Ben-Gurion University of the Negev Faculty of Humanities and Social Sciences Archaeological Division

BEN-GURION UNIVERSITY OF THE NEGEV FACULTY OF HUMANITIES AND SOCIAL SCIENCES ARCHAEOLOGICAL DIVISION

An Experimental Approach to Chalcolithic Sickle Blade Production:

The Case of Beit-Eshel Workshop

THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS

Emil Aladjem

UNDER THE SUPERVISION OF: Prof. Isaac Gilead

November 2009

BEN- GURION UNIVERSITY OF THE NEGEV FACULTY OF HUMANITIES AND SOCIAL SCIENCES ARCHAEOLOGICAL DIVISION

THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS

An Experimental Approach to Chalcolithic Sickle Blade Production:

The Case of Beit-Eshel Workshop

UNDER THE SUPERVISION OF : Prof. Isaac Gilead

Signature of student: ______Date: ______Signature of supervisor: ______Date: ______Signature of chairperson Of the committee for graduate studies: ______Date: ______

November 2009

ABSTRACT

The identification of craft specialization in the Chalcolithic period is a cardinal part of our understanding of the socio-economic structure of this culture in the southern Levant. Several unique craft specialization techniques were developed during the Chalcolithic period some that may have been developed as a direct result of the secondary products revolution which occurred during this time. One of the indicators for craft specialization are specified production areas, an example of which is the case presented in this study of the Beit Eshel flint blade workshop. The Beit Eshel site was excavated during the year 2003 and was identified as the most prolific flint blade production industry in our region. Since its discovery a number of studies were carried out with the aim of discerning the skill and proficiency of the craftsmen and the various techniques employed by them at the site. These studies have focused primarily on the understanding of the methodology used for the production of the flint blades and their reduction sequences based solely on the ancient finds. Alternatively, the present study is an attempt to enhance our knowledge of the techniques used at the site not only through the examination of the actual artifacts, but also by utilizing an experimental approach. A variety of techniques have been identified for the production of flint tools during prehistoric times. The complexity of a technique is indicative of the level of efficiency, specialization and craftsmanship of the flint knapper. The three major techniques of flint knapping techniques which have been defined are: the hard hammer percussion, the soft hammer percussion and the indirect percussion. These techniques are recognizable through specific attributes that appear on the resulting artifacts. Examples can be seen in patterns on the striking platforms and the profiles of the blades. Unfortunately these attributes can, at times, be difficult to recognize and may be deceptive. Therefore, additional ways of analyzing the methods employed in the production of flint tools are crucial for our understanding of the knapping techniques used at the site of Beit Eshel. The field of experimental archaeology offers us widely accepted solutions for this problem by replication of ancient tools and techniques. In this thesis a number of flint knapping experiments were carried out on the basis of the most common flint knapping techniques, which have been noted above. The blades and flakes that were experimentally knapped utilizing the hard hammer percussion, soft hammer percussion and the indirect percussion techniques were compared to the sample from the actual flint blade and

flake assemblage from the site of Beit Eshel. At the first stage of the comparative study the metric attribute averages (width, length and thickness) were measured and compared to the same attributes in the actual assemblage. The second stage of the study focused on the correlations of the non-metric attributes. In this stage the affects of the size of the striking platform on the metric attributes of the blades and flakes was analyzed and compared to the ancient assemblage. The results of the comparative study indicate that there are extreme difficulties in identifying the actual technique use in the flint knapping industry at the Beit Eshel site. The results of the experiments were not conclusive, for instance, they showed that one set of attributes favored one of the techniques and yet another set of attributes gave preference to another, on the same flint tool produced. There is however, a slight prevalence towards the hard hammer direct percussion technique, but the evidence does not entirely support the conclusion that it was the main technique used at the site. However, other evidence in supports the use of the hard hammer direct percussion technique at the site of Beit Eshel. This evidence was discovered on the face of numerous limestone cobbles that were identified as hammerstones. As part of the research process a classification of hammerstones discovered at Beit Eshel was carried out. This classification, which was based on size and shape indicated, that in all probability, different types of hammerstones were used at different stages of the operational sequence. It has been shown that the big hammerstones were used firstly for the opening of flakes; the middle sized hammerstones were used for the removal of the blades and the smallest hammerstones for retouching the blades and transforming them into sickle blades. The specific shape of the hammerstones was most likely a personal choice of the flint knapper. The preference of the shape of the hammerstone could be an indication of the level of proficiency of the various flint knappers working at the site. The most striking example is the presence of abrading stones whose circumference was shaped by abrading marks; these stones were used for the preparation of the striking platform and are most likely a tool used only by specialist. All of the knocked-off flakes that originated from the use of the hammerstones of the Beit Eshel assemblage were collected, sorted and measured. Certain types were probably produced by specialist craftsmen while others were more likely the result of the work of an apprentice. The breakage pattern of the hammerstones may be a possible indication for the efficiency and professional level of the flint knappers of the site.

The high skill level of the flint knappers from Beit Eshel is attested to in the high quality of the blade assemblage found at the site. Nevertheless, according to the hammerstone assemblage, it is hard to believe, that a sophisticated flint knapping technique was utilized at the site, hence it is hard to support the idea that a full time specialization was present at that site. Possibly it was a part time seasonal activity that took place incongruence with the harvesting season and required a big number of sickle blades. The methodology and analysis conducted in the present study could be instrumental for future works on the topic of recognition and classification of hammerstone and hammerstone knocked-off flakes dated to the Chalcolithic as well as other prehistoric periods.

Table of contents Abstract------3 List of Figures------8 List of Tables------10 1. Introduction------11 1.1. Goal of research------11 1.2. Beit Eshel Chalcolithic site------11 1.3. Chalcolithic flint tools research background------13 1.4. Experimental approach history------14 2. Methodology------15 2.1. Question of research------15 2.2. Forces which affect the formation of flakes and studied attributes------15 2.3. Experimental flint knapping------17 3. Experimental flint knapping------20 3.1. Knapping methods------20 3.1.1. Raw material acquisition------20 3.1.2. Hard hammer direct percussion------22 3.1.3. Soft hammer percussion------22 3.1.4. Indirect percussion------24 3.2. Element of Attribute Analysis------25 3.2.1. Length------26 3.2.2. Witdth------26 3.2.3. Thickness------26 3.2.4. Striking platform size------26 3.2.5. Profile------28 3.2.6. Lipping------28 3.2.7. Striking platform pattern------28 3.2.8. Bulb of percussion magnitude------29 3.3.1. Comparison of length, width and thickness------29 3.3.2. Striking platform size------33 3.3.3. Shape of profile------33

3.3.4. Lipping------35 3.3.5. Striking platform pattern------37 3.3.6. Bulb of percussion magnitude------38 3.4.1. Synthesis of the attribute ralations------39 4. Beit Eshel Hammerstone assemblage------50 4.1. Terminology and parallel assamblages------50 4.2. Beit Eshel hammerstone assemblage------51 4.2.1. Hammerstone types------53 4.2.1.1. Big hammerstones------55 4.2.1.2. Medium sized hammerstones------55 -Medium flat hammerstones------55 -Medium round hammerstones------56 -Medium elongated hammerstones------57 -Abrading stones------57 4.2.1.3. Small hammerstones ------58 -Small round hammerstones------59 -Small flat hammerstones------60 4.3. "Knocked-off flakes"------60 4.3.1. Abrading flakes------63 4.3.2. Massive flakes------65 4.3.3. Chips------64 4.3.4. Chunks------64 4.3.5. Natural backed knife (NBK) flakes------64 4.3.6. Hinge flakes------64 4.3.7. Feather termination flakes------65 4.4. Anvils------65 4.5. Spatial distribution and intra-site organization of the site------66 5. Disscusion------72 5.1. Technique recognition ------72 5.2. Suggested operational sequence and tool kit------72 5.2.1. Operational sequence------73

-The tool kit of the Beit Eshel flint knapper------75 5.3. Suggested skill level recognition and workshop organization according to the hammerstone assemblage------78 6. Conclusion------82 Appendices------86 References------86

------106תקציר

List of Figures

Figure 1. Cross section of the site of Beit Eshel------12

Figure 2. Limestone pebbles which were used during the experiment.------18

Figure 3. Antler hammers ------19

Figure 4. Beer Sheva river bed ------21

Figure 5. Measurements of the items ------27

Figure 6. Profile of the items ------30

Figure 7. Striking platform patterns ------31

Figure 9. Abrading stones ------59

Figure 10.Knocked-off scars and flakes ------62

Figure 12. Illustration of bipolar reduction of a core ------67

Figure 13. Block-on-block technique------68

Figure 14. Distribution of hammerstone chunks and flakes------69

Figure 15. Distribution of the flint cores from the site of Beit Eshel------70

Figure 16. Preparation of the striking platform edge------74

Figure 17. The tool kit------77

Graph 1. Graph of the striking platform size to length ratio of the hard hammer flakes from the experimental------41

Graph 2. Graph of the striking platform size to thickness ratio of the flakes from the Beit Eshel assemblage------41

Graph 3. Graph of the length to width ratio of the hard hammer flakes from the experimental assemblage------42

Graph 4. Graph of the length to width ratio of the flakes from the Beit Eshel assemblage------42

Graph 5. Graph of the length to width ratio of the blades from the Beit Eshel assemblage------43

Graph 6. Graph of the length to width ratio of the hard hammer blades from the experimental assemblage------43

Graph 7. Graph of the striking platform size to thickness ratio of the soft hammer flakes from

the experimental assemblage------44

Graph 8. Graph of the striking platform size to thickness ratio of the flakes from the Beit Eshel assemblage------45

Graph 9. Graph of the striking platform size to length ratio of the soft hammer flakes from the experimental assemblage------45

Graph 10. Graph of the striking platform size to length ratio of the flakes from the Beit Eshel assemblage------46

Graph 11. Graph of the length to width ratio of the soft hammer flakes from the experimental assemblage------46

Graph 12. Graph of the length to width ratio of the flakes from the Beit Eshel assemblage----47

Graph 13. Graph of the length to width ratio of the indirect percussion flakes from the experimental assemblage------48

Graph 14. Graph of the length to width ratio of the indirect percussion blades from the experimental assemblage------51

List of Tables

Table 1. The averages length, thickness and width of blades from the experimental assemblage

compared to blades of the original assemblage------32

Table 2 The averages length, thickness and width of flakes from the experimental assemblage

compared to the original assemblage. ------32

Table 3. The sizes of the striking platforms of the flakes and blades from the experimental and the

original assemblages ------33

Table 4. The results of the shape observation of the experimental blades and flakes compared to

the original assemblage. ------33

Table 5. Shape of profile of the flakes------33

Table 6. Blades - lipping frequencies.------36

Table 7. Flakes - lipping frequencies ------36

Table 8. Striking platforms patterns of the blades ------35

Table 9. Striking platforms patterns of the flakes ------38

Table 10. Bulb of percussion magnitude for blades ------39

Table 11. Bulb of percussion magnitude for flakes ------39

Table 12. Hamerstone types from the Beit Eshel hammerstone assamblage ------54

Table 13. Flake types found at Beit Eshel------63

1. Introduction

1.1 Goal of research

Studying of the organization of production areas and the level of craft specialization of flint tools is vital for understanding the socio-economic structure of the Stone Age societies. The identification and study of production areas or workshops is based on a variety of parameters such as: level of intensity, standardization, presence of microdebitage, site organization and identified tool kits.

Recent research is focused on recognition of workshops and the production areas and their integration with the rest of the sites of the period (Winter 2006). For better understanding of the craft specialization level, a more systematic research of the intra site structure of the recognized production sites and the skill level is needed. The present thesis is an experimental approach which attempts to determine the sophistication and skill level of flint blade production techniques used in the Chalcolithic site of Beit Eshel. The purpose of the research is to define the level of craft specialization and chaine operatoire of blade production at that site in particular and to better understand aspects of craft specialization of the Chalcolithic period in general.

1.2 Beit-Eshel Chalcolithic site

The site of Beit Eshel was discovered in 2003 (Gilead et al. 2004) and recognized as a flint workshop, where thousand of cores, thousands of sickle blades, sickle blade blanks and related debitage, were uncovered. The site is a pit, dug in the loess sediment. It dimensions are 3.5 x 3.5m; the southern section of the site is slightly higher than the north, where it ends with a small depression

(Fig.1). During the excavation three layers were recognized, which were separated by layers and lenses of ashes.

Figure 1. The cross section of the site of Beit Eshel (Gilead at all. 2004; Figure 5).

The floor of the pit slopes from west to east, which some of the flint distribution follows (Gilead et al. 2004:250, Fig.5).

Many of the flint artifacts are currently being refitted by Angela Davidson. The preliminary results of which showed that all the stages of production took place at the site, the production appear to be standardized and finally a variation of knapping skills are evident (Davidzon and Gilead in

press). The evidence of Micro-burin Technique used at the site was studied recently by Y. Vardi

(Vardi and Gilead in press(b)).

Along with the flint artifacts, a stone assemblage who consisted almost entirely of limestone pebbles was also uncovered. A total of 248 limestone pebbles were identified as hammerstones used for flint knappings. The hammerstones were identified on the basis of pounding signs and typical hammerstones flake scars. Uncovered were also 297 limestone flakes, mostly knapped-off from the hammerstones, as well as number of broken flint slabs, which were probably used as anvils. This assemblage is unique, because almost the entire assemblage origin from a clear workshop context. It is a rare opportunity of studying the whole sequence of the flint blade production along with the entire tool kit used in the process by the Chalcolithic flint knappers.

The present work is an attempt to understanding the Chalcolithic knappers techniques based on experiments conducted by the author. We attempt comparing indicative attributes of artifacts produced in a controlled experiment assemblage and a random sample from the original Beit Eshel assemblage.

1.3. The Chalcolithic flint tools research background

The research of Chalcolithic flint assemblages has a long history. It ranges from the earliest publications of Macdonald (1932) and Neuville (1934) through the publications of the site of

Shiqmim (Levy and Rosen 1987; Rosen 1987) and the renewed study of the Macdonald excavation assemblages from Nahal Besor (Roshwalb 1981), as well as the recent studies of Grar, Gilat (Rowan

2006) and Beer Sheva (Hermon 2008; Gilead et al. 2004; Gilead and Hermon in press). It has been demonstrated that locally available flint was utilized as raw material for the production of the

Chalcolithic sickle blades and that they tend to show a high degree of standardization whether they

found in a workshop or in domestic contexts (Gilead at al. 1995; Hermon 2008). Beside the Beit

Eshel site, several other workshop sites from the Chalcolithic context were reported. The only site which comes close to the quantity of the Beit Eshel assemblage is site A in the Nahal Besor area excavated by Macdonald in the early thirties (Macdonald 1932), but this site is of a different date and nature compared to the blade cores of Beit Eshel , since the most commomelement is the bladelet core (Roshwalb 1981; Rosen 1997). Another Nahal Beer Sheba site is Bir es-Safadi, where the same techniques as those of Beit Eshel were used, as indicated by the similar use of the microburin technique for reducing the length of the blade blanks (Vardi and Gilead 2008). Beside the Negev sites, three workshops were reported from the site of Abu Hamid in Transjordan.

However, the assemblages seem to be of a domestic nature and only small amount of artifacts were uncovered (Navaro I Barberan 1997).

1.4. Experimental approach history

Using experiments to explain flint technology is not new to the research of prehistoric cultures in the southern Levant. The skills of the professional flint knapper D. Ben-Ami were instrumental during experimental studies of Acheulian hand axe manufacture (Gonnen and Goren-Inbar 1999;

Dag and Goren-Inbar 2001; Goren-Inbar et al.1991) and Neolithic bifacial knifes (Gonnen and

Goring-Morris 2000). Neolithic blade technology was studied by Quintero and Wilke (1995) who replicated the reduction sequence of the Naviform technique. Thus, the experimental approach is sometimes an important way of understanding prehistoric technology.

2.Methodology

2.1. Questions of research

The analysis of the limestone artifacts of Beit Eshel, the flint knapping experiment and the comparative study, are aimed towards answering the following questions related to the Chalcolithic sickle blade manufacture from the perspective of the of Beit Eshel workshop:

1. Is it possible to distinguish between different techniques and means of flint knapping, by caring out experiments?

2. What was the role of hammerstones in the blanks production, according to the technique recognized by the experiment and the hammerstone assemblage of Beit Eshel?

3. What were the limestone tools used in process of producing the sickle blade and is it possible to identify a specific tool kit that was involved in the process?

4. Is it possible to detect level of flintknapping skill and chain operatoire on the basis of hammerstone assemblage.

2.2. Forces which affect the formation of flakes and studied attributes

Most lithic specialists and flint knappers agree that technology and variety of hammers used in the process of flint knapping produce recognizable differences in curtain attributes of the chipped flint products and debitage (Whitaker 2000; Hayden and Hutchingson 1989; Crabtree 1967 and 1972).

Hammers of different raw materials affect in different ways the cracks and brakeage patterns.

Lithic research, field observations and experimental flint-knapping produced a list of flake attributes, briefly introduced by Pelcin (1997), that may help in distinguishing different hammers used in the flint knapping processes.

Most of the attributes suggested (Pelcin 1997; Bordes 1947; Speth 1975, 1981, Dibble and Pelcin

1995; Ohnuma and Bergman 1982) are results of controlled experiments using special mechanism in order to single out factors which are believed to have affection on the flake morphology. The velocity and angle of the blow that resulted in the flake, and also the knapping skill, were not considered in these experiments, although we believe such factors are important when considering different knapping techniques.

To give an example, the relation of platform thickness to other flake attributes such as thickness, length, and width proved to be evocative for describing different flint knapping techniques (Hayden and Hutchigson 1989). On the one hand, it is agreed that antler and other soft materials produce longer and thinner flakes relative to their platform size. They have small bulbs of percussion and the profile of the striking platform shows a small lip of the outside angle of the platform. Hard hammers, on the other hand, are characterized by shorter, thicker flakes with pronounced bulbs of percussion and no lips on the striking platform ( Hayden and Hutchingson 1989; Whitaker 2000). On the other hand, others had not found differences between the attributes of flakes removed by soft and hard hammers (Bonnichsen 1977; Patterson and Sollberger 1978; Patterson 1982). Some even claim that differentiation of techniques solely on the basis of flake attributes is virtually impossible

(Mewheney; 1964). Occasionally characteristics which are believed to be typical of flakes produced by antler, or other soft hammer indenters, are, in fact, results of blow angle and velocity and have little to do with the material of the indenter. Lower blow angles might increase the chance of producing curved flakes, which are flakes where the crack appears away from the point of impact, and results in a peeling of the flake off the core rather than a chipping. Such flakes lack bulbs of percussion and bear a pronounced lip on their striking platforms - not necessary related to the material of the hammer used (Cotterel and Kaminga 1987).

2.3. The experimental flint knapping

The experimental approach we adopt is based on the well known assumption that different hammers and techniques were used in antiquity for producing flint tool: soft hammers (antler, bone, wood and various soft stones), hard hammers (hard limestone pebbles and basalts), indirect percussion and pressure flaking (where use of a special apparatus and assistant is often required)

(e.g. Crabtree 1972; Lewis-Johnson 1978). Each sort of hammer produces specific attributes found on the flake.

The experiment was carried out in three phases:

1. Flint and limestone pebbles were collected from the channel of Nahal Beer Sheva, near the site of Beit Eshel. The limestone pebbles were collected from the same area in order to be used as hammers during the hard hammer percussion experiment.

2. A series of flint knapping sessions for obtaining blades were carried out using different hammers and techniques:

- Hard hammer (hard limestone and flint pebbles) (Figure 1)

- Soft hammer ( Dama antler) (Figure 2a and b)

- Indirect percussion, while using antler and bone punch (Figure 2c).

- Pressure flaking, using a pole with an antler tip, a device which holds a core, and assistance of another person. This experiment was performed following Crabtree’s (1968) instructions. He presented a method of construction and use of such a pressure flaking device. This method appears to be too complicated, time consuming and in most of the cases almost impossible to apply due to

the small size of the cores and their course texture. The platform preparation and the non-prismatic morphology of most cores, suggest that this method was not in use at the Beit Eshel workshop.

Blades and flakes detached during the flint knapping sessions were collected for comparative analysis. Indicative attributes such as shape, width, thickness and striking platform patterns of blades and flakes were recorded and compared to samples from Beit Eshel, following Hayden and

Hutchingson (1989) and Pelcin (1996). The signs and the pattern of brakeage of the hammers were also recorded and compared to the hammers found at Beit Eshel. All the complete hammers and hammer struck flakes and chips found at the site were studied and measured.

Figure 1. Limestone pebbles which were used during the experiment.

a b c

Figure 2. Antler (Dama) hammers (a and b) and punch (c).

3. Experimental flint knapping

3.1. Knapping methods

Three experiments, using hammers of different raw materials and methods, were conducted in order to produce blades and flakes for comparison with the Beit Eshel assemblage. To obtain the optimal samples for analysis, selected were only whole blades and flakes from the original and the experimental assemblages. The same reduction sequence was used in all the experiments, in order to prevent undesirable variability which could be caused by differences in technique.

3.1.1. Raw material acquisition

Flint and limestone pebbles are still found in the Beer Sheva river bed, approximately 300 m. south-west of the site of Beit Eshel (Figure 4). This is the closest source of raw material and therefore pebbles used in the experiments were collected there.

Elongated and flat flint pebbles flint pebbles, 15- 30 cm in size, as those found in the original assemblage, were collected from the river bed for the experiment. The Chalcolithic knappers avoided round pebbles probably since the removal of the opening flake appears to be extremely difficult (Davidzon and Gilead in press).

The selection of limestone pebbles for hammers was made on the basis of the hammerstones from the Beit Eshel original assemblage, where the most common type appears to be round and flat shaped pebbles, 5-10 cm in diameter. These sizes later proved to be the most convenient size for flint knapping.

Figure 4. Abundance of flint and limestone pebbles in the Nahal Beer Sheva river bed.

3.1.2. Hard hammer direct percussion

In our study, we used both soft and hard hammers for direct percussion in an attempt to find attributes that will allow us to distinguish between the products of the two techniques. Hard hammer direct percussion is accomplished when the striking platform of the core is directly hit by a hard object, in our case, a hard limestone pebble (Figure 2).

Hard hammer direct percussion is the first recognized technology used for producing stone tools and it is the simplest and most intuitive of all the methods presented. Although the common perception is that the tools produced by this technique are crude and rough, experiments have shown that well-formed tools could also be made with this technique (Newcomer 1971; 1975, Mewhiney

1964).

One of the most important stages in the reduction sequence is the removal of the opening flake of the pebble. This stage is common to all of the knapping methods presented in the experiment. The scar resulting from the opening flake served as a striking platform of the core. There is evidence to suggest that these thick primary flakes were used at the Beit Eshel for the preparation of scrapers known as the Beer Sheva type (Hermon 2003).

Flint hammerstones, rather than limestone, were employed in the early stages of the experiment, because flint examples are often recognized as hammerstones and relate to flint manufacture

(Gophna and Friedman 1993; Gilead 1995). Knapping with flint hammers produced huge amounts of chunks, chips and smaller debitage which where the result of the large number of shattered and pulverized striking platforms. With the exception of one half of a flint hammer, flint hammers were not present at the site of Beit Eshel, far less then it should be expected from the site of the magnitude of Beit Eshel. Those hammers were not used in the flint knapping manufacture and are more likely used in preparation of food (Rosen 1995; 101).

Two cores were knapped with limestone pebbles and during the process a total of 40 blades, 30 flakes (Appendix 1) and a relatively small amount of chips were produced while chunks were not at all present. It is worth noting that the product never appeared with crushed or shattered striking platforms.

The striking edges of the cores were prepared in the experiment by grinding and abrading in order to remove overhangs . By edge preparation we mean only the edge abrasion of the platform and not the platform itself, unlike the common practice of polishing and faceting the platform before pressure flaking (Crabtree 1968).

Abrading of the striking platform's edge stabilizes and prevents crushing of the striking platform edge during the blow, applied with either stone or antler hammerstone. Thus, energy is not lost when penetrating the overhang of an unprepared platform edge (Whittaker 2000; Crabtree 1972). This striking edge preparation was performed with limestone pebbles, mostly flat and made of hard limestone. The preparation of the platform with the stone left specific signs of abrasion; such signs were discovered on several flat limestones from the Beit Eshel original hammerstone assemblage

(Figure 9). Usually their entire circumference is shaped by this activity, resulting in a profile which rather resembles a heavily used rubber eraser. Those stones are known as abrading stones or sometimes as abraders (Polhemus 1987; Cob 1998) and will later be discussed along with the hammerstone assamblge of Beit Eshel

3.1.3. Soft hammer percussion

Soft hammer percussion is carried out by directly knapping a core with a hammer made of soft material. The hammers used in the experiment are made of antler, bone and soft stone. It is worth noting that no antler or horn percussors were ever found in the site. The assumptions underlying the

choice of hammer raw material were based upon ethnographic research and archaeological research from other prehistoric sites (McCarthy 1967; Benson 1980). We used deer (Dama) antlers (Figure 3a and b), goat and cattle horn cores and splinters, and fragments of other large bones, usually extremities of a cow.

Two cores were knapped and during the process a total of 47 blades, 41 flakes were removed

(Appendix 1), a relatively small amount of chips appeared while chunks were not at all present.

The relative hardness and density of the Nahal Beer Sheva pebbles made it very difficult to work with these types of hammers. More power and skill are required for working with the organic materials. The use of an antler hammer requires greater skill and practice, as the basis of the technique is the use of an arched swing before hitting the striking platform. This method is difficult to learn, is less intuitive and requires greater commitment than using a simple hammer stone

(Whitaker 2000; Crabtree 1972). The disadvantage of such a hammer is the difficulty to obtain it, which turns such a hammer into a valuable tool which needs to be carried along (Hayden and

Hutchingson 1989). This fact renders the soft hammer made of antler into a special tool, and suggests the possible presence of craft specialization.

The heavy use of organic tools during the experiment left large numbers of amounts debris: broken fragments, splinters and chips. The presence of such debris in the site will indicate that organic hammers were extensively use.

3.1.4. Indirect percussion

This technique involved striking a punch-like object with an additional hammer, while the punch rests on the platform of the core. This technique requires very well-prepared platforms and perfect ridges for the crack to follow (Whittaker 1994).

For the application of this method, an antler (Dama) punch was used for the flaking and a Dama antler was used as a hammer (Figure 2a and c). During the experiment several punches were destroyed, because these splinter and crack often and reach a condition where further utilization is impossible. Two general techniques were used for this method. The first technique involves resting the core on the soft ground (Crabtree 1972, 1967 and Wittaker 2000), while the second involves holding the core and the punch in place with one hand and striking it with hammer with the other hand, as observed among the present-day Lacandon Indians (Michaels 1989).

Two cores were flaked and during the process a total of 40 blades and 35 flakes (Appendix 1) were produced, small amount of chips was also present.

It became immediately apparent that this method is far more demanding and requires considerably more skill than the direct percussion techniques. Further, the skill needed to perform this method is equivalent to the time needed to find the proper knapping tools.

3.2. Elements of Attribute Analysis

From the criteria present in literature (Bordes 1947; Crabtree 1967 and 1972; Hayden and

Hutchingson 1989; Wittaker 2000; Gonnen and Goren-Inbar 1999; Pelegrin 1996), shared knowledge of contemporary flint knappers, and practical experience, we present a list of four parametric and four non-parametric attributes which we believed offer the best method of distinguishing between different techniques and levels of flint working. As stated above, only the complete flakes and blades were measured (Figure 5a).

3.2.1. Length

Even though the longer blades does not seem to be the major attribute of the target blanks in the site of Beit Eshel (Daivdzon and Gilead in press), this attribute is consider while comparing the experimental assemblage to the original.

3.2.2. Width

In the present research the width was used to compute ratios with thickness and length for comparative purposes. This attribute was measured on the thickest part of the items, excluding the bulb of percussion area. (Hayden and Hutchingson 1989).

3.2.3. Thickness

This attribute was also used for computing ratios with other metric attributes of the blades and flakes. The wide was measured on the thickest section of the product while avoiding the bulb of percussion area, when present.

3.2.4. Striking platform size

This parameter (Figure 5b) is used along with others in creating proportional ratios between different categories in the same technique which will be later compared to the same ratio of other techniques.

a b

Striking platform size

length width h

Figure 5. Measurements of the items (a) and the striking platform (b).

3.2.5. Profile of the knapped artifact

In our experiment we distinguished four types of profiles: straight, s-shaped, curved and

overshoot (Figure 5). Profile types were subjectively determined by observing the side of the

item (Hayden and Hutchingson 1989; Pelcin 1997). The comparison between the experimental

and the original assemblage may indicate the relationships between types and techniques.

3.2.6. Lipping

The presence or absence of a lip was judged by passing a finger nail on the ventral side of the

flakes towards the striking platform; if the nail catches the lip formed in the area of the platform,

the artifact is recorded as lipped ( Hayden and Hutchingson 1989).

3.2.7. Striking platform pattern

The comparison between the experimental and the original assemblage may indicate which type

is related to which technique. We distinguished four striking platform patterns, which appear

within the experimental and original assamblages: cone, small (usually lipped), semi-cone, and

semicircle.

Cone (Figure 7d)- A conchoidal fracture, whose typical feature appears as a cone originating from

the point of impact on the proximal side of the outcome ( Crabtree 1972; Whitaker 2000).

Small (Figure 7b)- A small type of platform ( around 2-3 mm), bearing no signs of a cone and

usually appearing with a lip (usually referred amongst the flintknappers as elliptical (Pelegrin

personal communication)).

Semicircle (Figure 7c)- A type that usually appears in bending flakes (Coterel and Kaminga 1987)

where the crack appears away from the point of impact, and the striking platform is round and

bears no signs of bulb of percussion or cone.

Semi-cone (Figure 7a)- This class is a mixture of the cone and semicircle classes. This is a flake

which appears as a semicircle bending flake, but with the presence of a cone in the center of

the platform.

3.2.8. Bulb of percussion magnitude

The presence of a bulb of percussion, or areas of initiation, is usually related to hard hammer percussion. The bulb of percussion was observed subjectively by viewing it from the side of the outcome and than labeling it as pronounced, diffused or absent (Hayden and Hutchingson 1989:240;

Crabtree 1967:61,63).

3.3.1. Comparison of length, width and thickness

The total of 40 blades and 30 flakes were randomly selected from the pile of all the blades from all of the squares from the site of Beit Eshel (Appendix 1). Only unbroken pieces were selected, those were collected from all of the excavation squares of the site. Differences within variables were evaluated by comparing their measurement statistics and by correlating different ratios that were used as a basis of comparison between the assemblages.

The basic measurements of the blades from the original assemblage compared to the experimental assemblage are shown in Table 1. On the basis of the relations between the measurements, it is difficult to distinguish the techniques, because none of the presented technique

b c d a

Figure 6. Profile of the items; overhang (a), straight (b), curved (c) and s-shaped (d).

a b

c d

Figure 7. Striking platform patterns; semi-cone (a), small (b), semi-circle (c) and semi –cone (d).

results appear to be clearly distinct from the other. However, there is a slight tendency for the soft

hammer blades to be thicker and for the Beit Eshel blades to be longer, which could be related to the

relatively higher skill level of the Beit Eshel flint knappers.

Hard hammer Soft hammer Indirect Technique / Metric percussion percussion percussion Beit Eshel Attributes (N=40) (N=40) (N=40) (N=40) Length (cm) 5.8 5.85 5.4 7.5 Width (cm) 2.0 2.4 2.2 2.4 Thickness (cm) 0.63 0.84 0.63 0.76

Table 1. The averages length, thickness and width of blades from the experimental assemblage

compared to blades of the Beit Eshel assemblage.

The measurements of the flakes from the original assemblage compared to the experimental

assemblage are shown in Table 2. Among the flakes, the same tendency appear, where the attributes

are not distinctly different from one another and from the original assemblage.

In conclusion, we can say that in this set of attributes all of the presented technique could be the

one used in the Beit Eshel workshop.

Hard hammer Soft hammer Indirect Technique/Metric percussion percussion percussion Beit Eshel attributes (N=30) (N=30) (N=30) (N=30) Length (cm) 3.56 4.6 2.96 4.44 Width (cm) 2.73 3.8 2.36 4.0 Thickness (cm) 0.7 0.7 0.5 1.0

Table 2. The averages length, thickness and width of flakes from the experimental assemblage

compared to the Beit Eshel assemblage.

3.3.2. Striking platform size

The results of the comparison of striking platform sizes are presented in Table 3 and 4. The striking platform size of both the hard and soft hammer percussion blades appear to be closer to the original assemblage sample. The results of these two techniques appear to be almost identical, and the third - the indirect percussion technique - is not similar to any other technique, so it may not be considered the technique used according to that attribute. The striking platform size of the flakes appear to by significantly bigger then all the rest of the techniques used in the experiment. This could be explained in the different intention when the flake is struck in the experimental assemblages it is in an intention to produce a blade, while in the Beit Eshel assemblage it is probably struck in order to reshape the core.

Technique /Striking Hard hammer Soft hammer Indirect platform size percussion percussion percussion Original Mean of flakes N=30(cm) 0.5 0.5 0.46 0.7 Mean of blades N=40 (cm) 0.2 0.3 0.48 0.3

Table 3. The sizes of the striking platforms of the flakes and blades from the experimental and the original assemblages

3.3.3. Shape of Profile

The results of shape of the profile observations are presented in Table 4 and 5. The shapes of the experimental blades generally tend to have straight profiles, while the blades of the original assemblage have a tendency to be curved. The curved shape is also present within the experimental assemblage, but whereas in the original assemblage this is the dominant type, among the

experimental assemblage it is less pronounced. Another observation is the presence of relatively high percentage of overshoot blades in the original assemblage and only two examples present in the hard hammer percussion. The reason for the appearance of only two overshoot items of this type of shape in the experimental assemblage is that the original blades are relatively longer than the blades from the experimental assemblage. Usually the overshoot results when a hard, follow-through blow is applied to the core. The purpose of the overshoot is to remove series of hinged scars and to lengthen the knapping surface of the core in order to produce longer blades (Wittaker 2000; Mewhiney 1964:

204). Finally the existence of the overshoot blades excludes the option of utilization of the ground- supported indirect percussion technique, as the ground will prevent the formation of the overshoot.

The S-shaped type of profile is not present at all in the original assemblage, but it is one of the most frequent types produced by the indirect percussion and soft hammer direct percussion techniques. The indirect percussion flakes show the highest frequencies of the S-shaped profile, while this type is absent in all the flakes produced by the other techniques including the original assemblage.

This suggests that the technique closest to the original one, according to the attribute of the shape of profile is hard hammer percussion. A strong argument for that statement is that there is a significant occurrence of the curved type, which is the type that appears most frequently in the original assemblage

Shape of profile Hard hammer Soft hammer Indirect Original (blades) percussion percussion percussion assemblage

N % N % N % N % Straight 29 73 28 70 20 50 5 13 Curved 9 28 1 3 12 30 26 65 S-shaped 0 0 11 28 8 20 1 3 Overshoot 2 0 0 0 0 0 8 20 Total 40 100 40 100 30 100 40 100

Table 4. The results of the shape observation of the experimental blades and flakes compared to the original assemblage.

Shape of profile Hard hammer Soft hammer Indirect Original (flakes) percussion percussion percussion assemblage

N % N % N % N % Straight 29 73 29 97 10 33 19 63 Curved 11 28 1 3 3 10 8 27 S-shaped 0 0 6 0 17 57 3 10 Total 40 100 40 100 30 100 30 100

Table 5. Shape of profile of the flakes

3.3.4. Lipping

The frequencies of lipping are shown in the Table 6 and 7. The frequency of lipping in the direct percussion technique is closest to the original assemblage while the soft hammer percussion technique show less pronounced distinction and the indirect percussion differ clearly. Lipping is always present in the hard hammer percussion experimental blades and flakes, despite the common

assumption that lipping never occurs in this type of striking platforms (Sharon and Goren-Inbar

1999; Hayden and Hutchinson 1989:247; Crabtree 1970:150). All of the flakes show lipping in all of the techniques present.

It seems that, lipping produced by the direct percussion method is the closest to Beit Eshel original assemblage.

Lipping of the striking platform Hard hammer Soft hammer Indirect Original (blades) percussion percussion percussion assemblage N % N % N % N % Present 30 75 26 65 14 35 28 70 Absent 10 25 14 35 26 65 12 30 Total 40 100 40 100 40 100 40 100

Table 6. Blades - lipping frequencies.

Lipping of the striking platform Hard hammer Soft hammer Indirect Original (flakes) percussion percussion percussion assemblage N % N % N % N % Present 28 93 21 70 27 90 5 17 Absent 2 7 9 30 3 10 25 83 Total 30 100 30 100 30 100 30 100

Table 7. Flakes - lipping frequencies.

3.3.5. Striking platform pattern

The striking platform patterns of the experimental assemblages and the original assemblage are shown in Table 8 and 9. The striking platform of the original assemblage blades are clearly semicircular and small lipped only. The experimental blade assemblages, in contrast, are considerable diverse – excluding the indirect percussion, there is a clear preference for the semicircular type. Among the flakes, the semi-circle type is the most frequent in the original assemblage, while in the experimental assemblage the hard hammer direct percussion flakes show the greatest frequency in the cone type of striking platform. The soft hammer percussion flakes show a tendency for the semicircle type, like the original assemblage. In contrast, indirect percussion products are completely distinct from the original assemblage. According to this attribute the technique closer to the Beit Eshel assemblage is very hard to distinguish, because the results appear to be very diverse.

The only thing that could be said is that the indirect percussion method show to be different from the other techniques and the Beit Eshel assemblage, the fact represented by the lack of small platform type. In conclusion it could be said that according to the striking platform size attribute the indirect percussion technique show the most dissimilarity.

Striking platform pattern Hard hammer Soft hammer Indirect Original (blades) percussion percussion percussion assemblage

N % N % N % N % small 13 33 16 40 0 0 16 40 semicircle 11 28 12 30 27 68 20 50 semi-cone 14 35 12 30 13 33 4 10 cone 2 5 0 0 0 0 0 0 Total 40 100 40 100 40 100 40 100

Table 8.Striking platforms patterns of the blades

Striking platform pattern Hard hammer Soft hammer Indirect Original (flakes) percussion percussion percussion assemblage N % N % N % N % small lipped 10 33 5 16 0 0 3 10 semicircle 2 7 13 43 18 60 7 23 semi-cone 2 7 11 37 12 40 18 60 cone 16 53 1 3 0 0 2 7 Total 30 100 30 100 30 100 30 100

Table 9.Striking platforms patterns of the flakes

3.3.6. Bulb of percussion magnitude

The frequencies of the magnitude of the bulb of percussion between the experimental techniques and the original assemblage are shown in Tables 10 and 11. In the original assemblage of blades there are only two categories – a pronounced bulb of percussion or no bulb at all, with a majority of the blades lacking a bulb. In the hard hammer percussion technique, the blade assemblage is equally divided between those with diffused bulbs and those where bulbs are absent. In the soft hammer percussion technique the assemblage show the same distribution, only that the frequency is slightly in favor of the pronounced bulb of percussion, and this runs contrary to what is to be expected from the soft hammers (Hayden and Hutchinson 1989:240; Crabtree 1967:61). The majority indirect percussion blades show lack of bulbs of percussion.

The hard hammer percussion flakes tend to lack bulbs of percussion, the soft hammer percussion and the indirect percussion towards diffused bulbs and within the original assemblage the amount is almost evenly divided between the diffused and the absent types of bulbs of percussion magnitude.

According to that attribute both the hard hammer and the soft hammer percussion could be candidates to be the technique used in the original assemblage.

Bulb of percussion magnitude Hard hammer Soft hammer Indirect Original (blades) percussion percussion percussion assemblage

N % N % N % N % diffused 20 50 24 60 11 28 26 65 absent 18 45 12 30 29 73 14 35 pronounced 2 5 4 10 0 0 0 0 Total 40 100 40 100 40 100 40 100

Table 10. Bulb of percussion magnitude for blades.

Bulb of percussion magnitude Hard hammer Soft hammer Indirect Original (flakes) percussion percussion percussion assemblage

N % N % N % N % diffused 1 3 19 63 2 90 11 37 absent 28 93 6 20 27 7 5 17 pronounced 1 3 5 16 1 3 14 47 Total 30 100 30 100 30 100 30 100

Table 11. Bulb of percussion magnitude for flakes

3.4. Synthesis of the attribute relations

The relationship between two variables could be deceptive if we use only numerical summaries.

The linear regression method is one where one variable is plotted against the other, thus producing a

scatter diagram. This presentation has the advantage of being more easily comprehended because it is visually displayed (Shennan 1988: 114-131; Ott et al.1992: 464-497). In this stage of statistical analysis a comparison is made between two attributes and presented by a scatter diagram and a trend line. This correlation exemplifies how the striking platform size affects the metric attributes of the flint blanks (Hayden and Hutchingson 1989). The pattern of the scatter diagram, the slope value (the constant before the independent variable x) of the trend line and the coefficient of determination (R²) are compared to the original assemblage in order to find resemblance. In this thesis we are not attempting to explain the relationship itself, but we are trying to find similarities between the experimental and original assemblages, on the basis of that relationship.

The compared attributes of the flakes from the direct hard hammer percussion assemblage show a general similarity to the attributes of the flakes from the original assemblage. Amongst the flakes it is visible that the size of the striking platform mostly affects the dimensions of the flakes. Thus, it is evident that the smaller sized platforms will produce thinner flakes amidst both of the assemblages

(Graph 1 and 2).

The length to width ratio shows that the longer the flakes are the wider they tend to be in both of the presented assemblages (Graph 3 and 4). The dimensions of the blades, however, show almost no dependence on the size of the striking platform in both original and experimental assemblages. The length to width ratio, for the blades, represents a clustered scatter cloud, therefore illustrating that both of the assemblages show high standardization in their length and width dimensions (Graph 5 and 6).

Size of striking platform to thickness ratio of hard hammer percussion flakes (N.30)

2.5 y = 0.8303x + 0.2793 R2 = 0.2725 2

1.5

1 Thickness (cm)Thickness 0.5

0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Size of platform (cm)

Graph 1. Graph of the striking platform size to length ratio of the hard hammer flakes from the experimental assamblage.

Size of striking platform to thickness ratio of flakes from the original assamblage (N.30)

2.5 y = 0.6203x + 0.6306 R2 = 0.3816 2

1.5

(cm)Thickness

0.5

0 0 0.5 1 1.5 2 Size of platform (cm)

Graph 2. Graph of the striking platform size to thickness ratio of the flakes from the Beit Eshel assemblage.

Length to wide ratio of hard hammer percussion flakes (N.30)

6 y = 0.5384x + 0.8134 R2 = 0.6227 5

3 Width (cm) Width 2

0 0 1 2 3 4 5 6 7 8 Length (cm)

Graph 3. Graph of the length to width ratio of the hard hammer flakes from the experimental assemblage.

Length to wide ratio of flakes from the original assamblage (N.30)

7 y = 0.3549x + 2.6112 6 R2 = 0.2547

3 Wide (cm) Wide

0 0 1 2 3 4 5 6 7 8 9 Length (cm)

Graph 4. Graph of the length to width ratio of the flakes from the Beit Eshel assemblage.

Length to width ratio of blades from the original assamblage (N.40)

4 y = 0.0939x + 1.7656 2 3.5 R = 0.0468

2.5

2 Width (cm) Width 1.5

0.5

0 0 2 4 6 8 10 12 Length (cm)

Graph 5. Graph of the length to width ratio of the blades from the Beit Eshel assemblage

Length to width ratio of hard hammer percussion blades (N.40)

4.5 y = 0.3426x + 2E-05 R2 = 0.2883 4 3.5

2.5

2 Width(cm)

1.5

1 0.5

0 0 1 2 3 4 5 6 7 8 9 Length (cm)

Graph 6. Graph of the length to width ratio of the hard hammer blades from the experimental assemblage

In the soft hammer percussion flake population, the general tendency is that the dimensions of the flakes are affected by the size of the striking platform. The width of the soft hammer percussion flakes is the only one which shows no dependence to the striking platform size. The other two dimensions (thickness and length) for the flakes, show almost a straight line correlation, which is similar to the Beit Eshel assemblage (Graph 7, 8, 9 and 10). The length to width ratio of the soft hammer percussion flakes differs from the original assemblage in that there is no straight relationship to the length and the width variables of the flakes. The relationships of the blades from the soft hammer percussion experimental assemblage and the blades from the original assemblage generally appear to be close. The length to width ratio also show standardization which is presented by a clustered scatter cloud ( Graph 11 and 12).

Size of striking platform to thickness ratio of soft hammmer percussion flakes (N.30)

2.5 y = 0.8428x + 0.2175 R2 = 0.7461 2

1.5

1 (cm) Thickness

0.5

0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Size of platfrom (cm)

Graph 7. Graph of the striking platform size to thickness ratio of the soft hammer flakes from the experimental assemblage

Size of striking platform to thickness ratio of flakes from the original assamblage (N.30)

2.5 y = 0.6203x + 0.6306 R2 = 0.3816 2

1.5

(cm) platform of Size 0.5

0 0 0.5 1 1.5 2 Thickness (cm)

Graph 8. Graph of the striking platform size to thickness ratio of the flakes from the Beit Eshel assemblage

Size of striking platform size to length ratio of soft hammer percussion flakes (N.30)

y = 1.5079x + 3.8273 7 R2 = 0.2879 6

3 Length(cm) 2

0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Size of pllatform (cm)

Graph 9. Graph of the striking platform size to length ratio of the soft hammer flakes from the experimental assemblage

Size of striking platform to length ratio of flakes from the original assamblage (N.30) y = 1.4881x + 3.505 9 R2 = 0.2368

4 Length (cm) Length 3

1 0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Size of platfrom (cm)

Graph 10. Graph of the striking platform size to length ratio of the flakes from the Beit Eshel assemblage.

Length to width ratio of soft hammer percussion blades (N.40)

4 y = 0.6729x - 1.4993 2 3.5 R = 0.4658 3 2.5 2

Width(cm) 1.5 1 0.5 0 0 1 2 3 4 5 6 7 8 Length (cm)

Graph 11. Graph of the length to width ratio of the soft hammer flakes from the experimental assemblage

Length to width ratio of blades from the original assamblage (N.40)

4 y = 0.0939x + 1.7656 3.5 R2 = 0.0468

2.5

Width (cm) Width 1.5

0.5 0 0 2 4 6 8 10 12 Length (cm)

Graph 12. Graph of the length to width ratio of the flakes from the Beit Eshel assemblage

The graphed metric attributes of the flakes from the indirect percussion experimental assemblage that are affected by the striking platform size are different from those of the original assemblage.

The scatter diagram of the length to width ratio shows the general tendency that the longer the flakes are the wider they tend to be. Presently, indirect percussion flakes show the same slope of the trend line as in the original assemblage, but the samples are particularly clustered in the lower end of the line (Graph 13). The metric attribute ratios of the blades from the experimental assemblage do not show likeness to the attributes of the blades from the original assemblage, with the exception of the length to width ratio. This ratio appears to have the same position and clustering as are found in the original assemblage (Graph 14).

Length to width ratio of indirect percussion flakes (N.30)

4.5 y = 0.549x + 0.8629 4 R2 = 0.4964 3.5

3 2.5 2

Width (cm) 1.5 1

0.5 0 0 1 2 3 4 5 6 7 Length (cm)

Graph 13. Graph of the length to width ratio of the indirect percussion flakes from the experimental assemblage

Length to width ratio of indirect percussion blades (N.40)

3.5 y = 0.1512x + 1.4273 R2 = 0.1223 3

2.5

1.5 (cm) Width

0.5

0 0 1 2 3 4 5 6 7 Length (cm)

Graph 14. Graph of the length to width ratio of the indirect percussion blades from the experimental assemblage

In summary, the direct hammer and the soft hammer percussion technique emerge as the techniques which reflect the most correlation in the attributes of the final product between the experimental and the original assemblages. Conversely, the indirect percussion technique exhibited the most dissimilar results between the experimental and original assemblages. Consequently, judging from the statistical analysis the best results can be reached using the direct and the soft hammer percussion techniques.

4. Beit Eshel Hammerstone Assemblage

4.1. Terminology and similar assemblages

Hammerstones are one of the earliest stone tools used by ancient man, and date from the earliest

Paleolithic until the Late Byzantine period. They are defined in numerous typological lists (Leakey

1971; Bordes 1961; Willoughby 1985). Hammerstones are defined as round objects which were used for pounding and crushing, in butchering, or in food preparation; finally, they were an integral part of the flint production industry. The use of hammerstones during proto-historic periods (Neolithic,

Chalcolithic and Early Bronze Age) is discussed in a number of publications (Rosen 1995; Gilead and

Fabian 1995; Roshwalb 1981; Macdonald 1934). Rosen included the hammerstone type in his typological list of the late prehistoric periods, where he describes crushed or pecked pebbles in the range of 10-5 cm which preserve their natural round shape (Rosen 1995:101). In Rosen’s typology, these are separated into two types. The first type includes pieces called spheroids, that were made of flint with crushing and pounding marks on all sides. Rosen (1995;101) doubts the use of the flint spheroids in the manufacture of other flint implements, and suggests they function as pounding stones for preparation of food. During our experiment this claim was substantiated, because when used for flint knapping, the flint spheroids appeared to be very bad hammerstones. Knapping with flint spheroids always ended with crushed and shattered platforms of percussion and hinge flake terminations, which spoiled the core and required that it be discarded.

The second type Rosen's pecked stones, were made of limestone pebbles. In most cases these preserve their natural round shape, showing pecking and crushing signs on their entire circumference

(Rosen 1995; 101). Those, according to Rosen, are more likely to be involved in the flint knapping industry.

In the Nahal Besor Site A, which was described as a workshop site, limestone implements were used for flint blade manufacture (Macdonald 1932).

In their publication of the limestone tool assemblage from Grar, Gilead and Fabian (1995) describe what appears to be a tool type; in other words, the pieces fashioned by intentional flaking. By their definition, the limestone knapped series described in Grar were produced by the same technology as the flint Chalcolithic industries, suggesting the heavy use of hard hammer direct percussion, judging from the pronounced bulb of percussion on the flakes ventral surface (Gilead and Fabian 1995: 282).

On the other hand, the flakes and the flake scars from the Beit Eshel stone assemblage do not present any bulb of percussion whatsoever. They are, as stated above, the result of splitting the material as a result of 90 degrees straight blow. However, the authors mention the presence of 19 limestone pebbles with battered edges which, they suggest, could have been used as hammerstones in flint production

(ibid; 282).

Similar to the limestone assemblages at Grar, several assemblages, with hammerstones among them, were recently excavated in the sites of Gat Guvrin (Fabian in preparation) and Tell Gishron

(Nahshony and Aladjem in prep).

In the Beit Eshel site the stone pebbles were found along with a massive amount of flint workshop material, which clearly indicates the likelihood that they were closely involved in flint blade production (Gilead et al. 2004).

4.2. Beit Eshel hammerstone assemblage

The pebbles used at Beit Eshel were made of hard limestone, but chalk pebbles also appear in the assemblage; the ratio of the two raw materials is 1:74 in favor of the hard limestone ( Appendix 2).

The Beit Eshel pebble assemblage consists entirely of hammerstones which were used during the flint

knapping reduction sequence. Under the term hammerstone, we include every limestone or chalk pebble present in the assemblage, even if not showing pounding signs or scars from knocked-off flakes. 48 pebbles appeared in the assemblage with no pounding marks, with a ratio of 1:4.75 in favor of the battered and flaked hammers. These pebbles were included in the hammerstone assemblage they were deliberately brought to the site in order to be used in the flint knapping process. In addition, during the experiment it was observed that the pounding marks did not always appear immediately after the use of the hammer, but after a long usage. It was also observed that when the striking platform edge of the flint core is smoothed by abrading, in order to remove the sharp edging, little or no damage was caused to the hammerstone. Most of the wear appeared during repeated attempts to remove very thick flakes, where the resistance of the worked material is greater, or while working on an unprepared striking platform. Examples of such activity are the removal of a series of hinge scars from the striking face of the core, or attempts to remove the first opening flake (Davidzon and Gilead in press). A similar phenomenon was observed on the small retouching hammerstones; pounding marks were almost absent on their surfaces in spite of the fact that they were heavily used. The pounding marks are very specific, made by the sharp edge of the striking platform; thus, they differ from the signs usually found on the hand stones (manos) or grinding stones.

Unexperienced knapping during the same activity often results in heavily damaged or broken hammerstones, due to unprepared edge of the striking platform and the general inaccuracy of the blows. Thus, the skill level of the flint knapper can be inferred solely through examination of the hammerstones which were in use. Hammerstones used by novice flint knappers will produce a large amount of massive flakes and chunks. A skillful worker will produce less debitage, and the flakes struck off the hammer during the work will result from the general wear of the material; they will be thinner and less chunks will be present.

4.2.1. Hammerstone types

Some authors had attempted to classify the stone hammers and had suggested different levels of roundness in order to distinguish different types of hammers (Gonen 2000; 25). In the case of Beit

Eshel, the variety of roundness appears to be very diverse, and no clear pattern could be pointed out.

Therefore, the attribute of roundness was discarded, and the attributes of shape and size were the only utilized factors.

Similar study was carried on in the King Site, Georgia, were a flint workshop was found together with related pebbles which were used as a hammerstones. (Cob and Pope 1998). We divide the pebble hammerstones evident at the site of Beit Eshel into three types, following generally Cob and Pope typology:

1. Flat - this group consists of pebbles, whose width is at least twice their thickness. Type I

according to Cob and Pope (1998).

2. Round – this group represents pebbles whose width is less than twice their thickness or are

completely round. Type II according to Cob and Pope (1998).

3. Elongated - this group represents the hammers which are at least twice as long as they are wide,

with no regard to whether they appear flat or round in section. The members of this type show

pounding on the far edges, only, which brought us to a decision that they were used in a pestle like

fashion. Type III according to Cob and Pope (1998) represent lenticular pebble usually used as an

abrader. This type, as it is described by Cob and Pope, is not present in the Beit Eshel hammerstone

assamblage.

The pebbles within the site of Beit Eshel are found in different sizes. We suggest that the use of different sized pebbles represents an integral part and have different functions in the flint knapping

sequence. For that reason we created an additional, more relevant to the Beit Eshel assemblage, division of the hammers according to their size. The basic unit in our division is based on the ability of the pebbles to be hand-held (fist size):

1. Big - this group is represented by pebbles that measure more than 10cm in diameter.

2. Medium - this group is represented by pebbles that measure between 10 and 5cm in diameter.

The majority of the hammers fall within this range.

3. Small - this group is represented by pebbles that measure less then 5cm in diameter.

4. Abraders or preparators- this group is represented by mostly flat stones whose have signs of

heavy grinding on their circumference.

Big Medium Medium Medium Small Small Small Square Big Flat Abraders Elongated Total Flat Round Elongate Flat Round Elongated

1 0 2 1 0 0 0 0 0 0 3 2 4 8 2 0 0 0 0 0 0 14 3 0 0 1 0 1 0 0 1 0 3 4 1 0 1 0 3 0 1 3 0 9 5 4 12 3 0 2 0 1 2 0 24 6 5 10 3 0 3 0 1 8 0 30 7 4 2 0 0 1 0 0 1 0 8 8 1 0 1 0 0 2 0 0 0 4 9 1 2 0 0 1 0 1 1 0 6 10 3 4 0 0 0 0 0 1 0 8 11 2 2 0 0 0 0 0 0 0 4 12 3 3 1 0 3 0 1 1 0 12 13 1 1 1 1 1 1 0 2 0 8 14 5 4 1 1 1 0 1 4 0 17 15 0 0 0 0 0 0 1 0 0 1 16 0 0 0 0 0 0 0 0 0 0 17 6 7 1 4 3 0 3 8 0 32 18 4 22 2 0 2 0 1 5 0 36 19 3 5 0 1 0 0 0 0 0 9 Total 47 84 18 7 21 3 11 37 0 228 % 21 37 8 3 9 1 5 16 0 100

Table 12. Hammerstone types from the Beit Eshel hammerstone assamblage

4.2.1.1. Big hammerstones (10, 4%)

The results of the experiments demonstrate the fact that big hammerstones (more than 10 cm) were suitable only for striking a primary flake when opening small flint pebbles; even then, these usually break or flake severely. All attempts to strike a primary opening flake of a flint pebble with a hammerstone of a size equal to or smaller than the pebble were unsuccessful and resulted in a broken hammer. Their extraordinarily large size made them uncomfortable to grip and control. Probably that is the reason why the big hammerstones are rare in the original assemblage.

4.2.1.2. Medium sized hammerstones (149, 66%)

The medium sized hammers represent the majority of the hammers used in the site. They measure between 5 and 10 cm, which appears to be the most comfortable size when they are used in flint knapping. This group is divided into three groups, according to their shape: flat, elongated and round.

Medium flat hammerstones (47, 21%)

The flat hammers of that size were probably used only on special occasions, such as hard to reach places along the striking platform where the thin edge of the hammerstone is required. Alternatively, they may have been used due to a personal preference of the flint knapper. If so, they must represent the choice of specialists, because their usage demands more skill. Almost all of the medium-sized flat hammers are made of hard limestone (83%) (Appendix 2). Pounding is present on half of the stones in the assemblage and all of the pounding, when present, is located on the circumference of the stone.

Ten percent of the stones show abrading marks on their circumference along with pounding marks and scars. This is the same percentage as found on the round hammers, which shows that the stones were used randomly as abrading stones. The type of the scar termination most frequently found on the

stones is the feather type (21%) (Appendix 2), which, according to the experiment, was found to be indicative of specialist usage. As stated above, the flat stones in general are consistent with the presence of specialists. Another factor which supports the likelihood of specialist usage of the flat stones is the lower number of scars on their circumferences (47%) (Appendix 2).

Medium round hammerstones (No.84, 37%)

This group is the most common in the stone assemblage of Beit Eshel. The pounding marks are usually concentrated in one place on the circumference of the pebble, which indicates that the stone was fixed in the user’s hand and was not rotated during the work. The medium hammers (those between 10 and 5 cm) were found to be suitable for successful flaking and detaching blades, as well as de-cortication and removing core trimming elements.

The different sizes within a range of 5 to 10cm, were probably used according to the size of the core worked at the moment and the purpose of the blow applied to the core. For example, de-cortication and removing a thick series of unsuccessful hinge scars require bigger and stronger hammerstones then the simple removal of a thin blade. Those different sizes were used during the most of the knapping reduction sequence and provided the most of the tools in the tool kit of the knappers. For that reason, the stones of the range between 5 and 10 cm are the most common type in the Beit Eshel stone assemblage.

Almost all of the round stones are made of hard limestone (88%)(Appendix 2), which is the preferred material throughout the site. Pounding is found on the most of the stones (75%); abrading marks are also present but are not very frequent (7%), on the flat stones. On this type of stone, feather and hinge termination scar appear in an equal amount, but this observation will be discussed broadly in chapter 5.

Medium elongated hammerstones (No.18, 8%)

One may suggest intuitively that the elongated hammers were used in the same mode as the soft antler hammers - in an arch-like fashion. The experiment showed that the use of the elongated hammers in such a fashion results, in all cases, in a broken hammer. In cases where the hammer was used in an arc-like fashion, it is expected to find the pounding marks on the sides of the hammer. All of the pounding marks, however, are found on the extremities of the hammer. This is evidence of the use of the elongated hammers in a pestle-like fashion. Among the elongated hammerstones, 11 hammers were found without pounding marks. Half of the hammers in this category were made of chalk and the other half of hard limestone. Scars are rare (11%) on this type, hinge and feather types of scars, when present, show equal percentage. Two of the hammers show abrading marks, which indicates that they were occasionally used in the preparation of an abrading platform; thus they were also involved in flint blade production. Experiment reveal that elongated hammers are complicated to use; particularly when used in pestle-like action, but they generate large amounts of force, and probably were used while removing huge flakes, rather than obtaining thin blades. These consideration suggest that those hammers, were task-specific tool of the specialist’s tool kit.

Abrading stones (abraders) (21, 9%)

The abrading stones were used for preparation of the striking platforms of the blade cores by removing the overhang and especially blunting the edges by abrading it with the stone (Cob and Pope

1998;11). Archaeological evidence of using abrading stones is present in several site (Cob and Pope

1998: 6; Gallaghar 1977: 410). Such stones are made entirely from hard limestone; they are flat and within the range of 8 to 10 cm, the most comfortable size for being hand held. The abrading marks

reflect the way in which the stone was irregularly and unintentionally shaped along its entire circumference (Fig.9). On their flat sides, they show signs of polish over all of their surfaces, which we suggest are the result of continuous gripping of the stone during the working process.

This observation further supports our suggestion, as previously stated, that the stones were used in striking platform preparation (Cob and Pope 1998: 11) for abrading and scratching the sharp edge of the striking platform of the cores, in order to blunt and toughen it. Several small scars from knocked- off flakes show that the stones were used not only in abrading but that small blows were also involved during work. All of the cores that appear to be the result of a successful knapping sequence for removing a target blade show a prepared edge of the platform (Davidzon and Gilead in press). This type of stone we consider to be part of a tool kit - in other words a special task hammer, which could

be indicative of specialists working at the site.

4.2.1.3. Small hammerstones (48, 21%)

The small hammers form 20 percent of the whole assemblage. Their overall size measures less than

5 cm, according to our definition. Our suggestion, based on the experiment, is that they were used for shaping the actual sickle blades by retouching the truncation and the back of the blade while supporting it on an organic or stone anvil. They were also probably used in other fine retouching tasks as required, such as forming a core ridge or other core trimming element.

Like the previous size group of hammerstones, we divide the small hammers, according to their shape, into round and flat hammers. Elongated hammers were not found amongst the small hammers.

Figure 9. Abrading stones from the Beit Eshel hammerstone assemblage(a), note the deep striations that form the edge of the stones and the scar which appear on the right stone (b).

-Small round hammerstones (37, 16%)

Chalk as a raw material is not represented in the group of the small round hammerstones. This is because the small hammerstones need weight and consistency in order to perform well. Chalk is a

lighter, more porous and much more brittle material than limestone, hence the latter is the preferred material for small hammers.

Four of the stones show abrading marks, which show that platform preparation was also present in this advanced stage of the flint blade production. 15 hammer appear without pounding and only three hammers show scars from knocked-off flakes on their working face; this is because of the lesser amount of force applied to the flint and the fragile flint blades, which are more likely to break before the hammerstone itself. We can speculate that the stones which show scars on their surfaces were used in core shaping activity, which again points to a specialist’s signature.

Small flat hammerstones (11, 5%)

Only one of the hammerstones is made of chalk, which is an exception even though it shows marks of heavy usage (scars and pounding). Seven of the stones are found with pounding marks, all of which are found on the circumference of the stones. Two stones show abrading marks, and five stones (which comprise 50% of the flat small stones) scars from knocked-out flakes; three stones show hinge termination types of scars. The flat stones tend to show more scars on their circumference when compared to the group of the round stones, likely because of the more fragile shape of their profile.

4.3. The "Knocked-off Flakes"

The raw material of Beit Eshel was consists of river bed flint pebbles, which are characterized by its particular hardness and resistance to breakage. That fact was the reason why the hammerstones tend to break and flake often, producing a large amount of limestone flakes and chips. Flake is any fragment detached from a core, in our case hammerstone. Flakes bear specific attributes according to their initiation and termination phases (Cotterel and Kaminga 1987). These attributes are described

according to the flint and stone tool industries, where the flakes are intentionally struck from a core. In the Beit Eshel hammerstone assemblage, we relate the flakes which are accidentally struck off from the hammerstone surface during flintknapping work. The "knocked-off flakes" is a term used by

Gilead et al. (2004: 256); in this work we found it appropriate to use the same expression. In the limestone tool production series, if flake was struck intentionally, there is a conchoidal fracture. The presence of this type of fracture is characterized by a distinct striking platform and pronounced bulb of percussion (Gilead and Fabian 1995:282). The flakes which were accidentally knocked off the ends of the hammerstone during flint knapping results of splitting the material rather than chipping it. These flakes are defined as compression flakes caused by wedging initiation as a result of a blow at a 90 degree angle, like the angle of force usually applied to the core (Cotterel and Kaminga 1987: 688).

These appear flat on their ventral surface, have shattered or punctiform platforms of percussion, and the fissures which appear on the ventral side are concentrated in the point of the impact (Figure10a. and b.). However, at that point it is worth noting that in some lately excavated assemblages from

Nahal Komem (Fabian in prep.) and Tel Gishron (Nahshony and Aladjem in prep.), a number of limestone tools appear to be fashioned intentionally by this type of flaking in particular.

Figure 10. Knocked-off scars and flakes from Beit Eshel; a. Hammerstones with scars of knocked- off flakes on their face, the scars of the upper right hammer show series of knocked-off flake scars, which resemble the intentional flaking feature. b. Knocked-off flakes showing the typical flat ventral surface and the points of impact

Several different types of flakes were distinguished in the assemblage, which can be indicative of the comprehension of the method of work used at the site.

Square Massive Hinge Feathered NBK Abrading Chunks Chips Total 1 0 1 4 2 0 1 0 8 2 2 1 6 0 0 0 0 9 3 0 0 0 0 0 0 0 0 4 0 0 1 0 0 1 0 2 5 0 2 13 2 2 29 0 48 6 4 0 44 4 2 50 0 104 7 4 1 11 0 0 0 0 16 8 1 4 12 1 1 19 30 68 9 0 0 9 0 1 15 0 25 10 1 3 3 1 0 2 0 10 11 0 0 6 1 0 0 0 7 12 1 1 6 3 0 1 0 12 13 0 0 0 1 1 0 0 2 14 0 0 0 0 0 0 0 0 15 0 0 0 0 0 0 0 0 16 0 0 0 0 0 0 0 0 17 12 0 7 1 0 4 9 33 18 0 0 3 1 0 1 0 5 19 0 0 1 0 0 0 0 1 Total 25 13 126 17 7 123 39 350 % 7 4 36 5 2 35 11 100

Table 13. Flake types found at Beit Eshel

4.3.1. Abrading flakes (7, 2%)

The abrading flakes have abrading marks on their dorsal face. They show larger numbers than the scars on the abrading stones themselves found at the site. This fact suggests that some of the abrading stones were carried away, or were not found.

4.3.2. Massive flakes (25, 7%)

This group is represented by the massive, mostly primary flakes, sometimes appearing as almost half the size of a hammerstone. The sizes are between 7 and 10 cm and their thicknesses fall between 1-2 cm. They are usually produced of hard limestone.

4.3.3. Chips (39, 11%)

The chips represent the outcome of chipped small flakes from the hammerstone, while flint knapping.

4.3.4. Chunks (123, 35 %)

This group consists of huge chunks (between 10 and 5 cm.), which are result primarily from shattered and split hammerstones. On the basis of this category, it can be infer that the number of unskilled knappers was almost as large as, if not equal to, the number of specialists.

4.3.5. Natural backed knife (NBK) flake (17,5%)

The items of this group resembled the flakes with a cortical back, so named by Bordes (1961) these represent flakes that are struck from the same location where a previous flake has been struck. They are evidence of repeated use of the hammerstone even though it has been already ruined. This kind of action could mostly be characteristic of a novice flint knapper.

4.3.6. Hinge flakes (13, 4%)

This group is represented by flakes that are formed generally close to the point of impact when the crack turns abruptly to the face of the hammer, almost at a straight 90 degree angle, thus forming a

flake with a round section termination (Coterel and Kamminga 1987: 700 and Crabtree 1968: 466). In our case, these flakes are the result of hard blows on the core at a very blunt angle, where the force is insufficient to transfer to the core and initiate cracking, thus returning all of the force back to the hammerstone. The formation of those flakes is referred to the inappropriate and inefficient use of the hammer, thus represent work of inexperienced flint knapper.

4.3.7. Feather termination flakes (126, 36%)

The flakes of this group are formed when the crack is propagated almost parallel to the hammerstone surface and turns gently outwards to meet it at a very sharp angle, thus producing a very thin end of the outcome flake (Cotterel and Kaminga 1987: 699).

This type of flakes represents the largest flake class, indicating on extensive use of hammerstones took place at the site. During the experiment, it became clear that these flakes result from natural wear of the material, and are not a result that is related to inappropriate use of the hammer. For this reason we assume that these flakes are a sign of specialist work.

4.4. Anvils (N. 10)

The items of this group represent limestone pebbles that show pounding marks and depressions on

the flat surfaces, as a result of utilization. Our suggestion is that the anvils probably were used

during the shaping of the sickle segments. The blade was placed on the anvil for retouch, using the

small hammerstones. Three of the anvils from the site have deep pecking marks on their surfaces,

which result from a sharper object, and thus cannot be produced by simple pounding. These signs,

we believe, are signs of bipolar core reduction (Figure 12). When the cores become too small to be

knapped by free hand, the pointed distal end of the core was placed on the soft stone anvil and the

sequence continued. During the experiment, several cores were reduced according to this

technique, and the marks left on the anvil appear to be of the same pattern as the original. In

addition to the limestone anvils, 11 broken flint slabs were found at the site, which we believe

were used as anvils for the striking of the primary opening flake of the core, using block-on-block

technique (Fig.13). They also could have been used as anvils during retouching and forming the

sickle segments. They were broken into relatively small and comfortable to use and carry sizes

(20-40 cm.) , which led us to conclude that the slab fragments were deliberately broken in order to

be carried and used in the workshop

4.4. Spatial distribution and intra-site organization of the site

The site of Beit Eshel was excavated with a grid of 1x1m squares; in each square the amount and type of artifacts was recorded. This gave us the opportunity to model the spatial distribution of the stone artifacts. During that process the limestone artifacts were separated into two groups, complete hammerstones and hammerstone fragments, on the assumption that the tools should be found next to the workers and the fragments will be located further away, together with the waste from the flint production. The tools included all types of hammers and abrading stones, which were found unbroken.

The fragments were represented by chips and chunks knapped of the hammerstones.

The highest concentration of chips and chunks is found in squares 5 through 8. Most of the chips are found in the square number 8. The massive flakes are found mostly in square 17, which is the square of lowest elevation; therefore the massive flake concentration could also be related to the force of gravity.

The spatial distribution shows that a pattern can be observed in the site. It appears that the major distribution of tools is to be located on the periphery of the pit, with the center square ( 8) showing not

a single hammer present. The hammerstone fragments are found almost entirely in the center of the site, with the greatest concentration in the previously mentioned square 8 ( Figure 6; a and b).

. Figure 12. Illustration of bipolar reduction of a core placed on the original anvil found at the site

(note the depression marks on the item)

Figure 13. The suggested block-on-block technique of removing the primary flake of the core.

. 0 11 2 m

Figure 14. Distribution of: a. chunks and knocked-off flakes of the limestone hammers (N=311). b. limestone hammers (N=228).

0 11 2 m

Figure 15. Distribution of: a. flint cores (N=3800) b .complete blades (N=18 400) (Courtesy of Y.

Vardi)

The same spatial distribution of flint cores and blades was also observed. The assumption was that both of the flint categories will show the same pattern of distribution as the hammers, for the blades are the desired product and the cores are the source. However, the distribution of flint cores follows the same pattern as that of the hammers and the blades follow the same scatter pattern as the cores; however, blades are also found in the central and north-east section of the site, where the waste of the hammers was found. This could indicate the fact that not all of the blades (blanks) were collected in order to be turned into sickle blades.

Our suggestion of the site organization is that the flint knappers were sitting in a circle, around the location of square 8, facing each other, with not more than five people in the circle. The same organization of site activity areas was suggested by Binford (1987:449-512). The hammers and cores were placed close to the workers, and the small debitage were kicked toward the center of the pit.

Another possible scenario is that the small debitage was knapped on a rug and then discarded into the pits center.

5. Disscusion

5.1. Technique recognition

The objective of the experiments described previously in this thesis, was an attempt to recognize the means of flint knapping and technique used in the site of Beit Eshel and to estimate the level of craft specialization at that site. Those experiments showed no significant difference between the products, in terms of using different hammers. There is slight tendency, however, of some attributes of the hard hammer direct percussion technique, to resemble the original flint flake and blade found at the site (Chapter 2). Nevertheless, reconstructing the exact technique used at that site on the basis of flint knapping experiments only, appears to be virtually impossible.

Along with the flint implements, a limestone pebbles assemblage was found at the site. All of the stones found there are recognized as hammerstones (Gilead et al. 2004:260), which show signs of pecking and knocked-off flakes. The same features were observed on the hammerstones which were used in the experiment, a fact that allows us to regard these signs as made during the flint knapping process. Thus, based on the presence and analysis of the limestone assemblage, along with the slight prevalence of the hard hammer percussion technique, we can claim that this technique was used at the site of Beit Eshel .

5.2. Suggested operational sequence and tool kit

The operational sequence presented here is based only on the flint knapping experiments and the hammerstone assemblage found at the site. A variety of flint knapping methods used at the site are currently recognized by examining the entire flint assemblage (Gilead and Davidzon in press; Vardi and Gilead in press(b)) but they are beyond the scope of this thesis.

During the experiment, hammers of different materials were used to knapp flint pebbles in an attempt to replicate the technique used at the site. During the experiment it became clear that the entire knapping sequence for blade production could be done using limestone hammers only. The use of soft organic hammers, however, appeared to yield the same results as the limestone pebbles, when compared across some attributes, but remains of antler or bone hammers were nowhere present at the site of Beit Eshel. The experiments indicate that when organic hammers (antler and bone) were used by inexperienced knapper, they broke and splintered very often. We know that different skill level knappers are present at the site (Davidzon in press) consequently, we should expect to find some remains wherever the use of organic hammers took place. However, the only organic material found at the site of Beit Eshel consists of very small bone, which probably are remains of food which was consumed during the work in the site.

5.2.1. Operational sequence

After examining the original limestone assemblage and carrying out the experiments, we are able to suggest the following stages in the operational sequence:

Stage 1. Removing the primary opening flake, in order to form a striking platform ( Fig.13;

Chapter 4) by means of anvil block–on-block technique. Apparently a direct blow with a big hammerstone could produce the same result but it require greater skill in addition the large hammers which are appropriate to that use are rarely found (4%) within the Beit Eshel hamerstone assemblage (Chapter 4; Table12).

Stage 2. De-cortication and blade reduction sequence which is accomplished by knapping the striking platform using different size and shape limestone pebble hammer stones, according to the

Figure 16. Preparation of the edge of the striking platform of the core.

size of the core and the purpose of the strike . During the knapping sequence abrading stones were used frequently in order to remove overhang and blunt the edges of the striking platform (Figure 16).

Stage 3. When the core became too small to be held in freehand knapping fashion, we suggest that a bipolar anvil technique was performed (Figure 12). This technique allows the knapper to hold the core with only two fingers in desired position and continue the sequence, without risk of a finger injury. This technique was not obviously used in the reduction of all the cores, but we stress the possibility of the existence of this technique, according to the experiments and the presence of the limestone anvils which bear specific marks .

Stage 4. Sickle blade formation (shortening and backing) by using direct percussion with the small hammerstones on soft or hard anvils. At this stage suiFigure blades were selected, shortened by using the micro-burin and other techniques (Vardy in press), and formed into sickle segments by means of a small limestone or chalk pebble and an organic or stone anvil. In this case the size and the position of the anvil could vary from the personal decision of the performer.

The tool kit of the Beit Eshel flint knapper

The presence of tool kits is identified as an archaeological expression of craft specialization

(Michaels 1989; Clark 1986; 1991; Costin 1991). Those related to flint knapping manufacture are rarely documented or found in clear archaeological contexts. Generally, they are associated with burial sites (Benson 1980; Cob and Pope 1998; Polhemus 1987: 878-902; Seeman 1984). Most of our knowledge about tool kits is based on this scarce information, ethnological observation (Stout

2004; Galaghar 1977) and modern stone tool replication (Finlay 2008; Callahan 1979; Whittaker

2000; Crabtree 1972, 1970; Semenov 1964 and many others). Those sources show that main tool types used in the reduction sequence can be inferred. Beyond this, individual preferences, nature of the raw materials, desired products and, at last the choice of the tools in the tool kit could be determined also by cultural tradition.

In the tool kit presented here, we distinguish five types of tools ( Figure 17) which were used during the extensive flint working at the site of Beit Eshel, according to the stages of the reduction sequence presented above:

1. Flint slab anvils- The anvil was used during the first stage of the reduction sequence, where block-on-block technique was used for striking the primary opening flake of the flint core pebble, in

order to produce a striking platform. 11 flint slabs, all of them showing approximately the same dimensions (30X40 cm.), were found in the site.

2. Hammerstones- Most of the hammerstones found at the site were made of limestone pebbles, with a small percentage which were made of softer chalk pebbles. The different sizes, mostly between 5 and 10 cm in diameter, were used through the different stages of the reduction according to the size of the core and the knapper’s preference.

3. Abrading stones- These are flat stones which show deep striations from caused by rubbing and shaping the edge of the striking platform in order to blunt it for more successful flaking. The abrading stones were found in almost all of the squares with large amounts of flint implements, as stated in the spatial distribution chapter. This show that they were involved closely in blade production.

4. Small hammerstones- These were made entirely from limestone pebbles. They measure less than 5 cm in diameter, and our suggestion is that they were used during the final shaping of the flint blades and other light duty knapping tasks, such as forming ridges or other core trimming operations.

5. Soft stone light duty anvils- These anvils were usually made of whole or half large flat pebbles. They were used as a work tables and anvils during the final stages of blade removal, while using the bi-directional knapping technique (based on the pecking marks on the flat surface of the anvils Figure 12). The anvils used while shaping the final blades by means of the small hammerstones. Three anvils of that type were found at the site. The use of organic material anvils

4 2 1 3

6 7 5

10 8 9

Figure 17. The tool kit: Different size hammerstones (1-3), flint slab anvil (4), small retouching hammerstones (5-7), abrading stones (8,9) and a light duty limestone anvil (10).

was also considered, but it is very difficult to prove the use of organic anvils due to lack of preservation.

Since the tools described above were found in an actual workshop context and are described as elements of a flint knapper’s tool kit, any of these found in any Chalcolithic site suggests the presence of at least moderate flint knapping activity at the site.

5.3. Suggested skill level recognition and workshop organisation according to the hammerstone assemblage

The recognition of skill level, on the basis of flint knapping evidence, has been the topic of several recent works (Bamforth and Finlay 2008 and Finlay 2008); the objective is to recognize specialists’ work, thus allowing the possibility of identification of social organization according to skill and craft specialization.

Pelegrin (1990:118) has introduced two terms into the archaeological research of skill: knowledge

(connaissance) and know-how (savoir-faire). Knowledge consists of theoretical information only, while know-how is a muscle gesture memory which can be achieved only through practice (Apel

2008: 98). These two factors are integrated and expressed in the workshop debitage and tools. The general technique, such as for instance, the practice of using only one striking platform or the use of abrading stones, is connected to the knowledge factor, since it is a result of verbal communication; in other words, theoretical knowledge which is passed from the craftsman to the apprentice. This factor will be common to both the skilled and the novice, and it will be expressed at the same level in the site; all of the knappers, regardless of their skill level, will attempt to use single platform core and abrading stones. The know-how, on the other hand, distinguishes between the teacher and the apprentice because this factor is related to intuitive muscle memory, experience and repetition. We consider these two factors in the study of the skill level of the Beit Eshel flint knappers, as these are expressed in breaking and wearing patterns of the hammerstones. Several authors claim that knowledge and know-how are reflected by the raw materials or specific tool demand (Apel 2008: 99,

Hodder 1990: 155). The flint blade technology used in the site of Beit Eshel is the so-called

"ordinary production" ( Pelegrin 1990: 123), which is a technology based mostly on theoretical

knowledge; in other words one will be able to produce the desired product after understanding the logic behind the technique. Skill and experience could be achieved after a relatively short period of practice – unlike, for example, pressure flaking where experience and talent are crucial in achieving good results (Whitakker 2000 and Crabtree 1967).

The production of sickle blade blanks, despite its simplicity, is prone to mistakes, because knapping with a hard hammerstone is very difficult to control and every minor mistake will ruin the core. The experiments show that more capable knappers can be recognized from their cores, which show signs of attempts to repair the striking surface after a badly aimed blow, usually successfully.

The work of an unexperienced knapper is manifested in the large number of spoiled cores in very early stages and the large amount of broken hammerstones. The technique, however, improved after verbal explanation, and the skill level increased a very short time after that (Apel 2008:100).

We believe that the differences in the size of the limestone pebbles found at the site indicate differing functions in the flint knapping sequence. The most abundant group of stones is the medium size stones (between 5-10 cm; see Chapter 4), which show wide diversity in size. The different sizes are very difficult to sort into groups, because in our opinion they differ according to the flint knappers’ preferences and the specific purpose of the stone at the moment of its use, which are less standardized.

Earlier in this work it was mentioned that the skill level of the flint knapper could be recognized only through examination of both the finished product, the source the target blade and the core, respectively.

We also claim that the skill of the flint knappers could be recognized additionally by examination of their tools, especially their hammerstones. This will give a more accurate indication of the skill

level of the knapper, because the signs of the use signature will remain on the hammer even if a mistake was made on the core. The novice hammers are easy to recognize; they are simply broken to pieces, or show scars of large flakes usually showing hinge terminations. Amongst the Beit Eshel hammerstone assemblage the most frequent used hammerstones (the medium sized rounded hammers) showed equal amount of hinge and feather scars, indicating an equivalent use of these stones both of the skillful and novice flintknappers.

In most cases, the indicative attributes of the hammers will not be very different from their flint parallels. Usually the flakes which bear hinge terminations will be representative of the inexperienced flint knapper; the feather termination, on the other hand, will indicate the work of a more skillful and focused flint knapper (Bamfort and Finley 2008: Figure 2). In general, fewer flakes from the hammerstone and a majority of flakes showing feather termination found along with types of hammerstones (flat and abrading ;Chapter 4) which are considered to be specialists priority, will define a specialist work place. In contrast, a large amount of hinge flakes, natural baked knife flakes and large amounts of hammerstone chunks will define a less experienced artisan work place

In the examination of the complete hammerstones from the assemblage of the site of Beit Eshel, we notice that a low number of scars are present on the complete hammerstones; this phenomenon is related to the work of a specialist. One possible explanation of this fact is that probably only the stones used by specialists survived and the stones used by learners were completely broken into pieces, which explains the high percentage of chunks in the assemblage. In the hammerstone flake assemblage of Beit Eshel the percentage of feather termination flakes is higher than the hinged, natural baked knife flakes and the massive flakes. These two facts support the suggestion that the

Beit Eshel craftspeople were primarily specialists. The presence of high percentage of chips and feather terminated flakes also bear witness to the presence of specialists. Some have attempted to

recognize specific individuals through flint knapping signatures (Finlay 2008, Bamfort and Finlay

2008), while others have conducted experiments which show variation in the technique for the same replica artifact of two equally experienced knappers (Young and Bonichsen 1984). In regards to the hammerstone assemblage, the recognition of an individual style reflected in particular tools is even more blurred, due to the limited variability of the attributes that could be measured in order to recognize personal style.

The exact amount of hammers, used at the site, by novice and advanced knappers is impossible to claim due to a fact that most of the novice hammers probably broke and were reduced to chunks.

Nevertheless, we are at least able to claim the presence of different skill level on basis of the hammers found at the site. Thus we claim that hammerstones are a good measure of skill. The tools and blanks, however, naturally will be the best indicators of skill, when present.

6. Conclusion

The importance of the site of Beit Eshel is that it is a workshop site, or atelier (a smaller place associated with specific activity). It is the only Chalcolithic period site of that type and magnitude so far discovered in the Middle East, and probably elsewhere. The importance of special activity sites like burials, shrines and workshops is that, because of their public character, they could provide important clues for social organization and stratification. The flint assemblage of the site of Beit

Eshel has been extensively studied during the last years, including numerous refitting and experimental analyses (Gilead at all. 2004, Davidzon and Gilead in press). This thesis is part of that project research and an attempt to describe the level of craft specialization, based on the hammerstone assemblage evidence and knapping experiments which were carried out.

Craft specialization has been recognized as a factor of significant importance in the development of complex societies (Tosi 1984). The importance of craft specialization as a factor in the rise of the early states of the middle East was born from the assumption that the population had increased after a further labor division , which led the cities to be main centers of industry and trade (Childe 1950).

On the other hand, others have claimed that the growth of population is based in growth of agricultural productivity and craft specialists have minor impact on the process, even though it is an important element in recognizing the organization of the society (Trigger 1980). The presence of craft specialization indicates a growing society, wherein a larger number of individuals are busy with activities that are to the benefit of the whole society. This is in contrast to simpler societies, where people are living in small bands and are forced to posses individually the full range of essential skills needed in order to survive. In more complex societies, certain individuals perform a specific activity alone; the more diverse the range of these activities are, the more dependent are the members of the society upon their living environment. Such a society acts as a whole system, with widespread

mutual interdependence. Craft specialization is one of the signs of such a developed society. Some authors claim that craft specialization is dependent upon certain socioeconomic conditions and is related especially to stratified societies (Evans 1978 and Michaels 1989). Others have argued the same, while further claiming that craft specialization could be defined by part time specialists among the particular society who were mobilized to perform tasks for the benefit of their fellow man

(Quintero and Wilke 1995:26 and Cross 1993: 65).

Rodgers (1966:410) defines craft specialization is as follows:

"In my mind a specialist is an individual who holds a position or vocation because he controls a set of skills that most of his communal fellows do not control "

Another definition by Evans (1973: 55) is:

" I utilized the following points in my definition of craft specialization:

1. The manufacture of certain craft products is limited to a small percentage of the total number of individuals in any given community.

2. These individuals devote some of their productive time to the manufacture of these craft products.

3. Consequently, they must withdraw themselves from some or all of basic subsistence activities.

4. Thus, they must obtain some or all of their craft products "

It has already been stated that the development of craft specialization in general, and of the numerous blade technologies in particular, was developed following an increase in sedentary lifestyle (Wilke and Quintero 1995:26). The development of agriculture and other different crafts demanded the development of specific skills. During Chalcolithic periods, indicators of craft production are present in each dwelling unit or its dependent surroundings - little or no craft

production is allocated outside the immediate control of the household, a fact which points out that craft specialization was not an activity controlled by the society’s elite.

The recognition of the presence of craft specialization and its level of sophistication during the

Chalcolithic is a particularly important feature, since the organization of society in this period is still under debate (Levy 1995 and Gilead 1988).

During the Chalcolithic period different crafts developed: the beginning of metallurgy, with the introduction of copper work, and standardized clay and basalt vessels are found throughout

Chalcolithic sites. The question of craft specialization arises from the evidence of these crafts (Levy

1995; Gilead 1988; Gilead and Hermon in press; Davidzon and Gilead in press; Gilead 2003).

Activity areas are places where labor is located for performance of one or more stages of production sequence. Those areas are further defined as the atelier and the workshop. The difference between them is that with the atelier, a limited number of stages of the manufacture of a single craft is present, using only small portion of the site and usually associated with household activities. The workshop, conversely, is where the whole variety of stages of production is present and performed together, covering a considerable portion of the site ( Tosi 1984:24).

The site of Beit Eshel is defined as a specific activity area, because within its boundaries it consists solely of debitage and tools involved in flint blade production (Gilead at all. 2004; Davidzon and

Gilead in press; Winter 2006; 83-85). Archaeological indicators of craft specialization are also present in the form of tool kits at this site, together with related debitage and semifinished products which are used to witness at least partial craft specialization at that site.

It is found that flint knapping at the site was performed by specialists (Davidzon and Gilead in press), although novice flint knappers signatures are present too. The skill levels of the knappers at

Beit Eshel are reflected in the different breakage patterns of the hammerstones, from which we could

suggest that specialists and novices altogether were involved in the activity at the site. The fact that flint knappers of different skill levels were present at the site is evident for the fact that the activity of producing flint blades wasn’t only priority of craft specialists.

According to the working sequence presented in this paper, the technique used in Beit Eshel workshop appears to be very simple but highly productive (Davidzon and Gilead in press), using readily available materials both for manufacturing the desired product and the means which were involved in the process. The tool kit was no doubt used in the manufacture, but it was not something that a specialist carried always with them from site to site; it appears to be improvised rather then specifically crafted in order to be used in blade manufacture. It is something we should be calling an

Ad Hoc tool kit. The hammerstones were simply collected for the task from the nearby river bed, together with the flint pebbles used for blade production.

The different flint layers in the site, separated by sterile layers, suggests that the site was not used on a permanent basis. It is probably a seasonal temporary site, in order to answer the growing demand for sickle blades during the harvest season beginning in late spring (Watson 1979;

Turkowski 1969).

Appendix 1.

Blades and flakes selected from the experimental assemblage

Hard hammer Direct Percussion

Blades ID Width Thickness Length Shape Striking platform Size of Lipping Bulb of pattern platform percussion 1 1.6 0.7 5.4 straight cone 0.3 no big 2 2 1.3 6 straight small lipped 0.1 no small 3 2.7 1 6 straight semi-cone 0.5 no small 4 1.1 0.6 4.9 straight small lipped 0.1 no no 5 1.9 0.5 4.8 straight small lipped 0.1 no no 6 1.5 0.7 5.7 straight semi-cone 0.2 no no 7 1.7 0.8 6 curve small lipped 0.2 no small 8 1.3 0.5 5.2 curve semi-cone 0.1 no small 9 2.1 1 5 straight small lipped 0.1 no no 10 2.5 0.7 5 curve small lipped 0.2 present small 11 1.9 0.5 4.2 straight small lipped 0.1 no no 12 1.5 0.7 6.2 curve semi-cone 0.5 no small 13 1.5 0.4 4 straight semi-cone 0.3 no no 14 3 0.7 7.5 curve semicircle 0.2 no no 15 2.8 0.5 5.5 straight semicircle 0.3 no no 16 1.8 0.4 7 straight semicircle 0.3 no small 17 2 0.3 4.5 straight semicircle 0.5 no no 18 2.3 0.3 6 straight semicircle 0.1 no no 19 3 1 6 curve small lipped 0.5 present small 20 4.2 1.2 8.2 curve semicircle 0.1 present no 21 3.6 1 8.1 curve semi-cone 0.2 no no 22 3.2 0.8 7.9 straight semi-cone 0.3 no small 23 1.5 0.3 6 straight small lipped 0.2 present small 24 1.5 0.4 7 curve semicircle 0.7 present no 25 2 0.5 7 straight semi-cone 0.8 no no 26 2.2 0.6 6 straight semi-cone 0.5 no small 27 2 0.4 7 straight semi-cone 0.4 no small 28 1 0.4 4 straight semi-cone 0.1 no no 29 2 0.6 straight semicircle 0.1 present big 30 2 0.7 6 straight cone 0.1 no small 31 1.5 0.6 7 straight cone 0.4 present no 32 2.5 0.5 4 straight semicircle 0.2 present no 33 1.7 0.4 4.5 straight semicircle 0.6 present no

34 1.2 0.6 5.4 straight semi-cone 0.2 present small 35 1.6 0.8 6.2 curved semi-cone 0.3 no small 36 2.2 0.6 6 straight semi-cone 0.5 no small 37 1.4 0.6 5.2 straight punctiform 0.1 no small 38 1.8 0.6 6 curved punctiform 0.1 no small 39 1.2 0.5 5 straight small lipped 0.1 no no 40 1.4 0.5 6 straight small lipped 0.1 no no

Flakes ID Width Thickness Length Shape Striking platform Size of Lipping Bulb of pattern platform percussion 1 4 1.2 6 curve semi-cone 0.4 present big 2 2.2 0.5 4 curve semi-cone 0.4 no small 3 4 0.5 5 straight small lipped 0.8 present no 4 4.5 1.5 4.5 straight cone 1.2 present no 5 5.5 2 7 straight cone 0.8 no no 6 5 1.2 6.7 straight semicircle 0.4 present no 7 2.5 1 6 straight small lipped 0.6 present no 8 3 0.5 3.5 S semi-cone 0.4 present no 9 3 0.5 3.5 S semi-cone 0.8 present no 10 3.2 0.6 3.5 S semi-cone 0.6 present no 11 4 1.2 6 S semi-cone 0.7 present no 12 3.2 0.8 6 S semicircle 0.4 present no 13 3 1 4.5 S semi-cone 0.9 present no 14 3 0.5 4.5 S semi-cone 0.1 present no 15 4.2 0.5 5.2 S semicircle 0.6 present no 16 3.2 0.8 3 S semicircle 0.3 present no 17 2.5 0.8 3 S semicircle 0.5 present no 18 0.5 0.4 2.5 S semi-cone 0.4 present no 19 2.6 0.8 3 straight semicircle 0.5 present no 20 2 0.6 3.2 S semicircle 0.6 present no 21 1.2 0.4 2.8 straight semicircle 0.5 present no 22 1 0.5 2 straight semi-cone 0.6 present no 23 3 0.5 2 straight semicircle 0.5 present no 24 1.2 0.5 1.2 straight semicircle 0.3 present no 25 2 0.3 1 straight semicircle 0.3 present no 26 2.5 0.5 1 straight semicircle 0.8 present no 27 2 0.5 2 straight semicircle 0.2 present no 28 1.2 0.4 2 straight semicircle 0.2 present no 29 1.2 0.2 1.2 straight semicircle 0.3 present no 30 1.5 0.3 1 straight semicircle 0.1 present no

Soft hammer Direct percussion

Blades Striking platform Size striking Bulb of ID Width Thickness Length Shape pattern of platform Lipping percussion 1 3 0.5 5.5 straight semicircle 0.3 present medium 2 3 0.6 6.2 curved semicircle 0.4 3 2.5 1 6.5 straight semicircle 0.5 4 2.5 0.5 5.8 straight crushed 0.3 5 2.7 0.6 5.5 S semicircle 0.3 6 1.8 0.9 5.8 curved Semi +cone 0.4 small 7 2.8 1.2 6 S semi+cone 0.5 small 8 2 1 4.6 curved semicircle 0.6 9 2 0.5 4.9 straight semicircle 0.6 10 2.5 0.4 4.8 S semicircle 0.6 11 2 0.6 6.2 straight semi+cone 0.8 present 12 1.9 0.5 3 curved semi+cone 0.4 13 2 0.5 5.5 straight semicircle 0.3 present 14 2.2 0.5 6.2 straight semi+cone 0.4 present small 15 2.2 0.5 4.6 curved semi+cone 0.6 pronounced 16 1.7 0.5 4.9 straight semicircle 0.6 present 17 1.8 0.4 4.8 straight semi+cone 0.6 present small 18 1.9 0.5 6.2 S semi+cone 0.8 19 2 0.5 3 curved semicircle 0.4 small 20 2.2 0.5 6.5 S semicircle 0.5 present 21 2.2 0.5 5.8 curved semicircle 0.3 present 22 1.7 0.4 5.5 straight semi+cone 0.3 23 1.8 0.9 5.8 S semicircle 0.4 small 24 2 1.2 6 straight semi+cone 0.5 25 3 0.5 5.5 straight semicircle 0.3 present medium 26 3 0.6 6.2 curved semicircle 0.4 27 2.5 1 6.5 straight semicircle 0.5 28 2.5 0.5 5.8 straight crushed 0.3 29 2 0.5 5.5 straight semicircle 0.3 present 30 2.2 0.5 6.2 straight semi+cone 0.4 present small 31 2 0.5 3 curved semicircle 0.4 small 32 2.2 0.5 6.5 S semicircle 0.5 present 33 2.2 0.5 5.8 curved semicircle 0.3 present 34 1.7 0.4 5.5 straight semi+cone 0.3 35 2 1 4.6 curved semicircle 0.6 36 2 0.5 4.9 straight semicircle 0.6 37 2.5 0.4 4.8 S semicircle 0.6 38 3 0.6 6.2 curved semicircle 0.4 39 2.5 1 6.5 straight semicircle 0.5 40 2.5 0.5 5.8 straight crushed 0.3 41 1.8 0.9 5.8 S semicircle 0.4 small 42 2 1.2 6 straight semi+cone 0.5 43 2 0.5 4.9 straight semicircle 0.6

44 2.5 0.4 4.8 S semicircle 0.6 45 2 0.6 6.2 straight semi+cone 0.8 present 46 1.9 0.5 3 curved semi+cone 0.4 47 2 0.5 5.5 straight semicircle 0.3 present

Flakes Striking platform Size of Bulb of ID Width Thickness Length Shape pattern platform Lipping percussion 1 3 0.5 3.5 S semi+cone 0.7 present big 2 3 0.5 3.5 S semi+cone 0.4 small 3 3.2 0.6 S semi+cone 0.8 present 4 4 1.2 6 S semi+cone 0.6 present 5 3.2 0.8 straight semicircle 0.7 6 3 1 4.5 S semi+cone 0.4 present 7 3 0.5 straight semi+cone 0.9 present 8 4.2 0.5 5.2 curved semicircle 0.1 present 9 3.2 0.8 3 straight semicircle 0.6 present 10 2.5 0.8 3 S semicircle 0.3 present 11 0.5 0.4 2.5 straight semi+cone 0.5 present 12 2.6 0.8 3 S semicircle 0.4 present 13 2 0.6 3.2 straight semicircle 0.5 14 1.2 0.4 2.8 straight semi+cone 0.6 present small 15 1 0.5 2 straight semicircle 0.5 present 16 3 0.5 2 straight semicircle 0.3 present 17 1.2 0.5 1.2 straight semicircle 0.3 present 18 2 0.3 1 straight semicircle 0.8 present 19 2.5 0.5 1 straight semicircle 0.2 present 20 2 0.5 straight semicircle 0.2 present 21 1.2 0.4 straight semicircle 0.3 present 22 1.2 0.2 straight semicircle 0.1 present 23 1.5 0.3 straight semicircle 0.4 present 24 4.2 0.5 5.2 curved semicircle 0.1 present 25 3.2 0.8 3 straight semicircle 0.6 present 26 2.5 0.8 3 S semicircle 0.3 present 27 0.5 0.4 2.5 straight semi+cone 0.5 present 28 1 0.5 2 straight semicircle 0.5 present 29 3 0.5 2 straight semicircle 0.3 present 30 1.2 0.5 1.2 straight semicircle 0.3 present 31 2.5 0.8 3 S semicircle 0.3 present 32 0.5 0.4 2.5 straight semi+cone 0.5 present 33 2.6 0.8 3 S semicircle 0.4 present 34 2 0.6 3.2 straight semicircle 0.5 35 1 0.5 2 straight semicircle 0.5 present 36 3 0.5 2 straight semicircle 0.3 present 37 1.2 0.5 1.2 straight semicircle 0.3 present 38 2 0.3 1 straight semicircle 0.8 present 39 2.5 0.5 1 straight semicircle 0.2 present 40 2 0.5 straight semicircle 0.2 present

Blades and flakes selected from the original assemblage

Blades Striking platform Size of Bulb of ID Basket Width Thickness Length Shape pattern platform Lipping percussion 1 2 0.4 5 curve small lipped 0.1 present small 2 1 2.2 1.2 9.5 S no no 3 6 1.8 0.5 8.2 curve small lipped 0.2 present no 4 14 3 1.5 7.8 overshoot small lipped 0.2 present no 5 9 2.8 0.5 6.5 curve small lipped 0.3 present small 6 12 2.6 1 9.4 curve small lipped 0.2 present no 7 5 3.5 0.9 8 straight semicircle 0.5 no small 8 13 3.8 0.8 8.2 curve semicircle 0.3 present small 9 6 2.6 0.7 6.4 curve semi-cone 0.4 no small 10 6 2.4 1.2 7.8 overshoot semicircle 0.3 present small 11 6 3.3 0.9 8.7 straight semi-cone 0.3 no small 12 9 1.5 0.3 5.2 curve small lipped 0.2 no small 13 9 2 0.6 7 curve semicircle 0.5 present no 14 12 3 1.1 6.1 curve semi-cone 0.5 no small 15 sur 1.6 0.6 6.8 straight small lipped 0.2 present no 16 10 2 1 7.3 curve semi-cone 0.4 no small 17 7 1.8 0.8 9 curve small lipped 0.1 present small 18 7 1.8 0.9 8 curve small lipped 0.2 present small 19 7 2.2 0.6 7.2 curve semicircle 0.5 present small 20 1 2.2 0.7 6.9 straight semicircle 0.5 present small 21 sur 3 1 6.5 overshoot semicircle 0.3 present small 22 10 1.8 0.4 6.2 curve semicircle 0.2 present small 23 1 2 0.6 6.5 overshoot small lipped 0.1 present small 24 1 2.4 0.5 5.2 curve semicircle 0.3 present small 25 12 3 0.6 8.5 overshoot small lipped 0.1 present small 26 9 3.2 0.5 8.5 overshoot semicircle 0.3 present no 27 6 2.6 0.8 10.4 curve semicircle 0.3 present small 28 17 3 0.4 8.7 curve semicircle 0.6 no no 29 9 2.3 1.2 7.8 curve semicircle 0.4 no no 30 7 1.6 0.5 curve 31 7 2.5 0.8 9 curve semicircle 0.3 present small 32 8 1.9 1 9 overshoot small lipped 0.2 no no 33 9 3.1 1.3 8.6 curve semicircle 0.4 no no 34 2 2.4 1.3 7.3 overshoot small lipped 0.2 present small 35 17 2.4 0.6 9.2 curve semicircle 0.4 present no 36 14 3 0.4 8 curve small lipped 0.1 present no 37 9 2 0.6 8.8 curve small lipped 0.1 present small 38 6 2.9 0.7 6.5 curve semicircle 0.3 present small 39 7 2.8 0.6 7 curve semicircle 0.3 present small

40 12 2.4 0.5 6.8 straight semicircle 0.3 present small

Flakes Striking platform Size of Bulb of ID Basket Width Thickness Length Shape pattern platform Lipping percussion 1 9 6 1.5 5 S semi-cone 1.2 no big 2 12 4 1.2 5.5 straight semi-cone 0.8 no big 3 7 4.5 1.2 8 straight semi-cone 0.5 no big 4 6 4.2 0.7 2.8 straight cone 0.4 no big 5 12 3.5 0.7 4.6 straight small lipped 0.1 present big 6 6 3.5 1 5 curve 7 7 4.5 0.7 4 straight semi-cone 0.5 no no 8 7 5 1 5 S 1 no big 9 8 6 1 4 straight semi-cone 1 no small 10 7 4 1.3 3 straight semi-cone 1 no big 11 7 3.8 1 4.5 straight semi-cone 0.9 present big 12 7 3 0.8 5 curve semi-cone 0.5 no small 13 6 4 2 3.5 curve semi-cone 0.2 present small 14 7 3 0.7 3.5 straight semicircle 0.6 present small 15 7 3 0.4 4 straight semicircle 0.3 no no 16 7 2 0.5 2.5 straight small lipped 0.1 present no 17 6 6 1.8 7 straight semi-cone 1.9 no big 18 6 6 2 6.5 straight cone 1.3 no big 19 6 4.2 1.2 5 curve semi-cone 0.9 no small 20 9 4.2 1.1 5 straight semicircle 0.6 no no 21 6 4.5 0.6 3.5 curve semicircle 0.6 no small 22 9 4.7 0.7 6.4 curve semi-cone 0.3 no small 23 6 5 2 7 S semi-cone 1.7 no big 24 9 4.2 1.3 4.5 straight semi-cone 1.3 no big 25 7 4.5 1.5 4.5 straight semi-cone 0.8 no small 26 1 4 0.8 3.6 straight semi-cone 0.4 no small 27 6 4.5 1.3 4.3 straight semicircle 0.2 no small 28 7 3.2 1 3.5 curve semi-cone 0.5 no big 29 7 3 1 semicircle 1.1 no small 30 9 4 0.5 2.5 straight semicircle 0.6 no no

Appendix 2

Scar patterns and material of the hammerstones used in Beit Eshel

Medium Elongated Hammers No. of Abrading Location of No. Square Type of scars Pounding Material Scars marks pounding 1 6 2 Hinge Present Edge Chalk 2 5 2 Hinge Present Present Edge Chalk 3 2 5 Hinge Present Edge Limestone 4 2 2 Feather Present Present Edge Chalk 5 8 1 Feather Present Edge Limestone 6 5 Present Edge Limestone 7 5 2 Hinge+Feather Present Edge Limestone

Medium Flat Hammers Location of No. Square No. of Scars Type of scars Abrading marks Pounding pounding Material 1 10 Present Edge+Flat side Limestone 2 2 7 Hinge+Feather Present Edge Limestone 3 18 Limestone 4 14 2 Hinge+Feather Chalk 5 18 Present Edge Limestone 6 10 3 Hinge+Feather Limestone 7 14 Present Edge Limestone 8 17 2 Feather Present Edge Limestone 9 17 1 Feather Present Limestone 10 17 10 Hinge+Feather Chalk 11 10 2 Hinge+Feather Present Edge Limestone 12 3 4 Hinge Present Edge Chalk 13 13 2 Hinge Present Edge Limestone 14 9 2 Feather Limestone 15 9 Present Edge Limestone 16 17 2 Hinge Present Edge Chalk 17 13 1 Hinge Limestone 18 6 Limestone 19 5 1 Feather Present Edge Limestone 20 5 5 Hinge+Feather Present Edge Limestone 21 5 Limestone 22 12 Limestone 23 6 Present Present Edge Limestone

24 6 Limestone 25 6 Present Present Edge Limestone 26 17 1 Feather Present Present Edge Limestone 27 19 Limestone 28 14 Present Edge Limestone 29 13 Limestone 30 19 4 Hinge Limestone 31 4 Present Edge Limestone 32 9 Limestone 33 7 1 Feather Limestone 34 7 Limestone 35 9 Limestone 36 12 4 Feather Limestone 37 18 5 Feather Limestone 38 18 3 Feather Present Edge Limestone 39 12 1 Feather Present Edge Limestone

Small flat Hammers Abrading Location of No. Square No. of Scars Type of scars marks Pounding pounding Material 1 14 1 Feather Present Edge Chalk 2 14 1 Feather Present Edge Limestone 3 17 2 Hinge Limestone 4 7 0 Present Edge Limestone 5 7 0 Present Edge Limestone 6 17 0 Limestone 7 14 0 Limestone 8 5 0 Limestone 9 6 1 Hinge Present Present Edge Limestone 10 17 0 Present Edge Limestone 11 4 1 Hinge Present Present Edge Limestone

Medium Round Hammers No. Square No. of Scars Type of scars Abrading marks Pounding Material 1 3 4 Hinge+Feather Present Chalk 2 13 2 Hinge+Feather Present Limestone 3 9 2 Feather Limestone 4 9 Present Limestone 5 2 1 Hinge Present Limestone 6 17 Limestone 7 17 Limestone

8 18 Present Present Limestone 9 14 3 Feather Present Limestone 10 6 2 Hinge+fFeather Present Limestone 11 17 Present Limestone 12 6 Limestone 13 18 Present Limestone 14 17 1 Hinge Present Limestone 15 19 Limestone 16 18 2 Hinge Present Limestone 17 17 Limestone 18 5 3 Feather Present Limestone 19 6 2 Feather Present Present Limestone 20 6 Present Limestone 21 7 1 Hinge Present Limestone 22 6 3 Hinge+Feather Present Limestone 23 12 Limestone 24 17 Limestone 25 6 2 Hinge+Feather Present Present Limestone 26 19 Present Limestone 27 5 Limestone 28 12 Present Limestone 29 5 Present Limestone 30 5 Present Present Limestone 31 5 3 Feather Present Limestone 32 10 Present Limestone 33 6 3 Hinge Present Chalk 34 6 Present Limestone 35 6 Present Present Limestone 36 18 3 Hinge+Feather Present Limestone 37 17 2 Hinge Present Limestone 38 11 Present Limestone 39 18 1 Feather Present Limestone 40 17 Limestone 41 18 Present Limestone 42 18 1 Feather Present Limestone 43 12 2 Hinge+Feather Present Chalk 44 1 2 Hinge Present Limestone 45 6 3 Hinge+Feather Present Limestone 46 17 3 Hinge Present Chalk 47 5 Present Limestone 48 10 Present Limestone 49 18 Present Limestone 50 10 Present Limestone 51 10 Limestone 52 18 Limestone 53 12 Limestone 54 17 1 Hinge Present Limestone 55 7 Present Limestone

56 10 1 Hinge Limestone 57 2 Present Limestone 58 18 1 Hinge Present Limestone 59 17 Limestone 60 10 Present Limestone 61 18 1 Hinge Present Limestone 62 5 1 Hinge Present Limestone 63 6 Present Limestone 64 17 4 Hinge Present Limestone 65 14 Present Limestone 66 17 1 Feather Present Limestone 67 10 Limestone 68 5 4 Hinge Present Limestone 69 5 Limestone 70 5 Present Limestone 71 1 Present Limestone 72 1 Present Limestone

Small round Hammers No. Square No. of Scars Type of scars Abrading marks Pounding Material 1 6 Limestone 2 4 3 Hinge+Feather Present Limestone 3 5 Present Limestone 4 6 Present Limestone 5 18 Present Limestone 6 6 1 Hinge Present Limestone 7 6 Present Limestone 8 18 Present Limestone 9 6 Present Limestone 10 14 1 Hinge Present Present Limestone 11 18 Present Limestone 12 17 Limestone 13 7 Limestone 14 14 Present Limestone 15 6 Present Limestone 16 17 Limestone 17 17 Limestone 18 14 Present Limestone 19 4 Limestone 20 4 Present Limestone 21 18 Limestone 22 9 Limestone

References

Apel, J. 2008 Knowledge, Know-How and Raw Material. Journal of Archaeological Method and Theory 15:101-111

Bamfort, B. and Finlay., N. 2008 Introduction: Archaeological Aprproaches to Lithic Production Skill and Craft Learning. Journal of Archaeological Method and Theory 15:1-27.

Benson, M. 1980 A Prehistoric Subsistence Kit from CastleValley, Central Utah. Flint Knappers Exchange 3:6-15.

Binford, L. 1987 Researching Ambiguity: Frames of Research and Site Structure. In Methods and Theory for Activity Areas Research edited by S. Kent, pp. 449-512. Columbia University Press, New York.

Bonnichsen, R. 1977 Models for Deriving Cultural Information From Stone Tools. 60. National Museum of Man Mercury Series, Archaeological Survey of Canada., Ottawa.

Bordes, F. 1947 Etude comperative des differentes techniques de taille du silex et des roches dures. . L'Anthropologie 51:1-29.

-1961 Typologie du Paleolithique Ancient et Moyen. Memoire 1. Institute de Prehistoire deUniversite de Bordaux, Bordaux.

Clark, J. E. 1986 From Mountains to Molehills: A critical Review of the Teotihuacan's Obsidian Industry. In Economic Aspects of Pre-hispanic Highland Mexico, Research in Economic Anthropology Supplement., edited by B. L. Isaac, pp. 23-74. vol. 2. JAI Press Inc., Greenwich.

-1991 Modern Lacandon Lithic Technology and Blade Workshops. In Maya Tools, edited by T. R. a. S. Hester, J. H. Monographs in World Archaeology. vol. 1. Prehistory Press.

Cob, R. C. and Pope, M. 1998 Sixteen-Century Flintknapping Kits from the King Site, Georgia. Journal of Field Archaeology 25(1):1-18.

Costin, C. L. 1991 Craft Specialization: Issues in Defining, Documenting and Explaining the Organization of Production. In Archaeological Method and Theory, edited by M. Schiffer, pp. 1-18. vol. 3. University of Arizona Press, Tucson.

Cotterel, B. and. Kaminga., J. 1987 The Formation of Flakes. American Antiquity 52(4):675-707.

Crabtree, D. E. 1967 Note on Experiments in Flintknapping 4: Tools Used for Making Flaked Stone Artifacts. Tebiwa 10:60-73

-1968 Mesoamerican Polyhedral Cores and Prismatic Blades. American Antiquity 33(4):446-478.

-1970 Flaking Stone With Wooden Implements Science 169:146-153.

-1972 An Introduction to Flinknapping. Occasional Papers of the Idaho State University Miseum 28, Pocatello.

Cross, J. R. 1993 Craft Specialisation in non Stratified Societies. Research in Economic Anthropology 14:61-84.

Dag, D. and Goren.-Inbar N. 2001 An Actualistic Study of Dorsaly Plain Flakes: A Technological Note. Lithic Technology 26(2):105-117

Davidzon, A. and Gilead, I. in press The Chalcolithic Workshop at Beit Eshel: Preliminary Refitting Studies and Possible Socio- Economic Implications.

Dibble, H. and Pelcin A. 1995 The Effect of Hammer Mass and Velocity on Flake Mass. Journal of Archaeological Science 22:429-439

Evans, R. K. 1980 Craft Specialization in the Chalcolithic Period of the Eastern Portion of the Balkan Peninsular. Ph. D. Dissertation, University of California

Gallaghar, J. R. 1977 Contemporary Stone Tools in Ethiopia: Implications for Archaeology. Journal of Field Archaeology 4(4):407-414

Gilead, I. 1984 The Micro-Endscraper: A New Tool Type of the Chalcolithic Period. Tel Aviv 11:3-10.

-1988 The Chalcolithic Period in the Levant. Journal of World Prehistory 2:397-445.

-1995 Grar: A Chalcolithic site in the Northern Negev. Beer Sheva VII, Studies by the Department of Bible and Ancient Near East. Ben Gurion University, Beer Sheva.

Gilead, I. 2003 Review of the Late Neolithic and Chalcolithic Periods in the Southern Levant by J.L.Lovell. Journal of the Israel prehistoric Society 33; 218-224

Gilead, I. and. Fabian., P. 1995 The Knapped Stone Series.Grar: A Chalcolithic Site in the Northern Negev; 81-306. Ben-Gurion University Press, Beer Sheva.

Gilead, I. and Hermon. S. in press The Lithic Assamblages from Abu-Matar and Bir-es-Safady(Beer Sheva), Two Chalcolithic SItes in the Northern Negev. CRFJ, Jerusalem.

Gilead, I., Marder, O., Khalaila, H., Fabian, P., Abady, Y. and Israel, Y. 2004 The Beit-Eshel Chalcolithic Flint Workshop in Beer Sheva: A Preliminary Report. . Journal of the Israel Prehistoric Society 34:245-263.

Gonnen, S. and Goren.-Inbar, N. 1999 Soft Percussor Use at the Gesher Benot Ya'aqov Acheulian Site? Journal of the Israel Prehistoric Society 28:5-79.

Gonnen, S. and Goring.-Moris, N. A. 2000 Knifes, Bifaces and Hammers: A Study in Technology from the Southern Levant. Eurasian Prehistory 2(1):53-76.

Gophna, R. a. Friedman., N. 1993 The Flint Impliments from Tel En Besor. Tel Aviv 20:147-63.

Goren-Inbar, N., Zohar, I. and Ben- Ami, D. 1991 A new look at old cleavers –Gesher Benot Ya'aqov Mitekufat Haeven 24:7-33.

Hayden, B., and Hutchingson, K.W. 1989 Whither The Billet Flake? In Experiments in Lithic Technology 528: 235-259, edited by D. S. a. M. R. P. Amick,. BAR International Series.

Hermon, S. 2008 Socio-Economic Aspects of Chalcolithic (4500-3500) Societies in the Southern Levant- A Lithic Perspective,. BAR Series

Hodder, I. 1990 Commentary to the Technology in Humanities. Archaeological Review of Cambridge 9(1):154- 157.

Leakey, M. 1971 Olduvai Gorge, Excavations in Beds I and II 1960-1963. . Cambridge University Press, Cambridge.

Levy, T. E. 1995 Cult, Metalurgy and Ranked Societies- Chalcolithic Period. In The Archaeology of Society in the Holy Land., edited by T. E. Levy. Leicester University Press, London.

100

Levy, T. E. and Rosen, S. A. 1987 The Chipped Stone Industry In Shiqmim I: Studies Concerning Chalcolithic Societies In the Northern Negev Desert: 356; 281-294, edited by T. E. Levy, BAR International series, Oxford.

Lewis-Johnson, C. 1978 A History of Flint Knapping Experimentation. Current Anthropology 19(2): 337-372.

Macdonald, E. 1932 Prehistoric Fara, Beit-Pelet II. The British School of Archaeology in Egypt, London.

McCarthy, F. 1967 Asutralian Aboriginal Stone Impliments. Australian Museum, Sidney.

Mewhiney, H. 1964 A Skeptic Views the Billet Flake. American Antiquity 30:203-205.

Michaels, G. H. 1989 Craft Specialization in the Early Postclassic Colha. In Prehistoric Maya Economies of Belise, Research in Economic Anthropology, Supplement 4:139-183., edited by P. A. and I. McAnny. JAI Press, Greenwich.

Moore, A. M. T. 1985 The Development of the Neolithic Societies in the Near East. . In Advances in World Archaeology:1-69, edited by F. a. C. Wendorf, A.E., Academic Press, Orlando.

Navaro, I. a. Barberan., C. 1997 Trois Ateliers de Taille de Silex a Abu Hammid (Jordanie). In The Prehistory of Jordan II, Perspectives from 1997, edited by G. K. Gebel, Kafafi, Z. and Rollefson, G. O., pp. 383-393, Berlin.

101

Neuville, R, Mallon, A. and Koeppel, R.. 1934 L'Outilage en Silex: . In Tleilet Ghassul. Institut Biblique Pontificial, Rome.

Newcomer, M. H. 1971 Some Quantitative Experiments in Hand Axe Manufacture. World Archaeology 3(1):85-94.

-1975 "Punch Technique" and the Upper Paleolithic Blades. In Lithic Technology Making and Using Stone Tools, edited by E. Swanson, pp. 97-102. Mouton Publishers Hague.

Ohuma, K. and Bergman, C. 1982 Experimental Study in the Determination of the Flaking Mode. Bulletin of the Institute of Archaeology 19:161-170.

Ott, L.R., Larson,R., Rexroat, S. and Mendenhall, W. 1992. Statistics; A tool for the social sciences. Duxbury Press. Belmont, California.

Patterson, L. W. 1982 Replication and Classification of Large Size Lithic Debitage. Plains Archaeologist 11:50-58.

Patterson, L. W. and Sollberger, J. B. 1978 Replication and Classification of Small Size Lithic Debitage. Plains Archaeologist 23:103-112.

Pelcin, A. 1997 The Effect of Indentor Type on Flake Attributes. Journal of Archaeological Science 24:613-621.

Pelegrin, J. 1990 Prehistoric Lithic Technology: Some Aspects of Research. Archaeological Review of Cambridge 9:116-125.

102

Polhemus, R. R. 1987 The Toqua Site-40MR6: A Late Mississipian, Dallas Phase Town. 41. Report of Investigation, Department of Athropology, University of Tennessee, Knoxville.

Quintero, L. A. and Wilke, P. J. 1995 Evolution and Economic Significance of Naviform Core-and-blade Technology in the Southern Levant. Paleorient 21(1):17-33.

Rodgers, W. B. 1966 Development and Specialization: A Case From the Bahamas. Ethnology V:409-414.

Rosen, S. A. 1987 The Potentials of Lithic Analysis in the Chalcolithic of Northern Negev. In Shiqmim I, Studies Concerning Chalcolithic Societies in the Northern Negev Desert 356: 295-312, edited by T. E. Levy, British Archaeological Reports International Series, Oxford.

-1987 The Tabular Scraper Trade: A Model for Material Culture Dispersion. . Buletin of American Schools of Oriental Research 294:79-86.

-1997 Lithics After the Stone Age. Altamira Press, Walnut Creek.

Roshwalb, A. F. 1981 Prehistory of the Wady Gazzeh, A Typological and Technological Study Based on Macdonald Excavations Unpublished Ph. D. Dissertation, London University.

Sahlins, M. 1972 Stone Age Economics. Polity Press, London

103

Seeman, M. F. 1984 Craft Specialization and Tool Kit Structure: A Systematic Perspective on the Moidcontinental Flint Knapper. In Lithic resource Peocurmetn: Proceedings from the Second Conference on Prehistoric Chert Exploitation, Occasional Paper, Cenetr for Archaeological Investiagation, Southern Ilinois University at Casrbondale., edited by S. C. Vehik. Southern Ilinois University, Carbondale.

Shennan, S. 1988 Quantifying Archaeology. Harcourt Brace Jovanovich, Publishers San Diego, California.

Speth, J. 1975 Miscellaneous Studies in Hard Hammer Percussion flaking: the Effects of Oblique Impact. American Antiquity 40; 203-207.

Stout, D. 2002 Skill and Cognition in Stone Tool Production: An Ethnographic Case Study From Irian Jaya. Current Anthropology 43(5):693-722.

Tosi, M. 1984 The Notion of Craft Specialization and its Representation in the Archaeological record in the Early States in the Turanian Basin. In Marxist Perspectives in Archaeology: 22-52, edited by M. Spriggs,. Cambridge University Cambridge

Turkowski, L. 1969. Peasant Agriculture in the Judean Hills. Palestine Exploration Quarterly. vol.101, pp.21-33, 101-112

Vardi, J. and Gilead I. in press(a). "Side Blow - Blade Flakes from the Ghassulian Sickle Blade Workshop of Beit Eshel: a Chalcolithic Solution to a Neolithic Riddle" in 6TH Conference on PPN Chipped and Ground Stone Industries of the Fertile Crescent S.T.E.P.S (Studies in Technology, Environment, Production and Society)

104

Vardi, J. and Gilead, I. in press(b). On the Definition of Errors in Contexts of Craft Specialization: Krukowski Microburins from the Beit Eshel Chalcolithic Flint Workshop in Techniques and People: anthropological perspectives on technology in the archaeology of the proto-historic and early historic periods in the Southern Levant. Editors Rosen S. A. and Roux V

Watson, P. J. 1979. Archaeological Ethnography in Western Iran. The University of Arizona Press. Tucson.

Whittaker, J. C. 2000 Flint Knapping: The Making and Undesrtanding Stone Tools. . University of Texas Press, Austin Texas.

Wiloughby, P. 1985 . Spheroids and Battered Stones. World Archaeology 17:44-60.

Winter, R. 2006 Stone Tool Production Areas in the Southern Levant During the Proto-Historic Periods: Definition and Social Meaning. Unpublished MA Thesis, Ben-Gurion University of the Negev, Beer Sheva, Israel. (In Hebrew).

Young, D., E, and Bonichsen, R. 1984 Understanding Stone Tools: A Cognititve Approach. Centre of the Study of Early Man, Maine.

105

תקציר

לזיהוי של התמחות המקצועית בתקופה הכלקוליתית חשיבות רבה להבנת המבנה החברתי/כלכלי של ישובי התקופה. ידוע שבתקופה זו התפתחו לראשונה מספר התמחויות ייחודיות, , כגון יצור להבי מגל, תעשיית כלי חרס ובזלת. התמחות מקצועית ניתן לזהות על ידי גילוי כלים מיוחדים שאיכותם גבוהה יותר מכלים שכיחים שנמצאו בקונטקסט ביתי ושימשו לחיי היום- יום. אחד מסימני ההתמחות המקצועית הינו קיום של אזור הפקה של כלים מסוימים, ובמקרה המוצג בעבודת גמר זאת- אזור הפקה של כלי צור. הגדרת רמת היעילות של הסתתים, איכות התוצר ומגוון הכלים ששימשו לייצור כלי הצור, הם מרכיבים חשובים להבנת רמת ההתמחות המקצועית בתקופה הכלקוליתית. עבודה זו מבוססת על ממצאי האתר בית אשל שנחפר בשנת 2003 והוגדר כבית מלאכה )Workshop( להפקת להבי צור. מאז נערכים מספר מחקרים שמטרתם לקבוע מהן טכניקות הסיתות השונות בהן השתמשו ואת רמת המומחיות של הסתתים. עבודות אלו ממוקדות בשיטות הפקתם ועיבודם של הלהבים במונחים של רצפי הפחתה הבאים לידי ביטוי בפריטי הצור. העבודה הנוכחית היא ניסיון לבדוק את השיטות והאמצעים ששימשו להפקתם של להבי הצור באתר זה על בסיס חקר פריטי אבן הגיר ששימשו לסיתות. בתקופות הפרהיסטוריות הופקו כלי צור בשיטות רבות. מורכבות השיטה יכולה להעיד על יעילותה ועל רמת ההתמחות בהפקת כלי הצור. בתקופות הפרהיסטוריות נעשה שימוש בשלוש שיטות התזה עיקריות: במקבת קשה בהכאה ישירה )hard hammer direct percussion(, במקבת רכה בהכאה ישירה ) soft hammer direct percussion( ובהכאה בלתי ישירה )indirect percussion(. כל אחת מהשיטות האלו ניתן לזהות לפי הסימנים שהותירו על תוצרי ההפקה. למשל, צורת משטח הנקישה של הלהבים או צורת הפרופיל שלהם. למרבה הצער, הסימנים הללו לא תמיד ניתנים לזיהוי, לכן יש לנקוט בדרכים נוספות כדי לשחזר את שיטות הסיתות השונות. אחד הדרכים המקובלים והנפוצים בעולם הטכנולוגיות של כלי צור, הוא הפקה נסיונית של כלי צור. במסגרת עבודה זו הופקו להבים ונתזים מחלוקי צור שנאספו באפיק נחל באר שבע. התזת הפריטים מהגרעינים נעשתה בשלוש השיטות הנזכרות למעלה. הלהבים והנתזים שהופקו הושוו לדגימה של נתזים ולהבים ממכלול המקורי שנחשף בבית אשל. בשלב הראשון הושוו נתונים סטטיסטיים בסיסים – ממוצעי רוחב, אורך ועובי. בשלב השני הושוו יחסים משולבים בין משתנים. במסגרת זאת נבדקה השפעת גודל משטח הנקישה על אורך, רוחב ועובי הלהבים והנתזים. בנוסף, נבדק היחס בין האורך לרוחב של התוצרים. על סמך תוצאות של המבחנים הללו קשה להוכיח בוודאות מה הייתה הטכניקה בשימוש באתר של בית אשל. תוצאות המבחנים לא היו חד-משמעיות. מהשוואת קבוצת נתונים נסיוניים אחת ניתן היה ללמוד שנעשה שימוש בטכניקה מסויימת אך מהשוואת המדגם הארכיאולוגי לקבוצת נתונים נסיוניים אחרת, ניתן היה להסיק על שימוש בטכניקה אחרת. ככלל, נראה שיש העדפה לשימוש בהכאה ישירה במקבת קשה, אך בדיוק לא מספק בשביל לקבוע שזאת השיטה ששימשה באתר זה. לסיכום, השוואת מכלולי הניסוי למכלול הארכיאולוגי לא הניבה הוכחות חותכות דיין כדי לקבוע מהי שיטת ההכאה שהייתה בשימוש. למרבה המזל, נמצאה באתר ראיה אחרת המלמדת על השימוש במקבת הקשה ובהכאה ישירה: באתר נמצאו מספר רב של חלוקי גיר

106

שזוהו כמקבות ששימשו ליצור פריטי הצור. הגדרתם כפטישים נעשתה על ידי זיהוי סימני שימוש מסויימיים על היקפם כגון צלקות של נתזים שהותזו ללא כוונה במהלך העבודה וסימני נקישה. מקבות אבן הגיר נמצאו בכל שכבות האתר ובכל שטח האתר, דבר המוכיח שנעשה בהם שימוש נרחב. בנוסף, ניתן להבחין בדגם פיזור המעיד, כנראה, על סיתות שנעשה מסביב ללב איזור הפעילות ועל פסולת שהושלכה לכוון המרכז. במהלך המחקר נספרו סוגים שונים של מקבות האבן שכנראה שימשו לצרכים שונים במהלך הפקת כלי הצור. הפטישים חולקו לפי גודלם ולפי צורתם, ונמצא שלגודל השונה יש משמעות בהבחנת שלבים השונים במהלך הפקת פריטי הצור. פטישים גדולים שימשו להתזת נתז פותח/ראשוני, פטישים בגודל בינוני שימשו להפקת להבי המטרה הפטישים והקטנים ביותר שימשו לשיברור הלהבים. צורת הפטישים היא כנראה פועל יוצא של בחירה אישית או שימוש מוגדר, דבר המעיד על יעילות גבוהה ומקצועיות שאיפיינה את הסתתים באתר. דוגמה בולטת היא מציאתם של אבנים שטוחות, בדרך כלל בגודל אחיד, אשר כל היקפם מכוסה בחריתות שהם תוצאה של שימושם ככלי המכין את קצה של משטח הנקישה. כל הצלקות שנמצאו על הפטישים והנתזים שהותזו מהם נבדקו וחולקו לפי טיפוסים. טיפוסים מסויימים מציינים עבודה של סתת מקצועי וטיפוסים אחרים זאת של מתלמד. לפי הפטישים אפשר להעריך את רמת המומחיות של הסתתים שפעלו באתר. מכיוון שמדובר באתר שללא ספק שימש להפקת להבי מגל, חקר הפטישים שנמצאו בו,יכול לשמש כדוגמה להגדרת סוגי הפטישים ונתזי הפטישים באתרים אחרים מהתקופה הכלקוליתית. לבסוף, אוסף הפטישים מאפשר להגדיר מהי בקירוב ערכת כלים בה עשה שימוש הסתת הממוצע שפעל באתר. ניתן לאמור בהכללה שהסתתים באתר השתמשו בכלים פשוטים ומאולתרים. למרות שהלהבים באתר הופקו באופן מקצועי, על בסיס אוסף הפטישים והטכניקה שהייתה בשימוש לא ניתן לקבוע שבמקום פעלו מומחים מקצועיים בהיקף מלא. כנראה באתר בית אשל פעלו מומחים בהיקף חלקי בעונת האביב, בזמן הקציר כהיה צורך לייצר כמויות גדולות של להבי מגל. על בסיס חקר פטישי האבן ניתן לכן לקבוע שבחברה הכלקוליתית לא פעלו מומחים שעסקו רק בהפקת כלי הצור.

107

תוכן עניינים

תקציר באנגלית------3 רשימת איורים8------רשימת טבלאות 10------1. מבוא11------.1.1 מטרת המחקר11------.1.2 האתר הכלקוליתי בבית אשל11------.1.3 תולדות המחקר של כלי הצור מהתקופה הכלקוליתית 13------.1.4 גישה הניסיונית במחקר כלי הצור14------2. מתודולוגיה15------.2.1 שאלת המחקר15------.2.2 כוחות המשפיעים על היווצרות נתזי הצור ומאפיינים נלמדים15------.2.3 סיתות צור ניסיוני17------3. סיתות צור ניסיוני20------.3.1 שיטות סיתות20------.3.1.1 בחירת חומרי הגלם20------.3.1.2 הקשה ישירה במקבת קשה22------.3.1.3 הקשה ישירה במקבת רכה23------.3.1.4 הקשה לא ישירה 24------.3.2 ניתוח מאפיינים של פריטי הצור25------.3.2.1 אורך26------.3.2.2 רוחב26------.3.2.3 עובי 26------.3.2.4 גודל משטח הנקישה26------.3.2.5 פרופיל28------.3.2.6 ליפינג28------.3.2.7 צורת משטח הנקישה28------.3.2.8 גודל גבשושית הנקישה29------.3.3.1 השווה בין אורך, רוחב ועובי29------.3.3.2 השווה בין גודלי משטחי הנקישה33------

108

.3.3.3 השווה בין צורות הפרופיל33------.3.3.4 ליפינג35------.3.3.5 דגמי משטחי הנקישה37------.3.3.6 גודל גבשושית הנקישה38------.3.4.1 סיכום של היחסים שבין המאפיינים39------4. מכלול המקבות מבית אשל50------.4.1 מונחים ומכלולים מקבילים50------.4.2 מכלול המקבות מבית אשל51------.4.2.1 סוגי מקבות53------.4.2.1.1 מקבות גדולות55------.4.2.1.2 מקבות בגודל בינוני55------מקבות שטוחות55------מקבות עגולות56------מקבות ארוכות 57------מקבות לעיבוד משטחי הנקישה57------.4.2.1.3 מקבות קטנות58------מקבות קטנות עגולות59------מקבות קטנות שטוחות60------.4.3 נתזים שהותזו במהלך הסיתות60------.4.3.1 נתזים שהותזו ממקבות לעיבוד משטחי הנקישה60------.4.3.2 נתזים מסיבים64------.4.3.3 צ'יפס64------.4.3.4 גושים64------4.3.5 נתזים דמויי סכין טבעי בעל גב (Natural backed knife) 64------.4.3.6 נתזי הינצ'64------.4.3.7 נתזי דמויי נוצה65------.4.4 סדנים65------.4.5 פיזור מרחבי וארגון הפנים של האתר66------.5 דיון72------.5.1 זיהוי של טכניקת הסיתות לפי מכלול המקבות72------.5.2 סדר פעולה משוער ו''ארגז הכלים'' 72------109

.5.2.1 סדר הפעולה73------''ארגז הכלים'' של הסתת מבית אשל75------.5.3 רמת מיומנות וארגון בית המלכה בבית אשל בהתאם למכלול המקבות78------.6 סיכום82------נספחים85------הפניות 95------תקציר העברית------106 תודות------111

110

תודות למשפחה שלי למנחה שלי יצחק גלעד, על זה שהיה לו את הסבלנות לקרא ואפילו לתקן שגיות כתיב וניסוח לאנג'לה דוידזון וקובי ורדי על זה שחלקו איתי את כל הידע הנרחב שלהם לגבי בית אשל ותק' הכלכוליתית בכלל לארין סטאפני, ריצ'רד פלטצ'ר וקרני גולן על זה ששברו את הראש בניסיון לנסות להבין ולסדר את האנגלית שלי

111

אוניברסיטת בן- גוריון בנגב הפקולטה למדעי הרוח והחברה המחלקה למקרא ומזרח קדום מסלול לארכיאולוגיה

גישה ניסיונית לגבי תעשיית להבי המגל מתקופה הכלקוליתית: המקרה של בית המלאכה מבית אשל

חיבור זה מהווה חלק מהדרישות לקבלת התואר "מוסמך למדעי הרוח והחברה" )M.A(

מאת: אמיל אלג'ם מנחה: פרופ' יצחק גלעד

חתימת הסטודנט: ______תאריך:______חתימת המנחה: ______תאריך:______חתימת יו"ר הועדה המחלקתית: ______תאריך:______

נובמבר 2009

112

אוניברסיטת בן- גוריון בנגב הפקולטה למדעי הרוח והחברה המחלקה למקרא ומזרח קדום מסלול לארכיאולוגיה

גישה ניסיונית לגבי תעשיית להבי המגל מתקופה הכלקוליתית: המקרה של בית המלאכה מבית אשל

חיבור זה מהווה חלק מהדרישות לקבלת התואר "מוסמך למדעי הרוח והחברה" )M.A(

מאת: אמיל אלג'ם בהנחיית: פרופ' יצחק גלעד

חשון תש''ע נובמבר 2009

113