Stone toolmaking difficulty and the evolution of hominin technological skills

Abstract

Stone tools are a manifestation of the complex cognitive and dexterous skills of our hominin ancestors. As such, much research has been devoted to understanding the skill requirements of individual lithic technologies. Yet, comparing skill across different technologies, and thus across the vast timespan of the Palaeolithic, is an elusive goal. We seek to quantify a series of commensurable metrics of knapping skill across four different lithic technologies (discoids, handaxes, Levallois, and prismatic blades). To compare the requisite dexterity, coordination, and care involved in each technology, we analysed video footage and lithic material from a series of replicative knapping experiments to quantify deliberation (strike time), precision (platform area), intricacy (flake size relative to core size), and success (relative blank length). According to these four metrics, discoidal knapping appears to be easiest among the sample. Levallois knapping involved an intricate reduction sequence, but did not require as much motor control as handaxes and especially prismatic blades. Compared with the other Palaeolithic technologies, we conclude that prismatic blade knapping is set apart by being a skill intensive means of producing numerous standardised elongate end-products.

Figure 1

The experimental setup and examples of the lithics.(a) Knapping footage. (b) Discoidal core and flakes. (c) Earlier (left) and later (right) handaxes. (d) Preferential Levallois core and flake. (e) Recurrent Levallois core and flakes. (f) and (g) Prismatic blade cores and blades.

Figure 2

Results of the proxy measures for knapping difficulty. (a) Boxplot of the strike times (seconds) taken for Levallois (N = 1757, median = 0.53s), handaxe (N = 649, median = 0.60s), discoidal (N = 250, median = 0.60s), and prismatic blade (N = 469, median = 0.77s) knapping. (b) Boxplot of the flake platform area values for the discoidal (N = 150, med = 160.08mm²), Levallois (N = 543, med = 79.91mm²), handaxe (N = 175, med = 38.27mm²), and prismatic blade (N = 184, med = 38.53mm²) flakes. (c) Boxplot of individual flake mass versus total flake mass from the same core for the discoidal (N = 165, med = 0.0067), handaxe (N = 357, med = 0.0039), prismatic blade (N = 289, med = 0.0036), and Levallois (N = 698, med = 0.0024) iterations of this experiment. A logarithmic scale on the y-axis in b and c is used to better display the positively skewed data. (d) Histograms showing the percent of blank length to core length for the discoidal (N = 118, med = 90), Levallois (N = 18, med = 85) and prismatic blade (N = 169, med = 70) iterations.

Figure 3

Visual approximation of the inherent traits of the technologies investigated here. 3D scatter plot of the skill, efficiency and hierarchical complexity involved in (a) discoidal, (b) handaxe, (c) Levallois, and (d) prismatic blade knapping.