An Introduction to Stone Artifact Analysis

Home , Archaeological site, Eraillure, Ethnoarchaeology, Experimental archaeology, Feature (archaeology), Lithic

Chris Clarkson and Sue O’Connor

6 An Introduction to Stone Artifact Analysis

Introduction

Lithic analysis is a fundamental and often key component of contemporary archaeological practice of relevance to any region or time period where stone tools were employed in past technologies. For this reason, acquiring familiarity with the identification and analysis of stone artifacts is an important component of archaeological training and can be an important professional skill. Needless to say, there are numerous approaches to analyzing stone artifacts tailored to the vast range of topics being researched, and the one presented in this chapter may differ from those in use in some parts of the world or for particular time periods or assemblage types. Rather than review the huge diversity of approaches to lithic analysis, this chapter aims to arm the student of lithic technology with a set of principles to guide the construction of their research design, alert them to the philosophical underpinnings of various kinds of stone analysis, point to some simple but frequently overlooked issues of data management, provide an overview of some common laboratory techniques and analyses, and provide case studies and suggested readings that offer insight into both the process of actually doing stone analysis as well as drawing meaningful conclusions from the results. It takes a question-and-answer format in the hope that some frequently asked questions might be addressed in a straightforward manner.

There are many good reasons why archaeologists study stone artifacts. Primary An overview among them is the fact that stone artifacts are typically the most abundant and durable traces of past human activity of any of the artifactual remains archae- Why study stone ologists have available to study. In many cases, stone artifacts actually constitute artifacts? the only surviving traces of the behavior of people and our hominin ancestors

Archaeology in Practice: A Student Guide to Archaeological Analyses, Second Edition. Edited by Jane Balme and Alistair Paterson. Ap © 2013 John Wiley & Sons Inc. Published 2013 by John Wiley & Sons, Inc.

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 151 5/16/2013 10:15:36 AM that lived hundreds, thousands, and even millions of years ago (Semaw et al. 1997; McPherron et al. 2010). Stone tools continued to be used by some human groups until very recently or indeed to the present day (Sillitoe & Hardy 2003; Weedman 2006), and stone tools were often not immediately replaced by the introduction of metals due to the advantages they possessed as easily obtained and highly functional tools (Lechtman 1984; Rosen 1996; Greenfield 1999). Because stone artifacts survive under conditions that typically destroy most other human creations and castoffs, stone artifacts are ubiquitous in the landscape. Another reason for studying stone artifacts is that for most of human history, stone tools played a vital role in our day-to-day survival, in shaping the physical world to our various needs, and in signifying to others our identity and place in the world. They therefore constitute a vast and invaluable record of the diversity of strategies people devised to make a living, to solve problems, to communicate, and to live and compete with one another. As this chapter deals mainly with methodological issues and laboratory techniques, it offers little discussion of the sorts of theoretical frameworks that might employ lithic assemblages to answer some of the “big questions” in archaeology. Nevertheless, a great deal of thought has been given to such questions, including the place of technology as an integral aspect of cultural variability, adaptation and change (Lemonnier 1986; Pfaffenberger 1992; Bleed 1997; Schiffer & Skibo 1997), and the social, demographic, and evolutionary mechanisms giving rise to technological innovation (van der Leeuw & Tor- rence 1989; Bamforth & Bleed 1997; Shennan 2001; Kline & Boyd 2010). Like- wise, much research is directed at understanding the behavioral and physical factors governing variation within individual artifacts (e.g., fracture mechanics and the effects of reduction intensity) (Dibble & Whittaker 1981; Cotterell & Kamminga 1987; Dibble & Pelcin 1995; Pelcin 1997a, 1998; Shott et al. 2000; Macgregor 2005) as well as whole assemblages (such as patterns of artifact procurement, transport, use and discard) (Binford 1979; Shott 1989; Torrence 1989a; Nelson 1991; Kuhn 1995; Clarkson 2007). Another ongoing focus of research is the cognitive and selective underpinnings of technological evolution (Kohn & Mithen 1999; Hallos 2005; Stout et al. 2008), as well as the role of skill (Stout 2002, 2005; Finlay 2008; Nonaka et al. 2010; Eren et al. 2010a) and cultural transmission (Shott 1997; MacDonald 1998; Bettinger & Eerkens 1999, 2008). Numerous researchers have also explored the symbolic role of stone in com- municating social, political, and ideological relationships or differences (Wiess- ner 1983; Ingold 1990; Sinclair 1995; Wurz 1999; Harrison 2002), the role of social agency in stone artifact manufacture and use (Dobres 2000; Sinclair 2000), as well as stone artifacts as markers of gender (Gero 1991; Sassaman 1992; Dobres 1995; Walthall & Holley 1997). Much thought has also been given to the technological signatures of various mechanisms of trade and exchange (Renfrew et al. 1968; Ericson & Earle 1982; Zeitlin 1982; Torrence 1986; Peter- son et al. 1997; Torrence & Summerhayes 1997 among many more) as well as the markers and dynamics of lithic craft specialization (Charlton et al. 1991; Kenoyer et al. 1991; Shafer & Hester 1991; Hiscock 2005). Archaeologists have recently begun to explore technological variability using formal optimality Ap models drawn from evolutionary ecology (Bright et al. 2002; Brantingham 2003;

152 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 152 5/16/2013 10:15:36 AM Ugan et al. 2003; Clarkson 2007; Surovell 2009b; Clarkson et al., in press). Many of these studies are moving toward the development of new and innovative approaches to explaining assemblage variation.

A stone artifact is any piece of rock modified by human behavior, whether What are stone artifacts? intentionally or unintentionally. Although this definition could be applied to extreme and even ridiculous cases such as humanly modified landscapes, aque- ducts, or open-cut mines, it is most often used to signify portable, chipped, ground, or pecked stone objects created by a single or small group of individuals, and usually in the context of hunter–gatherer, pastoralist, early agricultural, early metal-using, or other nonindustrialized societies.

Most people are familiar with the simplest form of stone artifact manufacture How are they made? commonly portrayed in depictions of our early ancestors banging two rocks together. While this is, generally speaking, the way most stone artifacts were made, there is nothing simple about controlling the process to the degree that allows artifacts of specific shapes to be accurately and repeatedly produced from a block of stone, as was achieved by prehistoric artisans with sometimes startling finesse. The symmetry and regularity of some of the highest known forms of flintknapping can be astounding, as seen for instance in the fluted Folsom points of North American Paleo-Indians, the Solutrian points of Upper Paleolithic Europe, the flint daggers of the Danish Neolithic, the Gerzian ripple-flaked knives of Late Stone Age Egypt, or the obsidian eccentrics and polyhedral blades of Mayan and Aztec artisans (Figure 6.1). In reality though, most stoneworking tends to be far less sophisticated than these examples suggest (in terms of the precision and investment of labor), and literally involved the striking of flakes of varying shapes and sizes from a block of stone (a core), using a stone pebble (hammerstone) or some hard object (an indentor) such as a piece of bone, antler, or hard wood. Removing a flake from a block of stone creates a positive scar or ventral surface, on the flake, and leaves behind a negative flake scar on the core. The opposite side to the ventral surface on the resulting flake is called the dorsal surface. Cores are artifacts that possess only negative flake scars. Flakes that have had other flakes removed from their surfaces after they were struck from the core are called retouched flakes. Because flakes can be removed from the dorsal surface of a flake before or after it is struck from a core, the term retouched flake is reserved only for artifacts that show clear signs of flakes having been detached after the creation of the ventral surface, and hence scars must either derive from or modify the ventral surface in some way to be treated as retouch. The term nucleus will be used in the following discussion to refer to any stone from which flakes have been removed, whether flakes or cores. The process of fracture propagation that underlies flaked stone artifact manufacture is complex, and the effects of various core morphologies on the fracture path are not well understood. Yet it is the fracture path that ulti- Ap mately determines the morphology of flakes and cores, and archaeologists

An Introduction to Stone Artifact Analysis 153

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 153 5/16/2013 10:15:36 AM (a) (b) (c) (d)

2 cm 2 cm

(e) 2 cm 2 cm (f)

2 cm

Figure 6.1. Examples of some of the highest achievements in prehistoric stone artifact manufacture: (a) a fluted Folsom point; (b) an Upper Paleolithic Solutrean point; (c) a Danish Neolithic flint dagger; (d) a Late Stone Age Egyptian Gerzian ripple-flaked knife; (e) a Mayan chert eccentric; (f) Aztec obsidian pressure blades and cores (from Whittaker 1994, copyright © 1994, by permission of the University of Texas Press). From Flintknapping: Making & Understanding Stone Tools by John C. Whittaker, Copyright © 1994. By permission of the University of Texas Press.

have therefore begun to try and understand this process in detail. Due to the complexity of this subject, readers are directed to a number of papers that provide detailed overviews of fracture mechanics for archaeologists (Cotterell & Kamminga 1977, 1987; Phagan 1985), as well as more focused experimental investigations (Dibble & Whittaker 1981; Phagan 1985; Dibble & Pelcin 1995; Dibble 1997; Pelcin 1997a,c, 1998; Shott et al. 2000; Macgregor 2005; Dibble & Ap Rezek 2009; Clarkson & Hiscock 2011). Without delving into the details, it is

154 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 154 5/16/2013 10:15:37 AM possible to briefly describe some of the main principles and the most common fracture features that result. First of all, only a limited number of stone types are well suited to making flaked stone artifacts, and these generally possess three qualities: they areelastic , in that they will temporarily deform when force is applied to them; they are brittle, in the sense that they will fracture if the applied force exceeds the capac- ity of the material to deform elastically; and they are isotropic, meaning they are equally susceptible to fracture in any direction and will not preferentially fracture along particular planes. Cryptocrystalline or amorphous silicates (such as chert, chalcedony, flint), monocrystalline or microcrystalline silicates (crystal quartz and “milky” quartz), acrystalline silicates (such as glass and obsidian), and some larger grained and less homogeneous materials such as silcrete and quartzite all possess these qualities to varying degrees and are commonly employed in flaked stone artifact manufacture (Cotterell & Kamminga 1987; Kooyman 2000). In most forms of flaking, force is directed into the platform (i.e., any surface receiving force) of a nucleus with an indentor (any object imparting force to a nucleus) using one of three techniques: striking the nucleus at high velocity with either a hard indentor such as a hard hammerstone (hard hammer percussion) or a soft indentor such as a piece of wood, bone, soft stone, copper, or antler (soft hammer percussion), slowly applying pressure through a process called dynamic loading (pressure flaking), striking a positioned punch (indirect percussion), or applying compressive force by placing the nucleus on an anvil and striking it from above (bipolar technique) (Cotterell & Kamminga 1987; Kooyman 2000). Skilled flintknappers observe that in most flaking, force is generally directed into the nucleus using both an inward and outward motion (Crabtree 1972a; Whittaker 1994), creating both “opening” and “shearing” stresses in the nucleus (Figure 6.2a). Fracture occurs when stresses within the nucleus reach a critical threshold and break the molecular bonds holding the nucleus together. The most common form of fracture is known as conchoidal fracture, which begins from preexisting flaws in the surface of the nucleus close to the point of impact and creates what is known as a Hertzian cone, as illustrated in Figure 6.2b. The Hertzian cone propagates in a circle around the contact area and expands down into the nucleus in a cone shape at an angle partly dependent on the angle of applied force. If the nucleus is struck close to the edge, only a partial cone will be visible on the flake (Figure 6.2b). Whether or not a fracture will continue to propagate through the nucleus once a cone is formed (i.e., and not just leave an incipient cone in the nucleus) depends on whether the force of the blow is sufficient to accelerate and overcome the inertia of the material that is to be removed. Once fracture is initiated, a number of counteracting stresses created by the magnitude and direction of force (tensile, bending, and compressive stresses) will influence the path it then takes through the core, as well as the location of the free face of the core (Macgregor 2010). In conchoidal fracture, the path will typically first head into the core before diving back toward the free face, creating the bulb of force, and then stabilizing on a path that is more Ap or less parallel to the free surface.

An Introduction to Stone Artifact Analysis 155

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 155 5/16/2013 10:15:37 AM (a) Direction of force Opening Force Shearing Force

Platform Opening force

Shearing force

Fracture Exterior platform angle Nucleus

(b)

Complete Hertzian cone formed by spherical indentor near the center of a rectangular prism, and a partial cone formed near the edge

Platform

Lip Free Bulb of surface force

Fracture

(d) Feather Step Hinge Plunging (outrépassé)

(i) (ii)

Figure 6.2. Types and features of initiation and termination: (a) fracture forces; (b) Hertzian cones; (c) fracture initiations; (d) termination types. With kind permission from Chris Clarkson. Adapted from Cotterell and Kamminga (1987); Andrefsky and Bindon (1995).

Conchoidal flakes (i.e., those with Hertzian initiations) often retain a ring crack at the point of force application (PFA), and an eraillure scar just below the point of percussion on the bulb of force (Figure 6.3). Undulations in the fracture path also often leave compression waves on the ventral surface of Ap flakes. Fissures radiating out from the point of percussion are also often found

156 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 156 5/16/2013 10:15:38 AM Proximal Platform Ring crack (PFA)

Cone of force Bulb of force Erraillure scar

Dorsal Ventral

Medial Fissures

Lateral margins

Compression waves

Distal

Figure 6.3. Fracture features often found on the ventral and dorsal surfaces of a conchoidal flake (reproduced courtesy of the Trustees of the British Museum). With kind permission from Chris Clarkson.

on the ventral surfaces of flakes, but are most often seen on fine-grained materials. Force eventually exits the nucleus either gradually and at a low angle, creating a feather termination, or more rapidly and at around 90°, creating a step or hinge termination (Figure 6.2d). Not all fractures follow this path, however, and the fracture path sometimes travels away from the free surface and exits on the other side of the nucleus, creating a plunging or outrépassé termination (Figure 6.2d). Pelcin’s (1997c: 1111) controlled experiments have shown that when all other variables are held constant, increasing platform thickness will produce regular changes in termination type from feather through to hinge terminations, as the force becomes insufficient to run the length of the free face. The direction of force is also often implicated, as a determinant of either hinge or step terminations, but this proposition has not been tested under controlled circumstances. Others have suggested that thick platforms and inward-directed force are more likely to produce outrépassé terminations given sufficient force to initiate a fracture (Crabtree 1968; Phagan 1985: 237, 243). Less commonly, fracture will initiate behind the point of percussion, creating a bending initiation, which dives rapidly toward the free face without forming a Hertzian cone, and leaves a pronounced “lip” on the ventral edge of the platform (Figure 6.2c). Bending initiations are most common on flakes from nuclei with low-angled platforms and may sometimes have a fracture surface that often resembles a diffuse bulb, even though no bulb is present (Cotterell & Kamminga 1987: 689). Although it has long been thought that bending initiations are typically produced by soft hammer and pressure flaking, Pelcin (1997c: 1111) found that bending initiations were repeatedly created on cores with Ap low platform angles when blows were placed relatively far in from the edge,

An Introduction to Stone Artifact Analysis 157

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 157 5/16/2013 10:15:38 AM suggesting that their frequent association with soft hammer and pressure flaking is more likely a factor of the common use of these techniques in knapping cores with low platform angles (e.g., bifaces) than it is of either force or indentor type. Pelcin (1997b) was able to show that soft hammer flakes were on average longer and thinner than hard hammer flakes, and that this technique was therefore better suited to bifacial thinning than hard hammer percussion. Hence, the association between soft hammer/pressure and bending initiations is likely to be coincidental rather than causally linked. Nevertheless, experimental manufacture of punch blades from cores with a steep external platform angle (EPA) has also shown that certain force and indentor types can consistently produce bending initiations at high edge angles. Compression fractures created by bidirectional forces produce a wedging initiation that results in flattish fracture surfaces without a bulb of force (Figure 6.2b) (Cotterell & Kamminga 1987). Because compression fractures are typically initiated by particles driven into existing percussion cracks, flakes created through this process often exhibit battered or crushed platforms with cascading step scars on the platform edge (Cotterell & Kamminga 1987). Bipolar cores and flakes that have been rested on an anvil most commonly display this form of initiation. Because the anvil on which the nucleus is supported can also act like an indentor, bipolar flakes can at times exhibit platform and initiation features at both ends, such as crushing, dual bulbs of force, and bidirectional compression waves. When nuclei are stabilized on an anvil, problems of momentum and inertia can be overcome, increasing the likelihood of detaching a flake rather than simply moving the core away when it is stuck. This technique is therefore ideally suited to working very small cores (Hiscock 1982). Recent controlled fracture experiments have revealed that the closer the Hertzian cone is to the edge of the nucleus, and the lower the EPA, the less material needs to be accelerated away from the core, and hence the less force will be required to initiate a fracture (Dibble & Whittaker 1981; Speth 1974, 1981; Dibble & Pelcin 1995; Pelcin 1997a–c). The more these variables are reduced, however, the smaller the resulting flake will be. This relationship is illustrated in Figure 6.4a and can be seen to be a simple result of changing core geometry. Alternatively, increasing platform angle and striking further from the edge requires greater force input to initiate a fracture, but also results in larger flakes (Figure 6.4b). Increasing force input by too much can result in longitudinal splitting of the flake or crushing of the platform edge. At some point, increasing EPA and/or platform angle will reach a threshold at which the amount of force required to detach a flake will exceed the inertia of the nucleus itself and will result in moving the nucleus rather than detaching a flake (Phagan 1985: 247). At this point, force requirements can be reduced by decreasing EPA, platform thickness, or both, or by stabilizing the core on an anvil. Macgregor’s (2005) experiments have demonstrated that removing some of the mass of the free face (such as might occur through overhang removal for instance) allows a blow to be placed further from the platform edge (given the same amount of force) than would have been possible were it not removed (thereby detaching a larger flake). Furthermore, Macgregor found that the Ap morphology of the free face directly affected the morphology of the resulting

158 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 158 5/16/2013 10:15:38 AM (a) (b) Increasing platform thickness Increasing platform thickness

5 4 3 2 1 5 4 3 2 1

High exterior Low exterior platform angle platform angle

Increasing length

Figure 6.4. The effects of increasing or decreasing platform angle and platform thickness: (a) low exterior platform angle; (b) high exterior platform angle. With kind permission from Chris Clarkson.

flakes. His experiments demonstrated that features such as large preexisting step or hinge terminations on the free face will decrease the viable platform area at which fractures can be successfully initiated. In the case of preexisting step and hinge fractures, more force and the placement of blows further into the nucleus were required to successfully remove a preexisting step or hinge termination without adding another one. It can be expected then that as more step and hinge terminations build up on the dorsal surface, it will become increasingly difficult to remove them from the free face, as the viable platform area will become too small and the amount of force required too excessive to strike off a flake without shattering the platform, adding new step terminations, splitting the flake longitudinally, creating an outrépassé termination, or failing to initiate a flake altogether. Studies by Pelcin (1997a) and Dibble and Rezek (2009) have also demonstrated that varying the shape of the free face morphology affected the dimensions of the resulting flakes. These findings confirm the observations of flintknappers that setting up ridges running the length of the core face aids the production of longer, thinner, and more parallel-sided flakes (Crabtree 1972a: 31; Whittaker 1994: 106). A number of trade-offs therefore exist between the interdependent variables of platform size, platform angle, core inertia, force input, and nucleus morphology that knappers must manipulate to gain control over the fracture path and to extend the reduction of raw materials. A large number of strategies were employed in the past to modify force variables, to rectify problematic morphologies, and to prevent prematurely damaging the nucleus (Macgregor 2010). Ap Some of these strategies are listed in Table 6.1. These focus on variables that

An Introduction to Stone Artifact Analysis 159

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 159 5/16/2013 10:15:39 AM D ibble and D ibble and D ibble and D ibble and Pelcin K ooyman (2000) Reference D ibble and Whittaker (1981), Phagan (1985: 237), Pelcin (1995) Phagan (1985: 237) Speth (1974, 1981), Whittaker (1981), (1995), Pelcin (1997a–c) D ibble and Whittaker (1981), Phagan (1985: 237), Pelcin (1995) Phagan (1985: 247) H iscock (1982, 1996), Cotterell and K amminga (1987) Phagan (1985: 247) Speth (1972: 38), Phagan (1985) Macgregor (2005) Macgregor (2005) Crabtree (1968, 1972b: 60), K obayashi (1985), Pelcin (1997b) D ibble (1997) Crabtree (1968, 1972b), Phagan (1985), Pelcin (1997a), Andrefsky (1998), Negative effect R educes control over the fracture path by complicating platform morphology May result in smaller flakes; if blow is placed too close to the edge, the platform may shatter R educes the size, mass, and inertia of the nucleus May create step or hinge terminations to the free face Increases force requirements, removes mass more quickly from the nucleus, increases platform angle by removing more material from the platform end of the nucleus May result in smaller flakes; if blow is placed too close to the edge, the platform may shatter R educes control over the fracture path by complicating platform morphology L ess control over force delivery May shatter the nucleus through excessive force H arder to initiate fracture F lake may terminate abruptly if insufficient force; excessive force may result in a plunging termination or shattering the flake F lake may terminate abruptly if insufficient force to overcome the irregularity; excessive force may result in a plunging termination or shattering the flake Can increase curvature of the free face, resulting in more curved flakes; new platforms can encounter irregularities left by knapping from previous platforms T hinner flakes will have a greater chance of transverse snapping due to “end shock” Greater chance of longitudinally splitting the flake Increases the probability of transverse breaks due to “end shock” As above E PA and force requirements by removing E PA and force requirements by removing Positive effect R educes flakes from the platform surface R educes force requirements by reducing the amount of mass that must be accelerated Creates a new platform with lower angles Increases platform angle and strength; allows blows to be placed further in from the edge, creating larger flakes R educes chances of platform crushing and results in bigger (heavier) flakes R educes force input requirements by reducing the amount of mass that must be accelerated R educes flakes from the platform surface Increases the inertia of the nucleus by supporting it against a larger object Increases the inertia of the nucleus by resting it on an anvil and imparting compressive force using by (e.g., input force of speed the By increasing a longer or indentor, lighter and/or swing a faster the from leverage greater enables that indentor faster nucleus the to imparted be can force wrist), movement through overcome be can inertia its than Increases the coefficient of friction and creates microflaws in the surface R emoves projections, irregularities or preexisting step or hinge terminations from the free face by removing larger, thicker flakes R emoves problematic features gradually Projections, irregularities, or preexisting step or hinge terminations are removed from the free face from the opposite end R esults in thinner bulbs, and hence thinner flakes, creating higher cutting edge to weight ratios Produces a higher cutting edge to weight ratio while minimizing increases in platform angle Produces longer, thinner flakes with a higher cutting edge to weight ratio U se several of the strategies listed above, such as core rotation, stabilization or bipolar working, preparing the platform (faceting and overhang removal), change indentor type (e.g., soft hammer), and adjusting platform size (increasing or decreasing platform thickness and width) Strategy F aceting D ecrease platform area Core rotation O verhang removal Increase platform thickness D ecrease platform area F aceting Stabilize core Bipolar technique Increase speed of force input Grinding and/or faceting Increase platform thickness, platform angle, and force input Position blow to left or right R otate nucleus Soft hammer technique Increase platform width relative to thickness Setup arises on core face E xtend reduction of nucleus Common problems, solutions, and negative effects of various stoneworking procedures (with kind permission from Chris Clarkson).

Ap Table 6.1. Problem H igh platform angles (excessive force requirements) L ow platform angles (decreased flake size and increased platform crushing) L ow nucleus inertia Insufficient platform friction Poor free face morphology F lakes have insufficient cutting edge for weight N o replacement raw material

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 160 5/16/2013 10:15:39 AM D ibble and D ibble and D ibble and D ibble and Pelcin K ooyman (2000) Reference D ibble and Whittaker (1981), Phagan (1985: 237), Pelcin (1995) Phagan (1985: 237) Speth (1974, 1981), Whittaker (1981), (1995), Pelcin (1997a–c) D ibble and Whittaker (1981), Phagan (1985: 237), Pelcin (1995) Phagan (1985: 247) H iscock (1982, 1996), Cotterell and K amminga (1987) Phagan (1985: 247) Speth (1972: 38), Phagan (1985) Macgregor (2005) Macgregor (2005) Crabtree (1968, 1972b: 60), K obayashi (1985), Pelcin (1997b) D ibble (1997) Crabtree (1968, 1972b), Phagan (1985), Pelcin (1997a), Andrefsky (1998), Negative effect R educes control over the fracture path by complicating platform morphology May result in smaller flakes; if blow is placed too close to the edge, the platform may shatter R educes the size, mass, and inertia of the nucleus May create step or hinge terminations to the free face Increases force requirements, removes mass more quickly from the nucleus, increases platform angle by removing more material from the platform end of the nucleus May result in smaller flakes; if blow is placed too close to the edge, the platform may shatter R educes control over the fracture path by complicating platform morphology L ess control over force delivery May shatter the nucleus through excessive force H arder to initiate fracture F lake may terminate abruptly if insufficient force; excessive force may result in a plunging termination or shattering the flake F lake may terminate abruptly if insufficient force to overcome the irregularity; excessive force may result in a plunging termination or shattering the flake Can increase curvature of the free face, resulting in more curved flakes; new platforms can encounter irregularities left by knapping from previous platforms T hinner flakes will have a greater chance of transverse snapping due to “end shock” Greater chance of longitudinally splitting the flake Increases the probability of transverse breaks due to “end shock” As above E PA and force requirements by removing E PA and force requirements by removing Positive effect R educes flakes from the platform surface R educes force requirements by reducing the amount of mass that must be accelerated Creates a new platform with lower angles Increases platform angle and strength; allows blows to be placed further in from the edge, creating larger flakes R educes chances of platform crushing and results in bigger (heavier) flakes R educes force input requirements by reducing the amount of mass that must be accelerated R educes flakes from the platform surface Increases the inertia of the nucleus by supporting it against a larger object Increases the inertia of the nucleus by resting it on an anvil and imparting compressive force using by (e.g., input force of speed the By increasing a longer or indentor, lighter and/or swing a faster the from leverage greater enables that indentor faster nucleus the to imparted be can force wrist), movement through overcome be can inertia its than Increases the coefficient of friction and creates microflaws in the surface R emoves projections, irregularities or preexisting step or hinge terminations from the free face by removing larger, thicker flakes R emoves problematic features gradually Projections, irregularities, or preexisting step or hinge terminations are removed from the free face from the opposite end R esults in thinner bulbs, and hence thinner flakes, creating higher cutting edge to weight ratios Produces a higher cutting edge to weight ratio while minimizing increases in platform angle Produces longer, thinner flakes with a higher cutting edge to weight ratio U se several of the strategies listed above, such as core rotation, stabilization or bipolar working, preparing the platform (faceting and overhang removal), change indentor type (e.g., soft hammer), and adjusting platform size (increasing or decreasing platform thickness and width) Strategy F aceting D ecrease platform area Core rotation O verhang removal Increase platform thickness D ecrease platform area F aceting Stabilize core Bipolar technique Increase speed of force input Grinding and/or faceting Increase platform thickness, platform angle, and force input Position blow to left or right R otate nucleus Soft hammer technique Increase platform width relative to thickness Setup arises on core face E xtend reduction of nucleus Common problems, solutions, and negative effects of various stoneworking procedures (with kind permission from Chris Clarkson).

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 161 5/16/2013 10:15:39 AM are under the direct control of the knapper and tend to be visible archaeologi- cally. As should be apparent by now, fracture mechanics plays a preeminent role in shaping each individual artifact. It is important to keep this in mind when inferring the meaning of variation in flake and core form. While different forms could be interpreted as having stylistic or functional meaning, they might just as well relate to the methods employed in working various raw materials, to create flakes of different shapes, to prolong reduction, or to overcome certain difficulties. Pecking and grinding are quite different manufacturing processes to flaking. Pecking involves either dislodging grains or small pieces of material from the surface of a nucleus, or creating small and intersecting impact pits (incipient cones of force) over the surface of the nucleus until a specific shape is attained (Crabtree 1972a). Grinding, either on or with an abrasive material, likewise gradually wears away the surface of an artifact and usually results in the formation of many parallel striations (sometimes microscopic) aligned in the direction of the grinding motion that may blur preexisting fracture features or polish high points on the surface of the artifact.

How do I recognize Recognizing fracture features and the various techniques employed by past different techniques? knappers to rectify problems or improve their control over the fracture path, above all requires experience. Replicative flintknapping also provides a fast way of improving one’s identification skills by generating large number of flakes and cores showing a range of features created using known techniques. Flintknapping can also provide a means of generating hypotheses about how an assemblage might have been created, although analogical arguments of this kind do not provide tests in themselves of the various procedures used in the past. Only the archaeological record itself can provide such tests (e.g., refitting and attribute analysis) (Schindler et al. 1984: 176). Numerous resources now exist on the Internet, especially in the form of burgeoning YouTube videos, to help develop your understanding of reduction techniques and associated debris (Eren et al. 2010). Developing consistent sets of criteria to reliably identify specific procedures is often difficult, as the case of soft hammer percussion discussed earlier demonstrates, but fortunately, the recognition of some of the most common techniques is quite straightforward. A list of some of the commonly employed features used to identify various techniques is compiled from the observations of archaeologists and flintknappers and presented in Table 6.2 (e.g., Crabtree 1972a; Cotterell & Kamminga 1987; Ahler 1989; Hayden & Hutchings 1989; Whittaker 1994), although such features should be used with extreme caution. Entire assemblages should also provide a better “feel” for the use of dominant techniques than should individual specimens (Kooyman 2000: 78). According to replicative flintknappers (Crabtree 1972a; Newcomer 1975; Whittaker 1994), hard hammer techniques more frequently produce pronounced bulbs of force, compression waves and ring cracks, and expanding flake margins, whereas soft hammer techniques produce more diffuse bulbs, Ap sheared cones or shattered bulbs, flatter fracture surfaces, and narrower flakes.

162 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 162 5/16/2013 10:15:39 AM Pecking Pecked implement N /A N /A N /A N /A Variable N /A N /A N /A N /A Present Abrasion Ground implement N /A N /A N /A N /A Variable N /A N /A N /A Present Absent Bipolar flake F lat/pronounced N one/pronounced Absent Absent Parallel Crushed F lat Crushed Absent Platform Pressure flake Small D iffuse Subdued and widely spaced May be present R are/deep R are Common T hin and parallel F acetted Pronounced Much smaller Variable Platform Absent Soft hammer flake Variable size, tend to plano-convex D iffuse Subdued and widely spaced R are L ess common/shallow R are (5–10%) Common (20–60%) T hin and expanding F acetted/crushed Pronounced T hinner than hard hammer T end to feather Platform Absent 1%) < Flaking Hard hammer flake T end to large size and triangular Pronounced Pronounced and closely spaced May be present Common (95%)/shallow Common (60–80%) R are ( Variable Variable Variable T hicker than soft hammer Variable N /A N /A L ist of features and their supposed frequency in various forms of stone artifact manufacture (with kind permission from Chris Clarkson).

Ap Table 6.2. Feature Platform Bulb of force Compression waves Bulbar fissures E rraillure scar R ing crack Bending initiation Shape Platform scarring Ventral curvature T hickness T ermination Striations Impact pitting N /A, not applicable.

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 163 5/16/2013 10:15:39 AM These observations are borne out to some degree by controlled experiments (Cotterell & Kamminga 1987: 686; Pelcin 1997b), although it is difficult to know how well-controlled observations translate to archaeological assemblages in which a wide range of variables have presumably varied freely (Dibble 1997: 151). Studies of replicative flintknapping debitage, or the by-products of flaking, have produced arguments both for and against feasible identification of soft hammer working in archaeological assemblages (Mewhinney 1964; Touhy 1987). Pressure techniques can sometimes be quite distinctive and recognizable on retouched implements (Akerman & Bindon 1995). Likewise, some analysts believe they can recognize the flakes produced during pressure flaking from a combination of size, thinness, bending initiations, high ventral curvature, and a complex platform and dorsal morphology. As flintknappers point out, however, these same features can be created by percussion flaking, and cannot be considered diagnostic of any one technique in and of themselves (Touhy 1987; Whittaker 1994). Fine, parallel rows of elongate flake scars with small, discrete, and indented initiations are often the best way to identify pressure retouch on a piece. Bipolar flaking also presents difficulties for consistent identification (Jeske & Lurie 1992). Crushing of the platform edge, together with a flattish fracture surface and a battered distal end, is the usual criterion employed in identifying bipolar flakes, although not all flakes removed from bipolar cores possess these features (Cotterell & Kamminga 1987), and some possess platform features at both ends, or crushing in addition to fully formed Hertzian initiations. Negative scars can sometimes also appear on the ventral surfaces of bipolar flakes directed from either end as a result of the crushing blow. Bipolar flakes also are not easily separated from bipolar cores, but the presence of a single flat scar on one face may serve as a guide, whereas bipolar cores may tend to exhibit a number of scars on all faces. Identifying stoneworking techniques such as overhang removal, faceting, core rotation, retouching, and burination is generally more straightforward. Figure 6.5 illustrates the characteristics of overhang removal and faceting. Overhang removal can be identified by the presence of a series of smallish scars initiated from the platform surface onto the dorsal surface of flakes or the front edge of core platforms. Overhang removal is performed by firmly rubbing or gently tapping the edge of the core to remove the lip remaining after previous flake removals. Faceting looks much like overhang removal, but is oriented in the reverse direction, with smallish flake scars initiated from the dorsal surface onto the platform surface of cores and flakes. There is no real size cutoff between overhang removal or faceting flakes and other dorsal flake scars, and most analysts either employ an arbitrary cutoff (we use 15 mm), or simply use their intuition. Attempts have also been made to identify the distinctive features of overhang removal and faceting flakes so that they may be identified in archaeological assemblages (Newcomer & Karlin 1987). Faceting may serve different purposes or signify different things in different contexts. In some contexts, such as certain kinds of core reduction, it may signify the creation of Ap strong, steep platforms from which to strike large flakes (as in Levallois or blade

164 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 164 5/16/2013 10:15:40 AM Flake Core

Faceting Faceting

(Platform view)

Overhang removal Overhang removal

Flake Core (Dorsal and core face view)

Figure 6.5. Platform features indicative of various preparatory techniques (arrows indicate the location and direction of blows). With kind permission from Chris Clarkson.

production), whereas in other cases, it may be a platform maintenance strategy (such as to reduce EPA) or even an unintended consequence of core rotation. Core rotation is identified simply by the presence of a number of platforms on cores, or by the existence of truncated flake scars that originate from a point where a platform no longer exists. Core rotation can also be detected by the presence of redirecting flakes that preserve old platform edges on their dorsal surfaces at different orientations to the current platform (Figure 6.6a). Not every rotation of a core will result in a redirecting flake, and many rotations simply result in striking cortical flakes or flakes with complex platform morphologies (see “How do I measure flake reduction?”). Some of the potential uses of these three stoneworking techniques are listed in Table 6.1. Retouching is also easily identified if flake scars can clearly be seen to initiate from or modify the ventral surface (Figure 6.6c), but in cases where flaking is initiated from the dorsal surface without clearly modifying the ventral surface, it is often hard to be sure whether it is retouch or preexisting dorsal scars that are present. A classic case of this problem occurs in Australia where redirection flakes with old steep platform edges on their dorsal surfaces are misidentified as backing retouch. The key to the proper identification of retouch therefore is to locate the actual point of initiation of scars in order to determine whether they were formed before or after the creation of the ventral face. Lateral spalling of the margins, or burination, is another form of retouching that can be misidentified as preexisting dorsal scarring or old platforms (Figure 6.6b). Bifacial reduction is recognized on cores and flakes as flaking that is directed from either side of the platform edge or lateral margin (Figure 6.6e). Modern flintknappers have identified a set of criteria that they believe can be used to consistently recognize the debris resulting from reduction of bifacial cores Ap and bifacially retouched flakes. These include the high prevalence of bending

An Introduction to Stone Artifact Analysis 165

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 165 5/16/2013 10:15:40 AM (a) (b) Burinate retouch

Redirecting flake Burin spall Old burin scars

Dorsal Ventral

(d) Ventral retouch

Dorsal Ventral

(e) Bifacial retouch

Dorsal Ventral

Figure 6.6. Various reduction techniques and forms of retouched: (a) platform redirection; (b) burination; (c) dorsal retouch; (d) ventral retouch; (e) bifacial retouch. With kind permission from Chris Clarkson.

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 166 5/16/2013 10:15:41 AM initiations, pronounced curvature along the percussion axis, low platform angles, faceted and or ground platforms, and complex dorsal scar patterns that remove a portion of the opposite margin (Bordes 1972; Crabtree 1972a,b; Touhy 1987; Patterson 1990; Whittaker 1994: 196). Once again, it is unclear what proportion of bifacial debitage displays some or all of these features. A large number of recent studies have employed a range of techniques, such as mass and attribute analysis on replicated debitage analysis (Patterson & Sollberger 1978; Patterson 1982, 1990; Stahle & Dunn 1984; Ahler 1989; Odell 1989; Shott 1996; Austin 1997; Steffen et al. 1998), breakage patterns (Sullivan & Rozen 1985; Baulmer & Downum 1989; Prentiss & Romanski 1989), or a combination of these (Morrow 1997a; Bradbury 1998), to try and differentiate the various techniques used to create archaeological assemblages (such as hard and soft hammer, pressure, bifacial reduction and core vs. flake reduction), but with varying degrees of success (Prentiss & Romanski 1989; Shott 1994; Pren- tiss 1998; Bradbury & Carr 1999). A final technique worthy of mention is heat treating. Although the manner in which heat treating works is still not well understood, this technique aims to improve the strength of stone (Purdy & Brooks 1971) or ease of flaking (Crabtree & Butler 1964), and can result in dramatic changes to color, texture, and flaking properties. Heat treating is often discussed in the technological literature, but discriminating between deliberate and accidental thermal alteration is almost impossible, and requires careful attention to the context of heat application and the range of assemblage elements affected (Mercieca & Hiscock 2008). Thermal alteration often causes fine-grained materials to acquire a “greasy luster” or to change color. Alteration to the homogeneity of the stone can also be seen directly via electron microscopy when samples of the same stone with and without heating are compared (Purdy & Brooks 1971; Flenniken & White 1983). Excessive or rapid heating and cooling can result in the formation of pot lid scarring, crenated fractures, crazing, spalling, and color alteration, but the presence of these features does not necessarily imply that heating was unintended.

The first step in any analysis should be to determine what it is one is trying to Analyzing Stone Artifacts find out, and what analytical techniques will best provide the answers. We use the word “should” because no project can ever anticipate the full range of pos- Research design sibilities that will eventuate, and as new problems may spring up in the course of the analysis, these may require a different set of techniques or even the What am I trying to find development of novel methods. out?

Research questions can come from many sources: They may spring from the How do I develop a imagination fully formed, or coalesce gradually as one digests the literature research question? and examines its strengths and weaknesses. Good questions stand to shed new light on important issues in archaeology and can be answered through empirical research (i.e., stone artifact analysis) that can be undertaken within the time Ap frame available (Odell 2001a).

An Introduction to Stone Artifact Analysis 167

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 167 5/16/2013 10:15:41 AM Are some lithic analyses This depends entirely on whether a good match exists between the questions more meaningful than posed and the methods and data used to address them. Beyond this, there is no others? “right” way to analyze stone that will guarantee more meaningful results. The philosophical position taken, however, often leads us to choose various forms of analysis over others for the particular advantages they offer. The section below on classification provides an example of one such situation where our underlying “views of reality” may influence the sorts of data we collect and the types of classifications we employ.

When are experiments Experiments in lithic technology are a common and sometimes fundamental likely to be useful? means of replicating and understanding the factors generating certain lithic forms, creating patterning in lithic assemblages and testing certain hypotheses (Carr & Bradbury 2010). Indeed, controlled replicative flintknapping is reemerg- ing as one of the key means of determining past reduction techniques, rediscovering important and even counterintuitive steps and techniques in producing certain forms, and in answering a host of questions associated with efficiency, raw materials, skill, cognition, and the distinctive products and debitage created using various reduction strategies (Stafford 2003; Aubry et al. 2008; Eren et al. 2008, 2011; Marwick 2008; Stout et al. 2008; Bradley et al. 2010). Your research might benefit from performing some focused and controlled experiments in lithic technology that test aspects of your methods such as experiments that test indices of reduction tailored to your assemblages (Eren et al. 2005, 2008; Clarkson & Hiscock 2008, 2011; Eren 2009), test aspects of performance and efficiency (Mathieu & Meyer 1997; Eren et al. 2008; Sisk & Shea 2009; Clarkson et al., in press), examine taphonomic processes (McBrearty et al. 1998; Eren et al. 2010b), explore the experiential or even musical qualities of flintknapping (Cross et al. 2002), explore cognition and skill (Stafford 2003; Stout 2005; Stout et al. 2008; Eren et al. 2011), or replicate objects, processes, or debitage (Aubry et al. 2008; Marwick 2008; Mercieca & Hiscock 2008; Adams 2010; Bradley et al. 2010; Jeske et al. 2010), to name but a few potential uses for experimentation.

Classifying an Classification in archaeology, as in all fields, really only serves two purposes. assemblage of The first is to structure our observations into a limited set of groupings that stone artifacts can be said to be alike in a defined way. Grouping our observations in this way allows our results to be compared, contrasted, and explained. The second Why classify? purpose is to provide a set of terminological conventions, usually a set of named groupings or “classes,” that allow us to communicate about the world in a simplified and understandable fashion (Lyman et al. 1997: 15).

Are there rules of There are three basic rules on which successful classifications are based. The classification? first is that classifications should be based on sets of variables whose importance and means of combination is somehow determined from a body of theory. The second is that there must be recognizable similarities and differences between Ap the phenomena being observed in relation to the variables on which the clas-

168 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 168 5/16/2013 10:15:41 AM sification is based (Hill & Evans 1972; Dunnell 1986; Bailey 1994: 232). The third rule is that the classification must be exhaustive, or in other words, it must encompass all of the observed variation. Many classifications fail on these three counts, and particularly in the case of exhaustiveness. For instance, many classifications adopt the use of “miscellaneous” categories in which to place specimens that do not meet any of the classificatory criteria, rather than revising the classification to include unique objects. Obviously, the variables employed in a classification, as well as their means of combination, are of prime importance in determining its utility for a particular research design, its comparability to alternative systems, its sensitivity to variation, and its sufficiency as an exhaustive and unambiguous description of variation.

There are numerous forms of classification ranging from ad hoc folk classifica- What are the different ?tion to systematic forms, and a potentially infinite range of variables on which types of classification to base any system of division. We can usefully distinguish three elements of classification that all find their way into lithic classification to some degree. A first principle relates to the criteria used to assign objects to a particular class, and it is possible to differentiate between monothetic and polythetic class construction. In monothetic class construction, objects belong to a certain class only if they possess all of the specified attributes (or properties) that define that class (Figure 6.7). The implication of this type of classification is that an object can be assigned to a particular class according to the presence of any single attribute, because it is assumed that if it possesses one, it must possess them all. Polythetic classification, on the other hand, is better suited to dealing with variation in that it requires that an artifact possess only one or more of the total number of defining properties to belong to a class, and that no artifacts possess all of them. The implication of this form of classification is that a single property does not always provide an accurate basis on which to assign an artifact to its proper class, and classification must instead take into account the total combination of attributes and their overall weighting in the system. Polythetic classifications require explicit definition of each defining property so that -dif ferent analysts do not accidentally produce different classifications. To give an example of the kinds of classes that each system might produce, as well as the ways in which properties can be combined to form distinct classes, two monothetic and two polythetic classes are shown in Figure 6.7. The first three specimens are assigned to the monothetic class “High-angled concave and nosed end scrapers,” on the basis that all members possess high edge angles, concave edges, nosed projections, and distal retouch. In this system, the presence of any one of those features will identify the artifact as belonging to that class, as they are mutually exclusive and do not occur in any other monotheti- cally defined class. The second class is also a monothetic class with a different set of attributes that are also held in common by all its members. In contrast, the next two classes are constructed using a polythetic system in which not all properties are held in common by all members, though at least one property is held in common by all (e.g., a distal point for “Points,” and both pronounced Ap bulbs and dorsal bulbar trimming for “Tulas”).

An Introduction to Stone Artifact Analysis 169

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 169 5/16/2013 10:15:41 AM Retouch properties Angle Edge Shape Feature Location Surface Ventral Distal Low High Straight edge Concave edge Convex edge Notched Specimen Nosed Distal point Lateral Distal and lateral proximal and Distal Dorsal only Ventral only Bifacial Pronounced bulb Bulbar trimming High-angled concave and nosed end scraper 1 X X X X 2 X X X X 1 3 X X X X Low-angled convex and notched side and end scraper 4 X X X X

Monothetic 5 X X X X 2 6 X X X X Point 7 XX X X X 8 X X XX X 3 9 X X X X X X Tula 10

X X X X XX Polythetic 11 X X X X XX 4 12 X X X X XX X

Retouched flakes

Figure 6.7. An illustration of some typical classifying variables and their means of combination under monothetic or polythetic classification. With kind permission from Chris Clarkson.

Each system has its own strengths and weaknesses. For instance, monothetic classifications are simple and straightforward to design and implement, but suffer difficulties when dealing with variation and complexity. It must be acknowledged that monothetic classes are also high-level abstractions in the sense that they impose rigid boundaries around phenomena that may in fact form a continuum. Polythetic classifications are better able to deal with variation, but may embody too much flexibility. Unless the various defining properties are rigorously defined and weighted, there is great potential for each researcher to come up with a different set of assignments. A second principle of classification is that objects can either belong exclusively to a certain class and no other, or they can belong to many classes, sometimes with a “membership weighting.” This division again determines how variation is dealt with. Exclusivity means that variation is suppressed to fit unique objects into a limited set of classes, as is the case for Classes 1 and 2 in Figure 6.7. Alternatively, overlapping classifications mean that variation is allowed expression and that unique objects are recognized for their potential to fit into any number of classes depending on which attributes are given prominence in the classificatory scheme. For example, Classes 3 and 4 in Figure 6.7 share features in common with specimens found in other classes, and we Ap could theoretically assign them to all of the classes with which they overlap – if

170 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 170 5/16/2013 10:15:42 AM we were to find this useful for some purpose. By placing a weighting on specific variables, however, it would be equally possible to narrow the range of classes into which they fit. Tinkering with the choice and weighting of variables allows the degree of overlap between each type to be expanded or narrowed. A third principle centers on whether classes have some sort of structure imposed on them, such as a hierarchical arrangement of the sort seen in Linnean biological classification. Unordered classifications impose no - prece dence or structure and treat each class as though it is “on the same level.” In stone artifact classification, hierarchical classification is best seen in classifications that attempt to order each stage of the reduction process in terms of the chronological sequence in which it takes place (Hiscock 2001). For instance, all of the specimens in Figure 6.7 also belong to the higher-order grouping “Retouched flakes.” Andrefsky (1998: 65) provides an example of the way in which either monothetic and polythetic methods could be used in the construction of a hierarchical classificatory system.

Classifications can be undertaken by manually allocating objects to a class using What are the different ?a set of variables whose importance is deduced from theory, or using statistical methods of classification techniques that find clusters within the data. It has sometimes been claimed that these techniques can provide an objective means of “discovering” natural types (Spaulding 1953), but as Dunnell (1971) points out, while statistical techniques may indeed derive attribute clusters from empirical data that are of utility for certain problems, they cannot discover types with an independent reality, as the robustness of class divisions ultimately rests on the value and weighting of the attributes employed. Statistical types are therefore always constructed at some level.

Choosing between classificatory systems is not straightforward, but depends How do I choose between ?on the sorts of data you want to collect and the types of questions being classificatory systems addressed. Our theoretical position may also sway our decision to use one form of classification over another, as these tend to suite certain approaches more than others. Most classifications in use today are built around one or other of two alternative views of the world that have important consequences for the way things are classified. The first was discussed at least as long ago as Plato’s time and is today called “essentialism.” This idea holds that the world is divided into real, discontinuous, and immutable “kinds.” This notion underlies most typological constructions, which hold that artifacts, particularly retouched implements and certain types of flakes and cores, can be separated into discrete and mutually exclusive kinds (Dunnell 1986; Dibble 1995; O’Brien 1996; Lyman et al. 1997; Hiscock 2001, 2002b). In the context of stone artifact manufacture, essential forms are often thought of as “mental templates,” or combinations of traits that are favored by the maker. Variation is seen as a consequence of the imper- fect realization of the conceptually perfect form and is usually attributed to Ap differences in raw material properties or individual skill levels (Dunnell 1986).

An Introduction to Stone Artifact Analysis 171

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 171 5/16/2013 10:15:42 AM In practice, individual artifacts are usually assigned through comparison with illustrated “type specimens,” often with an accompanying description, or sets of artifacts that exemplify the ideal forms for each class. The essentialist metaphysic lends itself to the use of mutually exclusive, unordered, monothetic classes of the sort typically employed in older but commonly used typologies, such as Francois Bordes’s (1972) 63 Mousterian tool types. An alternative view of reality is called “materialism” and holds that all phenomena are unique, often arranged as continuums, and that “kinds” are illusory and imposed on reality rather than extracted from it. Materialist classifications therefore set out to find ways of depicting variation as well as central tendency, and treat observational units as units of measurement rather than real kinds. In archaeological classification, the materialist metaphysic has been particularly embraced by evolutionary and processual schools. The processual school has argued the position that there is no natural, single, or “best” typology and no inherent meanings to be discovered in an assemblage of artifacts (Hill & Evans 1972). Rather, the meaning imposed on archaeological phenomena derives from a priori problems, hypotheses, and other interests (Hill & Evans 1972: 252). Hence, processual archaeology encouraged selection of attributes derived from the discipline’s problems that will lead to classifications that are useful in addressing those problems. Evolutionary archaeologists make the additional claim that most phenomena are in a state of constant change (as in cultural phenomena and artifacts themselves), and that classifications may be enhanced by somehow factoring time, distance, and/or historical relatedness into their formulations (Lyman & O’Brien 2000). This can be clearly seen for instance in the changes that take place in the form of an artifact as reduction continues (Dibble 1995; Hiscock & Attenbrow 2002, 2003, 2005b; Clarkson 2005b).

How do I build my own By this stage, we hope we have convinced the reader that classification requires classification? some thought, that no classification is “real” or fixed, and that it is most useful if approached as a tool for measurement, description, and problem solving. Important points as far as stone artifacts are concerned are that different levels of classification can exist, that the same artifacts can be assigned to different classes according to the weighting and combination of variables used, and that all classifications will create a certain level of abstraction and ambiguity, but that this can be reduced by being explicit about the choice of variables and the weightings given to each. This also increases the ease with which each classification can be replicated by other researchers (Andrefsky 1998: 62). Building your own problem-oriented classifications therefore requires attention to these factors. Once they have been dealt with, however, virtually any set of groupings based on a potentially infinite range of variables is conceivable. Another approach we would advocate in classifying stone artifacts is that some categories should always remain exclusive, whereas others might be allowed to overlap, as in the case of the three chronologically separate and mutually exclusive categories of cores, flakes, and retouched flakes. Classifica- tions that set out to describe and order manufacturing processes and/or prod- Ap

172 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 172 5/16/2013 10:15:42 AM ucts should generally seek to keep these classes distinct. Unfortunately, many classifications blur these categories and draw an initial division within assemblages between “tools” and “debitage” (e.g., Andrefsky 1998). “Tools” are all those artifacts believed to represent the intended “end products” of the process, while “debitage” constitutes all the waste left over from tool production, use, and maintenance (Dibble 1995; Hiscock 2001). These divisions are based on propositions about the intentions of long-dead knappers that cannot be verified empirically, and it is therefore safest when building classifications to start with basic observational categories, and if other higher-level categories are required, to build on them as required.

The selection of variables to record and measure in an analysis is clearly one Choosing attributes of the most important decisions you will make. As Hiscock and Clarkson (2000: to record and 99) state: measure

the most crucial consideration must be the analytical power of the attribute and What attributes should I its relevance to the questions posed.. . .the application of a single standardised choose? method of analysis, including the use of a standard set of attributes, is not an appropriate response because different observations will be needed for each new question and in each archaeological context. However, for any particular question there may be a number of relevant attributes, and it is valuable to also consider the power of equally relevant variables.

Our advice for choosing the most powerful attributes is to read widely within the technological literature and identify attributes that help address the questions you have posed. Compiling a table of justifications and references to successful uses of each attribute can also provide a useful starting point, as in Table 6.3. Phagan (1985) provides a fairly extensive list of attributes (and some justi- fication for each) commonly employed in the analysis of flakes and cores, as does Clarkson and David (1995) and Soriano et al. (2007). Attributes relevant to the recording of retouch are detailed by Dibble (1995), Clarkson (2002a, 2005b), Hiscock and Attenbrow (2002, 2003), and Hiscock and Clarkson (2005). Many of the standard introductory texts on lithic analysis listed in the introduction to this chapter provide guidelines on common axial measurements and variables recorded on flakes and cores.

A basic analysis, if such as thing could be said to exist, would probably try and What is a “basic” incorporate some description (whether quantitative or qualitative) of the size, analysis? shape, level of reduction, raw material, and technological and typological cat- egory for each artifact in an assemblage in the hope that the broadest possible range of questions could be addressed using those attributes. It may be possible to cover each of these aspects and yet still record only a small number of variables. As stated earlier, it will be the power of each attribute to address each criteria that will determine how streamlined the analysis can be. Ap

An Introduction to Stone Artifact Analysis 173

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 173 5/16/2013 10:15:42 AM 2 1989a) R oche (2005) (2008)

E dmonds (2005) (2008) (2002) H ester (1991) (2008, 2011) K elly (1987) L ove (1985) et al. et al. et al. et al. Examples and key references T echno-typological analysis: Boëda (1995) O hnuma and Bergman (1990) H iscock (1993) R eduction sequence analysis: Bleed (2001, 2002) Clarkson (2002a) D ibble (1995) R efitting: D elagnes and Van Peer (2007) R eplication and experimentation: Aubry Crabtree (1968) F lenniken (1978) E ren Stout Mobility and the organization of technology: K elly (1988) Binford (1979) Shott (1986) Bleed (1986) N elson (1991) K uhn (1992, 1995) Surovell (2009a) Clarkson (2006) T orrence (1983, T ransport and reduction: Beck Byrne (1980) Goodyear (1989) Interaction and migration: E erkens and Bettinger (2008) Stark (1998) T ostevin (2007) Clarkson (2010, in press) Buchannan and Collard (2008) T rade and exchange: T orrence and Summerhayes (1997) T orrence (1986) Sheppard (1993) Bradley and Craft specialization: Shafer and H iscock (2005) H iscock (1981) K uhn (1995) Parry and H all and Jung (1992) Petraglia (2002) uhn uhn K R eduction R efitting and T echniques such as L ikewise, the R eduction sequence analysis orrence examined the issue of issue the examined orrence T T he presence of artifact forms T rade and exchange become more U nderstanding local patterns of T rampling can also provide a guide to occupational II, GI UR , or M T I R for retouched flakes, or number of rotations or ools intended for long-distance transport also may tend to show signs signs show to tend may also transport long-distance for intended ools T R eplication and experimentation are also valuable for reconstructing T echno-typological analyses examine constellations of morphological and R efitting and size sorting can also help identify scuffage and site clearance, R efitting is a powerful means of studying reduction sequences to determine the odd considered the ways certain technological strategies might be advantageous in advantageous be might strategies technological certain ways the considered odd F requently and predictably used places were more often stockpiled with raw T elly and and elly Key ideas/possible approaches R econstructions of past lithic technologies can be achieved in many ways, including techno-typological analysis, reduction sequence analysis, refitting or replication, and experimentation. technological characteristics to infer past reduction systems. using reduction measures like flake scars for cores, to order artifacts in terms of reduction intensity, allowing changes in morphological, technological, and typological features to be analyzed as artifact reduction increases. steps and stages, problems, and solutions, and to discard criteria for individual blocks of stone or implements. past technologies by relearning lost techniques and skills, comparing products with archaeological assemblages, and experiencing the process such that complexity, skill, degree of forward planning, and perhaps even past goals and intentions can be inferred. lithic from mobility and use land past determining in helpful are concepts key Several residential frequent to according vary might technology ways the explored Binford curated and assemblages. expedient of concepts the developing mobility, logistical periodic versus mobility that mobility demonstrated Shott structuring. resource to relationship their and technologies diversity. toolkit lithic on effects predictable have also may versus frequency and individuals magnitude mobile provisioning for strategies identifying provisioning, of idea the developed the and design toolkit of issue the examined Bleed places. used frequently mobility. and provisioning use land of forms certain for reliability with maintainability balancing of advantages K landscapes. unfamiliar of use the or mobility high of of situations periods to techtime scheduling by risk reduce foragers mobile high that ways and scenarios scheduling economic and ecological of range a wide explores Surovell sites. residential in downtime these of many employed Clarkson formation. assemblage and design tool on effects their rankings and patch different given use land differential about hypotheses test and formulate to or ideas periods long for transported been have that toolkits Stone availability. material raw and of proportions changing show may and reduction of levels higher show to tend also distances materials. raw transported transported. are components useful only ensure to preprocessing of Patterning in the techniques of stone artifact manufacture, or “technological style,” can provide a means of studying transmission and group interaction and can form the basis of studies of exchange, trade, diffusion, and migration. manufacture and their chronology as well as the availability of local raw materials versus exotics is an important step in the analysis of interaction. without local precedents may be a guide to interaction and exchange. appearance of exotic, high-quality, rare, and standardized items that may be highly curated or treated differently (i.e., smashed or deposited in specific contexts) can be a good guide to long-distance interaction and exchange. Many kinds of stones can be chemically or visually sourced, enabling origins and transport to be determined. easily identified when formal markets and dedicated workshops producing standardized goods are involved, likely involving some degree of craft specialization. cladistics also exist to explore phylogenetic relationships between stoneworking traditions that help determine common ancestry versus intrusion. Stone artifact densities can be calculated per unit volume and time to measure changing occupational intensity over space and time. Caution is warranted in using this method however as changing technology, raw materials, and/or reduction intensity may alter the rate of discard independent of changes in occupational intensity. Another method of inferring occupational intensity involves consideration of the kinds of provisioning strategies employed at a site. materials (lumps of stone or cores and hammerstones) in anticipation of future use, whereas sites that saw only ephemeral occupation typically show traces of only the maintenance or use of the transported stone supply. intensity, measured as rates of edge damage and transverse snaps caused by treadage. T hermal alteration of lithics can also be a guide to firing intensity and site use. intensity may als L ithic T his is a L ike language, lithic L ithic technology can T his can generate D esign, procurement, T his involved planning, scheduling, Rationale R econstructing past lithic technology allows us to understand the sequence and technical aspects of stone artifact manufacture, including the level of complexity and skill required, the production tools needed, the manufacturing stages and debitage produced, the efficiencies of labor and raw material use, and the processes and thresholds that lead to discard. valuable first step in addressing the questions listed below and in understanding variability and change. L ithic technology was often closely attuned to landscape use because people needed to ensure they had the right tools where and when they were needed. and design and maintenance of suitable tools and raw materials. rationing, and transport will all be conditioned by the availability of stone in the landscape and the type, structure, and temporal availability of crucial resources. Successfully reconstructing past land use and mobility therefore requires a good understanding of lithic sources, ecology, and climate as well as archaeological data on occupational intensity, subsistence patterns, and foraging range. Spatial variability in material culture can inform us about social interaction as well as processes of cultural change, diffusion, conflict, trade, exchange, and migration. technology is a culturally inherited practice that was usually passed from parents to offspring, generation to generation. therefore be highly patterned by transmission of stoneworking knowledge. ethnically circumscribed and diverse practices when transmission is socially and geographically restricted, or homogeneous and widespread practices when technologies are widely shared and copied. Archaeologists are often looking for ways to infer changing group size, how long sites were used for or how frequently revisited. Measuring occupational intensity in this way can be helpful for understanding changes in site use, demography, and settlement pattern. assemblages are durable, and flaking and use of stone generate chipped and ground implements and debitage in quantities that may be proportional to the levels of occupation or activity witnessed at a site.

Ap K ey questions and possible approaches in lithic analysis.

H ow do I H ow can I study H ow can I study H ow can I study Table 6.3. Question 1. reconstruct past lithic technologies? 2. land use and mobility? 3. cultural interaction? 4. occupational intensity?

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 174 5/16/2013 10:15:43 AM 2 1989a) R oche (2005) (2008)

E dmonds (2005) (2008) (2002) H ester (1991) (2008, 2011) K elly (1987) L ove (1985) et al. et al. et al. et al. Examples and key references T echno-typological analysis: Boëda (1995) O hnuma and Bergman (1990) H iscock (1993) R eduction sequence analysis: Bleed (2001, 2002) Clarkson (2002a) D ibble (1995) R efitting: D elagnes and Van Peer (2007) R eplication and experimentation: Aubry Crabtree (1968) F lenniken (1978) E ren Stout Mobility and the organization of technology: K elly (1988) Binford (1979) Shott (1986) Bleed (1986) N elson (1991) K uhn (1992, 1995) Surovell (2009a) Clarkson (2006) T orrence (1983, T ransport and reduction: Beck Byrne (1980) Goodyear (1989) Interaction and migration: E erkens and Bettinger (2008) Stark (1998) T ostevin (2007) Clarkson (2010, in press) Buchannan and Collard (2008) T rade and exchange: T orrence and Summerhayes (1997) T orrence (1986) Sheppard (1993) Bradley and Craft specialization: Shafer and H iscock (2005) H iscock (1981) K uhn (1995) Parry and H all and Jung (1992) Petraglia (2002) uhn uhn K R eduction R efitting and T echniques such as L ikewise, the R eduction sequence analysis orrence examined the issue of issue the examined orrence T T he presence of artifact forms T rade and exchange become more U nderstanding local patterns of T rampling can also provide a guide to occupational II, GI UR , or M T I R for retouched flakes, or number of rotations or ools intended for long-distance transport also may tend to show signs signs show to tend may also transport long-distance for intended ools T R eplication and experimentation are also valuable for reconstructing T echno-typological analyses examine constellations of morphological and R efitting and size sorting can also help identify scuffage and site clearance, R efitting is a powerful means of studying reduction sequences to determine the odd considered the ways certain technological strategies might be advantageous in advantageous be might strategies technological certain ways the considered odd F requently and predictably used places were more often stockpiled with raw T elly and and elly Key ideas/possible approaches R econstructions of past lithic technologies can be achieved in many ways, including techno-typological analysis, reduction sequence analysis, refitting or replication, and experimentation. technological characteristics to infer past reduction systems. using reduction measures like flake scars for cores, to order artifacts in terms of reduction intensity, allowing changes in morphological, technological, and typological features to be analyzed as artifact reduction increases. steps and stages, problems, and solutions, and to discard criteria for individual blocks of stone or implements. past technologies by relearning lost techniques and skills, comparing products with archaeological assemblages, and experiencing the process such that complexity, skill, degree of forward planning, and perhaps even past goals and intentions can be inferred. lithic from mobility and use land past determining in helpful are concepts key Several residential frequent to according vary might technology ways the explored Binford curated and assemblages. expedient of concepts the developing mobility, logistical periodic versus mobility that mobility demonstrated Shott structuring. resource to relationship their and technologies diversity. toolkit lithic on effects predictable have also may versus frequency and individuals magnitude mobile provisioning for strategies identifying provisioning, of idea the developed the and design toolkit of issue the examined Bleed places. used frequently mobility. and provisioning use land of forms certain for reliability with maintainability balancing of advantages K landscapes. unfamiliar of use the or mobility high of of situations periods to techtime scheduling by risk reduce foragers mobile high that ways and scenarios scheduling economic and ecological of range a wide explores Surovell sites. residential in downtime these of many employed Clarkson formation. assemblage and design tool on effects their rankings and patch different given use land differential about hypotheses test and formulate to or ideas periods long for transported been have that toolkits Stone availability. material raw and of proportions changing show may and reduction of levels higher show to tend also distances materials. raw transported transported. are components useful only ensure to preprocessing of Patterning in the techniques of stone artifact manufacture, or “technological style,” can provide a means of studying transmission and group interaction and can form the basis of studies of exchange, trade, diffusion, and migration. manufacture and their chronology as well as the availability of local raw materials versus exotics is an important step in the analysis of interaction. without local precedents may be a guide to interaction and exchange. appearance of exotic, high-quality, rare, and standardized items that may be highly curated or treated differently (i.e., smashed or deposited in specific contexts) can be a good guide to long-distance interaction and exchange. Many kinds of stones can be chemically or visually sourced, enabling origins and transport to be determined. easily identified when formal markets and dedicated workshops producing standardized goods are involved, likely involving some degree of craft specialization. cladistics also exist to explore phylogenetic relationships between stoneworking traditions that help determine common ancestry versus intrusion. Stone artifact densities can be calculated per unit volume and time to measure changing occupational intensity over space and time. Caution is warranted in using this method however as changing technology, raw materials, and/or reduction intensity may alter the rate of discard independent of changes in occupational intensity. Another method of inferring occupational intensity involves consideration of the kinds of provisioning strategies employed at a site. materials (lumps of stone or cores and hammerstones) in anticipation of future use, whereas sites that saw only ephemeral occupation typically show traces of only the maintenance or use of the transported stone supply. intensity, measured as rates of edge damage and transverse snaps caused by treadage. T hermal alteration of lithics can also be a guide to firing intensity and site use. intensity may also be a rough guide to occupational intensity if people spend more time a L ithic T his is a L ike language, lithic L ithic technology can T his can generate D esign, procurement, T his involved planning, scheduling, Rationale R econstructing past lithic technology allows us to understand the sequence and technical aspects of stone artifact manufacture, including the level of complexity and skill required, the production tools needed, the manufacturing stages and debitage produced, the efficiencies of labor and raw material use, and the processes and thresholds that lead to discard. valuable first step in addressing the questions listed below and in understanding variability and change. L ithic technology was often closely attuned to landscape use because people needed to ensure they had the right tools where and when they were needed. and design and maintenance of suitable tools and raw materials. rationing, and transport will all be conditioned by the availability of stone in the landscape and the type, structure, and temporal availability of crucial resources. Successfully reconstructing past land use and mobility therefore requires a good understanding of lithic sources, ecology, and climate as well as archaeological data on occupational intensity, subsistence patterns, and foraging range. Spatial variability in material culture can inform us about social interaction as well as processes of cultural change, diffusion, conflict, trade, exchange, and migration. technology is a culturally inherited practice that was usually passed from parents to offspring, generation to generation. therefore be highly patterned by transmission of stoneworking knowledge. ethnically circumscribed and diverse practices when transmission is socially and geographically restricted, or homogeneous and widespread practices when technologies are widely shared and copied. Archaeologists are often looking for ways to infer changing group size, how long sites were used for or how frequently revisited. Measuring occupational intensity in this way can be helpful for understanding changes in site use, demography, and settlement pattern. assemblages are durable, and flaking and use of stone generate chipped and ground implements and debitage in quantities that may be proportional to the levels of occupation or activity witnessed at a site.

K ey questions and possible approaches in lithic analysis. Ap

H ow do I H ow can I study H ow can I study H ow can I study Table 6.3. Question 1. reconstruct past lithic technologies? 2. land use and mobility? 3. cultural interaction? 4. occupational intensity?

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 175 5/16/2013 10:15:43 AM What is more powerful or As mentioned earlier, there are many kinds of lithic analysis and each may be appropriate: Individual best suited to particular questions, assemblages, and time constraints. This artifact analysis or assemblage (debitage) chapter is mainly concerned with individual artifact analysis wherein a series analysis? of attributes is recorded on each artifact. We argue that this is the most powerful and flexible system of analysis as it allows patterning in associated technological traits to be explored for each artifact as well as for whole assemblages, but it is time-consuming! An alternative set of analyses focuses not on individual artifacts but on whole assemblages, usually for specifically comparative purposes. Such techniques typically aim to answer questions about differences between sites or time periods in reduction strategy (e.g., biface reduction vs. single platform core reduction), identify likely stages of reduction (early vs. late core reduction, or core reduction vs. flake retouching and maintenance), or identify the use of certain techniques of reduction (Andrefsky 2001; Odell 2003). The use of screening (or other rapid artifact size sorting techniques) to aggregate debitage into size categories and counting artifact features rather than individually recording each artifacts aims to process large batches of artifacts quickly. This type of “debitage analysis” typically looks at patterning in artifact size (mass analysis) (Patterson & Sollberger 1978; Patterson 1982, 1990; Stahle & Dunn 1984; Ahler 1989; Odell 1989; Shott 1996; Austin 1997; Steffen et al. 1998), breakage patterns (aka “SRT” or Sympathetic Resonance Technol- ogy) (Sullivan & Rozen 1985; Baulmer & Downum 1989; Prentiss & Romanski 1989), or a combination of size, breakage, and diagnostic platform and dorsal characteristics to infer the use of various techniques (such as hard and soft hammer, pressure, bifacial reduction, and core vs. flake reduction) (Shott 1996; Morrow 1997a; Bradbury 1998). Problems have been identified with some of these techniques, particularly highlighting their reduced success in cases where assemblages are mixed (Prentiss & Romanski 1989; Shott 1994; Prentiss 1998; Bradbury & Carr 1999). The best results will always be obtained when such techniques are tailored to specific regions and raw materials, and in our view, will be most successful when used in combination with individual artifact analysis.

Managing data Because laboratory analysis is slow and painstaking, there is a good argument to be made for reducing data handling time by entering information straight How should I record my into the computer as it is gathered. Although some archaeologists still prefer attributes? to use spreadsheets, there is no doubt that a database provides a far superior means of entering, storing, and retrieving data about individual specimens. Computer data entry may not be practical in some field situations where the use of recording forms may still be the most suitable option, but handheld computer devices, often coupled with global positioning system (GPS) and advanced mapping software (such as the NOMAD series of devices), are increasingly used in field settings.

When do I need to use Statistics are typically used in stone analysis to provide a means of seeking statistics and what independent confirmation that the patterns observed in the data are not simply Ap statistics are most useful? a result of the vagaries of sampling (i.e., random effects), small sample size, or

176 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 176 5/16/2013 10:15:43 AM the result of a complex interaction of several variables that makes the important variables or patterns difficult to determine. We cannot advise what statistics should be used to analyze the data as each question and analysis lends itself to different techniques and tests. Nevertheless, several tests tend to be used over and again in lithic analysis. These include chi-square tests, t-tests, Spearman’s rho, and regression analysis. These are all basic techniques for working with the kinds of data that archaeologists use in lithic analysis, such as proportions, weights, and axial measurements. A good introduction to all of these techniques, as well as a demystification of concepts like significance and sampling, and useful suggestions for identifying and working with skewed populations (which most lithic assemblages tend to be), is provided for archaeologists by Drennan (2009). These basic tests are all that is typically needed to make comparisons between assemblages, confirm order- ing in the data, and determine whether a relationship between two variables is strong and significant. Statisticians can be helpful in identifying the techniques that best address your questions, and in navigating and interpreting the more complex world of multivariate statistics and cluster analysis, but you should be capable of performing most simple tests yourself with the help of statistics software.

As stoneworking is a reductive technology, the measurement of the degree to Measuring extent which this process has progressed often forms the basis of many modern analy- of reduction ses. Quantifying the extent of reduction allows estimations to be made of the amount of time and energy invested in the production of an artifact, the level Why measure reduction? of departure of the observed form from its original form, the amount of material likely to have been created as a product of the process, and the position in the sequence at which changes in manufacturing strategies took place and their likely effects on artifact morphology. At a higher interpretive level, many archaeologists see measures of reduction as critical to the testing of behavioral models that hypothesize the place of stone artifacts in broader systems of time budgeting, mobility, and land use. Consequently, measures of reduction have come to be associated, at least implicitly, with discussions of risk, cost, and efficiency in past technological systems (Bleed 2001). These discussions build on the assumption that the differential distribution of sequential steps and stages through space and time will reflect aspects of planning, land use, ecology, and settlement and subsistence patterns affecting people’s daily lives (Nelson 1991; Kuhn 1995). Measures of reduction have consequently become a central component of contemporary lithic analysis. Determining stages of reduction also forms the basis of the European chaîne opératoire approach (Leroi-Gourhan 1964; Meignen 1988; Roebroeks et al. 1988; Geneste 1990), although this system differs from North American and Australian approaches in some important philosophical and methodological respects (Sellet 1993; Shott 2003; Clarkson 2005a). The chaîne opératoire approach places emphasis on the technical “choices” people make during lithic reduction between a variety of possible solutions and the context of these decisions within broader cultural values and social relations Ap (Lemonnier 1986; Dobres 2000).

An Introduction to Stone Artifact Analysis 177

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 177 5/16/2013 10:15:43 AM How do I measure core Both fracture mechanics and basic engineering principles would suggest that reduction? striking more and more mass from a core will affect its size and geometry, which will have direct consequences for the nature of force input, the viability of different reduction strategies, and the size and morphology of the flakes produced over the sequence. We can speculate for instance that the gradual reduction of cores will result in more flake scars and less cortex, that continued use of a platform will result in a decrease in platform size, and that as more mass is struck from a core, the size of the core and resulting flakes might also decrease. If cores are rotated during this process to create fresh platforms once old ones become damaged or unproductive, cores should begin to preserve signs of former flaking on the platform surfaces as well as indications of the existence of old platforms. To provide an example of the sorts of procedures that can be used to track morphological changes and the use of different technological strategies over the sequence of core reduction, a number of variables are plotted against increasing numbers of core rotations in Figure 6.8. These changes are documented from a set of 87 small, locally occurring, river-rounded chert cobbles found in a site near Wollongong on the southern coast of New South Wales, Australia. This diagram shows that many core characteristics show an increase over the sequence of reduction, while others decrease. For instance, as might be expected, the number of scars found on cores increases with each rotation, as does the percentage of platforms that has more than one conchoidal scar (resulting from former use as a core face). The percentage of scars found on the core showing step and hinge terminations also increases as core rotation proceeds, as does the external angle of the last platform used on the core. Overhang removal increases early on and remains high throughout the remainder of the reduction sequence. Overhang removal was presumably used to strengthen the platform to better receive the forceful blows required to remove flakes from small cores with increasingly high-angled platform edges. In contrast, cortex diminishes at a fairly consistent rate throughout the sequence, indicating that similar amounts of material were likely removed from each platform with each rotation. The used portion of the platform edge first increases and then decreases as irregularities left on the core face and platform by previous rotations reduce the usable platform perimeter. The number of cores from each stage of reduction also clearly indicates that most cores were abandoned in early stages of reduction, although a small proportion continued into later stages of reduction, by which time cores were heavily rotated and generally lacking cortex, or were subjected to bipolar reduction. Changes in core morphology over the reduction sequence are illustrated as a reduction flowchart in Figure 6.9, which depicts a number of the ways of flaking small spherical nodules followed at the site. While archaeologists have sometimes used this type of chart to illustrate normative reduction sequences through which most forms are argued to pass, this chart ascribes frequencies to each stage in each sequence as determined from the assemblage itself. Reduc- tion begins with a single flake removed from a cortical platform. In the left-hand sequence (Sequence 1), new platforms are always created from the previous Ap flaked surface via 90° core rotations. In the middle sequence (Sequence 2), new

178 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 178 5/16/2013 10:15:44 AM 80 60 40 Number of scars 20 0 90 80 60 %multi scarred platforms 40 20 0 60 50 40 %step/hinge 30 100

90 Final platform angle

80 80 70 %overhang removal 60 50 60 40 20 %cortex 0 100 75

50 %platform perimeter used 25 40 30 20 Number of cores 10 0 0 1 2 3 ≥4 Number of rotations

Figure 6.8. Changes in core morphology over the sequence of reduction. With kind permission from Chris Clarkson.

platforms are always created from cortical surfaces. In the right-hand sequence (Sequence 3), a single large scar is removed from each surface, which then becomes the platform for the next single large flake removal. Also illustrated in Figure 6.9 are late-stage rotated and bipolar cores with and without cortex, Ap which represent the very end stages of all sequences.

An Introduction to Stone Artifact Analysis 179

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 179 5/16/2013 10:15:44 AM 5%

10%

32% 5% 3%

10% 4% 2%

8% 1% 1%

2% 1% 1%

Late stage with cortex Late stage no cortex Bipolar, no cortex Bipolar with cortex

2% 10% 2% 1%

Figure 6.9. The reduction flowchart and the frequencies with which various core reduction sequences were employed at Sandon Point, New South Wales coast. With kind permission from Chris Clarkson.

From this diagram, it can be seen that Sequence 1 was most commonly followed at the site, but that Sequences 2 and 3 also formed common alternatives. Mapping reduction sequences in this way allows variation as well as the central tendency to be explored, and also demonstrates that core reduction was a Ap highly variable process, with knappers responding to the results of each suc-

180 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 180 5/16/2013 10:15:45 AM cessful or unsuccessful blow in a flexible fashion, in which the options for rotation, discard, or strategy switching (such as to bipolar reduction) were appraised at various points along the way. This simple case study indicates that a number of variables are likely to be useful measures of reduction intensity, at least for this type of core reduction. Used in combination, these attributes would provide a reasonably sound basis on which to infer the level of reduction reached by any individual core. This could be done by dividing the continuum into a number of intervals (such as early, medium, and late) or using a continuous ranking system. Assigning each core its own degree of reduction then allows the intensity of core reduction to be traced across space and time. It should be kept in mind, however, that different forms of reduction might well require the use of different variables than those employed here.

Lithic analysts employ various means of assigning flakes from archaeological How do I measure flake assemblages into reduction stages, but most of these tend to involve compari- reduction? son of archaeological specimens with experimentally produced assemblages. To avoid this analogical approach, our case study ranks flake reduction according to simple and universal changes in flake morphology that are deduced from the analysis of changing core morphology presented earlier, as reflected in dorsal and platform scar morphology. This type of analysis is called “diacritical analysis” (Sellet 1993) and aids in the construction of hypothetical reduction models. In this case, diacritical analysis allows changes in flake morphology to be examined for Sequence 1 of the pebble core reduction sequence illustrated in Figure 6.9. The reduction process can be modeled by examining stages in flake scar superimposition on the platform and dorsal surfaces of flakes and the stages of decortication present. Nine stages of flake production can be envisaged. As in Figure 6.9, the first phase involves the creation of an initial flake scar on the core to serve as a platform for the next stage of reduction. This results in the production of Stage 1 flakes that possess primary cortex (i.e., 100%) on all surfaces. The second stage involves the rotation of the core 90°, so that flakes can be struck from the first scar. These Stage 2 flakes will possess a single conchoidal scar on the platform and primary cortex on the dorsal surface. Stage 3 flakes result from continued reduction of this second face and will have single conchoidal platforms but only secondary cortex (<100%) remaining on the dorsal surface. Stage 4 flakes result from the final stages of reduction on this face and will possess the same type of platform, but will have no cortex remaining on the dorsal surface (tertiary decortication). At some point, reduction is likely to end on this second face as nonfeather terminations increase, or platforms become unproductive due to high platform angles. This will result in either discarding the core or the formation of another platform by rotating the core again. Redirection flakes, or flakes that remove the edge of an old platform, are sometimes created by this process, and these are here labeled Stage 5 flakes. At this early stage of reduction, redirection flakes should preserve cortex Ap on one or more of their surfaces (and this is the criteria used for determining

An Introduction to Stone Artifact Analysis 181

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 181 5/16/2013 10:15:45 AM Stage 5 redirection flakes). The process of reduction then begins anew on the third face, with cortical flakes with multiple scarred platforms (as a result of striking from previously flaked surfaces) produced first (Stage 6), followed by flakes with secondary cortex (Stage 7), and then by flakes with no cortex (Stage 8). Once this face also begins to encounter difficulties for further reduction, the core may be discarded or rotated again, and another redirection flake may be produced. As cores enter increasingly later stages of reduction, cortex is likely to have been entirely removed from all surfaces, and hence redirection flakes from this stage would show no cortex on their surfaces. These flakes are assigned to Stage 9 (or 3 to N rotations). Although not included in this hypothetical reduction model, another option for knappers is to switch to the use of a bipolar reduction technique once cores become too small to continue freehand percussion, and this was frequently undertaken at the site. Figure 6.10 maps out the sorts of changes in flake characteristics that accom- pany each of the stages of Sequence 1 reduction as deduced from dorsal and platform scar patterns. Interestingly, these changes are largely cyclical, with the gradual increases or decreases in characteristics taking place throughout the first phase of flake removals (Stages 2–4) often repeated in the second phase (Stages 6–8). The basic series of changes is as follows: Flakes are at first rather squat but become increasingly elongate toward the end of each phase as parallel ridges become more common, platform area decreases, platform preparation becomes common, the proportion of nonfeather terminations increases, and the size of flakes measured by width and thickness decreases (weight does not show a sequential decrease due to differences in initial nodule size). Redirecting flakes stand out in terms of their larger size and their apparent nonconformity to the trends otherwise seen for most characteristics. This is not surprising since striking off old platforms often requires delivering large amounts of force to the core from a less than optimal platform. Hence, striking off old platforms is unlike other forms of flaking, and the resulting flakes are often larger and distinctive. The final line of Figure 6.10 shows the- fre quency with which flakes at each stage of reduction are found in the assemblage, and indicates that the greatest proportion of flakes in this assemblage belong to early stages of reduction. This is consistent with the cores from the site, as shown in the figure, which were rarely taken into later stages of reduction.

How can I explore blank Archaeologists are often interested in the process of blank selection – or the selection? selection of a subset of flakes from the total population produced at a site – for further use, retouching, and transport away from the site. Blank selection is of interest as it has the potential to inform us about design considerations (such as tool performance, reliability/maintainability, suitability to prehension and hafting, and multifunctionality), a range of environmental and cultural constraints (functional, material, technological, socioeconomic, and ideological) (Hayden et al. 1996), and the level of standardization in the production system, both in terms of overall flake production and selection from the larger pool of Ap flake variation.

182 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 182 5/16/2013 10:15:45 AM 80 60 %Non feather 40 terminations 20 0 60

%Elongate 40 flakes 20 0 80 60 %Overhang 40 removal 20 0 14 12 10 Thickness 8 6 4 33 30 27 Platform creation flake Width 24 21 Early stage 1R reduction 18 15 Mid stage 1R reduction 160 Late stage 1R reduction 120 Platform area 80 Early stage redirection flake 40 Early stage NR reduction 0 Mid stage NR reduction 60 Late stage NR reduction 40 Number Late stage redirection flake 20 0 Stage 1 2 3 4 5 6 7 8 9

Figure 6.10. Changes in flake morphology over Reduction Sequence 1. With kind permission from Chris Clarkson.

An example of one approach to examining the pool of variation in flake forms produced at the same site near Wollongong, and the range of blank shapes selected for various forms of retouching, is shown in Figure 6.11. Here, two measures of flake shape are plotted against one another to illustrate the spread of flake shapes found at the site. Plotted on they -axis is the angle of the retouched margins expressed in degrees, with 0 indicating parallel-sided Ap margins, positive values indicating contraction of the margins along their

An Introduction to Stone Artifact Analysis 183

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 183 5/16/2013 10:15:46 AM 60

20 Backed artifacts

0 “Blades” Flakes

–20 Retouched flakes Angle of the lateral margin s

–40

–60 1 1.5 2 2.5 3 3.5 4 4.5 Elongation

Figure 6.11. A graph illustrating the range of variation in flake shape employed in the different forms of retouching. With kind permission from Chris Clarkson.

length, and negative values indicating expansion of the lateral margins (Clark- son & David 1995). The x-axis plots elongation (length/width), with values ranging from very squat flakes (e.g., values of 1) through to extremely elongate flakes (e.g., values of 4.5). The resulting graph shows a wide range of flake shapes produced at the site, with the majority proving to be squat flakes of varying contraction and expansion along their length that likely derive from earlier stages of core reduction (see Figure 6.11). Retouched flakes clearly represent a much smaller subset of the total range of flake shapes, while backed artifacts (large and small symmetrics and asym- metrics) (Hiscock 2002a) make up an even narrower range. Overall, retouched and backed flake shapes mirror the broader pattern of squat flakes with variable marginal angles produced at the site, although backed artifacts tend more frequently to be parallel sided with contracting margins. The graph indicates that flake production and blank selection were far from standardized at the site, suggesting that few design considerations affected the types of blanks chosen for further modification. The greater constriction of variation seen in backed artifacts, however, points to tighter constraints on the design of these implements than was the case for retouched artifacts more generally, and these may be related to hafting requirements, functional efficiency, potential for multiple uses, or all of these factors.

How do I measure Retouched flakes are most commonly the subject of detailed lithic analysis, and Ap retouch? many techniques have been proposed and tested to measure how much retouch

184 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 184 5/16/2013 10:15:46 AM they have received (Dibble & Whittaker 1981; Barton 1988; Kuhn 1990; Gordon 1993; Dibble 1995; Dibble & Pelcin 1995; Holdaway et al. 1996; Pelcin 1997a–c, 1998; Shott et al. 2000; Clarkson 2002b; Blades 2003; Eren et al. 2005; Hiscock & Clarkson 2005, 2009; Clarkson & Hiscock 2008, 2011; Eren 2009). Several techniques have entered the literature as standard procedures for measuring retouch, such as the index of invasiveness (Clarkson 2002b), the geometric index of unifacial reduction (GIUR) (Kuhn 1990), and the ratio of platform to ventral surface area (Dibble 1995). The index of invasiveness provides a measure of retouch coverage over both the dorsal and ventral surfaces of an artifact that is suited to measuring unifacial and bifacial retouch (Clarkson 2002b). It is best suited to the measurement of artifacts that tend to become more invasively retouched over the sequence of reduction. The index of invasiveness calculates intensity of retouch as a value between 0 and 1 by estimating the extent of retouching around the perimeter of a flake as well as the degree to which it encroaches onto the dorsal and ventral surfaces (see Clarkson 2002b for procedures). Another measure, the GIUR, was developed by Kuhn (1990) and calculates edge attrition as the ratio of the height of retouch to the maximum thickness of the flake. This technique is designed for the measurement of unifacial retouch, as the name suggests, but is also best suited to the measurement of steep and marginal retouch. Again, a score of 0 indicates no retouch, while a score of 1 indicates that retouch height is equal to flake thickness. Used in conjunction, these two techniques are capable of describing almost any form of retouching, and of quantifying the degree to which retouch in steep and marginal or acute and invasive. Although retouched flakes (commonly termed “scrapers”) have often been treated as stylistically irregular artifacts shaped simply to meet immediate needs (Hayden 1979; White & O’Connell 1982), recent studies (Clarkson 2002a, 2007: 2007; Hiscock & Attenbrow 2002, 2003, 2005a,b) have demonstrated that this group of implements can display marked internal consistency when examined in light of increasing retouch intensity, with a regular series of changes noted to the shape, extent, and type of retouch found on their margins as retouch increases. These changes can be depicted using a number of indices of retouch extent, shape, and type. These are the percentage of the perimeter of an artifact that is retouched, the curvature of the retouched edge, the angle of retouch, and the invasiveness of retouch. The percentage perimeter of retouch is calculated by dividing length of retouch by the perimeter of the flake. The angle of the retouched edge is calculated as the mean of several edge angle measurements taken at regular intervals along retouched edge. Edge curvature is measured by dividing the maximum diameter of retouch by the total depth of retouch (Clarkson 2002a). Negative results indicate concave edges, while positive ones indicate convex edges. An example of the power of these measures of flake reduction is illustrated in Figure 6.12 by plotting the morphological changes that occur in a population of 128 retouched flakes from the same site near Wollongong as retouching Ap increases. Figure 6.12a, for instance, indicates that retouching usually starts out

An Introduction to Stone Artifact Analysis 185

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 185 5/16/2013 10:15:46 AM (a) 50 (b) 100

80 40 60

40 30

%margin retouched %margin retouched 20

20 0 0.25 0.50 0.75 1.00 –0.4–0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Kuhn index Index of edge curvature

Index of invasiveness 20 10 0 0.25 0.50 0.75 1.00 0.25 0.50 0.75 1.00 Kuhn index Kuhn index

Figure 6.12. Changes in the flake morphology as retouching increases: (a) the percentage of the margin retouched; (b) the edge curvature; (c) the mean retouched angle; (d) the index of invasiveness. With kind permission from Chris Clarkson.

covering only a short section of the lateral margins, but as Kuhn’s GIUR measure increases, eventually extends to cover around 50% of the margin. As retouch spreads around the margin, the curvature of the edge also changes as shown in Figure 6.12b. Retouch usually starts out slightly concave, but then straightens out before finally becoming quite convex. Similarly, Figure 6.12c and 6.12d indicates that retouch generally starts out quite low angled and marginal, but ends up steep and more invasive. This sequence of changes accounts for much of the variation in retouched flake morphology and underlies the differences in form that have sometimes been enshrined in formal scraper typologies (such as that of McCarthy et al. 1946). Much like the situation for cores argued earlier, once these changes in flake form are documented and the reduction sequence is understood, individual artifacts can be assigned their position in a particular reduction sequence. Understanding reduction sequences also enables the construction of classifications that divide the continuum in sensible places, rather than jumbling together artifacts on the basis of features that are disconnected from the mechanisms actually creating that variation. The resulting classes could be treated as meas- Ap urement units (in this case of intensity of reduction) that expressed temporal

186 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 186 5/16/2013 10:15:47 AM and historical relationships (in this case departure from an original form), thereby meeting some of the expectations of materialist classificatory systems. Other studies have also explored stage and continuum models in artifact reduction, and some of these employ quite different measures of reduction intensity to those employed here (Barton 1988; Hiscock & Veth 1991; Gordon 1993; Marcy 1993; Neeley & Barton 1994; Dibble 1995; Holdaway et al. 1996; Morrow 1997b; Yvorra 2000; Blades 2003).

So far we have dealt almost exclusively with chipped stone artifacts, but of What should I do with course many artifacts such as axes, adzes, maces, clubs, and grindstones show ground artifacts? extensive grinding, often in combination with flaking and/or pecking. Because grinding can dramatically alter an artifact and create distinctive and even elabo- rate forms, ground implements are the focus of many typologies (Duff 1970; Smith 1985; Wright 1991; Gorecki et al. 1997). As valued and sometimes highly curated and traded goods, they are also often subjected to detailed sourcing studies, functional studies (see Chapter 8 for a review of recent functional analyses), and even technological studies. The latter seeks to understand the nature of manufacture, maintenance, reworking, use-life, and reasons for discard. Such studies might typically include some description of the extent of surface coverage of grinding, the nature of ground surface (flat, projecting, or dished), the form of the parent rock or piece from which the object was made (i.e., a slab, boulder, flake, or core), the nature of flaking and the superimposition of grinding, evidence for the maintenance of ground edges, ground edge angles, the size and nature of shaft holes or tangs, and the number of ground depressions and their depth in the case of grindstones, as well as factors that may have caused discard, such as a grindstone dish becoming too thin, or the edge of an adze catastrophically failing beyond repair. Flakes from ground implements are also readily identified in archaeological assemblages, and point to maintenance activities having been carried out at a site. In some cases, reduction analysis can also be profitably applied to ground adzes and axes to explore artifact variability, use-life, and changes to shape as they are reworked and reground, particularly when grinding and flaking are both present on a piece. Examining the proportion of the surface covered in grinding in relation to flake scar coverage, and/or the number of ground facets, is one way of examining reduction intensity on axes and adzes that has proved successful for Australian hatchets and Lapita adzes (Ulm et al. 2005; Clarkson & Schmidt 2009).

This can be a big problem for any analyst working on assemblages that have Dealing with been subjected to intense heating and trampling, extensive reduction and use, difficult or some other postdepositional process. Even broken and damaged artifacts assemblages can preserve information relevant to the reconstruction of the overall manufacturing system, however, and assemblage attrition can also reveal much about What should I do if most site formation, disturbance, and occupational intensity (Flenniken & Haggarty of my artifacts are broken or damaged? Ap 1979; Mallouf 1982; Hall & Love 1985; Hiscock 1985; Jung 1992).

An Introduction to Stone Artifact Analysis 187

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 187 5/16/2013 10:15:47 AM Figure 6.13 illustrates a range of fragment types that are commonly found in flaked stone assemblages. Different kinds of technological information can be recovered from each fragment type. For instance, transversely broken fragments such as proximal pieces preserve information about the platform surface, the presence and type of platform preparation, and the platform angle, while distal pieces preserve information about the flake termination type. Medial pieces can be informative about the cross section of the flake, while all of these fragments may reveal something of the original width and thickness of the flake, its dorsal scar morphology, or the presence, type, and amount of cortex. Longitudinally split flakes may preserve original length and thickness as well as some features of the platform and termination. Once transverse and longitudinal fractures occur together on the same artifact, however, information loss increases dramatically. Surface fragments and flaked pieces often yield little technological information at all. Combining the relevant information from each type of fragment with that gained from complete flakes may be a useful way of increasing the number of observations if assemblages are highly frag- mented, especially for categories that suffer high breakage rates and may be underrepresented in the assemblage (such as very thin artifacts or retouched flakes). Recovering information from broken artifacts is often a necessary means of increasing sample size (see “What if I only have a small number of stone artifacts?”). One problem that results from fragmentation can be accurately determining how much flaking took place at a site. Simple measures like weight or number of artifacts can be unreliable in cases where an assemblage contains only a few heavy artifacts or where postdepositional processes such as burning and trampling have caused severe artifact fragmentation. Hiscock (2002b) explored this problem and suggested some counting procedures that may assist in the quantification of numbers of flaking events and in assessing the effects of breakage and weathering. Hiscock’s technique allows a number of different indices to be calculated, including the minimum number of flakes (MNF) present in an assemblage. This is derived by dividing the assemblage into raw material types and then adding the number of complete flakes of each raw material to which- ever is the greater number of proximal or distal fragments, the greater number of left or right fragments, and the greater number of left or right proximal or distal fragments. MNF, or better still MNA (i.e., minimum number of artifacts including both flakes and cores), provides a superior means of assessing and comparing the intensity of flaking within or between sites than weight of raw numbers. Shott (2000) also offers various techniques (MNT, ETE, and TIE) for 1 estimating the original number of artifacts in an assemblage. To determine how serious a problem fragmentation is, a number of simple techniques exist that provide quantitative assessments of artifact fragmentation. The simplest index is calculated by dividing the number of broken fragments and flaked pieces by the number of complete artifacts. A measure of information loss, on the other hand, can be calculated by dividing the total number of fragments from which little or no technological information can be derived (such as surface fragments, marginal fragments, flaked pieces, and Ap longitudinally and transversely broken fragments) by those that provide a

188 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 188 5/16/2013 10:15:47 AM Longitudinal fragments Left Right

Marginal Surface (pot lid)

Tansverse and Tansverse and longitudinal Tansverse longitudinal Left proximal Proximal Right proximal

Left medial Medial Right medial

Left distal Distal Right distal

Figure 6.13. An illustration of the range of flake fragment types found in many assemblages. With kind permission from Chris Clarkson.

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 189 5/16/2013 10:15:47 AM great deal more technological information (such as proximal, medial, distal, right, and left fragments). These two techniques can be particularly useful in assessing the severity of attritional processes and their impact on information loss at a site.

What should I do if I This dilemma is quite common in archaeology and was central to the long- cannot tell artifacts from running naturefact/artifact debate in which claims were made for very old but natural rocks? dubious artifacts (Schnurrenberger & Bryan 1985; Peacock 1991). Pseudo- artifacts (often called eoliths or geofacts) can result from a host of natural processes, such as “natural soil movements, glaciation, wave action, high velocity water movement, gravity (such as alluvial fans or steep inclines), rapid tem- perature changes, internal pressure (such as starch fractures and pot lids), exfoliation, tectonic movements, diastrophism, solifluction, foot trampling and other unintentional activity caused by nature” (Crabtree 1972b: 78). As well as the difficulties faced in differentiating natural fractures from artifacts, certain types of stone, such as vein quartz, fracture in such a way that identification of artifacts becomes difficult. Weathering of artifact surfaces can also have severe effects on assemblages and can obscure or obliterate the diagnostic features of stone artifacts (Hiscock 2002b: 251). In cases where natural fracture is common or stone type or weathering renders artifacts difficult to distinguish from nonartifactual rocks, it may be helpful to try and determine the presence or absence of the fracture features identified earlier for each specimen, as well as those for grinding or pecking. Peacock’s (1991) analysis of natural and artifactual stone assemblages found many of these features to be reliable indicators of human manufacture. Com- piling lists of fracture features for each piece will enable the ranking of specimens in degrees of certainty (i.e., number of features present), with the aim of rejecting those lower in the rank order and accepting those higher up. Addi- tionally, certain fracture features tend to occur more commonly among natu- rally fractured pieces (due to the type and magnitude of forces) than among humanly derived assemblages (such as obtuse platform angles, edge rounding, and microedge and -ridge fracturing) (see figure 5.1 and table 5.1 in Schnur- renberger & Bryan 1985). The context of finds also provides a key to whether artifacts would be expected in a given location or not (e.g., rockshelters and middens vs. graded tracks and garden beds).

What if I only have a Sample size can have a profound effect on the sorts of information retrieved small number of stone from an assemblage, and sampling is therefore a key issue in understanding artifacts? stone artifact assemblage patterning through time and space. The probability of drawing a representative sample of the original population drops dramatically as sample size decreases, and this is particularly true in lithic assemblages where many objects of interest, such as retouched flakes, tend to be rare (i.e., less than 5% of the assemblage) and therefore unlikely to turn up in small samples. The diversity of an assemblage – or the number of different elements Ap found in the assemblage – is also sensitive to assemblage size, as the number

190 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 190 5/16/2013 10:15:47 AM of assemblage elements present cannot exceed the number of artifacts in the assemblage, and hence small assemblages will always have low diversity. It is therefore generally worthwhile enlarging the sample at the expense of other components of the research (e.g., number of attributes recorded) in order to obtain a useful and representative sample.

Unfortunately, there is no magic minimum number that will always overcome sample size effects as every assemblage or region is likely to be different. The How can I overcome only general rule as far as stone artifacts are concerned is likely to be “more is sample size effects? better” (within practical limits of course). It is possible to determine something like a minimum sample size for each assemblage, however, by studying the effects of increasing sample size by repeatedly drawing random samples from a large assemblage and examining the variation and deviations in percentages of classes, mean characteristics, and so on, between each sample and the parent assemblage. Of course, this can defeat the purpose since a large sample must already have been collected in order to perform such a test, although it may be helpful in determining minimum sample sizes that should ideally be obtained from sites within a region or survey area. A number of assemblage descriptions are more robust in relation to sample size effects. Assemblage richness, for example, provides a measure of assemblage composition that is not affected by sample size, and mathematically determines the number of classes expected to occur for a given sample size (Leonard & Jones 1989). Assemblage richness is calculated by dividing assemblage diversity, however measured (e.g., raw materials, classes), by sample size. Differences in assemblage richness between populations can be determined by plotting diversity against sample size (using a scatterplot) and comparing the gradients of slope for each population (higher gradients mean richer assemblages). Richness therefore provides a more robust comparative measure of assemblage complexity than does diversity. Assemblage richness also has real interpretive applications. For instance, archaeologists often interpret differences in the richness of raw materials and assemblage elements as a reflection of the range and diversity of activities carried out at a site (Binford 1979; Shott 1989). Theoretically, it is possible to determine how large a sample is needed before all classes recognized in an analysis will be present in an assemblage (or even a number of each class), by increasing sample size until the line of best fit flattens out, such that further increases in assemblage size no longer yield new classes. Obviously, the number of classes used in the analysis will affect the point at which this takes place. Such a technique could be used, say, to identify the ideal sample size to be recovered from a site threatened with destruction.

Archaeology increasingly draws on specialist information from fields outside Archaeometry of its own core knowledge and practice, and the incorporation of such technical analyses (usually involving measurement), particularly from the physical When do I need specialist Ap sciences, is often called archaeometry. Archaeometry can play a vital role in analyses?

An Introduction to Stone Artifact Analysis 191

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 191 5/16/2013 10:15:48 AM stone artifact analyses, and common examples of applications include using geochemical or elemental analyses to identify stone types, their age, origins, and alterations to the composition of artifacts due to time, weathering, and heat.

Determining the While the range of materials used to make stone artifacts often tends to be type and faking fairly limited (depending on local geology), the ability to differentiate reliably properties of stone between materials depends on a combination of expertise and experience, and archaeologists should solicit the help of geologists for identification of all but How do I identify different the most common types. A useful first step in identifying the range of raw raw material types? materials in an assemblage is to consult a geology map of the region. Unfor- tunately, geology maps are rarely drawn at the scale that archaeologists desire, and usually do not pinpoint small sources of flakable stone. A second step might involve sorting artifacts into broad material categories (based on texture, grain size, and color). Without significant experience in stone identification, a next step would be to take a sample of the various types to a geologist for identification. Once materials have been identified, it is often useful to know the original size in which they were available and their shape and original cortex, as this may influence the types of reduction strategies used. Often archaeologists tackle these questions by looking for cores at earlier stages of reduction, to determine the size, shape, and nature of cortex for each raw material type.

How do I determine if it Ethnographic literature often points to a range of cultural factors determining is good or poor quality the types of stones selected to make various implements, and these were often stone? strongly influenced by associations with ancestral beings, totemic affiliations, or powerful substances (Gould 1968; Jones & White 1988; Tacon 1991). In the absence of information about social and ritual value, however, archaeologists often look to observable measures of the properties of stone that might have influenced the choices of prehistoric knappers. These might help quantify the ease with which stone fractures, how well it holds its fracture path, its suitability for particular functions, its durability, and so on. The quality of raw materials has also been given prominence in modeling raw material selection in times and places where greater demands are placed on toolkit performance (Good- year 1989). Material testing laboratories offer a number of tests that can be useful in determining how suited certain materials are to conchoidal fracture. These include tests of tensile strength, compressive strength, elasticity, hard- ness, and so on.

Sourcing stone Sources of stone with good flaking qualities are not evenly distributed in the artifacts landscape and were often so highly valued that they were exchanged over many hundreds of kilometers. Sourcing stone artifacts from archaeological sites can Why is sourcing therefore provide insights into past exchange networks or interaction spheres, Ap important? changes in territoriality, or changing access to stone resources.

192 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 192 5/16/2013 10:15:48 AM Sourcing analysis is to some extent a misnomer because as Shackley (1998b: How do I find out which 261) points out, “nothing is ever really ‘sourced’. The best we can do is provide source an artifact came from? a probable fit to known source data.” How good this fit is will depend on how good the source data is, which in turn will rely on the location and sampling of potential sources. If no source localities have been mapped or sampled, archaeologists may have to do the field source sampling of geological outcrops themselves. Luedtke (1992) and Shackley (1998b: 262) outline useful hierarchical procedural guidelines for the archaeologist embarking on an artifact sourcing study. In the past, sourcing proceeded by visual inspection and comparison of stone artifacts and samples of stone from potential sources, either in the form of hand specimens, or using microscopic characterization of the structure and texture of ground thin sections mounted on slides. Today. archaeologists rou- tinely use chemical composition analyses such as X-ray fluorescence spectrometry (XRF), neutron activation analysis (NAA or INAA), inductively coupled plasma (ICP)–atomic emission spectrometry, and proton-induced X-ray–proton- induced gamma ray emission (PIXE/PIGME). Quantitative statistical analyses are then used to determine best fit from the results of each analysis. While in some respects this eliminates qualitative assessments and allows vast numbers of specimens to be analysed, sourcing studies will still only be as good as the field sampling that underpins them.

In order to have confidence in the match of artifact to source, it is necessary What is an adequate to have an adequate sample of material from potential source locations as well sample? as an adequate sample of the artifacts from the assemblage, which are to be compared against the source locations. In theory, the more homogeneous the material, the smaller is the sample size that should be necessary to be confident that a representative sample has been obtained. Conversely, the greater the heterogeneity of a geological outcrop, the larger the sample size required to obtain an accurate representation of its variability. While obsidian is relatively homogeneous, cherts and other secondary siliceous sediments can be extremely heterogeneous even within a single small outcrop. When undertaking field sampling, it is also necessary to keep in mind that secondary depositional contexts (such as stream beds and moraines) may have been equally important sources of stone as primary geological sources (such as quarries), and to ignore them in field sampling can result in serious misassignments (Shackley 1998a: 6). These in turn may result in major interpretative errors where stone from an archaeological site is argued to have been traded or exchanged when in fact the raw material could have been derived from a secondary source close to the site.

So far, this chapter has dealt exclusively with traditional 2D (i.e., caliper-based) Is 3D the future of approaches to lithic measurement. Since the first edition of this book was lithic analysis? published, enormous inroads have been made into developing techniques of Ap 3D lithic analysis. 3D analysis is an obvious next step in the sense that stone tools

An Introduction to Stone Artifact Analysis 193

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 193 5/16/2013 10:15:48 AM are complex 3D objects and archaeologists have long expressed dissatisfaction with the limitations inherent in capturing shape, size, surface area, perimeter, or volume using calipers or other kinds of measurement techniques alone. Early attempts made use of 3D capture of data points using 3D point digitizers and specially designed calipers. Clarkson et al. (2006) for instance used a 3D point digitizer (a Microscribe) to calculate an index of scar patterning on cores that could be used to differentiate radial cores, blade cores, and polyhedron or multiplatform cores. Lycett (2007; Lycett et al. 2007) developed his own caliper system to obtain 3D homologous landmark data on the surface of cores for use in sophisticated shape analysis. In the last few years, archaeologists have turned to 3D scanning due to the availability of cheap, high-resolution, desktop scanners that are now available on the market. Grosman et al. (2008) was among the first to present an application of 3D scanning to stone tool analysis by illustrating its ability to create very accurate 3D models of hand axes and to use software to automatically orient and take a series of measurements on each scan. Most recently, archaeologists have begun to look to 3D scanning to explore the potential to create more reliable measures of artifact reduction. For example, Braun et al. (2008) and Clarkson and Hiscock (2011) have devised indices of flake reduction using the 3D surface area of flake platforms to accurately calculate original flake mass and hence determine the amount of stone removed from modified flakes through retouch or breakage. Lin et al. (2010) have also employed 3D scanning to improve on Dibble et al.’s (2005) methods of calculating core reduction intensity by measuring the actual versus the expected proportion of cortex present on flakes in an assemblage by precisely quantifying cortex surface coverage on 3D scans of flakes. Development of new applications in 3D analysis is taking place quickly, and this could well turn out to be the next major direction in lithic analysis. Scan- ning remains time-consuming and expensive at present, however, and may not quickly replace some forms of traditional caliper-based analysis, especially where very large assemblages must be recorded quickly.

Conclusion Material culture, technology, and technological strategies were vital in the operation of all cultural and social processes in the past. For much of human history, stone artifacts constitute a large part of the record of what humans accomplished, how they behaved, and how they interacted with one another. Happily, archaeology now seems set on investing a great deal more effort into rethinking and advancing lithic studies. Many old and current debates require the development of new analytical frameworks for their resolution, and these are also likely to be aided by new advances in archaeometry. New practitioners in lithic studies will witness and take part in the development of new applications that will engage with disciplinary theory to an unprecedented degree. This chapter has attempted to provide a glimpse of the range of approaches employed in the subject today, and a baseline from which the readers may begin to explore the diversity of stone artifact analyses for themselves. Ap

194 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 194 5/16/2013 10:15:48 AM Navin Officer Heritage Consultants are thanked for access to the assemblages Acknowledgments used in our case studies.

Readers in search of critical reviews of the diversity of approaches in lithic Further Reading studies are directed to Odell (2000, 2001b) for a review with a largely North American focus, Dibble (1995) for a view of continental schools of thought, Hiscock and Clarkson (2000) for a review of issues in Australian lithic studies, and Andrefsky (2008) for the latest global research in stone artifact reduction analysis. Another source of such overviews is the growing number of introductory manuals on lithic analysis. Useful texts include Lithics (2006) and Lithic Debitage Analysis (2001) by William Andrefsky, Jr., Lithic Analysis by George Odell (2003), Written in Stone by Nick Kardulius and Yerkes (2003), Understand- ing Stone Tools and Archaeological Sites by Brian Kooyman, and A Record in Stone by Holdaway and Stern (2004). Many more specialized texts exist dealing with theoretical approaches to lithic analysis (Surovell 2009b; Goodale & Andrefksy, in press), the organization of lithic technology in the context of land use and subsistence (Torrence 1989b; Kuhn 1995; Clarkson 2007), the identification and evolution of reduction strategies over the course of human evolution (Dibble & Debenath 1994; Dibble & Bar-Yosef 1995; Hirth 2003; Bradley et al. 2010), trade and exchange (Ericson & Earle 1982; Torrence 1986; Torrence & Sum- merhayes 1997), and of course regionally focused or thematic volumes that are too numerous to list here (but see Clarkson & Lamb 2005; Hiscock & Atten- brow 2005b; Andrefsky 2008; Adams & Blades 2009; Lycett & Chauhan 2010; Goodale & Andrefksy, in press for some recent volumes).

Adams, B. and Blades, B. (eds) (2009) Lithic Materials and Paleolithic Societies. Hoboken, NJ: References Wiley-Blackwell. Adams, J. L. (2010) Understanding grinding technology through experimentation. In J. R. Ferguson (ed.), Designing Experimental Research in Archaeology: Examining Technology through Production and Use. Boulder, CO: University of Colorado Press, pp. 129–52. Ahler, S. A. (1989) Mass analysis of flaking debris: studying the forest rather than the trees. In D. O. Henry and G. H. Odell (eds), Alternative Approaches to Lithic Analysis. Archaeologi- cal Papers of the American Anthropological Association 1, pp. 85–118. Akerman, K. and Bindon, P. (1995) Dentate and related stone biface points from northern Australia. The Beagle, 12, 89–99. Andrefsky, W. (1998) Lithics – Macroscopic Approaches to Analysis. Cambridge, UK: Cambridge University Press. Andrefsky, W. (ed.) (2001) Lithic Debitage: Context, Form, Meaning. Salt Lake City, UT: Uni- versity of Utah Press. Andrefsky, W. (ed.) (2008) Lithic Technology: Measures of Production, Use, and Curation. Cam- bridge, UK: Cambridge University Press. Aubry, T., Bradley, B., Almeida, M., Walter, B., Neves, M. J., Pelegrin, J., Lenoir, M., and Tiffagom, M. (2008) Solutrean laurel leaf production at Maîtreaux: an experimental approach guided by techno-economic analysis. World Archaeology, 40, 48–66. Austin, R. J. (1997) Technological characterization of lithic waste-flake assemblages: multivariate analysis of experimental and archaeological data. Lithic Technology, 24, 53–68. Ap

An Introduction to Stone Artifact Analysis 195

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 195 5/16/2013 10:15:49 AM Bailey, K. D. (1994) Typologies and Taxonomies: An Introduction to Classification Techniques. Thousand Oaks, CA: Sage Publications. Bamforth, D. B. and Bleed, P. (1997) Technology, flaked stone technology, and risk. In C. M. Barton and G. A. Clark (eds), Rediscovering Darwin: Evolutionary Theory in Archaeology. Archaeological Papers of the American Anthropological Association, No. 7. Washington, DC: American Anthropological Association, pp. 109–40. Barton, C. M. (1988) Lithic Variability and Middle Paleolithic Behaviour: New Evidence from the Iberian Peninsula. Oxford: British Archaeological Reports. Baulmer, M. F. and Downum, C. E. (1989) Between micro and macro: a study in the interpretation of small-sized lithic debitage. In D. S. Amick and R. P. Mauldin (eds), Experiments in Lithic Technology. Oxford: British Archaeological Reports, pp. 110–6. Beck, C., Taylor, A. K., Jones, G. T., Fadem, C. M., Cook, C. R., and Millward, S. A. (2002) Rocks are heavy: transport costs and Paleoarchaic quarry behaviour in the Great Basin. Journal of Anthropological Archaeology, 21, 481–507. Bettinger, R. L. and Eerkens, J. (1999) Point typologies, cultural transmission, and the spread of bow-and-arrow technology in the prehistoric Great Basin. American Antiquity, 64, 231–42. Bettinger, R. L. and Eerkens, J. (2008) Evolutionary implications of metrical variation in Great Basin projectile points. Archeological Papers of the American Anthropological Associa- tion, 7(1), 177–91. Binford, L. R. (1979) Organizational and formation processes: looking at curated technologies. Journal of Anthropological Research, 35, 255–73. Blades, B. (2003) End scraper reduction and hunter-gatherer mobility. American Antiquity, 68, 141–56. Bleed, P. (1986) The optimal design of hunting weapons: maintainability or reliability. American Antiquity, 51, 737–47. Bleed, P. (1997) Content as variability, result as selection: toward a behavioural definition of technology. In G. A. Clark and C. M. Barton (eds), Rediscovering Darwin: Evolutionary Theory in Archaeology. Archaeological Papers of the American Anthropological Associa- tion, No. 7. Washington, DC: American Anthropological Association, pp. 95–104. Bleed, P. (2001) Trees or chains, links or branches: conceptual alternatives for consideration of stone tool production and other sequential activities. Journal of Archaeological Method and Theory, 8(1), 101–27. Bleed, P. (2002) Obviously sequential, but continuous or staged? Refits and cognition in three late Paleolithic assemblages from Japan. Journal of Anthropological Archaeology, 21, 329–43. Boëda, E. (1995) Levallois: a volumetric construction, methods, a technique. In H. Dibble and O. Bar-Yosef (eds), The Definition and Interpretation of Levallois Technology. Mono- graphs in World Prehistory No. 23. Madison, WI: Prehistory Press, pp. 41–69. Bordes, F. (1972) A Tale of Two Caves. New York: Harper and Row. Bradbury, A. P. (1998) The examination of lithic artefacts from an Early Archaic assemblage: strengthening inferences through multiple lines of evidence. Midcontinental Journal of Archaeology, 23, 263–88. Bradbury, A. P. and Carr, P. J. (1999) Examining stage and continuum models of flake debris analysis: an experimental approach. Journal of Archaeological Science, 26, 105–16. Bradley, B. A., Collins, M. B., and Hemmings, A. (2010) Clovis Technology. Ann Arbor, MI: International Monographs in Prehistory. Bradley, R. and Edmonds, M. (2005) Interpreting the Axe Trade: Production and Exchange in Neolithic Britain. Cambridge, UK: Cambridge University Press. Brantingham, P. J. (2003) A neutral model of stone raw material procurement. American Antiquity, 68(3), 487–509. Braun, D. R., Rogers, M. J., Harris, J. W. K., and Walker, S. J. (2008) Landscape-scale variation in hominin tool use: evidence from the developed Oldowan. Journal of Human Evolu- Ap tion, 55, 1053–63.

196 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 196 5/16/2013 10:15:49 AM Bright, J., Ugan, A., and Hunskar, L. (2002) The effect of handling time on subsistence technology. World Archaeology, 34(1), 164–81. Buchannan, B. and Collard, M. (2008) Testing models of early Palaeoindian colonisation and adaptation using cladistics. In M. O’Brien (ed.), Cultural Transmission and Archaeology. Washington, DC: Society for American Archaeology, pp. 59–76. Byrne, D. (1980) Dynamics of dispersion: the place of silcrete in archaeological assemblages from the Lower Murchison, Western Australia. Archaeology and Physical Anthropology in Oceania, 15, 110–9. Carr, P. and Bradbury, A. P. (2010) Flake debris and flintknapping experimentation. In J. R. Ferguson (ed.), Designing Experimental Research in Archaeology: Examining Technology through Production and Use. Boulder, CO: University of Colorado Press, pp. 71–92. Charlton, T. H., Nichols, D. L., and Charlton, C. O. (1991) Aztec craft production and specialization: archaeological evidence from the city-state of Otumba, Mexico. World Archae- ology, 23(1), 98–114. Clarkson, C. (2002a) Holocene scraper reduction, technological organization and landuse at Ingaladdi Rockshelter, Northern Australia. Archaeology in Oceania, 37, 79–86. Clarkson, C. (2002b) An index of invasiveness for the measurement of unifacial and bifacial retouch: a theoretical, experimental and archaeological verification.Journal of Archaeologi- cal Science, 1(1), 65–75. Clarkson, C. (2005a) Introduction. In C. Clarkson and L. Lamb (eds), Lithics “Down Under”: Australian Approaches to Lithic Reduction, Use and Classification. British Archaeological Reports International Monograph Series S1408. Oxford: Archaeopress, pp. 1–5. Clarkson, C. (2005b) Tenuous types: scraper reduction continuums in the Eastern Victoria River Region, Northern Territory. In C. Clarkson and L. Lamb (eds), Lithics “Down Under”: Australian Approaches to Lithic Reduction, Use and Classification. British Archaeologi- cal Reports International Monograph Series S1408. Oxford: Archaeopress, pp. 21–34. Clarkson, C. (2006) Interpreting surface assemblage variation in the eastern Victoria River region. In I. Lilley and S. Ulm (eds), An Archaeological Life: Essays in Honour of Jay Hall. Aboriginal and Torres Strait Islander Research Report Series 7. Brisbane: The University of Queensland, pp. 179–92. Clarkson, C. (2007) Lithics in the Land of the Lightning Brothers: The Archaeology of Wardaman Country, Northern Territory. Terra Australis 25. Canberra: ANU E Press. Clarkson, C. (2010) Regional diversity within the core technology of the Howiesons Poort Techno-Complex. In S. Lycett and P. Chauhan (eds), New Perspectives on Old Stones: Analyti- cal Approaches to Palaeolithic Technologies. New York: Springer Press, pp. 43–60. Clarkson, C. (in press) East of Eden: founder effects and the archaeological signature of modern human dispersal. In M. Porr and R. Dennell (eds), East of Africa: Southern Asia, Australia and Human Origins. Cambridge, UK: Cambridge University Press. Clarkson, C. and David, B. (1995) The antiquity of blades and points revisited: investigating the emergence of systematic blade production south-west of Arnhem Land, northern Australia. The Artefact, 18, 22–44. Clarkson, C. and Hiscock, P. (2008) Tapping into the past: exploring Palaeolithic reduction patterns through experimentation. Lithic Technology, 33(2), 1–15. Clarkson, C. and Hiscock, P. (2011) Predicting original flake mass from 3-D platform area. Journal of Archaeological Science, 38, 1062–8. Clarkson, C. and Lamb, L. (eds) (2005) Lithics “Down Under”: Australian Approaches to Lithic Reduction, Use and Classification. British Archaeological Reports International Monograph Series S1408. Oxford: Archaeopress. Clarkson, C. and Schmidt, L. (2009) Stone artefact manufacture at the Natunuku, Votua, Kulu and Ugaga sites, Fiji. In G. Clark and A. Anderson (eds), The Early Prehistory of Fiji. Terra Australis 31. Canberra: ANU E Press, pp. 345–72. Clarkson, C., Vinicius, L., and Lahr, M. M. (2006) Quantifying flake scar patterning on cores using 3D recording techniques. Journal of Archaeological Science, 33, 132–42. Ap

An Introduction to Stone Artifact Analysis 197

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 197 5/16/2013 10:15:49 AM Clarkson, C., Haslam, M., and Harris, C. (in press) When to retouch, haft, or discard? Experimental examinations of the importance of manufacturing costs and use- maintenance schedules in optimal tool use. In N. Goodale and W. Andrefsky (eds), Lithic Technological Systems: Stone, Human Behaviour, Evolution. Cambridge, UK: Cambridge Uni- versity Press. Cotterell, B. and Kamminga, J. (1977) The mechanics of flaking. In B. Hayden (ed.), Lithic Use-Wear Analysis. New York: Academic Press, pp. 97–112. Cotterell, B. and Kamminga, J. (1987) The formation of flakes. American Antiquity, 52(4), 675–708. Crabtree, D. E. (1968) Mesoamerican polyhedral cores and prismatic blades. American Antiq- uity, 33(4), 446–78. Crabtree, D. E. (1972a) The cone fracture principle and the manufacture of lithic materials. Tebiwa, 15, 29–42. Crabtree, D. E. (1972b) An Introduction to Flintworking. Occasional Papers of the Idaho State University Museum. Pocatello, ID: Idaho State University Museum. Crabtree, D. E. and Butler, B. R. (1964) Notes on experiments in flintknapping: 1 heat treatment of silica materials. Tebiwa, 7(1), 1–6. Cross, I., Zubrow, E., and Cowan, F. (2002) Musical behaviours and the archaeological record: a preliminary study. In J. Mathieu (ed.), Experimental Archaeology. British Archaeo- logical Reports International Series 1035. Oxford: Archaeopress, pp. 25–34. Delagnes, A. and Roche, H. (2005) Late Pliocene hominid knapping skills: the case of Loka- lalei 2C, West Turkana, Kenya. Journal of Human Evolution, 48, 435–72. Dibble, H. L. (1995) Middle Paleolithic scraper reduction: background, clarification, and review of evidence to date. Journal of Archaeological Method and Theory, 2(4), 299–368. Dibble, H. L. (1997) Platform variability and flake morphology: a comparison of experimental and archaeological data and implications for interpreting prehistoric lithic technological strategies. Lithic Technology, 22(2), 150–70. Dibble, H. L. and Bar-Yosef, O. (eds) (1995) The Definition and Interpretation of Levallois Tech- nology. Madison, WI: Prehistoric Press. Dibble, H. L. and Debenath, A. (1994) Handbook of Paleolithic Typology: Lower and Middle Paleolithic of Europe. Philadelphia: University of Pennsylvania Museum. Dibble, H. L. and Pelcin, A. (1995) The effect of hammer mass and velocity on flake mass. Journal of Archaeological Science, 22, 429–39. Dibble, H. L. and Rezek, Z. (2009) Introducing a new experimental design for controlled studies of flake formation: results for exterior platform angle, platform depth, angle of blow, velocity and force. Journal of Archaeological Science, 36, 1945–54. Dibble, H. L. and Whittaker, J. (1981) New experimental evidence on the relation between percussion flaking and flake variation. Journal of Archaeological Science, 6, 283–96. Dibble, H. L., Schurmans, U. A., Iovita, R. P., and McLaughlinM, V. (2005) The measurement and interpretation of cortex in lithic assemblages. American Antiquity, 70(3), 545–60. Dobres, M.-A. (1995) Gender and prehistoric technology. World Archaeology, 27(1), 25–49. Dobres, M.-A. (2000) Technology and Social Agency. Oxford: Blackwell. Drennan, R. D. (2009) Statistics for Archaeologists, 2nd Edition: A Commonsense Approach. New York: Springer. Duff, R. (1970) Stone Adzes of Southeast Asia: An Illustrated Typology. Canterbury Museum Bulletin No. 3. Christchurch: Canterbury Museum Trust Board. Dunnell, R. C. (1971) Systematics in Prehistory. New York: The Free Press. Dunnell, R. C. (1986) Methodological issues in Americanist artifact classification. In M. B. Schiffer (ed.), Advances in Archaeological Method and Theory. New York: Academic Press, pp. 35–99. Eerkens, J. W. and Bettinger, R. L. (2008) Cultural transmission and the analysis of stylistic and functional variation. In M. O’Brien (ed.), Cultural Transmission and Archaeology. Wash- Ap ington, DC: Society for American Archaeology, pp. 21–39.

198 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 198 5/16/2013 10:15:49 AM Eren, M. I. (2009) Kuhn’s geometric index of unifacial stone tool reduction (GIUR): does it measure missing flake mass? Journal of Archaeological Science, 36, 1243–7. Eren, M. I., Dominguez-Rodrigo, M., Kuhn, S. L., Adler, D. S., Le, I., and Bar-Yosef, O. (2005) Defining and measuring reduction in unifacial stone tools.Journal of Archaeological Science, 32, 1190–201. Eren, M. I., Greenspan, A., and Sampson, C. G. (2008) Are Upper Paleolithic blade cores more productive than Middle Paleolithic discoidal cores? A replication experiment. Journal of Human Evolution, 55(6), 952–61. Eren, M.I., Kollecker, M. A., Clarkson, C., and Bradley, B. A. (2010a) Dynamic approaches to teaching lithic technology: using YouTube in the classroom. Ethnoarchaeology, 2, 223–33. Eren, M. I., Durant, A., Neudorf, C., Haslam, M., Shipton, C., Bora, J., Korisettar, R., and Petraglia, M. (2010b) Experimental examination of animal trampling effects on artifact movement in dry and water saturated substrates: a test case from South India. Journal of Archaeological Science, 37, 3010–21. Eren, M. I., Bradley, B. A., and Sampson, C. G. (2011) Middle Palaeolithic skill level and the individual knapper: an experiment. American Antiquity, 76(2), 229–51. Ericson, J. E. and Earle, T. K. (1982) Contexts for Prehistoric Exchange. New York: Academic Press. Finlay, N. (2008) Blank concerns: issues of skill and consistency in the replication of Scottish Later Mesolithic blades. Journal of Archaeological Method and Theory, 15, 68–90. Flenniken, J. J. (1978) Reevaluation of the Lindenmeier Folsom: a replication experiment in lithic technology. American Antiquity, 43, 473–9. Flenniken, J. J. and Haggarty, J. (1979) Trampling as an agency in the formation of edge damage: an experiment in lithic technology. Northwest Anthropological Research Notes, 13, 208–14. Flenniken, J. J. and White, J. P. (1983) Heat treatment of siliceous rocks and its implications for Australian prehistory. Australian Aboriginal Studies, 1983(1), 43–8. Geneste, J. M. (1990) Dévelopment des systèmes de production lithique au cours de Paléo- lithique moyen en Aquitaine septentrionale. In C. Farizy (ed.), Paléolithique Moyen Recent et Paléolithique Supérieur Ancien en Europe. Nemours: Mémoirs de Musée d’Ile de France 3, pp. 203–14. Gero, J. M. (1991) Genderlithics: women’s roles in stone tool production. In J. M. Gero (ed.), Engendering Archaeology: Women in Prehistory. Oxford: Basil Blackwell., pp. 163–93. Goodale, N. and Andrefksy, W. (in press) Lithic Technological Systems: Stone, Human Behavior, Evolution. Cambridge, UK: Cambridge University Press. Goodyear, A. C. (1989) A hypothesis for the use of crypto-crystalline raw materials among Paleoindian groups of North America. In C. G. Ellis and J. C. Lothrop (eds), Eastern Paleoindian Lithic Resource Use. Boulder, CO: Westview Press., pp. 1–9. Gordon, D. (1993) Mousterian tool selection, reduction, and discard at Ghar, Israel. Journal of Field Archaeology, 20, 205–18. Gorecki, P., Grant, M., O’Connor, S., and Veth, P. (1997) The morphology, function and antiquity of Australian grinding implements. Archaeology in Oceania, 32, 141–50. Gould, R. A. (1968) Chipping stones in the outback. Natural History, 77(2), 42–8. Greenfield, H. J. (1999) The origins of metallurgy: distinguishing stone from metal cut- marks on bones from archaeological sites. Journal of Archaeological Science, 26(7), 797–808. Grosman, L., Smikt, O., and Smilansky, U. (2008) On the application of 3-D scanning technology for the documentation and typology of lithic artifacts. Journal of Archaeological Science, 35, 3101–10. Hall, J. and Love, W. (1985) Prickly Bush, a site with backed blades on the Brisbane River: a pilot study towards the measurement of site “disturbance”. Queensland Archaeological Research, 2, 71–81. Ap

An Introduction to Stone Artifact Analysis 199

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 199 5/16/2013 10:15:50 AM Hallos, J. (2005) “15 minutes of fame”: exploring the temporal dimension of Middle Pleis- tocene lithic technology. Journal of Human Evolution, 49, 155–79. Harrison, R. (2002) Archaeology and the colonial encounter, Kimberley spearpoints, cultural identity and masculinity in the north of Australia. Journal of Social Archaeology, 2(3), 352–77. Hayden, B. (1979) Paleolithic Reflections: Lithic Technology and Ethnographic Excavations among Australian Aborigines. Canberra: Australian Institute of Aboriginal Studies. Hayden, B. and Hutchings, W. K. (1989) Whither the billet flake? In D. S. Amick and R. P. Mauldin (eds), Experiments in Lithic Technology. Oxford: British Archaeological Reports, pp. 235–57. Hayden, B., Franco, N., and Spafford, J. (1996) Evaluating lithic strategies and design criteria. In G. H. Odell (ed.), Stone Tools: Theoretical Insights into Human Prehistory. New York: Plenum Press, pp. 9–50. Hill, J. N. and Evans, R. K. (1972) A model for classification and typology. In D. L. Clarke (ed.), Models in Archaeology. London: Methuen, pp. 231–74. Hirth, K. (2003) Mesoamerican Lithic Technology. Salt Lake City, UT: University of Utah Press. Hiscock, P. (1981) Comments on the use of chipped stone artefacts as a measure of “intensity of site useage”. Australian Archaeology, 13, 20–34. Hiscock, P. (1982) A technological analysis of quartz assemblages from the south coast. In S. Bowdler (ed.), Coastal Archaeology in Eastern Australia. Canberra: Department of Prehis- tory, RSPAS, ANU, pp. 32–45. Hiscock, P. (1985) The need for a taphonomic perspective in stone artefact analysis. Queens- land Archaeological Research, 2, 82–95. Hiscock, P. (1993) Bondaian technology in the Hunter Valley, New South Wales. Archaeology in Oceania, 28, 64–75. Hiscock, P. (1996) Mobility and technology in the Kakadu coastal wetlands. Bulletin of the Indo-Pacific Prehistory Association, 15, 151–7. Hiscock, P. (2001) Looking the other way: a materialist/technological approach to classifying tools and implements, cores and retouched flakes, with examples from Australia. In S. McPherron and J. Lindley (eds), Tools or Cores? The Identification and Study of Alternative Core Technology in Lithic Assemblages. Philadelphia: University of Pennsylvania Museum. Hiscock, P. (2002a) Pattern and context in the Holocene proliferation of backed artefacts in Australia. In R. G. Elston and S. L. Kuhn (eds), Thinking Small: Global Perspectives on Microlithization. Archaeological Papers of the American Anthropological Association (AP3A) No. 12. Arlington, VA: American Anthropological Association, pp. 163–77. Hiscock, P. (2002b) Quantifying the size of artefact assemblages. Journal of Archaeological Science, 29(3), 251–8. Hiscock, P. (2005) Standardized axe manufacture at Mount Isa. In I. McFarlane, M.-J. Moun- tain, and R. Paton (eds), Many Exchanges: Archaeology, History, Community and the Work of Isabel McBryde. Aboriginal History Monograph 11. Canberra: Aboriginal History Inc., pp. 287–300. Hiscock, P. and Attenbrow, V. (2002) Early Australian implement variation: a reduction model. Journal of Archaeological Science, 30(2), 239–49. Hiscock, P. and Attenbrow, V. (2003) Morphological and reduction continuums in eastern Australia: measurement and implications at Capertee 3. Tempus, 7, 167–74. Hiscock, P. and Attenbrow, V. (2005a) Reduction continuums and tool use. In C. Clarkson and L. Lamb (eds), Lithics “Down Under”: Australian Approaches to Lithic Reduction, Use and Classification. British Archaeological Reports International Monograph Series S1408. Oxford: Archaeopress, pp. 43–56. Hiscock, P. and Attenbrow, V. (2005b) Australia’s Eastern Regional Sequence Revisited: Technol- ogy and Change at Capertee 3. BAR International Series 1397. Oxford: Archaeopress. Hiscock, P. and Clarkson, C. (2000) Analysing Australian stone artefacts: an agenda for the Ap twenty first century. Australian Archaeology, 50(1), 98–108.

200 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 200 5/16/2013 10:15:50 AM Hiscock, P. and Clarkson, C. (2005) Experimental evaluation of Kuhn’s geometric index of reduction and the flat-flake problem. Journal of Archaeological Science, 32, 1015–22. Hiscock, P. and Clarkson, C. (2009) The reality of reduction experiments and the GIUR: reply to Eren and Sampson. Journal of Archaeological Science, 36(7), 1576–81. Hiscock, P. and Veth, P. (1991) Change in the Australian desert culture: a reanalysis of tulas from Puntutjarpa. World Archaeology, 22, 332–45. Holdaway, S. and Stern, N. (2004) A Record in Stone: The Study of Australia’s Flaked Stone Artefacts. Canberra: Aboriginal Studies Press. Holdaway, S., McPherron, S., and Roth, B. (1996) Notched tool reuse and raw material availability in French Middle Paleolithic sites. American Antiquity, 61, 377–87. Ingold, T. (1990) Society, nature and the concept of technology. Archaeological Review from Cambridge, 9, 5–17. Jeske, R. J. and Lurie, R. (1992) The archaeological visibility of bipolar technology: an example from the Koster site. Midcontinental Journal of Archaeology, 18, 131–60. Jeske, R. J., Winkler, D. M., and Blodgett, D. (2010) Experimental heat alteration of lithic raw materials. In J. R. Ferguson (ed.), Designing Experimental Research in Archaeology: Exam- ining Technology through Production and Use. Boulder, CO: University of Colorado Press, pp. 111–28. Jones, R. and White, N. (1988) Point blank: stone tool manufacture at the Ngilipitji Quarry, Arnhem Land, 1981. In R. Jones (ed.), Archaeology with Ethnography: An Australian Perspec- tive. Canberra: Department of Prehistory, Research School of Pacific Studies, The Aus- tralian National University, pp. 51–87. Jung, S. (1992) Trample damage of stone flakes as an index of occupation intensity: a case study from Magnificent Gallery. Queensland Archaeological Research, 9, 26–8. Kardulius, P. N. and Yerkes, R. W. (eds) (2003) Written in Stone: The Multiple Dimensions of Lithic Analysis. Lanham, MD: Lexington Books. Kelly, R. L. (1988) The three sides of a biface. American Antiquity, 53, 717–34. Kenoyer, M., Vidale, M., and Kumar Bhan, K. (1991) Contemporary stone beadmaking in Khambhat, India: patterns of craft specialization and organization of production as reflected in the archaeological record. World Archaeology, 23, 44–63. Kline, M. and Boyd, R. (2010) Population size predicts technological complexity in Oceania. Proceedings of the Royal Society B: Biological Sciences, 277, 2559–64. Kobayashi, H. (1985) The accidental breakage of backed blades. Lithic Technology, 14, 16–25. Kohn, M. and Mithen, S. (1999) Handaxes: products of sexual selection? Antiquity, 73, 518–26. Kooyman, B. P. (2000) Understanding Stone Tools and Archaeological Sites. Calgary, Canada: University of Calgary Press and University of New Mexico Press. Kuhn, S. L. (1990) A geometric index of reduction for unifacial stone tools. Journal of Archaeological Science, 17, 585–93. Kuhn, S. L. (1992) On planning and curated technologies in the Middle Paleolithic. Journal of Anthropological Research, 48, 185–207. Kuhn, S. L. (1995) Mousterian Lithic Technology. Princeton, NJ: Princeton University Press. Lechtman, H. (1984) Andean value systems and the development of prehistoric metallurgy. Technology and Culture, 25, 1–36. Lemonnier, P. (1986) The study of material culture today: toward an anthropology of technical systems. Journal of Anthropological Archaeology, 5, 147–86. Leonard, R. D. and Jones, G. T. (1989) Quantifying Diversity in Archaeology. Cambridge, UK: Cambridge University Press. Leroi-Gourhan, A. (1964) Le Geste et la Parole 1: Technique et Language. Paris: Albin Michal. Lin, S. C. H., Douglass, M. J., Holdaway, S. J., and Floyd, B. (2010) The application of 3D laser scanning technology to the assessment of ordinal and mechanical cortex quantification in lithic analysis. Journal of Archaeological Science, 37, 694–702. Luedtke, B. E. (1992) An Archaeologist’s Guide to Chert and Flint. Archaeological Research Tools, No. 7. Los Angeles, CA: Institute of Archaeology, University of California. Ap

An Introduction to Stone Artifact Analysis 201

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 201 5/16/2013 10:15:50 AM Lycett, S. J. (2007) Is the Soanian techno-complex a Mode 1 or Mode 3 phenomenon? A morphometric assessment. Journal of Archaeological Science, 34, 1434–40. Lycett, S. J. and Chauhan, P. R. (eds) (2010) New Perspectives on Old Stones: Analytical Approaches to Paleolithic Technologies. New York: Springer. Lycett, S. J., von Cramon-Taubadel, N., and Foley, R. A. (2007) A crossbeam co-ordinate caliper for the morphometric analysis of lithic nuclei: a description, test and empirical examples of application. Journal of Archaeological Science, 33, 847–61. Lyman, R. L. and O’Brien, M. J. (2000) Measuring and explaining change in artefact variation with clade-diversity diagrams. Journal of Anthropological Archaeology, 19, 39–74. Lyman, R. L., O’Brien, M. J., and Dunnell, R. C. (1997) The Rise and Fall of Culture History. New York: Plenum Press. MacDonald, D. H. (1998) Subsistence, sex, and cultural transmission in folsom culture. Journal of Anthropological Archaeology, 17, 217–39. Macgregor, O. J. (2005) Abrupt terminations and stone artefact reduction potential. In C. Clarkson and L. Lamb (eds), Lithics “Down Under”: Australian Approaches to Lithic Reduc- tion, Use and Classification. British Archaeological Reports International Monograph Series S1408. Oxford: Archaeopress, pp. 57–66. Macgregor, O. J. (2010) Experimental investigations of mechanical problems in the flake production process, and the identification of problem avoidance strategies in archaeological assemblages, applied to Middle and Upper Palaeolithic sites at Mount Carmel. Unpub- lished PhD Thesis, Australian National University. Mallouf, R. J. (1982) An analysis of plough-damaged chert artefacts: the Brookeen Creek cache (41H186), Hill County, Texas. Journal of Field Archaeology, 9(1), 79–98. Marcy, J. L. (1993) Aperçu sur les stratégies de producion des raclois du niveau Riencourt- lés-Baupaume (Pas-de-Calais):Un Gisement du Paléolithique Moyen. Documents d’Achéologie Français, No. 37, 87–94. Marwick, B. (2008) What attributes are important for the measurement of assemblage reduction intensity? Results from an experimental stone artefact assemblage with relevance to the Hoabinhian of mainland Southeast Asia. Journal of Archaeological Science, 35, 1189–200. Mathieu, J. R. and Meyer, D. A. (1997) Comparing axe heads of stone, bronze, and steel: studies in experimental archaeology. Journal of Field Archaeology, 24, 333–51. McBrearty, S., Bishop, L., Plummer, T., Dewar, R., and Connard, N. (1998) Tools underfoot: human trampling as an agent of lithic artifact edge modification. American Antiquity, 63, 108–29. McCarthy, F. D., Brammell, E., and Noone, H. V. V. (1946) The stone implements of Aus- tralia. Memoirs of the Australian Museum, 9, 1–94. McPherron, S. P., Alemseged, A., Marean, C. W., Wynn, J. G., Reed, D., Geraads, D., Bobe, R., and Béarat, H. A. (2010) Evidence for stone-tool-assisted consumption of animal tissues before 3.39 million years ago at Dikika, Ethiopia. Nature, 466, 857–60. Meignen, L. (1988) Variabilite technologique au proche orient: l’exemple de Kebara. In J. P. Rigaud (ed.), L’homme de Néandertal. Liège: Université de Liège, pp. 87–95. Mercieca, A. and Hiscock, P. (2008) Experimental insights into alternative strategies of lithic heat treatment. Journal of Archaeological Science, 35, 2634–9. Mewhinney, H. (1964) A sceptic views the billet flake. American Antiquity, 30, 203–5. Morrow, T. A. (1997a) A chip off the old block: alternative approaches to debitage analysis. Lithic Technology, 22, 51–69. Morrow, T. A. (1997b) End scraper morphology and use-life: an approach for studying Paleoindian lithic technology and mobility. Lithic Technology, 22(1), 51–69. Neeley, M. P. and Barton, C. M. (1994) A new approach to interpreting late Pleistocene microlith industries in southwest Asia. Antiquity, 68, 275–88. Nelson, M. C. (1991) The study of technological organization. Archaeological Method and Theory, 3, 57–100. Ap

202 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 202 5/16/2013 10:15:50 AM Newcomer, M. H. (1975) “Punch technique” and Upper Paleolithic blades. In E. H. Swanson (ed.), Lithic Technology: Making and Using Stone Tools. The Hague, the Netherlands: Mouton Publishers, pp. 97–102. Newcomer, M. H. and Karlin, C. (1987) Flint chips from Pincevent. In G. D. G. Sieveking and M. H. Newcomer (eds), The Human Uses of Flint and Chert: Proceedings of the Fourth International Flint Symposium, Held at Brighton Polytechnic, 10–15 April 1983. Cambridge, UK: Cambridge University Press, pp. 33–6. Nonaka, T., Bril, B., and Rein, R. (2010) How do stone knappers predict and control the outcome of flaking? Implications for understanding early stone tool technology. Journal of Human Evolution, 59, 155–67. O’Brien, M. J. (1996) Evolutionary archaeology: an introduction. In M. J. O’Brien (ed.), Evolutionary Archaeology: Theory and Application. Salt Lake City, UT: University of Utah Press, pp. 1–15. Odell, G. H. (1989) Experiments in lithic reduction. In D. S. Amick and R. P. Mauldin (eds), Experiments in Lithic Technology. Oxford: British Archaeological Reports, pp. 163–98. Odell, G. H. (2000) Stone tool research at the end of the millennium: procurement and technology. Journal of Archaeological Research, 8(4), 269–331. Odell, G. H. (2001a) Research questions R us. American Antiquity, 66(4), 679–85. Odell, G. H. (2001b) Stone tool research at the end of the millennium: classification, function and behavior. Journal of Archaeological Research, 9(1), 45–100. Odell, G. H. (2003) Lithic Analysis. New York: Springer. Ohnuma, K. and Bergman, C. A. (1990) A technological analysis of the Upper Palaeolithic levels (XXV–VI) of Ksar Akil, Lebanon. In P. Mellars (ed.), The Emergence of Modern Humans: An Archaeological Perspective. Cornell: Cornell University. Parry, W. J. and Kelly, R. L. (1987) Expedient core technology and sedentism. In C. A. Morrow (ed.), The Organization of Core Technology. Boulder, CO: Westview Press, pp. 285–304. Patterson, L. W. (1982) Replication and classification of large sized lithic debitage. Lithic Technology, 11, 50–8. Patterson, L. W. (1990) Characteristics of bifacial-reduction flake-size distribution. American Antiquity, 55, 550–8. Patterson, L. W. and Sollberger, J. B. (1978) Replication and classification of small size lithic debitage. Plains Anthropologist, 23, 103–12. Peacock, E. (1991) Distinguishing between artifacts and geofacts: a test case from Eastern England. Journal of Field Archaeology, 18(3), 345–61. Pelcin, A. (1997a) The effect of core surface morphology on flake attributes: evidence from a controlled experiment. Journal of Archaeological Science, 24, 749–56. Pelcin, A. (1997b) The effect of indentor type on flake attributes: evidence from a controlled experiment. Journal of Archaeological Science, 24, 1107–13. Pelcin, A. (1997c) The formation of flakes: the role of platform thickness and exterior platform angle in the production of flake initiations and terminations.Journal of Archaeological Science, 24, 1107–13. Pelcin, A. (1998) The threshold effect of platform width: a reply to Davis and Shea. Journal of Archaeological Science, 25, 615–20. Peterson, J., Mitchell, D. R., and Shackley, M. S. (1997) The social and economic contexts of lithic procurement: obsidian from Classic-period Hohokam sites. American Antiquity, 62, 231–59. Petraglia, M. D. (2002) The heated and the broken: thermally altered stone, human behavior, and archaeological site formation. North American Archaeologist, 23, 241–69. Pfaffenberger, B. (1992) Social anthropology of technology. Annual Review of Anthropology, 21, 491–516. Phagan, C. J. (1985) Lithic technology: flake analysis. In R. MacNeish, A. Nelken-Terner, C. J. Phagan, and R. Vierra (eds), Prehistory of the Ayacucho Basin, Peru. Ann Arbor, MI: The University of Michigan Press, pp. 233–81. Ap

An Introduction to Stone Artifact Analysis 203

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 203 5/16/2013 10:15:51 AM Prentiss, W. C. (1998) The reliability and validity of a lithic debitage typology: implications for archaeological interpretation. American Antiquity, 63(4), 635–50. Prentiss, W. C. and Romanski, E. J. (1989) Experimental evaluation of Sullivan and Rosen’s debitage typology. In D. S. Amick and R. P. Mauldin (eds), Experiments in Lithic Technology. Oxford: British Archaeological Reports, pp. 89–99. Purdy, B. A. and Brooks, H. K. (1971) Thermal alteration of silica materials: an archaeological approach. Science, 173, 322–5. Renfrew, C., Dixon, J. E., and Cann, J. R. (1968) Further analysis of Near Eastern obsidians. Proceedings of the Prehistoric Society, 34, 319–31. Roebroeks, W., Kolen, J., and Rensink, E. (1988) Planning depth, anticipation and the organization of technology: the “archaic natives” meet Eve’s descendants. Helinium, 28, 17–34. Rosen, S. (1996) The decline and fall of flint. In G. Odell (ed.),Stone Tools: Theoretical Insights into Human Prehistory. New York: Plenum Press, pp. 129–58. Sassaman, K. E. (1992) Lithic technology and the hunter-gatherer sexual division of labor. North American Archaeologist, 13, 249–62. Schiffer, M. B. and Skibo, J. M. (1997) The explanation of artefact variability. American Antiq- uity, 62, 27–50. Schindler, D. L., Hatch, J. W., Hay, C. A., and Bradt, R. C. (1984) Thermal alteration of Bald Eagle Jasper. American Antiquity, 49(1), 173–7. Schnurrenberger, D. and Bryan, A. L. (1985) A contribution to the study of the naturefact/ artifact controversy. In M. G. Plew, J. C. Woods, and M. G. Pavesic (eds), Stone Tools Analy- sis: Essays in Honor of Don E. Crabtree. Albuquerque, NM: University of New Mexico, pp. 133–59. Sellet, F. (1993) Chaîne opératoire; the concept and its applications. Lithic Technology, 18(1), 106–12. Semaw, S., Renne, P., Harris, J. W. K., Feibel, C. S., Bernor, R. L., Fesseha, N., and Mowbray, K. (1997) 2.5-million-year-old stone tools from Gona, Ethiopia. Nature, 385, 333–6. Shackley, M. S. (1998a) Current issues and future directions in archaeological volcanic glass studies. In M. S. Shackley (ed.), Archaeological Obsidian Studies: Method and Theory. New York: Plenum Press, pp. 1–14. Shackley, M. S. (1998b) Gamma rays, X-rays and stone tools: some recent advances in archaeological geochemistry. Journal of Archaeological Science, 25, 259–70. Shafer, H. J. and Hester, T. R. (1991) Lithic craft specialisation and product distribution at the Maya site of Colha, Belize. World Archaeology, 23, 79–97. Shennan, S. (2001) Demography and cultural innovation: a model and its implications for the emergence of modern human culture. Cambridge Archaeological Journal, 11, 5–16. Sheppard, P. (1993) Lapita lithics: trade/exchange and technology. A view from the reefs/ Santa Cruz. Archaeology in Oceania, 28(3), 121–37. Shott, M. J. (1986) Settlement mobility and technological organization. An ethnographic examination. Journal of Anthropological Research, 42, 15–51. Shott, M. J. (1989) On tool-class use lives and the formation of archaeological assemblages. American Antiquity, 54, 9–30. Shott, M. J. (1994) Size and form in the analysis of flake debris: review of recent approaches. Journal of Archaeological Method and Theory, 1, 69–110. Shott, M. J. (1996) Stage versus continuum models in the debris assemblage from production of a fluted biface. Lithic Technology, 21(1), 6–22. Shott, M. J. (1997) Transmission theory in the study of stone tools: a Midwestern North Ameri- can example. Archeological Papers of the American Anthropological Association, 7(1), 193–204. Shott, M. J. (2000) The quantification problem in stone-tool assemblages.American Antiquity, 65(4), 725–38. Shott, M. J. (2003) Chaîne opératoire and reduction sequence. Lithic Technology, 28, 95–105. Shott, M. J., Bradbury, A. P., Carr, P. J., and Odell, H. O. (2000) Flake size from platform Ap attributes: predictive and empirical approaches. Journal of Archaeological Science, 27, 877–94.

204 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 204 5/16/2013 10:15:51 AM Sillitoe, P. and Hardy, K. (2003) Living lithics: ethnoarchaeology in highland Papua New Guinea. Antiquity, 77, 555–66. Sinclair, A. (1995) The technique as symbol in late glacial Europe. World Archaeology, 27(1), 50–62. Sinclair, A. (2000) Constellations of knowledge: human agency and material affordance in lithic technology. In M.-A. Dobres and J. Robb (eds), Agency in Archaeology. London: Routledge, pp. 197–212. Sisk, M. L. and Shea, J. J. (2009) Experimental use and quantitative performance analysis of triangular flakes (Levallois points) used as arrowheads. Journal of Archaeological Science, 36, 2039–47. Smith, M. (1985) A morphological comparison of central Australian seedgrinding implements and Australian Pleistocene-age grindstones. The Beagle, 2, 23–38. Soriano, S., Villa, P., and Wadley, L. (2007) Blade technology and tool forms in the Middle Stone Age of South Africa: the Howiesons Poort and post-Howiesons Poort at Rose Cottage Cave. Journal of Archaeological Science, 34, 681–703. Spaulding, W. G. (1953) Statistical techniques for the discovery of artefact types. American Antiquity, 18, 305–13. Speth, J. D. (1972) Mechanical basis of percussion flaking. American Antiquity, 37, 34–60. Speth, J. D. (1974) Experimental investigations of hard-hammer percussion flaking. Tebiwa, 17, 7–36. Speth, J. D. (1981) The role of platform angle and core size in hard-hammer percussion flaking. Lithic Technology, 10(1), 16–21. Stafford, M. (2003) The parallel-flaked flint daggers of late Neolithic Denmark: an experimental perspective. Journal of Archaeological Science, 30, 1537–50. Stahle, D. W. and Dunn, J. E. (1984) An analysis and application of the size distribution of waste flakes from the manufacture of bifacial stone tools. World Archaeology, 14, 84–97. Stark, M. (1998) Technical choices and social boundaries in material culture patterning: an introduction. In M. Stark (ed.), The Archaeology of Social Boundaries. Washington, DC: Smithsonian Institution Press. Steffen, A., Skinner, E. J., and Ainsworth, P. W. (1998) A view to the core: technological units and debitage analysis. In A. Ramenofsky and A. Steffen (eds), Unit Issues in Archaeol- ogy. Salt Lake City, UT: University of Utah Press, pp. 131–46. Stout, D. (2002) Skill and cognition in stone tool production: an ethnographic case study from Irian Jaya. Current Anthropology, 43, 693–722. Stout, D. (2005) The social and cultural context of stone-knapping skill acquisition. In V. Roux and B. Brill (eds), Stone Knapping: The Necessary Conditions for a Uniquely Hominin Behaviour. Cambridge, UK: McDonald Institute, pp. 331–40. Stout, D., Toth, N., Schick, K., and Chaminade, T. (2008) Neural correlates of Early Stone Age toolmaking: technology, language and cognition in human evolution. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 363, 1939–49. Sullivan, A. P. and Rozen, K. C. (1985) Debitage analysis and archaeological interpretation. American Antiquity, 50, 755–79. Surovell, T. A. (2009a) Behavioural Ecology of Lithic Technology: Cases from Paleoindian Archaeol- ogy. Tucson, AZ: University of Arizona Press. Surovell, T. A. (2009b) Toward a Behavioral Ecology of Lithic Technology: Cases from Paleoindian Archaeology. Tucson, AZ: University of Arizona Press. Tacon, P. S. C. (1991) The power of stone: symbolic aspects of stone use and tool development in western Arnhem Land, Australia. Antiquity, 65, 192–207. Torrence, R. (1983) Time budgeting and hunter-gatherer technology. In G. Bailey (ed.), Hunter-Gatherer Economy in Prehistory. Cambridge, UK: Cambridge University Press, pp. 11–22. Torrence, R. (1986) Production and Exchange of Stone Tools: Prehistoric Obsidian in the Aegean. Cambridge, UK: Cambridge University Press. Ap

An Introduction to Stone Artifact Analysis 205

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 205 5/16/2013 10:15:51 AM Torrence, R. (1989a) Re-tooling: towards a behavioral theory of stone tools. In R. Torrence (ed.), Time, Energy and Stone Tools. Cambridge, UK: University of Cambridge, pp. 57–66. Torrence, R. (ed.) (1989b) Time, Energy and Stone Tools. Cambridge, UK: University of Cambridge. Torrence, R. and Summerhayes, G. (1997) Sociality and the short distance trader: interregional obsidian exchange in the Willaumez Peninsula region, Papua New Guinea. Archaeology in Oceania, 32(1), 74–8. Tostevin, G. B. (2007) Social intimacy, artefact visibility and acculturation models of Neanderthal–modern human interaction. In P. Mellars, K. Boyle, O. Bar-Yosef, and C. Stringer (eds), Rethinking the Human Revolution: New Behavioural and Biological Perspectives on the Origin and Dispersal of Modern Humans. Cambridge, UK: MacDonald Institute, pp. 341–57. Touhy, D. R. (1987) A comparison of pressure and percussion debitage from a Crabtree obsidian stoneworking demonstration. Tebiwa, 23, 23–30. Ugan, A., Bright, J., and Rogers, A. (2003) When is technology worth the trouble? Journal of Archaeological Science, 30, 1315–29. Ulm, S., Cotter, S., Cotter, M., Lilley, I., Clarkson, C., and Reid, J. (2005) Edge-ground hatchets on the southern Curtis Coast, central Queensland: a preliminary assessment of technology, chronology and provenance. In I. MacFarlane, M.-J. Mountain, and R. Paton (eds), Many Exchanges: Archaeology, History, Community and the Work of Isabel McBryde. Aboriginal History Monograph 11. Canberra: Aboriginal History Inc., pp. 323–42. van der Leeuw, S. E. and Torrence, R. (1989) What’s New? A Closer Look at the Process of Innovation. London: Unwin Hyman. Van Peer, P. (2007) Refitting of lithic reduction sequences, formal classification systems, and Middle Palaeolithic individuals at work. In U. Schurmans and M. De BieFitting (eds), Fitting Rocks: Lithic Refitting Examined. BAR 1596. Oxford: Archaeopress, pp. 91–104. Walthall, J. A. and Holley, G. R. (1997) Mobility and hunter-gatherer toolkit design: analysis of a Dalton lithic cache. Southeastern Archaeology, 16, 152–62. Weedman, K. (2006) An ethnoarchaeological study of hafting and stone tool diversity among the Gamo of Ethiopia. Journal of Archaeological Method and Theory, 13, 189–238. White, J. P. and O’Connell, J. F. (1982) A Prehistory of Australia, New Guinea and Sahul. Sydney: Academic Press. Whittaker, J. (1994) Flintknapping: Making and Understanding Stone Tools. Austin, TX: Univer- sity of Texas Press. Wiessner, P. (1983) Style and social information in Kalahari San projectile points. American Antiquity, 48, 253–76. Wright, K. (1991) The origins and development of ground stone assemblages in late Pleis- tocene southwest Asia. Paléorient, 17, 19–45. Wurz, S. (1999) The Howiesons Poort backed artefacts from Klaises River: an argument for symbolic behaviour. South African Archaeological Bulletin, 169, 38–50. Yvorra, P. (2000) Exploitation de l’analyse quantitative des retouches pour la caracterisation des industries lithiques du Mousterien: application au facies Quina de la vallee du Rhone. BAR International Series, No. 869. Oxford: Oxbow. Zeitlin, R. N. (1982) Toward a more comprehensive model of interregional commodity distribution: political variables and prehistoric obsidian procurement in Mesoamerica. American Antiquity, 47, 260–75.

206 Chris Clarkson and Sue O’Connor

Balme—Archaeology in Practice

Balme_7164_c06_main.indd 206 5/16/2013 10:15:51 AM