Isaac G.L. (Ed.), 143
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Koobi Fora Research Project, Volume 5: Plio-Pleistocene Archaeology. Clarendon Press, Oxford, pp. 262–362.
Kellerhals R., Bray D., 1971. Sampling procedures for coarse fluvial sediments. Journal of the Hydraulic Division 97, HY 8: 1165–1179.
Isaac G.L., Harris J.W.K., Kaufuli Z.M., Schick K.D., 1997. Application of the observations and experiments to the Koobi Fora cases. In: Isaac G.L. (Ed.), Koobi Fora Research Project, Volume 5: Plio-Pleistocene Archaeology. Clarendon Press, Oxford, pp. 256–261.
Kibunjia M., 1994. Pliocene archaeological occurrences in the Lake Turkana basin. Journal of Human Evolution 27: 159–171. Kimbel W.H., Walter R.C., Johanson D.C., Reed K.E., Aronson J.L., Assefa Z., Marean C.W., Eck G.G., Bobe R., Hovers E., Rak Y., Vondra C., Yemane T., York D., Chen Y., Evenson N.M., Smith P.E., 1996. Late Pliocene Homo and Oldowan tools from the Hadar Formation (Kada Hadar Member), Ethiopia. Journal of Human Evolution 31: 549– 561.
Issar A., 1961. Geology of sub-terranean water horizons of the Shephela and of the Sharon Regions. Unpublished Ph.D. dissertation, Hebrew University of Jerusalem. Itzhaki Y., 1961. Pleistocene shorelines in the coastal plain of Israel. Bulletin of Geological Survey of Israel 32: 1–9. Jagher R., 2011. Nadaouiyeh Aïn Askar – Acheulian variability in the Central Syrian Desert. In: LeTensorer, J.-M., Jagher R., Otte M. (Eds.), The Lower and Middle Palaeolithic in the Middle East and Neighboring Regions. Etudes et Recherches Archéologiques de l’Université de Liège (ERAUL), pp. 209–224.
Kleindienst M.R., 1961. Variability within the Late Acheulian assemblage in East Africa. South African Archaeological Bulletin 16: 35-52. Kleindienst M.R., 1962. Components of the East African Acheulian assemblage: an analytic approach. In: Mortelmans G., Nenquin J. (Eds.), Actes du IV Congrès Panafricain de Préhistoire et de l’Etude du Quaternaire, Leopoldville, 1959. Belgie Annalen, Musée Royal de l´Afrique Centrale, Tervuren, pp. 81–108.
Jelinek A.J., 1977. A preliminary study of flakes from the Tabun Cave, Mount Carmel. In: Arensburg A., Bar-Yosef O. (Eds.), Eretz-Israel. The Israel Exploration Society, Jerusalem, pp. 87–96.
Kobayashi H., 1975. The experimental study of bipolar flakes. In: Swanson E. (Ed.), Lithic Technology: Making and Using Stone Tools. Mouton, The Hague, Paris, pp. 115–127.
Jelinek A.J., 1982. The Middle Paleolithic in the Southern Levant with comments on the appearance of modern Homo sapiens. In: Ronen A. (Ed.), The Transition from Lower to Middle Paleolithic and the Origin of Modern Man. BAR International Series 151, Oxford, pp. 57–104.
Kocurek G., Dott R.H., 1981. Distinctions and uses of stratification types in the interpretation of eolian sand. Journal of Sedimentary Research 51: 579–595.
Jeske R.J., 1992. Energetic efficiency and lithic technology: an Upper Mississippian example. American Antiquity 57: 467–481.
Kolodny Y., 1967. Lithostratigraphy of the Mishash Formation, Northern Negev. Israeli Journal of Earth Sciences 16: 57–73.
Jeske R., Lurie R., 1993. The archaeological visibility of bipolar technology: an example from the Koster site. MidContinental Journal of Archaeology 18 (2): 131–160.
Kolodny Y., Nathan Y., Sass E., 1965. Porcellanite in the Mishash Formation, Negev, Southern Israel. Journal of Sedimentory Petrology 35: 454–463.
Karlin C., Julien M., 1994. Prehistoric technology: a cognitive science? In: Renfrew C., Zubrow B.W. (Eds.), The Ancient Mind: Elements of Cognitive Archaeology. Cambridge University Press, Cambridge, pp. 13–34.
Kretzoi M. and Vértes L., 1965. Upper Biharian (Intermindel) pebble-industry occupation site in western Hungary. Current Anthropology 6: 74–87.
Karlin C., Ploux S., Bodu P., Pigeot N., 1993. Some socioeconomic aspects of the knapping process among groups of hunter-gatherers in the Paris Basin area. In: Berthelet A., Chavaillon J. (Eds.), Use of Tools by Human and Nonhuman Primates. Clarendon Press, Oxford, pp. 318–337.
Kretzoi M. and Dobosi V.T., 1990. Vértesszölös: Site, Man and Culture. Akademiai Kiado, Budapest. Kroll E.M., 1997. Lithic and faunal distributions at eight archaeological excavations. In: Isaac G.L. (Ed.), Koobi Fora Research Project, Volume 5: Plio-Pleistocene Archaeology. Clarendon Press, Oxford, pp. 459–543.
Keeley L.H., Toth N., 1981. Microwear polishes on early stone tools from Koobi Fora, Kenya. Nature 293: 464–465.
Kroll E.M., Isaac G.L., 1984. Configurations of artifacts and 144
List of references
bones at Early Pleistocene sites in East Africa. In Hietala H.J. (Ed.), Intrasite Spatial Analysis in Archaeology. Cambridge University Press, Cambridge, pp. 4–31.
Lenoble A., 2005. Ruissellement et formation des sites préhistoriques: référentiel actualiste et exemples d’application au fossile. British Archaeological Report S. 1363, Oxford.
Kuhn S., 1995. Mousterian Lithic Technology: An Ecological Perspective. Princeton University Press, Princeton.
Leroi-Gourhan A., 1964. Le geste et la parole. I – technique et language. II – La mémoire et les rythmes. Albin Michel, Paris.
Kuijt I., Prentiss W.C., Pokotylo D.L., 1995. Bipolar reduction: an experimental study of debitage variability. Lithic Technology 20: 116–127.
Longo L., Peretto C., Sozzi M., Vannucci S., 1997. Artefacts, outils ou supports epuises? Une nouvelle approche pour l’étude des industries du pal’eolithique ancien: le cas d’Isernia La Pineta (Molise, Italie Centrale). L’Anthropologie 101: 579–596.
Kuman K., Field A.S., 2009. The Oldowan industry from Sterkfontein Caves, South Africa. In: Schick K., Toth N. (Eds.), The Cutting Edge: New Approaches to the Archaeology of Human Origins. Stone Age Institute Press Publications, Series 3, pp. 151–169.
Ludwig B.V., Harris J.W.K., 1998. Towards the technological reassessment of East African Plio-Pleistocene lithic assemblages. In: Petraglia M.D., Korisettar R. (Eds.), Early Human Behavior in Global Context. Routledge, London and New York, pp. 84–107.
Lamdan M., Ziffer D., Huster Y., Ronen A., 1977. Prehistoric Archaeological Survey of Nahal Shiqma. Local Council of Shaar Hanegev, Shaar Hanegev (in Hebrew). Laukhin S.A., Ronen A., Pospelova G.A., Sharonova Z.V., Ranov V.A., Burdukiewicz J.M., Volgina V.A., Tsatskin A., 2001. New data on the geology and geochronology of the Lower Palaeolithic site Bizat Ruhama in the southern Levant. Paléorient 27: 69–80.
Mace A., 1959. The excavation of a Late Paleolithic opensite on Hengistbury Head, Christchurch, Hants. Proceedings of the Prehistoric Society 25: 233–259. Magaritz M., 1986. Environmental changes recorded in the Upper Pleistocene along the desert boundary, southern Israel. Palaeogeography, Palaeoclimatology, Palaeoecology 53: 213–229.
Leakey M.D., 1967. Preliminary survey of the cultural material from Beds I and II, Olduvai Gorge, Tanzania. In Bishop W.W., Clark J.D. (Eds.), Background to Evolution in Africa. University of Chicago Press, Chicago, pp. 417–446.
Magaritz M., Goodfriend G.A., 1987. Movement of the desert boundary in the Levant from latest Pleistocene to Early Holocene. In: Berger W.H., Labeyrie L.D. (Eds.), Abrupt Climatic Change: Evidence and Implications. Reidel, Dordrecht, pp. 173–183.
Leakey M.D., 1971. Olduvai Gorge: Excavations in Beds 1 and 2, 1960–63. Cambridge University Press, Cambridge. Leakey M.D., 1975. Cultural patterns in the Olduvai sequence. In: Butzer K.W., Isaac G.L. (Eds.), After the Australopithecines. Mouton Publishers, The Hague, pp. 477–493.
Malinsky-Buller A., Grosman L., Marder O., 2011. A case of techno-typological lithic variability & continuity in the late Lower Palaeolithic. Before Farming 2011/1: 1–32.
Leakey M.G., Leakey R.E., 1978. Koobi Fora Research Project, Volume 1: The Fossil Hominids and Introduction to their Context, 1968–1974. Clarendon Press, Oxford.
Mallol C., VanNieuwenhuyse D., Zaidner Y., 2011. Depositional and paleoenvironmental setting of the Bizat Ruhama Early Pleistocene archaeological assemblages (Northern Negev, Israel): a microstratigraphic perspective. Geoarchaeology 26: 118–141.
Lemonnier P., 1976. La description des chaînes opératoires: contribution à l’analyse des systemes techniques. Techniques et Culture 1: 100–151.
Mania D., 1990. Auf den Spuren des Urmenschen: Die Funde aus der Steinrinne von Bilzingsleben. Deutscher Verlag der Wissenschaften, Berlin.
Lemonnier P., 1992. Elements for an Anthropology of Technology. Anthropological Papers 88. Museum of Anthropology, University of Michigan, Ann Arbor.
Mania D., Weber T., 1986. Bizingsleben III. Homo Erectus – seine Kultur und Umwelt. Veroffentl. Landesmus. Vorgesch. Halle 39, Berlin.
Lengyel G., 2005. Lithic Technology of the Upper Palaeolithic and Epipalaeolithic of Raqefet Cave, Mount Carmel, Israel. Unpublished Ph.D. dissertation, University of Haifa, Haifa.
Marder O., Gvirtzman G., Ron H., Khalaily H., Weider M., Bankirer R., Rabinovich R., Porat N., Saragusti I., 1999. 145
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
The Lower Palaeolithic site of Revadim Quarry, preliminary finds. Journal of the Israel Prehistoric Society 28: 21–53.
R.E. (Eds.), Earliest Man and Environments in the Lake Rudolf Basin. Chicago University Press, Chicago, pp. 574– 584.
Marder O., Milevski I., Matskevich Z., 2006. The handaxes of Revadim Quarry: typo-technological considerations and aspects of intra-site variability. In: Goren-Inbar N., Sharon G. (Eds.), Axe Age: Acheulian Tool-Making from Quarry to Discard. Equinox, London, pp. 223–242.
Mgeladze A., Lordkipanidze D., Moncel M.H., Despriée J., Chagelishvili R., Nioradze M., Nioradze, G., 2011. Hominin occupations at the Dmanisi site, Georgia, Southern Caucasus: raw materials and technical behaviours of Europe’s first hominins. Journal of Human Evolution 60: 571–596.
Marder O., Malinsky-Buller A., Shahack-Gross R., Ackerman O., Ayalon A., Bar-Matthews M., Goldsmith Y., Inbar M., Rabinovich R., Hovers E., 2011. Archaeological horizons and fluvial processes at the Lower Paleolithic openair site of Revadim (Israel). Journal of Human Evolution 60: 508–522.
Mourre V., 1996. Le débitage sur enclume au Paléolithique inférieur et moyen. Techniques, méthodes et schémas conceptuels. Article de D.E.A, Université de Paris X, Nanterre.
Martínez K., Garcia J., Carbonell E., Agustí J., Bahain J.J., Blain H.A., Burjachs F., Cáceres I., Duval M., Falguères C., Gómez M., Huguet R., 2010. A new Lower Pleistocene archaeological site in Europe (Vallparadís, Barcelona, Spain). PNAS 107: 5762–5767.
Mourre V., 2004. Le débitage sur enclume au Paleolithique moyen dans le Sudouest de la France. In: Van Peer P., Bonjean D., Semal P. (Eds.), Acts of the XIV UISPP Congress, Liège, pp. 29–38. Muckenhausen E., 1963. Le pseudogley. Sciences du Sol 1 : 21–29.
Martínez-Navarro B., 2004. Hippos, pigs, bovids, sabertoothed tigers, monkey, and hominids: dispersals through the Levantine corridor during late Pliocene and early Pleistocene times. In: Goren-Inbar N., Speth J.D. (Eds.), Human Paleoecology in the Levantine Corridor. Oxbow Books, Oxford, pp. 37–52.
Muttoni G., Scardia G., Kent D.V., 2013. A critique of evidence for human occupation of Europe older than the Jaramillo subchron (w1 Ma): comment on ‘The oldest human fossil in Europe from Orce (Spain)’ by Toro-Moyano et al. (2013). Journal of Human Evolution 65: 746–749.
Martínez-Navarro B., Rabinovich R., 2011. The fossil Bovidae (Artiodactyla, Mammalia) from Gesher Benot Ya‘aqov, Israel: Out of Africa during the Early-Middle Pleistocene transition. Journal of Human Evolution 60: 375–386.
Muhesen S., Jagher R., 2011. The Lower Paleolithic in Syria. In: LeTensorer J.-M., Jagher R., Otte M. (Eds.), The Lower and Middle Palaeolithic in the Middle East and Neighboring Regions. Etudes et Recherches Archéologiques de l’Université de Liège (ERAUL), Liège, pp. 35–48.
Martínez-Navarro B., Belmaker M., Bar-Yosef O., 2012. The bovid assemblage (Bovidae, Mammalia) from the Early Pleistocene site of ‘Ubeidiya, Israel: biochronological and paleoecological implications for the fossil and lithic bearing strata. Quaternary International 267: 78–97.
Munday F.C., 1976. Intersite variability in the Mousterian occupation of the Avdat/Aqev area. In: Marks A.E. (Ed.), Prehistory and Paleoenvironments in the Central Negev, Israel, Vol. 1. Southern Methodist University Press, Dallas, pp. 113–140.
Marwick B., 2003. Pleistocene exchange networks as evidence for the evolution of language. Cambridge Archaeological Journal 13: 67–81.
Mussi M., 1995. The earliest occupation of Europe: Italy. In: Roebroeks W., Van Kolfschoten T. (Eds.), The Earliest Occupation of Europe. University of Leiden, Leiden, pp. 263–269.
McBrearty S., Bishop L., Plummer T., Dewar R., Conard N., 1998. Tools underfoot: human trampling as an agent of lithic artifact edge modification. American Antiquity 63: 108–129.
Nadel D., Gordon D., 1993. Patination of flint artifacts: evidence from Bikta, a submerged prehistoric occurrence at the Sea of Galilee, Israel. Journal of the Israel Prehistoric Society 25: 145–162.
Menashe R., 2003. The Stratigraphy and Paleogeography of Tel-Sharuhen Section, North-Western Negev, Israel. Report GSI/35/02, Jerusalem.
Newcomer M., 1972. An analysis of a series of burins from Ksar Akil, Lebanon. Unpublished Ph.D. dissertation, University of London, London.
Merrick H.V., Merrick J.P.S., 1976. Recent archaeological occurrences of earlier Pleistocene age from the Shungura Formation. In: Coppens Y., Howell F.C., Isaac G.L., Leakey 146
List of references
Newcomer M., 1976. Spontaneous retouch. In: Second International Symposium on Flint: Maastricht: Nederlandse Geologische Vereniging (Staringia; 3), pp. 62–64.
Pelegrin J., Karlin C., Bodu P., 1988. Chaînes opératoires: un outil pour le préhistorien. In: Tixier, J. (Ed.), Technologie Préhistorique (Notes et monographies techniques 25), CNRS/CRA, Paris, pp. 55–62.
Newcomer M., 1980. Experimental flake scatter patterns: a new interpretive technique. Journal of Field Archaeology 7: 345–352.
Pei W.C., 1938. A preliminary study of a new Paleolithic station known as Locality 15 within the Choukoutien Region. Bulletin of the Geological Society of China 19: 147-187.
Nielsen A.E., 1991. Trampling the archaeological record: an experimental study. American Antiquity 56: 483–503.
Peretto C., 1994. Le industrie littiche del giacimento paleolitico di Isernia La Pineta, la typologia, le trace di utilizzazione, la sperimentazione. Istituto Regionale per gli Studi Storici del Molise “V. Cuoco”, C. Iannone, Isernia.
Neuville R., 1951. Le Paléolithique et le Mésolithique de Désert de Judée. Masson et Cie, Editeurs, Paris. Nir D., 1970. Notes on the Quaternary development of Nahal Shiqma basin. Studies in Geography 7: 1–12.
Perlès C., 1992. In search of lithic strategies: A cognitive approach to prehistoric chipped stone assemblages. In Gardin J.-C., Peebles C.S. (Eds.), Representations in Archaeology. Indiana University Press, Bloomington and Indianapolis, pp. 223–247.
Nir D., 1989. Geomorphology of Israel. Academon, Jerusalem. Nir D., Bar-Yosef O., 1976. Quaternary Man and Environment of Israel. Association for Protection of Nature, Tel Aviv (in Hebrew).
Perlès C., 1993. Ecological determinism, group strategies, and individual decisions in the conception of prehistoric stone assemblages. In: Berthelet A., Chavaillon J. (Eds.), Use of Tools by Human and Non-human Primates. Clarendon Press, Oxford, pp. 267–277.
Ohel M.Y., 1976. Upper Acheulian handaxes from Ruhama, Israel. Tel Aviv 3: 49–56.
Petraglia M.D., Nash D.T., 1987. The impact of fluvial processes on experimental sites. In: Nash D.T, Petraglia M.D. (Eds.), Natural Formation Processes and the Archaeological Record. BAR International Series 352, Oxford.
Oms O., Parés J.M., Martínez-Navarro B., Agustí J., Toro I., Martínez-Fernández G., Turq A., 2000. Early human occupation of Western Europe: paleomagnetic dates for two Paleolithic sites in Spain. PNAS 97: 10666–10670.
Petraglia M.D., Potts R., 1994. Water flow and the formation of Early Pleistocene artifact sites in Olduvai Gorge, Tanzania. Journal of Anthropological Archaeology 13: 228–254.
Owen L., 1982. Analysis of experimental breaks on flint blades and flakes. Studia Praehistorica Belgica. FNRS, pp. 77–87.
Pigeot N., 1990. Technical and social actors: flint knapping specialists and apprentices at Magdalenian Etiolles. Archaeological Review from Cambridge 9: 126–141.
Pelegrin J., 1985. Réflexions sur le comportement technique. In: Otte M. (Ed.), La signification culturelle des industries lithiques. Studia Prehistorica Belgica 4. BAR International Series 239: pp. 72–88. Pelegrin J., 1986. Technologie lithique: une méthode appliquée à l’étude de deux séries du Périgordien ancien (Roc de Combe, couche 8 – La Cote, niveau III). Thèse de doctorat. Université de Paris X, Paris.
Piperno M., Collina C., Galloti R., Raynal J.-P., Kieffer G., Bourdonnec F.-X., Poupeau G., Geraads D., 2009. Obsidian exploitation and utilization during the Oldowan at Melka Kunture (Ethiopia). In: Hovers E., Braun D.R. (Eds.), Interdisciplinary Approaches to the Oldowan. Springer, Dordrecht, pp. 111–128.
Pelegrin J., 1990. Prehistoric lithic technology: some aspects of research. Archaeological Review from Cambridge 9: 116–125.
PiPujol M.D., Buurman P., 1998. Analyzing groundwater gley and surface-water gley (pseudogley) effects in paleosols. Quaternary International 51–52: 77–79.
Pelegrin J., 1993. A framework for analysing prehistoric stone tool manufacture and a tentative application to some early stone industries. In: Berthelet A., Chavaillon J. (Eds.), The Use of Tools by Human and Non-Human Primates. Clarendon Press, Oxford, pp. 302–317.
Plummer T., 2004. Flaked stones and old bones: biological and cultural evolution at the dawn of technology. Yearbook of Physical Anthropology 47: 118–164. Plummer T.W., Bisop L.C., Ditchfield P.W., Ferraro J.V., Kingstone J.D., Hertel F., Braun D.R., 2009. The 147
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
environmental context of Oldowan hominin activities at Kanjera South, Kenya. In: Hovers E., Braun D.R. (Eds.), Interdisciplinary Approaches to the Oldowan. Springer, Dordrecht, pp. 149–160.
Ron H., Porat N., Ronen A., Tchernov E., Kolska Horwitz L., 2003. Magnetostratigraphy of the Evron Member – implications for the age of the Middle Acheulian site of Evron Quarry. Journal of Human Evolution 44: 633–639.
Potts R. 1984. Home bases and early homininds. American Scientist 72: 338-347.
Ronen A., 1975. The Paleolithic archaeology and chronology of Israel. In: Wendorf F., Marks A.E. (Eds.), Problems in Prehistory: North Africa and the Levant. SMU Press, Dallas, pp. 229–248.
Potts R., 1988. Early Hominid Activities at Olduvai. Aldine and Gruyter, New York.
Ronen A., 1979. Paleolithic industries. In: Horowitz A. (Ed.), The Quaternary of Israel. Academic Press, New York, pp. 296–307.
Potts R., 1991. Why the Oldowan? Plio-Pleistocene toolmaking and the transport of resources. Journal of Anthropological Research 47: 153–176.
Ronen A., 1991a. The Yiron Gravel lithic assemblage: artifacts older than 2.4 My in Israel. Archäologisches Korrespondenzblatt 21: 159–164.
Pryor J.H., 1988. The effects of human trample damage on lithics: a model of crucial variables. Lithic Technology 17: 45–50.
Ronen A., 1991b. The Lower Palaeolithic site Evron-Quarry in Western Galilee, Israel. Sonderveröffenttlichungen Geologisches Institut der Universität zu Köln 82: 187–212.
Pye K., 2009. Aeolian Sand and Sand Dunes. Springer, New-York. Raynal J.-P., Kieffer G., Bardin G., 2004. Garba IV and the Melka Kunture Formation. A preliminary lithostratigraphic approach. In: Chavaillon J., Piperno M. (Eds.), Studies on the Early Paleolithic site of Melka Kunture, Ethiopia. Origines, Istituto Italiano di Preistoria e Protostoria, pp. 137–166.
Ronen A., Gilead D., Schachnai E., Saul A., 1972. Upper Acheulian in the Kissufim region. Proceedings of the American Philosophic Society 116: 68–96. Ronen A., Burdukiewicz J.-M., Laukhin S.A., Winter Y., Tsatskin A., Dayan T., Kulikov O.A., Vlasov V.K., Semenov V.V., 1998. The Lower Palaeolithic site Bizat Ruhama in the northern Negev, Israel. Archäologisches Korrespondenzblatt 28: 163–173.
Roche H., 2005. From simple flaking to shaping: stoneknapping evolution among early hominins. In: Roux V., Bril B. (Eds), Stone Knapping: The Necessary Conditions for a Uniquely Hominin Behavior. McDonald Institute for Archaeological Research, Cambridge, pp. 35-53.
Rosen A., 1986. Quaternary Alluvial Stratigraphy of the Shephela and Its Paleoclimatic Implication. Report GSI/25/86, Jerusalem.
Roche H., Texier P.-J., 1996. Evaluation of technical competence of Homo erectus in East Africa during the Middle Pleistocene. In: Bower J.R.F., Sartorno S. (Eds.), Human Evolution in its Ecological Context. Royal Netherlands Academy of Arts and Sciences, Leiden, pp. 153–167.
Sahnouni M., Hadjouis D., de Made J.V., Derradji A.-e.-K., Canals A., Medig M., Belahrech H., Harichane Z., Rabhi M., 2002. Further research at the Oldowan site of Ain Hanech, North-Eastern Algeria. Journal of Human Evolution 43: 925–937.
Roche H., Delagnes A., Brugal J.P., Feibel C., Kibunjia M., Mourre V., Tixier P.J., 1999. Early hominid stone tool production and technical skill 2.34 Myr ago in West Turkana, Kenia. Nature 399: 57–60.
Sahnouni M., 2006. The North African Early Stone Age and the sites at Ain Hanech, Algeria. In: Toth N., Schick K. (Eds.), The Oldowan: Case Studies into the Earliest Stone Age. Stone Age Institute Publication Series 1, Gosport and Bloomongton, pp. 77–112.
Roche H., Brugal J.-P., Delagnes A., Feibel C., Harmand S., Kibunjia M., Prat S., Texier P.J., 2003. Les sites archéologiques plio-pléistocènes de la Formation de Nachukui (Ouest Turkana, Kenya): bilan préliminaire 1996– 2000. Comptes Rendus Palévol 2: 663–673.
Santonja M., Villa P., 2006. The Acheulian of Western Europe. In: Goren-Inbar N., Sharon G. (Eds.), Axe Age, Acheulian Toolmaking from Quarry to Discard. Equinox, London, pp. 429–478.
Ron H., Gvirtzman G., 2001. Magnetostratigraphy of Ruhama badland Quaternary deposits: a new age of the Lower Paleolithic site. Israel Geological Society Annual Meeting p. 95 (abstract).
Schick K.D., 1986. Stone Age sites in the making: experiments in formation and transformation of archaeological occurrences. BAR International Series 319, 148
List of references
Oxford.
Shackley M.L., 1978. The behavior of artifacts as sedimentary particles in fluviatile environment. Archaeometry 20: 55–61.
Schick K.D., 1987. Modeling the formation of early Stone Age artifact concentrations. Journal of Human Evolution 16: 789–807.
Shea J.J., 1999. Artifact abrasion, fluvial processes, and “Living Floors” at the Early Paleolithic site of ‘Ubeidiya (Jordan Valley, Israel). Geoarchaeology 14: 191–207.
Schick K.D., 1992. Geoarchaeological analysis of an Acheulian site at Kalambo Falls, Zambia. Geoarchaeology 7: 1–26.
Shea J.J., Klenck J.D., 1993. An experimental investigation of the effects of trampling on the results of lithic microwear analysis. Journal of Archaeological Science 20: 175–194.
Schick K.D., Toth N., Wei Q., Clark J.D., Etler D., 1991. Archaeological perspectives in the Nihewan Basin, China. Journal of Human Evolution 21: 13–26.
Shea J.J, Bar-Yosef O., 1998. Lithic assemblages from new (1988–1994) excavations at ‘Ubeidiya: a preliminary report. Journal of the Israel Prehistoric Society 28: 5–21.
Schick K., Toth N., 2006. An overview of the Oldowan industrial complex: the sites and the nature of their evidence. In: Toth N., Schick K. (Eds.), The Oldowan: Case Studies into the Earliest Stone Age. Stone Age Institute Publication Series 1, Gosport and Bloomongton , pp. 3–42.
Shifroni A., Ronen A., 2000. The “Tabun snap” from the Yabrudian/Acheulean interface at Tabun. Praehistoria 1: 109: 116. Shott M., 1989. Bipolar industries: ethnographic evidence and archaeological implication. North American Archaeologist 10: 1–24.
Semaw S., 2000. The world’s oldest stone artefacts from Gona, Ethiopia: their implications for understanding stone technology and patterns of human evolution between 2.6– 1.5 million years ago. Journal of Archaeological Science 27: 1197–1214.
Shott M., 1999. On bipolar reduction and splintered pieces. North American Archaeologist 20: 217–238.
Semaw S., 2006. The oldest stone artifacts from Gona (2.62.5 Ma), Afar, Ethiopia: implications for understanding the earliest stages of stone knapping. In Toth N., Schick K. (Eds.), The Oldowan: Case Studies into the Earliest Stone Age. Stone Age Institute Publication Series 1, Gosport and Bloomongton, pp. 43–76.
Sikes N., 1994. Early hominid habitat preferences in East Africa: paleosol carbon isotopic evidence. Journal of Human Evolution 2: 25–45.
Semaw S., Renne P., Harris J.W.K., Feibel C., Bernor R.L., Fesseha N., Mowbray K., 1997. 2.5–million-year-old stone tools from Gona, Ethiopia. Nature 385: 333–336.
Sneh A., Buchbinder B., 1984. Miocene to Pleistocene surfaces and their associated sediments in the Shephela region, Israel. GSI Current Research 1983–84, pp. 60–64.
Semaw S., Rogers M. J., Quade, J., Renne P.R., Butler R.F., Domínguez-Rodrigo M., Stout D., Hart W.S., Pickering T., Simpson S.W., 2003. 2.6-million-year-old stone tools and associated bones from OGS-6 and OGS-7, Gona, Afar, Ethiopia. Journal of Human Evolution 45: 169–177.
Solecki R.L., Solecki R., 1970. A new secondary flaking technique at the Nahr Ibrahim cave site. Bulletin du Musée de Beyrouth 23: 137–142.
Sneh A., Bartov Y., Rosensaft M., 1998. Geological Map of Israel. Geological Survey of Israel.
Soudry D., Nathan Y., Roded R., 1985. The Ashosh-Haroz Facies and their significance for the Mishash palaeography and phosphorite accumulation in the Northern and Central Negev, Southern Israel. Israel Journal of Earth Sciences 34: 211–220.
Semaw S., Rogers M.J., Stout D., Quade J., Levine N., Renee P.R., Butler R., Kidane T., Simpson S.W., 2008. The Oldowan-Acheulian transition: new insights from Gona, Ethiopia. Paleoanthropology 2007: A28. Semaw S., Rogers M., Stout D., 2009. The Oldowan– Acheulian transition: is there a ‘‘Developed Oldowan’’ artifact tradition? In: Camps M., Chauhan P. (Eds.), Sourcebook of Paleolithic Transitions. Springer, New York, pp. 173–193. Shackley M.L., 1974. Stream abrasion of chert implements. Nature 248: 501–502. 149
Stekelis M., Gilead D., 1966. Ma‘ayan Barukh: a Lower Palaeolithic site in Upper Galilee. Center of Prehistoric Research, Jerusalem. Stern N., 1993. The structure of the Lower Pleistocene archaeological record: a case study from the Koobi Fora Formation in Northwest Kenya. Current Anthropology 34: 201–225.
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
Stiles D., 1979. Recent archaeological findings at the Sterkfontein site. Nature 277: 381–382.
(Eds.), Evolución Humana en Europa y los Yacimientos de la Sierra de Atapuerca, Vol. 2. Junta de Castilla y León, Valladolid, pp. 403–420.
Stiles D., 1991. Early hominid behaviour and culture tradition: raw material studies in Bed II, Olduvai Gorge. African Archaeological Review 9: 1–19.
Toro-Moyano I., Martínez-Navarro B., Agustí J., Souday C., Bermúdez de Castro J.M., Martinón-Torres M., Fajardo B., Duval M., Falguères C., Oms O., Parés J.M., Anadón P., Julià R., García-Aguilar J.M., Moigne A.-M., Espigares M.P., Ros-Montoya S., Palmqvist P., 2013. The oldest human fossil in Europe dated to ca. 1.4 Ma at Orce (Spain). Journal of Human Evolution 65: 1–9.
Stiles D., 1998. Raw material as evidence for human behaviour in the Lower Pleistocene: the Olduvai case. In: Petraglia M., Korisetter R. (Eds.), Early Human Behavior in the Global Context: The Rise and Diversity of the Lower Paleolithic Period. Routledge, New York, pp. 133–150.
Toth N., 1982. The Stone Technologies of Early Hominids at Koobi Fora, Kenya: An Experimental Approach. Unpublished Ph.D. dissertation, University of California, Berkeley.
Stockton E.D., 1973. Shaw’s Creek shelter: human displacement of artifacts and its significance. Mankind 9: 112–117. Stoops G., Eswaran H., 1985. Morphological characteristics of wet soils. In Wetland Soils: Characterization, Classification and Utilization. International Rice Research Institute, Manila, pp. 187–189.
Toth N., 1985. The Oldowan reassessed: a close look at early stone artifacts. Journal of Archaeological Science 12: 101–120. Toth N., 1987. Behavioral inferences from early stone artifact assemblages: an experimental model. Journal of Human Evolution 16: 763–787.
Stout D., Quade J., Semaw S., Rogers M.J., Levin N.E., 2005. Raw material selectivity of the earliest stone toolmakers at Gona, Afar, Ethiopia. Journal of Human Evolution 48: 365– 380.
Toth N., 1997. The artifact assemblages in the light of experimental studies. In Isaac G. (Ed.), Koobi Fora Research Project, Volume 5: Plio-Pleistocene Archaeology, Chapter 7. Clarendon Press, Oxford, pp. 363–401.
Strathern M., 1969. Stone axes and flake tools: evaluations from two New Guinea Highland societies. Proceedings of the Prehistoric Society 35: 311–329.
Toth N., Schick K.D., Semaw S., 2006. A comparative study of the stone tool-making skills of Pan, Australopithecus and Homo sapiens. In: Toth N., Schick K.D. (Eds.), The Oldowan: Case Studies into the Earliest Stone Age. Stone Age Institute Publication Series 1, Gosport and Bloomongton, pp. 155– 222.
Svoboda J., 1987. Lithic industries of Arago, Vértesszölös, and Bilzingsleben hominids: comparison and evolutionary interpretation. Current Anthropology 28: 219–227. Sullivan A.P. III, Rosen K.C., 1985. Debitage analysis and archaeological interpretation. American Antiquity 50: 755– 779.
Vaks A., Bar-Matthews M., Ayalon A., Matthews A., Frumkin A., Dayan U., Halicz L., Almogi-Labin A., Schilman B., 2006. Paleoclimate and location of the border between Mediterranean climate region and the Saharo-Arabian Desert as revealed by speleothems from the northern Negev Desert, Israel. Earth and Planetary Science Letters 249, 3–4: 384–399.
Tchernov E., 1987. The age of ‘Ubeidiya Formation, an Early Pleistocene hominid site in the Jordan Valley, Israel. Israel Journal of Earth Sciences 36: 3–30. Tchernov E., Kolska Horwitz L., Ronen A., Lister A., 1994. The faunal remains from Evron Quarry in relation to other Lower Paleolithic hominid sites in the southern Levant. Quaternary Research 42: 328–339.
Vaks A., Bar-Matthews M., Ayalon A., Matthews A., Halicz L., Frumkin A., 2007. Desert speleothems reveal climatic window for African exodus of early modern humans. Geology 35: 831–834.
Texier P.J., 1995. The Oldowan assemblage from NY 18 site at Nyabusosi (Toro-Uganda). Comptes Rendus de l’Académie des Sciences Paris 320: 647–653.
Valoch K., 1989. Zur Kulturentwicklung im Altpaläolithikum in Europa und Berűcksichtigung Afrikas und Asiens. Ethnographisch-Archäolische Zeitschrift 30: 573–578.
Texier P.J., Roche H., 1995. The impact of predetermination on the development of some Acheulean chaînes opératoires. In: Bermúdez de Castro J.M., Arsuaga J.L., Carbonell E.
Van Riet Lowe C., 1946. The Coastal Smithfield and bipolar technique. South African Journal of Science XLII: 240–246. 150
List of references
Vergès J.M., Ollé A., 2011. Technical microwear and residues in identifying bipolar knapping on an anvil: experimental data. Journal of Archaeological Science 38: 1016–1025.
Yeshurun, R., Zaidner Y., Eisenmann V., Martínez-Navarro B., Bar-Oz G., 2011. Lower Paleolithic hominin ecology at the fringe of the desert: faunal remains from Bizat Ruhama and Nahal Hesi, Northern Negev, Israel. Journal of Human Evolution 60: 492–507.
Villa P., Courtin J., 1983. The interpretation of stratified sites: a view from the underground. Journal of Archaeological Science 10: 267–281.
Zaidner Y., 2003a. The use of raw material at the Lower Paleolithic site of Bizat Ruhama, Israel. In: Burdukiewicz J.M., Ronen A. (Eds.), Lower Palaeolithic Small Tools in Europe and the Levant. BAR International Series 1115, Oxford, pp. 121–132.
White J.P., 1967. Ethno-archaeology in New Guinea: two examples. Mankind 6: 409–414. White J.P., 1968. Fabricators, outils écailles or scalar cores? Mankind 6: 658–666.
Zaidner Y., 2003b. The lithic industry of Bizat Ruhama: a Lower Paleolithic site in southern coastal plain of Israel. Unpublished M.A. Thesis, University of Haifa, Haifa (in Hebrew).
White J.P., Modjeska N., Hipuya I., 1977. Group definitions and mental templates: an ethnographic experiment. In: Wright R.V.S. (Ed.), Stone Tools as Cultural Markers: Change, Evolution and Complexity. Australian Institute of Aboriginal Studies, Canberra, pp. 380–390.
Zaidner Y., Ronen A., Burdukiewicz J.-M., 2003. The Lower Paleolithic microlithic industry of Bizat Ruhama, Israel. L’Anthropologie 107: 203–222.
Wieder M., Gvirtzman G., Porat N., Dassa M., 2008. Paleosols of the southern coastal plain of Israel. Journal of Plant Nutrition and Soil Science 171: 533–541.
Zaidner Y., Yeshurun R., Mallol C. 2010. Early Pleistocene hominins outside of Africa: Recent excavations at Bizat Ruhama, Israel. PaleoAnthropology 2010: 162-195.
Wynn T., 1981. The intelligence of Oldowan hominids. Journal of Human Evolution 10: 529–541. Wynn T., 1985. Piaget, stone tools and the evolution of human intelligence. World Archaeology 17: 32–43. Yaalon D.H., Dan J., 1967. Factors controlling soil formation and distribution in the Mediterranean coastal plain of Israel during the Quaternary. Paper presented at the 7th INQUA Congress 1965, pp. 321–338. Yaalon D.H., Dan J., 1974. Accumulation and distribution of loess-derived deposits in the semi-desert and desert fringe areas of Israel. Zeitschrift für Geomorphologie N.F. 20: 91– 105.
Zhu R.X., Potts R., Xie F., Hoffman K.A., Deng C.L., Shi C.D., Pan Y.X., Wang H.Q., Shi R.P., Wang Y.C., Shi G.H., Wu N.Q., 2004. New evidence on the earliest human presence at high northern latitudes in northeast Asia. Nature 431: 559–562. Zilberman E., 1984. The Neogene and the Quaternary in the Central Negev. In Begin Z. B. (Ed.), Outlines of the Geology of the Northwestern Negev. Geological Survey of Israel, GSI/19/84, Jerusalem. Zilberman E., 1986. Pliocene-Early Pleistocene surfaces in the northwestern Negev – paleogeography and tectonic implications. Geological Survey of Israel, GSI/26/86, Jerusalem.
Yaalon D.H., Ganor E., 1975. Rates of aeolian dust accretion in the Mediterranean and desert fringe environments of Israel. Proceedings of the Ninth International Congress of Sedimentology, pp. 169–174.
151
APPENDIX APPENDIX 1. DESCRIPTION OF RAW MATERIAL OUTCROPS SAMPLED DURING THE SURVEY Sample 1
along 10 meters section. The conglomerate is lying on chalk of Maresha Formation (Sneh et al. 1998). The morphology of the pebbles and the matrix indicate fluvial deposition. The outcrop probably belongs to the Ahuzam Formation.
Coordinates 0124798/4104404; elevation 142 m.a.s.l.
Sample 4
Conglomerate exposures on both sides of a small channel about 500 meters from its conjunction with Nahal Shiqma. The height of the exposure is about 5 m above the channel floor. The exposure is characterized by unsorted pebbles from several to more than 20 cm in length. The matrix is uncemented quartzitic sand. Pebbles are rounded and smooth, some are flat. According to the characteristics of the sediment and the shape of the pebbles the outcrop was identified as Pleshet Formation (Sneh et al. 1998).
Coordinates 0126074/4101040; elevation 184 m.a.s.l.
Sample 2
On the eastern foothills of Tel Nagila conglomerates are exposed along a few hundred meters. The pebbles are rounded and smoothed, some of them are flat. Matrix is cemented quartzitic sand. The outcrop was identified as Pleshet Formation (Sneh et al. 1998).
An additional sample from Nahal Sad terrace. Large rounded and smoothed pebbles within uncemented quartzitic sand matrix. Pleshet Formation (Sneh et al. 1998). Sample 5 Coordinates 0127890/4100935; elevation 200 m.a.s.l.
Coordinates 0126032/4101076; elevation 185 m.a.s.l. About 100 meters from the main channel of Nahal Sad pebbles are exposed in two different units. The lower one includes large rounded and smoothed pebbles within uncemented quartzitic sand matrix. The upper one is distinguished by a high degree of size sorting. It includes small flat and smoothed pebbles in uncemented quartzitic sand matrix. The morphology of the pebbles from the upper unit indicates deposition in coastal environment. The transition between both units is not clear-cut. Sometimes small flat pebbles occur among large pebbles of the lower unit as well.
Sample 6 Coordinates: 0127726/4105364; elevation 163 m.a.s.l Outcrop of Ahuzam Formation near Tel Keshet identified by Issar (1961) as prototype of the formation. The pebbles are angular in cemented calcic matrix. Sample 7
Sample 3
Coordinates: 0124928/4102178; elevation 168 m.a.s.l.
Coordinates 0127890/4100935; elevation 200 m.a.s.l
Exposure containing flat and smoothed pebbles on the bank of Nahal Umm-Qida. The pebbles are found within slightly cemented sand. According to the location, elevation, sandy matrix and shape of the pebbles, the exposure belongs to Pleshet Formation.
The outcrop includes pebbles of a variety of shapes and varying level of roundedness in uncemented calcic matrix. Thickness of the layer is about 1 meter and it is exposed
152
Appendix
APPENDIX 2. RECORDS ON ARTIFACTS
Records on fractured flat pebbles Measurements Length Width Thickness Angle between fracture and cortex surfaces
Signs of impact on the fracture surface Crushing Small scars on fracture surface
Shape of the fracture surface Flat Ripple marks
Signs of further knapping Yes No
Records on cores Measurements Length, maximum Width* Thickness** Length, debitage surface Width, debitage surface Flaking angle*** Length of the largest complete scar Width of the largest complete scar
Total number of scars Number of debitage surfaces Number of scars on each surface Number of series of the removals on orthogonal cores**** Frequency of removals that remove the core distal edge Frequency of removals that remove the core lateral edge Number of hinged scars Number of incipient cones
Raw material shape Pebble, spheroid, subspheroid Pebble, subdiscoid Indeterminate
Direction of removals on each debitage surface
Distal end features
Unidirectional Bidirectional Unidirectional and side One removal per surface
Crushing Wedge-shaped fracture lines Both
Striking platform Cortical Flat (one removal) Chopper-like Prepared Indeterminate
Type of last removal on each surface Hinge Ordinary Removing the distal end of the core
Cortex 0-25% 26-50% 51-75% >75%
* At the broadest point, perpendicular to the length. ** At the thickest point, perpendicular to the length. *** Angle between striking platform and debitage surface. **** Debitage surface with at least two successive scars showing same direction. Records on exhausted cores Measurements Length Width Thickness Largest scar length Largest Scar width
Cortex 0-25% 26-50% 51-75%
Number of scars with identifiable negative of the bulb of percussion
Records on Detached Pieces The majority of the observations were made on complete flakes alone. Proximal fragments were subjected to observations III, IV and X only. Other categories in the group were only counted and measured. When possible, flaked flakes, anvil flakes 153
Lithic production strategies at the Early Pleistocene site of Bizat Ruhama, Israel
and flakes with modified edge were also subjected to these observations. For example, if they preserved proximal edge intact, observations III, IV were made and measurements of the butt were taken; if they preserved distal end intact,
observation VII was taken. Scar pattern was recorded only on a very few pieces that preserved the most of the dorsal face intact.
Measurements Maximum length Length Width Thickness Butt length Butt width Butt angle Edge angle
I. Cortex 0-25% 26-50% 51-75% 76-99% 100% First flake
II. Cortex location Distal Proximal Left Right Mesial Covering Right back None Left/distal Right/distal Right/proximal Left/proximal Distal back Left back
III. Butt Plain Cortical Dihedral Removed Broken Indeterminate Shattered
IV. Bulb Flat Prominent Indeterminate With ridge Pronounced cone Crushed
VI. Scars number
VII. Distal end features Scars on ventral face Fracture lines Crushing Opposite bulbs Combination of 1/2/3 Hinge Overshot
VIII. Edges
X. Number of incipient cones Butt Other
Patination Patinated Slightly patinated Patinated Preservation Fresh Slightly abraded Abraded
Measurements Maximum length is maximum distance between two points on the artifact’s edges. Length is measured along the technological axis of the item. Width is measured at the widest point of the flake perpendicular to its technological axis. Thickness is measured at the thickest point of the item. Length of the butt is measured between lateral extremities of the butt. Thickness of the butt is measured at the thickest part of the butt. Angle of the edge is measured on the most acute edge.
Oval Parallel Convergent Divergent Irregular Indeterminate
Pattern of scars on dorsal face
154
V. Scar pattern Multidirectional Unidirectional Bidirectional Side Indeterminate Cortical Unidirectional/side
Appendix
Pattern of scars is identified as cortical when 75% or more of the dorsal face of the flake is covered by the cortex. Additional records taken for flaked flakes Measurements Scar length Scar width
Scars location Dorsal face Ventral face Both faces
Number of scars
Additional records taken for anvil flakes • Signs of impact on dorsal faces. • Signs of impact at the intersection between ventral and broken/lateral surface. • Shape of the scars at the intersection between the ventral and broken/lateral surfaces. • Broken surface features. Additional records taken for flakes with modified edge A. Distribution* B. Delineation Continuous Partial Discontinuous Indeterminate
Rectilinear Convex Concave Notch Denticulate Irregular Indeterminate One tooth
C. Angle of removals** Flat Semi-abrupt Abrupt Cross-abrupt
D. Scar morphology*** Conchoidal fracture Clactonian notch Step-like fracture Marginal
E. Edge modification E. invasiveness type**** Scaled Long Step-like fracture Short Marginal Irregular Isolated removals Clactonian notch Clct. Notch and removals
*
Continuous trimming – scars that cover at least 2/3 or of one of the flake edges Partial trimming – scars that cover less than 2/3 of the edge Discontinuous – scars that occur on different edges, in each case covering less than 2/3 of the flake edge
**
flat - ≤ 45°; semi-abrupt 45°-70°; abrupt