Current Anthropology Volume 60, Number 1, February 2019 000

Origins of the Human Predatory Pattern

The Transition to Large-Animal Exploitation by Early Hominins

by Jessica C. Thompson, Susana Carvalho, Curtis W. Marean,

and Zeresenay Alemseged

Online enhancements: supplemental appendix

The habitual consumption of large-animal resources (e.g., similar sized or larger than the consumer) separates human
and nonhuman primate behavior. Flaked stone tool use, another important hominin behavior, is often portrayed as
being functionally related to this by the necessity of a sharp edge for cutting animal tissue. However, most research on
both issues emphasizes sites that postdate ca. 2.0 million years ago. This paper critically examines the theoretical
significance of the earlier origins of these two behaviors, their proposed interrelationship, and the nature of the em-
pirical record. We argue that concepts of meat-eating and tool use are too loosely defined: outside-bone nutrients (e.g.,
meat) and inside-bone nutrients (e.g., marrow and brains) have different macronutrient characteristics (protein vs. fat),
mechanical requirements for access (cutting vs. percussion), search, handling and competitive costs, encounter rates,
and net returns. Thus, they would have demanded distinct technological and behavioral solutions. We propose that the
regular exploitation of large-animal resources—the “human predatory pattern”—began with an emphasis on percussion-
based scavenging of inside-bone nutrients, independent of the emergence of flaked stone tool use. This leads to a series of
empirical test implications that differ from previous “meat-eating” origins scenarios.

Many primates consume animal resources, but humans are ford and Bunn 2001). Large animals represent concentrated
the only living primates that regularly exploit animals the packages of easily digestible and calorically rich fat and protein.
same body size or larger than themselves (Butynski 1982).1 Thus, they are often described as part of the high-quality diet
Although this behavior may have been present in some extinct necessary for major biological changes such as the larger brains
hominins, we term this the “human predatory pattern” (HPP). and bodies of some early Pleistocene hominins (Antón, Potts,
This specifically separates the behavior from that of other and Aiello 2014). However, even scavenging meat from large-
living primates while remaining neutral about when and how it animal carcasses comes with significant costs in terms of predator
began and how such resources were actually acquired. Within exposure, bacteria load, and chewing energetics (Smith et al.
paleoanthropology, the exploitation of larger animals has tra- 2015; Treves and Naughton-Treves 1999; Zink and Lieberman
ditionally fallen under the umbrella term “meat-eating” (Stan- 2016). Dietary quality can be enhanced through consumption of
alternative fat and protein sources (e.g., insects and small prey)
or through premasticatory processing of foods that may or may
Jessica C. Thompson is Assistant Professor in the Department of not be animal derived (Carmody and Wrangham 2009; Hardy
Anthropology at Yale University (10 Sachem Street, New Haven, Con- et al. 2015; Rothman et al. 2014). Thus, there is a need to crit-
necticut, 06511, USA [[email protected]]). Susana Carvalho is ically examine the concept of meat-eating as it has been used in
Assistant Professor in the Institute of Cognitive and Evolutionary An-
paleoanthropology and, specifically, to understand the evolu-
thropology at the University of Oxford (64 Banbury Road, Oxford OX2
tionary context of the emergence of the HPP as a behavior
6PN, United Kingdom), and Director of Paleontology and Primatology
at Gorongosa Restoration Project, Gorongosa National Park, Mozam- unique to the hominin lineage.
bique. Curtis W. Marean is Professor in the School of Human Evolu-
tion and Social Change and Institute of Human Origins of the Arizona Background
State University (PO Box 874101, Tempe, Arizona 85287-4101, USA),
and Honorary Professor at the Centre for Coastal Palaeoscience of the For Dart (1953), the violent act of prey acquisition made meat-
Nelson Mandela Metropolitan University (Port Elizabeth, Eastern Cape eating a fundamental milestone in human evolution. He argued
6031, South Africa). Zeresenay Alemseged is Professor in the Depart-
ment of Organismal Biology and Anatomy of the University of Chicago 1. Because this paper draws on many different bodies of research, in-
(1027 East 57th Street, Chicago, Illinois 60637, USA). This paper was text citations have been limited to the most relevant and/or recent
submitted 27 IX 16, accepted 4 X 17, and electronically published 5 works. For a more comprehensive citation list, refer to the online ap-
II 19. pendix.

q 2019 by The Wenner-Gren Foundation for Anthropological Research. All rights reserved. 0011-3204/2019/6001-00XX$10.00. DOI: 10.1086/701477

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 000 Current Anthropology Volume 60, Number 1, February 2019

via his osteodontokeratic hypothesis that australopith carnivory stone tool use through analogy, inferring that “it is likely that
in South Africa marked the “predatory transition from ape to they acquired prey in much the same manner as that employed
man.” Subsequent taphonomic work by Brain (1981) revealed by modern chimpanzees, i.e., capturing it by hand on the ground
that accumulations of broken animal bones in South African or in trees” (145).
caves did not represent the remains of australopith meals. The use of these chimpanzee models makes several assump-
Concepts of “humanness” and “meat-eating” continued their tions about early hominin diet: (1) there is a long history of
entanglement, however, with the naming of Homo habilis (the animal resource exploitation and dietary flexibility, (2) there was
“handyman”), from East Africa, in 1964. Unlike Australopithe- a focus on small prey before moving to larger prey, (3) most prey
cus, this hominin was associated with both stone tools and fossil was cooperatively hunted, and (4) animal resources formed a
animal bones, and the tool-use/meat-eating package became small but persistent dietary element. The implication from this
inherent parts of the definition of our genus Homo. model is that the transition to the HPP occurred through ex-
As in South Africa, critical assessment of these fossil and pansion of existing omnivory to also include larger prey, drawing
artifact accumulations resulted in the understanding that these on existing cooperative hunting behavior and augmented by new
sites were not the sole residues of hominin activities (Binford emphasis on tool use. Regardless of inferred mode of acquisition
1981). This research led to a voluminous and ongoing debate (e.g., hunting, passive scavenging, or confrontational scaveng-
about the mode of early hominin carcass acquisition, and ing), this work is always undertaken with the assumption that
whether it was predominately through hunting or scavenging large-animal resources offered benefits that outweighed their
(Domínguez-Rodrigo et al. 2014; Pante et al. 2015). This debate costs to later Homo but not to earlier hominins such as Austra-
has centered on localities in Olduvai Gorge, Tanzania, and other lopithecus (Stanford and Bunn 2001). Here, we critique the model
sites dating maximally to ca. 2.0 million years ago (2.0 Ma; under which large-animal exploitation emerges as an extension
Domínguez-Rodrigo, Barba, and Egeland 2007), and follows a of cooperative hunting of small prey and suggest an alternative
general pattern in the literature in which meat-eating has been hypothesis for the initial transition to the HPP. We argue that
given most attention in the context of the behavioral and bi- Pliocene hominins began their trajectory toward the HPP by
ological evolution of Homo (Bunn 2007). exploiting mainly the inside-bone nutrients of large-animal car-
Many of the presumed defining characteristics of later Homo casses using percussion technology—but not necessarily flaked
are explicable in terms of energetic trade-offs involving dietary stone tools.
change. These include dentognathic transformations, encepha- This paper has three goals. We examine the Pliocene hominin
lization, enhanced cooperation with nonkin, and life histories record beyond the earliest date for Homo at 2.8 Ma, where
that include prolonged childhood, female postreproductive pe- hominins such as Australopithecus exhibit a brain size that is
riods, and shorter interbirth intervals (Antón et al. 2014). The ∼30% larger than a chimpanzee of comparable body size, canine
energetic cost of these changes would have been paid through size reduction, a significant degree of terrestriality, and a hand
consumption of higher quality foods, new processing methods that shows adaptations for tool use (Kimbel and Villmoare
for existing foods, food sharing, or other partnerships that 2016). We also reject further use of the term meat-eating, as it
reduced individual energetic costs (Gettler 2010). By 1.5 Ma, prioritizes outside-bone nutrients (meat) at the expense of
many of these characteristics had coalesced into a successful inside-bone nutrients (e.g., the fat in brains and marrow). These
adaptive suite that would continue into later hominin evolution, have different macronutrient characteristics; mechanical re-
arguably with roots fueled by earlier dietary change. quirements for access, search, competition, and handling costs;
However, understanding the origins of meat-eating is a sep- encounter rates; and net returns that would have demanded
arate issue from its later evolutionary ramifications. Less effort distinct technological and behavioral solutions. Similarly, al-
has been put to exploring the initial transition to obligate om- though small mammals and reptiles, birds, fish, and insects do
nivory, and most has relied on models derived from living ape provide proteins and fats, and are therefore often treated under
ecology, especially chimpanzees (Stanford 1996). Domínguez- the rubric of “animal resources” (Lupo 2012), they also differ
Rodrigo and Pickering (2017) argue that “it is most parsi- from large animals in their foraging costs. Therefore, here we
monious to consider this shared behavioural module as a investigate the origins of the HPP specifically, rather than
synapomorphy, rather than a homoplasy, of these sister taxa” meat-eating broadly construed.
(11), but they do not differentiate between body sizes or nutri-
tional components (e.g., meat vs. fat) of prey. This approach has Ecology of Ape Omnivory and Use
been applied as far back in time as Ardipithecus, ca. 4.4 Ma, by
of the Chimpanzee Model
interpreting them as herbivores that opportunistically ate meat
(Stanford 2012). Tappen (2001) also suggests that “early austra- Both of our closest living relatives—chimpanzees (Pan troglo-
lopithecines may be expected to eat meat at least to the degree dytes) and bonobos (Pan paniscus)—hunt, but never for prey
that living chimpanzees do (by argument from phylogeny and larger than themselves. Small ungulates such as duiker (Tribe
parsimony)” (14). Because of the lack of flaked stone tools from Cephalophini) are one of the most commonly taken prey by
this time period, Stanford (2012) decouples meat-eating from bonobos (Surbeck and Hohmann 2008), whereas small-bodied

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
Thompson et al. Origins of Large-Animal Exploitation 000

primates represent the majority of vertebrate prey for chim- sources may be homologous, there is little phylogenetic basis
panzees (Watts and Mitani 2002). This shared pursuit of animal for considering that the specific form of prey acquisition and
resources has led to the inference that a basic need or desire for processing observed in modern chimpanzees is a suitable model
them should have also been present in our last common an- for early hominins. It is equally plausible that this behavior is a
cestor and persisted throughout subsequent hominin evolution derived trait for chimpanzees, which is a common problem with
(McGrew 2010). The reconstruction of early hominins taking ape analogies (Sayers and Lovejoy 2014). The key inference is
small prey and later transitioning to larger prey (or scavengable that a long precedent of cooperative hunting of small prey as a
carcasses) also stems from analogy to living apes (Stanford 2001). main mode of acquisition was unlikely to have been a neces-
However, their cooperative hunts rely on biological adaptations sary or even likely preadaptation for hominins to begin ex-
for moving quickly through tree canopies or (less frequently) on ploiting large-bodied animals.
the ground and using sharp canines to dispatch prey (Stanford
1995).
For both bonobos and chimpanzees, higher-ranking indi- Behavioral Ecological Approaches
viduals more frequently consume animal protein—even though
to Understanding the HPP
it does not provide a large caloric contribution (Tennie, O’Malley,
and Gilby 2014). In spite of the attraction of animal resources, The principles of optimal foraging theory provide a useful
chimpanzees are reluctant to scavenge (Watts 2008), and they framework for examining the subsistence decisions of early
do not take some low-cost hunting opportunities such as mon- hominins. Specifically, the diet breadth model allows for the
itor lizards (Varanus spp.; McGrew 2015). This indicates that development of predictions about the conditions under which
hunting of some small- and medium-bodied prey does not a novel resource (e.g., large animals) would have been added to
automatically translate into a universal motivation to exploit the existing dietary repertoire. Food items are ranked on their
any available animals. net return, enabling prediction of an optimal diet breadth (list
Although small prey may be less dangerous than large prey, of prey that are always pursued upon encounter). The currency
many are fast moving, and thus energetically costly in the ab- of choice is normally calories, although macro- and micronu-
sence of biological adaptations for pursuit and capture. In the trient composition can also influence prey ranking (Hill 1988).
case of forest-dwelling animals, this often also requires a high Costs are typically divided into search and handling costs
degree of arboreality. Size of prey is an important correlate for (Codding and Bird 2015). The highest-ranking foods should
group size and cooperative hunting even within the same species always be taken on encounter, whereas lower-ranking foods may
of ape; chimpanzees in Fongoli hunt nocturnal primates with be bypassed, even if abundant, to continue pursuit of higher-
tools (small prey, no cooperative hunting), in Bossou they eat ranking items. When something changes about the ecology or
tree pangolin (Phataginus tricuspis; small prey, no cooperative abundance of the forager, the novel resource, or other higher-
hunting), and in Gombe they hunt colobus monkeys (Colobus ranked resources in the environment, then there may be expan-
and Procolobus), a larger prey for which cooperative hunting is sions or contractions in dietary breadth. Significant changes in
required. Such variability in response to local ecology might also technology or social behavior can change prey costs, resulting in
be inferred for early hominins, such that some small terrestrial a change in its rank and thus its pursuit probability upon en-
game may have been hunted in a manner analogous to apes. counter.
However, such resources differ in key ways from large animals, The origins of the HPP can be conceptualized in terms of the
and thus occasional omnivory of small terrestrial animals in costs and returns associated with prey and prey body parts, as
forested settings does not provide an obvious bridge to the ex- well as those resources that rank above them. Large-animal
ploitation of large animals in the mosaic habitats inhabited by resources contain energy-rich and nutrient-dense packages of
early hominins, particularly Australopithecus. macronutrients (protein and fat) and micronutrients (heme
For highly terrestrial modern hunter-gatherers, small prey iron, zinc, vitamin A, and some B vitamins) that are rare in
are rarely taken unless investment is made in specific and often most plant-based diets (Murphy and Allen 2003). Energy yields
elaborate technology to facilitate their capture (Ugan 2005). In from animal resources can be high, especially if they contain fat
addition to being difficult to acquire, small prey are not easily (Speth 2010). However, processing costs can also be high with-
shared. Smaller-bodied animals also lack the large fat reservoirs out biological adaptations or technological interventions (Zink
of brain and marrow present in larger animals (Speth 1989). and Lieberman 2016). Inside-bone nutrients such as brains and
Thus, increasing terrestriality without complex technology would marrow are the fattiest body parts on lean wild game, and the
have increased the cost of small, mobile prey relative to larger precursor to docosahexaenoic fatty acids (Langdon 2006) and
or less mobile options. oleic acids (Morin 2007) that play an important role in eye and
Although Domínguez-Rodrigo and Pickering (2017) sum- brain development.
marize examples in which apes hunt small prey that are not A group of medium-bodied hominins may have been ca-
fast moving, the majority of chimpanzee and bonobo hunts pable of dispatching an animal the same size or larger than
occur in this way. Although our shared desire for animal re- themselves. However, because cooperative hunting of large

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 000 Current Anthropology Volume 60, Number 1, February 2019

animals invokes a suite of assumptions about early hominin body size (Blumenschine et al. 1987). Using models derived from
social structure that are difficult to test, we focus on the costs observations of modern “Serengeti-like” ecosystems (both dry
and returns associated with the most basic mode of acquisi- and seasonal), scavenging opportunities are most abundant
tion: scavenging in a Pliocene context. These include poten- during the dry season. When carcass availability is low, these
tially high search and handling costs, the chance of bacterial kills are quickly stripped of meat (e.g., outside-bone nutrients)
infection from rotten meat, conspecific competition, and expo- in a predictable sequence (Blumenschine 1986).
sure to large-bodied carnivores. Reduction of one or more of In modern Serengeti-like ecosystems, leopard kills have been
these costs could explain expansion of the diet to include a found to provide some of the most persistent and predictable
novel resource. flesh-scavenging opportunities and are more readily encoun-
tered in riparian woodlands (Cavallo and Blumenschine 1989).
Such environments are also inhabited by crocodiles, which may
Changing Costs and Returns have presented unique scavenging opportunities in the form of
thrown-off limbs (Davidson and Solomon 1990). However, in
of Large-Animal Resources
wetter and more closed environments, such as at Parc National
Risk of exposure to predators likely increased as hominins des Virunga, (adjacent to the central African rain forest), wooded
began to exploit large animals in more open environments. areas provide less frequent and higher-risk scavenging oppor-
Tappen (2001) found that exposure risk during passive scav- tunities than more open areas (Tappen 2001). There, Tappen
enging differed substantially between environments; thus, the showed that passive scavenging of carcasses encountered dur-
decision to scavenge is best modeled along a continuum of ing other activities was more productive than active searching,
risk. New risk mitigation strategies would have already been nec- with most scavengable remains being inside-bone nutrients
essary as hominins became increasingly terrestrial after 4.2 Ma, from medium-sized carcasses that each provided a median of
with group size likely mediated by this risk (Markham et al. ca. 2,000 kcal. Pobiner (2015) further demonstrates the eco-
2015). In hominins, medium-sized multimale groups are im- logical contingency of scavenging opportunities, especially
plied from the fossil record where multiple individuals have relative to carnivore guild composition.
been found together (Johanson 2004), even though body mass Although scavenging niches have been given much attention
estimates show a high degree of body size dimorphism in for Pleistocene hominins (Blumenschine 1989; Domínguez-
Australopithecus (Grabowski et al. 2015). Rodrigo 2001), less work has explored the implications of car-
Large group sizes may mitigate risk of predation by carni- nivore paleo-guild composition prior to ca. 2.6 Ma. The period
vores but also increase competition at scavenging sites. Scav- before ∼3.9 Ma is especially poorly sampled, but patterns in
enging also carries health risks. Outside-bone nutrients acquire species richness in the east African carnivore guild show a peak
harmful quantities of bacteria within 24 hours of exposure at around 3.6 Ma, with the first appearance of several new
(Smith et al. 2015), receiving bacteria from carnivore mouths, species (Turner 1999). Starting ca. 2.0–1.5 Ma, there is a decline
insects, soil, scavenging birds, and fecal matter (Ragir, Rosen- in carnivore richness to the present, with specialists going ex-
berg, and Tierno 2000). Chimpanzees avoid carrion, apparently tinct in favor of more generalist taxa (Werdelin and Lewis 2005).
deterred by risk of infection (Watts 2008). Gut microbiome The late Pliocene included species that do not have modern
composition—an unknown in early hominins—is dependent ecological analogues, including giant hyenids (Werdelin 1999),
on diet and can alter an organism’s ability to cope with food- sabertooth felids (Marean and Ehrhardt 1995), and extinct
borne illness (Josephs-Spaulding, Beeler, and Singh 2016). Where crocodiles (Brochu and Storrs 2012).
flesh was encountered in an early stage of carcass access, a mixed This diversity of Pliocene carnivores points to more fre-
strategy of outside-bone and within-bone nutrients might have quent and varied scavenging opportunities, as well as greater
been facilitated by this. However, unbreached bone marrow danger from predators at fresh kills, than are seen in modern
retains low bacteria counts for much longer than exposed flesh African ecosystems (Blumenschine et al. 1987; Lewis 1997).
or exposed bone marrow (Smith et al. 2015). This undoubtedly Specifically, sabertooth felids were flesh specialists, and most
extended the amount of time it persisted in a fresh state, thus had adaptations to closed-environment ambush hunting of
increasing encounter rates relative to edible flesh and de- large-bodied animals that are also now extinct (Parkinson,
creasing risk from carnivores that remain near fresher kills. Plummer, and Hartstone-Rose 2015). Bone-cracking morphol-
In comparison to the outside-bone nutrients usually con- ogy evolved as a specialized adaptation for accessing inside-bone
sidered as the primary motivator for scavenging, inside-bone nutrients from large-bodied carcasses. In Pliocene Africa, this
nutrients from scavenged carcasses are more likely to be highly morphology first appears ca. 3.6 Ma (Turner 1999), suggesting
ranked because of their persistence, palatability, and high fat the expansion of scavenging opportunities focused on inside-
content. Most small-bodied carcasses are divested of meat and bone nutrients that remained after flesh was removed. Austra-
marrow within a day, but marrow and brains can persist for lopiths were one of many groups jostling for emerging niche
several days in medium to large carcasses. Four major factors space at this time as resources changed in their distributions and
influence carcass persistence and nutritional quality: (1) carni- open environments became more commonly exploited (Alem-
vore guild composition, (2) habitat type, (3) season, and (4) prey seged 2015). Through use of basic technology, Pliocene homi-

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
Thompson et al. Origins of Large-Animal Exploitation 000

nins could have rapidly taken advantage of the resources bone- and Boesch 1990). In captivity, chimpanzees taught to smash
cracking carnivores were evolving to exploit. open long bones will extract and eat it (Kitahara-Frisch, Nori-
koshi, and Hara 1987). Ungulate long bones filled with dried
A Percussive Adaptation fruit are also opened by bonobos that recognize there is a treat
inside (Roffman et al. 2015).
to Large-Animal Exploitation
Increased encounter rates with scavengable inside-bone re-
Although living apes are not direct analogues for early hom- sources can explain how early hominins came to recognize this
inins, primate archaeology can provide predictions about the as a potential food. In chimpanzees, increased environmental
foraging circumstances under which the intersection of meat- variability has been shown to link with novel behaviors (Hock-
eating and tool use should occur. Because stone is a raw material ings et al. 2015). However, in modern, heavily forested envi-
with a high likelihood of preservation, here we focus on use of ronments, encounter rates with within-bone nutrients are low,
stone tools. Where stone is available, tool use in primates often and movable stones are not often readily available (Mercader,
includes a percussive element, which appears to have evolved Panger, and Boesch 2002). Both apes and monkeys are heavily
convergently multiple times across primate lineages (Haslam constrained in their tool use by the distribution and quality of
et al. 2009). Chimpanzees, capuchin monkeys (Sapajus spp.), suitable raw materials (Carvalho et al. 2008; Visalberghi, Spag-
and long-tailed macaques (Macaca fascicularis aurea) carefully noletti, et al. 2009). Although it requires only simple technology,
select unmodified stones to smash open nuts or other tough inside-bone resource scavenging does demand that tools be
food packages (Carvalho et al. 2008; Gumert and Malaivijitnond carried in case of encounter or that tool locations be remem-
2009; Visalberghi, Addessi, et al. 2009). This demonstrates that bered and carcass portions transported to those locations.
the cognitive foundations and anatomical requirements for re- Chimpanzees conform their tool use behaviors as they move
peated, socially influenced percussive technology is deeply rooted into new social or environmental settings (Luncz, Wittig, and
in the primate lineage, mainly within the context of accessing Boesch 2015). They curate tools for later use (Mulcahy and
tough, high-value packages of proteins and lipids. Often, such Call 2006), and they return to the same tool composites (anvils
extractive foraging is also associated with larger brain sizes and hammers) many times. They remember the locations of
(DeCasien, Williams, and Higham 2017). tools such as nut-cracking stones but will transport stone only
Use of percussion in extractive foraging has been seen as a up to a few hundred meters (Mercader et al. 2002). Over time,
bridge to lithic reduction (Whiten 2015), and lithic production these stones could be transported several kilometers, providing
again as a necessary requisite for meat-eating (Zink and Lieber- the appearance of longer-term transport (Luncz et al. 2016)—
man 2016). However, the focus on outside-bone nutrients deflects but always remaining in proximity to the resource. Scavengable
from the potential for hominins to begin exploiting large-bodied carcasses are more widely distributed than nut-bearing trees.
animals in the absence of flaked stone. Flaked stone produc- Percussive behaviors also have the potential to facilitate the
tion is a case of secondary tool production; it requires at least production of “flakes,” as unintentional by-products of anvil and
one or two tools (a hammerstone and a nodule or an anvil and hammerstone use (Carvalho et al. 2009). When bearded capu-
a nodule) to create another tool (a flaked stone). The scenario chin monkeys (Sapajus libidinosus) smash stones for nondietary
in which flaked stones must predate large-animal exploitation purposes, they also ignore the resultant flakes (Proffitt et al.
is unnecessarily complex, demanding creation of a new tech- 2016). Early hominins would have been far less constrained in
nology (flaked stone) to be applied to an equally novel purpose their ability to locate and carry stones, as they ranged in different
(cutting meat). Fewer steps are invoked if hominins first began kinds of environments and had anatomical adaptations for more
exploiting large-animal resources through hammerstone percus- energy-efficient load carrying (Carvalho et al. 2012).
sion, which transfers an existing extractive technology from one
similarly packaged resource (nuts or other encased foods) to
Paleoecology and Pliocene Hominin Diet
another (bones). Chimpanzee use of stick tools follow this pat-
tern, with tools originally used for one task repurposed to an- Reconstructions of australopith paleoenvironmental settings
other (Wilfried and Yamagiwa 2014). show an increase in the use of open-habitat environments over
The appeal of bone marrow is common to both humans and the course of the Plio-Pleistocene (Behrensmeyer and Reed
our closest relatives. Bone marrow is dense in both calories and 2013). At this time, climate also became more intensely variable,
nutrients but poses similar problems to other encased foods: creating larger regions of rapidly changing mosaic environ-
seasonal fluctuation (because marrow quality and fat content ments with multiple ecotones (Potts 2013). Current evidence
declines significantly when an animal is stressed; Blumenschine also suggests that a shift in diet occurred in the hominin lineage
and Madrigal 1993), patchy distribution, and a hard outer shell. by about 3.76 Ma (Lee-Thorp et al. 2010; Sponheimer et al. 2013).
Among chimpanzees, unpredictability in space adds value to At this time, hominins began to more intensively exploit C4
some resources upon encounter (Carvalho et al. 2012). The resources in the relatively open environments that had been ex-
marrow of colobus monkeys and other prey is regularly eaten by panding across Africa since the Miocene. Australopithecus afa-
wild chimpanzees by chewing off the extremities of long bones rensis in Ethiopia and Kenyanthropus platyops in Kenya were
and sucking on or extracting the marrow with a probe (Boesch the first hominins known to expand their dietary ranges away

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 000 Current Anthropology Volume 60, Number 1, February 2019

from that of extant apes and earlier hominins (Cerling et al. The problem is compounded because cut marks are prone to
2013; Wynn et al. 2013). By ca. 3.5 Ma, with later australopiths, equifinality, where similar-appearing linear marks can also be
brain size had undergone an increase proportional to the in- caused by other agents such as trampling (Domínguez-Rodrigo
crease it would later see with Homo (Kimbel and Villmoare et al. 2009). This must be considered as a potential problem at
2016), and canine size had already long been significantly re- any site, even if there are associated stone tools. Another prob-
duced in both males and females. Hand proportions were also lem is that many different butchery processes can create cut
more humanlike than apelike, and adaptations for terrestriality marks that have a wide range of forms (Domínguez-Rodrigo
were present throughout the skeleton (Ward, Kimbel, and and Yravedra 2009). Mark distribution and form can relate to
Johanson 2011). Diversity in habitat preference characterized the kind of butchery (e.g., filleting, disarticulation, chopping,
these and other later australopiths, with habitat reconstructions slicing) but does not seem to relate to how much flesh was being
that range from relatively wet, closed woodlands to much more removed (Merritt 2015). Moreover, chopping at carcasses with
open conditions (Rowan and Reed 2015). a flaked stone or a stone with a naturally sharp edge would leave
Carbon isotopic evidence from Au. afarensis (Wynn et al. a cutting trace even if the intent and action are percussive. For
2013) and Au. africanus (van der Merwe et al. 2003) indicates this reason, cut marks can overlap in morphology and size with
increased use of open habitats but a diet highly variable be- marks more indicative of percussion (Blumenschine, Marean,
tween individuals. At least some individuals had diets largely and Capaldo 1996). Amorphous marks created by hammerstone
based on foods that use the C4 photosynthetic pathway (tropical percussion have not to date been subject to the same critique
grasses and sedges) or herbivores/insects that fed upon those about equifinality as have cut marks, but they have also not been
foods. In contrast, earlier hominins such as Au. anamensis or the given the same scrutiny.
even earlier Ardipithecus ramidus (White et al. 2009) included Marrow scavenging has been most intensively examined at
relatively few C4/CAM foods in their diet, showing stronger C3 Early Pleistocene sites, usually where flaked stone tools were
signals more consistent with feeding off parts of shrubs and trees widely employed and cut marks on fossils are common (Blu-
(Cerling et al. 2013). This provides further support for later menschine et al. 2012). Percussion that is not related to stone
australopiths as versatile omnivores that could exploit a range knapping has also been reported from these contexts, occur-
of resources across many habitats (Alemseged 2015). ring sometimes in volumes of stone material that far exceed
volumes reported for flaked stone artifacts (Mora and de la
Taphonomic and Archaeological Evidence Torre 2005). However, even at Oldowan sites with abundant
pounding artifacts, both lithic and faunal analyses do not sug-
for Emergence of the HPP
gest that marrow extraction was the main activity taking place—
As reviewed earlier, activities falling under the rubric of “butch- the percussive tools were being extensively used for other pro-
ery” compose two main—and fundamentally different—ac- cessing (Diez-Martín et al. 2009). However, work has only
tivities: cutting and percussion. Cutting behaviors focus on recently begun on how to identify and/or analyze prospective
processing of outside-bone nutrients such as meat, which are pounding tools from earlier deposits (Benito-Calvo et al. 2015;
low in fat, putrefy quickly, involve higher risk of carnivore pres- Caruana et al. 2014).
ence, and have low overall encounter rates because they occur Currently, the earliest reported butchery marks are from
on more complete carcasses. Percussion exploits inside-bone two ca. 3.4 Ma specimens recovered from the DIK-55 site at
nutrients that are high in fat, persistent, transportable, low in Dikika, in Ethiopia (McPherron et al. 2010). These marks have
bacteria loads, and less risky to exploit and that potentially been alternatively interpreted as resembling linear trampling
have higher encounter rates. These two activities also leave ta- marks (Domínguez-Rodrigo, Pickering, and Bunn 2012) or
phonomic signatures that, although variable, are collectively dif- crocodile tooth marks (Sahle, El Zaatari, and White 2017).
ferent from one another in archaeologically detectable ways: However, the DIK-55 marks are outliers in terms of both size and
the marks they leave on bones, skeletal element representation, shape when compared to a large sample of other marks on fossils
and the degree and nature of skeletal part fragmentation. from the same deposits and to experimentally produced trample
Cut marks are traces left on bone surfaces by a sharp-edged marks (Thompson et al. 2015). This ambiguity in effector may at
object. They can be created by a variety of raw materials but are least partially be because many of the DIK-55 marks are quite
most often considered the by-product of flaked stone tool use. deep and large, and in appearance they “transition from cut-
Flaked stone preserves well and provides contextual evidence marks to percussion marks in morphology” (McPherron et al.
that a locality was used by hominins, as well as a plausible ex- 2010:22 in supplementary information).
planation for how associated bones with linear marks became Although the DIK-55 specimens may represent early per-
modified (Domínguez-Rodrigo et al. 2005; Njau 2012). Because cussive butchery behavior, the next documented evidence de-
flaked stone is robust and readily recognizable, there has argu- rives from nine specimens (two of which conjoin) with only
ably been undue bias on flaked stone tools and cut marks rather cut damage reported from the 2.6 Ma site of Gona, Ethiopia
than percussive tools and percussion traces, which may be robust (Domínguez-Rodrigo et al. 2005). These are followed closely
but not as easily recognized, as the primary forms of evidence in time by three specimens, reported with both cut and per-
for identifying butchered bone localities. cussion damage, from the 2.5 Ma locality of Bouri in the Middle

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
Thompson et al. Origins of Large-Animal Exploitation 000

Awash deposits of Ethiopia (de Heinzelin et al. 1999). Cut- date hominins for the origins of both behaviors appear to
marked fossil bovid elements have also been reported from have been australopiths living at least 700,000 years earlier.
controversial 2.6 Ma deposits in the Masol region of the
Siwalik Range of India (Malassé et al. 2016). However, all these Percussion Scavenging
cases represent very small samples that are widely separated in
both time and space; systematic evidence of butchery does not In light of the theoretical and empirical evidence just reviewed,
become apparent in the archaeological record anywhere in the we argue that the transition to the HPP was not reliant on con-
world until after about 2.0 Ma, at localities such as Kanjera current flaked stone tool technology to cut meat but rather that
South in East Africa (Ferraro et al. 2013). percussive extraction of inside-bone nutrients offers a more
The timing of appearance and pattern of abundance of the parsimonious pathway by which hominins first made regular use
flaked stone record provides only circumstantial evidence that of large-animal resources (fig. 1). This is anticipated to leave a
it emerged in tandem with the HPP, based mainly on the fact different archaeological and taphonomic signature than outside-
that it follows a similar chronological and geographic pattern. bone nutrient extraction, thus rendering traditional emphasis on
Stone tools have been reported from the Lomekwi 3 (LOM-3) “cut marks” caused by meat processing less informative for
site in Kenya, dating to ca. 3.3 Ma (Harmand et al. 2015). These understanding how large-animal exploitation first emerged.
tools were flaked and heavily used as anvils, appearing to have Under the percussion scavenging model of the origins of the
been produced through percussive techniques distinct from HPP, both percussive extraction and flaked stone tool manu-
the production of later Oldowan tools—perhaps produced facture represent fundamentally different behaviors applied to
through a process similar to chimpanzee nut-cracking. Their substrates with different spatial distributions, characteristics,
function is unknown, but they could have been useful for both costs, benefits, and processing requirements. They need not
cutting and percussive activities. As with the DIK-55 fossils, have been behaviorally linked nor have emerged together.
this early date for flaked stone tool technology has been ques- Paleoanthropologists must resist being constrained by the
tioned (Domínguez-Rodrigo and Alcalá 2016). known archaeological record. For example, percussion scav-
By 2.6 Ma, again at Gona, hominins had begun to make enging does not require the use of any stone tools. Perishable
Oldowan tools (Rogers and Semaw 2009). These and other early materials such as tree boles and friable lateritic soil have been
Oldowan assemblages show complexity in understanding of used by chimpanzees to crack nuts, especially in areas poor in
fracture mechanics and preferred transport of specific raw ma- stone (Marchant and McGrew 2005; Mercader et al. 2002).
terials (Delagnes and Roche 2005; Stout et al. 2010), implying that Experimental work may also reveal that bone hammers are ef-
their predecessors had accumulated experience with stone artifact fective for breaking other bones, as later in time they became
manufacture that predated the Oldowan (Panger et al. 2002). important percussors for working stone (Rosell et al. 2011).
Simple flaked stone artifacts are also reported to date to ca. 2.6 Ma Within this theoretical framework, we make several predictions
at the Masol locality in India (Gaillard et al. 2016). However, like about the conditions under which the HPP emerged and what
evidence for the HPP, flaked stone tools remain uncommon until evidence should be associated with it.
ca. 2.0 Ma (Braun et al. 2010). In Africa after this time, they be-
come a frequent component of the archaeological landscape
Spatial Distribution and Frequency
(Plummer et al. 2009), where they were carefully selected and
carried up to several kilometers from their sources (Braun, Harris, Because of the patchy and unpredictable distribution of scav-
and Maina 2009). This inconsistent record may best be inter- engable inside-bone resources, their exploitation should have
preted as the product of occasional, rather than obligatory, stone been generally opportunistic, seasonal, and contingent on local
tool use that convergently emerged several times in many dif- ecology. Thus, evidence for the behavior should be overall rare
ferent places (Shea 2017a). but with concentrations near rivers, streams, and potentially
Until the recent work reviewed here, the earliest examples within areas that were within the home range of a resident
of the HPP and stone tool making were both known from the carnivore—especially felids. We predict the behavior to be-
2.6 Ma sites at Gona (Domínguez-Rodrigo et al. 2005). The come more common and widely distributed when hominins
prevailing view was that this time coincided with major speci- began to more regularly challenge large scavengers and pred-
ation events in the hominin lineage (Ambrose 2001), a position ators for fully fleshed carcasses. Although this prediction is not
reinforced by an example of early Homo from the nearby 2.3 Ma reliant on knowing the degree of risk involved in scavenging
deposits at Hadar, Ethiopia (Kimbel 2009). Now that the ear- within extinct Pliocene ecosystems, scavenging risk could be
liest Homo is reported at 2.8 Ma from Ledi-Geraru, Ethiopia quantitatively modeled to provide more precise predictions for
(Villmoare et al. 2015), the earliest evidence of butchery at when and where such a transition took place.
3.4 Ma from DIK-55, and the earliest flaked stone tools at
3.3 Ma from LOM-3, there has been a breakdown in this con-
Direct Evidence
venient and long-lived triumvirate: early Homo, the HPP, and
the first flaked stone tools do not all appear to have emerged Percussion scavenging should be evidenced primarily by per-
simultaneously at the start of the Pleistocene. Now, the candi- cussion marks and notches on highly fragmented long bone

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 000 Current Anthropology Volume 60, Number 1, February 2019

Figure 1. A, The former meat-eating scenario, in which hominins create accumulations of flaked stone tools at focal points on the
landscape, in association with the partially fleshed animal carcasses they are consuming. This places emphasis on consumption of
outside-bone nutrients, production of cut marks, and the formation of “sites” containing flaked stone. B, The percussion scavenging
scenario, in which hominins focus on percussion-based extraction of inside-bone nutrients from carcass portions that persist on the
landscape. This does not result in large accumulations of flaked stone and bone, and most damage signatures are percussion marks
and bone fragmentation.

shafts, mandibles (which also contain a marrow cavity), and Discussion

cranial fragments. These should be found in localities that of-
fered opportunities for scavenging and raw materials for per- Disparate lines of evidence have begun to increasingly coalesce
cussive implements, along with a high likelihood of rapid burial on a picture of the late Pliocene as a period marked by oppor-
and therefore good bone surface preservation (Thompson et al. tunistic and eclectic foraging by hominins that were manually
2015). Alternatively, accumulations of marked bones may occur dexterous and behaviorally flexible and that regularly exploited
together where they were transported to safer localities. Such a varied diet from a range of both open and closed habitats. Early
sites may or may not also have discarded hammerstone or anvil hominins such as Australopithecus had relatively large brains,
elements with macroscopic and microscopic evidence of bat- humanlike hand proportions, small canines, and a high degree
tering and other use wear (Benito-Calvo et al. 2015). of terrestriality. These features strongly suggest that extractive
foraging and tool use were already present and that these factors
were the substrate upon which the HPP was built. At some
Associated Evidence
point in evolutionary time, the benefits of large-animal ex-
If percussion tools were transported, then evidence for per- ploitation began to outweigh the costs. We propose that this
cussion scavenging is likely to also be found at “Type M” sites, was achieved in the late Pliocene largely through the appli-
where modified bones are found without association with cation of percussion to extract inside-bone nutrients. This has
flaked stone artifacts (Bunn 1994). Marked bone surfaces need implications for the reconstruction of early hominin paleo-
not be restricted to sites with nonperishable or recognizable biology and ecology, as well as an opportunity for a fresh look
percussors but should usually occur on or in close spatial at the behavioral implications of the beginnings of the HPP.
proximity to other indicators of percussion activity, such as Although research surrounding the HPP revolves almost
bone flakes and notched long bone shafts (Marean and Bertino exclusively around male strategies, this perspective suggests a
1994). stronger female role in at least the initial movement into a

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
Thompson et al. Origins of Large-Animal Exploitation 000

scavenging niche (Zihlman 2012). Among chimpanzees, fe- the process by which large-animal tissues first became incor-
males and juveniles are the subsets that most frequently use porated into the hominin diet, how important they were prior
tools, and most social transmission about tool use skills takes to the emergence of Oldowan technology, and how this be-
place between mother-offspring dyads (Lonsdorf 2006). Active havior should have been manifested in recognizable ways on
hunting is normally a male activity (Fahy et al. 2013; Stanford Pliocene paleolandscapes.
1995), which shares a tempting parallel with modern hunter- The likelihood of discovering relevant sites has arguably
gatherer social organization. However, female and juvenile been impacted by a lack of systematic survey in Pliocene depos-
chimpanzees hunt for small-bodied primates using tools to its, as well as a search image aligned mainly to flaked stone.
probe them from tree trunks (Pruetz and Bertolani 2007), and However, evidence may also simply be very rare because of one
female bonobos also participate in hunts (Hohmann and Fruth or more other possibilities (Panger et al. 2002): (1) these be-
2008). Percussion scavenging, especially in large social groups, haviors were deployed by early hominins inconsistently and
would have opened similar opportunities to females and ju- played a marginal role in day-to-day foraging, (2) activities may
veniles already engaged in tool-assisted extractive foraging. have been done while “foraging on the go” and not concen-
Unlike scenarios of human evolution in which females are trated at central places that form the “sites” composing most
passively provisioned with meat, early large-animal exploita- archaeological investigation, (3) tools unmodified prior to use
tion may have been at least initially driven by female strate- are difficult to identify, (4) diagnostic characteristics of pound-
gies. ing tools may weather readily over long periods, and (5) group
These and other higher-order interpretations are reliant on sizes of tool-using hominins may have been smaller and thus
concordance with the empirical record. Reports of flaked stone left smaller quantities of concentrated debris.
artifacts and butchered bones from deposits dating between Only fieldwork explicitly designed to test these scenarios
3.3 and 3.4 Ma suggest that both behaviors likely arose earlier can provide an answer. In light of the literature reviewed here,
than what was previously known, at 2.6 Ma. Until more older we provide the following empirical expectations:
localities are reported, it is premature to assign a specific func- 1. Evidence for the earliest consumption of animal tissues
tion to the tools or to interpret the stone flaking and large- should be rare.
animal butchery as either a widespread or a critical part of the 2. Evidence for the earliest consumption of animal tissues
early hominin adaptation. It is also difficult to interpret where should be primarily from percussive activities aimed at inside-
they fit in the overall sequence of behavioral change. Even tools bone tissue.
such as at LOM-3 may represent a relatively late manifestation 3. Evidence for percussion scavenging should primarily be
of stone tool use that began long after percussive extractive in the form of modified fossil surfaces with notches, associ-
foraging behaviors such as those used by chimps were already ated bone flakes, and fragmented long bones.
in place. At present, the LOM-3 and DIK-55 cases are both 4. Bone surface modifications should exhibit a wider range
singular and demand much further work (Domínguez-Rodrigo of sizes and morphologies than has been reported in the exist-
and Alcalá 2016). ing taphonomic literature, and marks should be deeper and
However, such discoveries have pushed paleoanthropology more often amorphous than linear.
into fresh directions for understanding the origins of the HPP. 5. Fossil evidence should be spatially associated with pro-
Most important, they illustrate the need for a shift in our the- spective percussors, or with microhabitats that offered per-
oretical framework and systematic survey for evidence of Plio- ishable percussors (e.g., large trees).
cene butchery through a range of proxies (stone percussors as 6. Percussors should exhibit a well-defined damage type,
well as flakes, and percussed as well as cut bone). Rather than which may be subtle.
disassembling familiar paradigms about the significance of 7. Evidence for percussion scavenging should be preferen-
these behaviors in human evolution (Aiello and Antón 2012; tially found where opportunities for transport of scavenged
Domínguez-Rodrigo et al. 2014), these new data point to two resources and bone surface preservation intersected, for ex-
thresholds of inquiry: (1) when and why the behaviors first ample, lake margins and riparian woodlands where hominins
began and (2) when and why they later became more complex cached stone or used trees as refuge.
and systematic. We advocate a new methodology of field research to test
The realities of the archaeological record are that it pre- these predictions that is essentially an archaeological approach
serves stones well, and bones far better than traces of other to deposits previously thought to be purely paleontological.
potentially important aspects of diet, such as starchy plants We also provide recommendations for development of an in-
(Hardy et al. 2015). Modified bone surfaces provide direct terpretive framework for these methods (table 1). We have
evidence that stone tools were at least sometimes employed in begun to implement some of these approaches through field-
butchery activities, but preservation and research bias have work, museum research, and methods development—starting
arguably both contributed to the impression that flaked stone with a taphonomic study of sieved fossil collections from Di-
tools and the HPP are dependent upon each other. By con- kika (Thompson et al. 2015), systematic survey and collection
ceptually stepping away from the traditional pairing of meat- of modified bones at Hadar, Ethiopia (Thompson et al. 2016),
eating and flaked stone tool manufacture, we critically examine and deployment of novel methods for the identification and

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 000 Current Anthropology Volume 60, Number 1, February 2019

Table 1. Summary of key gaps in data to be addressed with new field approaches and explicit development
of interpretive frameworks

Problem Solution
Field method:
Bias toward flaked stone tool localities Systematic survey for percussors and flake stones in Pliocene deposits
Training of workers and students to recognize potential percussive tools
Experimental replication of potential stone percussors
Lack of microscale data about taphonomic Sieved samples of fossils that map microhabitat occurrences likely to preserve bone
alterations Systematic documentation of scavenging opportunities via paired sedimentary and
taphonomic variables
Understanding the distribution of bones broken Systematic collection and refitting of fragmented long bones
by different taphonomic agents Systematic documentation of bone surface modification
Lack of diversity in referential models of An expanded experimental taphonomy program that emphasizes the material traces
percussion traces of percussive scenarios
Different raw materials (different types of rocks, bone, wood, hardened sediment)
Different technology types (rounded stones, sharp stones, Lomekwian tools) and animal
body sizes (medium and large)
Interpretive framework:
Poor resolution about scavenging opportunities Increase paleoecological research on Pliocene carnivore guild structure and scavenging
niches
Refine neotaphonomic models to emphasize how predator-prey events at the moment of
carcass acquisition influence nutrient availability
Additional work on the relationships between microbiome and dietary constraints/choices
Need for specificity in foraging constraints Develop nutritional data and foraging models that quantify the costs, returns, and spatial
distribution of marrow exploitation opportunities relative to other resources
Lack of data on energetics and biological Examine percussive biomechanics and energetics in apes, humans, and extinct hominins
constraints on percussion

analysis of bone marks (Harris et al. 2017; Otárola-Castillo Acknowledgments

et al. 2018).
The ideas presented here were generated through discussions
with Bill Kimbel, Andrew Barr, Denné Reed, Shannon Mc-
Pherron, and René Bobe. Bill Kimbel provided valuable com-
Conclusion ments on an earlier draft. J.C.T. and C.W.M. recognize the
support of the John Templeton Foundation (to the Institute of
A simple argument from the temporal coincidence of the Human Origins). The opinions expressed here are those of the
earliest known flaked stone tools and evidence for large-animal author(s) and do not necessarily reflect the views of the John
resource exploitation lacks explanatory power about the emer- Templeton Foundation. We are very grateful to Luis da Silva
gence, role, and relationship of these behaviors in early hom- for providing the illustrated alternative scenarios. Three anon-
inin ecology and evolution. Although some elements of ana- ymous reviewers and the editorial team at Current Anthropology
tomical and behavioral changes appear in our lineage by provided much constructive critique and greatly improved the
ca. 2.8–2.4 Ma, these do not become consistently expressed for article. This article was conceptualized while the lead author
at least another 400,000 years—or potentially longer. The was based at the University of Queensland and written while at
paradigm under which the emergence of Oldowan stone tool Emory University.
manufacture coincides with consistent, repeated large-animal
butchery by members of the genus Homo (among potentially,
other species) still appears useful and coherent. However, cur-
rent evidence does not support the interrelationship of these Comments
three events (advent of flaked stone tools, large-animal exploi-
tation, and speciation into Homo) as an explanation for their David R. Braun
origins. Instead, new theoretical and empirical approaches must Center for the Advanced Study of Human Paleobiology, George
Washington University, 2110 G Street NW, Washington, DC 20052,
be developed to understand the context of their earliest, and
USA ([email protected]).
potentially decoupled, emergence. Here, we propose a new
model: the regular exploitation of large-animal resources—the
Pliocene Hominin Behavior: New Insights
HPP—began with an emphasis on percussion-based scavenging
of inside-bone nutrients, independently of the emergence of Pliocene hominin behavior has gone from the relatively un-
flaked stone tool use. known to one of the most exciting new components of our

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
Thompson et al. Origins of Large-Animal Exploitation 000

understanding of the archaeological record. This is in part 2012; Mora and de la Torre 2005), the recognition of percussive
because of new data detailing both the appearance of stone tools remains somewhat elusive. Considering the difficulties
artifacts (Harmand et al. 2015) and evidence of cut marks on faced with identifying bone surface modifications (Harris et al.
bone surfaces (McPherron et al. 2010). Regardless of your 2017; James and Thompson 2015; Thompson et al. 2011), the
stance on the earliest evidence of these behaviors (Domínguez- identification of percussive tools will pose new challenges.
Rodrigo, Pickering, and Bunn 2011; McPherron et al. 2011; A further complication in the identification of percussive
Sahle et al. 2017; Toth 2017), most paleoanthropologists rec- technology relates to the use-lives of these tools. Percussive
ognize that our current data set for this time frame (12.6 Ma) tools frequently have relatively longer use-lives compared to
requires substantially more data before we can make definitive chipped stone artefacts (Carvalho et al. 2008; Shott and Sillitoe
claims about hominin behavior. However, these discoveries 2005). The implication for Pliocene assemblages is that we
combined new insights from the study of nonhuman primate should not expect to find large concentrations of these tools.
tool use suggest that our current paradigms likely need to shift If the identification of percussive tools requires microscopic
(Biro, Carvalho, and Matsuzawa 2010; Carvalho et al. 2008; scanning of surfaces, then collection procedures will need to
Haslam et al. 2009, 2013). I applaud Thompson and colleagues radically change at Pliocene sites. Making the argument that
for incorporating this new data to create a new perspective on numerous isolated stones need to be collected from Pliocene
hominin carnivory in the Pliocene. localities will certainly draw ire from museums that already
I agree with their assertion that the zooarchaeological record have difficulty storing the voluminous paleoanthropological
as well as our knowledge of Pliocene hominin diet (Sponheimer record. Finally, the identification of percussive technology
et al. 2013) does not seem to record major changes at the time from the remains of bones that were broken open for within
when Oldowan stone artifacts are first recovered from the ar- bone nutrients poses similar difficulties (Ferraro et al. 2018).
chaeological record (Domínguez-Rodrigo 2009; Robinson et al. If we are to test the HPP, dramatic new increases in our ability
2017). Although there are good reasons to believe that the zoo- to identify subtle traces of behavior must be developed.
archaeological record does not always faithfully reflect the pat- The premise of the HPP hypothesis is interesting, and the
terns of resource acquisition (Domínguez-Rodrigo and Yravedra authors should be commended for the testable expectations
2009), the large assemblages with clear evidence of human ac- of this hypothesis that they provide. One concern is the as-
tivity that appear after 2 Ma (Ferraro et al. 2013) are conspicu- sertion that small animals would not have ranked highly in
ously absent prior to 2 Ma. Why do we not see localities that the diet of Pliocene hominins. Thompson et al. suggest that
have single carcasses associated with large numbers of artifacts, because chimpanzees often forego hunting of low-cost items
such as Site 15 at Olorgesailie (Potts 1989), in time frames be- (e.g., monitor lizards), and other prey items require specialized
tween 2.55 and 2.0 Ma? A possible answer is that stone artifacts adaptations (e.g., levels of arboreality that allow for the hunt-
were not necessarily a frequent part of the hominin tool kit ing of smaller primates), small animals were an unlikely re-
prior 2 Ma (Shea 2017a). If that is the case, then maybe our source for Pliocene hominins. However, as they also note,
inferences about the nature of hominin carnivory should also these high-dietary-quality items rarely provide substantial ca-
be shifted. loric contributions (Tennie, O’Malley, and Gilby 2014). In-
A major insight by Thompson and colleagues is the ability deed, hunting among chimpanzees appears to be more related
to view the archaeological record from the perspective of new to certain social and ecological conditions (Mitani and Watts
data on the behavioral record of other primates (Carvalho 1999). Some small prey may have higher acquisition costs
et al. 2008). Thompson and colleagues are careful to use this because they require specialized technology to capture them
record to develop inferences while recognizing the potential (Clark and Plug 2008). However, many smaller prey items can
pitfalls of this kind of comparison. One of the implications be captured with little to no technology (Archer et al. 2014;
of the human predatory pattern, as they describe it, is for Stewart 1994). Some small aquatic resources have relatively
percussive tools to play a much bigger role in the Pliocene high return rates and maintain high fat content in times of the
archaeological record. Many primates use percussive tools, year when large mammals are relatively depleted of fat (even
and as such it makes sense that these would be major com- within bone nutrients; Caruana et al. 2014; Hawkes, O’Connell,
ponent of the earliest technology (Haslam et al. 2013; Luncz, and Jones 2014; Madrigal and Blumenschine 2000; Stewart
Mundry, and Boesch 2012). The frequency of these tool and Gifford-Gonzalez 1994). Although hominins may have
forms in the Lomekwi record would seem to support this increased return rates by focusing on within-bone nutrients in
assertion (Harmand et al. 2015). However, the relative lack large animals, they also would have been moving into this niche
of these tools in even the next youngest archaeological as- at a time when bone-crunching carnivores were numerous
semblages (e.g., Gona; Stout et al. 2010) suggests a possible (Lewis 1997; Werdelin and Lewis 2013;Werdelin, Lewis, and
disjoint between Pliocene and Pleistocene technologies. Further, Haile-Selassie 2014). Small prey may have been a difficult prey
if we use the tool use of primates as a phylogenetic analogy, then item to capture, but evidence from both the ethnographic and
it is possible that the percussive tools we see in the modern archaeological record suggests that hunters with relatively basic
primates represent a relatively young phenomenon (Haslam technology can access this resource (Ferraro et al. 2013; Hawkes
2014). Furthermore, despite extensive experimental studies 1990; Hawkes et al. 2014). It is possible that small prey items
(Benito-Calvo et al. 2018; Caruana et al. 2014; de la Torre et al. were a part of the hominin dietary pattern for much of our

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 000 Current Anthropology Volume 60, Number 1, February 2019

lineage, yet the evidence of these resources may be difficult to Our current appreciation for the signatures left by the ex-
identify (Behrensmeyer, Kidwell, and Gastaldo 2000). traction of within-bone nutrients and their potential to be
Much of our understanding of hominin behavioral ecology mimicked by other biostratinomic and diagenetic processes is
rests on the substantial investigations on this topic beginning extremely limited. Actualistic research on trace evidence for
several decades ago (Blumenschine, Cavallo, and Capaldo 1994; within-bone nutrient extraction found on artifacts and bones
Oliver 1994; Plummer and Bishop 1994; Rogers, Harris, and lags far behind that for outer-bone nutrients, where there has
Feibel 1994). Since those landmark studies, our knowledge of been a recent explosion of innovative methods proposed to
hominin behavior (Harmand et al. 2015; McPherron et al. improve the reliability of cut mark identification (Harris et al.
2010) and biology (Sponheimer et al. 2013; Ungar, Grine, and 2017; Otárola-Castillo et al. 2018; Pante et al. 2017). None of
Teaford 2006; Villmoare et al. 2015) has changed substantially. these techniques have been applied to percussion damage,
I agree with Thompson and colleagues that it is time to view and researchers are left to rely on qualitative criteria, first
this record with a different perspective. As always, the devil is published decades ago (Blumenschine and Selvaggio 1988) to
in the details. identify percussion marks on bones or distinguish hammer-
stone impact notches from those inflicted by carnivore teeth
(Capaldo and Blumenschine 1994). Just as problematic as
the identification of traces on bone are the use-wear traces on
the tools used for exposing within-bone nutrients, which have
Michael Pante
Department of Anthropology, Colorado State University, 1787 recently been shown to be minimal and difficult to distinguish
Campus Delivery, Fort Collins, Colorado 80523, USA (Michael from those produced by stone tool manufacture (Benito-Calvo
[email protected]). 10 II 18 et al. 2018). Given the limitations of our knowledge, the pa-
leoanthropologist is incredibly ill-equipped to find early evi-
Over the last few years, Thompson and colleagues have taken a dence of the HPP, especially if it is in the form of percussion
systematic approach to improving the methods and standards damage on bone or tools.
that we employ in investigating the earliest archaeological Even the more extensively studied trace evidence of outer-
record. Here they present a theoretical and semantic rebrand- bone nutrient extraction in the form of cut marks or flaked
ing of human carnivory and its uniqueness when compared stone tools is difficult to reliably identify in the patchy Pliocene
with the carnivorous components to the diets of our nonhu- archaeological record. The 3.4 Ma marks from Dikika have
man primate relatives. Central to their argument is what they been differentially interpreted as cut marks (McPherron et al.
believe is an overemphasis on meat-eating and unjustified 2010), trample marks (Domínguez-Rodrigo et al. 2012), and
reliance on models derived from living ape ecology that pos- crocodile tooth marks (Sahle et al. 2017), all using methods
tulate the consumption of large prey in our ancestors emerged that are qualitative and impossible to evaluate for accuracy.
from cooperative hunting of smaller prey, a behavior observed Here, Thompson et al. suggest that the apparently anomalous
in our closest living relative, the chimpanzee. They suggest the depth of the marks on these specimens may indicate a per-
earliest evidence of what they term the “human predatory cussive origin. However, the depth and morphology of the
pattern” (HPP) should be in the form of trace evidence of per- marks is not outside the range of those inflicted by the pow-
cussion activities, such as percussive technology and percussion erful jaws of crocodiles that use dynamic impacts, like those
damage inflicted on bones during the extraction of within-bone inflicted by hominins wielding hammerstones, to kill and dis-
nutrients, which are rich in fat and remain edible for far longer member their prey (Njau and Gilbert 2016). The traces left by
than flesh after the death of an animal. this distinct feeding behavior can produce not only V-shaped
Others (Binford 1988; Blumenschine 1995; Capaldo 1997; pseudocut marks but also pits with associated microstriations
Selvaggio 1998) have emphasized the importance of within- that are easily mistaken for percussion (Njau and Gilbert 2016).
bone nutrients in early hominin subsistence strategies, sug- My own observations of crocodile-modified bones indicate that
gesting our ancestors likely acquired large mammal carcasses they can also create notches and flake scars that are similar to
that were mainly devoid of flesh through passive scavenging those inflicted during hammerstone percussion. Evidence of
from the kills of carnivores. However, these hypotheses were crocodiles at Plio-Pleistocene archaeological sites is common,
based on the 1.8 Ma FLK 22 Zinjanthropus assemblage from but their potential to mimic the feeding traces of human an-
Olduvai Gorge, Tanzania, and at the time, the potentially much cestors has only begun to be recognized. As such, any identifi-
deeper antiquity of the evidence for human consumption of cation of human predatory behavior in the Pliocene is prema-
large prey was not yet appreciated. Given the substantial tem- ture and should not be accepted until we meet methodological
poral expansion of the archaeological record since the debate standards of inquiry that are currently well beyond our capa-
over the FLK Zinjanthropus assemblage began, the reframing of bilities.
our hypotheses is long overdue. However, investigators of the If we are to identify the earliest evidence of the HPP in the
emergence of the HPP must overcome considerable challenges, archaeological record, we need to completely revolutionize
especially if this behavior began in the form of percussive ac- and standardize our approach to this investigation. We need to
tivities, as Thompson et al suggest. work together to develop a greatly expanded and open-access

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
Thompson et al. Origins of Large-Animal Exploitation 000

actualistic database aimed at understanding the potential for faunal assemblage. (Remember, evolution’s opponents read our
equifinalities between traces left by human carnivory on bones professional disagreements as epistemological uncertainty.)
and tools and those left by any bone or stone modifying pro- Inasmuch as this paper offers predictions for the zoo-
cess that we can replicate. We need to further invest in the new archaeological evidence, the rest of my remarks focus on its
identification methods that are in development (Harris et al. implications for the lithic record. For humans and our hominin
2017; Otárola-Castillo et al. 2018; Pante et al. 2017) and apply precursors, stone cutting tools are, in essence, artificial teeth and
them to our actualistic samples, for it is in this application that nails. Chimpanzees use their teeth and nails to kill and eat
we can potentially break equifinalities perceived through tra- smaller animals, but they also use their teeth and nails to shape
ditional qualitative methods of analysis. We need to imple- wooden artifacts. Humans use stone cutting tools as butchery
ment blind testing to ensure that these methods are accurate aids, and we use them to shape wooden tools, things for which
and replicable between research teams, and we need to antic- we do not generally use our teeth or fingernails. Because fos-
ipate the effects of postdepositional processes on the criteria silized bones preserve better than wood, and because some Plio-
used by our models. These standards are difficult to meet, and Pleistocene bones have stone tool cut-marks on them, paleo-
it will take years if not decades to adequately satisfy the out- anthropologists have long assumed that the first stone cutting
lined expectations. However, we can no longer rely on the ap- tools were butchery aids. In fact, no prior theory justifies making
proaches that have led to a complete lack of consensus among this assumption. Hominins might have used stone cutting tools
researchers. Thompson and colleagues have already begun to as aids to carpentry for eons before repurposing them as butch-
undertake these challenges, and we have reason to be hopeful ery tools. Which of these is the more primitive or derived ho-
that paleoanthropologists will ultimately meet the higher pro- minin behavior remains unknown, but it is an issue on which
posed standards in our investigations of the HPP. only archaeology can shed light. Thus far, nearly every Pleisto-
cene sedimentary deposit in which archaeologists have found
preserved wood also preserves wooden artifacts with stone tool
marks on them. Unless the Early Pleistocene archaeological rec-
ord hides some as-yet-undiscovered “Carpentry Revolution,”
John J. Shea one expects that if we ever find Plio-Pleistocene-age wood, we
Anthropology Department, Stony Brook University, Stony Brook, will also find purposefully shaped and cut-marked wooden arti-
New York 11794-4364, USA ([email protected]). 10 I 18 facts.
One agrees with Thompson and colleagues that “paleo-
I congratulate Thompson and colleagues on a substantive and anthropologists must resist being constrained by the known
thoughtful contribution to the Plio-Pleistocene hunting versus archaeological record,” but neither should we limit our hy-
scavenging debate. Using the same term for different things is potheses about early hominin tool use to what apes and mon-
never a good idea. Meat-eating and stone tool use are just such keys do with rocks and sticks. Archaeologists welcome insights
“two-for-one” concepts. Their description of the HPP teases from ethology and experiments with captive primates, but many
these concepts apart, contrasting the costs, benefits, and risks such studies emphasize broadly defined behavioral similarities
of using stone cutting tools to detach meat-protein from large- among “technological primates” rather than important differ-
animal carcasses with those involved in using percussive tools ences (Shea 2017b). Focusing on similarities distracts us from
to extract marrow-fat from inside large-animal bones. The paper evolutionary anthropology’s most important question: how dif-
argues that percussive fat extraction preceded stone cutting ferences between humans and other primates evolved. In evolu-
tool use as the HPP’s foundational strategy. tion, only differences matter.
Thompson and colleagues make explicit predictions about Let us consider stone percussors. Thompson and colleagues
what we should find if their hypothesis is correct, and they write, “Even at Oldowan sites with abundant pounding ar-
call for increased scrutiny of those Pliocene sedimentary de- tifacts, both lithic and faunal analyses do not suggest that
posits likely to contain relevant evidence. For this research to marrow extraction was the main activity taking place—the
be successful, it has to be done mindfully. Paleoanthropolog- percussive tools were being extensively used for other pro-
ical claims of the earliest anything invariably ignite contro- cessing.” These other percussive activities might have included
versy. Much controversy about early stone tool use focuses nut cracking or other plant food-processing tasks similar to
on the criteria for recognizing hominin agency in modifica- what some living nonhuman primates do with stone per-
tions to stone or bone. All too often, archaeologists develop cussors, but they could include activities only living humans do
these criteria after the fact of discovery, and the criteria in together with quintessentially hominin qualities of behavioral
question include subjective, visually assessed morphological variability. Percussion makes noise, and percussion instru-
analogies and appeals to authority. This must change. Paleo- ments are either human universals or nearly so. Depending on
anthropologists need to reach consensus about objective terrain, vegetation, and atmospheric conditions, noise from
measurement-based criteria for recognizing percussion-damaged stone-on-stone percussion carry for hundreds of meters or
bone and stone before we invest much time and energy de- more. Humans are more prosocial than apes are. Perhaps our
fending our preferred interpretation of one or another lithic or hominin ancestors used stone percussion to signal and locate

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 000 Current Anthropology Volume 60, Number 1, February 2019

one another in dense vegetation so that they could find and and softer body parts (brain, marrow, viscera) allows the in-
groom one another and unite for common defense as night dividual to learn to prefer the latter by means of their better
fell and carnivores stirred to life. Stone-on-stone percussion payoff. In turn, this preference may motivate the individual to
produces sharp reports that can drive away carnivores. (If access encased brain and marrow of large preys by using
this does not work, thrown projectiles effectively clarify the pounding tools to exploit inside-bone nutrients. Let us now
“humans 1 noise p pain” equation.) Early hominins had consider the evidence supporting this view.
larger brains than living apes do and, it follows, led more Chimpanzees and bonobos “hunt, but never for prey larger
complex social lives (Dunbar 2016). Stone-on-stone percus- than themselves” (Thompson et al.). The same holds true for
sion might have been part of social performances, precursors capuchin monkeys (genus Cebus and Sapajus; even though I
to such uniquely human institutions as music and song observed four young bearded capuchin monkeys together
(Mithen 2005). Might the oldest stone percussors and flaked hunting an iguana twice as big as each of them—the iguana
tools have been by-products of such noise making later co- was not killed, but its tail was torn off and each monkey got its
opted into use as aids to resource extraction and tool making? share; Visalberghi and Albani 2014). Chimpanzees and ca-
Perhaps they served all these purposes simultaneously, or puchins eat the different body parts of the prey often starting
differently in response to situational variables. Can we identify from, or privileging, the viscera and the brain, that is, the softer
these different activities’ lithic signatures? Not yet, but we will parts (Prieto 2013; Tennie et al. 2014). The caloric contents of
never know whether we can unless we start asking more ex- these food sources are not strikingly different, although micro-
pansive questions about early stone tools. Primitive means and macronutrients are.2
“ancestral.” Just because Plio-Pleistocene hominins made stone Animals learn what to feed upon from the feedback pro-
tools long ago does not mean their strategies for using those vided by each food (Galef 1996; Visalberghi et al. 2003). Ex-
tools were simple. (After all, those ancestral hominins “invented perimental evidence demonstrates that in capuchins, food
inventing.”) The more we study nonhuman primate tool use, preferences are significantly related to the energy intake rate
the more complex it appears. We should anticipate the same (kilojoule ingested per second), that is, to the amount of energy
learning curve in research on early hominin technology. ingested per unit of time (Stammati, Sabbatini, and Visal-
berghi 2008). However, food preferences are not significantly
related to ingestion rates (grams of food ingested per unit of
time) and to food energy content alone (kilojoule per gram).
These findings may well account for a stronger preference for
soft body parts (e.g., brain, viscera, and bone marrow) than for
Elisabetta Visalberghi meat, the consumption of which requires longer mastication
Istituto di Scienze e Tecnologie della Cognizione, National Research and provides less energy per unit of time.
Council of Italy (CNR), via Aldrovandi 16 b, 00197 Rome, Italy
([email protected]). 27 II 18 The preceding findings on living nonhuman primates
strengthen the argument that early hominins, well before
Meat Exploitation: Reflections flaking to cut meat, used pounding tools to access brain and
bone marrow and, why not, to tenderize meat and reduce
by a Cognitive Primatologist
mastication time.
The article “Origins of the Human Predatory Pattern: The
Transition to Large-Animal Exploitation by Early Hominins”
brings an innovative approach to investigate the predatory
pattern(s) of the hominin lineage. A critical examination of the
state of the art has led Thompson et al. to argue that “the Lars Werdelin
Department of Palaeobiology, Swedish Museum of Natural History,
transition to the HPP was not reliant on concurrent flaked Box 50007, SE-10405 Stockholm, Sweden ([email protected]).
stone tool technology to cut meat, but rather that percussive 4 III 18
extraction of inside-bone nutrients offers a more parsimonious
pathway by which hominin first made regular use of large- I am in complete agreement with the general principles and
animal resources.” reasoning behind the paper by Thompson et al. on the HPP.
This new focus opens a refreshing window to look at the Nevertheless, it does require some comment, and perhaps
costs and benefits of extracting inside-bone nutrients (e.g., clarification of a sort, from the perspective of the carnivoran
bone marrow and brain) versus outside-bone nutrients (e.g., mammals (members of the Order Carnivora—distinct from
meat). As a cognitive primatologist, I consider that the prox-
imate mechanisms of extant nonhuman primate behavior are
inspirational to draw possible scenarios of meat consumption 2. US Department of Agriculture, National Nutrients Database for
by early hominins. Specifically, I argue that capturing small Standard References, https://www.ars.usda.gov/northeast-area/beltsville-md
prey allows the individual easy access (i.e., without the use of -bhnrc/beltsville-human-nutrition-research-center/nutrient-data-laboratory
tools) to brain and marrow and that the consumption of meat /docs/usda-national-nutrient-database-for-standard-reference/.

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
Thompson et al. Origins of Large-Animal Exploitation 000

carnivores, animals that exploit animal dietary resources) that around 3.6 Ma (or slightly earlier given the latest dates of some
early hominins came into conflict with due to the develop- critical sites), this can largely be attributed to poor sampling
ment of the HPP. of the record prior to this time and to the “Laetoli effect,” in
Looking back into the Pliocene and Pleistocene, hominoids which one of the richest sites for Pliocene carnivorans in
and carnivorans are unusual among nonobligate herbivores eastern Africa is also one of the earliest, greatly inflating the
in that both groups include members reaching a large body size number of new species in its time slice. This applies to
(here identified as species with a mean body mass greater than carnivorans, but one must assume that it applies equally to
21 kg). This is important because 21 kg represents a threshold herbivores. Second, bone cracking as an advanced adaptation
value among carnivorans at which the dietary pattern changes did not first appear in Africa ca. 3.6 Ma (with Crocuta). It was
from species feeding mainly or exclusively on prey with a body present in Africa from at least the late Miocene. In fact, specific
mass much less than their own to species that feed mainly or dental adaptations had evolved as a synapomorphy of all de-
exclusively on prey of equal body mass or greater than their rived Hyaenidae by 11 Ma at least and are thus shared by all
own (Carbone et al. 1999). This is a general energetic con- African hyenas, living and extinct, with the exception of the
straint and is not limited to carnivorans. Both chimpanzees aardwolf, Proteles (Ferretti 2007; Werdelin and Solounias
and early hominins exceed this mass threshold, yet chim- 1991). Thus, the hyenid fossil record does not support eco-
panzees do not exploit large-bodied prey, whereas early logical change around 3.6 Ma that involved increased scav-
hominins did (although when this began is a moot point, as enging opportunities. If there was any specific time at which
outlined in the paper). This means that, from the carnivoran early hominin exploitation of inside-bone resources became
perspective, chimpanzees are not carnivores, merely animals possible, this would have been due to changes that allowed
that occasionally exploit animal resources for reasons that may these early hominids access to a resource that hyenas already
have nothing to do with nutrition. Early hominins, on the were exploiting.
other hand, would be considered carnivores, although this In summary, from the carnivoran perspective the hypoth-
perspective is silent on the amount of animal protein ingested esis of Thompson et al. regarding the HPP represents a sig-
relative to other foodstuffs. Important to note, in agreement nificant and important change in focus in the study of the
with the Thompson et al. paper, this means that chimpanzees process of “dietary hominization.” It more clearly places
are an unsuitable model for investigating the HPP and the HPP hominins as a marginal but significant member of the large
cannot be extrapolated from chimpanzee feeding behavior. carnivore guild and helps place prior hypotheses of carnivore
Therefore, in my carnivoran-biased view, Thompson et al. are extinction in east Africa (Werdelin and Lewis 2013) in a
quite correct in proposing a new paradigm, divorced from the broader context. Finally, the authors speak of a new method-
chimpanzee-based model. ology of field research taking an archaeological approach to
As carnivores, early hominins shared a trait that gave them a questions thought purely paleontological. From the carni-
competitive advantage over contemporary carnivorans: they voran perspective I would also (self-servingly) encourage re-
had available to them a diversity of fallback foods, including searchers to adopt a more broadly biological approach to
vegetable matter as well as animal protein not obtained from questions thought purely anthropological. In the final reck-
mammals. Carnivoran mortality is often associated with a oning, it is evolution that drives it all.
reduction in the amount of available animal protein during
times of environmental stress. Early hominins, on the other
hand, had less risk of starvation than other large-bodied car-
nivores, as they could sustain their existence on a diet that did
Richard Wrangham
not include protein from large mammals. This gave early ho- Peabody Museum, Harvard University, 11 Divinity Avenue,
minins a competitive edge that would have allowed them Cambridge, Massachusetts 02138, USA ([email protected]
to increasingly encroach on carnivoran ecospace, eventually .edu). 6 II 18
leading to a collapse of the large carnivoran guild in eastern
Africa (Werdelin and Lewis 2013). It is further clear that the Thompson et al. constructively propose that a critical inter-
more spectacular phase of this collapse, dated to ca. 2 Ma, was mediate step between hominins killing small and large prey
the tipping point of a long-term trend of increasing hominin was the exploitation of large-animal carcasses for marrow and
competition with carnivorans, to the detriment of the latter. brains. The key activity, percussive fat gathering, would have
Issues of habitat change and the risk of confrontation are im- differed importantly from eating meat.
portant in this context. In the moment of confrontation the risk The idea that percussive fat gathering would have been a
would have been to the hominins, but on an evolutionary time regular Pliocene activity seems plausible and heuristically
scale the risk was to the carnivorans. A discussion of these im- valuable. As Thompson et al. note, it would have been a rel-
portant issues is beyond the scope of this commentary, however. atively low-cost and high-benefit activity. Marrow and bone
Some aspects of carnivoran evolution in Africa mentioned persist longer, in better condition for a consumer, than outside-
in the paper should be clarified. First, although the record bone products and would therefore have been more readily
shows a number of first appearances of carnivoran species at available. Percussive technology has arisen several times in

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 000 Current Anthropology Volume 60, Number 1, February 2019

nonhuman primates, and chimpanzees readily eat marrow need to cut meat away to reach areas of fat as much as to ex-
(and, it should be noted, brains; Goodall 1986). ploit the meat itself? Fat areas were not confined to within-
I suggest an additional argument not mentioned by bone regions. For instance, in impala Aepyceros melampus the
Thompson et al. The raw meat of large adult animals was not volume of kidney fat varies widely over the year and is closely
necessarily very valuable. When chimpanzees kill infant prey correlated with the percentage of fat in marrow (Dunham and
or infant chimpanzees, they normally eat them (Arcadi and Murray 1982). Hominins’ ability to reach fat both outside and
Wrangham 1999; Stanford 1998). By contrast, no cases of within bones might have been an important stimulus for using
chimpanzees eating adult chimpanzee meat have been re- flakes.
corded, despite dozens of opportunities to do so (Wilson et al. A final point is a sidenote on a specific datum cited from
2014). When victims are killed in a territorial boundary area, Kimbel and Villmoare (2016), who reported the striking claim
the lack of meat-eating could be explained by the killers being that australopithecines had a brain size about 30% larger than
motivated to return to the safety of the more central part of a chimpanzee. The 30% figure applies only to Pan troglodytes
their community range. That explanation does not apply to troglodytes, which according to Isler et al. (2008) has the
victims killed within a community, however. Chimpanzees are smallest recorded endocranial volume (ECV) among chim-
evidently not averse to cannibalism or to tasting their adult panzee subspecies (female: 347.2 cc, n p 41; male: 378.9 cc,
victims, as they have occasionally ingested small amounts of n p 38). For P. t. schweinfurthii, measurements of adult ECV
blood from open wounds. These points suggest that the rea- from Kibale National Park in Uganda (five female, four male),
sons why chimpanzees avoid eating the meat of adults are combined with data from Isler et al. (2008; two female, two
“economic.” Difficulties could include not only meat being male) yield higher means (female: 362.9 cc, n p 7; male:
difficult to detach but also its having a low value due to its 436.9 cc, n p 6). The mean ECV of females and males is thus
being difficult to chew raw, more liable to pathogens, and 363.1 cc for P. t. troglodytes versus 399.9 cc for P. t. schwein-
putatively less easily digested than the meat of younger animals furthii. If australopithecines are given a mean ECV of 470 cc
(due to higher levels of collagen). This could also explain why (Kimbel and Villmoare 2016), their ECVs are larger than chim-
chimpanzees sometimes leave the carcasses of medium-sized panzees by 29.5% for P. t. troglodytes, versus only 17.5% for
prey (monkeys or ungulates up to about 10 kg) that they have P. t. schweinfurthii. Although the sample size for P. t. schwein-
killed (Goodall 1986). In the early stages of exploiting large furthii is small, this note suggests that it is premature to con-
carcasses, therefore, “inside-bone” fat may have been routinely clude that australopithecines had brains 30% larger than those
preferable to meat because the meat could have been low value. of chimpanzees in general.
Investigation of the costs and benefits of chewing the raw meat
of large wild adults would be helpful to assess this issue for the
early stages of increased carnivory. Later, the adoption of
cooking would have increased the relative benefits of meat-
eating compared to fat-eating (Wrangham 2017). In short, the
relative nutritional value of meat compared to fat may have Reply
been lower prior to the adoption of cooking compared to af-
terward. This contribution aimed to establish a new theoretical frame-
If percussive fat gathering was an important strategy for work and open empirical research avenues on a major dietary
Pliocene hominins, brains would have had a different signifi- transition in hominin evolution. Our approach was twofold:
cance from marrow. The fat content of mammalian brains (1) critical deconstruction of the terms and underlying assump-
never changes significantly from around 50%–60%, whereas tions that have limited previous discourse and (2) presentation of
the fat content in ungulate marrow varies over the year be- a new model for this dietary transition, with test implications,
tween about 20% (when the animal is starving) to more than that relocates percussive within-bone nutrient extraction to an
90% (Lupo 1998; Wrangham 2017). This means that hominins explicitly central position. We then offered a road map for future
are expected to have exhibited a distinct seasonal pattern in the work that must be interdisciplinary in nature. We were thus
use of these resources: marrow should have been eaten more pleased to see responses from a range of perspectives, including
during seasons of high ungulate quality (i.e., good grazing con- archaeology, primatology, and paleoecology-paleontology. Here,
ditions). Brains would always have been valuable foods, as they we reply to issues the commentators have thoughtfully raised
are among chimpanzees, which often prioritize eating them concerning the nature of the empirical record, the methodolog-
(Goodall 1986). They should have been particularly important ical tool kits we must develop to advance research in this area,
targets during bad periods, when marrow quality was poor. and the theoretical framing of our percussion scavenging model.
If fat was a significantly higher quality and more valuable
food than meat, especially prior to cooking, hominins may
Pattern and Constraint of the Empirical Record
have been faced with the need to cut into carcasses more be-
cause meat was an obstruction than because it was a desirable Braun emphasizes the absence of large archaeological and
item. Is it possible that early butchery could have reflected the zooarchaeological assemblages prior to ∼2.0 Ma, also noting

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
Thompson et al. Origins of Large-Animal Exploitation 000

how (so far) there are no reported instances of single-carcass informative? Where can we store all the additional rocks and
butchery events dating to this time. This archaeological in- fossils that must be collected? Others are more abstract. How
visibility may adequately be explained by scarcity and lack of can we identify subtle anthropogenic traces in the field, and
investigation of appropriate-aged deposits or by absence of discern these from other traces, to guide collection? Where on a
the behavior. This directly addresses one of the main archae- vast landscape of possibilities should we look for them?
ological hurdles, which is how to interpret absence of evidence Pante takes a pessimistic view of the practical obstacle of
at a given time period. In cases where the emergence of some- identifying traces associated with within-bone nutrient ex-
thing new is under investigation, verifying its absence is as traction, emphasizing the ongoing uphill climb that taphono-
critical as documenting its presence. This places a heavy bur- mists face in interpreting behavioral traces on bone and stone
den of proof on researchers to demonstrate that their methods surfaces. However, in reiterating our assertion that within-bone
are adequate for finding the requisite evidence, if such evidence nutrient extraction has been a persistent but underemphasized
does in fact exist. element in this work, he offers a nugget of optimism. Many of
We argue that work done thus far in paleoanthropology to the new technical and inferential approaches, such as 3D scan-
understand the transition to large-game foraging does not yet ning and Bayesian modeling, have yet to be comprehensively
meet this standard, but it is equally important to recognize applied to such traces. These approaches offer fine-resolution
that even the most comprehensive new survey methods may data that can be subjected to more objective and probabilistic
not be enough. The earliest interactions of hominins with car- statements on the agent surface modification. By placing per-
casses may have initially been constrained by environments— cussion traces as central test implications in a new model of the
thus, specific depositional settings—where fossils simply may transition to large-carcass foraging, we hope that such work will
not preserve. They may have been conducted with no tools, with receive the attention it merits. It is an exciting moment in ta-
perishable tools, or with tools that do not preserve diagnostic phonomy; even between the time of Pante’s writing and this
traces. This is an old problem in paleoanthropology; some of the reply, percussion damage to bone has begun to enter into these
milestones in our evolution may have occurred well outside the analyses (Yravedra et al. 2018).
boundaries of the small localities from which we derive most This rapid escalation of new experimental and analytical
fossil data. However, a general pattern is still discernable. We work reflects a renewed, broadly applicable cycle of tapho-
take a similarly optimistic view of the Pliocene archaeological nomic investigation (James and Thompson 2015). As during
record but argue that its pattern will not be discernable while the previous cycles, data from Plio-Pleistocene assemblages have
field is dominated by individual and often serendipitous dis- been instrumental in advancing this work, but now there is a
coveries of the presence of a given behavior (flaked stone tool key difference. This one is fueled by questions that, until re-
use, carcass butchery, etc.). Now is the time to begin the much cently, few researchers would have thought to ask (McPherron
harder task of systematically demonstrating the timing and lo- et al. 2010). The DIK-55 marks have been controversial, but
cation of both their presence and absence. they have played a critical role in stimulating a new arena of
Shea notes that we may need to reconsider other key as- research. First, there is recognition that new objective and
sumptions, as there is no a priori reason to consider that the probabilistic techniques for assigning agent to surface modi-
earliest use for flaked stone tools was in fact for butchery. fication need to be developed for these challenging fossil as-
Because bones and stones preserve better than other classes of semblages, and as Pante notes, several research groups are
evidence, and because sometimes they are found in associa- vigorously making headway. Second, fieldwork targeting pre-
tion, an undertheorized coevolutionary explanation has been Oldowan sediments is required, and that fieldwork must be
forged. What we imagine early hominins doing with pounding equipped with new methods, largely developed in archaeology,
tools is similarly constrained by our modern interpretive mod- that target fragmented fossil bone ignored by past field studies.
els. Shea argues that knowing essentially nothing about early Finally, we need fresh theoretical perspectives to help guide
percussive tool use in our own lineage does not give us license this research, and as the commentators note, our paper tackles
to use living nonhuman primates as the default model. Instead, this need.
he looks to percussive behaviors unique to hominins as suffi- The DIK-55 marks offered impetus for researchers to revisit
cient explanation for their emergence. This does not, however, other collections that long resided in the literature as early
negate the value of extant primate studies in establishing what instances of stone-tool-assisted butchery (Sahle et al. 2017).
is and is not unique within our lineage. More important, they have stimulated a critical rethinking of
methods of identifying bone surface modifications that were—
and still are—common fare in zooarchaeology. This crisis of
Methodology
faith in bone surface modification studies, at least when ap-
The commentators raise rightful concerns with a departure plied to small pre-Oldowan samples, has exposed several key
from traditional approaches to paleontological and archaeo- problems. For example, there has been an overreliance on
logical survey of Pliocene deposits. Some are inherently prac- expert knowledge systems in mark identification (Harris et al.
tical. How can we justify dedicating large quantities of field 2017), which penetrates even to the level of individuals trained
time to documenting minute traces that may not ultimately be in the same research tradition (Domínguez-Rodrigo et al.

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 000 Current Anthropology Volume 60, Number 1, February 2019

2017). Expert knowledge of blind-tested analysts was once our view of available evidence from the perspective of changes
the accepted approach, but with very small mark samples in in the hominin lineage to that of entire ecosystem transfor-
contexts where there is ambiguity about which tools were used, mations.
it is clear that new objective and probabilistic approaches are The impact of hominin competition on the carnivore guild
needed. These may include, for example, machine learning is also worth considering in light of the variability selection
(Domínguez-Rodrigo and Baquedano 2018) and Bayesian hypothesis, which is typically applied to explanations of spe-
inference (Harris et al. 2017; Otárola-Castillo et al. 2018) to ciation and innovation within the hominin lineage itself (Potts
reduce as much as possible the influence of individual experts and Faith 2015). It is well known that large-bodied carnivores
in mark diagnosis. are most vulnerable to extinction during periods of ecosystem
These approaches also address a second problem: marks change (Ripple et al. 2014) and that generalists typically fare
have traditionally been diagnosed in a binary manner (either it better than specialists (Clavel, Julliard, and Devictor 2010).
“is” or it “is not” a cut mark, trample mark, tooth mark, etc.). The implications are that some hominins, as the ultimate gen-
By coding suites of characteristics of marks, probability can eralists, have shaped the course of faunal communities even as
be assigned to the overall goodness of fit against modern ex- they themselves have adapted to environmental change. How-
perimental analogues. The next step will be to establish—and ever, in advocating for a “more broadly biological approach,”
quantify—the physical variables that underlie production of Werdelin narrows his own view to exclude other key advantages
these characteristics. Lithic analysts have developed tools for unique to the hominin lineage: technology, culture, and ex-
standardizing the experimental production of stone artifacts panded forms of cooperation.
so that they know how changes in striking angle, force, core Biological sciences alone cannot explain the origins and
morphology, and other variables affect the outcome of a knap- evolution of a species the ultimate adaptation of which is
ping event (Dibble and Rezek 2009; Rezek et al. 2011). In ta- cultural, and we argue for an approach that fuses biology and
phonomy, we remain decades behind in understanding and culture in an effective manner. We do not yet know when these
constraining comparable individual variables (James 2018). features evolved and became significant, but together they gave
Pante and Shea both carefully articulate the many places hominins a potent lethality against prey and competitors. As
such work has yet to go before taphonomists are fully equipped we note in our paper, the hand of Australopithecus afarensis
to identify the traces we expect to be associated with the tran- already suggests regular tool use. Fallback foods often require
sition to the HPP. Shea emphasizes the importance of agreement relatively intense extractive foraging, and tool use may have
not only in advancing the science of paleoanthropology but in been the key pathway into the efficient use of those foods.
demonstrating epistemological coherence outside of our disci- Werdelin also offers important clarification on the emer-
pline. Pante focuses on specific attributes of the marked fossils gence of bone-cracking morphology in some carnivoran line-
from DIK-55 to argue that until we have better methodological ages, noting that in Africa such morphologies have been present
tools for untangling the traces left by different agents, we cannot since 11 Ma. Thus, hominin entry into the bone-cracking niche
accept this as definitive evidence of stone-tool-assisted butchery was not a response to increased scavenging opportunities but
in the Pliocene. We concur that advances in taphonomic method rather a shortcut in the evolutionary pathway necessary to ac-
must be an integrated part of research into the origins of the cess within-bone nutrients. The niche was already occupied, and
HPP, and we argue that therefore it is equally premature to as- this makes intrusion by hominins all the more impressive.
sign a new interpretation of the marks until we have all those Again, a key insight is that the intrusion into this bone-cracking
tools at hand. Fortunately, and thanks to the efforts of several niche was via technology and culture, not biology, and this
research groups, that future seems increasingly near. percussive technology may have ultimately been the preadap-
tation to stone tool technology itself.
Visalberghi offers a more proximate perspective, arguing for
Theoretical Framing
how preferences may have developed for specific carcass por-
Werdelin situates the transition to the HPP within a tions (within-bone vs. outside-bone nutrients). She uses evi-
“carnivoran-view” perspective of the encroachment of hom- dence from outside the ape clade to argue that small prey
inins into their competitive ecospace between ∼3.6 and 2.0 Ma. exploitation is requisite to developing taste awareness of within-
He argues that over evolutionary timescales, the flexible nature bone nutrients, because in small prey these can be exploited
of hominin diets were deadly to carnivorans, even if individual without tools. Energetics form an important part of this argu-
encounters were deadlier to hominins. Thus, the transition to ment, because the costs of chewing raw meat are relatively high
the HPP should be heralded by a general increase in dietary (Zink and Lieberman 2016). From this, Visalberghi offers an
breadth and flexibility, both of which have test implications for additional use for pounding tools as meat processors, which
lines of evidence such as zooarchaeology, dietary isotopes, and would have expanded their utility to outside-bone nutrients and
dental microwear. Important for behavioral models, hominins made meat more attractive on both small- and large-bodied
need not have engaged in frequent direct competition over prey.
carcasses to have waged a war of slow attrition that ultimately Wrangham takes a similar stance on the value of meat versus
facilitated even more regular access. Thus, Werdelin widens softer parts, citing chimpanzee data showing that meat from

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
Thompson et al. Origins of Large-Animal Exploitation 000

infant chimpanzees is eaten, but meat from adults—the only Benito-Calvo, A., A. Arroyo, L. Sánchez- Romero, M. C. Pante, and I. de la
Torre. 2018. Quantifying 3D micro-surface changes on experimental stones
putative “large prey” of other chimpanzees—is avoided. Wrang- used to break bones and their implications for the analysis of Early Stone
ham argues that this is because meat is not per se an energeti- Age pounding tools. Archaeometry 60(3):419–436. [DRB, MP]
cally profitable resource (in his view, it did become profitable Benito-Calvo, Alfonso, Susana Carvalho, Adrian Arroyo, Tetsuro Matsuzawa,
and Ignacio de la Torre. 2015. First GIS analysis of modern stone tools used
after the advent of cooking). A similar critique has been leveled by wild chimpanzees (Pan troglodytes verus) in Bossou, Guinea, West
at this assumption by researchers who note the very low fat Africa. PLoS ONE 10(3):e0121613.
content of wild game meat and the metabolic difficulties that Binford, L. R. 1981. Bones: ancient men and modern myths. New York: Academic.
———. 1988. Fact and fiction about the Zinjanthropus floor: data, arguments,
overconsumption would convey to hominin consumers (Speth and interpretations. Current Anthropology 29:123–135. [MP]
2010). The low profitability of meat is a critical point because its Biro, D., S. Carvalho, and T. Matsuzawa. 2010. Tools, traditions, and tech-
purported superiority in terms of “dietary quality” is perhaps nologies: interdisciplinary approaches to chimpanzee nut-cracking. In The
mind of the chimpanzee: ecological and experimental perspectives. Elizabeth
one of the most dearly held stances in paleoanthropology. V. Lonsdorf, Stephen R. Ross, and Tetsuro Matsuzawa, eds. Pp. 141–155.
Wrangham even suggests that early cutting tools were used Chicago: University of Chicago Press. [DRB]
primarily as a means of removing the largely useless meat in Blumenschine, R. J. 1986. Carcass consumption sequences and the archaeo-
logical distinction of scavenging and hunting. Journal of Human Evolution
order to access fat-rich bones and organs. 15:639–659.
Using a similar energetic argument, Braun notes that not all ———. 1989. A landscape taphonomic model of the scale of prehistoric
small prey are fat poor and fast moving. They can also be scavenging opportunities. Journal of Human Evolution 18:345–371.
———. 1995. Percussion marks, tooth marks, and experimental determi-
subject to seasonality effects in both their costs and returns. nations of the timing of hominid and carnivore access to long bones at
Specifically, aquatic resources are more easily exploited during FLK Zinjanthropus, Olduvai Gorge, Tanzania. Journal of Human Evolution
dry seasons, when terrestrial animals are most stressed and 29:21–51. [MP]
Blumenschine, R. J., J. A. Cavallo, and S. D. Capaldo. 1994. Competition for
their fat content lowest. Similarly, Wrangham notes that brain carcasses and early hominid behavioral ecology: a case study and concep-
experiences much less seasonal fluctuation in fat content than tual framework. Journal of Human Evolution 27(1):197–213. [DRB]
does marrow, although it occurs in overall smaller quantities Blumenschine, R. J., and T. C. Madrigal. 1993. Variability in long bone
marrow yields of East African ungulates and its zooarchaeological impli-
(Lupo 1998). Together, these arguments reinforce comments cations. Journal of Archaeological Science 20:555–587.
by Werdelin that it was the generalist and flexible dietary Blumenschine, R. J., and M. M. Selvaggio. 1988. Percussion marks on bone
strategy of early hominins—although we add to this technol- surfaces as a new diagnostic of hominid behaviour. Nature 333:763–765. [MP]
Blumenschine, Robert J., Henry T. Bunn, Valerius Geist, Fumiko Ikawa-
ogy and culture—that afforded them their evolutionary edge. Smith, Curtis W. Marean, Anthony G. Payne, John Tooby, and Nikolaas J.
It is perhaps this very complexity that has made it so difficult van der Merwe. 1987. Characteristics of an early hominid scavenging niche.
for paleoanthropologists to model—or indeed even agree Current Anthropology 28:383–407, comments and reply.
Blumenschine, Robert J., Curtis W. Marean, and Salvatore D. Capaldo. 1996.
upon a definition for—the origins of the HPP. With a suite of Blind tests of inter-analyst correspondence and accuracy in the identifi-
emerging tool kits and interdisciplinary intersections of data, cation of cut marks, percussion marks, and carnivore tooth marks on bone
we are optimistic that Pliocene deposits will soon begin to surfaces. Journal of Archaeological Science 23:493–507.
Blumenschine, Robert J., Ian G. Stanistreet, Jackson K. Njau, Marion K.
divulge their more subtle traces of this behavior. Bamford, Fidelis T. Masao, Rosa M. Albert, Harald Stollhofen, et al. 2012.
—Jessica C. Thompson, Susana Carvalho, Environments and hominin activities across the FLK Peninsula during
Curtis W. Marean, and Zeresenay Alemseged Zinjanthropus times (1.84 Ma), Olduvai Gorge, Tanzania. Journal of Hu-
man Evolution 63:364–383.
Boesch, C., and H. Boesch. 1990. Tool use and tool making in wild chim-
panzees. Folia Primatologica 54:86–99.
Brain, C. K. 1981. The hunters or the hunted? an introduction to African cave
References Cited taphonomy. Chicago: University of Chicago Press.
Braun, D. R., J. W. K. Harris, and D. N. Maina. 2009. Oldowan raw material
Aiello, Leslie C., and Susan C. Antón. 2012. Human biology and the origins of procurement and use: evidence from the Koobi Fora Formation. Archaeo-
Homo: an introduction to supplement 6. Current Anthropology 53(suppl. metry 51(1):26–42.
6):S269–S277. Braun, David R., John W. K. Harris, Naomi E. Levin, Jack T. McCoy, Andy
Alemseged, Zeresenay. 2015. Stable isotopes serving as a checkpoint. Pro- I. R. Herries, Marion K. Bamford, Laura C. Bishop, Brian G. Richmond,
ceedings of the National Academy of Sciences 112(40):12232–12233. and Mzalendo Kibunjia. 2010. Early hominin diet included diverse ter-
Ambrose, Stanley H. 2001. Paleolithic technology and human evolution. restrial and aquatic animals 1.95 Ma in East Turkana, Kenya. Proceedings of
Science 291(5509):1748–1753. the National Academy of Sciences 107(22):10002–10007.
Antón, Susan C., Richard Potts, and Leslie C. Aiello. 2014. Evolution of early Brochu, Christopher A., and Glenn W. Storrs. 2012. A giant crocodile from
Homo: an integrated biological perspective. Science 345(6192):1236828. the Plio-Pleistocene of Kenya, the phylogenetic relationships of Neogene
Arcadi, Adam C., and Richard W. Wrangham. 1999. Infanticide in chim- African crocodylines, and the antiquity of Crocodylus in Africa. Journal of
panzees: review of cases and a new within-group observation from the Vertebrate Paleontology 32(3):587–602.
Kanyawara study group in Kibale National Park. Primates 40:337–351. Bunn, H. T. 1994. Early Pleistocene hominid foraging strategies along the
[RW] ancestral Omo River at Koobi Fora, Kenya. Journal of Human Evolution
Archer, W., D. R. Braun, J. W. Harris, J. T. McCoy, and B. G. Richmond. 27:247–266.
2014. Early Pleistocene aquatic resource use in the Turkana Basin. Journal Bunn, Henry T. 2007. Meat made us human. In Evolution of the human diet:
of Human Evolution 77:74–87. [DRB] the known, the unknown, and the unknowable. P. Ungar, ed. Pp. 191–211.
Behrensmeyer, Anna K., Susan M. Kidwell, and Robert A. Gastaldo. 2000. Oxford: Oxford University Press.
Taphonomy and paleobiology. Paleobiology 26(suppl. 4):103–147. [DRB] Butynski, Thomas M. 1982. Vertebrate predation by primates: a review of
Behrensmeyer, Anna K., and Kaye E. Reed. 2013. Reconstructing the habitats hunting patterns and prey. Journal of Human Evolution 11(5):421–430.
of Australopithecus: paleoenvironments, site taphonomy, and faunas. In Capaldo, S. D. 1997. Experimental determinations of carcass processing by Plio-
The paleobiology of Australopithecus. K. E. Reed, J. G. Fleagle, and R. E. Pleistocene hominids and carnivores at FLK 22 (Zinjanthropus), Olduvai
Leakey, eds. Pp. 41–60. Dordrecht: Springer. Gorge, Tanzania. Journal of Human Evolution 33:555–597. [MP]

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 000 Current Anthropology Volume 60, Number 1, February 2019

Capaldo, S. D., and R. J. Blumenschine. 1994. A quantitative diagnosis of Domínguez-Rodrigo, M., and L. Alcalá. 2016. 3.3-million-year-old stone tools
notches made by hammerstone percussion and carnivore gnawing on bovid and butchery traces? more evidence needed. PaleoAnthropology 2016:46–
long bones. American Antiquity 59:724–748. [MP] 53.
Carbone, C., G. M. Mace, S. C. Roberts, and D. W. Macdonald. 1999. Ener- Domínguez-Rodrigo, M., R. Barba, and Charles P. Egeland. 2007. Decon-
getic constraints on the diet of terrestrial carnivores. Nature 402:286–288. structing Olduvai: a taphonomic study of the Bed I sites. Dordrecht:
[LW] Springer.
Carmody, Rachel N., and Richard W. Wrangham. 2009. The energetic sig- Domínguez-Rodrigo, M., H. T. Bunn, A. Z. P. Mabulla, E. Baquedano, D.
nificance of cooking. Journal of Human Evolution 57(4):379–391. Uribelarrea, A. Pérez-González, A. Gidna, et al. 2014. On meat eating and
Caruana, Matthew V., Susana Carvalho, David R. Braun, Darya Presnyakova, human evolution: a taphonomic analysis of BK4b (Upper Bed II, Olduvai
Michael Haslam, Will Archer, Rene Bobe, and John W. K. Harris. 2014. Gorge, Tanzania), and its bearing on hominin megafaunal consumption.
Quantifying traces of tool use: a novel morphometric analysis of damage Quaternary International 322:129–152.
patterns on percussive tools. PLoS ONE 9(11):e113856. Domínguez-Rodrigo, M., S. de Juana, A. B. Galán, and M. Rodríguez. 2009. A
Carvalho, Susana, Dora Biro, Eugénia Cunha, Kimberley Hockings, William new protocol to differentiate trampling marks from butchery cut marks.
C. McGrew, Brian G. Richmond, and Tetsuro Matsuzawa. 2012. Chim- Journal of Archaeological Science 36(12):2643–2654.
panzee carrying behaviour and the origins of human bipedality. Current Domínguez-Rodrigo, M., and T. R. Pickering. 2017. The meat of the matter:
Biology 22(6):R180–R181. an evolutionary perspective on human carnivory. Azania 52(1):4–32.
Carvalho, Susana, Dora Biro, William C. McGrew, and Tetsuro Matsuzawa. Domínguez-Rodrigo, M., T. R. Pickering, and H. T. Bunn. 2011. Reply to
2009. Tool-composite reuse in wild chimpanzees (Pan troglodytes): ar- McPherron et al.: doubting Dikika is about data, not paradigms. Pro-
chaeologically invisible steps in the technological evolution of early ho- ceedings of the National Academy of Sciences 108(21):E117–E117. [DRB]
minins? Animal Cognition 12(1):103–114. Domínguez-Rodrigo, M., T. Pickering, and H. T. Bunn. 2012. Experimental
Carvalho, Susana, Eugénia Cunha, Cláudia Sousa, and Tetsuro Matsuzawa. study of cut marks made with rocks unmodified by human flaking and its
2008. Chaînes opératoires and resource-exploitation strategies in chimpan- bearing on claims of ∼3.4-million-year-old butchery evidence from Dikika,
zee (Pan troglodytes) nut cracking. Journal of Human Evolution 55(1):148– Ethiopia. Journal of Archaeological Science 39:205–214.
163. Domínguez-Rodrigo, M., T. R. Pickering, S. Semaw, and M. J. Rogers. 2005.
Cavallo, J. A., and R. J. Blumenschine. 1989. Tree-stored leopard kills: Cutmarked bones from Pliocene archaeological sites at Gona, Afar,
expanding the hominid scavenging niche. Journal of Human Evolution Ethiopia: Implications for the function of the world’s oldest stone tools.
18:393–399. Journal of Human Evolution 48(2):109–121.
Cerling, Thure E., Fredrick Kyalo Manthi, Emma N. Mbua, Louise N. Leakey, Domínguez-Rodrigo, M., and J. Yravedra. 2009. Why are cut mark frequen-
Meave G. Leakey, Richard E. Leakey, Francis H. Brown, Frederick E. Grine, cies in archaeofaunal assemblages so variable? a multivariate analysis.
John A. Hart, and Prince Kaleme. 2013. Stable isotope-based diet recon- Journal of Archaeological Science 36:884–894.
structions of Turkana Basin hominins. Proceedings of the National Acad- Domínguez-Rodrigo, Manuel. 2001. A study of carnivore competition in ri-
emy of Sciences 110(26):10501–10506. parian and open habitats of modern savannas and its implications for
Clark, J. L., and I. Plug. 2008. Animal exploitation strategies during the south hominid behavioral modelling. Journal of Human Evolution 40(2):77–98.
African middle stone Age: Howiesons Poort and post-Howiesons Poort Domínguez-Rodrigo, Manuel, and Enrique Baquedano. 2018. Distinguishing
fauna from Sibudu Cave. Journal of Human Evolution 54(6):886–898. butchery cut marks from crocodile bite marks through machine learning
[DRB] methods. Scientific Reports 8(1):5786.
Clavel, Joanne, Romain Julliard, and Vincent Devictor. 2010. Worldwide Domínguez-Rodrigo, Manuel, Palmira Saladié, Isabel Cáceres, Rosa Huguet,
decline of specialist species: toward a global functional homogenization? José Yravedra, Antonio Rodríguez-Hidalgo, Patricia Martín, et al. 2017. Use
Frontiers in Ecology and the Environment 9(4):222–228. and abuse of cut mark analyses: the Rorschach effect. Journal of Archae-
Codding, Brian F., and Douglas W. Bird. 2015. Behavioral ecology and the ological Science 86:14–23.
future of archaeological science. Journal of Archaeological Science 56:9–20. Dunbar, Robin. 2016. Human evolution: our brains and our behavior. New
Dart, R. 1953. The predatory transition from ape to man. International An- York: Oxford University Press. [JJS]
thropological and Linguistic Review 1:201–218. Dunham, K. M., and Martyn G. Murray. 1982. The fat reserves of impala,
Davidson, Ian, and Sue Solomon. 1990. Was OH 7 the victim of a crocodile Aepyceros melampus. African Journal of Ecology 20:81–87. [RW]
attack? In Problem solving in taphonomy: archaeological and palaeontological Fahy, Geraldine E., Michael Richards, Julia Riedel, Jean-Jacques Hublin, and
studies from Europe, Africa and Oceania, vol. 2. S. Solomon, I. Davidson, and Christophe Boesch. 2013. Stable isotope evidence of meat eating and
D. Watson, eds. St. Lucia: Tempus. hunting specialization in adult male chimpanzees. Proceedings of the Na-
DeCasien, Alex R., Scott A. Williams, and James P. Higham. 2017. Primate tional Academy of Sciences 110(15):5829–5833.
brain size is predicted by diet but not sociality. Nature Ecology & Evolution Ferraro, J. V., K. M. Binetti, L. A. Wiest, D. Esker, L. E. Baker, and S. L.
1:0112. Forman. 2018. Contesting early archaeology in California. Nature 554
de Heinzelin, J., J. D. Clark, T. D. White, W. K. Hart, P. Reene, G. (7691):E1. [DRB]
WoldeGabriel, Y. Beyene, and E. Vrba. 1999. Environment and behavior of Ferraro, Joseph V., Thomas W. Plummer, Briana L. Pobiner, James S. Oliver,
2.5-million-year-old Bouri hominids. Science 284:625–635. Laura C. Bishop, David R. Braun, Peter W. Ditchfield, et al. 2013. Earliest
Delagnes, Anne, and Hélène Roche. 2005. Late Pliocene hominid knapping archaeological evidence of persistent hominin carnivory. PLoS ONE 8(4):
skills: the case of Lokalalei 2C, West Turkana, Kenya. Journal of Human e62174.
Evolution 48(5):435–472. Ferretti, Marco P. 2007. Evolution of bone-cracking adaptations in hyaenids
de la Torre, I., A. Benito-Calvo, A. Arroyo, A. Zupancich, and T. Proffitt. (Mammalia, Carnivora). Swiss Journal of Geosciences 100(1):41–52. [LW]
2012. Experimental protocols for the study of battered stone anvils from Gaillard, Claire, Mukesh Singh, Anne Dambricourt Malassé, Vipnesh Bhard-
Olduvai Gorge (Tanzania). Journal of Archaeological Science 40(1):313–332. waj, Baldev Karir, Amandeep Kaur, Surinder Pal, et al. 2016. The lithic
[DRB] industries on the fossiliferous outcrops of the Late Pliocene Masol For-
Dibble, Harold L., and Zeljko Rezek. 2009. Introducing a new experimental mation, Siwalik Frontal Range, northwestern India (Punjab). Comptes Ren-
design for controlled studies of flake formation: results for exterior platform dus Palevol 15(3):341–357.
angle, platform depth, angle of blow, velocity, and force. Journal of Ar- Galef, B. G., Jr. 1996. Food selection: problems in understanding how we
chaeological Science 36:1945–1954. choose foods to eat. Neuroscience and Biobehavioral Reviews 20(1):67–73.
Diez-Martín, Fernando, Policarpo Sánchez, Manuel Domínguez-Rodrigo, [EV]
Audax Mabulla, and Rebeca Barba. 2009. Were Olduvai Hominins making Gettler, Lee T. 2010. Direct male care and hominin evolution: why male–child
butchering tools or battering tools? analysis of a recently excavated lithic interaction is more than a nice social idea. American Anthropologist 112
assemblage from BK (Bed II, Olduvai Gorge, Tanzania). Journal of An- (1):7–21.
thropological Archaeology 28(3):274–289. Goodall, Jane. 1986. The chimpanzees of Gombe: patterns of behavior. Cam-
Domínguez-Rodrigo, M. 2009. Are all Oldowan Sites palimpsests? if so, what bridge, MA: Harvard University Press. [RW]
can they tell us about Hominid carnivory? In Interdisciplinary approaches Grabowski, Mark, Kevin G. Hatala, William L. Jungers, and Brian G. Rich-
to the Oldowan. Erella Hovers and David R. Braun, eds. Pp. 129–147. mond. 2015. Body mass estimates of hominin fossils and the evolution of
Dordrecht: Springer. [DRB] human body size. Journal of Human Evolution 85:75–93.

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
Thompson et al. Origins of Large-Animal Exploitation 000

Gumert, M. D., and K. S. Malaivijitnond. 2009. The physical characteristics Lonsdorf, Elizabeth V. 2006. What is the role of mothers in the acquisition of
and usage patterns of stone axe and pounding hammers used by long-tailed termite-fishing behaviors in wild chimpanzees (Pan troglodytes schwein-
macaques in the Andaman Sea region of Thailand. American Journal of furthii)? Animal Cognition 9(1):36–46.
Primatology 71(7):594–608. Luncz, L. V., R. Mundry, and C. Boesch. 2012. Evidence for cultural differ-
Hardy, Karen, Jennie Brand-Miller, Katherine D. Brown, Mark G. Thomas, ences between neighboring chimpanzee communities. Current Biology
and Les Copeland. 2015. The importance of dietary carbohydrate in human 22:922–926. [DRB]
evolution. Quarterly Review of Biology 90(3):251–268. Luncz, Lydia V., Tomos Proffitt, Lars Kulik, Michael Haslam, and Roman M.
Harmand, Sonia, Jason E. Lewis, Craig S. Feibel, Christopher J. Lepre, Wittig. 2016. Distance-decay effect in stone tool transport by wild chim-
Sandrine Prat, Arnaud Lenoble, Xavier Boës, et al. 2015. 3.3-million-year- panzees. Philosophical Transactions of the Royal Society of London B: Bio-
old stone tools from Lomekwi 3, West Turkana, Kenya. Nature 521(7552):310– logical Sciences 283(1845).
315. Luncz, Lydia V., Roman M. Wittig, and Christophe Boesch. 2015. Primate
Harris, Jacob A., Curtis W. Marean, Kiona Ogle, and Jessica Thompson. 2017. archaeology reveals cultural transmission in wild chimpanzees (Pan
The trajectory of bone surface modification studies in paleoanthropology troglodytes verus). Philosophical Transactions of the Royal Society of London
and a new Bayesian solution to the identification controversy. Journal of B: Biological Sciences 370(1682).
Human Evolution 110:69–81. Lupo, K. 2012. On early hominin meat-eating and carcass acquisition strat-
Haslam, M. 2014. On the tool use behavior of the bonobo-chimpanzee last egies: still relevant after all these years? In Stone tools and fossil bones:
common ancestor, and the origins of hominine stone tool use. American debates in the archaeology of human origins. M. Domínguez-Rodrigo, ed.
Journal of Primatology 76(10):910–918. [DRB] Pp. 115–151. Cambridge: Cambridge University Press.
Haslam, M., M. D. Gumert, D. Biro, S. Carvalho, and S. Malaivijitnond. 2013. Lupo, K. D. 1998. Experimentally derived extraction rates for marrow: im-
Use-wear patterns on wild macaque stone tools reveal their behavioural plications for body part exploitation strategies of Plio-Pleistocene Hominid
history. PLoS ONE 8(8):e72872. [DRB] scavengers. Journal of Archaeological Science 25(7):657–676.
Haslam, M., A. Hernandez-Aguilar, V. Ling, S. Carvalho, I. de la Torre, A. Madrigal, T. C., and R. J. Blumenschine. 2000. Preferential processing of high
DeStefano, A. Du, et al. 2009. Primate archaeology. Nature 460:339–344. return rate marrow bones by Oldowan hominids: a comment on Lupo.
Hawkes, K. 1990. Why do men hunt? some benefits for risky choices. In Risk Journal of Archaeological Science 27:739–741. [DRB]
and uncertainty in tribal and peasant economies. E. Cashdan, ed. Pp. 145– Malassé, Anne Dambricourt, Anne-Marie Moigne, Mukesh Singh, Thomas
166. Boulder, CO: Westview. [DRB] Calligaro, Baldev Karir, Claire Gaillard, Amandeep Kaur, Vipnesh Bhard-
Hawkes, K., J. F. O’Connell, and N. G. B. Jones. 2014. More lessons from the waj, Surinder Pal, and Salah Abdessadok. 2016. Intentional cut marks on
Hadza about men’s work. Human Nature 25(4):596–619. [DRB] bovid from the Quranwala zone, 2.6 Ma, Siwalik Frontal Range, north-
Hill, K. 1988. Macronutrient modifications of optimal foraging theory: an western India. Comptes Rendus Palevol 15(3):317–339.
approach using indifference curves applied to some modern foragers. Marchant, Linda F, and William C McGrew. 2005. Percussive technology:
Human Ecology 16:157–197. chimpanzee baobab smashing and the evolutionary modeling of hominid
Hockings, Kimberley J., Matthew R. McLennan, Susana Carvalho, Marc knapping. In Stone knapping the necessary conditions for an uniquely
Ancrenaz, Rene Bobe, Richard W. Byrne, Robin I. M. Dunbar, Tetsuro hominin behaviour. Blandine Bril and Valentine Roux, eds. Pp. 341–350.
Matsuzawa, William C. McGrew, and Elizabeth A. Williamson. 2015. Apes McDonald Institute Monograph Series. Cambridge: McDonald Institute for
in the Anthropocene: flexibility and survival. Trends in Ecology and Evo- Archaeological Research.
lution 30(4):215–222. Marean, C. W., and L. Bertino. 1994. Intrasite spatial analysis of bone:
Hohmann, G., and B. Fruth. 2008. New records on prey capture and meat subtracting the effect of secondary carnivore consumers. American Antiq-
eating by bonobos at Lui Kotale, Salonga National Park, Democratic Re- uity 59:748–768.
public of Congo. Folia Primatologica 79(2):103–110. Marean, C. W., and C. E. Ehrhardt. 1995. Paleoanthropological and paleo-
Isler, Karin, E., Christopher Kirk, Joseph M. A. Miller, Gene A. Albrecht, ecological implications of the taphonomy of a sabertooth’s lair. Journal of
Bruce R. Gelvin, and Robert D. Martin. 2008. Endocranial volumes of Human Evolution 28:515–547.
primate species: scaling analyses using a comprehensive and reliable data Markham, A. Catherine, Laurence R. Gesquiere, Susan C. Alberts, and Jeanne
set. Journal of Human Evolution 55:967–978. [RW] Altmann. 2015. Optimal group size in a highly social mammal. Proceedings
James, E. 2018. The relevance of basic science to stone tool use and meat of the National Academy of Sciences 112(48):14882–14887.
eating: standardisation of bone surface modification methods and inter- McGrew, W. C. 2010. In search of the last common ancestor: new findings on
pretations. PhD dissertation, University of Queensland, Brisbane. wild chimpanzees. Philosophical Transactions of the Royal Society of Lon-
James, Emma C., and Jessica C. Thompson. 2015. On bad terms: problems don B: Biological Sciences 365(1556):3267–3276.
and solutions within zooarchaeological bone surface modification studies. McGrew, William C. 2015. Why don’t chimpanzees eat monitor lizards?
Environmental Archaeology 20(1):89–103. African Primates 10:41–52.
Johanson, Donald C. 2004. Lucy, thirty years later: an expanded view of McPherron, Shannon P., Zeresenay Alemseged, Curtis Marean, Jonathan G.
Australopithecus afarensis. Journal of Anthropological Research 60(4):465– Wynn, Denné Reed, Denis Geraads, René Bobe, and Hamdallah Béarat.
486. 2011. Tool-marked bones from before the Oldowan change the paradigm.
Josephs-Spaulding, Jonathan, Erik Beeler, and Om V. Singh. 2016. Human Proceedings of the National Academy of Sciences 108:E116. [DRB]
microbiome versus food-borne pathogens: friend or foe. Applied Microbi- McPherron, Shannon P., Zeresenay Alemseged, Curtis W. Marean, Jonathan G.
ology and Biotechnology 100(11):4845–4863. Wynn, Denné Reed, Denis Geraads, René Bobe, and Hamdallah A. Béara.
Kimbel, B. 2009. The origin of Homo. In The first humans: origin and early 2010. Evidence for stone-tool-assisted consumption of animal tissues before
evolution of the genus Homo. F. E. Grine, J. G. Fleagle, and R. E. Leakey, eds. 3.39 million years ago at Dikika, Ethiopia. Nature 466(7308):857–860.
Pp. 31–37. New York: Springer. Mercader, Julio, Melissa Panger, and Christophe Boesch. 2002. Excavation of a
Kimbel, William H., and Brian Villmoare. 2016. From Australopithecus to chimpanzee stone tool site in the African rainforest. Science 296(5572):1452–
Homo: the transition that wasn’t. Philosophical Transactions of the Royal 1455.
Society B: Biological Sciences 371(1698). Merritt, S. R. 2015. Cut mark cluster geometry and equifinality in replicated Early
Kitahara-Frisch, J., K. Norikoshi, and K. Hara. 1987. Use of a bone fragment Stone Age butchery. International Journal of Osteoarchaeology 26:585–598.
as a step towards secondary tool use in captive chimpanzee. Primate Rep Mitani, J. C., and D. P. Watts. 1999. Demographic influences on the hunt-
18:33–37. ing behavior of chimpanzees. American Journal of Physical Anthropology
Langdon, J. 2006. Has an aquatic diet been necessary for hominin brain evo- 9:439–454. [DRB]
lution and functional development? British Journal of Nutrition 96:7–17. Mithen, S. J. 2005. The singing Neanderthal: the origins of music, language,
Lee-Thorp, Julia A., Matt Sponheimer, Benjamin H. Passey, Darryl J. de mind and body. London: Weidenfield & Nicholson. [JJS]
Ruiter, and Thure E. Cerling. 2010. Stable isotopes in fossil hominin tooth Mora, Rafael, and Ignacio de la Torre. 2005. Percussion tools in Olduvai Beds
enamel suggest a fundamental dietary shift in the Pliocene. Philosophical I and II (Tanzania): implications for early human activities. Journal of
Transactions of the Royal Society B: Biological Sciences 365(1556):3389– Anthropological Archaeology 24(2):179–192.
3396. Morin, E. 2007. Fat composition and Nunamiut decision-making: a new look
Lewis, M. E. 1997. Carnivoran paleoguilds of Africa: implications for hominid at the marrow and bone grease indices. Journal of Archaeological Science
food procurement strategies. Journal of Human Evolution 32:257–288. 34:69–82.

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 000 Current Anthropology Volume 60, Number 1, February 2019

Mulcahy, Nicholas J., and Josep Call. 2006. Apes save tools for future use. natural tools for extractive foraging by bonobos (Pan paniscus). American
Science 312(5776):1038–1040. Journal of Physical Anthropology 158(1):78–91.
Murphy, Suzanne P., and Lindsay H. Allen. 2003. Nutritional importance of Rogers, Michael J., John W. K. Harris, and Craig S. Feibel. 1994. Changing
animal source foods. Journal of Nutrition 133(11):3932S–3935S. patterns of land use by Plio-Pleistocene hominids in the Lake Turkana
Njau, Jackson. 2012. Reading Pliocene bones. Science 336(6077):46–47. Basin. Journal of Human Evolution 27(1–3):139–158. [DRB]
Njau, Jackson K., and William Gilbert. 2016. Standardizing terms for crocodile- Rogers, Michael J., and Sileshi Semaw. 2009. From nothing to something: the
induced bite marks on bone surfaces in light of the frequent bone modifi- appearance and context of the earliest archaeological record. In Sourcebook
cation equifinality found to result from crocodile feeding behavior, stone tool of paleolithic transitions. M. Camps and P. Chauhan, eds. Pp. 155–171.
modification, and trampling. Forensic Osteology Occasional Publications 3:1– New York: Springer.
13. [MP] Rosell, Jordi, Ruth Blasco, Gerard Campeny, J Carlos Díez, Rodrigo Alonso
Oliver, J. S. 1994. Estimates of hominid and carnivore involvement in the Flk Alcalde, Leticia Menéndez, Juan Luis Arsuaga, José M Bermúdez de Castro,
Zinjanthropus fossil assemblage: some socioecological implications. Journal and Eudald Carbonell. 2011. Bone as a technological raw material at the
of Human Evolution 27:267–294. [DRB] Gran Dolina site (Sierra de Atapuerca, Burgos, Spain). Journal of Human
Otárola-Castillo, E., M. Torquato, H. Hawkins, E. James, J. Harris, C. W. Evolution 61(1):125–131.
Marean, S. P. McPherron, and J. C. Thompson. 2018. Differentiating be- Rothman, Jessica M., David Raubenheimer, Margaret A. H. Bryer, Maressa
tween cutting actions on bone using 3D geometric morphometrics and Takahashi, and Christopher C. Gilbert. 2014. Nutritional contributions of
Bayesian analyses with implications to human evolution. Journal of Ar- insects to primate diets: implications for primate evolution. Journal of
chaeological Science 89:56–67. Human Evolution 71:59–69.
Panger, M. A., A. S. Brooks, B. G. Richmond, and B. Wood. 2002. Older than Rowan, John, and Kaye Reed. 2015. The paleoclimatic record and Plio-
the Oldowan? rethinking the emergence of hominin tool use. Evolutionary Pleistocene paleoenvironments. In Handbook of paleoanthropology. W.
Anthropology 11(6):235–245. Henke and I. Tattersall, eds. Pp. 465–491. New York: Springer.
Pante, M. C., M. Muttart, T. Keevil, R. J. Blumenschine, J. K. Njau, and S. M. Sahle, Yonatan, Sireen El Zaatari, and Tim D. White. 2017. Hominid butchers
Merritt. 2017. A new high-resolution 3-D quantitative method for identi- and biting crocodiles in the African Plio-Pleistocene. Proceedings of the
fying bone surface modifications with implications for the Early Stone Age National Academy of Sciences 114(50):13164–13169.
archaeological record. Journal of Human Evolution 102:1–11. [MP] Sayers, K., and C. O. Lovejoy. 2014. Blood, bulbs, and bunodonts: on evo-
Pante, Michael C., Robert S. Scott, Robert J. Blumenschine, and Salvatore D. lutionary ecology and the diets of Ardipithecus, Australopithecus, and early
Capaldo. 2015. Revalidation of bone surface modification models for in- Homo. Quarterly Review of Biology 89(4):319–357.
ferring fossil hominin and carnivore feeding interactions. Quaternary In- Selvaggio, M. M. 1998. Evidence for a three-stage sequence of hominid and
ternational 355:164–168. carnivore involvement with long bones at FLK Zinjanthropus, Olduvai
Parkinson, Jennifer A., Thomas Plummer, and Adam Hartstone-Rose. 2015. Gorge, Tanzania. Journal of Archaeological Science 25:191–202. [MP]
Characterizing felid tooth marking and gross bone damage patterns using Shea, John J. 2017a. Occasional, obligatory, and habitual stone tool use in
GIS image analysis: an experimental feeding study with large felids. Journal hominin evolution. Evolutionary Anthropology 26(5):200–217.
of Human Evolution 80:114–134. Shea, John J. 2017b. Stone tools in human evolution: behavioral differences
Plummer, T. W., and L. C. Bishop. 1994. Hominid paleoecology at Olduvai among technological primates. New York: Cambridge University Press. [JJS]
Gorge, Tanzania as indicated by antelope remains. Journal of Human Shott, M. J., and P. Sillitoe. 2005. Use life and curation in New Guinea ex-
Evolution 27:47–75. [DRB] perimental used flakes. Journal of Archaeological Science 32:653–663.
Plummer, Thomas W., Peter W. Ditchfield, Laura C. Bishop, John D. [DRB]
Kingston, Joseph V. Ferraro, David R. Braun, Fritz Hertel, and Richard Smith, Alex R., Rachel N. Carmody, Rachel J. Dutton, and Richard W.
Potts. 2009. Oldest evidence of toolmaking hominins in a grassland- Wrangham. 2015. The significance of cooking for early hominin scaveng-
dominated ecosystem. PLoS ONE 4(9):e7199. ing. Journal of Human Evolution 84:62–70.
Pobiner, Briana L. 2015. New actualistic data on the ecology and energetics of Speth, John D. 1989. Early hominid hunting and scavenging: the role of meat
hominin scavenging opportunities. Journal of Human Evolution 80:1–16. as an energy source. Journal of Human Evolution 18(4):329–343.
Potts, Richard. 1989. Olorgesailie: new excavations and findings in Early and ———. 2010. Were big-game hunters targeting fat? In The paleoanthropology
Middle Pleistocene contexts, southern Kenya Rift Valley. Journal of Human and archaeology of big-game hunting. John D. Speth, ed. Pp. 87–107. New
Evolution 18:477–484. [DRB] York: Springer.
———. 2013. Hominin evolution in settings of strong environmental vari- Sponheimer, Matt, Zeresenay Alemseged, Thure E. Cerling, Frederick E.
ability. Quaternary Science Reviews 73:1–13. Grine, William H. Kimbel, Meave G. Leakey, Julia A. Lee-Thorp, et al.
Potts, Richard, and J. Tyler Faith. 2015. Alternating high and low climate var- 2013. Isotopic evidence of early hominin diets. Proceedings of the National
iability: the context of natural selection and speciation in Plio-Pleistocene Academy of Sciences 110(26):10513–10518.
hominin evolution. Journal of Human Evolution 87:5–20. Stammati, M., G. Sabbatini, and E. Visalberghi. 2008. An experimental
Prieto, J. 2013. Hunting and vertebrate consumption in wild bearded capu- analysis of ingestion rates in an omnivorous species. American Journal of
chin monkeys (Sapajus libidinosus) from Fazenda Boa Vista, Brazil: an Primatology 70(5):510–513. [EV]
evolutionary approach. Master’s thesis, Universidade de Coimbra. [EV] Stanford, Craig B. 1995. Chimpanzee hunting behavior and human evolution.
Proffitt, Tomos, Lydia V. Luncz, Tiago Falótico, Eduardo B. Ottoni, Ignacio American Scientist 83:256–261.
de la Torre, and Michael Haslam. 2016. Wild monkeys flake stone tools. ———. 1996. The hunting ecology of wild chimpanzees: implications for
Nature 539:85–88. the evolutionary ecology of Pliocene hominids. American Anthropologist
Pruetz, Jill D., and Paco Bertolani. 2007. Savanna chimpanzees, Pan troglo- 98(1):96–113.
dytes verus, hunt with tools. Current Biology 17(5):412–417. ———. 1998. Chimpanzee and red colobus: the ecology of predator and prey.
Ragir, Sonia, Martin Rosenberg, and Philip Tierno. 2000. Gut morphology Cambridge, MA: Harvard University Press. [RW]
and the avoidance of carrion among chimpanzees, baboons, and early ———. 2001. A comparison of social meat-foraging by chimpanzees and
hominids. Journal of Anthropological Research 56(4):477–512. human foragers. In Meat-eating and human evolution. C. B. Stanford and
Rezek, Z., S. Lin, R. Iovita, and H. L. Dibble. 2011. The relative effects of core H. T. Bunn, eds. Pp. 122–140. Oxford: Oxford University Press.
surface morphology on flake shape and other attributes. Journal of Ar- ———. 2012. Chimpanzees and the behavior of Ardipithecus ramidus. An-
chaeological Science 38:1346–1359. nual Review of Anthropology 41:139–149.
Ripple, William J., James A. Estes, Robert L. Beschta, Christopher C. Wilmers, Stanford, Craig B., and Henry T. Bunn. 2001. Meat eating and human evo-
Euan G. Ritchie, Mark Hebblewhite, Joel Berger, et al. 2014. Status and eco- lution. New York: Oxford University Press.
logical effects of the world’s largest carnivores. Science 343(6167):1241484. Stewart, K. M. 1994. Early hominid utilization of fish resources and impli-
Robinson, J. R., J. Rowan, C. J. Campisano, J. G. Wynn, and K. E. Reed. 2017. cations for seasonality and behavior. Journal of Human Evolution 27:229–
Late Pliocene environmental change during the transition from Austra- 245. [DRB]
lopithecus to Homo. Nature Ecology and Evolution 1(6):0159. [DRB] Stewart, K. M., and D. Gifford-Gonzalez. 1994. An ethnoarchaeological
Roffman, Itai, Sue Savage-Rumbaugh, Elizabeth Rubert-Pugh, André Stadler, contribution to identifying hominid fish processing sites. Journal of Ar-
Avraham Ronen, and Eviatar Nevo. 2015. Preparation and use of varied chaeological Science 21:237–248. [DRB]

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
Thompson et al. Origins of Large-Animal Exploitation 000

Stout, Dietrich, Sileshi Semaw, Michael J. Rogers, and Dominique Cauche. Visalberghi, Elisabetta, Elsa Addessi, Valentina Truppa, Noemi Spagnoletti,
2010. Technological variation in the earliest Oldowan from Gona, Afar, Eduardo Ottoni, Patricia Izar, and Dorothy Fragaszy. 2009. Selection of
Ethiopia. Journal of Human Evolution 58(6):474–491. effective stone tools by wild bearded capuchin monkeys. Current Biology
Surbeck, Martin, and Gottfried Hohmann. 2008. Primate hunting by bonobos 19(3):213–217.
at LuiKotale, Salonga National Park. Current Biology 18(19):R906–R907. Ward, Carol V., William H. Kimbel, and Donald C. Johanson. 2011. Com-
Tappen, M. 2001. Deconstructing the Serengeti. In Meat eating and human plete fourth metatarsal and arches in the foot of Australopithecus afarensis.
evolution. C. B. Stanford and H. T. Bunn, eds. Pp. 13–31. Oxford: Oxford Science 331(6018):750–753.
University Press. Watts, David P. 2008. Scavenging by chimpanzees at Ngogo and the relevance
Tennie, Claudio, Robert C. O’Malley, and Ian C. Gilby. 2014. Why do of chimpanzee scavenging to early hominin behavioral ecology. Journal of
chimpanzees hunt? considering the benefits and costs of acquiring and Human Evolution 54(1):125–133.
consuming vertebrate versus invertebrate prey. Journal of Human Evolution Watts, David P., and John C. Mitani. 2002. Hunting behavior of chimpanzees
71:38–45. at Ngogo, Kibale national Park, Uganda. International Journal of Prima-
Thompson, J., S. Lansing, C. W. Marean, S. McPherron, and Z. Alemseged. tology 23(1):1–28.
2011. Experimental definition of bone surface signatures from natural Werdelin, Lars. 1999. Pachycrocuta (hyaenids) from the Pliocene of East
unmodified stones and implications for early hominin subsistence. Paleo- Africa. Paläontologische Zeitschrift 73(1–2):157–165.
Anthropology A38. [DRB] Werdelin, Lars, and Margaret E. Lewis. 2005. Plio-Pleistocene Carnivora of
Thompson, J. C., William H. Kimbel, Erella Hovers, and C. W. Marean. 2016. eastern Africa: species richness and turnover patterns. Zoological Journal of
New approaches to taphonomy and field survey of fossils across the Hadar the Linnean Society 144(2):121–144.
paleo-landscape at 3.3 Ma. Abstract presented at the Paleoanthropology Werdelin, Lars, and Margaret E. Lewis. 2013. Temporal change in functional
Society meeting, Atlanta, GA, April. richness and evenness in the eastern African Plio-Pleistocene carnivoran
Thompson, J. C., S. P. McPherron, René Bobe, D. Reed, W. A. Barr, J. Wynn, guild. PLoS ONE 8(3):e57944. [DRB, LW]
C. W. Marean, D. Geraads, and Z. Alemseged. 2015. Taphonomy of fossils Werdelin, Lars, Margaret E. Lewis, and Yohannes Haile-Selassie. 2014. Mid-
from the hominin-bearing deposits at Dikika, Ethiopia. Journal of Human Pliocene carnivora from the Woranso-Mille area, Afar region, Ethiopia.
Evolution 86:112–135. Journal of Mammalian Evolution 21(3):331–347. [DRB]
Toth, N. 2017. Revised taphonomic perspective on African Plio-Pleistocene Werdelin, Lars, and Nikos Solounias. 1991. The Hyaenidae: taxonomy, sys-
fauna. Proceedings of the National Academy of Sciences 114(50):13066–13077. tematics and evolution. Fossils and Strata 30:1–104. [LW]
[DRB] White, Tim D, Stanley H Ambrose, Gen Suwa, Denise F Su, David DeGusta,
Treves, Adrian, and L. Naughton-Treves. 1999. Risk and opportunity for Raymond L Bernor, Jean-Renaud Boisserie, Michel Brunet, Eric Delson,
humans coexisting with large carnivores. Journal of Human Evolution and Stephen Frost. 2009. Macrovertebrate paleontology and the Pliocene
36(3):275–282. habitat of Ardipithecus ramidus. Science 326(5949):67–93.
Turner, A. 1999. Evolution in African Plio-Pleistocene mammalian fauna: Whiten, Andrew. 2015. Experimental studies illuminate the cultural trans-
correlation and causation. In African biogeography, climate change, and mission of percussive technologies in Homo and Pan. Philosophical Trans-
early human evolution. Timothy G. Bromage and Friedemann Schrenk, eds. actions of the Royal Society of London B: Biological Sciences 370(1682):
Pp. 76–87. Oxford: Oxford University Press. 20140359.
Ugan, Andrew. 2005. Does size matter? body size, mass collecting, and their Wilfried, Ebang Ella Ghislain, and Juichi Yamagiwa. 2014. Use of tool sets by
implications for understanding prehistoric foraging behavior. American chimpanzees for multiple purposes in Moukalaba-Doudou National Park,
Antiquity 70(1):75–89. Gabon. Primates 55(4):467–472.
Ungar, P. S., F. E. Grine, and M. F. Teaford. 2006. Diet in early Homo: a Wilson, Michael L., C. Boesch, Barbara Fruth, Takeshi Furuichi, I. C. Gilby,
review of the evidence and a new model of adaptive versatility. Annual Chie Hashimoto, Catherine Hobaiter, et al. 2014. Lethal aggression in Pan
Review of Anthropology 35:209–228. [DRB] is better explained by adaptive strategies than human impacts. Nature
van der Merwe, Nikolaas J., J. Francis Thackeray, Julia A. Lee-Thorp, and Julie 513:414–417. [RW]
Luyt. 2003. The carbon isotope ecology and diet of Australopithecus afri- Wrangham, Richard. 2017. Control of fire in the Paleolithic: evaluating the
canus at Sterkfontein, South Africa. Journal of Human Evolution 44(5):581– cooking hypothesis. Current Anthropology 58(16):S303–S313. [RW]
597. Wynn, Jonathan G., Matt Sponheimer, William H. Kimbel, Zeresenay
Villmoare, Brian, William H. Kimbel, Chalachew Seyoum, Christopher J. Alemseged, Kaye Reed, Zelalem K. Bedaso, and Jessica N. Wilson. 2013.
Campisano, Erin N. DiMaggio, John Rowan, David R. Braun, J Ramón Diet of Australopithecus afarensis from the Pliocene Hadar Formation,
Arrowsmith, and Kaye E. Reed. 2015. Early Homo at 2.8 Ma from Ledi- Ethiopia. Proceedings of the National Academy of Sciences 110(26):10495–
Geraru, Afar, Ethiopia. Science 347(6228):1352–1355. 10500.
Visalberghi, E., and A. Albani. 2014. The bearded capuchin monkeys of Yravedra, José, Julia Aramendi, Miguel Ángel Maté-González, Lloyd Austin
Fazenda Boa Vista. DVD. [EV] Courtenay, and Diego González-Aguilera. 2018. Differentiating percussion
Visalberghi, E., G. Sabbatini, M. Stammati, and E. Addessi. 2003. Preferences pits and carnivore tooth pits using 3D reconstructions and geometric
towards novel foods in Cebus apella: the role of nutrients and social in- morphometrics. PLoS ONE 13(3):e0194324.
fluences. Physiology and Behavior 80(2–3):341–349. [EV] Zihlman, Adrienne L. 2012. The real females of human evolution. Evolu-
Visalberghi, E., N. Spagnoletti, E. D. Ramos da Silva, F. R. D. Andrade, E. tionary Anthropology 21(6):270–276.
Ottoni, P. Izar, and D. Fragaszy. 2009. Distribution of potential suitable Zink, Katherine D., and Daniel E. Lieberman. 2016. Impact of meat and
hammers and transport of hammer tools and nuts by wild capuchin Lower Palaeolithic food processing techniques on chewing in humans.
monkeys. Primates 50:95–104. Nature 531:500–503.

This content downloaded from 157.089.065.129 on February 05, 2019 13:30:42 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).