Darwin's monkey: Why baboons can't become human


  • Shirley C. Strum

    Corresponding author
    1. Department of Anthropology, University of California, San Diego, La Jolla, CA 92093-0532
    2. Uaso Ngiro Baboon Project, Box 62844, Nairobi, 00200, Kenya
    3. Institute of Primate Research, National Museums of Kenya, Box 24481-00502, Karen, Nairobi, Kenya
    • Department of Anthropology, University of California, 9500 Gillman Drive, San Diego, La Jolla, CA 92093-0532, USA
    Search for more papers by this author


Baboons were used in the past as models for human evolution. I utilize 40 years of data from my long-term study on baboons in Kenya to suggest that baboons are once again relevant for understanding human evolution, not as a referential model but to reset the starting conditions of the human experiment. The baboon data also offer a critique of widely held ideas about how natural selection might work by looking at real lives in real time. This situates competition in a matrix of collaboration and illustrates the critical role of chance, contingency, and history in baboon survival and success. I make three methodological moves to reach these conclusions. The first is to focus on process not just outcome. The second is to look at time scales longer than usual studies but shorter than evolutionary time as a way to open the black box that currently links behavior to evolutionary value. The third is to use comparative natural history, Darwin's method, as a way to capture and comprehend how complexity is generated and how baboons deal with it in their daily lives. These empirical and methodological turns lead to conclusions that run counter to widely held ideas about baboons, about primates, and about the determinism of natural selection. I follow my own research history to illustrate these points. The discussion ranges from alternative interpretations of the male and the female dominance hierarchies, to insights from a fission that happened when the foraging strategy of raiding and nonraiding baboons diverged, to evidence of adaptation after translocation, and finally to assessing two unusual fusions of baboon groups. Altogether, these natural histories yield baboon “principles of the social” with insights about cognition, cooperation, and culture and suggest why baboons can't become human. The data also support Weiss and Buchanan's framework (The Mermaid's Tale: Four Billion Years of Cooperation in the Making of Living Things. Cambridge, MA: Harvard University Press,2009. 305 p) with its alternative view of natural selection in which there is more slippage and tolerance, multiple solutions with larger acceptability spaces, and the possibility that an adaptive fit will be “good enough” rather than seamless. However, capturing behavioral complexity “in the wild” poses methodological challenges. Long-term field studies provide critical information but the current quantitative methods should be expanded also include natural history observations of behaviors and events across time, space, groups, and landscapes. Finally, the baboon natural histories illustrate how the evolutionary game has changed in the Anthropocene yielding a cautionary tale about the future for many primate species. Am J Phys Anthropol 2012. © 2012 Wiley Periodicals, Inc.

Baboons have played a central role both in our understanding of primates and as models for human evolution beginning with Zuckerman's1932 description of primate society. Based on his observations of Hamadryas baboons at the London Zoo, he argued for a society founded on sex, violence, and male domination of both females and males (Zuckerman,1932, 1933). Zuckerman can be forgiven for his mistakes because we knew very little about primates in nature at that time (Strum and Fedigan,2000b). Baboons continued to capture the scientific (and public) imagination three decades later through the early scientific work on savanna baboons done by Washburn, his student, DeVore, and the English psychologist, Hall (DeVore,1965). Once again baboons became emblematic of evolution; however, this time it was the fit between behavior and anatomy which told a convincing story about how natural selection worked (Washburn and DeVore,1961, 1963; DeVore and Hall,1965; Hall and DeVore,1965; Strum and Fedigan,2000b). Evolution gave male baboons the anatomy of aggression. Baboon males used those remarkable canines, large body size, impressive mantle of hair, and white eyelids in displays during sexual competition with other males and against predators. Fighting ability determined male rank which in turn governed reproductive success so that the best males were the most reproductive. The male dominance hierarchy also was the political structure of the group; it ordered male interactions reducing the need for frequent aggressive encounters. Males defended and policed; females raised babies. Males were central to all major group functions, whereas females were defined in terms of their relationships with males. Washburn and DeVore (1961) argued that we could infer the challenges that the savanna presented to a puny primate living away from the safety of the forest (read hominids) by looking through the lens of the only other primate that managed to do so successfully, baboons.

What follows is not a scholarly review of the baboon literature and its relevance to human evolution. Instead, I was invited to discuss my insights about baboons based on 40 years of observing them in the wild. Baboons have long since been dethroned as “the” preferred species, first by common chimpanzees (Goodall,1988/1971) and then by bonobos (Zihlman,1978; Kano,1992; Strum and Fedigan,2000b). In contrast, I will suggest that baboons are once again relevant for those interested in human evolution but in new ways. What follows is an argument that is mainly bottom up, grounded in baboon observations. In the end, I hope to illustrate that the baboons I have studied live in the most complex nonhuman primate society yet described and that they reside in the most socioecological complexity documented so far for a nonhuman primate. Furthermore, these baboon data suggest that focusing on real time and real lives provides a different view of how evolutionary processes work. Chance, contingency, and history play an important role in survival and success. This offers a critique of widely held ideas about natural selection, particularly the centrality of competition. The data also argue for the existence of multiple options (including “mistakes”) made possible by the intelligence and social skill that baboons use to build their primate society. I am often asked why, if baboons are so smart, they aren't human. There are many things wrong with the way this question is phrased. However, the perspective I present below does add a new factor to the list of obvious differences: the importance of managing complexity in daily life.

To reach these conclusions, I made three critical methodological moves. The first was to focus on process not just outcome. The second was to look at time scales longer than usual studies but not collapse everything into evolutionary time. The third was to use comparative natural history, Darwin's method. Primate studies rejected natural history observations when growing knowledge about primates grew and more sophisticated data collection methods were available (Strum and Fedigan,2000a). For now, however, natural history may be the only way to capture and comprehend how complexity is generated and how baboons deal with it in their daily lives.

These empirical and methodological turns lead to conclusions that run counter to widely held ideas about baboons, about primates, and about the determinism of natural selection. This is not simply an emphasis on proximate rather than ultimate levels of causal explanation (Tinbergen,1963). Instead, it tries to open the black box linking behavior and evolutionary outcome and raises the question of what can be legitimately assumed about shorter time scales when we make evolutionary time arguments. I do not review the extensive literature on all these topics. Instead, I include key references as a starting point for readers interested in exploring further.

My reference species is olive baboons, Papio anubis (Zinner et al.,2009). The reference material is the 40 years of my own research (see Strum and Fedigan,2000b) for a framework to situate most of these years). I track individuals, relationships, and social groups in two populations. The baboons lived in two locations because I translocated three groups in 1984 (Strum and Southwick,1987; Strum,2005). Translocation was an experiment that provided persuasive evidence about baboon adaptability. The new location, arid savanna occupied by nomadic pastoralists, has had both natural cycles of extreme events like droughts and in the last decade, human-induced changes that also created extreme events. Elsewhere (Strum and Western,1982), I have argued that the baboons don't distinguish between the sources (natural vs. human) of environmental change in their responses to change. This is particularly relevant to the scale and the processes I will explore. In this presentation, I touch on areas that are minefields of controversy, for example, what is culture or tradition and what is cooperation. Fortunately, my argument does not depend on resolving these controversies, a task I leave to others.

In this article, I move through a series of steps that reflect my own research history rather than their order of importance. During this time, I have gone from making baboons “almost human” by focusing on their social negotiation and politics (Strum,2001/1987a) to understanding better why baboons “aren't human.” I have collected quantitative data for four decades; however, I find these data inadequate to make sense of what the baboons are doing. Instead, I turn to comparative natural history for the missing insights. The title “Darwin's Monkey” refers both to the importance of Darwin's methodology as well as his prescient observations of baboons (Darwin,1871/2004).

My over-arching goal, in this article and in my research, is to recapture the complexity that exists in baboon lives and then to consider the implications of this complexity for evolutionary interpretations and for our current scientific practice. I end with a discussion of why understanding baboons is important for [physical/biological] anthropologists today.


Remodeling the male dominance hierarchy makes baboons “almost human” (1972–1980)

It is hard to believe in the 21st century how little we knew about primate behavior in the 1960s and how rudimentary were our methods. The pioneering work of the “classical ethologists” like Tinbergen, Lorenz, and von Frisch starting in the 1930s (Tinbergen,1942, 1963; Lorenz,1950) did two key things. It critiqued (previous) anthropomorphic interpretations of behavior and created a novel methodology that still relied on natural history [ethogram, fixed action patterns; critical periods; and interaction of field data with captive experiments (Tinbergen,1974/1951)].

When primate field studies resumed after World War 2, scientists still used natural history descriptions but added the “ethogram,” a careful and detailed description of the behavior of that species (or group, or population…these terms were interchangeable then). In the middle of last century, there were only dozens of studies of primates in the wild, not hundreds, and only a few species had been observed in more than one location (DeVore,1965; Altmann,1967; Jay,1968).

Variation in behavior seemed minimal, making broad generalizations about primate patterns seemed warranted (Washburn and DeVore,1961; Strum and Fedigan,2000b). However, by the early 1970s, variation in behavior and social organization began to accumulate. For example, patas monkeys solved the problems of savanna living with an anatomy adapted for rapid escape and behavior adapted for vigilant diversionary males and stealthy females and young (Hall,1968). While Washburn and DeVore's dominant males strode out to face a predator, Rowell's male baboons were the first to run to the trees, leaving the females and youngsters exposed (Rowell,1966). Natural history methods were necessary whenever a new species was studied; however, disagreements of “fact” were hard to resolve. Were the animals really different or were differences an artifact of observers and ways of observing?

When I began my baboon study on male and female roles in olive baboon society (1972–1974, Kekopey near Gilgil, Rift Valley, Kenya), there was already evidence from rhesus macaques on Cayo Santiago Island in the Caribbean that females, as well as males, had a dominance hierarchy and that matrilineal relationships were the core of the group (Sade,1967, 1972). As well, there was a move to use more systematic ways of data collection in the field [well summarized by Jeanne Altmann's often cited paper (Altmann,1974)]. All baboon projects from the early 1970s onward inherited a baboon ethogram from Hall, Washburn, DeVore, Rowell, and Ransom (Washburn and DeVore,1961; DeVore and Hall,1965; Hall and DeVore,1965; Rowell,1966; Ransom,1981). This meant that the data included the same basic elements of behavior. When possible (meaning in contexts where the primates could actually be seen), investigators relied on focal animal samples (follows for a fixed duration of randomly chosen individuals) as a way to minimize bias.

I collected my data this way. The results confirmed that matrilines (female based families) were the core of the group and that a matrilineal dominance hierarchy was the troop's main structure. Families, rather than just adult males, protected and policed the group on behalf of their own members. Females had “political” functions besides motherhood (see also Hausfater,1975; Ransom,1981).

However, the “males” I studied didn't behave as predicted. The category “males” includes large adolescents reaching the asymptote of their growth (Strum,1991) and fully adult males ranging in age from young (>10 years) to old (more than 20 years). Hausfater, on yellow baboons in Amboseli National Park in Kenya (Hausfater,1975; Hausfater et al.,1982), documented a relatively stable male dominance hierarchy that operated as predicted. The dominant male(s) had priority of access to limited resources and mating opportunities. The Hausfater study was a careful and quantitative refinement of the previous baboon model. By contrast, I found male rank relationships exceedingly dynamic (Strum,1982). Individuals spent time and energy working on their relationships with each other and tracking the changes in all male relationships. Processes of male maturation and male migration fueled the instability.

Besides its dynamic nature, the existing male hierarchy did not correlate with access to resources such as receptive females. Often rank did not even predict the winner of an aggressive encounter (Strum,1982). The reason was that baboons, including the males, had alternatives to aggression. These “social strategies of competition and defense” (Strum,1975a, 1976; Western and Strum,1983) could be as effective and were less risky than aggressive approaches. However, they depended on a number of factors for success. First, some relied on “special” social relationships. Ransom (1981) identified “special” relationships” between males and females and males and infants in the Gombe Stream baboons in his 1967 study but hadn't yet appreciated their function. Kekopey males created special relationships and later used them to advantage. For example, males used young infants as agonistic buffers against other males (Strum,1983c). Usually, it was the potential loser who got an infant, returning to the protagonist with the baby on his belly. Most often, the aggressive male backed off. A male could also use a female as a buffer (Strum,1983b). Why did it work?—because buffering allowed a male to mobilize troop support. Normally, other group members don't get involved in male–male interactions. However, the group will mob a male whose aggression causes distress to infants and females. A pivotal element in buffering as a successful tactic is trust. Trust is a social construct for humans and for baboons. It makes social life predictable, creates a sense of community, and makes it easier for individuals to work together (Holmes and Rempel,1989; Misztal,1996, 2001; Rousseau et al.,1998; Kappeler and van Schaik,2006; Henrich et al.,2009; Sueur et al.,2011, b). In economics, trust is a way to reduce transaction costs, creating new forms of collaboration and it is a form of social capital. For baboons in the context of agonistic buffering, trust implies a particular social history. It means that the male has been predictably affiliative with the infant. Both the infant and the male have expectations of each other based on their social history. If the buffer trusts the male friend, it will scream at the aggressive opponent. Without trust, a male runs the risk of himself becoming the object of mobbing if the infant screams at him rather than at the opponent. This implies that the social relationship comes first; using the relationships in a social strategy comes later.

Males got other benefits from friendships with females. Following ethological tradition, I refrained from calling special relationships “friendships” until I was convinced that baboon and human friendships shared many characteristics (the term special relationship was replaced by the term “friendship” in the literature beginning in the 1980s). There is no rape among baboons, and therefore, while males might fight to get a temporarily exclusive association with a sexually receptive female, or “consort,” the consort male needs the female's cooperation to successfully copulate with her. Consorts may attract many male “followers.” Friendship between the consort male and female also influenced whether a male can keep the consort in the face of pressure from follower males (Smuts,1985; Strum,2001/1987b).

Other social tactics relied on social knowledge and assessment skills. For example, older males with less aggressive potential still got consorts and copulations. One way was through a maneuver I called “sidelines.” In sidelines, a male monitors the consort from a distance. He makes his move when aggression breaks out between the consort male and his close followers. Then, the more distant monitoring male dashes to the female. By the time the male aggression is over, the older male has copulated with the female (becoming her new consort), has skillfully led her away from the other males, and is deftly grooming her. Ironically, another avenue for males with little aggressive potential is to manipulate aggression between consort male and followers. He can do this as a follower or from the sidelines, escalating aggression by his involvement but usually leaving before another male turns on him.

An individual's age and length of residency in the troop predicted whether he used aggression or social strategies as well as his success (Strum,1978, 1982). All males followed the same social trajectory through their life history from socially unskilled maturing male with growing physical powers (using aggressive strategies) to the socially skillful mature male whose success depends entirely on social expertise when he is old (using social strategies of competition and defense).

Males weren't the only beneficiaries of social strategies. Infants and females could rely on the support of their male friend when aggression was directed at them. Even the close proximity of a male friend reduced social interference that a female or infant would receive based on rank and age.

Complexity 1: Social strategies in addition to aggressive strategies create multiple options

Science makes simplifying assumptions to be able to begin studying complex phenomena such as social interactions. The problem comes when the results of those investigations slide into thinking that such systems are simple. Uncoupling the tight link between male dominance rank and male evolutionary payoffs was the first evidence I had of social complexity among baboons. Baboons created and managed special relationships. They adjusted their behavior based on expectations from past interactions (Strum,2008). Social strategies were also evidence of social collaboration (see section on collaboration later) and exchange. Among baboons, even the physically impressive males needed the help of less imposing individuals to succeed. Baboons, it seemed, had no option but to play by the “golden rule.”

At this point in my research, negotiations between baboons stood out in bold relief. Human negotiation is usually a “dialog” between two or more individuals. The dialog aims for a common understanding about a course of actions that satisfies the interests of the parties involved (Homans,1958; Festinger,1985/1957; Misztal,2001; Raiffa,2003/1982; Pearce and Hans-Werner,2008; West and Turner,2009). I saw baboon negotiations as behavioral conversations about investments in, and payoffs of, special relationships. Clearly, such interactions select for very different skills than would aggressive competition in, for example, mating success (see also Sussman and Garber,2011). The conclusion: a primate with a big brain can create more alternatives than previous evolutionary models suggested. I argued that these kinds of negotiations, social strategies, and sexual politics made baboons “almost human” (Strum,2001/1987b).

A brief note about my use of the term “complexity” is warranted. The study of social complexity has generated a huge literature since these first baboon observations (e.g., see Byrne and Whiten,1989; Whiten and Richard,1997; de Waal and Tyack,2003; Cheney,2007). Despite this, there is no clear definition of social complexity (Strum et al.,1997). Whiten (2000) suggested that social complexity must include the dimensions of variability of response, size of group, number of relationships, instability, and predictability. Dunbar (1998) found a strong correlation between group size. Ironically, Robert Hinde's 1970's model of interacting levels that moves from individuals to interactions to relationships to group structure is probably the best approximation to an operationalized definition of social complexity [Hinde,1976, 1987; and see Forster and Rodriguez (2006) for an example of how this would work]. In another section, I offer a specific definition of baboon social complexity. Until that point, I use data to build the dimensions of baboon complexity and use an everyday definition of complexity: being intricate, a system with numerous elements and many forms of relationships which are all “entwined (Latin “complexus”); complex systems have more parts and more connections than simple ones (Dictionary,1971).

Remodeling socioecology: Group size does not determine group fission (1981–1984)

Today, group size is thought to be a trade-off between the advantages (predator protection, resource discovery, and for primates, traditional knowledge of resources) and disadvantages (feeding competition and inefficiency) of living in a large group when compared with small group (Wrangham,1980; van Schaik,1983; Isbell,1994; Isbell and Young,2002). Group fission happens regularly among baboons (although there is a dearth of published information: for baboons, see Stoltz,1972; Nash,1976; Hamilton and Bulgar,1993; Henzi et al.,1997; Wasser et al.,2004; Van Horn et al.,2007; but for macaques, see Furuya,1969; Chepko-Sade,1974; Chepko-Sade and Sade,1979; Dittus,1988; Kuester and Paul,1997; Okamoto and Matsumura,2001; Kawamoto,2010). Theoretically, a group should split when it gets too large to forage effectively (increased feeding competition and/or has to travel too far to feed that many individuals), or too large to monitor and manage the many social interactions, or for both reasons.

In 1981, the main study group, Pumphouse, fissioned. The split centered on a disagreement about foraging strategies not about group size as Pumphouse was well below the numbers recorded in other baboon fissions (see above). An agricultural cooperative bought Kekopey Ranch, the study site, in 1976. Kekopey covered 45,000 acres. Although it was a cattle ranch, there were few cows, a large complement of wildlife, and minimal farming potential. By 1981, small-hold farms (shambas) appeared in the baboons' traditional home range. Then human–baboon conflict began (Strum,2001/1987b, 2010). Young adult and large adolescent Pumphouse males were the most motivated to raid. Troop movements away from the sleeping sites included extensive negotiations between these males and the rest of the group. Some days, the raiders pulled the rest of the troop with them but on other days they failed. Then only the raiders went to the fields. There was a dramatic increase in anthropogenic injuries and deaths because of the conflict (Strum,2010). Pumphouse spent over 6 months in the process of splitting. During this time, there were constant negotiations between raiders and nonraiders as well as among family members of raiders and nonraiders as they decided which group they would join. In the end, Pumphouse, the larger descendent group remained primarily natural foragers, whereas the smaller descendent group (Wabaya) became dedicated raiders.

However, these weren't the only responses. A third study group, Eburru Cliffs (larger than the other two and having the largest home range), tried human food, sustained the costs of conflict, and then completely abandoned the settlement area. They remained natural foragers using the part of their traditional home range that was as far away from the settled area as they could get (a distance of over 7 km as the crow flies).

Why raid? Human food offers significant benefits to baboons. Both males and females grew faster, reached adulthood earlier, and had heavier final weights (Strum,1991, 2010), a result found in other studies of food enhanced baboons (Altmann and Muruthi,1988; Muruthi et al.,1991; Altmann et al.,1993). Faster growth meant earlier female reproduction. Although the sample size is small, raider females reached menarche sooner and produced their first baby earlier than nonraider females. Interbirth intervals were significantly shorter in raider females when compared with nonraiders. An extreme example of this was the lowest ranking raider female. Her fecundity increased significantly when she became a raider, which reversed her previous age-related reproductive decline. Infants who grew up in food-enhanced conditions reached menarche and first birth earlier even though these landmarks occurred after they were translocated to a resource-poor environment. Infant survivorship did not differ between the raiders and nonraiders despite initially high rates of injury and death associated with raiding. With time, raiders adjusted to the risks of being close to humans so that costs declined (Strum,2010).

What were the baboons doing during those 6 months of the fission? The matriarch of the lowest ranking family provides one illustration. One day she would be with the raiders; the next day she was with the nonraiders. This went on for months. Her already independent infant, who still closely associated with his mother, moved back and forth with her unlike the female's adolescent sons who remained with the nonraiders. Eventually, after many migrations back and forth, that female became a permanent member of the raiding troop but her infant did not. He returned to the nonraiders and stayed with his brothers. Months later, the matriarch's subadult daughter also became a member of the raider group.

Females could have several reasons for joining the raiding group, Wabaya. Most often, the emigrating female was following her closest male friend who happened to be a “raider.” A female's rank in her matriline also played a role. The females who left Pumphouse for the raider group were most often the oldest daughters and hence the lowest ranking females in their matriline. By joining the raider group, they traded kin support for the benefits of male friendship and of raiding. They also avoided internal matrilineal disputes and reduced the number of females to compete with because Wabaya was smaller. Three pairs of sisters transferred to the raiding troop. The social framework that underpinned their previous rank relationship to each other disappeared and their rank reversed in Wabaya. The older sister became dominant to the younger one. By contrast, some things didn't change. Despite having only a few matrilines represented in Wabaya, the original Pumphouse female hierarchy was perfectly reproduced albeit with some females missing.

Complexity 2: Context becomes history as baboons negotiate socioecology

The study troop did not split because it was too large. What had happened during the fission added evidence about the socioecological complexity of baboon lives. Most individuals selected not to raid despite the major documented benefits of the raiding lifestyle. Individuals with the same characteristics chose different paths. To my surprise, other factors trumped evolutionary principles like optimal foraging and kinship as well as derived principles like hierarchy.

Why didn't all males become raiders? Large subadult and young adult males had the most to gain in terms of growth and weight (and reproduction?) yet not all of them made the switch. Not all females became raiders. I expect that the conservativeness of baboon females played an important role and for a few females a special social relationship (friendship) with a raider male was the defining influence. However, even in this smaller subset, there were exceptions: the lowest ranking female and her daughter were not friends with any raider male when they joined the raiding group. Equally important, not all troops with the opportunity chose to raid.

Paying attention to the process, to the natural history of events and individuals, showed me baboon complexity beyond social strategies and added to my appreciation of how evolution works at the level of real lives.

Translocation: Survival depends on others not on competition between individuals (1984–2000)

Darwin relied on the idea of “survival of the fittest” to explain evolution by natural selection (Darwin,1859). Subsequent evolutionary interpretations of animal (primate) behavior focused on competition as the key to individual survival. Sociobiology (Wilson,1975) enlarged the “individual” to include kin in the mid-1970s [inclusive fitness (Hamilton,1964)], and recently, group selection has been resuscitated as possible and important for humans (Richerson and Boyd,2005; Leigh,2010; Nowak et al.,2010; West et al.,2010). Usually, scientists infer what behavior is crucial to survival and reproductive success. By contrast, the translocation of the study baboons became an actual test of survival where the process could be documented and the outcome could be measured.

Despite great success with crop-raiding research and control techniques, in 1984, I had to translocate three troops of wild baboons (Pumphouse, Wabaya—the raiders, and a third troop called Cripple who were also in conflict with people), a total of 132 individuals. This was the first scientific translocation of a primate group (Strum and Southwick,1987). The translocation provided a unique opportunity to document adaptability (possibly adaptation) in action. I could compare the same individuals and the same troops in two radically different environments. The translocation site, on the Eastern Laikipia Plateau, is arid savanna with half the annual rainfall of Kekopey, the baboons' previous home. Serious droughts are common. The baboons were familiar with only about half of the foods available in the new location and the unknown ones turned out to be critical to survival during seasonal bottlenecks. The release site was not ideal; however, it was the only site I could get.

The descendants of the translocated baboons are alive today. The process by which they survived provided more evidence of the complexity of baboon life and illustrated evolutionary raw material.

Primates are social; it is a key characteristic of the Order. There are a number of theories about why most primates live in a social group (see above); however, it is difficult to document the value of group living during the course of most studies. The baboon translocation provided one chance. The benefits of living in an intact social group permeated almost every aspect of adjustment during the translocation.

To start, I manipulated existing social networks to ensure that the troops stayed where they were released. I held adult males captive at the release site, whereas the females and youngsters were set free. As expected, the females did not go far without the males. By the time the males were released, the females had already settled in and female conservatism checked the male tendency to wander (Strum,2002, 2005).

Each of the three troops was released in a different place and at different times. The troops maintained cohesion and coordination from the first days of freedom. The group represented a social learning resource in a more basic sense than we usually assume. Individuals in the translocated groups demonstrated flexibility in learning in novel settings. Social cues figured prominently in the process of acquiring information. For example, the translocated troops had to construct a diet in this unfamiliar place. Social facilitation played a role as was documented in another context before translocation (Strum,1983a). The incorporation of new foods into the diet was strongly related to specific social events. For example, adults in translocated troops watched and followed indigenous baboon troops. The strangers built their home range by tracking the path of local troops and foraging on some of same foods. In addition, when local males migrated into the translocated troops, they brought indigenous information. The translocated baboons' behavior tracked the immigrant male's behavior. In this way, they learned to find water in dry sand rivers and began using several novel large succulents (Sansevieria intermedia, Sansevieria abyssinica) for both moisture and food.

Another value of the social group could only be called “emotional support” (Hannah and McGrew,1991). This was evident during the accidental fission of one group at release and their subsequent reunion. Loud greetings, extensive embracing, and other amplifications of typical baboon reunion signals punctuated this meeting (Strum,2001/1987b). It was the exaggerations of these normal behaviors that pointed to the stress of separation and the relief of reunion.

A comparison of the baboon translocation with subsequent primate reintroductions and translocations add force to my conclusions. Survival skills are lost in captivity. Artificially created groups fall apart (Kawai,1960; de Vries,1991; Vie and Richard-Hansen,1997). Under normal circumstances (and in extraordinary conditions), the social group provides the context for appropriate learning and as a resource for that learning (King,1994; Russon,1999; Crast et al.,2010; Leca et al.,2010). Reintroduced captive primate groups fail to meet the learning challenges of their wild habitat even after extensive pre-release training (Kleiman and Rylands,2002).

The translocation gave other insights about adaptability, chance events, and the importance of the social. A night-time encounter with elephants was an “accident” that changed the future of one of the translocated troops (Strum,2001/1987b). Surrounded by elephants at their sleeping site, Pumphouse moved 5 km during the night, an unprecedented act for baboons and other diurnal primates. I know this because they marched past the lone human outpost in the area to reach another set of sleeping rocks at the edge of their home range. Subsequently, Pumphouse abandoned more than half of their range including the best rainy season portion. Following that move, group size declined as males and even mothers with infants left to join other, indigenous, baboon troops. Just when it seemed that Pumphouse might simply be absorbed into the local population, the exodus stopped. New males joined and the next stage in this troop's history began.

There were subsequent home range shifts but none as precipitous as the first. By 2000, Pumphouse ranged more than 20 km from their release site. By contrast, Wabaya, the previous raiders (now called Malaika trading “the bad guys” for the “the angels”) who were translocated to the same location stayed put from 1984 until 2007.

All but the first home-range shift tracked a resource gradient; however, information about and the attraction of food resources weren't enough to force a move. The shifts depended on social processes. Baboons are not territorial; adjacent groups have overlapping home ranges with undefended boundaries, although intensively used core areas are sometimes defended. Normally, the conservativeness of females prevents a baboon group from crossing home-range boundaries. Each subsequent Pumphouse range shift followed an influx of immigrant males [see Rowell's suggestions about sex-biased knowledge (Rowell,1972)]. These immigrants had firsthand experience of the resource gradient because they came from there. However, even with that knowledge, one male was ineffective. It took a cohort of four or more males, all indicating in the same direction during troop movement, to succeed. There was a time lag because the males had to first create friendships with females in order to influence troop direction. Clearly, knowledge was vital, but action relied on social processes.

Complexity 3: The social group is an organ of adaptation; accidents of history create different socioecological paths

The translocation was a real test not an imaginary one. If ever there was a situation where survival of the fittest should operate, where competition should have had an upper hand, this was it. Instead other baboons, either from within the translocated group or the indigenous population facilitated both learning and survival. The translocation, even more than the crop raiding, pointed out how evolutionary principles get embedded in a specific time and place and why context matters. Each group's “natural history” illustrated a variety of different paths. Chance (like an encounter with elephants), contingency (how many immigrant males from which area), and evolutionary principles interact to generate baboon options and baboon responses. The process was infinitely more complex than explanations collapsed into evolutionary time might suggest.

Fusion of two groups: Baboons negotiate the nature of their social group (2000–2002)

Fusion, when two groups merge to become one, is rare in baboons, so rare that it has been observed only three times, once in baboons in Amboseli National Park in 1960s (Altmann and Altmann,1970; Altmann et al.,1985), once among baboons in Mikumi National Park, Tanzania (Wasser et al.,2004), and once at De Hoop in South Africa (Henzi et al.,2000). There are only a few other cases reported for cercopithecine monkeys despite the hundreds of thousands of hours that cercopithecines have been studied in the wild. These include two vervet fusions (Hauser et al.,1986; Isbell et al.,2002; Isbell,2009), one toque macaque fusion (Dittus,1987), and several cases of fusion in Japanese macaques (Maruhashi,1992; Takahata et al.,1994; Sugiura et al.,2002). Descriptions suggest that the fusion happened almost instantaneously (or over a short time for toque macaques). Two separate monkey groups come together in the evening at a shared sleeping site and leave the next day as an integrated merged group.

This is quite different from fission/fusion societies among primates where coming together and splitting into smaller, usually foraging groups, happens on a daily or weekly basis (Aureli et al.,2008). Chimpanzees (Lehmann and Boesch,2004) and spider monkeys (Chapman et al.,1995) are well-known examples.

I have documented four baboon fusions in just over a decade. The first fusion was 16 years after translocation and involved one of the translocated troops, Malaika (the angels, previously the bad guys) and an indigenous group, Soit Oitashe, that had been studied since 1985 (Barton,1993). What made this (and subsequent fusions) so distinct is that the process took more than 2 years. During that time, individuals in each group engaged in extensive negotiations about merging both within the group and across the groups.

The impetus for the first fusion involved several ecological events including the creation of a wildlife sanctuary on communally owned land that overlapped the baboons' home range. The sanctuary led to an increase in prey and predator numbers. Later, the large mammal prey species left. This made local pastoralists' dogs and then the baboons the prey. Predation rates rose to unprecedented levels.

Who did what and when during the fusion process depended on specific events and on daily negotiations involving particular individuals and age-sex classes. The translocated troop, Malaika, was the smaller of the two troops. Malaika shadowed the indigenous group, Soit Oitashe, initially, but roles switched as Soit Oitashe, the indigenous group, lost members to heavy predation.

The sum of the forces of attraction and dispersal determined how mixed the two groups might be at any one time. For example, adult and subadult females were very attracted to new babies in the other troop. Sexually receptive females also approached the other troop's males. Several females from Malaika were drawn to a male friend who had recently emigrated back to his natal troop, Soit Oitashe. Large subadult and adult males, in both troops, herded their females away from the vicinity of the other troop's males trying to keep the two groups apart. This was only a small part of the daily push and pull of individuals, relationships, and classes of animals.

Leaving the sleeping rocks in the morning no longer followed a standard procedure because each group had its own indicator animals and there was no certainty about who was going to follow whom. Foraging during the day was also problematic because Malaika's smaller home range was entirely contained within Soit Oitashe's larger home range. Initially, Malaika refused to cross its own boundaries when Soit Oitashe moved on. Instead, Malaika waited, intercepting Soit Oitashe on its way back to the sleeping site. As the fusion progressed, Malaika did cross this invisible line, albeit hesitantly, thereby benefiting from access to new resources. Eventually, the home range of the fused troop (now called Nabo, Maasai for coming together) represented a compromise between the two previous home ranges.

Feeding competition also played a role during the fusion process. For example, acacia flowers are a preferred and high-quality food (Barton,1993). Flowers emerge during the dry season. Their distribution is limited both at the start and at the end of blooming. During these times, Soit Oitashe displaced Malaika from the few trees that had flowers. In response, Malaika would sneak away to feed undisturbed and out of sight. However, Malaika paid a high price for this maneuver because when Soit Oitashe noticed, they searched for them and aggressively mobbed Malaika when they found them.

For two groups to merge, the two female hierarchies needed to be integrated. Initially, Malaika's high-ranking females kept pushing all of Soit Oitashe's females. With the support of their male friends, Malaika females managed to be in the middle of the integrated dominance hierarchy and forcing their way upward. However, Malaika males stopped supporting their females. Without male support, Malaika females dropped to the bottom of the integrated female hierarchy while maintaining their original rankings among themselves.

Specific individuals had disproportionate impact on the degree of mixing between the two groups. This included two Malaika females and specific males from both groups. Low-ranking Soit Oitashe females also became the most active and gave the most aggressive resistance to any attempt by Malaika females to push up the integrated female hierarchy. An illustration of the influence of specific baboons was the effort that one Soit Oitashe male exerted to keep his troop's females from interacting with Malaika. The day he emigrated, there was a noticeable increase in intertroop interactions and in spatial proximity between members of the two groups.

Given the length and extent of the process, I had to develop criteria to use to decide when the two troops had become one group. The criteria reflected what I felt were aspects of normal group sociality. These included the absence of previous troop bias in proximity, grooming and other social interactions, and troop movement negotiations with indicators and followers responding across previous troop lines. Foraging subgroups should be mixed, and the fused group should react as an integrated troop in encounters with other groups, during predations by baboons, toward new male immigrants, to separations and reunions, and to social disruptions. Of course, the fused group had to have an integrated female dominance hierarchy and an integrated male hierarchy. It therefore should not be a surprise that integrating two social systems took such a long time.

Why fuse? Why fuse now? The answers to those questions offer a test of major socioecological theories of socialness (see above and Dittus,1986). Some of the fusions reported in the literature took place because one or both groups were too small to be viable in terms of social and sexual interactions and predator protection. Protection against predation likely played a role in the first baboon fusion I documented, although the sizes of both troops before merging were within the normal range for olive baboons and had all age–sex classes represented. Yet before the fusion, both lost many group members to predators. By contrast, after the fusion, the merged group lost none despite the continued presence of predators in the area.

The details of negotiations suggest, however, that the baboon reality was much more complex than simply defense against predation. Malaika traded off feeding competition and protection from danger on a daily basis until the fusion was complete. Once fused, it bore the cost of feeding competition but benefited from enhanced predator protection. Soit Oitashe benefited from better protection while remaining dominant in feeding competition. The costs and benefits of fusion, therefore, were different for Malaika and Soit Oitashe. Costs and benefits also diverged for different age–sex classes and even for individual baboons. Moreover, the benefit to cost ratio changed based on season, location, and day. This push/pull of fusion was even more elaborate than the process of fission I had documented during the crop-raiding era.

Complexity 4: Baboons negotiate almost everything, even the nature of their social group

Baboon negotiations during the fusion were about the nature of the social group to a degree not yet documented for a nonhuman primate species. The fact that the process took over 2 years and involved dyadic, triadic, and polyadic interactions in the context of every basic aspect of baboon life suggests that the term “negotiation,” in its human sense, is appropriate for baboons as well.

The negotiations were complex. The socioecology of the fusion was complex. The interaction of levels was complex. However, there was a ladder of complexity [building on Hinde's model of levels (Hinde,1976, 1987)] from dyadic interactions, cascading upward to relationships and subgroupings, and clusters of subgroupings to troops each with their own histories and negotiating interests relative to each other and to higher and lower levels. This was embedded in environmental complexity, which operated at different spatial and temporal scales. As ecological factors are often not correlated (e.g., the effects of rainfall, food availability, predation, disease, and human impacts are often independent of each other), one simple principle cannot govern what the appropriate response will be. One translocated troop, Pumphouse, reacted as if the elephant risk was greater than the loss of food resources. Another translocated troop, Malaika, accepted the cost of feeding competition in exchange for increased safety. This does not make selection random or arbitrary. Rather, it points out an important constraint and obstacle: selection must operate on a range of point responses to challenges that are sometimes themselves complex and independent, suggesting that evolutionary explanations need to incorporate, rather than eliminate, complexity, chance, and contingency. At many times in a study troop's history, it was possible to identify probable evolutionary causes of baboon responses; however, these did not operate alone. For example, although it is tempting to conclude that predation pressure explained the first fusion, another indigenous troop in the same area, Musul, lost members to predators and yet didn't try to fuse.

My argument so far can be summarized as follows. Looking at the scale of baboon daily life, focusing on process rather than only on outcome, unearths layers of complexity and illustrates how complexity is generated. Outcomes are neither random nor deterministic in a way often assumed when scaling up to make evolutionary arguments. Evolutionary time speculations of cause and effect assume rather than clarify how evolutionary principles are situated by chance, contingency, and history. This is because they either lack the information or ignore the importance of context and “black-box” the relationships between real behaviors and evolutionary outcomes.

My approach is not just a focus on proximate rather than ultimate explanations of behavior. The baboon data challenge some of the fundamental assumptions about how behavior evolves. Tinbergen (1963) highlighted four questions that must be addressed in any comprehensive explanation of a behavior. His framework identified both the “how” proximate level (mechanism and ontogeny) and the “why” ultimate level (function and phylogenetic history). Yet, the ethologists, like Darwin before them, had to assume the feedback between proximate and ultimate levels while studying “static” behaviors. Equally important, ethologists had to examine simplified contexts gaining limited insights about how behaviors fit together into a complex whole or how that complex whole can change during the lifetime of the individual. The baboon data offer a dynamic and integrated view of real-life behavioral complexity. From this vantage point, the inevitability of an individual's choices and therefore the inevitability of the link between proximate and ultimate causation seems less compelling.


Managing complexity: Redefining the meanings of social

In 1987, Bruno Latour and I proposed a different approach to compare human and nonhuman primate society (Strum and Latour,1987). We were struck by the unlikely similarities between his subjects, scientists in laboratories (see Latour and Woolgar,1986/1979), and mine, baboons in the wild.

Baboon negotiation, even at the minor level I had documented back then, raised a profound question: does society exist or is it created? At the time, we contrasted two major views about society. What holds society together? Traditional approaches to this question assume that society exists and actors enter society adhering to “rules and a structure that are already determined” (Strum and Latour,1987, p. 785). This is the definition of an “ostensive” society. For most descriptions of society, it certainly seems that way. By contrast, in a performative society (Goffman,1971/1959; Garfinkel,2003/1967), individuals actively create and define society for themselves and for others. Therefore society, rather than being a preexisting structure, is actually constructed through “the many efforts to define it; it is something achieved in practice by all actors” (Strum and Latour,1987, p. 785). We argued for a shift to examine how actors create (perform) their society. In hindsight, we were changing focus from outcome to process motivated by our data on negotiation in baboon and human social groups. Many fields have since embraced this shift (see below).

Although we argued that the process might be similar for baboons and humans, it was clear that the outcomes are different. Transforming baboons into active performers of their society doesn't put them “on a par with humans.” We proposed that the difference is to be found in the practical means actors have to “enforce their version of society or to organize others on a larger scale.” Baboons have only themselves, their bodies (including brains and minds), social skills, and whatever social strategies and relationships they can construct as resources. Some factors are embodied and change slowly; age, sex, perhaps kinship and dominance need not be continually reconsidered. However, social skills are “soft tools” and with them baboons can build only “soft” societies. Other factors and contingencies must be continually tracked and “performed.” Baboons become the archetypical “competent member” of the ethnomethodologists (Garfinkel,2003/1967) constantly subject to the interference of others who are trying to do the same thing. They live in a “complex” society.

Humans, by contrast, bring extrasomatic resources to bear in performing society. Ironically, according to this framework, material resources, language, and symbols simplify the task of creating society and of strengthening the social bond. Human society becomes a “complicated” latticework, which looks ostensive because so much can be held constant over space and time. These material and symbolic assets create a material and symbolic structure that human actors now enter (the traditional view of society). As a result, humans accomplish two new aspects of the social link. They can simplify social negotiations by using extrasomatic resources. This makes it possible to build a larger more complicated social structure from the complex baboon society. (Our terminology may be confusing in light of the developments of complexity science in the last three decades and the interchangeable use of the terms “complicated” and “complex” in everyday language. However, in the 1980s, we felt the contrast embedded in our use of the two terms was appropriate and I continue to draw on that distinction. Baboons live in a complex society where many things impinge simultaneously, needing attention constantly. Humans live in a complicated society which like a snowflake is built of simple units expanded out to make a large complicated but not complex entity.) Baboons are limited in the ways they can manage complexity because they can't simplify much of their negotiation. This in turn limits the size and scale of their society when compared with humans.

An amusing illustration of this contrast between baboons and humans can be found in “hats” (Fig. 1). If baboons wore hats, they could simplify their social negotiations saving a lot of time and effort. These material symbols contain information about roles and expectations of behavior. If only baboons had “hats,” they could begin to shift from complex society where almost everything impinges simultaneously and needs to be continually negotiated toward a complicated society where negotiations are simplified, more focused, and produce more stable outcomes that can travel over space and time.

Figure 1.

What if baboons wore hats? (Credit: Dr. Shirley C. Strum©) [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

In summary, redefining the social link suggests that baboons and humans engage in the same process of creating society but use different resources. This produces divergent societies. Baboon society is complex because the ways to simplify, hold constant, black box are limited. Baboon complexity often verges on the edge of chaos as so many factors impinge simultaneously on negotiations. The baboon evidence for complexity and negotiation that I've gathered since we wrote the article bolsters the argument. The volume and scale of negotiations that exist each day create the need for structure that goes beyond kinship. Structure creates relative stability and predictability, reducing what otherwise would be intolerable transaction costs of living in a large and heterogeneous social group.

Recently, Shultz et al. (2011) modeled the evolution of stable sociality in primates. Their conclusion that there was a step-wise evolution from solitary to multi-male/multi-female groups, which only later reverted to one-male and pair-living family groups, contracts previous speculations but fits our framework very well. One solution to the complexity problem is to revert to smaller and less heterogeneous groupings. In fact, recent events in one baboon troop illustrate just this point. This troop appears to be simplifying social complexity through permanent one-male subgroupings that are not ecological foraging units found in fission–fusion societies or in modular multilevel societies like hamadrayas baboons (Swedell and Plummer, in press).

Managing complexity: Remodeling the female hierarchy so that rank simplifies complexity but is not linked to reproductive success (2001–2010)

The evidence just presented (and more that I don't have space to review) raises the question of how baboons deal with complexity and how much complexity they can handle? Dunbar (1996) suggested that the transaction costs of a typical primate grooming encounter limit the number of grooming partners an individual can have hence humans switch to gossip in the service of a much larger social network. Economics and political science also see human interactions in terms of transaction costs (North,1987; Coase,1998; Beccevra and Gupta,1999; Putnam,2000; Rao,2002). I suggest that baboons, too, have transaction costs similar to those identified for humans: finding the information needed for a particular interaction, the bargaining costs of reaching an agreement with another individual, and the policing and enforcement costs involved in monitoring and reassessing the relationship and the interaction. These seem rudimentary to any complex primate social relationship and exchange.

Historically, those studying primates generally assumed that social structure and social grouping exist. For example, DeVore/Washburn/Hall proposed the male dominance hierarchy as the organizing principle of the baboon group (Washburn and DeVore,1961; Hall and DeVore,1965; Strum and Fedigan,2000b). At the same time, Chance (1967) suggested “attention” as the mechanism. The discovery of a female matrilineal hierarchy in rhesus macaques on Cayo Santiago (Sade,1967) provided an alternative “structure” for a monkey group where males leave and females stay. Sociobiology later injected kinship and the gene calculus of inclusive fitness, sexual selection, and parenting strategies to create socioecological determinants of group configurations (Williams,1966; Trivers,1972; Wilson,1975). These strengthened the idea that female matrilines and the female dominance hierarchy were evolutionary adaptations and the primary structure of a baboon group.

Arguments about hierarchy, female or otherwise, have assumed that rank will be strongly correlated to reproductive success (Hausfater,1975; Altmann et al.,1988; Cheney et al.,2006). Yet, data since the 1970s show an inconsistent and often poor fit between baboon female rank and parameters of female evolutionary success. Instead, other factors appear influential like female sociality (Silk et al.,2003, 2010b), female age-specific fecundity (Strum and Western,1982), environment (Wasser et al.,2004), or a variety of factors that seem to cancel each other out (Brown and Silk,2002; Wasser et al.,2004; Silk et al.,2005; Crockford et al.,2008; Silk and Strum,2010).

The imperfect fit between female rank and reproductive success is not surprising if attention is on what actually happens. Then evolutionary principles operate in a context also shaped by complexity, chance, contingency, and history. The surprise is, instead, that there is a stable female hierarchy at all despite limited evolutionary benefits.

I suggest that the female dominance hierarchy has a different purpose. Its function is primarily to create predictability (and stability) so that transaction costs for females and others can be reduced. The complexity of baboon social life needs to be simplified. If the transaction costs of grooming limit the number of grooming partners [and stress is lower when a female baboon has fewer rather than many grooming partners (Crockford et al.,2008; Wittig et al.,2008)], imagine the social transaction costs (and stress) involved in any part of baboon daily routine if an animal didn't know in advance its relationship to others. Individuals, and by analogy the troop, would be paralyzed by minute-to-minute negotiations involved in the process of where to feed, rest, or travel, whom to approach, and whom to avoid. There would be no time for the basics of survival or enough time and energy to meet any new challenges.

If I am correct that the female hierarchy is primarily “structure,” then there need not be a strong correlation between female rank and individual reproductive success and no conundrum about why low ranking females stay in a group. All individuals benefit from predictability and stability no matter their rank.

However, why use the female hierarchy instead of some other structure to generate predictability and stability? I'm not suggesting that the baboons think about the problem and consciously create the solution. Instead, it is obvious that the female hierarchy is easily generated from evolutionary principles such as kin selection and reproductive value (Vx) (Williams,1966). In baboons and other cercopithecines, a mother's intervention on behalf of her younger offspring in altercations with an older offspring creates an internal family hierarchy in which the youngest sibling is the highest ranking. The normal expansion of a family over time would fashion a group hierarchy where only those closest in rank are likely to be related enough for kin selection to operate (Seyfarth,1977). The female hierarchy thus easily fulfills the need for a structure to reduce transaction costs. As baboons practice female philopatry, this hierarchy also organizes the largest cohort of animals in the group. Of course, there are other evolutionarily generated structures in a baboon troop including the male dominance hierarchy, male–female friendships, and other affiliative bonds (like those between play group members), but all of these are more dynamic, more ephemeral, and incorporate fewer individuals than does the female hierarchy.

If the female hierarchy is co-opted as primary structure, then it is the conservative nature of females [based on reproductive constraints of female mammals (Trivers,1972, 1974)] that helps to keep this hierarchy relatively stable and predictable. A male baboon who loses an encounter to another male might continue to contest the outcome throughout the day or even the week. A female who loses an encounter to another female rarely disputes the results. This contrast results from sex differences in risks and benefits.

However, the female hierarchy isn't invariant. I've seen what happens when the female hierarchy is unstable. This evidence supports my argument. The female hierarchy (in the different study troops) has changed a few times during the last 40 years. These have been mostly adjustments in rank between mothers and daughters, disruptions but not of major concern to anyone but the members of the family. By contrast, the few major changes that have occurred in the overall female hierarchy are instructive. They elicited lots of aggression some resulting in injuries. The aggression also disrupted basic daily activities like foraging and travelling. Disorder quickly spread from a few females through the group embroiling individuals not initially involved. Group life came to a standstill for a good part of several days while the continued instability affected individuals and the troop for weeks and even months. It is not hard to imagine, from these periods of disruption, why a stable and predictable structure is crucial not just for females but for the rest of the troop.

The possibility exists that there may be a better fit between rank and individual reproductive success under specific conditions (Cheney et al.,2004, 2006). For example, high rankers have preferred access when it is feasible to monopolize key foods during a resource bottleneck (Barton,1993). However, during the decades of my study, such situations turned out to be few, short, and did not produce the expected fit between female rank and reproduction (Strum and Western,1982; Silk and Strum,2010). This was true even during critical periods such as droughts probably because the foods left then are small, dispersed, and not defensible.

Given the complexity of baboon reality, individuals (and groups) need a way to manage daily life. It makes more sense, in this context, to view the female dominance hierarchy as a transactional principle than as a genetic characteristic of individuals or as the genetic consequences of the behavior of individuals. It is the baboon version of what humans do; “…an economizing behavior in the sense that it greatly reduces the transaction costs of social interactions and permits efficient collective action” (Fukuyama,2011). In the same way that manners in human societies make life livable, workable, and usually prevent descent into anarchy, the female hierarchy creates a livable, even, civil social life for baboons. Reproductive success would not be possible otherwise; however, this is not the same as saying that a female's rank in the hierarchy is somehow directly connected to her reproductive success.

The importance of structure is intuitively obvious when complexity and process are taken seriously. There are analogous arguments about structure from other fields. For example, the value of structure in biological systems is well summarized by Weiss and Buchanan (2009). For humans, a variety of theories argue for the benefit of structure to reduce uncertainty, minimize cognitive dissonance, build social relationships, and facilitate social exchange (for cybernetics, see Bateson,1967; Watzlawick and Bevin,1967; Harries-Jones,1995). In fact, Festinger (1985/1957; Harmon-Jones and Harmon-Jones,2007) suggested that humans seek consistency and try to reduce dissonance in any new situation. Recent approaches to human cognition including distributed cognition (Hutchins,1995; Forster and Rodriguez,2006; Barnard,2010) and situated action (Suchman,1987; Rogoff,1990, 2003; Lave and Wegner,1991) rely on a nontraditional idea of “structure” that is similar to my argument. There may be indirect or downstream impacts of structure on female baboon reproduction, but I suggest that the value of structure is not directly reproductive success.

Managing complexity: Extinction, tolerance, slippage, and the evolution of the good enough (2007–2011)

An underlying assumption of evolutionary arguments is that there are winners and losers (Weiss and Buchanan,2009; Watts,2010). Yet, in the vast span of life on earth, more than 99% of all species have gone extinct (Raup,1986; Wilson,2002; Barnosky et al.,2011). As evolutionary history is continuous and the end point in evolutionary time is most probably extinction, calculating costs and benefits has to be sensitive to that history. Species who flourished at one point later disappeared. This is relevant to my argument in two ways. First, a particular point in time may or may not be representative of what has happened or what will happen. I recognized this about the baboons I'm watching and it will apply equally to other species. The power of evolutionary arguments will therefore depend on capturing as much of the (natural) history as possible. Second, the adaptationist paradigm (Gould and Lewontin,1979; Laland and Brown,2011/2002) often assumes that animals in nature are finely, even perfectly tuned to their environment and, at least until recently, that what we observe are the elements of this excellent fit. Yet, focusing on real lives in real time illustrates how difficult it is to know about the consequences of particular actions or to decide whether, for example, a baboon individual or group (since individuals don't exist outside of groups) is making a mistake.

I had the opportunity to consider this issue during a second fusion event. This time the impetus appeared to be food, not predators. The little group, Musul, that hadn't fused 7 years earlier, initiated a fusion for a different reason, to gain access to an important new food resource or so it seemed. This resource was dominated at the time by other, larger, baboon groups (who were not part of my study).

The resource they sought was the fruit of Opuntia stricta, an exotic invader which had only just appeared in Musul's home range. Ironically, before this fusion Musul had been my “control” as a test of the costs and benefits of “not fusing.” Musul targeted Nabo (itself the result of the previous fusion of Malaika and Soit Oitashe discussed previously). By this time, Nabo had operated as a well-integrated group for more than 7 years. Unlike the earlier fusion, Musul didn't shadow Nabo. Instead, Musul tried to lead Nabo to the area with the best cactus.

Initially, most of Nabo baboons resisted probably because the area was outside of Nabo's traditional home range and already hotly contested (baboons can hear intertroop interactions beyond where they normally range). As the Musul–Nabo fusion progressed, Musul females exerted more and more influence on Nabo through their growing network of friendships with Nabo males. Occasionally, the combined group made it to the cactus area only to be aggressively expelled by one resident baboon group in particular. Aggressive mobbing is usually a very effective tactic to displace another group from a location. In this case, while the intruders fled from the aggression, they kept coming back. Musul–Nabo endured repeated evictions over more than a year but also developed evasive tactics. For example, the invaders would leave the sleeping site at first light to avoid being mobbed. This early departure meant giving up resting and socializing time and even some feeding time clearly a disruption of their normal daily routine. However, leaving early didn't always work. Often the resident troop would track down the invaders and push them until they were away from the disputed area.

The Musul–Nabo fusion (the group was now called Namu) violated many rules of baboon life that I had documented during the preceding decades. I considered whether it might represent a baboon “mistake”. The quantitative data suggested that it was. No baboon was actually killed because of the confrontations between groups, but during the first 18 months of the range shift, most of Nabo's and Musul's babies did not survive the first year of life. Adult mortality also increased. Autopsies of adult females showed that previously nonfatal conditions had turned lethal. Nutritional analysis of the cactus fruits indicated that its value, and thus its appeal, might be the energy bonus provided by such a large package of calories [see also crop raiding (Strum,2010)]. Yet, the troop must have used up that bonus and more evading attack. Unlike during crop raiding, body condition didn't improve.

Do baboons make mistakes? Was I observing the process of extinction or at least extirpation, one death at a time? My answer is different now, 4 years later, than it was at the start of the fusion. Currently the troops have been fused (becoming Namu) for several years. Namu, as invaders, eventually created a home range in the cactus area despite all the aggression. Although Namu is still subordinate to the main resident troop, they aren't mobbed as often. Namu female reproductive parameters are steadily improving along with female physical condition. Age at menarche is earlier, and interbirth intervals are getting shorter. Namu's overall mortality has declined while survivorship is improving. I had to conclude, against my previous assessment, that Namu's persistence paid off.

The second fusion provides more evidence about the pitfalls of making evolutionary interpretations of behavior because these depend on history and time frame. Flexibility and adaptability are more vital than I had imagined. It is hard to do the evolutionary calculus without knowing the specific history even when popular evolutionary criteria like condition, reproduction, and survival are available as was evident in the Namu range shift. Too often, the link between the behavior and its evolutionary consequences passes through a black box where much of the work and most of the assumptions reside [see similar argument for the interpretation of the link between social complexity and the evolution of cognition (Strum et al.,1997)].

The Musul–Nabo fusion also clarifies that evolutionary processes may be more tolerant, have larger acceptability spaces (Weiss and Buchanan,2009), and allow for more variance than most evolutionary time arguments assume. What these baboons did for more than 4 years was not a perfect fit to the layers of complexity in their life but it seems to have been “good enough.” Are they on the road to success or to extirpation? It will take time to answer that question and require measures of success situated in their natural history.

Since complexity generates a variety of options, it is not surprising that individuals in a baboon troop disagree about what to do. These disagreements must be resolved because the group is constrained to move and act as a unit. Resolution requires negotiation. Therefore, managing the consequences of socioecological complexity is a serious daily challenge for baboons. This might be most apparent during the special circumstances I was able to document like the incursion of agriculture, periods of heavy predation, or the appearance of a new food. However, disagreements, “behavioral” discussions, happen during negotiations about the more ordinary parts of baboon life such as during troop movements, consorts, when a new infant is born, and the like.

Coming to grips with complexity

These are more than just interesting stories. They are evidence about how evolution actually works. The focus on process and history reveals the importance of complexity, chance, and contingency in the real lives of baboons. It also highlights the flexibility and adaptability (in response to complexity) that pervade baboon daily life.

Other disciplines like paleontology, ecology, genetics, and cellular biology have recently made similar arguments. The broad integrative approach of Weiss and Buchanan (2009; Weiss et al.,2011) documents the importance of chance, contingency, and history (and emergence) in the “history of life.” Their EcoDevoEvo perspective advocates going from evolutionary to developmental and ecological time scales [see Fig. 2.2 in Weiss and Buchanan (2009, p. 12) for an instructive diagram], and illustrates why chance is “always built into conception, inheritance, birth, death, and much that happens in between” (Weiss and Buchanan,2009, p. 197). Context and history, for Weiss and Buchanan, are essential to understanding how, for example, a gene network, cell, organism, or ecosystem gets from point a to point b. Viewing evolution at these scales suggests that there are multiple routes to success with more tolerance and slippage and therefore more flexibility than modern Darwinian interpretations often allow.

A recent example in paleontology is the work of Venditti et al. (2011) on body size in mammals. They contend that in the evolution of mammalian body sizes, diversification did not speed up and then slow down when ecological niches were invaded as previously assumed. Instead, body size diversification took multiple routes because ecological niches were “a constantly moving target.” They proposed “that speciation might itself be the outcome of unusual single events” (emphasized by the author) and that to understand so-called adaptive radiations, it is necessary to study the multiple events in a species' life that provide it with the opportunity to adapt, rather than studying wide and general processes” (Venditti et al.,2011, p. 393).

Lorenzen et al. (2011) offered another example. North American Pleistocene megafuana extinctions appear much more complex than previously assumed. Their analysis demonstrates that changes in the abundance of megafauna were “idiosyncratic” because “each species (and even continental populations within species) appeared to respond differently to the effects of climate change, habitat redistribution, and human encroachment” (Lorenzen et al.,2011, p. 364). Surprisingly, the species that went extinct did not have characteristics that set them apart from those that survived. This implies a complex and contingent process (with behavioral flexibility in some species) rather than a simple and deterministic one.

A recent study on the evolution of sociality in primates provides another parallel. Shultz et al. (2011) modeled the evolution of primate social grouping patterns using genetic and social information about 217 primate species. The conclusion contradicts previous speculations about the evolution of primate social groupings, which assumed that the original groups were small and simple and that larger more complex structures developed from these groupings. The analysis indicates that the initial change in social grouping was probably caused by a shift from nocturnal to diurnal activity that generated large multi-male and multi-female groups. Smaller social structures are derived later in a stepwise fashion that has some phylogenetic inertia. In this view, the evolution of primate sociality was neither simple nor gradual and likely involved multifaceted interactions of group size, social complexity, and cooperative sociality.

The argument for, and the value of, focusing on process and description is best illustrated by modern macroecology (Smith et al.,2011). Macroecologists look at complex interactions embedded in space and time by examining multiple factors and levels including morphology, physiology, behavior, ecology, phenology, and phylogeny. Taking their cue from past natural historians and biogeographers, macroecology uses natural history description as a critical scientific tool. Evolutionary change, they argue, is best understood in terms of complexity, history, chance, and contingency (and emergence) whether looking at the past or trying to predict the future.

Even studies of modern human behavior that usually begin with the assumption of complexity and history are moving further in that direction. For example, a recent study of the effectiveness of co-management of global fisheries identified 19 factors (summed into eight binary measures and their interactions) that were vital to success (Gutierrez et al.,2011). However, history and accident (not planned design) determined whether a fishery had any of the set of these critical features. Similarly, Black et al. (2011) explained the inadequacy of previous models of human migration that rely only on simple principles of causation. Instead, they argued that there are numerous drivers of migration that include social, economic, demographic, and political factors. Furthermore, each driver interacts with the others as well as each being influenced by environmental change. The result is that they can make better predictions about who migrates, when, and why than previous attempts.

Another recent example for humans is Fukuyama's work on the origins of human political order (Fukuyama,2011). He convincingly demonstrated how similar starting conditions result in radically different outcomes and that “The factors driving the development of any given political institution are multiple, complex, and often dependent on accidental or contingent events. Any causal factors one adduces for a given development are themselves caused by prior conditions that extend backward in time in an endless regression” (Fukuyama,2011, p. 23).

Of course, Complexity Theory as it has developed over the last three decades draws some of the same conclusions (Waldrop,1992; Thelen and Smith,1994; Capra,1997; Oyama et al.,2001; Ward,2002; Johnson,2011/2007). Complex systems models, including complex adaptive systems and dynamical systems, have recently been applied to primates using social network analysis and analysis of collective behavior (King and Sueur,2011a; Sueur and Deneubourg,2011; Sueur et al.,2011, a). Some of this work formally addresses the issues of nonlinearity and of history, chance, and contingency in producing specific outcomes. Most work on complex systems still falls short of addressing the context of behavior in the real world. For primate social networks, for example, this should include multiple interacting contexts that set up the starting conditions and continue to influence the operation of the system.

I began thinking about complexity in the early 1970s when I realized that male success rested on more than the male dominance hierarchy. Age, residency, context, and social relationships based on collaboration and behavioral exchange all played a role. This view of social complexity was my first step in expanding out from the powerful simplifying assumptions of the previous baboon models. Crop raiding (1976–1984) demonstrated the significance of ecology and how the interaction of social and ecological added to the complexity of baboon lives. The disagreement about whether to raid highlighted the way that social negotiations influenced choices about the basics of daily life. The translocation (1984) was both a provocative experiment and an unprecedented opportunity to see baboon adaptability in action. The value of the social group was never as obvious as after translocation; however, I also learned that baboons had multiple ways to succeed and that specific events and accidents of history set different groups on unique paths.

The stone of baboon “complexity” that I had slowly pushed up the hill in the first decades picked up incredible speed rolling down the other side in the last decade. The first fusion (1999–2001) showed the full extent of social negotiation about the nature of the social group, something scientists assume to be a given fact. The 2 years of negotiation embedded in specific social and ecological events made the fusion understandable but not predictable based on simple evolutionary rules. The second fusion (2005–2008) certainly looked like a monumental evolutionary mistake but turned out not to be. That fusion demonstrated the constantly changing balance of costs and benefits and the large list of factors that had to be considered. It also implied that evolution was more tolerant and allowed more slippage making simple evolutionary calculations problematic. Finally, the changes in the female dominance hierarchy (2008–2011) bolstered my argument about the challenge of managing complexity and offered an alternative meaning for the female dominance hierarchy.

As with all complex systems, the whole is greater than the sum of its parts (Johnson,2011/2007; Mitchell,2009) even though we still lack the methodology to capture behavioral emergence. However, baboon natural histories help. They are a resource for understanding how complexity is generated. They show how solutions to challenges develop. They illustrate the enormity of the task that baboons face in managing complexity. This type of evidence also underscores the role that context, chance, and contingency play in producing specific outcomes on the evolutionary stage. The baboon data also raise the possibility that there are multiple ways for individuals to succeed, not just one optimal evolutionary path.


Washburn and DeVore (1961, 1963) used baboons specifically to frame questions about human evolution. Ironically, the intuitive power of DeVore and Washburn's model did a disservice to their goals as the simplified version was applied to other primates and even to humans (Strum et al.,1999; Strum and Fedigan,2000b). Nearly four decades later, I once again suggest that baboons may be relevant in highlighting the challenges and solutions of early human evolution.

My first decade of baboon observations led me to declare them “almost human” (Strum,2001/1987b). Indeed, they exhibited remarkable social skills, social strategies, and even sexual politics (for a current definition of and reservations about using the word politics, see Watts,2010), we normally reserve for humans. However, the following decades of baboon quantitative data and detailed natural histories suggested to me why, despite having much in common in the process of building society, baboons and humans have diverged so much. This renders the baboon vantage point particularly useful both for understanding what makes us human and for refining concepts of how evolutionary processes actually work at the level of behavior.

Summary of baboon principles of the social

  • 1The social group is a major organ of adaptation. The group offers social resources useful in meeting new challenges (raiding, translocation, etc.) and in solving daily problems (translocation, Opuntia, etc.). Baboons illustrate how this works not just why it should work this way.
  • 2Social negotiation is critical to baboon life but it quickly reaches the limits of available time and comprehension. Baboons are also constrained by their inability to “hold” nonmaterial artifacts (relationships, structure, etc.) over time and space.
  • 3Baboon society is restricted in scale because it has few means to simplify social negotiations. Therefore, baboons cannot escape social complexity to build a complicated social structure.
  • 4Context and history matter to baboons. This includes social actions at the individual and group levels.
  • 5Collaboration (see below), not just competition, is a building block of baboon society.


Cognition in the world and in the head

Baboons illustrate why the social and the ecological cannot be separated in real life. The greatest cognitive challenge for baboons is probably how to navigate both simultaneously [see navigational intelligence (Strum et al.,1997)]. Because the two domains cannot be separated, baboons may reach their cognitive limits faster than predicted by either social or ecological hypotheses about the evolution of primate cognition (Byrne and Whiten,1988). By contrast, the human ability to simplify social negotiations and build a complicated larger scale society may reduce the cognitive demands of moment-to-moment life and effectively segregate the social from the ecological for humans compared to baboons.

Yet, the social is also a great baboon cognitive resource. Several recent frameworks for human cognition also seem relevant to baboons (Strum et al.,1997; Forster,2002). Situated action and distributed cognition extend the unit of cognition and cognitive analysis beyond the individual. They suggested that knowledge and cognitive processes can reside in the world (situated in time and space and distributed between social actors and their environment) rather than just inside the head of an individual. Hutchins (1995) argued that distributed cognition makes it possible for social actors to solve problems (and think thoughts) that an individual could never do alone. If Vygotsky (1978) is right, that there is a “mind in society,” then the social realm of baboons becomes a resource for problem solving. Individual baboons, like individual humans, can later appropriate socially created solutions. This offers a mechanism for how plans and schemas might get into the mind of an individual.

Although the social resources available to baboons may ameliorate some of the cognitive challenges of an integrated socioecology, baboons eventually reach the limits of their skills and abilities to manage complexity. Going beyond this requires additions to brain/mind and a social world that offers more resources for solving problems.

Competition versus collaboration versus cooperation

Recently, Weiss and Buchanan (2009) and Sussman and Garber (2011) have illustrated a growing shift in evolutionary arguments from an emphasis on competition to the importance of cooperation in daily life. Even in baboons, a species where males have acquired the evolutionary anatomy for aggression, competition is just one factor and often not the most critical. The iconic picture of baboon society should not be two males fighting; rather it is a big male baboon grooming his totally relaxed infant friend (Fig. 2). As Sussman and Garber suggested, competition is embedded in a larger social matrix (Sussman and Garber,2011).

Figure 2.

The iconic male baboon: friendship and trust rather than aggressive might. A: Male grooming his yearling friend (credit: Dr. Timothy Ransom©). B: Male resting with yearling friend (credit: Dr. Timothy Ransom©). C: Male with his infant and juvenile friends (credit: Dr. Shirley C. Strum ©). D: Male resting with one of his female friends and her juvenile daughter (credit Dr. Shirley C. Strum©). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Yet, cooperation is a controversial term when applied to nonhuman primates. The Oxford English Dictionary (Dictionary,1971) defines cooperation as “to work together,” a meaning that originated in 1616 (p. 963) which by 1830 also meant the practice of economic cooperation. For humans, it is used interchangeably with the term “collaboration” which first appeared in 1871. There is no scientific consensus about the definition of the term cooperation as it applies to animals and humans (Richerson and Boyd,2005; Kappeler and van Schaik,2006; van Schaik et al.,2006; Tomasello,2009; Weiss and Buchanan,2009; West et al.,2010). Definitions have ranged from complex and restrictive economic criteria with analysis based on game theory (e.g., see Noe et al.,2001), to more real-life criteria like “joint commitment to a goal” with “mutual support” and the ability to “reverse roles” (Tomasello and Moll,2010), and to the broadest possible meaning of cooperation: “components working together successfully by some criterion, including that of evolutionary viability” (Weiss and Buchanan,2009, p. 38). One cross-cutting claim, however, is that human cooperation is distinctly different from what other animals or even other primates do.

For the purpose of situating baboons, I find it useful to think in terms of a continuum that starts with the “geometry of the selfish herd.” Here, individuals aggregate and cohere by simple fact of trying to get away from something, for instance, a predator (Hamilton,1971). I suggest that what comes next is called “coordination,” which develops as the enhanced value of group living requires solutions to behaving as a collective in everyday life [see also “the collective action problem” (Nunn,2000; Noe et al.,2001; Sussman and Garber,2011)]. “Collaboration” is still further in the continuum from the selfish herd. Although cooperation and collaboration are used interchangeably in both the common and scientific lexicons, I make a distinction based on insights from baboons. In 1973, one baboon male started to “hunt” rather than opportunistically “collect” gazelle prey. His actions were purposeful and his intent was evident in the sequence of his behaviors. Then other males joined, initially alerted by his return to the troop covered in dry blood. Through a set of events, the males actually learned that collaborating in the direction of the chase paid dividends. Later, they practiced what they learned (Strum,1975b). I use the term “collaboration” in its original meaning, to work in conjunction with others, and not as a synonym for cooperation. Collaborating baboons do not have a predetermined common goal. Rather they are a collection of individuals with similar goals; in the case of predation, each wanted meat and realized the efficacy of chasing prey one way rather than another. Finally, in this scheme, “cooperation” is a special type of collective action in which participants have a preexisting common purpose (see also references above). Chimpanzee predatory hunts (Boesch and Boesch-Achermann,2000) appear a rudimentary form of primate cooperation, particularly by comparison with collaborative baboon hunting. The continuum ends with human cooperation, which entails a variety of additional aspects not found in other species (but see Wilson,2012), although the specifics depend on whose definition is used.

Although competition is not as central to the baboons, I have studied, as previously assumed, neither can baboons cooperate. That they are expert social collaborators is visible in their social strategies of competition and defense, in troop movements, intertroop encounters, home-range shifts, changes in diet, as well as in intricate multi-individual social and sexual encounters. Data from other long-term baboon research sites also provide evidence of social collaboration even if the investigators do not label the behaviors that way (e.g., Palombit et al.,1997; Silk et al.,2006, 2010a; Swedell and Leigh,2006). This means that baboons regularly engage in collective action that goes beyond the coordination of individuals. However, I propose that baboons have constraints. Without additional resources, baboons can't move from collaboration to cooperation just as they are limited in the type of society they can build.

On the other hand, would baboons benefit from being able to cooperate? The origins of cooperation in humans is often attributed to a change in lifestyle, particularly the development of gathering and hunting and its concomitant division of labor and necessity for sharing (Lee and DeVore,1968; Bird,1999; Wilson,2012). Without anything in their diet that could be shared, except the infrequent prey carcass (Strum,1975b), baboons may not have the necessary or sufficient conditions for the evolution of cooperation. In the end, although baboons are great collaborators, they have neither the means nor the reasons to cooperate.

Culture as an organizing principle

The performative view of society, augmented by the frameworks of situated action and distributed cognition, offers a novel perspective on tradition and culture among baboons. Baboons don't use or make material tools in the wild but they do skillfully craft friendships. These social tools are “nonmaterial artifacts” (Strum and Forster,2001) because making them requires the same cognitive achievement as making material artifacts: their shape and use isn't intuitively obvious on the surface of the material. However, here the similarity stops. Social tools, unlike material tools, have difficulty expanding across space and time. They are “soft” tools and difficult to stabilize. Still, since tool use is often a criterion for culture (see Fragaszy and Perry,2003), baboons do have some of the basic skills (Strum and Latour,1987).

Another criterion used to identify a primate cultural “tradition” is that the behavior should be invariant within a group but variant across groups (Tomasello,1999; Rendell and Whitehead,2001; Fragaszy and Perry,2003; Richerson and Boyd,2005). A number of baboon behaviors or behavioral systems may qualify. For example, a hunting tradition developed in only one group (Strum,1975b, 1976). The raiding lifestyle, the selection of sleeping sites after translocation, differences in diets between adjacent groups in the same location, and even variations in aspects of the friendship configuration might be more controversial suggestions as baboon traditions. I could go further and claim that every group's female dominance hierarchy is a traditional system. After all, each is unique to a group, carries forward across generations, and is relatively stable over time.

What sets baboons apart from humans, if we grant that these are social traditions, is the next step in the cultural process: the persistence and accumulation of innovations (Tomasello,1999; Richerson and Boyd,2005). I have argued that for baboons, the means of acquisition and transmission of a tradition depends on frequent exposure and visible actions. Thus, the hunting tradition disappeared when the innovator transferred out of the troop a year later. Why? Although the rate of baboon predatory behavior in Pumphouse was high when compared with other baboon and chimpanzee groups observed at that time, predation was still a rare event in daily life. Furthermore, hunting happened away from the troop so that most of the animals didn't see or participate in the actual behavior. As a result, hunting just didn't “stick.” Now contrast raiding. The opportunities to raid were frequent and after the fission, the entire raider troop, not just a few animals, raided. This frequency and visibility, I propose, were important contributors to the persistence of raiding behavior. Raiding even accumulated some modifications as raiders learned to avoid the new anthropogenic risks and human-induced mortality declined over time (Strum,2010).

Conversely, if culture is based, in part, on internalized rules for behavior (D'Andrade,1981, 1990; Byrne and Russon,1998), the same factors that may be important facilitators of traditions for baboons (visibility and involvement) could mitigate against the need for traditions. Baboons have constant access to each other as social information resources (Strum,1983a, 2010; King,1994; King and Cowlishaw,2007; King and Sueur,2011b; King et al., in press). They can rely on distributed cognition and situated action rather than on individually internalized operating principles (Forster and Strum,1994; Strum et al.,1997; Forster,2002; Forster and Rodriguez,2006). Recently, Roediger and McDermott (2011) made a similar suggestion for humans. They argued that “If one individual in a group forgets critical information (about food resources or dangers, for instance), then it is wise to get an updated memory from another group member” (Roediger and McDermott,2011, p. 48). This perspective could help explain why the two most “cultural” primates so far are chimpanzees (Whiten et al.,1999, 2007) and orangutans (van Schaik et al.,2003; Jaeggi et al.,2010). If a baboon forgets how to do something, it can just look around. Chimpanzees and orangutans don't have that luxury. Chimpanzee fission–fusion society (Chapman et al.,1995; Lehmann and Boesch,2004) and the nearly solitary existence of orangutans (van Schaik,1999) deprive them of each other as social resources. This should increase the value of traditions and possibly of culture. By analogy, if early humans lived in fission–fusion societies (Grove et al., in press), they would face challenges of “remembering” more like chimpanzees than like baboons.

In summary, I am proposing that baboons have a society at the limits of social complexity and that they lack the resources needed either to manage more complexity or to simplify negotiations and build a complicated large-scale society. Baboons have sophisticated cognition that tracks social, ecological, and socioecological interactions. It is likely that they appropriate solutions to new problems from the social context. If so, baboons may need less “in the head” because they have so much “in the world.” The performative process already creates some social traditions (culture?); however, baboons have difficulty with the persistence and accumulation of innovations because, I suggest, for baboons this depends on frequent exposure and actions that are visible to everyone. Although baboons are “almost human” in their negotiation and politics, baboons can't become human because they lack the ability to build a complicated society from a complex one. To do this, they would need the behaviors that are usually classified as hallmarks of humanness, minimally the new resources that material culture, symbols, language-related abilities provide, and a new subsistence base that both permits and requires division of labor and sharing.

Finally, I conclude that the four decades of baboon research and the framework outlined above is helpful in resetting the starting point for the human experiment. It is improbable that the earliest humans were less sophisticated or embedded in less complexity than these baboons. Moreover, “lowly” baboons also serve as a reminder that the group is a primary organ of adaptation for most primates, that collaboration/cooperation is likely as crucial as competition in daily life, and that social intelligence, social skills, and social management were at a premium long before humans or even the great apes.

Summary of baboon principles of evolution

  • 1Baboons demonstrate the value of examining process rather than just outcome and the pitfalls of uncritically collapsing behavior into evolutionary time. This is a critical methodological move. How evolutionary principles work in real time and real lives can only be identified this way.
  • 2History, chance, and contingency are crucial for understanding baboons embedded in social, ecological, and socioecological interactions. The extensive baboon negotiations imply that there are multiple options not just one “best” evolutionary solution. In fact, the baboons add support to the assertion that “behavioral adaptability” is a ubiquitous trait of life but that it is often missed because of the existing scientific emphasis on evolutionary time and on evolutionary outcomes (Weiss and Buchanan,2009, p. 231).
  • 3Thinking in terms of contingency and (natural) history helps to make sense of complexity. It is also easier to understand complexity by looking at its unfolding, how it is generated, than by puzzling about existing complex structures (Weiss and Buchanan,2009, p. 176).

Given these three principles, the baboon data that I have just presented don't fit the standard view of Darwinian natural selection (Laland and Brown,2011/2002). Instead, they correspond to Weiss and Buchanan's alternative evolutionary framework. Weiss and Buchanan argued that chance (action without direction that has major impacts; p. 36) is one of life's important regularities (p. 40), that “life is a contingent phenomenon” (p. 35), and that adaptability in the face of “changing circumstances” means that responses are flexible and not always prescribed in advance (p. 37). Predictability is the basis of life processes (p. 21), and history and context matter because these are the contingent conditions “upon which the future is built” (p. 40). Weiss and Buchanan demonstrates how this applies at the cellular, genetic, developmental, and ecological levels (Weiss and Buchanan,2009; Weiss et al.,2011). I suggest that it also applies to baboon behavior.


Anthropologists should be interested in baboons because of what they tell us about the reality of daily lives, about the challenge of complexity, and about how evolutionary principles get embedded in contexts determined by history, chance, and contingency. Baboons add evidence to the suggestion that evolutionary processes contain more tolerance, multiple solutions, larger acceptability spaces, and the possibility that an adaptive fit will be “good enough” rather seamless. When the straight line of evolutionary scale opens up to capture what actually happens, arguments become less reductionist and less deterministic [the pentagon (Weiss and Buchanan,2009, p. 9)].

Yet studying behavioral complexity in the wild poses methodological challenges. How do we document complexity and track the impact of chance, contingency, and context? I see no other way than by returning to comparative natural history as a scientific method (Dayton,2003; Smith et al.,2011). Although the plural of anecdote may not be data [but see recent shift in medicine (Aronson,2003)], the plural of comparative natural history certainly is. I call baboons, “Darwin's monkey,” not just because he made frequent reference to them (Darwin,1871/2004) but because baboon lives once again argue for the value of comparative natural history, the key method that Darwin used. Following Darwin's lead, natural history observations should include well-documented details tracked across time, space, groups, landscapes, and species. They should be aided, when possible, by quantitative data and experiment. Long-term studies become extremely vital because only they can generate natural histories and the kind of insights presented here.

Baboons are also useful to anthropologists for reasons that Darwin could not have guessed. Today, it is common practice to end an article about primates with a comment about conservation. The baboons offer a singular perspective on primate conservation because they illustrate how the evolutionary game has changed in the Anthropocene (Steffen et al.,2007; Zalasiewicz et al.,2008), the current era of human-dominated ecosystems and landscapes. The heart of the current biodiversity crisis is the presence of a super-dominant species (Western,2001), humans. People either directly or indirectly create the current threats (Wilson,2002; Barnosky et al.,2011). The baboon data suggest that the speed of human-induced changes and the diversity of their impacts are likely to create more complexity and introduce an even larger element of chance and contingency into evolutionary processes (as illustrated by some of the baboon natural histories). Baboons are not typical primates. Their flexibility and adaptability and their range of ecological and social strategies seem preadapted to new human-modified landscapes which have an acceleration of pace and fragmentation of space. This may create selective pressures for “weedy” species (Richard et al.,1989); however, the baboon evidence suggests that it also puts a premium on smart species [a benefit that may extend to other taxa: see trends in urban passerine birds (Maklakov et al.,2011)]. Human activity fragments baboon populations when it fragments space. These diversified but isolated conditions would select for new behavioral responses in baboons, perhaps even at the level of new traditions. Yet the baboon scenario is not all rosy. The history of the Cape baboons in South Africa (Hoffman and O'Riain,2010, 2012; Kaplan et al.,2011) illustrates the likely future trajectory for many baboons. The current reality for Cape baboons is both genetic isolation and unsustainable human–wildlife conflict. If the future of baboons is uncertain despite their special abilities, the situation may be even direr for the majority of species who are not as adaptable as baboons. Therefore, those interested in primate conservation will need to pay special attention to the accelerating human–wildlife conflict over old and new resources in the Anthropocene.

This article wouldn't be complete without a mention of what Darwin's thought about baboons. His summary is particularly apt: “For my own part I would as soon be descended from that heroic little monkey, who braved his dreaded enemy in order to save the life of his keeper, or from that old baboon, who descending from the mountains, carried away in triumph his young comrade from a crowd of astonished dogs—as from a savage… and I have given the evidence to the best of my ability” (Darwin,1871/2004, p. 791). I don't think Darwin would be surprised by the picture of baboons that I have just presented.


I thank the editor for his generous invitation to comment on the important lessons from my years of baboon research. My insights also rest on the work done by many other individuals associated first with the Gilgil Baboon Project and later with the Uaso Ngiro Baboon Project. Since 1981, Kenyan paraethologists have been the mainstay of the basic data collection. I thank them all for their contributions. Finally, I thank David Western for his distributed cognition; Katerina Semendeferi for useful comments on the article; and Bruno Latour who first helped me to see why complexity is so important although it took many decades to collect the evidence.