The evolution of human cooperation

Abstract

We argue that cooperation is instinctual. Human cooperation conferred advantages to individuals in the ancestral environment in which evolution occurred. Explanations of the evolution of cooperation for any species (human, pre-human, and non-human) have to be consistent with the biological, physiological, and environmental constraints that existed in the ancestral environment during which evolutionary selection occurred. Our explanation is consistent with: (1) the anatomical evolution of humanity; (2) the paleontological and chronological evidence; and (3) modern biology.

Full text

J Bioecon
DOI 10.1007/s10818-016-9213-z
The evolution of human cooperation
Philip R. P. Coelho1·James E. McClure1
© Springer Science+Business Media New York 2016
Abstract We argue that cooperation is instinctual. Human cooperation conferred
advantages to individuals in the ancestral environment in which evolution occurred.
Explanations of the evolution of cooperation for any species (human, pre-human, and
non-human) have to be consistent with the biological, physiological, and environmen-
tal constraints that existed in the ancestral environment during which evolutionary
selection occurred. Our explanation is consistent with: (1) the anatomical evolution of
humanity; (2) the paleontological and chronological evidence; and (3) modern biology.
Keywords Evolution ·Cooperation ·Protein and brain development
Actually evolution has no rules at all... the best thing we can say is that there are various ways to cheat,
and that some types of cheating are hard to get away with.
Michael Ghiselin (1974, p. 41)
The Mecca of the economist lies in economic biology rather than in economic dynamics.
Alfred Marshall (1890, 1949, p. xiv)
1 Introduction
The evolution of human cooperation is one of the most extensively researched and
debated topics in the social sciences.1Contra Marshall, economists have sought expla-
1A Google Scholar search of “the evolution of cooperation” produced 2.58 million hits on October
13, 2015. By contrast, a search for the term “ego” produced only 1.58 million hits; “the evolution of
man” had 3.58 million; and “human psychology” had 3 million.
BPhilip R. P. Coelho
00prcoelho@bsu.edu
1Ball State University, Muncie, USA
123
Author's personal copy

P. R. P. Coelho, J. E. McClure
nations for human cooperation in game theoretics where the pursuit of mathematical
consistency, elegance, and generality dominate the insights provided by: evolutionary
biology, anthropology, paleontology, history, physiology, and psychology. Eschewing
Ghiselin’s precept, economics has embraced rules and authorities in its explorations
of cooperation, with Prisoners’ Dilemma (PD) models (and related game theoretic
models) dominating its intellectual landscape.
In the PD models non-cooperation (defection) is the dominant (predicted) solution.
Yet in experimental settings the results continually confound theoretical predictions;
levels of cooperation observed substantially exceed predicted levels. A recent exam-
ple in the long-list of the failure of defection to dominate in PD experiments is the
work of Khadjavi and Lange (2013, p. 166); they found levels of cooperation among
their subjects that varied from 37 to 55.6 %.2Because PD theory predicts that levels
of cooperation should approach zero, the large and statistically significant levels of
observed cooperation suggest that something is fundamentally amiss.3The empiri-
cal failures of the PD in the initial straight-forward experiments led to a substantial
“refinements” literature, “refining” the basic PD model in attempts to obtain results that
were consistent with both the dominant solution and experimental evidence. Failing to
achieve these results, “the refinements literature is currently out of fashion.”4Group-
selection games and simulation studies of interactions/competitions within and across
groups made-up of stylized individuals are common among current explanations for
the evolution of cooperation.
While admitting the intellectual efforts that have gone into the creation of group-
selection models/simulations, evolutionary biologists and evolutionary psychologists
tend to reject group-selection for basically two reasons: (1) while plausible in the
abstract, group selection is empirically implausible: “Except in theoretically possible
but empirically unlikely circumstances in which groups bud off new groups faster
than their members have babies, any genetic tendency to risk life and limb that results
in a net decrease in individual inclusive fitness will be relentlessly selected against.”
Pinker (2012, p. 5); and (2) the conditions under which group selection could work, as
Ghiselin (2016) states, are: “stringent” and “so it [group selection] tends to be screened
out on the basis of parsimony.”5
2The experiment’s subjects were two groups of women: (1) university-level students, and (2) literally
incarcerated female prisoners who cooperated at higher levels than the students.
3Among the first recorded examples of cooperation contradicting theoretical priors of non-cooperation
were the unexpected levels of cooperation in PD experiments at RAND in the early 1950s (see Flood 1958).
In other experiments TIT-FOR-TAT dominated non-cooperative strategies in tournaments (see Axelrod
1984). “Refinements” to PD games spawned a substantial literature in an attempt to resolve unexpected
levels of cooperation (see Mailath 1998). In an extensive study Camer and Casari (2009) again found
surprising levels of cooperation in various experimental settings.
4“Originally, this (‘refinements’) literature was driven by the hope that theorists could identify the unique
“right” equilibrium. …we now understand that that hope, in principle, could neverbe met. … The refinements
literature is currently out of fashion because there were too many papers in which one example suggested a
minor modification of an existing refinement and no persuasive general refinement theory emerged. There
is a danger that evolutionary game theory could end up like refinements.” Mailath (1998, p. 1372)
5Group-selection remains a topic promoted primarily by non-biologists. The supposedly altruistic behavior
of individual organisms when investigated has found net benefits to the “altruistic” individual (Wessells
and Hopson 1988, p. 1048). This echoes Ghiselin’s prescient comments: “The real matter . . . is how the
123
Author's personal copy

The evolution of human cooperation
Rather than trying to explain the evolution of human cooperation by further modi-
fications of the PD model, or by invoking group-selection models,6we explain it from
biological and evolutionary perspectives; specifically we argue that: (1) cooperation is
instinctual in human beings as it is in all other social animals; (2) explanations of the
evolution of cooperation for any species (human, pre-human, and non-human) have
to be consistent with the biological, physiological, and environmental constraints that
existed in the ancestral environment during which evolutionary selection occurred; and
(3) the conceptualization of theoretic models unconnected to the ancestral environment
and constraints are unsuitable vehicles in explaining the literal course of biological
evolution. Explanations of the evolution of human cooperation (and by extension,
human evolution) in terms of mathematical models that ignore the constraints origi-
nating in the physical and the biological conditions of the ancestral environment are
likely to: (1) be inconsistent with data and other forms of evidence; (2) be anachronis-
tic, and (3) damage both our understanding of human cooperation and our appreciation
of the extraordinary benefits that mathematical models can provide in economic and
biological analyses.
Cooperation is instinctual; it was evolutionary favored because it produced net bene-
fits to individual organisms in the ancestral past. Cooperation is widespread throughout
the biological world,7for this reason we emphasize the biological origins of evolution
in the explanation of cooperation both within and outside the hominid line. In our
explanation we use a simple game theoretic model to put our thinking into the context
of current economic explanations. We do this to illustrate one of our messages, that
despite the repeated failures of PD models, game theoretic models can be useful in
explaining the evolution of human cooperation, but if, and only if, there is due con-
sideration of both the background conditions and the constraints of the evolutionary
setting. The problem with extant explanations of the evolution of cooperation is not
game theory per se, but rather the way it has been employed in attempts to explain the
evolution of cooperation.
Cooperation evolved in hominids as it did in other animals as an adaptation to the
physiological and environmental constraints that confronted species.8Cooperation is
Footnote 5 continued
system is controlled. Once the point is clear it becomes obvious that supposedly altruistic individuals may
act by compulsion.” (1974, p. 137) Alcock provides historical perspectives and more nuanced views of
why: “Most researchers [in biology] exploring ultimate questions about behavior look first to Darwinian
Theory [rather than group-selection] when producing their hypotheses” (2009, p. 22). Still, Ghiselin (2016)
cautions against unscientific closed-mindedness; he worries that the parsimony of selection at the organism
level “may result in real instances of it being overlooked.”
6See Bowles and Gintis (2011) for a comprehensive review of the recent literature attempting a resolution
between observed cooperation exceeding predicted levels.
7Cooperation occurs throughout biology; multicellular organisms exist because cooperation exists within
organisms at the cellular level. Abovethe cellular level, all sexual animals have to cooperate in reproduction.
Social animals are cooperators almost of necessity. Because cooperation is so widespread in the biological
world there must be benefits to the organisms involved in cooperation; otherwise, over evolutionary time,
non-benefiters of cooperation would have been eliminated from the gene pool.
8Khadjavi and Lange suggest that human cooperation derives from “social preferences.” Camera and
Casari speculate that “other oriented preferences” rather than cognition were more likely drivers behind the
123
Author's personal copy

P. R. P. Coelho, J. E. McClure
an adaptation that evolved in the distant past when it became instinctual among animals
with minimal levels of cognition. For this reason our explanation, unlike PD models,
does not require modern, or near modern, levels of cognition and culture.
2 The cooperative instinct in social animals and hominid brain evolution
Animals with minimal brain sizes cooperate; this strongly suggests that high levels of
cognition are not required to achieve cooperation. In the animal kingdom cooperation is
instinctual for social species because in the ancestral environment benefits to individual
organisms for cooperating exceeded the benefits of the alternatives to cooperation.
Evolutionarily processes selected traits that were beneficial to the organism’s survival
and reproduction. Examples of cooperation in small-brained animals are ubiquitous; an
example is the “mobbing behavior” (cooperativeattacks to fend off predators) observed
in birds,9baboons,10 squirrels,11 and wild dogs.12 Cooperation occurs in a wide array
of species ranging from sponges to mammals (mole rats, musk oxen, wolves, wild dogs,
baboons, hyenas, and dolphins), and others (birds and lichen).13 It is neither culture,
nor brains that are responsible for cooperation: it is an instinctual (arational) behavior
hard-wired (instinctive) because it was pro-adaptive in the ancestral environment.
The present-day environment does not affect evolved instincts. It is often com-
mented that humans are frequently frightened by snakes and spiders, but complacent
surrounded by automobiles; yet in contemporary societies deaths by automobiles out-
number deaths by snakes and spiders by thousands to one. The fear of snakes and
spiders is primal (instinctive) not cognitive; rather than irrational, it is arational; it is
a legacy of our ancestral past, “…an adaptive response to the input in environments
in which the input–output mechanism evolved.” [Price (2008), p. 235] In explaining
cooperation in human evolution we have to go back to the ancestral environments that
molded both the species that were both the progenitors of H. sapiens and H. sapiens
itself.
Relative to contemporary humanity, pre-hominids and early hominids had very
small brains. This implies that models, like the PD, that rely upon human cognition
in determining social institutions and norms are anachronistic because the tiny brains
that ancestral populations possessed could not have processed information and sen-
Footnote 8 continued
surprising levels of cooperation in their experiments. These scholars may well be correct; our interest is in
how and why such preferences came about.
9See Altman (1956)onWren-tits,Audubon Warblers,Anna Hummingbirds,Brewer Blackbirds, and
Redwinged Blackbirds;seeCully and Ligon (1976)onScrub and Mexican Jays;andseeShedd (1982)on
American Robins.
10 See Iwamoto et al. (1996).
11 See Tam u ra (1989).
12 See Estes and Goddard (1967). The absolute size of the brain is probably less important than its size
relative to its body mass. Elephants and whales have big brains, but the human brain is much larger relative
to its body size.
13 Besides biologists, some prominent economists have studied animal societies, notable among them were
Tullock (1994) who investigated cooperation in a variety of animal subjects (e.g. ants, termites, sponges,
and mole rates), and Landa (1986) who studied cooperation among social bees.
123
Author's personal copy

The evolution of human cooperation
sory inputs sufficient to design, or consciously choose, social behaviors, strategies
and norms. The ancestral environment which selected and molded hominid and pre-
hominid species (unlike assumptions in PD games) had no prosecuting attorneys, no
rules, no impartial adjudicators, and no paymasters. Neither the present-day envi-
ronment, nor PD games molded the instinctual behaviors of humanity; causation is
temporally uni-directional. The past affects the present, not vice-a-versa.
Two interrelated puzzles emerge in thinking about the conditions and constraints
prevailing in the ancestral environment in which human cooperation evolved: (1) how
did cooperation evolve in the absence of cognition; and (2) how did small brained
hominids acquire the resources (high-quality animal protein) necessary to facilitate the
evolution of the large brains of H. sapiens? The first puzzle is easily answered: Abstract
thinking and cognition are not pre-conditions for cooperation. As mentioned above,
cooperation occurs in an array of species with minute cognitive capacities (again,
examples include sponges, mole rats, birds, and lichens). When cooperation offers
evolutionary advantages (higher rates of survival and reproduction) organisms that
cooperate instinctively will be differentially benefitted by natural selection (evolution).
The second puzzle, how did large brains evolve in our species, is more com-
plex. Because pre-hominid (and early hominid) brains were too small to assess the
cost/benefits of complex strategies occurring over time, the relatively fragile pre-
hominids (and early hominids) were at a severe disadvantage in competition for the
animal protein (meat) required to allow larger brains to evolve. Essential for the evo-
lution of large brains is acquiring enough easily digestible protein to foster brain
development;14 more explicitly protein is a necessary condition for the evolution of
large brains, but it is not sufficient. However, in the ancestral environment easily
digestible protein was animal flesh; it was virtually the only source of protein avail-
able in sufficient quantities that would allow bigger brains to evolve.15 The ancestral
environment in which humanity evolved was replete with carnivorous animals in com-
petition with, and preying upon hominids. So the puzzle, restated more explicitly is:
How did primitive hominids having neither fangs, nor claws, nor big brains acquire
the necessary protein to allow the brain to expand?
Integrating both brain evolution and cooperation suggests a hypothesis: Hominid
cooperation preceded big-brains because large brains are neither necessary nor suf-
ficient to elicit cooperation. Pre-hominids (and early hominids) were social animals.
This means that the only way individuals could survive was in groups; cooperation is
innate among social species because without it they would not survive. Cooperation
14 The brains of modern humans require substantial proportions of the typical human’s nutritional intake.
In utero (and in newborns) the brain takes about 87 % of a nutritionally adequate metabolic diet to maintain
itself and to develop. The percentage declines as human beings age and grow larger: in a typical 5-year old
the brain requires 44 % of the metabolic budget, while in adults the brain’s metabolic demands fall to 25 %.
The decline in the percentage of the nutrition taken by the brain is a result of a larger body size; bigger
bodies demand more nutrients to move and sustain them. See Leonard et al. (2003) , Eppig et al. (2010)
and McGuire and Coelho (2011).
15 Developing and maintaining human brains make other demands on dietary inputs beyondthe caloric; also
required are amino acids, fats, proteins, and a host of assorted micronutrients. In the ancestral environment
animal source foods were virtually the only source of these nutrients that could be accessed in sufficient
quantities. See Leonard and Robertson (1992) and Milton (2003), Leonard et al. (2007), Krebs (2009), and
Milton (2015).
123
Author's personal copy

P. R. P. Coelho, J. E. McClure
benefits the individual both in the present and evolutionarily. A scenario illustrating
the interactions between cooperation and the evolution of larger brains is a situation
where pre-hominids obtained protein by scavenging the carcasses of animals that died
naturally or were killed by predators.16 Restating a point, to evolve larger brains pre-
hominids had to acquire animal protein;17 we envisage a group of early pre-humans
coming upon a carcass in competition with other animals. If an individual pre-hominid
cooperates with others in defending a carcass from other meat-eaters (imagine bran-
dishing sticks, shouting, and throwing rocks), the individual is more likely to survive
and obtain protein than if he defected and ran. In reality, running from predatory ani-
mals is a sure way to attract their attention; given that individual pre-hominids were
relatively defenseless against predators, then when turning their backs and running
from predators they became easy prey. Similarly opportunistic behaviors (such as
driving away conspecifics from the carcass to acquire a larger share) are eliminated
(both figuratively and literally) because predators that were driven from the carcass
typically linger in the bush waiting for an opportunity to feed.
This is consistent with a wealth of animal studies (see above) that show coop-
eration evolving in the absence of large brains. Cooperation among small-brained
pre-hominids in conflicts with other scavengers became instinctual because it was
proadaptive to individuals;18 the genes that harbored behaviors that fostered these
traits were evolutionarily successful—they survived in the ancestral gene pools. The
emergence of cooperation requires neither the presence of: culture, cognition, or group-
selection. Some key aspects of our hypothesis about cooperation among early hominids
are embedded in a passage from Richard Wrangham’s discussion of pre-hominid fossil
evidence from over two million years ago:
The first meat eaters certainly would have been slow, they had small bodies, their
teeth and limbs made feeble weapons, and their hunting tools were probably
little more than rocks and natural clubs. …Perhaps they found carcasses by
watching where vultures swooped down. Predators such as saber-toothed lions
brought further challenges. Teamwork might have been necessary, with some
individuals in a hunting party throwing rocks to keep fearsome animals at bay
while others quickly cut off hunks of meat before all retired to eat in a defensible
site. So it is easy to imagine that the rise of meat eating fostered various human
16 It is possible that the current human status at the top of the food chain has biased us against thinking about
early hominids scavenging for protein at the expense of animals endowed with greater natural weapons and
defenses (e.g. claws, fangs and/or body mass).
17 This was long before humans controlled fire and food was cooked; the control of fire and cooking
increased the ability to process and absorb protein. Raw (uncooked) foods are not easily absorbed by the
human gut. After fire was controlled and used, the human brain had the resources for continued growth and
evolution. Wrangham (2009, pp. 96–103) dates some cooking to Homo erectus whose beginnings predate
the present by approximately 1.9 million years. Wrangham also ascribes various changes in the anatomy
and physiology of the progenitors of H. sapiens to cooking.
18 A vivid example of instinctual cooperation is found in the African honeyguide bird; it deliberately
attracts human attention and guides humans to bee hives where the humans harvest the honey and the
birds eat the grubs and beeswax. We know that this behavior is instinctual rather than learned because the
honeyguides are brood parasites; like cuckoos, they do not raise their own chicks. Brood parasites deposit
their eggs in the nests of other birds, and the foster parents, hatch, feed and raise the honeyguide chicks.
123
Author's personal copy

The evolution of human cooperation
Tab l e 1 The hominid
scavenging game Hominid B fights Hominid B flees
Hominid A fights A gets 1.2 A gets 0.8
B gets 1.2 B gets 0
Hominid A flees A gets 0 A gets 0.4
B gets 0.8 B gets 0.4
characteristics such as long-distance travel, big bodies, rising intelligence, and
increased cooperation. (pp. 6–7)
3 An explanation for the evolution of human brains and cooperation
To illustrate the evolutionary benefits of cooperation, consider the game matrix19 that
would confront two tribe members when fending off a carnivore to acquire a carcass.
The game is dyadic: There are two individuals (A and B). Table 1contains the payoff
matrix that illustrates the choices that pre-hominids faced when competing with carni-
vores for meat (carcasses). Their “weapons” are as primitive as their brains; weapons
are simply sticks/clubs, stones, and shouting. When they contend with carnivores over
a carcasses they can either fight or flee in response to confrontations.
The matrix in Table 1illustrates the basic features of the hypothesis: The numbers
in the cells are the expected payoffs for each hominid given the action of the other.
In each cell the first entry represents the expected payoff to A, the second is B’s.
The numbers can be thought to represent a brain sustainability index that takes on a
value of unity for any dietary outcome that just maintains brain size. The 1.2 values
in the northwest cell indicate that each hominid will, on average, get a twenty percent
protein “profit” (relative to merely sustaining brain size) by cooperating in driving off
competitors.20 (Recall that larger brains cannot evolve without protein; above average
returns of protein allow larger brains to evolve.) The 0.8 value indicates insufficient
access to protein to maintain the current hominid brain.21
Consider the first column in Table 1, the column where B fights; given that B fights:
A gets 1.2 if he fights, but gets 0 if he flees. Now consider the first row in Table 1,
the column where A fights; given that A fights: B gets 1.2 if he fights, and gets 0 if he
flees. In this scenario, regardless of what the other does, fighting always yields a higher
expected payoff. Conversely fleeing always results in a lower expected payoff; if just
one flees, chances of survival are lowest (carnivores will find it easier to prey upon a
19 We frame this scenario in conjunction with the growth of hominid brains; however it is important to
realize this model, a positive-sum game, underlies the evolution of cooperation for all species. Organisms,
mindless or brainy, “cooperate” because it is beneficial to all cooperators; the benefits may, or may not, be
equally shared, but all cooperators are net benefiters. This is why cooperation evolved; alternatively it is
the evolutionary basis for cooperation.
20 In the cooperative case A and B fight together against competing carnivores.
21 The numeric values in the payoff matrix are largely heuristic. The value 1.2 could be replaced with any
number greater than one; the values 0.8 and 0.4 could be replaced with any values that had both values less
than one, but greater than zero, and had the former greater than the latter.
123
Author's personal copy

P. R. P. Coelho, J. E. McClure
hominid with his back turned than one standing his ground with club in hand and/or
stones to hurl). If both fled, the carnivore(s) may choose one of them as retreating
prey; this is why the expected payoff to A is rises from 0 to 0.4 if B also flees. Under
these conditions the dominant solution is cooperation.
Admittedly, this model is simplistic; it abstracts from evolutionary processes that,
over eons, produced the human species. An increase in protein does not immediately
increase brain size, nor does it necessarily, in and of itself, imply larger brains in the
long-run. An increase in protein merely opens the possibility for larger-sized brains to
evolve which would happen only if the reproductive success of individuals possess-
ing larger brains is greater than that of their conspecifics with smaller brains. More
explicitly, the acquisition of large amounts of protein is a necessary, but not a sufficient
condition for the evolution of large brains. The costs and benefits of larger brains differ
among species, as do the costs and benefits of other traits. The evolution of a large
human brain was favored by natural selection; this means that, on average, the pos-
sessors of large brains had more surviving descendants than conspecifics with smaller
brains. We must not be teleological; large brains were advantageous in the Homo line
because of some specific (as yet unidentified) conditions that made large brains repro-
ductively advantageous to their possessors.22 Interestingly, the advantages of large
brains are not apparent in the evolutionary record; paleontologists, anthropologists,
and evolutionary biologists believe that brains as complex as those of H. sapiens have
evolved only once. In contrast, vision, flight, and cooperation have evolved separately
in widely disparate species many times. This suggests that during the vast majority
of evolutionary time very large brains were not evolutionarily desirable; which means
that, outside of the hominid line, the reproductive success of individuals with brains
larger than the species mean was typically less than that of their small-brained con-
specifics.23 The aphorism that there are no free lunches applies to genetic heritability;
while humans are much brainier than other primates, humans are much weaker. Recent
research in comparative anatomy makes this clear. “The two species’ [chimpanzees and
humans] musculature is extremely similar, but somehow, pound-for-pound, chimps are
between two and three times stronger than humans.”24
22 Darwin speculated that it was sexual selection that made large brains attractive to the opposite sex; if
females found brainier mates more attractive, then brainier males would have more surviving descendants.
Following Darwin, Cronin (1991) argued that the human brain was analogous to the peacock’s tail in
that it was for to attracting mates, even though it does have some significant ecological costs among the
Peafowl. The reason why sexual selection is relevant is that it offers an explanation for the evolution of
brainier hominids in spite of the brain’s large costs (just as sexual selection provides an explanation for the
peacock’s tail despite the costs it entails in terms of predation).
23 An example of the cost/benefit calculus is that of brains versus brawn. Because the brain is so costly
to develop (see footnotes 14 and 15) and maintain, getting larger brains must entail reductions somewhere
else in the organism’s body. Big-brained but physically smaller carnivores (e.g., lions or wolves) face severe
disadvantages when competing against conspecifics with larger muscles and body mass. Lions and wolves
acquire and ingest large amounts of protein, but have not evolved the large brains that characterize human
beings. Again (as noted in the main text) the acquisition of large quantities of protein is a necessary, but
not a sufficient condition for the evolution of large brains. Again (as noted in the text and in Footnote 22
above) the evolution of the human brain is anomalous, akin to the peacock’s tail, with sexual selection also
being the likely explanation.
24 Wolchover (July 29, 2015); for a more complete scientific elaboration explaining the linkages between
big brains and weak muscles see Bozek et al. (2015).
123
Author's personal copy

The evolution of human cooperation
Our model should not be interpreted from the perspective of players “rationally”
comparing and individually choosing among strategies to maximize individual pay-
offs. Unlike PD games and all others involving strategizing players, ours makes no
presumptions that participants possess sufficient cognitive capacity to think and ratio-
nally choose among, alternatives.25 Rather, we are using the game matrix to illustrate
why the process of mindless evolution in the ancestral environment selected individ-
uals who cooperated.26 In the Hominid Scavenging Game individual defection (flight
while the other stays and fights) has the lowest payoff (no protein, increased risk of
death or injury), while cooperation (both fight) has the highest payoff (reduced risk
of death or injury, increased probability of protein). The alternatives described here
explain why cooperation became hard-wired. Given that both survival and additional
protein were necessary for the evolution and growth of brains, cooperation cannot be
predicated upon rationally weighing and comparing alternatives. Again, unlike the PD
in which defection is the dominant strategy, our scenario is consistent with both the
ancestral environment in which humanity evolved, and the physiological attributes of
the ancestral populations of hominids and pre-hominids.27
Within limits,28 the benefits of cooperation increase with group size. Referring
to the scavenging scenario, an absolutely larger number could increase the hail of
sticks, stones and noises that rain upon a pack of competing scavengers or carnivores.
Increased group size increased the probability of success (gains) and reduced the prob-
ability of injury/death (costs). Any animal running from carnivores faces an increased
probability of an early death; a single (pre) hominid defecting from the tribal group
and running from carnivores would face an almost certain death. Hunting animals
are most successful when they are able to separate individuals from herds (tribes);
their instinct is to isolate, chase and kill individuals that defect from groups.29 Again,
25 The stag game, like the PD, involves strategizing players. Unlike the PD, in the stag game there are
two Nash equilibria: (a) in the Pareto efficient equilibrium the two hunters cooperate to bag a stag yielding
each hunter say 3 units of meat, and (b) in the risk dominant equilibrium each hunter bags a rabbit yielding
just one unit of meat. In the stag game, players will cooperate (hunt the stag rather than rabbit) only if
they can be made to believe that the other player will cooperate because if one hunts stag while the other
hunts rabbit, then the stag hunter gets no meat while the rabbit hunter bags two rabbits. In our Hominid
Scavenging Game, there is, again, no presumption of strategizing players; rather it is the process of mindless
evolutionary selection in the ancestral environment that would probabilistically have favored the survival
of individuals who cooperated over those who did not.
26 Our model focuses upon inter-species competition. As an anonymous referee pointed out, discussions
of this type of competition were once frequent in the literature but have been crowded out to a large extent
by the over-emphasis of intra-species competition. Although we acknowledge that interactions between
these types of competition may exist, we see them as beyond the scope of this paper.
27 This scenario also explains the acquisition of protein before cultural developments (the acquisition of
fire, cooking, and increased cooperation) which further aided human evolution. This scenario underlies the
evolution of human cooperation in the actual environment(s) and temporal sequence in which it occurred.
28 As an anonymous referee pointed out primate research reveals that there are tradeoffs associated with
living in larger groups. In a study of the limits to primate group sizes, Dunbar (1992, p. 469) discovered
that “species will only be able to invade habitats that require larger groups than their current limit if they
evolve larger neocortices.”
29 Large herbivores (musk oxen, water buffalo) cooperate by creating defensive formations to repel preda-
tors; Alcock has numerous examples of defensive formations, mobbing behaviors, and other forms of
intra-species co-operation.
123
Author's personal copy

P. R. P. Coelho, J. E. McClure
this illustrates that cooperation does not require advanced cognition. In the ances-
tral environment, cooperation was instinctual for innumerable species because on an
individual basis there were net benefits from cooperation, and net costs from defec-
tion.
4 A virtuous cycle: cooperation, protein, and brain evolution
The acquisition of protein advances the hypothesis that the evolution of cooperation
and the development of the brain are linked. The self-reinforcing feedback cycle is
depicted in Fig. 1. In it the direction of the arrows indicates causality.
Starting with Box 1 of Fig. 1, cooperation allowed the ancestral forebears of the
Homo genus to acquire more protein (Box 2). Additional protein allowed: (1) the
brain to increase in size; and (2) alterations to other parts of the species anatomy
consistent with and, complementary to, larger brains (Box 3). Increased brain size and
changing physiology, musculoskeletal system,30 digestive system,31 and others that
are subsumed in Box 3 facilitated technological and cultural developments associated
with Box 4.32 These developments facilitated both increased protein acquisition and
increased cooperation. A major technological and cultural advance that considerably
enhanced the effective supply of protein was the control of fire: it made cooking
common, and cooking allowed for the evolution of a digestive system that processed
protein and other nutrients contained in meat without the caloric and physiologic
demands that the digestion of raw foods required.33
Box 4 contains “Technological and Cultural Advances;” among these are develop-
ments in communications, speech, and language that were outgrowths of physiological
and cultural evolution. These advances feedback to Box 1 (increasing cooperation):
this is where genes and culture coevolve. The literature on coevolution is expansive
30 Changes in the brain are nuanced. Calvin (2004, pp. 96–97) explains that most of the evolutionary data
suggest that the emergence of the cognitive planning associated with hitting moving targets did not arise
in a specialized “ …bump on the head that could be labeled Hand-Arm Planning Center. …Certainly there
is much evidence suggesting that oral-facial movement planning can overlap that for hand-arm—and with
that for language, both sensory and motor aspects.” That is, it is most likely that “major portions of the
brain [increased] together.”
31 A good example of the coevolution of genes and culture is in Wilson (2015, p. 64). Paraphrasing Wilson,
lactose intolerance is common among humans after weaning. Pastoralism and the herding of cattle, goats,
horses, and sheep made milk a source of adult nutrition. Over generations a genetic aberration that made the
human adult digestive system lactose tolerant became widespread among herders. The ability to utilize milk
products made the animals more valuable, thus increasing incentives to herd these animals; their numbers
and the population of lactose tolerant humans dependent upon milk-based nutrition both increased.
32 Box 4 includes a host of major and minor developments. Among these are: language, male-female
pair-bonding, fire, gender based division of labor, tool-making, problem-solving, and abstract thinking.
33 Wrangham explains that cooking allowed the human gut to evolve into a smaller organ that required fewer
calories to process and digest foods. This released calories and nutrients that had previously been claimed
by the gut’s processing demands to be reallocated to the evolving brain. Humans have by far the smallest
guts and the biggest brains in the primate family; these developments are synergistically intertwined.
123
Author's personal copy

The evolution of human cooperation
(1)
Cooperation
(2)
Protein
(3)
Bigger Brains, Cognitive
and Physiological Changes
(4)
Technical and Cultural
Advances and Coevolution
Fig. 1 Cooperation and human evolution
and debatable;34 while the nuances of coevolution are beyond the scope of this paper,
we note that our model can incorporate it and does not categorically exclude it.35
The arrow connecting Box 4 to Box 2 indicates that technological and cultural
changes (e. g. communication, speech, gender division of labor,36 and fire) enabled
more sophisticated hunting strategies and larger group sizes. Larger group sizes gave
rise to increased specialization, which led to increasing per capita output.37
The self-reinforcing feedback effects in the figure explain the growth of the brain
and the evolution of hominids into H. sapiens. Emphasizing a major point, cooperation
34 “Although culture has for many decades been envisioned as an evolutionary process, there is little
agreement about its precise nature, importance, or relationship to genetic evolution.” Wilson (2002; p. 28;
emphasis added)
35 An anonymous referee called attention to Chapais’ (2008) theory that pair bonding gave birth to human
society.As Chapais (2011, p. 1277) explains: “…pair bonding … brought about the multifamily composition
of human groups, with enduring associations between mothers and fathers enabling children to recognize
their fathers. This, in turn, made it possible for children to recognize their father’s relatives; that is, pair
bonding would reveal the underlying genealogical structure and create bilineal kinship.” This was “a factor
alleviating conflicts between male affines. Similarly, grandfathers, brothers, and uncles would recognize
their transferred kin and their affines [their mates] instigating a state of mutual tolerance.” (Chapais 2011;
p. 1277).
36 Chapais (2008, p. 183) argues that “the sexual division of labor was the outcome of a specific concate-
nation of unrelated events, namely, (1) bipedalism, which rendered gathering possible, (2) pair-bonding,
which created food-sharing biases among primary kin, and (3) a chimpanzee-like male hunting bias, which
operated as both a template and a spring-board for complete sexual specialization.” While the nuances of
this argument are beyond the scope of our model, it is not in conflict with the sexual division of labor in
Box 4 (Fig. 1) with its positive feedbacks to both Cooperation (Box 1) and Protein (Box 2).
37 Adam Smith famously wrote: “The division of labor is limited by the size of the market.” As the
absolute size of economic output increases, the returns to specialized economic activities increase, leading
to increasing (absolute and per capita) output. This generates a “virtuous cycle,” or, in the context of this
paper, coevolution. See Stigler (1951) and McGuire and Coelho (2011).
123
Author's personal copy

P. R. P. Coelho, J. E. McClure
had to have preceded the evolution of large brains because the acquisition of protein is
a necessary condition for the evolution of larger brains. Cooperation among animals
did not, and does not, require large brains;38 cooperation became biologically hard-
wired as a consequence of the benefits derived from cooperation and the costs of
non-cooperation.
5 Final remarks
Error is eternal, and wisdom consists in living with it, not letting our vanity tell
us that it has been transcended. Ghiselin (1974, p. 13)
It is futile to attempt an explanation of the evolution of human cooperation with
anachronistic models that are gross mischaracterizations of the ancestral environment
in which cooperation evolved. Evolution took place in a Ghiselin-like world; there were
no rules, no prosecuting attorneys, no jails, nor institutionally administered imprison-
ments. In the ancestral environment physical existence was at frequent risk; survival
depended upon access to resources and the ability to avoid being another organism’s
meal. Pervasive defection and strategic games involving bluffing and deception are
not indicative of an instinctual tendency to defect, but a consequence of the capacity
of the human brain to think strategically. Neither slime molds, nor mole rats, nor other
small-brained social animals continually employ strategic behavior in their dealings
with conspecifics; while humans habitually think strategically as vividly evidenced
by cultural pastimes (poker, chess, and gossiping).39 Strategic behaviors are a conse-
quence of large brains and cognition, while these in turn are the results of the success
of hominid cooperation in the ancestral environment that allowed the acquisition of
protein and the resultant growth in brain-power. Our explanation of the evolution of
human cooperation may be incorrect but it is consistent with: (1) the anatomical evolu-
tion of humanity; (2) the paleontological and chronological evidence; and (3) modern
biology. We are unaware of any other explanation that can make these claims.
References
Alcock, J. (2009). Animal behavior. Sunderland, MA: Sinauer Associates.
Altman, S. (1956). Avian mobbing behavior and predator recognition. Condor,58(4), 241–253.
Axelrod, R. (1984). The evolution of cooperation. New York: Basic Books.
Bowles, S., & Gintis, H. (2011). A cooperative species: Human reciprocity and its evolution. Princeton,
NJ: Princeton University Press.
38 As noted earlier, there is evidence of cooperation among species that have small brains. Certainly early
humans had brains, but cooperation has evolved throughout the animal kingdom among both creatures with
large brains and those with tiny ones; large brains are not necessary for the emergence of cooperation.
39 Strategic thinking should not be confused with deception or mimicry.Chameleons, puffer fish, butterflies
and a host of other species have behaviors that thwart predators and rivals. As examples, the chameleon
does not have to think about what color his skin is to be, neither do stick insects have to think about how
to look like sticks; these are innate actions accomplished without conscious thought. Strategic behaviors
have to be planned and thought over, so perhaps non-cooperation and strategic behavior is a result of the
large human brain. If so, this is paradoxical; game-playing and strategic non-cooperation were results of
the large brain which resulted from cooperation.
123
Author's personal copy

The evolution of human cooperation
Bozek, K., Wei, Y., Yan, Z., Liu, X., Xiong, J., Sugimoto, M., Tomita, M., Pääbo, S., Pieszek, R., Sher-
wood, C. C., Hof, P., Ely, J., Steinhauser, D., Willmitzer, L., Bangsbo, J., Hansson, O., Call, J.,
Giavalisco, P., Khaitovich, P. (2015). Exceptional evolutionary divergence of human muscle and brain
metabolomes parallels human cognitive and physical uniqueness. PLoS Biology. doi:10.1371/journal.
pbio.1001871. Retrieved October 25, 2014 from http://journals.plos.org/plosbiology/article?id=10.
1371/journal.pbio.1001871.
Calvin, W. (2004). A brief history of the mind: From apes to intellect and beyond. New York: Oxford
University Press.
Camera, G., & Casari, M. (2009). Cooperation among strangers under the shadow of the future. American
Economic Review,99(3), 979–1005.
Chapais, B. (2008). Primeval kinship: How pair-bonding gave birth to human society. Cambridge: Harvard
University Press.
Chapais, B. (2011). The deep social structure of humankind. Science,331, 1276–1277.
Cronin, H. (1991). The ant and the peacock: Altruism and sexual selection fromdarwin to today. Cambridge:
Cambridge University Press.
Cully, J. F., & Ligon, J. D. (1976). Comparative mobbing behavior of Scrub and Mexican jays. Auk,93(1),
116–125.
Dunbar, R. (1992). Neocortex size as a constraint on group size in primates. Journal of Human Evolution,
20, 469–493.
Eppig, C., Fincher, C. L., & Thornhill, R. (2010). Parasite prevalence and the worldwide distribution of
cognitive ability. Proceedings, Biological Sciences,277, 3801–3808.
Estes, R. D., & Goddard, J. (1967). Prey selection and hunting behavior in african wild dogs. Journal of
Wildlife Management,31(1), 52–70.
Flood, M. M. (1958). Some experimental games. Management Science,5(1), 5–26.
Ghiselin, M. (1974). The economy of nature and the evolution of sex. Oakland: University of California
Press.
Ghiselin, M. (2016). Editorial Communication.
Iwamoto, T., Mori, A., Kawai, M., & Bekele, A. (1996). Anti-predator behaviors of gelada baboons. Pri-
mates,37(4), 389–397.
Khadjavi, M., & Lange, A. (2013). Prisoners and their dilemma. Journal of Economic Behavior and Orga-
nization,92, 163–175.
Krebs, J. R. (2009). The gourmet ape: Evolution and human food preferences. The American Journal of
Clinical Nutrition,90(3), 7075–7115.
Landa, J. T. (1986). The political economy of swarming in honeybees: Voting-with-the-wings, decision-
making costs, and the unanimity rule. Public Choice,51(1), 25–38.
Leonard, W. R., & Robertson, M. L. (1992).Nutritional requirements and human evolution: A bioenergetics
model. American Journal of Human Biology,4, 179–195.
Leonard, W.R., Snodgrass, J. J., & Robertson, M. L. (2007). Effects of brain evolution on human metabolism.
Annual Review of Nutrition,27, 311–327.
Leonard, W. R., Robertson, M. L., Snodgrass, J. J., & Kuzawa, C. W. (2003). Metabolic correlates of
hominid brain evolution. Comparative Biochemistry and Physiology, Part A,135, 5–15.
Mailath, G. J. (1998). Do people play nash equilibrium? Lessons from evolutionary game theory. Journal
of Economic Literature,36, 1347–1374.
Marshall, Alfred. (1890/1949). Principles of economics. New York: McMillian.
McGuire, R., & Coelho, P. R. P. (2011). Parasites, pathogens, and progress. Cambridge, Massachusetts:
MIT Press.
Milton, K. (2015). A hypothesis to explain the role of meat-eating in human evolution. Evolutionary Anthro-
pology,8(1), 11–21.
Milton, K. (2003). The critical role played by animal foods in human (homo) evolution. The Journal of
Nutrition,133(11), 3886S–3892S.
Pinker, S. (2012). The false allure of group selection. Retrieved from http://edge.org/conversation/
the-false- allure-of- group-selection.
Price, M. E. (2008). The resurrection of group selection as a theory of human cooperation: A review
of foundations of human sociality: Economic experiments and ethnographic evidence from fifteen
small-scale societies. Social Justice Review,21, 228–240.
Shedd, D. H. (1982). Seasonal variation and function of mobbing and related antipredator behaviors of the
American Robin (Turdus migratorius). Auk,99(2), 342–346.
123
Author's personal copy

P. R. P. Coelho, J. E. McClure
Stigler, G. J. (1951). The division of labor is limited by the size of the market. Journal of Political Economy,
59(3), 185–193.
Tamura, N. (1989). Snake-directed mobbing by Formosan squirrel Callosciurus erythraeus thaiwanensis.
Behavioral Ecology and Sociobiology,24, 175–180.
Tullock, G. (1994). The economics of non-human societies. Tucson, Arizona: Pallas Press.
Wessells, N. K., & Hopson, J. L. (1988). Biology. New York: Random House.
Wilson, D. S. (2002). Darwin’s cathedral. Chicago: University of Chicago Press.
Wilson, D. S. (2015). Does altruism exist?. New Haven: Yale University Press.
Wolchover, N. (2015). Chimps vs. humans: How are we different? Live Science. Retrieved October 25,
2015 from http://www.livescience.com/15297-chimps-humans.html.
Wrangham, R. (2009). Catching fire: How cooking made us human. New York: Basic Books.
123
Author's personal copy