The Primate Origins of Human Nature

By Carel P. Van Schaik

The Primate Origins of Human Nature (Volume 3 in The Foundations of Human Biology series) blends several elements from evolutionary biology as applied to primate behavioral ecology and primate psychology, classical physical anthropology and evolutionary psychology of humans.  However, unlike similar books, it strives to define the human species relative to our living and extinct relatives, and thus highlights uniquely derived human features. The book features a truly multi-disciplinary, multi-theory, and comparative species approach to subjects not usually presented in textbooks focused on humans, such as the evolution of culture, life history, parenting, and social organization.

• Language: English
• Publisher: Wiley
• Release date: Jan 22, 2016
• ISBN: 9781119118183

Book preview

Preface

It has often and confidently been asserted that man's origins can never be known; but ignorance more frequently begets confidence than does knowledge; it is those who know very little, and not those who know much, who so positively assert that this or that problem will never be solved by science

Charles Darwin – The Descent of Man (1871)

Evolutionary Anthropology and Human Nature

Immanuel Kant, in his Lectures on Logic, famously argued that philosophy can be summarized as asking four major questions: (i) What can I know? (ii) What ought I to do? (iii) What may I hope for? and (iv) What is humanity? He then added that the first three, referring to Metaphysics, Ethics, and Religion, are in a way part of the all-important, and perhaps oldest, fourth question, which was the realm of anthropology. After Kant, philosophy and science went their separate ways and the central role of anthropology never materialized.

The rise of Darwinism should have revived Kant's insistence on a central role for anthropology. That did not happen. Not that numerous scientists did not ruminate on human nature. Indeed, during the past few decades, an increasing number of disciplines have discovered human nature as a topic worthy of study, perhaps most prominently Evolutionary Psychology. But Anthropology, Cultural or Physical, has been increasingly marginalized.

This is a real pity. Much recent work on the origin of human nature ignores two major contributions of anthropologists. First, it ignores the pervasive influence of culture on everything humans do, and even on how we think, and traditional cultural anthropology, now virtually extinct as an academic discipline, has much to offer. Second, and even more importantly for this book, the predominant approaches take place in an evolutionary and phylogenetic vacuum. Evolutionary anthropology, the broadly oriented descendant of physical anthropology, places human nature in an evolutionary, explicitly phylogenetic and comparative perspective. It can therefore help us unravel what is shared with other primates, especially the great apes, and thus requires a broad explanation, from what is new in human behavior and psychology and thus requires special explanation. It can also explain why some novelties evolved, by drawing on broad biological regularities that are backed up by comparative analyses. This book is built on these two cornerstones of evolutionary anthropology: finding shared features (or homologies) and finding general explanations for our novel features.

Why We Cannot Ignore Evolution

Biologically speaking, we are apes that are part of a mammalian radiation, primates. Mammals are a fairly recent group that originated from a minor group of reptiles. One of the main reasons we are here is that the dominant lineage, dinosaurs, became extinct in the last spectacular mass extinction event, some 65 Mya (millions of years ago), which ended the Cretaceous and started the Tertiary periods.

So we are primates. But how are we related to the other primates? About 100 years ago, the predominant view was as follows: we had split off from the rest of the primates some 30 Mya, and had therefore undergone a very long separate evolution. Not a ringing endorsement to run off to the jungle and study great apes, let alone other primates! In fact, this view justified the neglect of primate biology and its relevance to understanding human origins. It also explains why attempts to understand human nature in the past were so thoroughly non-biological.

Around 1960, a century after Darwin's revolution, this began to change. First, Louis Leakey decided, against prevailing opinion, that we should be able to learn more about the origins of our own behavior by studying great apes, and therefore helped to instigate the first field study of chimpanzees by Jane Goodall. This study, and a flood of field and zoo studies of apes and other primates since then, revealed numerous uncanny similarities between great apes and ourselves.

A second and equally influential development came from the fledgling science of molecular biology, which seemed to reconfirm that we were just another great ape – a conclusion already reached by Huxley in 1861 but subsequently largely ignored. Indeed, more refined analyses took this one step further by showing that the chimpanzee's closest relative is not the gorilla but us humans. We are the third chimpanzee! The genomes of both species have been sequenced, confirming this amazing genetic similarity. At the base-pair level, there is 98.8% similarity (chimps have fpref-math-0001 chromosomes, whereas we have fpref-math-0002 , because our chromosome 2 consists of two joined chromosomes).

The figure shows the modern classification of living (extant) species of great apes and humans based on molecular information. It shows that the hominin lineage, of which we are the only surviving member, is incredibly recent. If we can conclude anything from this, it is that we can no longer ignore primate behavior. Now anybody dealing with the origins of human nature has to take our evolutionary history seriously.

fprefuf001

Figure 1 Humans are African great apes. The actual divergence times are subject to debate, and may be somewhat older than suggested here when extant great ape life histories are assumed (Langergraber et al. 2012). After Robson & Wood (2008).

Wallace's Challenge

A major part of the explanation for the disconnect between evolutionary biology and the study of human nature is that most non-biologists do not really believe that evolution is relevant. Indeed, most scientists who study humans happily continue to ignore evolution. Either biology is irrelevant to human nature beyond the obvious morphological and physiological constraints, just like behaviorism used to claim, or human scientists are a bit like creationists, closing themselves off from information that might threaten their cherished beliefs. If the latter, they are in good company. Wallace (1870; quoted in Shermer 2002) was convinced that natural processes could not have produced our species. The brain of ‘savages’ was far too large for the kinds of intellectual challenges they faced in their simple societies with their simple subsistence. How was it possible that the brain, or indeed any organ, evolved to a size far larger than necessary? Had it evolved to anticipate future accomplishments not linked in any direct way to natural selection? That should be impossible under evolution. Baffled, Wallace turned mystic, arguing that for this origin [of intellectual and moral faculties] we can only find an adequate cause in the unseen universe of Spirit.

The challenge of providing a naturalistic explanation of our evolution is perhaps even greater than in Wallace's day. Not only have we learned in the meantime that we are much more closely related to the great apes than we thought, we also know by now that everything that makes us look really different from the other apes – fancy technology, large brains, hunting and gathering, the arts – is much more recent than even the split date suggested in the figure. In fact, behaviorally the hominin lineage began to diverge from the others as recently as about 2 Mya. This extremely recent behavioral divergence makes our task of relying on natural processes more difficult, but at the same time all the more pressing.

This book is written in the conviction that Wallace's challenge can be met (see Section 27.5), and that the seemingly miraculous evolution of humans can be explained invoking nothing but fully natural processes. An impressive amount of work has been done in the past few decades from which the outlines of a solution are gradually emerging. First, billions of years of shared ancestry has made primates, and especially great apes, much more similar to us than we used to believe, laying a firm biological foundation to human nature. Second, upon this substrate, a new, and increasingly dominant evolutionary force was grafted: cultural evolution, with its openness to developmental inputs, which was a fast and powerful force for change. This is not just another way of saying that culture took over and biology became irrelevant. Reality is much richer: we will encounter numerous examples of biologically-based processes interacting with cultural processes, and producing outcomes that are far more interesting than either behavioral biologists focusing exclusively on animals or cultural scientists focusing entirely on humans could have foreseen.

Organization of This Book

In this era of web-based search, where one thing leads to another in a network of connections, one may ask why one would still write textbooks with their enforced linear sequence. For a book that tries to explain the evolution of one particular species in all its variety this challenge is even greater. However, the major advantage of a textbook is that a patient reader can systematically build up the requisite background knowledge of both facts and concepts to fit the material offered in each empirical chapter into a larger framework.

This book is written for anyone interested in the evolution of human nature, with a basic background in natural sciences and a basic knowledge of biology, as provided in any introductory course in college or high-school advanced placement. But inevitably, some readers will be more familiar with certain topics than others. They can obviously skip parts. For instance, the book starts with a few general chapters that provide the basic concepts of evolutionary and behavioral biology. While readers with a background in basic evolutionary biology may not want to read them from beginning to end, Chapters 3 and 4 will also contain material that even those readers are unlikely to be familiar with.

After that, in Chapters 5 and 6, we give a thumbnail overview of nonhuman primates and human evolution, to provide the context in which the features of interest are embedded and which often affect their expression. This brief section is meant for those not familiar with primatology or the story of human evolution; those who are can skip it.

We then get to the meat of the book. Ideally, each section has a uniform structure. For each particular topic (say, sex or rearing offspring), we first introduce the general biological theory around a broad set of features, then examine their expression in the relevant nonhuman animals (primates for the shared features, selected non-primates to highlight independently derived similarities), followed by an examination of the proximate psychological mechanisms that regulate them, and finally their function and adaptive significance. Following that, we turn to humans. We must first ask whether the traits under consideration are identical or should be considered composed of different elements, each with a separate evolutionary history, mechanisms and functions. We then go through the same sequence of proximate mechanisms and adaptive significance.

For practical reasons, we cannot always follow this structure, because knowledge of the topic among animals is poor or because it is hard to delineate the topic in the first place. However, the reader should try to keep this approach, explained in detail in the introductory chapters, in mind at all times. But the major advantage of a textbook is that a patient reader can systematically build up the requisite background knowledge of both facts and concepts to fit the material offered in each empirical chapter into a larger framework. This task should be facilitated by the glossary at the end of this book.

A Plea for Tolerance

We will define human nature later, but in this book we will examine all those aspects that can reasonably be said or suspected to have biological foundations, including language, morality, religion, and even art. Many of these topics are highly contentious and scholars from different fields have radically different views. Any statement about human behavior, especially if cast in terms of generalities, tends to elicit acrimonious debate. This is not surprising, for the risk of political abuse is greater than when we study aardvarks or zebras, so we must walk a fine line. Largely because sensationalized or sloppy formulations may offend or, even worse, confound readers and fuel controversy, I kept the book's text as dry as possible. I hope I succeeded in doing this without putting off too many readers.

There is an inevitable link between scientific findings and political viewpoints, but we must have some faith that science can sort these things out and that people peddling different viewpoints are not necessarily politically motivated. None of us is entirely free of political biases, try as we may, but we can rely on others to point them out! Thus, collectively, the scientific community may find the right balance between tolerance for all scientific ideas and intolerance for inappropriate ideas that are meant to degrade or discriminate some class of individuals.

Some may consider this approach naïve, but science has a strong self-correcting tendency, because of our unshakeable respect for facts, as Charles Darwin (1871, p. 385) already knew: … false views, if supported by some evidence, do little harm, as everyone takes a salutary pleasure in proving their falseness; and when this is done, one path towards error is closed, and the road to truth is often at the same time opened. Thus, even if it may take a while for the correction to take hold, open scientific debate holds the key to progress.

I must also add a special plea for tolerance toward me. Every textbook author stands on the shoulders of giants, but many true giants have become anonymous due to the massive numbers of players and the need to summarize and simplify in a textbook. Because this book tries to cover a vast array of topics and thus integrate a huge literature, it is inevitable that I missed some key contributions, or worse, even misrepresented or misattributed viewpoints. I hope that readers will contact me to point out the more blatant cases, while forgiving me for the fact I had to keep the number of references within limits.

Series Editors' Preface

The Primate Origins of Human Nature is a timely and important book that coincides with something of a turning point in the history of anthropology. Anthropologists study both human biology and behavior; but there has for decades been no consensus among them about what the one thing has to do with the other. For the past 40 years or so, most anthropologists who describe and analyze human behavior have had little interest in trying to relate it to our biology. Many of them dismiss human biology as irrelevant or even inimical to an understanding of the way people act. This sort of dismissal has made fruitful dialogue between biological and social and cultural anthropologists nearly impossible. Worse, it has led to the expulsion of biological anthropology from departments of anthropology at many universities.

In recent years, there have been a number of signs that this odd dualism is beginning to evaporate. Today, we find anthropologists of all persuasions conceding that humans are very different from other animals in important ways, but that these human peculiarities originate in significant peculiarities of the human body, brain, and genome. As the title of this book implies, Carel van Schaik is such an anthropologist. He is committed to the common-sense proposition that the way people act has something to do with the kind of animals we are. He insists that the foundations of universally shared human behaviors must lie outside of human behavior— in the constraints and channels laid down by our primate prehistory, and in the overarching principles of evolutionary theory that explain that prehistory. In this book, he has erected a masterful synthesis on that foundation. The Primate Origins of Human Nature brings together and integrates established facts and theories from an extraordinary range of the life sciences to furnish explanations and explanation sketches of the ways in which human behaviors differ from and resemble those of other primates, with respect to our ecologies, social structures, group size, patterns of mating, breeding, cooperation, and parental investment, and our physical and cognitive growth and development. He is not afraid to seek and identify plausible nonhuman parallels and underpinnings even for such uniquely human traits as morality, religion, language, and esthetics, which have proved stubbornly resistant to biological approaches in the past. Throughout these searches, he maintains a careful skepticism, weighing and judging alternative accounts and theories with magisterial scholarship and frequent suspensions of judgment. Not every reader will agree with every one of his final assessments. But nobody who reads this book can fail to be impressed by the merit and importance of his inquiries and conclusions, or by the breadth of scholarship and depth of thought that inform and guide them.

We feel confident that The Primate Origins of Human Nature will prove to be a landmark work. As a textbook, it will provide students of primate behavior and ecology, evolutionary psychology, and biological and general anthropology with a host of provocative ideas and a comprehensive survey of a broad range of scientific literatures. As an innovative and masterful contribution to those literatures, it will direct and stimulate future inquiry. It is our hope that it will be read widely by social scientists as well as biologists, and that its influence on both students and scholars will ultimately help to bring about a reintegration of anthropology's scattered parts. We are proud and honored to welcome this superb book to our series on the Foundations of Human Biology.

Matt Cartmill

Kaye Brown

Acknowledgments

Science is a large collective enterprise and many more make crucial contributions than the few known to the public at large. I therefore particularly wish to acknowledge the important contributions of those colleagues whose work I could not cite to keep the reference lists manageable or of whose work I was not even aware, even though it fits very nicely. I hope they can forgive me for the biases in citation that are the inevitable consequence of the limited time and attention span any single author has.

A book like this is not written overnight and thus reflects the influence of many people. I owe much to the influence of my scientific heroes, Pauline Hogeweg, Jan van Hooff and John Terborgh, who each in their own way taught me how to observe, think, and write properly. I was influenced by many colleagues at Utrecht University, Duke University, and the University of Zurich, who without exception were generous with their advice or simply their enthusiasm. I thank the many colleagues, students and collaborators who influenced or inspired me: Natasha Arora, Filippo Aureli, Annie Bissonnette, Meredith Bastian, Balthasar Bickel, Diane Brockman, Redouan Bshary, Matt Cartmill, Steve Churchill, Parry Clarke, Laura Damerius, Rob Deaner, Roberto Delgado, Robin Dunbar, Lynda Dunkel, Charles Efferson, Ernst Fehr, Sofia Forss, Beth Fox, Herb Gintis, Sereina Graber, Hanjo Glock, Mike Griffiths, Thibaud Gruber, Cyril Grüter, Kirsten Hawkes, Sandra Heldstab, Sarah Hrdy, Bill Hylander, Charlie Janson, Rich Kay, Lukas Keller, Peter Klopfer, Cheryl Knott, Barbara König, Sonja Koski, Elissa Krakauer, Michael Krützen, Eric Kubli, Chris Kuzawa, Laurent Lehmann, Stephan Lehner, Becca Lewis, Angelika Linke, Marta Manser, Frank Marlowe, Ellen Meulman, Tatang Mitra Setia, Hans-Dieter Mutschler, Alex Nater, Ana Navarrete, Ronald Noë, Charlie Nunn, Ryne Palombit, Sagar Pandit, Stephanie Pandolfi, Andreas Paul, Mike Placvcan, Gauri Pradhan, Signe Preuschoft, Claudia Rudolf von Rohr, Anne Russon, Marcelo Sanchez, Peter Schaber, Matthijs Schilder, Caroline Schuppli, Dan Schmitt, Ian Singleton, Arnold Sitompul, Brigitte Spillmann, Romy Steenbeek, Jito Sugardjito, Claudio Tennie, Simon Townsend, Liesbeth Sterck, Erin Vogel, Han de Vries, Andy Whiten, Janneke van Woerden, Tony Weingrill, Gereon Wolters, Suci Utami, David Watts, Serge Wich, Erik Willems, and Klaus Zuberbühler.

A few people deserve special mention. Peter Kappeler on various occasions forced me to wade into topics that ended up enriching my take on animal behavior or hominin evolution. I am especially indebted to Judith Burkart, Karin Isler, Adrian Jaeggi, and Maria van Noordwijk for the insight that our initially separate strands of work on callitrichids, brain size evolution, and great apes turned out to complement each other miraculously. Kai Michel was a major source of inspiration and questions on all things cultural.

I thank Alie Ashbury for taming my grammar, and Piero Amodio, Judith Burkart, Adrian Jaeggi, Eric Kubli, Simon Townsend, Johanna van Schaik, and Kadrie Selmani for providing feedback on one or more chapters, and especially Maria van Noordwijk for going through most chapters. I am grateful to Kaye Brown and Matt Cartmill for giving me the opportunity to put this book together. My main gratitude goes to Perry van Duijnhoven: once again, he was there to illustrate and illuminate.

Section I

Evolution, Behavior and Culture

Chapter 1

Elements of Evolutionary Biology

1.1 Darwin's Argument

Charles Darwin's major contribution to science was that he gave us a mechanism for changes over time in the traits of organisms, that is, evolution, and so provided us with a framework for organizing and understanding the natural world. His theory of evolution through natural selection helps to provide a coherent explanation for the two most striking features of the natural world: complexity and diversity. These two features are linked through the concept of adaptation through natural selection.

Evolutionary biology explains evolution through natural selection using a five-step argument, as suggested by Endler (1986) (Table 1.1). First, the natural world is variable. This is a basic observation that one can make every day. Each and every trait of each and every organism shows some variation. This is obvious for many quantitative traits, such as the lengths of extremities or tails, as well as more qualitative traits such as the colors of eyes or hair. In many cases, this variation may be less obvious and more complex to measure, but it is still present; for instance, an organism's behavioral response to an environmental stimulus may differ from its conspecifics response to the same stimulus. This is known as behavioral variation, and it is an important example of variability in the natural world.

Table 1.1 Darwin's five-step argument for evolution through natural selection. After Endler (1986)

Second, the variation in the characters has a heritable basis. We now know this to be true for all characters of organisms, and it can be checked by experimental crosses of individuals with known different traits, while holding the environment as constant as possible or, where experiments are impossible, by studies of heritability or comparisons of monozygotic and dizygotic (identical and fraternal) twins reared together and twins reared apart. The heritable basis can be strong or weak (see Chapter 2), and the strength of the heritability merely affects the speed of the following step.

Third, it is rare that any given population of a species realizes its maximum potential reproductive output. There is, in Herbert Spencer's words, a struggle for life. This is an easy deduction to make: many more organisms are born than reach adulthood, and many that do never reproduce. If we were to supply organisms with a super-abundance of resources, individual reproduction would be high and mortality would be very low, thus population growth would continue to infinity. The fact that this does not happen must mean that there is something that restricts individual reproduction and longevity, and thus population growth.

Fourth, the nature of the characters affects success in the struggle for life. Even if much of the success is due to chance, there is usually some connection. This connection, known as the correlation between trait values and fitness, is due to the action of natural selection, inferred by this step. Historically, this step was summarized by the dictum survival of the fittest, meaning that those who do best do so because they are best equipped to deal with whatever problems are limiting fitness in that particular context. Thus, natural selection acts by favoring the variations of traits that allow for greater fitness in the particular conditions faced by the organism.

Fifth, as an inevitable result, adaptation ensues. Because natural selection favors those variations that are best suited to the current context, organisms gradually change over time – becoming increasingly more able to survive and reproduce in the reigning conditions. Traits that arose in this way through natural selection are called adaptations. They may seem immensely complex. Darwin's favorite example was the human eye, but virtually any natural structure, process, or behavior is an example of adaptation.

To sum up, there is variation in traits. This variation is at least partly heritable. Whenever the possession of different traits leads to differential fitness, selection ensues. As a result of such a selection process, organisms acquire adaptations. These adaptations change over time because environments, especially the biotic ones, change over time as well: there is evolution, the historical process of change, with its own attendant rules and regularities. The same process also creates historical contingency: the fact that the direction of future change depends on the current situation. As a result, evolution produces divergence among organisms, and so increases diversity.

There have been many studies testing one or more steps in Darwin's argument. They were generally successful (Freeman & Herron 2007). This means that the process of natural selection leading to adaptation, the core elements of evolutionary theory, has been tested and upheld over and over again.

A brief word is in order here about the term trait (or feature, or characteristic). A trait is whatever we decide to measure or characterize (for instance, size of brain, ability to build nests) and that is homogeneous enough to be subject to evolutionary change because it has a clearly delineated effect on the organism's functioning. When attempting to analyze and explain a trait's function, control, development, or phylogeny, traits can be lumped together or split into components. One can avoid much confusion by considering the optimal level of detail for the question at hand. Thus, ability to digest lactose while adult is a good trait by this definition, whereas religion is not because it is heterogeneous. However, we often do not know such details yet when we begin to examine a phenomenon, and the delineation is one of the outcomes of the investigation.

In this chapter, we will briefly examine the key concepts introduced above: natural selection, adaptation and complexity, and evolution, both at the level of individuals and populations (microevolution) and at higher levels, connecting organisms that shared common ancestors long ago (macroevolution). The next chapters examine how we can apply evolutionary biology to behavior in general (Chapter 2) and to human behavior in particular (Chapter 3).

1.2 Natural Selection and Fitness

In a world of limited resources, natural selection is inevitable. Natural selection is the process that causes, in Darwin's own words, the preservation of favorable variations and the rejection of injurious variations. It is therefore the process that causes the correlation between traits and fitness (Endler 1986).

A trait's fitness is defined as its positive or negative effect on the trait's frequency in the population in the next generation. It is usually expressed relative to the population's average, with the latter put at 1, so a fitness (often denoted as w) of <1 means the trait is a declining one, whereas w >1 means it is increasing. Estimating fitness requires estimating each individual's lifetime reproductive success, in offspring born or offspring that reach maturity, relative to the mean in a large sample of individuals. We can then extract the trait's fitness by comparing it with alternatives, allowing us to test the fitness of a trait or even of a genotype or a particular gene (for quantitative traits through regression analysis see Arnold & Wade 1984).

Natural selection is for a particular function (see below), by selecting on some particular mechanisms that produce the trait during development. It will thus favor some trait variants, however slightly, over others. In the process, over time it produces descent with modification in the population, which is evolution.

Selection must work with whatever variation already exists. The ultimate source of the variation is changes in the genetic material, collectively called mutations. These mutations are random with respect to the ultimate direction of change. (Some seeming directionality can be caused by the Baldwin effect, or genetic assimilation; see Section 2.4). Whether a particular selective response is possible depends on whether the requisite genetic variation happens to be available (constraint – discussed in more detail below).

In most organisms, the transmission of the variation on which natural selection works is genetic, but this is not strictly necessary. Other forms of heritability are also possible. Thus, cultural evolution occurs through a form of selection on cultural variants. This is examined in detail later (Section 3.4).

Selection can also work on traits that do not affect survival or reproductive success by improving the design of the organism to deal with the world around it, but instead by making individuals acquire more or better mates. This form of natural selection is called sexual selection (see Section 10.3).

Many critics have suggested that natural selection and fitness are circular concepts: the fittest survive, and fitness is ascertained by observing who survives. In part, this misunderstanding is due to the catchphrase survival of the fittest. However, fitness can be conceptualized and tested in two ways, both of which eliminate any circularity with natural selection. First, fitness can be seen as a design property: in this case, pre-determined fitness criteria are used to predict the fitness of the trait, and the trait can then be measured in observation or experiment. Second, and more commonly, fitness is deduced from the observed success: in this case, one derives hypotheses as to why a particular design is adaptive, that is, leads to above-average fitness. These hypotheses are subsequently tested in various ways, as discussed below. To summarize, evolution through natural selection can be predicted or explained after the fact (retrodicted), but in either case there is no circularity, provided one is willing to test predictions independently from the initial observation.

In most descriptions we refer to selection moving a trait toward another value or state, a process called directional selection. Most of the time, however, natural selection serves to remove unfavorable variation and keep the trait where it is: stabilizing selection. At the molecular level, where the object of selection is a discrete genetic variant, say a particular allele, selection can be for or against it, and is therefore referred to as positive or negative.

1.3 Adaptation

Definitions of Adaptation

One important outcome of the selection process is adaptation. We must therefore ask what an adaptation is and how we can test whether a trait is indeed adaptive. Adaptation can refer to both the product and the historical process that first produced it.

The definition of adaptation as a product is straightforward: it is a trait's current function or utility. Reeve & Sherman (1993) define it as a phenotypic variant that results in the highest fitness among a specified set of variants in a given environment. This operationalization allows us to test whether a trait is actually adaptive in its current genetic context and environment.

A definition that also acknowledges the historical process that produced the adaptation is given by Stearns (1992): a change in phenotype that occurs in response to a specific environmental signal and that has a clear functional relationship to that signal. This is a more demanding definition because it requires that we can show that the trait actually evolved to serve its current function, that is, that the effect was brought about by selection rather than by some other process. The conditions surrounding the origin of the trait, however, are usually buried in deep time and may therefore be difficult to identify.

The requirement to show that a trait actually evolved for its current function makes it difficult to convincingly demonstrate its status as an adaptation. Indeed, sometimes an existing trait may turn out to be useful for newly arising functions and so may be subsequently modified to optimize its role in this function. Such modified traits are sometimes called exaptations to differentiate them from traits that arose de novo. For instance, feathers are thought to have evolved in some reptilian groups for their function in thermoregulation, and only secondarily acquired the function of supporting flight in birds.

It is therefore likely that many adaptations began their evolutionary career as unselected byproducts. Their unselected novel functions are called spandrels (Gould & Lewontin 1979), after the spandrels in medieval cathedrals that may appear to be specifically designed to support elaborate paintings, but instead are simply there because arches were needed to bridge open space, producing a surface that could be filled in. Spandrels are probably not very common among animals because selection should quickly modify them to optimally serve their new function (Dennett 1995), but they are bound to be far more common in humans as a result of recent cultural evolution. For instance, our noses can support eyeglasses, but we can be sure that selection did not favor noses for this function before glasses were invented.

In practice, it is difficult to test such historical scenarios for non-morphological traits, making exaptation much harder for behavioral biologists to demonstrate. For instance, it is conceivable that group living originally arose in response to the threat of predation, but that it is now maintained in organisms no longer subject to predation (or at least not to the extent that grouping beyond a certain level is required) by advantages that could only have arisen secondarily as a result of gregarious habits, for example, communal hunting or learning skilled behaviors essential for foraging success. Indeed, we should expect that such novel functions would arise quite often.

Why We Must Test Adaptive Explanations

A trait is not necessarily adaptive, in the sense of having current utility, that is, providing a fitness advantage in the current environment. Non-adaptive traits generally arise through three possible processes: (i) The non-adaptive trait could be the byproduct of selection on another trait, but not disadvantageous enough to have been strongly selected against; (ii) the non-adaptive trait reflects a developmental mishap, or (iii) the non-adaptive trait was once adaptive but no longer is, due to changes in the environment. In the latter case, the trait is often still called an adaptation but it is not longer adaptive. When discussing animal behavior, the trait in question, say the motivation to dominate others in the same population or local group, is usually both an adaptation and adaptive. However, the same behavioral disposition in humans, while almost certainly an adaptation (because of the continuity with our relatives), no longer needs to be adaptive in modern societies (post-demographic transition: see Section 4.3).

As a result of these possibilities, we must ideally show that a trait is currently adaptive. It would appear that demonstrating a trait's current utility should not be too difficult. The most obvious method is to estimate the fitness of the individuals bearing the adaptation relative to those who do not. This can be accomplished by manipulating the trait or by looking for natural variation in the trait and linking it to fitness variation within a population. For social behavior in particular, this may look simpler than it is, because manipulations of social behavior are likely to produce unanticipated confounding consequences. For instance, one may be interested in assessing whether a particular dominance-acquisition strategy of males is adaptive. But how is one to manipulate this strategy? One may be able to manipulate the social context and make predictions, but such experimentally induced changes may alter much more than just the variable of interest.

Because of these difficulties, some have taken a less rigorous approach. One can assess the degree to which the trait betrays design: a sign of being unusual and complex. Take, for example, the acoustical extraction technique called tap foraging by aye-ayes (Daubentonia madagascariensis). Aye-aye extractive foraging (Erickson 1991) involves a sequence of tapping on a branch, listening, then – upon having acoustically located a larva – gnawing a narrow hole and finally extracting the larva using their third finger (Figure 1.1). Not surprisingly, aye-ayes have exquisite hearing ability, large mobile external ears, a clearly elongated middle finger, and large, sharp and ever-growing incisors. Thus, tap foraging depends on the presence in aye-ayes of costly morphological parts that are not seen in any of its relatives. Few people would insist on a formal test that this complex of characters is indeed an adaptation, especially when, as is the case here, the trait is universal in the species.

c1f001

Figure 1.1 An aye-aye (Daubentonia madagascariensis) engaged in tap foraging. Notice its visual and auditory focus on the hole it has gnawed with its ever-growing incisors, and the use of the elongated middle finger to fish contents from the hole.

When the argument is applied to humans, however, it is not always so clear where to draw the line, especially where alternative interpretations are possible (see Section 4.4). For instance, Bulbulia (2007) argues that religion is adaptive because it is a very costly trait (and an unusual one, not shared with any other species), which would be unlikely to exist unless it created a countervailing benefit. However, because cultural evolution has produced so many new, sometimes quite costly phenomena (think of wingsuit base-jumping, celibacy or suicide bombing), reliance on the design argument can lead to the wrong conclusions.

Adaptive Function

An even more pressing issue is whether we can identify the function of the adaptation (Andrews et al. 2002). Design arguments or even demonstrations of a fitness advantage due to the presence of a trait remain incomplete if they cannot also demonstrate the actual function of the trait in question. In other words, we must always examine what the trait is an adaptation for. In many cases, identifying this function is obvious, as with the aye-aye's tap foraging, but this is not always the case. Adaptations often arose in response to some external biotic force, and if they are successful the negative impact of this force is rarely observed – so rarely perhaps that one may never see an example of it during a normal field study.

Instructive examples of such adaptations whose functions are difficult to identify are grouping in response to the risk of predation and year-round male–female association in response to the risk of infanticide by males. If group living is effective, very few individuals, at least in the species with the slowest life-history pace like primates (see Section 18.3), may actually end up as prey to predators. Thus, observers may conclude that predation was not very important in shaping the trait. Similarly, infanticide by males in some species is so rare that one may doubt that it is responsible for the continuous presence of likely sires in female societies (see Section 18.4). However, the extreme negative fitness impact of falling victim to predators or infanticide means that the mere threats of these forces are strong enough to give rise to adaptations working against them – adaptations, therefore, whose functions are not necessarily obvious.

This problem has been christened the White Knight Problem, after the white knight in Alice in Wonderland. The white knight had a horse with spiky anklets on its feet, which, upon inquiry by Alice, turned out to be there to prevent shark bites. When Alice noticed that there were no sharks around, she was confidently told that the anklets were extremely effective. Thus, hypotheses must be developed based on careful behavioral observations or on the impact of rare events, the serendipitous breaks one occasionally gets when observing animals for very long times. Subsequently, these hypotheses must be tested.

Adaptive hypotheses must therefore always tie a trait to a specific function. Testing the function of an adaptation can be done using experiments in which the functional consequences are manipulated. We can remove the breeding male from a primate group and see whether infanticide ensues (it does! – but note that these lethal experiments were done before this effect was known or even suspected). Alternatively, or in addition, we can compare the association between the presence of the trait in different species and the presence of the putative selective agents. We are then not estimating the fitness of the bearers of the trait, but we assume that the trait evolved as an adaptation if across species the presence of the trait is associated with particular conditions that were hypothesized to have favored its evolution through natural selection. This approach is called the comparative method, which can be discussed in more detail once we have discussed phylogeny and character reconstruction.

Cause and Effect in Adaptation: The Role of Constraint

When we deal with adaptations to the physical world, such as a thick coat of fur in cold climates, it is usually quite clear which is the selective agent (the climate) and which is the adaptive response (long hair). When dealing with behavioral adaptations this is not always so easy, because there is rarely an obvious external factor. Yet, adaptive hypotheses often force us to identify one trait as a selective agent and the other, the adaptation, as the evolutionary response. For instance, one adaptive hypothesis is that primate females often face the problem of infanticide by males because they cannot conceive during their long period of post-partum amenorrhea (as a result of their slow-paced life history). In response, they live in permanent association with males that can protect them from infanticidal attacks by other males, and also show various adaptations in their reproductive biology. We will take the presence of post-partum amenorrhea (see Section 10.7) as selective agent or cause, and the behavior, association with males, as the adaptation or consequence.

Why do we assume the selective process went in this direction, when all we have is the correlation of these two traits across species? There are four methods through which we can deduce which trait is the causal variable and which the response variable.

First, as a general rule, the trait with the earlier origin constrains the newer one, rather than the other way around. These older traits tend to be more deeply embedded in the design of the organism and are therefore harder for natural selection to change. In the infanticide example, we would not expect selection to increase the pace of life history, so as to move the organisms out of the range of pace in which they become vulnerable to infanticide. Instead, we assume that life history is the trait that is more difficult for natural selection to modify, because doing so would involve changing numerous other features of the organisms, often including their body size, with too many repercussions on physiology, ecology, and so on to be feasible for natural selection. So selection takes the easier route, which is also fast and therefore brings an immediate improvement in fitness, and instead modifies the other trait (here: year-round male-female association), which, retrospectively, is considered the response variable.

Second, another fairly safe shortcut is to assume that behavior is more often responsive to morphology than the other way around. Behavior tends to have a wider norm of reaction than other traits, allowing more flexible ontogenetic responses to conditions. What is developmentally flexible can also respond more easily to selection. Thus, when we ask whether group living is a response to predation risk in mammals, we assume that animals could not have changed their morphology to reduce predation risk, for example, by producing body armor or spiny quills that make them invulnerable to predators. (Of course, there are animals that did this, and interestingly they tend not to live in permanent groups). But we tend to ask this question about animals that are already known to live in groups, and thus have picked organisms that did have the grouping option available to them, while ignoring the ones that did not, for whatever reason, opt to respond to predation risk by becoming gregarious.

Third, one can examine the degree of phylogenetic lability of traits. Some traits are omnipresent once they have evolved (e.g., lactation), others are somewhat flexible (e.g., body size), and yet others are often even more flexible (e.g., social organization). The more other traits depend on the presence of a given trait, the more likely it is to act as a constraint, simply because changing the given trait would therefore imply that many other aspects of the organism must change as well, causing a cascade of necessary changes to maintain homeostasis.

Finally, some organs or functions are apparently impossible to change once they have come into existence. Brains appear to be one such example, since they are constrained in so many ways. An especially difficult feature of brains that selection must work with is that brains cannot be temporarily starved of fuel. Natural selection has apparently never found a way to undo this constraint (except perhaps to some extent in hibernators).

The alternative to one trait changing in response to another unchanging one is that two traits are about equally responsive. In such cases we expect the traits to mutually influence each other's evolutionary changes in a causal feedback loop. For instance, one prominent hypothesis for the evolution of larger brains is the expensive tissue hypothesis (see Section 24.2). This idea posits that, when for some reason an energetically expensive organ such as the gastrointestinal tract is reduced in size, this frees up energy. Selection can then favor a change in energy allocation that leads to increased investment in brains, which can therefore become larger. If this change then leads to an improvement in diet (for example, an increase in cognitive capacity leads to an increase in extractive foraging and thus to a higher-quality diet), then it can lead to a further reduction in the size of the gastrointestinal tract, and so on. This type of evolutionary feedback loop is called coevolution, or correlated evolution.

Comparative analysis can, in principle, distinguish between correlated evolution and evolution in response to an external signal or internal constraint. In practice, however, this distinction is difficult and is often made conceptually by the researcher and then examined more closely in other ways.

Understanding Genetic Constraint

The difficulty of distinguishing between the selective agent and the adaptive response requires that we develop a good grasp of the concept of (genetic) constraint. The key point is that organisms are not collections of loose traits but instead tightly integrated trait-clusters. The challenge for selection is to find a path from one cluster to the next, if such a path exists. When too many changes are involved, the organism may never be able to collect them together in one generation.

This point is appropriately illustrated with the classic hill-climbing metaphor in the genetic landscape (Figure 1.2). Selection can push a population uphill in the landscape (toward greater fitness), but never downhill. Selection also cannot look more than one generation into the future when considering which uphill path to take. As a result, distant peaks may be higher than the one currently occupied by a population, but those higher peaks are unattainable without first crossing a valley of reduced fitness, and so the population remains stuck where it is. Because history determined where in the genetic landscape the population happens to find itself, history constrains evolution: it places limits on the kinds of responses that can evolve. Put more poetically: adaptive pathways taken at one time may cast very long shadows into the future of a lineage (Seger & Stubblefield 1996).

c1f002

Figure 1.2 Illustration of the concept of genetic constraint. A population at the foot of the hill in this landscape must always climb up (a population's fitness cannot be reduced, if there is an opportunity to go up), and, given the maximum mutational step size, will go to the peak in (a) but get stuck at a lower peak in (b).

The phenomenon of constraint therefore has profound consequences for evolution. Once a lineage has committed to a particular bauplan, subsequent changes can no longer take place in all directions with equal probability: they are constrained. As a result, a species' phylogenic position strongly constrains what it can further evolve into (Futuyma 1998). Moreover, once lineages have diverged beyond a certain level, they will rarely converge again. Genetic constraint is therefore largely responsible for the amazing diversity in form and function that we see in nature, because it produces historical contingency – the fact that we can read a species' evolutionary history from its list of traits. All this allows us to conclude that natural selection can explain adaptation, complexity, and divergence all at the same time.

In some cases, the constraint is seemingly absolute. The deeper traits of an organism, those that are functionally integrated with many others, cannot, in practice, be modified by selection. Classic examples include the absence of a return to egg laying in eutherian mammals, or the absence of lactation in male mammals, even in species in which males have acquired a serious parenting role. Many morphological examples of suboptimality can also be given, where traits remain unchanged even if they have become maladaptive. An obvious example is that in land vertebrates food and air must cross on their way to stomach and lungs, respectively, with – in humans at least – the occasional choking death as a result. This suboptimal design is a leftover from their origin in fish-like organisms, prior to the evolution of the current mammalian respiratory system (Shubin 2008). Adaptations can still arise, of course, but they can be seen as a local solution, with a more global solution being prevented by this deep historical constraint.

One is inclined to believe such strong constraints are only found for morphology, but there are nice examples of constraint where the traits are purely behavioral. Consider the evolution of cooperative breeding. A detailed analysis in birds found that cooperative breeding almost always requires as a precondition that the species lives in family systems, that is, that fledged young remain with the parents for a considerable time (Drobniak et al. 2015). Once family groups exist, only one change is necessary: the independent young need to start helping. The direct route from pair living to cooperative breeding requires two steps: retention of the young and helping by these retained young. Comparative analysis reveals no cases of evolution having ever made these two such steps simultaneously. Thus, cooperative breeding is constrained by the presence of family groups.

In other cases, it is not that a particular new trait cannot evolve, but that another one evolves first and then takes care of business. This process could explain the evolution of adult food sharing among primates (Figure 1.3). In many primate species, adults with valuable relationships tend to show behaviors that help to maintain (service) their social bonds. One such behavior is food sharing. Given its social benefits, one might therefore predict that food sharing would be ubiquitous across species with strong adult social bonds. Yet this is not so: only a minority of the species with strong adult social bonds exhibit food sharing. Indeed, we only see food sharing among adults in those species where mothers also share food with their offspring. Selection picked those behaviors to service bonds that already happened to be available, albeit used in another context, and in species without mother-infant food sharing, selection picked another behavior, such as grooming. Constraint is therefore the reason for the prominent explanatory role of exaptation in evolution: without constraints, exaptations (such as mother-infant food sharing being extrapolated to adult-adult food sharing) would not need to be invoked.

c1f003

Figure 1.3 Constraints on evolutionary change in behavior: evolution of food sharing among adult primates. The 17 species with adult food sharing turn out to be a nested subset of the 38 species with mother-infant food sharing, indicative of the operation of constraint. After Jaeggi and van Schaik (2011).

Cost is Not Constraint

The alternative to invoking constraint as an explanation for why a trait did not evolve is to invoke high costs. This alternative implicitly assumes that there is enough genetic variation for selection to find a way to overcome the constraint. For example, we will see that a larger brain, which tends to allow a better response to physical, biotic, or social challenges, requires so much energy that it will hamper growth or reproduction, and thus may not provide a net fitness benefit, especially in organisms with fast life history and thus short life expectancy (see Section 24.3). Thus, high energetic and therefore, ultimately, fitness cost may prevent the evolution of larger brains.

It is not always easy to distinguish between constraint and high costs. Neuroscientists, for instance, often invoke the idea that the limited abundance of omega fatty acids, which are essential for adequate brain development in human children, has prevented the evolution of larger brains in early hominins (and perhaps other species as well). Human brains, so the argument goes, could only really expand once we had learned how to catch fish. This is a constraint explanation because it assumes that the factor limiting the developmental process is also limiting the trait on evolutionary time scales. The alternative is that other costs prevented this from happening, and that selection could have found other ways to solve this problem, for instance by synthesizing these compounds when they turn out to be essential, finding other dietary sources of it, or finding other compounds to use in nervous tissue with the same function. Many evolutionary biologists reflexively favor the second kind of explanation, but we generally need additional evidence to reject constraint explanations.

The evolution literature also often mentions physical constraints, but they are actually an example of limitations due to costs, since they are dictated by the physical laws of space and time, rather than the absence of sufficient genetic variation for selection to act upon. Thus, in the case of physical constraints, selection generates a compromise solution simply because a more optimal solution would defy basic physical laws. For instance, there is a limit on how much fat a bird can store because of the enormous impairment of flight and especially take-off velocity and thus potential to escape from predators (Creswell 2008). Similarly, there is a limit on the size of infants at birth due to physical limits on the birth canal. In humans this has become a real issue, and selection is forced to work around such a constraint by limiting the size increase of the infant at birth.

1.4 Evolution

Evolution refers to the historical changes in the traits of organisms. Evolution is a fact, as shown by the presence of fossils of species and inferred from the presence of highly divergent lineages of living organisms. Evolution can have multiple causes: natural or sexual selection, drift or migration. All of these can lead to changes in gene frequencies associated with the changes in traits of a population, but selection is the more persistent of these forces.

Microevolution is the change that happens in local populations over relatively short periods of time, often under the influence of natural selection. If we follow these changes over long periods of time, we see that populations become separated and diverge, and rarely become united again. Thus, the same selection process that produces adaptation also, over long periods of time, produces ever-increasing diversity. This sprawling tree of life arose from what was either a single or very few origins.

When evolving units of the same population, for whatever reason, become isolated from each other they will gradually diverge, and probably – at least to some extent – adapt to the slight differences in their local environments. If these populations meet again at some future point, they may no longer be reproductively compatible, or if they are, produce offspring with depressed fitness. Thus, the combination of adaptation and separation leads to new species, and so gradually to ever increasing diversity in forms, functions and sizes.

Although species can usually be recognized without too much trouble, species are an artificial category, even though a very useful one in practice (see Box 1.1). In particular, when dealing with changes over time, as paleontologists must do, it is simply impossible to clearly delineate one species from the next.

Speciation marks the boundary between micro- and macro-evolution, because species usually do not exchange genes and so evolve independently. As species give rise to other species they form lineages or clades, groups of species unified by being descended from a common ancestor (because of this the species within one lineage are called monophyletic). Macroevolution thus refers to longer-term changes involving features of species and clades. Macroevolution has emergent properties not directly predictable from knowledge of micro-evolutionary processes.

Box 1.1: What are species?

Humans divide the observed variability in natural organisms into discrete categories called species. We have done so at least as long as we have been humans, as indicated by the good correspondence between folk taxonomies and scientific taxonomies, which probably reflects features of our perceptual system. The early taxonomists, who by describing the species He created were trying to divine God's plan, also adopted this categorical view. Darwin was among the first to question the ‘reality’ of species, and wrote: I look at the term ‘species’ as one arbitrarily given, for the sake of convenience, to a set of individuals closely resembling each other (Zimmer 2008), this view being a direct consequence of a dynamic, evolutionary view, where species gradually arise (and may also become extinct again).

The current majority view of species is that they are distinct, evolving lineages. Thus, in most cases one can assign individuals to species, but these species are statistical entities, not real discrete classes in the way that individuals are. Where populations or species are more isolated by impassable regions or mating barriers, or where this isolation was longer, they will be more distinct from related lineages. The confusion, therefore, is hardly about what species are, but instead about how to recognize them (Zimmer 2008). Helpful criteria to determine species status are: (i) the depth of history as a recognizable lineage; (ii) the degree of gene flow with other lineages (especially through hybridization); and (iii) the presence of a distinct ecological niche, broadly defined (thus also including for instance the social system).

Paleontologists face a particularly tough problem: species change over time, and this can be quite gradual, but those who deal with fossils must somehow name them, if only to enable communication. The obvious practical solution is to draw lines where morphological differences are similar to those among contemporary species, but this does not solve the problem of what to do with intermediate forms. Neontologists face a similar problem: ring species circle the globe and where the ends meet again, they may be incompatible. The absence of a solution to the problem of chronospecies (or ring species) explains why paleoanthropologists cannot seem to agree on the taxonomy of hominins (e.g. Cartmill & Smith 2009). This problem becomes more acute as more fossils are known for a group, and thus gradual changes in traits become more likely.

Rules and Trends in Macroevolution

Evolution is a historical process, and like any such process can have a net direction in the long run, even if selection is very short-term, merely favoring those variants that in the current conditions outperform the alternatives. Yet, clear trends exist in the history of life. A basic explanation for the presence of an overall trend is expressed in Biology's First Law (McShea & Brandon 2010): a general increase in diversity and complexity.

Here, we are less concerned with the diversity and focus on the complexity. Over time, assuming that both increased and reduced complexity are equally likely to arise, when life historically started out simple some complexification is bound to arise. It is not inevitable, however, because more complex systems are more vulnerable to developmental mishaps or general malfunctioning. Thus, it could be that as complexity increases, changes toward higher complexity are less likely to confer fitness benefits than changes in the opposite direction. Yet, complexity has increased over time, and there are two major explanations.