Levallois economics: an examination of ‘waste’ production in experimentally produced Levallois reduction sequences

Mathematical modelling has suggested that Levallois core morphology represents a reduction strategy driven by economic considerations; particularly the minimization of ‘waste’ while aiming to maximize cutting edge length of flakes obtained from cores of a given size. Such models are elegant in that they facilitate formal modelling of economic considerations that potentially motivate patterns seen in prehistoric data. However, the abstract nature of such models means that they do not take full account of all the practical difficulties and material challenges involved in reproducing Levallois-style reductions in stone. In particular, such models have only examined nodule morphology in two-dimensions, and did not take account of the fact that in the case of classic (lineal) Levallois reduction, core surfaces must be re-prepared between successive stages of flake removal. Hence, the potential economic implications of these factors are currently unknown, potentially undermining the significance of models that assume specific economic conditions. Here, we undertook to examine these factors using a series of experimentally produced Levallois reduction sequences. A total of 3957 flaking events were considered in our analyses, and we used six specific measures of economy to examine Levallois reduction across successive phases. Our analyses found that key assumptions of mathematical models suggesting that Levallois core morphology was driven by economic considerations (i.e. conservation of raw material when attempting to remove flakes with long cutting edges) can be upheld under the practical challenges of replicating Levallois-style reduction in stone. In supporting the notion that Levallois reduction has advantageous economic properties, our results emphasize the importance of considering why Levallois reduction did not emerge earlier in the archaeological record, and indeed, why even during the later Pleistocene the temporal and geographic distribution of Levallois technology varies. Our results also re-emphasize the value of formally modelling lithic reduction strategies in specific economic terms.     

Milky Quartz Bipolar Reduction and Lithic Miniaturization: Experimental Results and Archaeological Implications

We present experimental data examining the energetics and identification of axial bipolar reduction in contexts of lithic miniaturization on milky quartz. These experiments answer two specific questions. First, does bipolar reduction provide any benefits over freehand reduction? Second, can axial bipolar reduction be distinguished from freehand reduction? Our data show that bipolar reduction requires significantly less time to reduce a percentage unit of core mass and to produce a millimeter of cutting edge on milky quartz than freehand reduction. Milky quartz bipolar reduction surpasses even the cutting edge production efficiency of obsidian pressure blades. We outline a series of quantitative criteria for identifying bipolar cores and flakes. Our results show that bipolar cores and flakes can be distinguished from those produced using freehand reduction by quantifying platform crushing, distal flake rebound scars, bulb shearing, as well as axial, bipolar, and splintered flake terminations. Our results challenge the widely held perceptions about the wastefulness of bipolar reduction and provide clear guidelines for identifying this reduction strategy in archaeological milky quartz assemblages.     

A 3D morphometric analysis of surface geometry in Levallois cores: patterns of stability and variability across regions and their implications

Levallois cores and products were manufactured by hominin populations distributed across wide regions of Africa and Eurasia. Levallois technology remains an important focus for research in Palaeolithic archaeology, yet quantitative morphological comparisons of Levallois core morphology from different regions remain rare. Here, utilizing Levallois cores from Africa, the Near East, Europe, and the Indian subcontinent, patterns of morphological variability in the shape of the Levallois flaking surface and core outline (margin) shape were examined for patterns of variability and stability across regions using 3D geometric morphometrics. The multivariate statistical shape analyses undertaken revealed a clear pattern: that is, the greatest levels of shape variability in Levallois cores is evident in the form of their outline (planform) shape. Conversely, the geometrical relationship between the margin of the Levallois cores and their topological surface morphology was relatively uniform. This pattern of variability was evident in terms of variation both across regions and between cores from the same locality. These results indicate that the outline form of such cores was a less important variable than the geometric/topological properties of the surface morphology and, in particular, the relationship between the margin of the core and those variables. These results have implications for why it has been reported that replicating such cores in modern experiments is a particularly difficult task. The specific interrelationship between the geometric properties of the core and the core margin provide further evidence that Levallois core technology would be unlikely to emerge from the context of opportunistic migrating platform reduction strategies (such as those seen in many Mode 1 industries). If, as is widely suggested, Levallois cores were deliberate products in Pleistocene contexts, these results also hint that relatively sophisticated means of social transmission (i.e. teaching) may have been required to sustain their production over time and space.     

Edge Angle as a Variably Influential Factor in Flake Cutting Efficiency: An Experimental Investigation of Its Relationship with Tool Size and Loading

Simple flake cutting tools were utilized across broad chronological and geographical ranges during prehistory. Fundamental to their functional utility is the presence of a relatively acute working edge. The acuteness of this ‘edge angle’ is widely hypothesized to be a primary determinant of cutting efficiency and, subsequently, of potential consequence to prehistoric peoples. However, the influence of the cutting edge angle in flake tools on the ability (efficiency) of tool users to cut through objects has not been empirically investigated under explicitly stated experimental conditions. Moreover, no consideration has been given to whether this relationship is dependent upon the size of the tool. Here, the influence that edge angle exerts on human stone tool users is examined experimentally in terms of efficiency during a cutting task, while also considering the relationship between edge angle, loading (i.e., the force applied) and overall flake size. The results demonstrate that there is a highly significant relationship between more acute working edges and increased cutting efficiency in the smallest flake tools tested. Above a certain flake‐size threshold, however, the working edge angle has no influence on cutting efficiency because larger flakes appear to facilitate the application of greater working loads by tool users. These results have important implications for potential flake selection criteria by prehistoric peoples, especially in relation to utility, function and the changing effects of edge angle through a sequence of retouch.     

Introducing a new experimental design for controlled studies of flake formation: results for exterior platform angle,platform depth,angle of blow,velocity, and force

Potential variables that underlie variation in flake size, and in some instances shape, are investigated in a newly designed experiment. This new design, which utilizes glass cores molded to a specific shape, results in flakes that are identical to archaeological ones. Variation in exterior platform angle, platform depth and angle of blow all directly affect flake size, and in the case of exterior platform angle, flake shape as well (in spite of constant core surface morphology). In treating velocity and force independently, neither is found to affect flake size or shape. These results have implications for understanding different strategies that flintknappers may employ to control the size of their flake products.     

Experimental production of bending and radial flake fractures and implications for lithic technologies

Bend and radially broken flake tools have been identified in Paleolithic and Paleoindian assemblages, and their presence raises important questions. Were these breaks intentionally produced to serve as tool edges or were broken flakes simply scavenged? More importantly, can we distinguish between intentionally produced breaks and those produced incidentally? Experimental archaeology can help answer these questions. In this paper, three sets of experimentally produced bend and radial flake breaks were compared. Flakes were intentionally broken by percussion, and these breaks were compared to those produced during bifacial core reduction and by flake trampling. The presence of point of impact markers, near ninety degree break angles, and an assemblage with high percentages of bend and radial breaks distinguish intentional fracture from incidental fractures produced during bifacial reduction. High percentages of radial breaks distinguish intentional fracture from trampling. Finally, it may not be possible to identify intentional breaks in a bifacial reduction assemblage severely affected by flake-on-flake trampling.     

The Spatial Organization of Bone Crafting During the Middle and Late Mesolithic at Ringsjöholm and Strandvägen in Sweden

This paper focuses on the spatial distribution of bone tool production waste from two Mesolithic sites in Sweden, Ringsjöholm and Strandvägen, with well-preserved faunal remains including bone and antler artifacts. Local production on both sites has generated a variety of identifiable waste products deriving from complete chains of production, including unmodified bones, debitage and finished products. Identified categories include: blanks, removed epiphyses, bone flakes, and preforms. Identification of species shows that antler and bone from red deer were the preferred raw materials. Spatial statistical analyses confirm that different stages of bone tool production were organized within separate areas of the sites and that larger items were discarded in the water along the shorelines. Interestingly, blanks and preforms seem to have been stored under water for future use and demarcated clusters of bone flakes in association with dwellings represent “bone knapping floors” where production was more intense than in other areas.     

Technology based evolution? A biometric test of the effects of handsize versus tool form on efficiency in an experimental cutting task

The use of stone cutting tools opened a novel adaptive niche for hominins. Hence, it has been hypothesised that biomechanical adaptations evolved to maximise efficiency when using such tools. Here, we test experimentally whether biometric variation influences the efficiency of simple cutting tools (n = 60 participants). Grip strength and handsize were measured in each participant. 30 participants used flint flakes, while the other 30 used small (unhafted) steel blades. Variations in basic parameters of tool form (length, width, thickness, cutting edge length) were recorded for the 30 flint flakes. It was ensured that mean handsize and strength in each participant group were not significantly different to investigate the effect of tool variation. The experimental task required cutting through a 10 mm-diameter hessian rope. Cutting efficiency was measured using both ‘Number of cutting strokes required’ and ‘Total time taken’. Results show that both efficiency measures were significantly correlated with handsize using all 60 participants. However, no significant differences were found between the flake and blade groups in terms of mean efficiency. Nor was any significant relationship found between tool form parameters and efficiency in the flint flake group. We stress that our results do not imply that tool form has no impact on tool efficiency, but rather that – all other things being equal – biometric variation has a statistically significant influence on efficiency variation when using simple cutting tools. These results demonstrate that biomechanical parameters related directly to efficiency of use, may plausibly have been subject to selection in the earliest stone tool-using hominins.     

The oldowan reassessed: A close look at early stone artifacts

Early Stone Age assemblages called “Oldowan” and early “Developed Oldowan” are discussed, based on the results of a long-term study of Plio-Pleistocene sites at Koobi Fora, Kenya and an extensive experimental research program of replicating and using early stone artifact forms. Five major conclusions are drawn from this investigation: (1) many Oldowan core forms (“core-tools”) are probably simple by-products of flake manufacture rather than representations of stylistic norms; (2) flakes and retouched flakes - were essential tools in Oldowan technology, particularly for activities involving cutting; (3) this simple technology does not necessarily reflect the cognitive abilities of the early hominids that manufactured the stone artifacts; (4) there is evidence to show that Oldowan technology can be viewed as a simple curated one, in which raw material was intentionally carried from place to place for future use; (5) early hominid populations that made and used stone implements were not necessarily dependent upon them for their survival.     

The relative effects of core surface morphology on flake shape and other attributes

It is long been thought that many flake attributes, including both size and shape, are largely due to the morphology of a core’s flaking surface, yet this has never been tested under strictly controlled conditions. Using molded glass cores with surface morphologies that highly resemble prehistoric ones, this experiment demonstrates that while core surface morphology does exhibit some influence on flake size and shape, a high degree of variation in flakes produced with the same core surface morphology shows that the effects of other independent variables, such as exterior platform angle and platform depth, have an even stronger effect. A major implication of these results is that current approaches to reconstruct prehistoric knapping strategies are overlooking significant sources of variation.     

Estimating original flake mass from 3D scans of platform area

This study confirms the increased capacity to predict flake mass that arises from more accurately measuring surface area in three dimensions using a digital scanner. We also reveal the existence of significantly different relationships between platform area and flake mass for flakes with different platform types and ventral and dorsal morphologies. These different relationships between platform surface area and mass have not been previously identified, and reveal the complexity of platform/mass relationships. Using multivariate regression of 3D platform surface area and external platform angle we improve the accuracy of predictions of original flake mass. We propose a method for studying reduction intensity on retouched flakes, based on comparing predictions of the initial mass of the flake with measurements of the mass following retouching to estimate the amount of mass removed through retouching. We name this approach the Initial-/Terminal-Mass Comparison, or ITMC. Our experiments demonstrate that the capacity of the 3D platform scans to predict flake mass, and by implication the capacity of the ITMC to estimate mass loss, rivals or exceeds the capacity demonstrated for existing reduction measures.     

New experimental evidence on the relation between percussion flaking and flake variation

Controlled experiments in percussion flaking allowed for objective analysis of relationships between variables of flake production and those variables which are attributes of the final result. The independent variables, those controlled by the flintknapper in the production of stone tools, include force and angle of blow, platform thickness and exterior platform angle. The dependent variables are those attributes of the flakes which are often used in current lithic analyses and include interior platform angle, length, thickness and flake termination. The results clearly show that exterior platform angle is highly significant for understanding many aspects of flake production. These and other relationships between the independent and dependent variables are also discussed.     

The technology of Stone Age colonization: an empirical,regional-scale examination of Clovis unifacial stone tool reduction,allometry, and edge angle from the North American Lower Great Lakes region

There is now broad consensus that the appearance of Clovis in Northeastern North America (Great Lakes, New England) represents a colonization pulse into recently deglaciated landscapes. Due to the increased resource uncertainty that comes with colonizing unfamiliar landscapes, it was hypothesized that the majority tool component of Clovis assemblages, unifacial stone tools, should have been knapped on tool blanks possessing the design properties of longevity and functional flexibility to facilitate exploration mobility and guard against the absence of toolstone sources in the new landscape. These properties are optimized by large, flat flakes, possessing large surface area relative to flake thickness. Since discarded and, at times, exhausted unifacial stone tools do not preserve the original dimensions of the blank upon which they were created – necessary items for a true test of blank morphology selection – this study presents a set of predictions for inferring whether Clovis unifacial stone tool blanks were selected for the properties of longevity and functional flexibility based on evidence that Clovis people actually capitalized on those properties. Due to the nature of Clovis unifacial stone tools, tool size was of necessity used as a proxy for tool reduction, on the grounds that smaller tools are more likely to have been resharpened than larger tools, at least in the case of unifacial flake tools. The results showed that less resharpened tools possessed flatter, less spherical shapes than the more resharpened tools, which possessed more globular, spherical shapes, suggesting Clovis foragers exploited the retouch potential afforded by the larger, flatter blanks. Edge angles showed no relationship with tool reduction, suggesting that Clovis foragers exploited the functional flexibility afforded by flatter blanks by adjusting the edge angle to be either higher or lower as needed. These results are consistent with the notion that human colonizers, who did not know the abundance or location of stone outcrops prior to settling an unfamiliar territory, not only “geared up” before leaving a stone source, but geared up as efficiently as possible by carefully selecting the blanks they chose to carry. Broader implications for such careful unifacial stone tool blank selection are discussed.     

Constraints on Levallois Core Technology: A Mathematical Model

Recent volumetric definitions of Levallois core technology are amenable to mathematical modelling. We present a simple geometric model that permits controlled manipulation of a few of the key parameters defining Levallois core morphology. The models indicate that Levallois cores are relatively efficient at minimizing raw material waste while at the same time maximizing productivity in terms of total number of tool blanks and amount of cutting edge produced. Deviations from an ideal Levallois geometry produce significant declines in both efficiency and productivity. These results implicate mechanical and economic constraints as factors underlying the broad geographic distribution and temporal persistence of Levallois core technologies during the Middle and Late Pleistocene.     

Measuring core reduction using 3D flake scar density: a test case of changing core reduction at Klasies River Mouth,South Africa

A new core reduction index is presented, calculated as the ratio of flake scar number to 3D surface area (SDI). The index is tested experimentally on five types of core (blade, discoid, Levallois, biface and multiplatform cores) and then applied to the core assemblages from five sub-stages of the Middle Stone Age at Klasies River Mouth, South Africa. Preliminary results indicate that the SDI possesses the desirable attributes of a successful reduction index and is a significant improvement on traditional proxy measures of core reduction. The results of the archaeological case study confirm previous untested observations that cores from the Howiesons Poort and MSAIII sub-stages are more heavily reduced than preceding and following stages, and that local and exotic raw materials as well as different types of cores are all more heavily reduced during these periods. The SDI fills a significant lacuna in available core reduction measures.     

The Formation of Flakes: The Role of Platform Thickness and Exterior Platform Angle in the Production of Flake Initiations and Terminations

This paper presents the results of several controlled experiments which examined Cotterell & Kamminga's (1987,American Antiquity52, 675–708) model of flake initiations and terminations in relation to platform thickness and exterior platform angle. The results contribute to their model by demonstrating that for a given exterior platform angle, as platform thickness was increased, predictable changes occurred in the type of initiation and terminations produced. In addition, low exterior platform angles were found to produce changes similar to those usually attributed to different indentor types. These experimental results suggest that the differences between bending and conchoidal flakes must be re-evaluated.     

A biface and blade core efficiency experiment: implications for Early Paleoindian technological organization

Early Paleoindians often are described as highly mobile hunter–gatherers who employed lithic technologies designed to minimize stone transport costs. We experimentally reduced blade and bifacial cores and found both reduction strategies to be equally efficient for the production of useable flake blanks. Further, when compared to similar core reduction experiments, the results of this study showed no significant differences in core efficiency between bifacial, prismatic blade, and wedge-shaped blade core reduction. Biface and blade cores with initial weights greater than 1000 g produced useable flakes as efficiently as informal cores. However, bifacial and blade core efficiency decreased with initial core weight. When considered in terms of Early Paleoindian technological organization, differences in core efficiencies suggest that Folsom groups employed core reduction strategies designed to minimize stone transport costs, but Clovis groups did not.     

Investigating interrelationships between Lower Palaeolithic stone tool effectiveness and tool user biometric variation: implications for technological and evolutionary changes

Lower Palaeolithic hominins are thought to have been dependent upon stone tools during the acquisition and processing of food resources. Hence, it is hypothesized that the evolutionary advantages provided by efficient stone tool use may have selected for anatomical changes observed in the hand during this period. Similarly, hominin manipulative capabilities are suggested to have been of consequence to Lower Palaeolithic technological choices and tool use capabilities. The extent and character of these relationships are not, however, fully understood and it is not known whether these hypothesized co-evolutionary and co-dependent relationships are consistent across varying technological and task-type conditions. Here, six key biometric parameters of the hand are investigated in terms of their statistical relationship with cutting efficiency using both flakes and handaxes over extended periods of use and in multiple types of cutting task. Results indicate that (1) both handaxe and flake cutting efficiencies are significantly related with biometric variation of individual tool users, (2) relationships between biometric parameters and efficiency are consistent across extended durations but vary dependent upon task-type conditions, (3) manipulative strength is the most significant biometric trait in terms of predicting flake efficiency, while (4) hand size is the strongest predictor of handaxe cutting efficiency. These results demonstrate the long-term impact that stone tool use likely had on the evolution of hominin biometric variation during the Lower Palaeolithic, while also highlighting the variable influence of different tool use contexts. Most notably, results indicate that the onset of the Acheulean may have been dependent, a priori, upon hand dimensions that are close to the modern human range, and that prior to the appearance of this anatomy, handaxe use would have been an impractical (i.e. inefficient) tool use behaviour compared to the use of flakes.     

Is the Soanian techno-complex a Mode 1 or Mode 3 phenomenon? A morphometric assessment

The Soanian is traditionally seen as one of the major (non-Acheulean) Palaeolithic techno-complexes of the Indian subcontinent. Over several decades comparisons of Soanian assemblages have been made with the non-bifacial industries of East Asia and north-west Europe. The chronological status and typo-technological relationship(s) of the Soanian to other Palaeolithic industries have been the subject of much debate. When first named and described the Soanian was considered to contain evidence of Mode 3 Levallois-style core reduction. However, in recent years, the potential Mode 3 component of the Soanian has largely been ignored, and the techno-complex is described under various guises as a core/flake or ‘Mode 1’ techno-complex. Here, a comparative morphometric assessment of selected Soanian cores and other Palaeolithic nuclei is undertaken, to test the hypothesis that this industry contains a definite Mode 3 Levallois element. Discriminant Function Analyses (DFA) of morphometric variables provide robust evidence that at least part of the Soanian techno-complex contains Mode 3 Levallois cores. The implications of these analyses for the relationship between the Soanian and the Acheulean, and the relevance of the Soanian in considerations of the Movius Line are also discussed.