How Monkeys See the World: Inside the Mind of Another Species

by Dorothy L. Cheney, Robert M. Seyfarth

Overview

Cheney and Seyfarth enter the minds of vervet monkeys and other primates to explore the nature of primate intelligence and the evolution of cognition.

"This reviewer had to be restrained from stopping people in the street to urge them to read it: They would learn something of the way science is done, something about how monkeys see their world, and something about themselves, the mental models they inhabit."—Roger Lewin, Washington Post Book World

"A fascinating intellectual odyssey and a superb summary of where science stands."—Geoffrey Cowley, Newsweek

"A once-in-the-history-of-science enterprise."—Duane M. Rumbaugh, Quarterly Review of Biology

Product Details

ISBN-13:	9780226102467
Publisher:	University of Chicago Press
Publication date:	04/15/1992
Edition description:	1
Pages:	388
Product dimensions:	6.00(w) x 9.00(h) x 1.00(d)

About the Author

Robert M. Seyfarth is professor of psychology at the University of Pennsylvania. He is coauthor with Dorothy L. Cheney authors of How Monkeys See the World, also published by the University of Chicago Press.

Read an Excerpt

CHAPTER 1

WHAT IS IT LIKE TO BE A MONKEY?

On the 5th of September, 1379, as two herds of swine, one belonging to the commune and the other to the priory of Saint-Marcel-le-Jeussey, were feeding together near that town, three sows of the communal herd, excited and enraged by the squealing of one of the porklings, rushed upon Perrinot Muet, the son of the swinekeeper, and before his father could come to his rescue, threw him to the ground and so severely injured him that he died soon afterwards. The three sows, after due process of law, were condemned to death; and as both the herds had hastened to the scene of the murder and by their cries and aggressive actions showed that they approved of the assault, and were ready and even eager to become participes criminis, they were arrested as accomplices and sentenced by the court to suffer the same penalty (Evans 1906 / 1987:144).

The preceding account is by no means unusual. Throughout Europe during the Middle Ages animals ranging from insects to horses were commonly brought to trial, provided with defense lawyers, and charged with crimes such as the willful destruction of crops, murder, and sodomy. People in the Middle Ages were not careless anthropomorphizers; they clearly recognized that animals were not people. Nevertheless, the behavior of their animals often led people to believe that animals could be aware of what they did and held accountable for their acts. Why are we sometimes willing to attribute intent, beliefs, and consciousness to animals? What aspects of their behavior suggest the operation of complex mental processes?

We borrowed the title for this chapter from an article written in 1974 by the philosopher Thomas Nagel, who asks: "What is it like to be a bat?" Not a particularly auspicious beginning, since Nagel's conclusion was, essentially, "We can never know." Bats use sonar echoes to navigate through their environment and regularly attend to high-pitched sounds that fall outside the range of human hearing. Because their sensory world is so different from our own, Nagel argues, we can never really understand what it is like to be a bat, no matter how much we may learn about the biological processes that underlie a bat's existence. Likewise, Ludwig Wittgenstein (1958) felt that the close intermingling of customs, concepts, and meaning in any society would always create barriers between different cultures or between different species. "If a lion could talk," he said, "we would not understand him."

From a quite different perspective, other philosophers have argued that, even if there are things we cannot learn about the mind of another animal (or another person), the very inaccessibility of these presumably deeper facts renders them scientifically negligible — or even nonexistent. In his book Word and Object, W. V. O. Quine (1960) imagines that a linguist has entered an unknown land where none of the sounds people make are familiar. The linguist's goal is to construct a dictionary of the local language, to learn what each word means, and in this way to begin to understand how the people think. But if a native shouts "Gavagai!" when a rabbit appears, can the linguist conclude that gavagai means the same as rabbit in English? Quine believes that with no words in common and only gestures like pointing to go on, the linguist can never know. He might be able to specify the conditions under which gavagai is uttered, but his assessment of precisely what this sound means to the native will always be an approximation. Of course this residual uncertainty, Quine adds in mitigation, is no worse than the uncertainty we have about what our own words mean. We can never be sure that the others who (apparently) speak our language match objects and words in exactly the same way we do. So Quine concludes that the barriers to understanding another species are virtually insurmountable, but in principle no greater than the barriers to understanding our friends and neighbors.

We are, obviously, different from other animals, and these differences inevitably complicate any attempt to understand how animals communicate and how they see the world. Nevertheless, we hope to convince you that Nagel and Wittgenstein have been too pessimistic and have declared impossible what is merely difficult — and fascinating. Their reservations need not preclude research; on the contrary, we hope to show that we can learn a great deal about the mind of another species. We do this by describing the work of a number of ethologists, including ourselves, who — like Quine's imaginary linguist — have gone into the field intent on deciphering the gestures and sounds made by different creatures, particularly monkeys and apes.

Humans have always been curious about the behavior of nonhuman primates. During the past 30 years, however, there has been renewed interest in primate behavior and communication and in what such research might reveal about how animals think. The motivation to probe into the minds of monkeys and apes has come from at least four sources.

First, the richly detailed, descriptive accounts of primate behavior produced during the past 20 years (e.g., Goodall 1983, 1986; de Waal 1982, 1989; Fossey 1983; and Smuts 1985) raise important — and largely neglected — questions about how animals communicate and how they might deceive, plot strategies, or attribute motives to others. Much of the material in these books is anecdotal, but there is a growing awareness among ethologists that such anecdotes should be taken seriously and can, at the very least, serve as working hypotheses for further observation and experimentation (Byrne and Whiten 1988b, 1988c; Whiten and Byrne 1988c).

Second, the well-known ape language projects (e.g., Gardner and Gardner 1969; Premack 1976; Rumbaugh 1977; Patterson 1978; Terrace et al. 1979; Miles 1983; Savage-Rumbaugh 1986) have demonstrated clearly that apes in the laboratory can learn elements of human language. In some cases, research on captive apes has also revealed cognitive abilities not yet demonstrated for the same species in their natural habitat. Some people involved in these projects explain the disparity between performance in the laboratory and the field simply as a function of training: experience in the use of signs provides animals with skills they would, under natural circumstances, neither need nor possess (e.g. Premack 1983b; Savage-Rumbaugh et al. 1983). This conclusion, however, may be premature, because no one has yet gone into the wild and systematically presented apes with the same sort of problems they face in the laboratory.

To evolutionarily oriented ethologists, by contrast, the disparities between field and laboratory are puzzling and provide an intriguing challenge. Evolutionary theory holds that general skills, such as the ability to use signs as representations of objects or to make judgments based on analogical reasoning, do not evolve unless they serve some adaptive function (e.g., Humphrey 1976; Dawkins 1986). The evolutionary view assumes as its working hypothesis that the skills uncovered by research on captive apes are part of the animals' natural behavior and are used regularly under natural conditions. The lack of systematic data on such abilities simply means that primatologists have not looked closely enough at the problems their subjects face in the wild and the skills they use to solve them. At the very least, this evolutionary view provides a framework within which field and laboratory studies can be compared and interesting questions can be addressed.

Third, spurred on by developments in evolutionary theory, ethologists have demonstrated that many animals, from insects to apes, have an extraordinarily sophisticated knowledge of certain aspects of their environment and their social companions (reviewed in, for example, Griffin 1984). In their natural habitats animals act as if they can recognize kin, assess the suitability of mates or the size and aggressiveness of opponents, remember past social interactions, and solve complex problems that allow them to locate and harvest food efficiently (see, e.g., Holmes and Sherman 1983; Cheney, Seyfarth, and Smuts 1986; Kamil, Krebs, and Pulliam 1987; Waldman, Frumhoff, and Sherman 1988 for reviews). The picture of animal intelligence that emerges from fieldwork is richer and more complex than earlier studies, conducted mainly on laboratory animals, would have led us to expect. This very richness, however, makes it all the more challenging to specify how animals and humans differ. Toads, ground squirrels, and monkeys may respond differently to full- and half-siblings, but their understanding of kinship is unlikely to be the same as our own. How should we characterize the difference?

Fourth, there is increasing awareness that studies of animal social behavior can contribute to the growing area of research that has been broadly labeled cognitive science. Here the ultimate object of interest — for philosophers, linguists, computer scientists, anthropologists, and psychologists — is the human mind. How does the mind represent and process information? Can we build a machine, find another species, or write a computer program that duplicates the mind's performance? What would it mean if we could? One research strategy, pursued in a variety of forms, has been to investigate "almost minds," such as the minds of children or the "minds" of computer programs, to see what makes them different from our own and what would be needed to elevate them to adult human status. As we hope to demonstrate, the social behavior of nonhuman primates offers a glimpse of almost minds at work — a glimpse that may ultimately tell us how, in the course of our own evolution, some minds gained an advantage over others.

How We Approach The Problem

Many scientists implicitly assume that, among all animals, the behavior and intelligence of nonhuman primates are most like our own. Nonhuman primates have relatively larger brains and proportionately more neocortex than other species (e.g., Passingham 1982; Martin 1983; Jerison 1985), and it now seems likely that humans, chimpanzees, and gorillas shared a common ancestor as recently as 5 to 7 million years ago (Weiss 1987). This assumption about the unique status of primate intelligence is, however, just that: an assumption. The relation between intelligence and measures of brain size is poorly understood, and evolutionary affinity does not always ensure behavioral similarity. Moreover, the view that nonhuman primates are the animals most like ourselves coexists uneasily in our minds with the equally pervasive view that primates differ fundamentally from us because they lack language; lacking language, they also lack many of the capacities necessary for reasoning and abstract thought (Premack 1983b).

To make our own position clear at the outset, we accept the broad evolutionary view which argues that, just as one can learn about the origin of the human hand or brain by studying the anatomy of nonhuman primates, so research on nonhuman primate behavior can shed light on the origins of human language, cognition, and self-awareness. This does not mean that modern monkeys and apes present us with a complete and accurate picture of the ancestral human condition, nor that similar behavior in human and nonhuman species is necessarily caused by similar underlying mechanisms. At a purely descriptive level, for example, there are many similarities between the interactions that occur within monkey matrilines and the interactions that occur within human families. Such data do not, however, prove that human and nonhuman primates recognize kin or compare social relationships in exactly the same way (e.g., Hinde 1987). An evolutionary perspective draws our attention to the many apparent similarities between human and nonhuman primate behavior. Given these observations, the goal of our (as well as many other scientists') research is to probe more deeply and establish where such similarities break down.

Perhaps more important, taking an evolutionary perspective means that we approach the study of primate intelligence from a practical, functional perspective. What are the problems monkeys face in their daily lives? What do they need to know, and how might one method of obtaining and storing knowledge give certain individuals a reproductive advantage over others? Done with care, such analysis can suggest not only how human intelligence evolved but why.

Regardless of the theoretical view one adopts, the problem of investigating intelligence empirically remains. The difficulty arises in part because field and laboratory studies of intelligence have traditionally used different methods to address fundamentally different questions.

In the laboratory, learning and intelligence have usually been measured by discrete and quite specific experiments (see Skinner 1974; Rescorla 1985; Herrnstein 1970; Essock-Vitale and Seyfarth 1987; Kamil 1987; and Roitblat 1987 for reviews). Whether they measure performance in the formation of associations, "learning sets," oddity discrimination, analogical reasoning, or language learning, such experiments are attractive for a number of reasons. Their precision and control, for example, are unlikely to be matched by any study conducted in the field. Different experiments can focus precisely on different cognitive skills and allow one to state explicitly what would constitute evidence that an individual possesses a particular ability. Laboratory tests provide one means of comparing intelligence across species (but see Warren 1973) and, since similar tests can be conducted on humans, also allow comparison between human performance and that of closely related species.

At the same time, the very controlled, objective nature of laboratory experiments creates problems. For example, the goal of many studies of captive primates has been to formulate one measure or set of measures that permits comparisons of intelligence in different species (e.g., Harlow 1949; Schrier 1984). Although they are tightly controlled and systematic, such experiments are typically divorced from questions of evolutionary function, and their relevance to the animals' natural social behavior is often unclear. Moreover, cross-species comparisons generally ignore the fact that different species have evolved in different social and ecological environments. Given any one task or set of tasks, some species will invariably perform better than others, and it will always be possible to argue that one species is predisposed to learn certain tasks more easily than others (Seligman and Hager 1972; Hinde and Stevenson-Hinde 1973), that one species is simply more familiar with the stimuli in question, or that one species has more experience with the type of question being posed (e.g., Bitterman 1965; see also Kamil 1987). Of course, this kind of "species relativism" easily becomes tautological; to compare species fairly, one must control all possible contextual features that might result in a performance difference, which in turn makes it almost impossible to determine if a species difference exists. As Niko Tinbergen (1951: 12) remarked, "One should not use identical experimental techniques to compare two species, because they would almost certainly not be the same to them."

In their search for an objective measure of performance, laboratory psychologists have traditionally designed tests using arbitrary stimuli like lights, tones, or different-shaped blocks that animals would never encounter in their natural habitat. This method allows tests on different species to be compared directly, but it also increases the likelihood that results will underestimate or fail to reveal a subject's true ability. An animal may not understand the problem or it may simply lack the motivation to perform in what it regards as an unfamiliar or even hostile environment (e.g., Terrace et al. 1979).

Finally, despite overwhelming evidence that primates are social creatures, laboratory tests have generally ignored the social dimension of primate intelligence. With some notable exceptions (which we discuss in subsequent chapters), few tests have been designed to mimic the social problems faced by group-living animals and few tests have measured an individual's performance using social stimuli, like the pictures or voices of familiar conspecifics.

The research we describe here takes a different approach to the study of how animals think. Drawing on our own work with East African vervet monkeys (Cercopithecus aethiops), and the research of scientists studying other species, we document what monkeys and apes do in their natural habitats and consider what sorts of underlying mental operations might possibly account for this behavior. Although the virtue of this approach is that it considers behavior and communication within an evolutionary framework, its drawback is its heavy reliance on observations, experiments, and anecdotes that may suggest, but can never definitively prove, what an animal's mental abilities are.

To put our approach in perspective, we begin by outlining a philosophical debate that we are not really qualified to discuss and will not pursue at great length. For more detailed reviews of the history of ethology and comparative psychology, together with the debates that have punctuated their development, we refer the reader to Hinde (1982), Dewsbury (1984), Boakes (1984), and Roitblat (1987).

(Continues…)

---

Excerpted from "How Monkeys See the World"
by .
Copyright © 1990 University of Chicago.
Excerpted by permission of The University of Chicago Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

---

Table of Contents

Acknowledgments
1. What Is It Like to be a Monkey?
2. Social Behavior
3. Social Knowledge
4. Vocal Communication
5. What the Vocalizations of Monkeys Mean
6. Summarizing the Mental Representations of Vocalizations and Social Relationships
7. Deception
8. Attribution
9. Social and Nonsocial Intelligence
10. How Monkeys See the World
Appendix
References
Index

Editorial Reviews

An exploration of communication and intelligence in free-ranging primates. Much of the text is devoted to the authors' field studies on velvet monkeys in East Africa. Theories of modern cognitive science are applied in order to discover the similarity and differences between the cognition of monkeys and humans. Some topics discussed are: social behavior, vocal communication, deception, and attribution. Annotation c. Book News, Inc., Portland, OR (booknews.com)

Booknews