Exploring variability in lithic armature discard in the archaeological record

https://doi.org/10.1016/j.jhevol.2021.102981Get rights and content

Abstract

The invention of projectile technology had important ramifications for hominin evolution. However, the number of stone points that could have been used as projectiles fluctuates in archaeological assemblages, making it difficult to define when projectile technology was first widely adopted and how its usage changed over time. Here we use an agent-based model to simulate a hunter-gatherer foraging system where armatures are dropped according to their usage. We explore the impact of interactions between human behaviors and the environmental constraints of a data-informed landscape on the distribution and number of lithic armatures found in archaeological assemblages. We ran 2400 simulations modeling different population sizes, rates of hunting with projectiles, and tool curation levels. For each simulation, we recorded the location of dropped armatures and calculated the number and percentage of used armatures that were discarded at habitation camps vs. lost during hunting. We used linear regression to identify the demographic, behavioral, and environmental factor(s) that best explained changes in these numbers and percentages. The model results show that in a well-controlled environment, most armatures used as projectile weapons are lost or discarded at hunting sites; only ∼4.5% of used armatures (or ∼2 armatures per year of simulation) are discarded in habitation camps where they would likely be excavated. These findings suggest that even rare hafted armatures found in the Early and Middle Stone Age could indicate a well-established use of such tools. Our model shows that interactions between reoccupation of archaeological sites, population size, rate of hunting with projectile weapons, and tool curation levels strongly influence the count of lithic armatures found in archaeological assemblages. Therefore, we argue that fluctuations in the counts of armatures documented at archaeological sites should be evaluated within their demographic and environmental contexts to better understand if they reflect spatiotemporal changes in hunting behavior.

Introduction

The development of projectile technology is integral to research on human origins for several reasons. Technological changes and the development of complex technologies are some of the earliest archaeologically visible markers of early hominins’ cognitive evolution. The invention of projectile technology, either simple or complex (as defined by Shea and Sisk, 2010; Sisk and Shea, 2010, 2011), had important ramifications for Pleistocene hominin evolution. Projectile technology improved hunting success, expanded prehistoric hominins’ ecological niche (Sisk and Shea, 2011) and may have played a part in the emergence of important cognitive achievements such as foresight and planning (Thieme, 1997). Therefore, better understanding the archaeological record of hafted armatures can provide insight into our species’ broad evolutionary adaptations.

The number of stone points—which remain the best evidence of projectile technology—fluctuates highly across archaeological assemblages. Retouched Still Bay points are rare or even absent in certain Middle Stone Age (MSA) archaeological sites such as Pinnacle Point 5–6 (Wilkins et al., 2017) and Klasies River Mouth (Singer and Wymer, 1982), but they are found in relatively high numbers at Apollo 11 Rock Shelter (n = 25; Lombard and Högberg, 2018), Umhlatuzana Rock Shelter (n = 102; Kaplan, 1990; Högberg and Lombard, 2016), and Blombos Cave (n = 371; Villa et al., 2009). Similarly, the number of unretouched points also fluctuates among MSA archaeological sites (e.g., n = 212 at Holley Shelter, Bader et al., 2015; n = 380 at Pinnacle Point 13B, Thompson et al., 2010; and n > 3000 at Klasies River Mouth, Singer and Wymer, 1982). Although it can be tempting to argue that those numbers indicate different frequencies of projectile weapons usage, recent research has shown that high numbers may instead result from widespread misclassifications of the function of the points, based solely on morphology (Schoville, 2010; Bouzouggar and Barton, 2012; Tomasso and Rots, 2018). Given the importance of quantifying the use of hafted armatures to understand human evolution, we use an agent-based model (ABM) to present an alternative perspective on this topic. Our ABM explores the impact of human and environmental factors—hunting behavior, tool curation, population size, and biome size and productivity—on the count and spatial distribution of simple projectile armatures found in archaeological sites.

Archaeological ABMs are becoming robust middle-range research tools (as advocated by Binford, 1981) because they allow us to connect the archaeological record to the behaviors that could have created that record (see, for example, Mithen, 1990; Lake, 2000; Brantingham, 2003; Barton and Riel-Salvatore, 2014; Davies et al., 2016; Oestmo et al., 2016; Coco et al., 2020). This is important because the archaeological record is not made of behavior but rather of discarded materials. In addition, human behavior is spatially continuous (Foley, 1981), but the location of archaeological sites is constrained by taphonomy and visibility, making the traces of such behavior spatially imbalanced (Dunnell, 1992).

Inferring behavior from the patchy archaeological record using traditional archaeological methods relies on assumptions and interpretations that are prone to biases. Here, we use an ABM to explore the link between behavior and the archaeological record in a controlled setting, where our assumptions are clearly explained and where the artifact record is not affected by taphonomy or restricted sampling. We take advantage of the capacity of ABMs to create complex, data-informed, realistic models to test hypotheses (Lake, 2014) in ways that are similar to the Artificial Anasazi (Axtell et al., 2002) or the MedLand project (Barton et al., 2010). This approach enables us to simulate hunter-gatherers’ daily subsistence practices with controlled variables, probabilities, and constraints that can be altered to explore the effects of different interactions between humans and the environment. Through this, we can explore the complex patterns that occur when human decisions affect common resources, which cannot easily be modeled using simple mathematical equations (Janssen and Hill, 2014). Furthermore, this approach allows us to visualize the outcome of our simulations on a reconstructed paleo-landscape and uncover spatial patterns that may be archaeologically invisible due to the patchy nature of the MSA archaeological record. However, we must clarify that our goal is not to reproduce lithic assemblage composition across South African MSA sites but rather to explore the impact of different human and environmental factors on general counts of simple projectile armatures.

Simple projectile technology—projectile weapons that use human energy for propulsion (Shea and Sisk, 2010)—was found in both Europe and Africa during the late Lower Paleolithic and Earlier Stone Age (ESA; Thieme, 1997; Villa and Lenoir, 2009; Wilkins et al., 2012; Sahle et al., 2013). To date, the ∼500 ka points used for hafted trusting spears from Kathu Pan, South Africa, constitute the earliest evidence of such simple projectiles (Wilkins et al., 2012; but see Rots and Plisson, 2014), whereas ∼400 and ∼300 ka wooden spears and sticks have been found in Mousterian assemblages at Clacton-on-Sea, the UK, and Schöningen, Germany, respectively (Thieme, 1997; Milks et al., 2019; Conard et al., 2020). Given that simple projectile technology has been documented in both modern humans and Neandertals archaeological contexts, it may have been invented by their common ancestor (Wilkins et al., 2012), thus pushing the timing of its conception to before the appearance of major ‘behavioral modernity’ traits. Therefore, as the invention of projectile technology likely had important impacts on ESA hominins’ hunting success, meat consumption, and thus, social behavior, it is not surprising that studying projectile tools and their evolution is a major part of paleoanthropology. Unfortunately, documenting the frequency and changes in lithic armatures is complicated by methodological and taphonomic constraints, which are discussed below.

Many MSA cave and rock shelter sites report numbers of various types of points (e.g., unretouched, retouched, and tanged [points with basal retouch]). However, as lithic coding systems vary between regions, sites, and even researchers (Will et al., 2019), it is often difficult to compare assemblages. This is best illustrated by Högberg and Lombard’s (2016) restudy of the Umlhatuzana lithic assemblage previously published by Kaplan (1990), where they state: “The numbers of points in our study and in Kaplan’s vary because our definition of a point differs from his” (Högberg and Lombard, 2016: p. 5). Differing excavation protocols further complicate the issue, as the extent and quality of excavations have a strong impact on artifact counts. Therefore, it is problematic to compare point counts of sites that had widely different quantities of sediment removed or were excavated using widely different plotting size thresholds.

In addition to the challenges faced when trying to compile and compare point counts over time and space, there is uncertainty about the function of most reported points because use-wear and residue studies—which would help define tool function—rarely accompany site reports. This is important because recent research shows that points previously thought to be used for projectile weapons might, in fact, have been handheld, e.g., the tanged Aterian tools of North Africa primarily thought to be hafted projectile points (Clark, 1970) are now assumed to have been handheld hafted scrapers (Iovita, 2011; Bouzouggar and Barton, 2012; Tomasso and Rots, 2018). Similarly, Still Bay points appear to have been used for both knives and projectile weapons (Lombard, 2006, 2007; Villa et al., 2009; Högberg and Larsson, 2011; Soriano et al., 2015), and experimental edge damage studies have shown that most MSA unretouched points found at Pinnacle Point 13B (PP13B), South Africa, were used as knives rather than projectiles (Schoville, 2010; Schoville et al., 2016). These studies show that single tool shapes likely had more than one use, making tool classification difficult. This suggests that we should not rely solely on tool morphology to document the evolution of projectile technology (Douze et al., 2020). Because of these challenges, it is difficult to evaluate the timing and intensity at which projectile weapons were used in the past. In turn, this leads us to wonder: do the fluctuations in ‘point’ numbers reported in published articles reflect fluctuations in the usage of projectiles?

Here, we use an ABM to explore this question. We test two hypotheses: H1: there are high counts of armatures in archaeological sites because most of the armatures used for hunting are discarded at archaeologically visible habitation sites, and H2: the counts of armatures found at MSA archaeological sites reflect the frequency with which projectiles were used for hunting. The alternative hypothesis for H2 is that other factors such as hunting behavior, tool curation, population size, and biome size and productivity have an impact on those counts. We predict that the results of the ABM will reject both null hypotheses for two reasons. First, recent research has shown that points found in archaeological sites were more often used as handheld or for multipurpose activities (Lombard, 2006; Schoville, 2010; Högberg and Larsson, 2011; Bouzouggar and Barton, 2012; Tomasso and Rots, 2018). Second, ethnographic research shows that human behavior is spatially continuous (Foley, 1981; Dunnell, 1992); therefore, the number of armatures used primarily for hunting and discarded at habitation sites should correlate with environmental productivity and hunting behavior.

Section snippets

Materials and methods

This research focuses on the results of a discard module added to the PaleoscapeABM, which is a component of the Paleoscape model explained in Marean et al. (2015) and illustrated in a recent special issue of Quaternary Science Review (e.g., Cleghorn et al., 2020; Cowling et al., 2020; Kraaij et al., 2020; Marean et al., 2020; Wren et al., 2020). The different models i.e., Paleoscape, PaleoscapeABM, and projectile submodel, are explained below.

Summary statistics of projectiles discarded at habitation camps

The results of the simulations show that, on average, 4.5% of armatures used in a simulation are brought back to habitation camps while the remaining 95.5% are lost or discarded where the hunt took place. As each simulation runs for one year and produces an average of ∼40 armatures, this suggests that ∼2 armatures are discarded at habitation camps every year. However, as simulations produce an average of ∼205 sites per year—either multicomponent or used only as habitation camps—most of those

Low counts of projectile armatures in archaeological sites are expected

Our modeling results show that low counts of hafted armatures in archaeological sites should be viewed as normal if prehistoric hunter-gatherers followed the OFT principles modeled here. The model results show that most projectile armatures end up in locations with low archaeological visibility (either lost at hunting sites or discarded at rarely reoccupied habitation sites). For the MSA archaeological record, this suggests that low counts could still be interpreted as a sign that projectile

Conclusions

This paper presents an ABM that focuses on the connection between past behavior and its archaeological traces. Our modeling results show that hafted armatures were most probably used regularly during the MSA, despite their low counts in archaeological sites. In fact, this paper shows that the equifinality created by the interaction of multiple socio-behavioral and ecological factors suggests that hunting tool frequency at habitation sites may not directly reflect the rate at which hunting took

Declaration of competing interest

The authors declare there is no conflict of interest.

Acknowledgments

The authors recognize the support of a grant from the National Science Foundation (BCS-1138073), Hyde Family Foundations, the Institute of Human Origins at Arizona State University, and the John Templeton Foundation to the Institute of Human Origins at Arizona State University. The opinions expressed in this publication are those of the authors and do not necessarily reflect the views of any of these funding organizations. Sincere thanks to Erik Otárola-Castillo (Department of Anthropology,

References (108)

  • F.A. Engelbrecht et al.

    Downscaling Last Glacial Maximum climate over southern Africa

    Quat. Sci. Rev.

    (2019)
  • V. Grimm et al.

    The ODD protocol: a review and first update

    Ecol. Model.

    (2010)
  • K. Hawkes

    Showing off: tests of an hypothesis about men’s foraging goals

    Ethol. Sociobiol.

    (1991)
  • A. Högberg et al.

    Lithic technology and behavioural modernity: New results from the Still Bay site, Hollow Rock Shelter, Western Cape Province, South Africa

    J. Hum. Evol.

    (2011)
  • Z. Jacobs

    An OSL chronology for the sedimentary deposits from Pinnacle Point Cave 13B—a punctuated presence

    J. Hum. Evol.

    (2010)
  • T. Kraaij et al.

    A fiery past: A comparison of glacial and contemporary fire regimes on the Palaeo-Agulhas Plain, Cape Floristic Region

    Quat. Sci. Rev.

    (2020)
  • C.W. Marean

    Pinnacle Point Cave 13B (Western Cape Province, South Africa) in context: The Cape Floral Kingdom, Shellfish, and modern human origins

    J. Hum. Evol.

    (2010)
  • C.W. Marean et al.

    The stratigraphy of the Middle Stone Age sediments at Pinnacle Point Cave 13B (Mossel Bay, Western Cape Province, South Africa)

    J. Hum. Evol.

    (2010)
  • C.W. Marean et al.

    The Palaeo-Agulhas Plain: Temporal and spatial variation in an extraordinary extinct ecosystem of the Pleistocene of the Cape Floristic Region

    Quat. Sci. Rev.

    (2020)
  • C.M. Monahan

    The Hadza carcass transport debate revisited and its archaeological implications

    J. Archaeol. Sci.

    (1998)
  • S. Oestmo et al.

    Applying Brantingham’s neutral model of stone raw material procurement to the Pinnacle Point Middle Stone Age record, Western Cape, South Africa

    Quat. Sci. Rev.

    (2020)
  • G. Porraz et al.

    Technological successions in the Middle Stone Age sequence of Diepkloof Rock Shelter, Western Cape, South Africa

    J. Archaeol. Sci.

    (2013)
  • A.G. Rebelo et al.

    Plant communities of the Riversdale Plain

    S. Afr. J. Bot.

    (1991)
  • V. Rots et al.

    Projectiles and the abuse of the use-wear method in a search for impact

    J. Archaeol. Sci.

    (2014)
  • B.J. Schoville

    Frequency and distribution of edge damage on Middle Stone Age lithic points, Pinnacle Point 13B, South Africa

    J. Hum. Evol.

    (2010)
  • B.J. Schoville et al.

    A model of hunter-gatherer skeletal element transport: The effect of prey body size, carriers, and distance

    J. Hum. Evol.

    (2014)
  • E. Singels et al.

    Foraging potential of underground storage organ plants in the southern Cape, South Africa

    J. Hum. Evol.

    (2016)
  • E. Thompson et al.

    Middle and Late Pleistocene Middle Stone Age Lithic Technology from Pinnacle Point 13B (Mossel Bay, Western Cape Province, South Africa)

    J. Hum. Evol.

    (2010)
  • J.A. Venter et al.

    Large mammals of the Palaeo-Agulhas Plain showed resilience to extreme climate change but vulnerability to modern human impacts

    Quat. Sci. Rev.

    (2020)
  • P. Villa et al.

    The Still Bay points of Blombos Cave (South Africa)

    J. Archaeol. Sci.

    (2009)
  • L. Wadley

    Announcing a Still Bay industry at Sibudu Cave, South Africa

    J. Hum. Evol.

    (2007)
  • C.J.H. Ames et al.

    Evaluating erosional impacts on open-air archaeological sites along the Doring River, South Africa: Methods and implications for research prioritization

    Archaeol. Anthropol. Sci.

    (2020)
  • C.J.H. Ames et al.

    A Multi-user mobile GIS solution for documenting large surface scatters: An example from the Doring River, South Africa

    J. Field Archaeol.

    (2020)
  • J.B. Arnold

    ggthemes: Extra Themes, Scales and Geoms for “ggplot2.”

    (2019)
  • R.L. Axtell et al.

    Population growth and collapse in a multiagent model of the Kayenta Anasazi in Long House Valley

    Proc. Natl. Acad. Sci. USA

    (2002)
  • G.D. Bader et al.

    The lithic technology of Holley Shelter, Kwazulu-Natal, and its place within the MSA of southern Africa

    S. Afr. Archaeol. Bull.

    (2015)
  • C.M. Barton et al.

    Land-use, water and Mediterranean landscapes: Modeling long-term dynamics of complex socioecological systems

    Philos. Trans. R. Soc. London, Ser. A

    (2010)
  • L.E. Bartram

    An Ethnoarchaeological Analysis of Kua San (Botswana) Bone Food Refuse

    (1993)
  • L.R. Binford

    Bones: Ancient Men and Modern Myths

    (1981)
  • M.S. Botha

    A Gatherer’s Paradise? Early Humans and Plant Foraging in the Cape South Coast, South Africa

    (2018)
  • A. Bouzouggar et al.

    The identity and timing of the Aterian in Morocco

  • P.J. Brantingham

    A neutral model of stone raw material procurement

    Am. Antiq.

    (2003)
  • J.D. Clark

    The Prehistory of Africa

    (1970)
  • E. Coco et al.

    The effects of secondary recycling on the technological character of lithic assemblages

    J. Paleolit. Archaeol.

    (2020)
  • N.J. Conard et al.

    A 300,000-year-old throwing stick from Schöningen, northern Germany, documents the evolution of human hunting

    Nat. Ecol. Evol.

    (2020)
  • B. Davies et al.

    Modelling the palimpsest: An exploratory agent-based model of surface archaeological deposit formation in a fluvial arid Australian landscape

    Holocene

    (2016)
  • J.C. De Vynck et al.

    Seasonal availability of edible underground and aboveground carbohydrate resources to human foragers on the Cape south coast, South Africa

    PeerJ

    (2016)
  • K. Douze et al.

    Technology and function of Middle Stone Age Points. Insights from a combined approach at Bushman Rock Shelter, South Africa

  • R.C. Dunnell

    The Notion Site

  • A. Fischer et al.

    Macro and micro wear traces on lithic projectile points: experimental results and prehistoric examples

    J. Dan. Archaeol.

    (1984)
  • Cited by (8)

    • Were prehistoric coastal sites more intensively occupied than inland sites? Using an agent-based model to understand the intensity of prehistoric coastal occupation, and what it means for studies on the evolution of the coastal adaptation

      2022, Quaternary International
      Citation Excerpt :

      The estimates on shellfish and plant foraging returns are from modern foraging experimental studies (De Vynck et al., 2016, 2020; Singels et al., 2016; Botha et al., 2020). The foraging behavior aspect of the model is based on Optimal Foraging Theory (OFT) principles (see Gravel-Miguel et al., 2021; Janssen and Hill, 2014; Wren et al., 2020, In press for more details). In the model, forager groups are modeled as band agents that use an algorithm to make mobility decisions to set camp in specific locations based on expected foraging returns.

    • Holding your shape: Controlled tip fracture experiments on cast porcelain points

      2022, Journal of Archaeological Science: Reports
      Citation Excerpt :

      It is likely then that spin-off fractures on triangular unifacial points > 10 mm are caused through low velocity conditions with a kinetic energy range conducive of thrusting and throwing spears. In an archaeological context, the rate of discard for projectile and thrusted weaponry may be affected by factors such as the repairability, point size, raw material type and abundance, and frequency of use or usability (Gravel-Miguel et al., 2021). Consequently, the use life of a stone projectile or weapon tip is a complicated process and reflects the demands of the physical, social and economic environment.

    View all citing articles on Scopus
    View full text