Content uploaded by Paul L Hooper
Author content
All content in this area was uploaded by Paul L Hooper on Dec 12, 2017
Content may be subject to copyright.
Review
Leadership in Mammalian
Societies: Emergence,
Distribution, Power, and Payoff
Jennifer E. Smith,
1,
* Sergey Gavrilets,
2
Monique Borgerhoff Mulder,
3
Paul L. Hooper,
4
Claire El Mouden,
5
Daniel Nettle,
6
Christoph Hauert,
7
Kim Hill,
8
Susan Perry,
9
Anne E. Pusey,
10
Mark van Vugt,
11,12
and Eric Alden Smith
13,
*
Leadership is an active area of research in both the biological and social
sciences. This review provides a transdisciplinary synthesis of biological and
social-science views of leadership from an evolutionary perspective, and exam-
ines patterns of leadership in a set of small-scale human and non-human
mammalian societies. We review empirical and theoretical work on leadership
in four domains: movement, food acquisition, within-group conflict mediation,
and between-group interactions. We categorize patterns of variation in leader-
ship in five dimensions: distribution (across individuals), emergence (achieved
versus inherited), power, relative payoff to leadership, and generality (across
domains). We find that human leadership exhibits commonalities with and
differences from the broader mammalian pattern, raising interesting theoretical
and empirical issues.
Evolution of Leadership Across Mammalian Societies
Leadership (see Glossary) is a pervasive phenomenon in social species, organizing behaviors
ranging from group movement to complex patterns of cooperation and conflict. Despite being
an active topic in biological and social sciences, little in the way of theoretical and empirical
synthesis exists. Theory and data indicate that leadership is relevant in a variety of contexts of
current biological interest, including reproductive skew, the evolution of cooperation, intergroup
conflict, and evolved personality differences [1–4]. Biologists have produced a substantial
theoretical and empirical literature on hierarchy and decision-making in non-human societies,
but focus primarily on leadership during movement [5–9]. By contrast, social scientists have
written extensively about leadership in a wide variety of contexts [10,11], but often lack
evolutionary perspectives and formal models (but see [12–15]).
This review offers a transdisciplinary synthesis of biological and social-science models of
leadership from an evolutionary perspective, and analyzes leadership patterns across a sample
of mammalian species. We review evolutionary models of leadership in four domains: group
movement, food acquisition, conflict mediation, and between-group interaction (Box 1). We
draw on theoretical work to analyze how leadership varies across these domains for a sample of
human and non-human groups. With evolutionary questions in mind, we classify this variation
into five dimensions that can be identified across species: distribution across individuals
(skew), leadership emergence, differential power, relative payoff (to leaders versus followers),
Trends
Leadership is an active research area in
both biological and social sciences, but
there has been limited synthesis within
or across these areas; evolutionary the-
ory can assist with such synthesis, but
additional elements are needed for a
robust comparative framework.
Variation in leadership can be mea-
sured in multiple dimensions, including
emergence (how does one become
a leader?), distribution (how widely
shared is leadership?), power (how
much power do leaders wield over fol-
lowers?), relative benefit (do leaders
gain more or less than followers?),
and generality (how likely are leaders
in one domain, such as movement or
conflict resolution, to lead in other
domains?).
A comparative framework based on
these dimensions can reveal common-
alities and differences among leaders in
mammalian societies, including human
societies.
1
Department of Biology, Mills College,
Oakland, CA 94613, USA
2
Department of Ecology and
Evolutionary Biology, and Department
of Mathematics, University of
Tennessee, and National Institute for
Mathematical and Biological
Synthesis, Knoxville, TN 37996, USA
3
Department of Anthropology, and
Center for Population Biology,
University of California at Davis, Davis,
TREE 2008 No. of Pages 13
Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy http://dx.doi.org/10.1016/j.tree.2015.09.013 1
© 2015 Elsevier Ltd. All rights reserved.
TREE 2008 No. of Pages 13
and generality across domains (Box 1). We advance the study of leadership by synthesizing[30_TD$DIFF] its
theoretical predictions regarding emergence and maintenance, and establish measurable
criteria for comparative study to stimulate empirical and theoretical progress. The hypothesis
that humans have evolved unique leadership attributes not found among other mammals
(e.g., [3,15]) remains largely untested because we currently lack suitable comparative frame-
works for examining leadership across mammalian societies. One aim of this review is to apply
such a framework in a preliminary examination of this hypothesis.
Definitions
The study of collective decision-making suffers from lack of agreement on terminology [16].
Biologists often define leadership as differential influence on group activity, however generated.
The usual domain of interest is movement [7], with some additional focus on the duration of
activities [6]. By contrast, few social science uses of this term highlight being the first to initiate
movement or change activity because this might reflect impatience or need rather than
differential influence [17]. Instead, they commonly point to differences in social power, meaning
the ability of an individual to induce others to act or believe in a particular way [18].Wedefine
leadership here as non-random differential influence on collective behavior. This differential
influence might vary based on visibility, knowledge, or other factors affecting voluntary decisions
to follow or emulate. Alternatively, it might depend on coercion to control the behavior of
subordinates, for which the term ‘dominance’is often employed. This variation in causal
mechanisms is one issue we address below.
Models of Leadership
Members of social species regularly make decisions to coordinate activities with group-mates,
including where and when to forage, rest, or move. Formal evolutionary models help reveal the
evolutionary causes and consequences of these decisions. Theory predicts that, if accurate
information is to be maximized and the cost of differences in individual preferences is to be
CA 95616, USA
4
Department of Anthropology, Emory
University, Atlanta, GA 30322, USA
5
Department of Zoology, University of
Oxford, Oxford OX1 3PS, UK
6
Centre for Behavior and Evolution,
and Institute of Neuroscience,
Newcastle University, Newcastle NE2
4HH, UK
7
Departments of [28_TD$DIFF]Mathematics and
[29_TD$DIFF]Zoology, University of British
Columbia, Vancouver, BC V6T 1Z4,
Canada
8
Institute for Human Origins, and
School of Human Evolution and Social
Change, Arizona State University,
Tempe, AZ 85281, USA
9
Department of Anthropology,
University of California Los Angeles,
Los Angeles, CA 90095, USA
10
Department of Evolutionary
Anthropology, Duke University,
Durham, NC 27708, USA
11
Department of Experimental and
Applied Psychology, VU University
Amsterdam, the Netherlands
12
Institute for Cognitive and
Evolutionary Anthropology, University
of Oxford, UK
13
Department of Anthropology,
University of Washington, Seattle, WA
98195-3100, USA
*Correspondence: jesmith@mills.edu
(J.E. Smith) and easmith@uw.edu
(E.A. Smith).
Box 1. Leadership Domains and Dimensions
Leadership and alternative forms of decision-making (e.g., via quorum-sensing) can occur in one or more domains or
contexts within social groups. As noted in the main text, collective movement is the domain most com monly treated in the
biological literature (both theoretical and empirical), but social scientists as well as some biologists are concerned with
leadership phenomena in other domains that have significant fitness implications. For our comparative analysis of
mammalian (including human) leadership, we chose to focus on four domains: (i) collective movement, (ii) food
acquisition, (iii) within-group conflict resolution, and (iv) between-group interactions (whether peaceful or hostile). The
text discusses examples of leadership in each of these domains for a variety of social groups, as well as related theoretical
models concerning leadership evolution.
Leadership dimensions involve a more abstract categorization. This can be approa ched in the form of a set of questions.
For any given domain, are there leaders? In a purely egalitarian or democratic decision regime, involving either consensus
decision-making or absence of collective behavior, the answer will be ‘no’. We recognize that this ‘pure’form of
consensus may be rare given the pervasiveness of often subtle forms of leadership and influence. If the answer is ‘yes’for
any given domain, the following questions can be used to characterize different dimensions along which leadership
varies:
(i) How distributed is leadership? In other words, what proportion of group membership typically makes a decision in a
given domain?
(ii) Emergence. Is leadership achieved (by leader's actions or qualities) or ascribed (by inheritance or some other
process not dependent on the individual's actions or qualities)?
(iii) How powerful is the leader? Our definition of power [23_TD$DIFF](see Glossary[24_TD$DIFF])[25_TD$DIFF]does not necessarily [26_TD$DIFF]entail coercion.
(iv) Relative benefit. Do leaders reap fitness-correlated benefits from their actions equal to, greater than, or less than,
that averaged by other group members?
(v) Generality. To what extent does leadership in one domain predict leadership in other domains?
As discussed further in the text, evolutionary theory as well as empirical evidence suggests that these dimensions are not
independent of each other; for example, greater power should be (and often is) associated with higher relative payoff to
leadership.
2Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy
TREE 2008 No. of Pages 13
minimized, [31_TD$DIFF]fully shared decision-making[5_TD$DIFF] should emerge[6_TD$DIFF] (i.e., no consistent leaders) [19,20].
This raises the puzzle of why leaders often do emerge. However, if time is a crucial factor,
consensus decisions can be inefficient, thereby increasing costs for group members. For
example, in the presence of significant within-group heterogeneity (defined as variation in
preferences, personality traits, and reputation), group decisions by a single leader can be most
efficient [21]. Moreover, recent models suggest that leadership can also emerge under the
influence of differential information, differential relatedness to group-mates, differential power or
resource control, personality variation, imitation, punishment, and volunteerism, with varying
effects on fitness.
Our ultimate goal is to [32_TD$DIFF]derive [33_TD$DIFF]general [34_TD$DIFF]principles [35_TD$DIFF]from [36_TD$DIFF]models [37_TD$DIFF]of leadership[8_TD$DIFF].Whereasthescope
of inference from any single mathematical model is limited, comparison of multiple models
allows for true insights about biological principles. Although some models span multiple
domains, each model depends upon its specific assumptions, which are often domain-
specific. Therefore we first organize our review of models according to the four domains noted
above before concluding with the overarching conceptual insights to be gained from existing
theoretical models.
Movement
Couzin et al. [7] studied group movement when only a few members have information on the
location of a resource. If members lack a preferred travel direction, and all individuals attempt to
avoid collision while simultaneously maintaining group cohesion, a small proportion (e.g., 5%) of
informed individuals can efficiently lead others towards the resource. In a related model [22],
individuals can invest in obtaining personal or social information about the resource location.
With fitness tradeoffs, the population evolves to be dimorphic, with a small proportion of
individuals (leaders) actively acquiring information from the environment but ignoring social
information from their group-mates, and the remainder simply follow the leaders. These dynam-
ics are similar to those in other producer–scrounger models [23,24]. Stable differences in
propensities to lead dispersal attempts arise when dispersal decreases competition between
kin; the net fitness benefits of leadership depend upon the costs of dispersal as well as on the
degrees of relatedness and reproductive skew within dispersal groups [25].
Johnstone and Manica [4] modeled an N-player coordination game with conflicts of interest, and
analyzed the emergence and maintenance of two personality types: (stubborn) leaders and
(agreeable) followers (see followership). Leaders are maintained by frequency-dependent
selection[38_TD$DIFF].[39_TD$DIFF]They [40_TD$DIFF]benefit[41_TD$DIFF]from imposing their preferences on followers, but fail to coordinate
effectively when interacting with other leaders. The fraction of leaders in the population increases
with the degree of conflict among group members, with both types being common only at
intermediate levels of conflict.
Resource Acquisition
There are several ways leaders improve the efficiency of resource acquisition, thereby increasing
the fitness of group members.
Leaders as Role Models
In mathematical models of social interactions with a possibility of imitation, leaders (i.e.,
individuals whose strategies are imitated by the rest of the group) emerge naturally. For example,
in a version of the minority game [26,27], individuals chose between two different resources of
equal total value, but being in the minority provides an advantage (e.g., because the resource is
shared among a smaller number of group-mates). Individuals remember group choices over
several rounds and base their decisions on past experiences (individual learning) or choices of
others (social learning). Rapid emergence of influence-networks can occur when followers copy
Glossary
General Terms
Coercion: control of the behavior of
others through threats or attacks.
Collective action problem: any
situation in which multiple individuals
would all benefit from a particular
action, but difficulties of coordination
or of ensuring fair contribution to the
costs of the action create obstacles;
also known as a ‘social dilemma’.
Coordination problem: situations in
which individual success requires
collective action to achieve a goal.
These range from contexts of pure
coordination, where individuals have
the same preference or fitness ranking
across outcomes, to cases where
individuals have different rankings
but still achieve higher payoffs by
coordinating on one choice.
Dominance: ability to win dyadic
agonistic interactions (e.g., contests
or unsolicited appeasements), with
outcomes determining priority of
access to resources or mating.
Egalitarian group: a society in
which the individuals of the same
age-sex category have equal access
to and control over resources and/or
other individuals within the group
(in contrast to a stratified/despotic
society).
Leadership: non-random differential
effect on group behavior of
conspecifics through actions evolved
or intended to elicit this effect (a
glossary of leadership types is given
below)
Power: the ability of leaders to
motivate followers to behave in ways
they would otherwise not do, often
but not necessarily through coercion.
Prestige: influence or deference that
is freely granted (i.e., not generated
by the use of threat or force, in
contrast to dominance).
Shared decision-making: multiple
group members decide on an
outcome (via consensus or quorum-
sensing), in contrast to unshared
decision-making.
Small-scale society: a human
society consisting of one or a few
local communities (several hundred to
a few thousand members in total);
these tend to be egalitarian, but with
some notable exceptions.
Stratified society: one in which
segments have differential access to
and/or control over resources; also
known as a despotic society, or a
society structured by a dominance
hierarchy.
Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy 3
TREE 2008 No. of Pages 13
the choices of a small number of individuals with the best strategies (leaders) [27,28]. In a version
of the Prisoner Dilemma game with dynamical adjustment of social ties and social learning,
adaptation dynamics favor the emergence of highly connected cooperators with high payoffs
[29]. These ‘leaders’sustain the global cooperative state because their social partners imitate
their cooperative strategies. A model of coordination on networks showed that group-mates
often emulate individuals (leaders) with privileged information [30]. In a model of how leadership
influences group success, players in a multi-player weak-link game become leaders after
publicly choosing to contribute before all other players [31,32]; thus, leadership can increase
efficiency but not always guarantee success.
Leaders as Punishers
Punishment has attracted substantial attention as a way to stabilize the production of public
goods [33,34]. However, if punishment is costly, it raises a second-order social dilemma
because mild cooperators who shirk the costs of punishment will restore the original dilemma
to the detriment of all [35]. One solution is to compensate specialized actors (leaders) to administer
punishment [36].Buildingonpatron–client and managerial mutualism models [37], Hooper et al.
[13] showed that a combination of leaders and followers will outcompete acephalous strategies
if followers are taxed to pay leaders to efficiently enforce collective action in circumstances
where blindly cooperative strategies, peer punishment, or other alternative mechanisms are
incapable of efficiently ensuring cooperation. The amount extracted by leaders in the model
depends on the cost of the leaders’efforts, and the efficiency of the process of selecting leaders.
This model was extended to the specific public good of success in contests over territory [38]
as well as applied to agent-based simulations of pre-Hispanic Pueblo societies [39,40].
Leaders as Managers
A related model [42_TD$DIFF]assumed that[43_TD$DIFF] the presence of a leader increases a group's productivity in
collective action [41]. Leaders gain more of the reward than the remainder of the group, but they
are tolerated because of increased absolute productivity. Leaders are constrained by the fitness
followers would gain from leaving the group; as a result, where leadership is efficient and costs
of dispersal are high, leaders are able to claim unequal shares of production. This class of models
[13,41] is analogous to those of reproductive skew [42–44].
Leaders as Volunteers
In many public goods games in which the total group benefit is a nonlinear increasing function of
the number of contributors, cooperation can sometimes be maintained without genetic relat-
edness or other forms of assortment [45]. For example, in an N-person volunteer's dilemma [46],
a public good is produced if there are at least kvolunteers (who can be viewed as group leaders);
frequency-dependent selection favors a stable mix of volunteers (leaders) and free-riders
(followers) [47]. If the benefit of a collective action declines in time [48], altruistic leaders can
emerge in groups of non-relatives, and larger group sizes and higher costs of volunteering delay
but do not prevent their emergence. In contrast to producer–scrounger models where the
producers and scroungers have equal fitness at equilibrium, volunteers (that is, producers) have
lower fitness than free-riding group-mates.
Within-Group Conflict
There is a substantial body of theoretical work on within-group conflicts in which dominant
individuals (‘leaders’) participate in or are targets of coalitionary attacks [49,50]. Coalitions can
strongly affect the power of both dominant and subordinate individuals. Within-group competi-
tion and conflict can reduce the absolute amount of resources available to the group. [44_TD$DIFF]In [45_TD$DIFF]this
case, individuals can benefit from investing in a trait (‘policing’) that simultaneously reduces
the relative amount of resource subject to within-group competition and the costs of this
competition [51]. If group members vary in relative cost of policing, stronger individuals (‘leaders’)
Unshared decision-making: one or
a very few group members make
decisions for the group.
Types of Leadership
Acephalous group: leaderless
group; in the human case, one
without institutionalized leadership.
Achieved status: determined by an
individual's abilities, performance or
effort (in contrast to ascribed status).
Ascribed: assigned at birth or
assumed involuntarily later in life (e.g.,
through maternal rank inheritance
based on maternal interventions or
bequeathal of resources down a
lineage).
Attribute-based leadership:
determined by the specific traits
possessed by a leader (rank, age,
tenure, sex, physiological state) when
that individual assumes role of leader
or follower.
Delegated leadership: when a
leader has tacit or explicit group
consent to act without seeking
consensus on each decisions (thus
intermediate between pure
consensus-based decisions and
despotism).
Despotic leadership: consistent
leadership by a powerful individual or
small set of individuals; also termed
‘unshared’or ‘personal’leadership.
Distributed leadership: different
individuals lead group action on
different occasions, for a given
domain, due to fluctuations in
motivation, knowledge, or a turn-
taking convention; also known as
shared leadership.
Domain-specific leadership: when
an individual's leadership is limited to
a specific domain (contrasting with
domain-general leadership, as in
generalized dominance).
Followership: leadership presumed
on the basis of others following (e.g.,
one individual initiates movement and
others follow).
Institutionalized leadership:
consists of durable positions
(regardless of length of tenure of
individuals occupying those
positions), in contrast to situational
leadership.
Situational leadership: leadership
that arises opportunistically and
occurs only in specific situations of
short duration; contrasting with
institutionalized leadership.
Volunteerism: the act of freely
proffering services to a group (e.g.,
agreeing to assume a costly
leadership role).
4Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy
TREE 2008 No. of Pages 13
are favored to take on most or all of the policing [52]. For example, game theoretic models used
to analyze field data on East African pastoralists showed that wealthy households can afford to
enforce conservation of grazing areas, incurring the costs of policing other households from
overgrazing [53].
Between-Group Interactions
Because both conflict and peaceful interactions between groups can produce public goods
(e.g., territorial gains), the models reviewed above concerning leaders in collective action can
apply in this context. Leaders can also be modeled as volunteers in between-group conflicts.
One model [2] predicts that high-ranking individuals (leaders) who usurp a disproportionate
share of resources from group-mates will act seemingly altruistically in between-group conflict,
expending more effort and often having lower reproductive success than low-ranking group-
mates. Similar behavior is expected for individuals with[46_TD$DIFF] the greatest motivation, strength, or
endowments as well as for those paying the lowest costs. Analysis of conflict between
egalitarian groups showed that, if the collaborative abilities of individuals are relatively low,
then the population becomes dimorphic, with a small proportion of ‘leaders’contributing to
public goods and the rest free-riding [21].
General Principles
Several general lessons emerge from our theoretical synthesis. First, leaders and followers can
emerge naturally as a result of heterogeneity in preferences, motivation, personality, physical
characteristics, information available,and other features affecting individual performance in fitness-
related activities. Under some conditions, this heterogeneity is predicted to be favored by natural
selection. Several models predict relatively low frequencies of leaders, but the results depend on
modeling details. Some models also predict that leaders’preferences and choices will strongly
control those of followers, but theoretical work on this is very limited. To the extent that differentia-
tion of group members into leaders and followers, and subsequent division of labor between them,
is driven by natural selection, leadership tends to be beneficial to both social groups and individual
members of these groups. Although sometimes followers can have lower fitness than leaders, this
is most likely when the alternatives to accepting the follower's position (e.g., leaving the group
or challenging the leader) are more costly. In some situations, group differentiation into leaders
and followers is predicted to be maintained by negative frequency-dependent selection, such that
both types will have equal fitness at equilibrium. Under other conditions, leaders behave as
volunteers and can have lower payoffs than their largely free-riding followers. In sum, there are
many ways in which leadership can evolve, some less intuitive than others.
Empirical Patterns in Leadership
To focus the empirical portion of our review, we compared leadership patterns across mam-
malian societies by examining leadership in a sample of non-human and human societies
(Table 1). We restrict our sample of non-human mammals (NHM) to species for which sufficient
data on social interactions of free-living individuals are available. For comparability, we included
only small-scale societies (SSS) of humans lacking complex political institutions (e.g., more
than two formal administrative levels) [11,54]. Our rationale for comparing human and non-
human leadership is to help to determine the extent to which a unified evolutionary analysis of
mammalian leadership can reveal and explain similarities as well as differences. Although some
may assume that leadership among humans is intrinsically unique, we have no a priori position
on this question, nor on whether we expect greater variation in leadership within or across these
groupings (NHM and SSS).
The domains and dimensions delineated in Box 1 provide a useful framework for describing and
comparing leadership patterns across societies. We coded each dimension for each domain in
our 16 societies according to a five-point rating system (Table S1 in the [16_TD$DIFF]supplemental [17_TD$DIFF]information
Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy 5
TREE 2008 No. of Pages 13
online), drawing on observational data collected by us or in the published literature (supple-
mented by consultation with the original field researchers when needed). Each rating is akin to
an item on a Likert scale, a measurement tool utilized extensively by psychologists and
marketing researchers.
This method is clearly preferable to purely qualitative comparisons, but has significant limitations.
One could raise epistemological concerns: do different raters have different implicit norms of
reference, for example about what might constitute ‘weak’versus ‘strong’levels of power? We
attempted to minimize this problem by explicit definitions of the possible ratings for each
dimension (Table 2), scoring each of the examples in dialogue within our group, and consulting
with experts in the field for those groups not studied directly by one of us. We view the empirical
portion of this review as a foundational contribution towards a unified approach to studying
leadership. Methodologically, it is a survey of current expert opinion on a sample of societies,
with the aim of stimulating further research.
Because leadership dimensions (other than generality) can vary independently across domains
even in the same society, and leadership can be absent in some domains for a given society, we
present the results by domain. For each dimension other than generality, we analyzed ratings to
determine whether these varied significantly by leadership domain or society type (i.e., SSS
versus NHM) (Table S2).
Leadership Distribution
The distribution of decision-making within a group (i.e., the proportion of group membership that
makes decisions) is of particular interest to biological and social scientists. In the models
reviewed above, this can range from individual autonomy through consensus decision-making
to despotic control.
Table 1. Sixteen Societies Included in the Comparative Analysis
a
Society Units for Collective [11_TD$DIFF]Action ([12_TD$DIFF]Typical [13_TD$DIFF]Number of [14_TD$DIFF]Individuals)
[15_TD$DIFF]Non-humans
African elephant (Loxodonta africana) Multi-family bond group (25–50), matrilineal family groups (10–20)
Bottlenose dolphin (Tursiops sp.) Pod (10–30), alliances (5–14), other subgroups
Chimpanzee (Pan troglodytes) Community (46), other subgroups
Lion (Panthera leo) Pride (4–18), other subgroups
Meerkat (Suricata suricatta) Extended family group (‘clans’) (20)
Plains zebra (Equus quagga) Herd (40), harems (4 adults + offspring)
Spotted hyena (Crocuta crocuta) Clans (45–90), other subgroups
White-faced capuchin (Cebus capucinus) Multi-male group (5–39)
Humans
Ache (Paraguay forest) Residential band (35)
Cheyenne (North American plains) Tribe (4000), other subgroups
Inuit (Canadian arctic) Residential band (60)
Kipsigis (Kenyan savannah) Community (200–300)
Nootka (Canadian Pacific coast) Winter village (200)
Pimbwe (Tanzanian woodlands) Tribe (25 000), village (dozens to hundreds)
Shoshone (North American Great Basin) Camp (10–50), larger temporary aggregations
Tsimane (Bolivian Amazon) Village (dozens to hundreds)
a
See the [16_TD$DIFF]supplemental [17_TD$DIFF]information online for references and details of the societies sampled.
6Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy
TREE 2008 No. of Pages 13
We found a similar overall pattern for SSS and NHM societies, with the majority of societies
featuring moderately shared leadership in most domains (Figure 1 and Table S2 for details).
Leadership was rated as significantly more concentrated (less shared) in the domains of within-
group conflict resolution and between-group interactions than in the other domains. There was
no significant effect of society type (SSS vs NHM). However, leadership in domains other than
food acquisition is less-evenly distributed in NHM than SSS. Food acquisition and consumption
was done individually in most of the NHM societies, but was a cooperative matter within most of
the SSS societies [55,56].
Leadership Emergence
In some cases, leaders emerge through a process of competition or performance. In others, they
are born, not made –by inheritance or some other process independent of their own actions or
qualities. Social scientists typically term this contrast achieved versus ascribed leadership
status, labels we adopt here. This ascribed–achieved dimension is a continuum which for
convenience we divide into five steps (Table 2). Even in the same society, leadership in one
domain can be primarily achieved even though in another domain it is ascribed. In common with
many other status differences, leadership is often correlated with age, as one builds up
experience, alliances, and/or competitive ability; we classify this as achieved status, unless
there is an invariant seniority rule (e.g., the oldest female always leads)[47_TD$DIFF]. Mathematical models
reviewed above, however, focus exclusively on achieved leadership status.
Table 2. Rating Codes for Leadership on Mammalian Societies
a
[10_TD$DIFF]
Dimension Rating [18_TD$DIFF]Codes and [19_TD$DIFF]Definitions
Distribution 1 = no leadership (all adults are autonomous in domain X)
2 = semi-autonomy: leadership in domain X is widely shared among a restricted age or sex
category (e.g., females usually lead, older adults usually lead)
3 = moderately unshared (roughly the midpoint between 1 and 5)
4 = very unshared (a small number of adults lead in this domain)
5 = highly unshared (one individual usually leads in this domain)
Emergence 1 = leadership fully achievement-bas ed (includes cases where adults are autonomous)
2 = primarily achievement-based (e.g., adults of one age or sex category usually lead, but
otherwise leadership is achievement-based)
3 = roughly equal mix of achieved and ascribed (e.g., most competent senior males lead)
4 = primarily ascribed (e.g., leadership inherited by a senior member of chief's family)
5 = fully ascribed (e.g., senior female always leads)
Power 1 = weak or non-existent (adult autonomy, highly democratic decision-making, etc.)
2 = delegated leadership (leaders coordinate or execute decisions most or all agree to)
3 = moderate power (roughly the midpoint between codes 1 and 5)
4 = leaders can coerce or persuade many but not all, or often but not consistently
5 = despotic (leaders consistently coerce or persuade others to follow)
Relative benefit n.a. = not applicable because no leadership (e.g., full autonomy)
1 = leadership very costly (small payoff relative to that of followers)
2 = leadership somewhat costly, on average
3 = equality (payoff to leadership is roughly the same as the payoff to followers)
4 = leaders gain moderately relative to followers, on average
5 = relative payoff is heavily skewed to leaders
Generality 1 = leadership in one domain rarely predicts leadership in other domains
2 = some correlation[1_TD$DIFF] (e.g., [20_TD$DIFF]individuals [21_TD$DIFF]belonging to one age or sex category usually or always
lead in every domain[22_TD$DIFF], but the precise identity of leaders varies)
3 = moderate correlation (e.g., consistent across two domains)
4 = strong correlation (e.g., consistent across three domains)
5 = leaders are consistently the same across all domains, independently of distribution
(including both cases with consistent autonomy and ones with generalized dominance)
a
See Box 1 for explanations of each dimension.
Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy 7
TREE 2008 No. of Pages 13
Our ratings of the 16 societies indicate that most exhibit achieved leadership in the majority of
domains (Figure 1). There are no systematic differences across domains in this dimension. We
found greater variability within categories (SSS, NHM) than between them, although NHM
societies are somewhat more achievement-based than SSS in domains other than movement.
Spotted hyenas and the Nootka (Northwest Coast Indians) are clear outliers with highly ascribed
(inherited) leadership in all domains.
Differential Power
We define power as the ability to motivate others to behave in ways they would otherwise not via
mechanisms of coercion, persuasion, or prestige. Leaders in SSS are often particularly skilled at
using rhetoric to express common goals and to persuade and inspire followers [57]. Few
mathematical models explicitly allow for evolving power of leaders (see above), predicting that
leaders’actions will have substantial effects on followers unless the latter form coalitions.
The ratings suggest that leaders are generally less powerful in SSS than NHM except in the food-
acquisition domain, where adult NHMs are often autonomous (Figure 1). Anthropological
research indicates that the smallest-scale societies of low-density foragers or horticulturalists
have less-powerful leaders than other human societies, a generalization borne out in our sample
(see ratings for Ache, Inuit, Shoshone, and Tsimane, Table S2). There are also significant
differences across domains, with leaders being rated more powerful in within- and between-
group interactions, and less powerful for movement and food.
Relative Benefit
Do leaders reap fitness payoffs from their actions equal to, greater than, or less than those
averaged by other group members? Although intuition might suggest that leaders gain dispro-
portionate payoffs, several models predict otherwise. If payoffs are frequency-dependent (e.g.,
leadership traits are favored when rare, but deleterious when common), fitnesses of leaders and
Movement
Within-group conflict resoluon
Distribuon
Distribuon
Emergence
Emergence
Power
Power
Relave benefit
Relave benefit
Rang
Dimension
6
5
4
3
2
1
0
6
5
Human
Key:
Non-human
4
3
2
1
0
Food
Between-group interacon
Figure 1. Ratings for Leadership Dimensions in Non-Human Mammals (white) and Small-Scale Human
Societies (shaded) in Each of the Four Leadership Domains. In each case, the central dark bar represents the
median and the box the inter-quartile range.
8Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy
TREE 2008 No. of Pages 13
followers will be equal at equilibrium. Self-serving leaders can be thwarted by coalitions of the
less powerful, or (particularly in human societies) by rules that remove them from power; one
mechanism to ensure that this is through delegated leadership, by which leaders serve the
larger group and have only limited autonomy to pursue their own interests.
The evidence from our 16 societies shows greater similarity between NHM and SSS in this
dimension than in any other (Figure 1). Notably, across most of our cases, leaders gain fitness-
related payoffs equal to or only modestly better than followers. A handful of cases exhibit
negative returns to leadership (i.e., apparent altruism or exploitation by followers), as docu-
mented among lions [58]. Neither domain nor society type has a significant effect on ratings for
relative benefit (Table S2).
Generality
Leadership is domain-specificin some groups, but fairly generalized in others [59].We
therefore asked if individuals who lead in one domain are likely to exercise leadership in other
domains. We expected societies with strong dominance hierarchies or charisma-based leader-
ship would exhibit higher generality than societies with widely distributed leadership. We only
located one modeling study allowing for the effects of leaders over multiple domains (within- and
between-group conflicts [2]), and therefore no model-based generalizations are possible yet.
We found that leadership is somewhat more generalized in NHM societies than in SSS; this was
true for both median ratings (median = 4 for NHM, versus 3 for SSS) as well as the range across
societies (three NHM but no SSS societies were rated 5, whereas three SSS but no NHM
societies rated 2; Table S2). As expected, societies with marked dominance hierarchies (e.g.,
capuchins, meerkats, hyenas) exhibit greater leadership generality across domains, reflecting
the fact that high-ranking individuals typically occupy most leadership roles.
Cross-Domain Synthesis
Because several of the rating dimensions are intercorrelated, we employed principal compo-
nents analysis to reduce the dimensions of the correlation matrix. The first two principal
components accounted for 55% of the variation. The first (39% of the variation) loaded
substantially on all variables, but was specifically associated with leaders being powerful,
leadership being concentrated, and the benefits of leadership being relatively large across
domains. We thus named this principal component ‘power’. The second principal component
(16% of the variation) specifically identified variables relating to ‘emergence’: societies scoring
highly on this principal component had leaders whose role was more ascribed, whereas those
with low scores had leadership that was more achievement-based. Figure 2 reveals no
systematic divergence between human and non-human mammals; some SSS have low
leadership power scores while others have some of the highest. There is similar variation
amongst SSS for the emergence dimension.
Towards a Unifying View of Leadership
Examination of leadership in a sample of NHM and human SSS in a cross-species framework
revealed some suggestive patterns. We found significant variation across domains for two
dimensions, distribution and power, with leadership being the most concentrated and powerful
in conflict mediation and between-group interactions. Comparing humans to other mammalian
species in our sample, we found both similarities and differences. Despite common assumptions
often made around the notion of human uniqueness, interestingly, we found no clear divide
between human and non-human social mammals with respect to the emergence of leadership in
this exploratory analysis. Instead, both NHM and SSS leadership is based more on individual
achievement than on ascribed (inherited) status, although exceptions to this pattern should
motivate future theoretical and empirical analyses. In addition, both NHM and SSS leaders tend
Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy 9
TREE 2008 No. of Pages 13
to wield more power in the domains of within-group conflict-resolution and between-group
interactions, consistent with theoretical analyses of the importance of multilevel selection [60] as
well as within-group dynamics [2]. Furthermore, the paucity of coercive leadership in small-scale
(but not complex) human societies is arguably similar to the way NHM use communicative
or passive mechanisms, but rarely coercion, to recruit followers [9]. These continuities between
humans and other mammals in leadership patterns likely reflect shared evolved cognitive mecha-
nisms governing dominance–subordination, alliance formation, and decision-making [3,15].
On the other hand, SSS leaders in our sample generally wield less power than do NHM leaders,
and leadership is marginally more domain-general in the societies of NHM than in SSS. These
differences reflect a tendency for humans to develop role specialization and depend less on
dominance as the basis for leadership, which arguably indicates significant evolutionary diver-
gences between humans and NHM [61,62]. Even in the least complex human societies, the
scale of collective action (for food acquisition and distribution, conflict mediation, and between-
group interactions) is greater and presumably more crucial for survival and reproduction
[[48_TD$DIFF]13,54,63].
Pimbwe
Hyena
Elephant Nootka
Meerkat
Cheyenne
Dolphin
More powerful
Principal component 1
Principal component 2
Chimpanzee
Zebra
Capuchin
Kipsigis
Inuit
Tsimane
Ache
Shoshone
Lion
8
4
0
–4
–8
–8 –4 0 4 8
More
ascribed
Figure 2. Position of Each Sample Society in the 2D Space Defined by the First Two Principal Components
Extracted from the Set of Ratings for Four Dimensions (Excluding Generality) in All Four Domains.
10 Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy
TREE 2008 No. of Pages 13
We hypothesize that the factors causing human leadership to potentially diverge from the
prevailing mammalian patterns are those responsible for other aspects of human uniqueness
[64], but additional empirical tests are clearly warranted to investigate this possibility. Biologically
evolved characteristics facilitating syntactic language and high-volume cumulative cultural
transmission underpin the development of norms and institutions shaping all aspects of human
social behavior [63,65–67], including leadership [68]. Human socioeconomic systems are
characterized by massive flows of goods and services among large numbers of individuals
who are often non-kin. Such a social system appears to favor the emergence of mechanisms to
solve complex coordination problems [69]. Human cognitive capacities for planning and
communication facilitate solution of more challenging collective action problems that benefit
most members of the cooperative social units [21,67]. In such contexts, coercion is not
necessary to motivate group members to follow leaders [13].
Our review reveals notable gaps in leadership studies that differ for SSS and NHM (see
Outstanding Questions[9_TD$DIFF]). Leadership has rarely been quantified in domains beyond group
movement for even the best-studied NHM. Future studies should investigate the degree to
which this lacuna in the literature reflects actual patterns of leadership in NHM. In SSS,
leadership is poorly described in the domains of movement and food acquisition, perhaps
because of the rarity of formal leadership roles; here again new studies will be necessary to
examine the dimensions of collective decision-making in these domains. We also have little
systematic understanding of the variable fitness costs and benefits associated with leadership
status in SSS [68] or NHM societies [9]. Existing theoretical work spans multiple domains of
leadership, but most focuses on only three dimensions of leadership: distribution, emergence,
and payoffs. An important direction for future theoretical work is to provide modeling guidance
on evolutionary processes driving variation in the power of leaders and the generality of
leadership across multiple domains. We hope this review helps to stimulate greater integration
of empirical work and model-building on this emerging research topic.
Acknowledgments
This project was sponsored by the National Institute for Mathematical and Biological Synthesis, supported throug hNational
Science Foundation awards EF-0832858 and DBI-1300426, with additional support from The University of Tennessee,
Knoxville. S.G. was supported by the US Army Research Laboratory and the US Army Research Office under grant number
W911NF-14-1-0637. P.H. thanks C. von Rueden, H. Kaplan, and M. Gurven for helpful discussion regarding the Tsimane.
J.E.S. was supported by funds from Faculty Development Funds and from the Provost's Office at Mills College. M.v.V.
received funding from the Netherlands Science foundation (NWO). C.H. was supported by the National Science and
Engineering Research Council of Canada (NSERC, RGPIN-2015-05795) and the Foundational Questions in Evolutionary
Biology Fund (RFP-12-10). We are grateful to Richard Connor, Craig Packer, Elizabeth Archie, and Daniel Rubenstein for
advice on rating particular species, and Britt Pimental for bibliographic assistance.
[49_TD$DIFF]Supplemental Information
[49_TD$DIFF]Supplemental [50_TD$DIFF]information associated with this article can be found, in the online version, at http://dx.doi.org/10. 1016/j.tree.
2015.09.013.
References
1. Hamilton, I.M. (2013) The effects of behavioral plasticity and lead-
ership on the predictions of optimal skew models. Behav. Ecol. 24,
444–456
2. Gavrilets, S. and Fortunato, L. (2014) A solution to the collective
action problem in between-group conflict with within-group
inequality. Nat. Commun. 5, 3526
3. King, A.J. et al. (2009) The origins and evolution of leadership.
Curr. Biol. 19, 1591–1682
4. Johnstone, R.A. and Manica,A. (2011) Evolution of personality differ-
ences in leadership. Proc. Natl. Acad. Sci. U.S.A. 108, 8373–8378
5. Boinski, S. and Garber, P.A. (2000) On the Move: How and Why
Animals Travel in Groups, University of Chicago Press
6. Conradt, L. and Roper, T.J. (2005) Consensus decision making in
animals. Trends. Ecol. Evol. 20, 449–456
7. Couzin, I.D. et al. (2005) Effective leadership and decision making
in animal groups on the move. Nature 433, 513–516
8. Petit, O. and Bon, R. (2010) Decision-making processes: the case
of collective movements. Behav. Process. 84, 635–647
9. Smith, J.E. et al. (2015) Collective movements, leadership and
consensus costs at reunions in spotted hyenas. Anim. Behav.
105, 187–200
10. Anderson, C. and Brown, C.E. (2010) The functions and dysfunc-
tions of hierarchy. Res. Organ. Behav. 30, 55–89
Outstanding Questions
The evolution of leadership is a rapidly-
growing area of interest in both biolog-
ical and social sciences (and particu-
larly in transdisciplinary work that cuts
across the boundary between them).
Here we simply list some of the out-
standing questions that have yet to be
addressed in a systematic way.
What drives variation in leadership pat-
terns across species or cultures? For
the distribution dimension, it seems
clear that neither consensus-based
nor concentrated leadership constitute
inevitable equilibria, but rather reflect
different points on a dynamic contin-
uum maintained by some combination
of internal and external tradeoffs and
opportunities. Although widely distrib-
uted decision-making may maximize
efficiency under some conditions, it
can be very inefficient under others.
In addition, heterogeneity among indi-
viduals in ability, motivation, and inher-
ited status or resources may
concentrate leadership.
Is the predominance of female leader-
ship by adults seen in our sample of
NHM generalizable to other species? Is
the bias towards male leadership seen
in our SSS sample a robust contrast?
Salient attributes such as age and sex
have yet to be systematically quantified
across species within a phylogenetically
controlled context, despite our efforts
here. Understanding attribute-based
leadership should facilitate the devel-
opment of mechanistically inspired
hypotheses, and illuminate the cognitive
basis of leadership traits.
We summarize evidence that benefits
are fairly evenly divided between lead-
ers and other group members. How-
ever, this might reflect the fact that we
differentiate leadership from domi-
nance, as well as restricting human
cases to small-scale societies. Under
what conditions can individuals convert
leadership into despotism? Many
social scientists are currently interested
in the dark side of leadership, when
leaders exploit followers (see next
point). There is also keen interest in
the evolutionary origins of charismatic
leadership, where followers voluntarily
obey leaders because of some quality
they signal (which immediately sug-
gests the relevance of signaling theory,
as yet barely exploited in leadership
studies).
Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy 11
TREE 2008 No. of Pages 13
11. Flannery, K. and Marcus, J. (2012) The Creation of Inequality: How
Our Prehistoric Ancestors Set the Stage for Monarchy, Slavery,
and Empire, Harvard University Press
12. Dyer, J.R.G. et al. (2009) Leadership, consensus decision making
and collective behaviour in human crowds. Philos. Trans. R. Soc.
Lond. B: Biol. Sci. 364, 781–789
13. Hooper, P.L. et al. (2010) A theory of leadership in human coop-
erative groups. J. Theor. Biol. 265, 633–646
14. Eerkens, J.W. (2010) Privatization of resources and the evolution
of prehistoric leadership strategies. In The Evolution of Leadership:
Transitions in Decision Making from Small-scale to Middle-range
Societies (Vaughn, K. et al., eds), pp. 73–94, SAR Press
15. Van Vugt, M. and Ronay, R. (2013) The evolutionary psychology of
leadership: Theory, review, and roadmap. Organ. Psychol. Rev. 4,
74–95
16. Pyritz, L.W. et al. (2011) Reaching a consensus: terminology and
concepts used in coordination and decision-making research. Int.
J. Primatol. 32, 1268–1278
17. Rands, S.A. et al. (2003) Spontaneous emergence of leaders and
followers in foraging pairs. Nature 423, 432–434
18. Bowles, S. and Gintis, H. (2008) Power. In The New Palgrave
Dictionary of Economics (Durlauf, S.N. and Blume, L.E., eds),
Palgrave Macmillan
19. Conradt, L. and Roper, T.J. (2003) Group decision-making in
animals. Nature 155–157
20. Conradt, L. (2012) Models in animal collective decision-making:
information uncertainty and conflicting preferences. Interf. Focus
2, 226–240
21. Gavrilets, S. (2015) Collective action and the collaborative brain. J.
R. Soc. Interf. 12, 20141067
22. Guttal, V. and Couzin, I.D. (2010) Social interactions, information
use, and the evolution of collective migration. Proc. Natl. Acad. Sci.
U.S.A. 107, 16172–16177
23. Barta, Z. and Giraldeau, L.A. (1998) The effect of dominance
hierarchy on the use of alternative Foraging tactics: a pheno-
type-limited producing-scrounging game. Behav. Ecol. Sociobiol.
42, 217–223
24. Mathot, K.J. and Giraldeau, L.A. (2010) Within group relatedness
can lead to higher levels of exploitation: a model and empirical test.
Behav. Ecol. 21, 843–850
25. Koykka, C. and Wild, G. (2015) The evolution of group dispersal
with leaders and followers. J. Theor. Biol. 371, 117–126
26. Challet, D. and Zhang, Y.C. (1997) Emergence of cooperation
and organization in an evolutionary game. Physica A 246, 407–
418
27. Anghel, M. et al. (2004) Competition-driven network dynamics:
emergence of a scale-free leadership structure and collective
efficiency. Phys. Rev. Lett. 92, 058701
28. Clemson, T. and Evans, T.S. (2012) The emergence of leadership
in social networks. Physica A 391, 1434–1444
29. Zimmerman, M.G. and Eguíluz, V.M. (2005) Cooperation, social
networks, and the emergence of leadership in a prisoner's
dilemma with adaptive local interactions. Phys. Rev. Lett. 72,
056118
30. Dávid-Barrett, T. and Dunbar, R. (2013) Social elites can emerge
naturally when interaction in networks is restricted. Behav. Ecol.
25, 58–68
31. Cartwright, E. et al. (2013) Leadership by example in the weak-link
game. Econ. Aff. 51, 2028–2043
32. Hirschleifer (1983) From weakest-link to best-shot: the voluntary
provision of public goods. Public Choice 41, 371–386
33. Olson, M. (1965) The Logic of Collective Action: Public Goods and
the Theory of Groups, Harvard University Press
34. Boyd, R. and Richerson, P.J. (1992) Punishment allows the evo-
lution of cooperation (or anything else) in sizable groups. Ethol.
Sociobiol. 13, 171–195
35. Sigmund, K. et al. (2001) Reward and punishment. Proc. Natl.
Acad. Sci. U.S.A. 98, 757–762
36. O’Gorman, R. et al. (2009) Constraining free riding in public goods
games: designated solitary punishers can sustain human cooper-
ation. Proc. Biol. Sci. 276, 323–329
37. Smith, E.A. and Choi, J-K. (2007) The emergence of inequality in
small-scale societies: simple scenarios and agent-based simula-
tions. In The Model-based Archaeology of Socionatural Systems
(Kohler, T. and van der Leeuw, S., eds), pp. 105–119, SAR Press
38. Hooper, P.L. et al. Ecological and social dynamics of territoriality
and hierarchy formation. In Principles of Complexity: An Introduc-
tion to Complex Adaptive Systems and Human Society, Princeton
University Press (in press)
39. Kohler, T.A. et al. (2012) An agent-based public-goods model for
the emergence of leadership in prehispanic Pueblo societies. Adv.
Complex Syst. 15, 1–29
40. Kohler, T.A. et al. Sociopolitical evolution in midrange societies:
The prehispanic Pueblo Case. In Principles of Complexity: An
Introduction to Complex Adaptive Systems and Human Society,
Princeton University Press (in press)
41. Powers, S.T. and Lehmann, L. (2014) An evolutionary model
explaining the Neolithic transition from egalitarianism to leadership
and despotism. Philos. Trans. R. Soc. Lond. B: Biol. Sci. 281,
20141349
42. Johnstone, R.A. (2000) Models of reproductive skew: a review and
a synthesis. Ethology 106, 5–26
43. Summers, K. (2005) The evolutionary ecology of despotism. Evol.
Hum. Behav. 26, 106–135
44. Buston, P.M. et al. (2007) Reproductive skew and the evolution of
group dissolution tactics: a synthesis of concession and restraint
models. Anim. Behav. 74, 1643–1654
45. Archetti, M. and Scheuring, I. (2012) Review: game theory of public
goods in one-shot social dilemmas without assortment. J. Theor.
Biol. 299, 9–20
46. Diekmann, J. (1985)Volunteer'sdilemma. Confl.Resol. 29, 605–610
47. Archetti, M. and Scheuring, I. (2011) Coexistence of cooperation
and defection in public goods games. Evolution 65, 1140–1148
48. Shen, S.F. et al. (2010) The brave leader game and the timing of
altruism among nonkin. Am. Nat. 176, 242–248
49. Mesterton-Gibbons, M. et al. (2011) Models of coalition or alliance
formation. J. Theor. Biol. 274, 187–204
50. Bisonnette, A. et al. (2015) Coalitions in theory and reality: a review
of pertinent variables and processes. Behaviour 152, 1–56
51. Frank, S.A. (1995) Mutual policing and repression of competition in
the evolution of cooperative groups. Nature 377, 520–522
52. Frank, S.A. (1996) Policing and group cohesion when resources
vary. Anim. Behav. 52, 1163–1169
53. Ruttan, L.M. and Borgerhoff Mulder, M. (1999) Are East African
pastoralists truly conservationists? Curr. Anthropol. 40, 621–652
54. Vaughn, K. et al. (2010) The Evolution of Leadership: Transitions
in Decision Making from Small-scale to Middle-range Societies,
SAR Press
55. Gurven, M. (2004) To give or not to give: an evolutionary ecology of
human food transfers. Behav. Brain. Sci. 27, 543–560
56. Hooper, P.L. et al. (2015) Inclusive fitness and differential produc-
tivity across the life course determine intergenerational transfers in
a small-scale human society. Philos. Trans. R. Soc. Lond. B: Biol.
Sci. 282, 20142808
57. Price, M.E. and Van Vugt, M. (2014) The evolution of leader-
follower reciprocity: The theory of service-for-prestige. Front.
Hum. Neurosci. 8, 1–17
58. Heinsohn, R. and Packer, C. (1995) Complex cooperative strate-
gies in group-territorial African lions. Science 269, 1260–1262
59. Lewis, J.S. et al. (2011) Highly dynamic fission-fusion species can
exhibit leadership when traveling. Behav. Ecol. Sociobol. 65,
1061–1069
60. Van Vugt, M. (2006) Evolutionary origins of leadership and follow-
ership. Pers. Soc. Psychol. Rev. 10, 354–371
61. Chapais, B. (2014) Complex kinship patterns as evolutionary
constructions, and the origins of sociocultural universals. Curr.
Anthropol. 55, 751–783
62. Hill, K.R. et al. (2011) Co-residence patterns in hunter-gatherer
societies show unique human social structure. Science 331,
1286–1289
63. Bowles, S. and Gintis, H. (2011) A Cooperative Species: Human
Reciprocity and Its Evolution, Princeton University Press
If, as we argue, leadership evolves by
natural selection, one might expect the
existence of special, conditionally
expressed adaptations that increase
the efficiency of both leaders and fol-
lowers in their respective roles. For
example, followers might evolve to be
less ‘rebellious’(or more willing to
accept their role) and leaders more
pro-social. Alternatively, a different tra-
jectory might arise, where leaders
evolve to be more powerful and manip-
ulative (‘Machiavellian’), and followers
more suspicious and stubborn. One
can arguably find evidence for both
trajectories in the historic and ethno-
graphic record of humans, and also in
studies of other mammals.
What accounts for the transition from
acephalous societies to more-complex
and stratified societies in human his-
tory? This question has preoccupied
scholars from Hobbes and Marx to
more contemporary analyses drawing
on evolutionary game theory and cul-
tural transmission dynamics. Do com-
parisons with (generally acephalous)
mammalian cases offer any insights?
Are any of the models of leadership
summarized in our review sufficient to
address this question? What are the
causes and consequences of the shift
from informal or situational leader-
ship to institutionalized forms in
human history, and do these have
any non-human analogs?
12 Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy
TREE 2008 No. of Pages 13
64. Hill,K. et al. (2009) The emergence of human uniqueness: characters
underlying behavioral modernity. Evol. Anthropol. 18, 187–200
65. Boyd, R. and Richerson, P.J. (2005) The Origin and Evolution of
Cultures, Oxford University Press
66. Kaplan, H. et al. (2005) The natural history of human food sharing
and cooperation: a review and a new multi-individual approach to
the negotiation of norms. In Moral Sentiments and Material Inter-
ests: The Foundations of Cooperation in Economic Life (Gintis, H.
et al., eds), pp. 75–113, The MIT Press
67. Smith, E.A. (2010) Communication and collective action: language
and the evolution of human cooperation. Evol. Hum. Behav. 31,
231–245
68. von Rueden, C.R. (2014) The roots and fruits of social status in
small-scale human societies. In The Psychology of Social Status
(Cheng, J. et al., eds), pp. 179–200, Springer
69. Gillet, J. et al. (2010) Selfish or servant leadership? Evolutionary
predictions on leadership personalities in coordination games.
Pers. Indiv. Differ. 51, 231–236
Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy 13