How Dogs Work

How Dogs Work

How Dogs Work

Raymond Coppinger Mark Feinstein

Foreword by Gordon M. Burghardt

The University of Chicago Press : Chicago and London

RAYMOND COPPINGER is professor emeritus of biology at Hampshire College. His books include Dogs: A New Understanding of Canine Origin, Behavior, and Evolution, also published by the University of Chicago Press. MARK FEINSTEIN is professor of cognitive science at Hampshire College.

The University of Chicago Press, Chicago 60637

The University of Chicago Press, Ltd., London

© 2015 by The University of Chicago

Foreword © 2015 by Gordon M. Burghardt

All rights reserved. Published 2015.

Printed in the United States of America

24 23 22 21 20 19 18 17 16 15 1 2 3 4 5

ISBN-13: 978–0-226–12813–9 (cloth)

ISBN-13: 978–0-226–32270–4 (e-book)

DOI: 10.7208/chicago/9780226322704.001.0001

Library of Congress Cataloging-in-Publication Data

Coppinger, Raymond, author.

How dogs work / Raymond Coppinger and Mark Feinstein.

pages cm

Includes bibliographical references and index.

ISBN 978-0-226-12813-9 (cloth : alkaline paper) — ISBN 978-0-226-32270-4 (ebook) 1. Dogs—Behavior. 2. Canis—Behavior. I. Feinstein, Mark H., author. II. Title.

QL737.C22C64 2015

636.7—dc23

2015015809

♾ This paper meets the requirements of ANSI/NISO Z39.48–1992 (Permanence of Paper).

In memory of Erich Klinghammer, the founder of Wolf Park

Contents

Foreword by Gordon M. Burghardt

1. What Are Dogs Like?

2. What Makes Ethologists Tick?

3. The Shape of a Dog Is What Makes It Tick

4. The Shape of Behavior

5. The Rules of Foraging

6. Intrinsic Behavior

7. Accommodation and Behavior

8. Emergent Behavior

9. Play

10. Minding the Dog

A Last Word

Acknowledgments

References

Index

Plates

Foreword

It is a great pleasure for me to compose the foreword to this book by Ray Coppinger and Mark Feinstein devoted to the ethology of dogs. It is doubly a pleasure as the book is dedicated to both Wolf Park in Indiana and the late Dr. Erich Klinghammer, who founded it. Erich Klinghammer was one of my teachers, a member of my dissertation committee at the University of Chicago and, subsequently, a long-time friend. Although he began his career by carrying out some of the first studies on sexual imprinting in altricial bird species, he always loved dogs, especially German shepherds. It was his dog Gitta who, over fifty years ago, found a pregnant garter snake on his research farm in northern Indiana. Knowing my interest in snakes, Erich gave her to me; the babies born to this snake led to my first snake research, resulted in my dissertation topic, and were a key factor in my subsequent career in serpentine ethology. Both Erich and Gitta were acknowledged in the first publication (Burghardt 1966); that Gitta was a dog seemed irrelevant to mention. So dogs and Erich greatly influenced my career. When Erich developed serious allergies to birds, he changed his scholarly interest to the behavior of dogs and other canids, advocated for proper captive conditions for wolves, championed wolf conservation, and promoted ethology by translating important books from German.

Ray Coppinger, first author of this book, I have known for almost as long, and with great fondness, since his sense of humor broke the tension of perhaps the most embarrassing event in my life as a scientist. This occurred at an American Association for the Advancement of Science/Animal Behavior Society meeting in Dallas in 1968 when he followed me in the same session, presenting his dissertation work on birds. So Ray is also a refugee from the avian world, who with students and colleagues has studied the detailed workings of dog behavior. His important, but underappreciated, theoretical writings (Coppinger and Smith 1989) greatly informed my own thinking on the evolution of animal behavior, especially play.

This book on the ethology of dog behavior rides the crest of a renewed scientific interest in dogs and their evolution, behavior, cognition, and domestication. The great apes, humans’ closest relatives, have been iconic mirrors for human behavior going back to the nineteenth century, including the famous World War I–era insight experiments of Wolfgang Kohler, captive studies of Robert Yerkes in the 1920s, the pioneering field research of Jane Goodall in the 1960s, the language trained apes of the 1970s, and the cognitively and socially adept apes so widespread today in both popular books and the burgeoning field termed comparative cognition. But dogs are pushing their way into the august pantheon of super smart animals—to the extent that ape researchers are switching some of their efforts to our canine cousins.

Actually, however, the history of dogs as useful entries into the mysteries and origins of human behavior is quite ancient. Charles Darwin, who loved dogs more than any other animal, discussed breed differences in behavior and their scientific importance in his famous chapter on instinct in the Origin of Species in 1859. In his later Descent of Man and Selection in Relation to Sex (Darwin 1871) and in his book on emotions (Darwin 1872), he used dogs as exemplars of his three principles of emotions as well as arguing that dogs have attributes we think most distinctive of humanity such as loyalty, love, jealousy, pride, shame, imagination, reason, abstraction, and rudiments of language. In the seminal book on mental evolution by Darwin’s protégé, George John Romanes, dogs, but not other carnivores, were ranked with apes as most humanlike in their mental attainments (Romanes 1883).

Konrad Lorenz, a key founder of ethology, also had a lifelong love for dogs, resulting in the first popular ethological book on dog behavior and evolution, Man Meets Dog (Lorenz 1954). Here a more scientifically measured and biologically objective treatment of dog behavior was presented. Interestingly, it took the cognitive revolution in comparative psychology and ethology to bring dogs back to the forefront of research, accompanied by modern molecular genetics working out the relationships among the various breeds, to make it timely to look closely once again at dogs through an ethological lens. How Dogs Work is a true milestone and worthy successor to Lorenz. It gathers together much of the massive amount of new biological and ethological findings on dogs, wolves, and their relatives and is authoritatively written by canine behavior researchers with a long history of seminal contributions. Furthermore, the authors focus on particular breeds rather than the generic dog, maintain an ethological approach throughout, and, like Lorenz, present provocative, not consensus, views on many important aspects of dog behavior. Thus, not only will dog owners and aficionados learn much from this wide-ranging book about both dogs and science but dog professionals and canine scientists will have some persistent beliefs challenged as well.

This last claim derives from the fact that current approaches to comparative animal cognition using ingenuous methods not only highlight the cognitive complexity and problem-solving abilities of apes, monkeys, dogs, and other species but also facilitate the entry of uncritical anthropomorphic thinking into the interpretation of behavior. Such anthropomorphism, often considered sins that only pet owners and nonscientists commit, can actually enter the work and vocabulary of professional scientists themselves. This happens due to competition among researchers to show that apes, dogs, and other animals have humanlike abilities based on a tendency to view other animals through our psychology, not theirs. Such competition has led to continued debates between the promoters of animals having human smarts and killjoys who try to keep researchers grounded on the most parsimonious, even if strained and unlikely, interpretations.

Enter this stimulating book, which negotiates between cognitive and behaviorist extremes and discusses and interprets the behavior of dogs with the rich conceptual tools of ethology and basic animal learning processes. Although focusing on the fascinating behavior of sled, guard, and herding dogs, as well as wolves, the modes of thinking about behavior illustrated here can and should be extended to many more canine breeds and relatives. This book—unabashedly adapting and applying central tenets and methods of comparative ethology to understanding dogs—actually sets forth new ways of looking at the behavior of any and all species, including our own. It takes seriously Lorenz’s insight that behavior is as much a characteristic trait of animals’ species as are their anatomy and physiology. The wonderful photographs accompanying the text enable readers to interpret the postures, expressions, and behavioral dynamics of dogs and wolves, greatly enhancing understanding and enjoyment of these animals, much as being able to identify trees in a forest enriches the experience of hiking in the woods.

The authors, provocatively viewing dogs and other animals as complex machines, firmly tie the study of specific types and sequences of behavior, as well as learning, development, emotion, and cognition, to an intimate understanding of body, brain, and the evolution and modification of instinctive mechanisms. Today, molecular genetics is providing details of evolutionary origins, and neuroscience is providing insight into how brain processes underlie behavior. Alone they are often disconnected topics, but through behavior, and especially with an ethological approach, they can be more effectively connected than through any other behavioral science.

Gordon M. Burghardt

References

Burghardt, G. M. 1966. Stimulus control of the prey attack response in naive garter snakes. Psychonomic Science 4:37–38.

Coppinger, R. P., and C. K. Smith. 1989. A model for understanding the evolution of mammalian behavior. In Current Mammalogy, ed. H. Genoways, 2:335–74. New York: Plenum.

Darwin, C. 1859. On the Origin of Species by Means of Natural Selection. London: Murray.

———. 1871. The Descent of Man and Selection in Relation to Sex. London: Murray.

———. 1872. The Expression of the Emotions in Man and Animals. London: Murray.

Lorenz, K. 1954. Man Meets Dog. Translated by M. K. Wilson. London: Methuen.

Romanes, G. J. 1883. Mental Life of Animals. London: Kegan, Paul, Trench, Trübner, & Co.

1

What Are Dogs Like?

This book is about the behavior of animals, and in particular about how dogs and other canids (like wolves and coyotes) make a living—what a biological organism like the dog actually does, and how and why it does what it does. We want to understand the forces and mechanisms that enable a dog to tick as it moves and acts in the world: why Border collies chase after sheep but livestock-guarding dogs don’t; why greyhounds make good racing dogs but dachshunds do not; why a newborn pup behaves differently from an adult dog.

For us as ethologists—scientists who systematically investigate the biological bases of behavior—the notion of an animal ticking away, something like a clockwork machine, isn’t just a clever metaphor. A machine is a device that does work by converting energy into action. Like any machine, a dog’s behavior results from the translation of energy into patterns of movement (and ultimately, in the case of biological organisms, into offspring). How it is built, the shape and organization of its parts, and how it acquires the energy it needs to work all determine what a machine will do and set limits on what it can do. In this book, we will ask you to think about dogs and other animals in something of the same way.

You may well immediately and fiercely object that a dog isn’t just some kind of mechanical wind-up toy. Surely, plenty of us believe, dogs have personalities and desires that we would never ascribe to a machine. Indeed, it is probably true that dogs and other animals do have minds that are at least something like our own. This is an exciting perspective; it is much in vogue in the popular media and it’s the focus of a great deal of research in a new field that has come to be known as cognitive ethology. We’ll look more closely at some of this work in a later chapter—but we will not often appeal to cognitive explanations in this book. Our aim is to see how much we can understand about why and how animals behave from the standpoint of traditional ethology: by considering how the bodies of organisms are constructed and how the shape of that biological machinery determines the patterns of movement and activity that are so important in their lives.

Since the Darwinian revolution, virtually all biologists—and most thoughtful people—have understood that all life on earth is related in an evolutionary web stretching over billions of years. The myriad properties of biological machines—the cells, tissues, and body parts and the processes that wire them together—are the result of evolutionary forces that have shaped and reshaped the genetic mechanisms that ultimately are responsible for translating energy into purposeful activity. Darwin’s great idea was that evolution—by means of natural selection favoring beneficial variations in form—results in adaptations that enable an animal to acquire energy by eating, to avoid hazards (like being eaten by others), and to reproduce themselves. The central insight of ethology is that an animal’s behavior, just like the organic parts that make up the physical form of a biological machine, is itself an adaptive product of those evolutionary forces.

If animals are indeed like machines in some sense, however, it goes without saying that they aren’t just simple mechanical devices. Brains, for instance, are surely a critical component of the biological machinery that leads to behavior in higher organisms like dogs (and us)—and the vertebrate brain is quite possibly one of the most complex objects on earth, if not in the entire universe. There are some four hundred billion stars in the Milky Way galaxy—and sixty trillion neural connections in a human brain. The brain of a dog is not quite so cosmically large, but it is nonetheless a formidably complex organ—and it is only one part of an intricately integrated biological machine. Without bones and guts, skin and musculature, eyes and ears, and other organ systems—all products of the animal’s genes, honed by evolution—an animal like a dog can’t adaptively convert energy into effective action. So behavior must be a consequence of the animal’s whole shape, the complex totality of interacting mechanisms that are constructed by its genes.

That said, the notion that animals are just (something like) simple machines does have a long intellectual history. A few centuries ago, the philosopher René Descartes famously articulated the doctrine of dualism, arguing that body and mind are two different kinds of thing, neither reducible to the other. He saw human beings as possessing both properties. But nonhuman animals, Descartes insisted, were essentially just like cleverly constructed clockwork devices that possess only mechanical bodies and no soul-stuff. We are not going to take on this ages-old and still-ongoing philosophical debate about the relationship between body and mind. But it is instructive to think about why Descartes was able to view animals through the metaphorical lens of a ticking machine like the works of a clock—and how that idea might help us in understanding behavior.

Clocks are machines that mark out time. Inventive humans have come up with a wonderful range of ways to do this. Sundials indicate the passage of time by a shadow that relates to the position of the sun in the sky. Candle clocks, water clocks, and hourglasses can do the job because they use up various materials at a predictable rate. The mechanical clocks that appeared in the Middle Ages, and were much refined by Descartes’s time, worked on a different principle. These devices transferred mechanical movement—the swinging of a pendulum or the deformation of a spring controlling an oscillating wheel—to moving gears. Moreover, early clockmakers and other tinkerers soon discovered that their intricate mechanisms could do much more than just tell time—they could also cause other complex movements of many sorts. Eighteenth-century inventors thus delighted in building remarkable automata, clockwork machines that were intended to be realistic models of people and animals that could simulate lovers kissing, soldiers firing guns, or dogs chasing their own tails. Constructed only with gears that turned other gears, wires that pulled pieces into place, and pendulums that swung them away again, their machines could be made to appear to act in eerily but wonderfully recognizable ways. The tradition—now enhanced by sophisticated digital computational devices—continues today with the animatronic robots you’ll find at theme parks around the world. Over the centuries, countless audiences have marveled at these amazing self-operating machines: How lifelike they are! Some of these robots can be hard to distinguish, at least at first glance, from the real thing. But the fact is that even a very simple mechanical wind-up toy may have something of the behavioral flavor of a real organism (see, e.g., fig. 1).

Fig. 1 A wind-up mechanical toy dog ticking away. Drawing by Carol Gomez Feinstein.

Two things make these artificial machines and devices seem like they are (almost) alive. They have the overt physical form of animals (or people); and they move like them. In effect, we see in these mechanical toys, these automata, the fundamental properties of animal behavior—which we define as the shape of an organism moving in space and time. This may seem to you like an overly simple definition. But we think it’s the right way to characterize what a natural organism is doing when we say that it behaves. With that definition in mind, it makes perfect sense to say that a machine—even one that is human-made—exhibits behavior.

How it behaves is a matter of how it is built and how its shape changes as the machine engages with the world. When a mechanical cuckoo clock tells time by the movement of gears driven by a wound spring, it is behaving by the same definition that characterizes the behavior of a biological organism. The shape of its gear mechanisms, the relation in space of one gear and its teeth with respect to another, changes over time as gears turn and engage others. In one gear-shape configuration, the clock will chime four times. A little figure in a Tyrolean hat may also pop out, circle, and bow four times because a pulley and lever have been engaged by the shape of the gears, by their position and movement in space. As time passes and the spring imparts energy to the gears, the shape of the clock’s innards changes—and its time-signaling behavior changes as well.

The analogy between human-built ticking machines and real animals can be quite compelling. Like a cuckoo clock that only strikes precisely on the hour and only on the hour, many birds exhibit courtship behaviors in which the male makes stereotyped (ritualized) movements in order to attract a female—for instance, a specific and fixed number of head-bobbing motions—that they perform only at a precise time of the year. Wolves come into season only once a year, in early winter. In any given year and area, their reproductive activity is synchronized and all of their pups will be born at just about the same time in early spring after an average gestation period of sixty-three days. Just as a striking clock is quiet between the hour marks, the wolf’s stereotyped courtship behavior is dormant between seasons.

We must emphasize again that real animals are—of course!—astonishingly more complicated and more subtly constructed than any cuckoo clock or cleverly designed mechanical toy dog. Actual biological organisms are made of quite different materials. They have remarkable parts that can sense and respond to aspects of the world beyond them (though these days artificial automata can be built with this kind of ability too). A clockwork dog gets its energy from a human hand winding a spring or perhaps from a battery; real dogs get their energy from food that people provide for them. Wild animals need to find their own energy sources somewhere out in the world in competition with countless others. Real animals also change over time: a chicken starts out as an egg. A newborn pup and an adult dog have very different shapes. Artificial devices, however, don’t remodel themselves into something new. And, perhaps most important, animals can replicate themselves. This is something no human-made machine can do (yet). Reproduction is an all-important part of the biological story; it’s at the heart of what we mean by life. It’s also an essential element in evolution and plays a crucial role in animal behavior.

It goes without saying, we hope, that animals weren’t designed by a clever inventor who put parts together simply and logically. Rather, the shapes of organisms and their capacities for movement are complex outcomes of the forces of natural selection and other evolutionary and developmental processes. These led to a vast variety of solutions to the challenges of life; how and why these solutions work at all is often perplexing. In contrast, understanding how a clock or an automaton functions is a relatively simple task: you can patiently disassemble them, identify their discrete parts and work out how they function together. Biological machines are much more opaque. Taking an animal apart, whether anatomically, physiologically, or behaviorally, is a daunting enterprise. The problem is that it’s often unclear what the components actually are, what they’re for, and how they fit together. From cellular mechanisms to neural organization, biological systems can be infuriatingly complicated; figuring them out has been the life’s work of many generations of many kinds of scientists.

Genes and the Behavior of Biological Machines

One thing that we do now understand with certainty, however, is that genes, the heritable chemical instructions encoded in DNA, are essential elementary components of the machine. They underlie its fundamental ability to replicate. An animal’s genetic information plays a key role in establishing the initial basic plan of the machine (often in close interaction with the environment in which the genes operate). Over a lifetime, and on a daily basis as well, the genes build and rebuild the organism, specifying the character and limits of its body shape and its ability to move at any given time.

From that point of view, how an animal behaves is necessarily—and always—shaped by the genes that govern its construction. A dog behaves like a dog because it has dog genes—because it is built like a dog and not like something else. In this sense, all behavior is genetic. How an animal is affected by the world around it, the degree to which its behavior might be modifiable by training and learning, even the ways in which it can represent and use information (its mind, if you will)—all of these are fundamentally limited by species-specific genetic characteristics that are, at its heart, the real subject matter of ethology.

We do have to be very careful and precise, however, when we say that all behavior is genetic. The actual patterns of movement that an animal exhibits are never explicitly written directly in the language of DNA itself; they are not laid out as such in the molecular code of a single gene. What the genes do is no more (and no less) than to trigger cellular mechanisms that give rise to proteins, building the animal’s body and regulating bodily processes that enable it to move and act in certain ways. In this sense, all behavior does indeed have—and must have—a genetic basis. But at the same time, paradoxically, it’s also right to say that there are no genes for behavior. What we mean is that there is no single gene for mate selection, no one gene that by itself controls the intricate motor patterns of predation. There are only whole bodies (and brains), built by the totality of gene expression, whose form allows for particular kinds of behavior.

So when we observe that racing greyhounds tend to run faster than dachshunds, does this mean it is a genetic property of the breed? In an important sense, the answer is yes. It is a result of the fact that a greyhound has a genome that builds an animal with the size, bone structure, musculature, and nervous system of a greyhound, giving rise to a shape that supports running fast. However, slow dachshunds don’t differ from greyhounds because they have different genes for speed: it is because dachshund genes build a different body with a different capacity for movement.

Go to a dog track, moreover, and you’ll see that some greyhounds are clearly faster than others. An individual dog that is ill-fed or poorly