Greg Radick, 2006. “What’s in a Name? The Vervet Predator Calls and the Limits of the Washburnian Synthesis.” Studies in History and Philosophy of Biological and Biomedical Sciences 37: 334-362.


After the Second World War, a renaissance in field primatology took place in the United States under the aegis of the ‘new physical anthropology’. Its leader, Sherwood Washburn, envisioned a science uniting studies of hominid fossils with Darwinian population genetics, experimental functional anatomy, and field observation of non-human primates and human hunter–gatherers. Thanks to Washburn’s stimulus, his colleague at Berkeley, the bird ethologist Peter Marler, took up the study of the natural communicative behaviour of apes and monkeys. When Marler’s first primatological student, Thomas Struhsaker, reported in the mid-1960s that the vervet monkeys of Amboseli, Kenya, give acoustically distinct alarm calls to different predators, and respond to alarm calls as if to the sight of those predators, a debate broke out over whether the vervet calls thus function as names, translating as ‘leopard’, ‘eagle’ and ‘python’. Washburn and his students argued that no matter what the behavioural evidence, vervet calls could not be predator names, since monkeys had been shown to lack the neuroanatomical basis of naming. This controversy thus reveals, first, the persistence of older patterns of disciplinary allegiance within the new, synthetic physical anthropology; and second, the impotence of adaptationist Darwinism — common to both sides of the debate — as a force for unity.


  • Sherwood Washburn
  • Peter Marler
  • Physical anthropology
  • Primate communication
  • Language
  • Neuroscience

1. Introduction

Since the 1960s, discussion of how animal communication relates to human language has turned again and again to the predator calls of the vervet monkeys of Amboseli, Kenya.1 The attraction is threefold. First, the calls are tantalizingly word-like. Amboseli vervets, it is said, give acoustically distinct alarm calls to three different predators — leopards, eagles and pythons — and respond to these calls as if informed about which predator is present. ‘Leopard’ calls send vervets running from the ground to the trees, where leopards are no threat. ‘Eagle’ calls send tree-bound vervets running to the ground and into the dense cover that protects against eagles (but that can conceal leopards). ‘Python’ calls elicit looks down at the ground: a less dramatic response, but then pythons are less of a threat to vervets than are leopards and eagles.2

Second, the fact that these are East African monkeys makes them plausible placeholders for pre- or proto-linguistic human ancestors. A major contrast here is with honeybees. Since the Austrian zoologist Karl von Frisch’s famous studies in the first half of the twentieth century, the honeybee ‘dance language’, apparently conveying detailed information about food sources, has stood as perhaps the best known example of symbolic communication among non-human animals.3 But bees are extremely remote from humans evolutionarily. Savannah-dwelling vervets, by comparison, are near kin. Accordingly, the vervet predator calls are much more serious candidates for attention from anyone concerned to trace the origins of human language from its animal antecedents.4

Third, the calls have been the subject of consistently innovative scientific work. In the 1960s, a behavioural zoologist, Thomas Struhsaker, published the first description of vervet alarm behaviour as part of a richly detailed and much admired survey of vocal communication among the Amboseli vervets.5 In the 1970s, Struhsaker’s former supervisor, the ethologist Peter Marler, together with two postdoctoral students, the husband and wife team of Robert Seyfarth and Dorothy Cheney, followed with field experiments designed to test whether the vervet alarm calls truly convey the predator-specific information claimed for them.6 These experiments, involving the ‘playback’ of previously recorded alarms to vervets when there were no predators around, were the first of their kind since the late nineteenth century.7 The finding of evidence for ‘rudimentary semantic signals’ in vervets made the news internationally.8 In the 1980s, Seyfarth and Cheney, now regarded as standard-bearers for a new ‘cognitive ethology’, pushed the use of experimental field playback further still in pursuit of the meaning of the vervet predator calls, notably by introducing variations inspired by studies of preverbal infants.9

It is the earliest, 1960s portion of this scientific trajectory that this paper seeks to understand. The story of how the vervet predator calls came to scientific attention in the first place, and how they came to prominence as, potentially, names given by non-human animals to objects in their environment, is one that provides a new angle of approach to the history of several modern behavioural sciences. As we shall see, Marler was a participant–observer among the first generation of postwar primatological fieldworkers—an expert on bird songs and calls who brought to the primate data a highly developed perspective on animal vocal communication as information transfer. It was this older, ethological preoccupation that led Marler to pluck the vervet predator calls from Struhsaker’s field data and exhibit them as deserving special notice. At the same time, the manner of their display, as test cases on the borderline between affective and symbolic signalling, reflects the recent addition to the intellectual mix of new research in the neuroscience of language, most strongly associated with the aphasiologist Norman Geschwind.

One aim of this paper is to recall the circumstances under which the vervet predator calls came to be discovered and then discussed in the way they were. Another aim is to use this history to clarify the nature and limits of a synthetic project widely credited — in Donna Haraway’s Primate visions (1989) as elsewhere—with helping to bring modern primatology into being: the ‘new physical anthropology’ of Sherwood Washburn. Washburn and his students at Berkeley were instrumental two times over in the story that follows. It was they who brought Marler into the primatological fold, as part of Washburn’s campaign to align physical anthropology with the best in behavioural field studies of animals. It was, however, another of Washburn’s attempted alignments, with functional anatomy, that prompted a sceptical interpretation of the vervet calls not as names for predators but as non-symbolic expressions of arousal. When the ethological, field-derived evidence pointed one way, towards the vervet calls as names, and the neuroscientific, laboratory-derived evidence pointed in the opposite direction, the Washburnians came down unreservedly on the side of neuroscience. The early debate over the vervet predator calls thus reveals a crucial but little-noted asymmetry among the disciplines making up the Washburnian synthesis. The fact that adaptationist Darwinism was common cause on both sides of the debate suggests further that, for all Darwinism’s attraction as a unifying principle, it has the vices as well as the virtues of interpretative flexibility.

2. Washburn, Marler and the new field primatology

‘Sherry’ Washburn was a one-man disciplinary revolution. For him, the old physical anthropology meant the kind he had learned at Harvard in the 1930s: typological, race-obsessed, and devoted above all to measuring in exhaustive detail the hard parts of human and hominid bodies. He had sensed early on how estranged this project was from the best of modern scientific thinking. The new physical anthropology as he promoted it dealt instead in populations and the functional demands that coordinated bones, tissues and behaviour into adaptive wholes. It avoided even raising questions of ‘racial quality’. Changes in practice went with these changes in orientation. Alongside measurement, according to Washburn, there needed to be experiments, of the sort he had done during the war years at Columbia medical school, cutting and reconnecting muscles in the heads of young rats in order to understand the forces shaping hominid skulls. There also needed to be observations of social behaviour in living hunter–gatherers and non-human primates. Evolutionary functionalism was the intellectual glue holding these diverse activities together. The dynamic processes available to scientists now — the growth of muscle and bone in a rat’s head, the grooming habits of a troop of monkeys — could be regarded as surrogates for the processes at work in the past because, in Washburn’s view, Darwinian natural selection adapts organic form to function, whenever and wherever it operates. ‘The task of the anthropologist’, he affirmed, in a paper delivered before G. G. Simpson, Ernst Mayr, Theodosius Dobzhansky and other proponents of the new synthetic evolutionary biology, gathered at a famous 1950 symposium on human origins, ‘is to fit knowledge of the primates into the framework of modern evolutionary theory’.10

The infusion of functional thinking and experimental tinkering into anthropological anatomy was the central reform, since anatomical studies were the main business of physical anthropology, and non-functional, non-experimental approaches had long dominated. As Washburn wrote in 1952:

Traditional physical anthropology was based on the study of skulls. Measurements were devised to describe certain features of the bones, and, when the technique was extended to the living, the measurements were kept as close to those taken on the skeleton as possible. From a comparative and classificatory point of view, this was reasonable, and for a while it yielded useful results, but it brought the limitations of death to the study of the living. Whereas the new physical anthropology aims to enrich the study of the past by study of the present, to understand bone in terms of function and life, the old tried to reduce the living to a series of measurements designed to describe bones.11

Physical anthropology as an exclusively descriptive science was coming to an end. Washburn hailed the power of a functional, experimental approach to give meaning to the measurements which otherwise accumulated so pointlessly. To the traditional anthropometrist, a nose was just an opportunity for anatomical measurement, along this, that and the other dimension, with patterns then sought in the results. But considered functionally, a nose was part of a face adapted to the demands of chewing—demands revealed through experimentation and other non-biometrical studies:

Far from being an independent structure which can be described by itself, the nose is an integrated part of the face, and variations in its form can be interpreted only as a part of the functioning face. The form of the nose is the result of a variety of factors. Just how many and how they are interrelated can be discovered only by research, but it seems clear that the most important ones, as far as gross form is concerned, are the teeth and forces of mastication. But these are not included in the traditional descriptions of the nose, nor will looking at skulls or measuring them give this kind of information.12

The usefulness of behavioural studies, in addition to experimental ones, was something he had glimpsed as a graduate student, when he had accompanied the Asiatic Primate Expedition to Siam, charged with assisting the anatomist Adolph Schultz with dissecting dead gibbons and the psychologist C. Ray Carpenter with observing lives ones.13 It was not until the mid-Fifties, however, and some chance observations while in east Africa of free-ranging baboons — more conspicuously intelligent and complexly social than gibbons, and much more easily observed — that Washburn began to incorporate behavioural studies in a serious way into his disciplinary reforms.14 By now, he commanded the institutional resources to do it very effectively. A professor in the stellar anthropology department at the University of Chicago, with positions on all the major professional organizations and journal editorial boards, he was well placed to extract large sums from granting bodies, including the deep-pocketed Wenner–Gren Foundation, set up recently to advance anthropological understanding. The first beneficiary of Washburn’s well-funded behavioural turn was Irven DeVore, a Chicago graduate student whose background in social rather than biological anthropology was, in Washburn’s view, just what a decent observer of baboons needed. At first reluctant, DeVore quickly understood that this was not an invitation open for him to refuse; and within a few weeks, Washburn’s intelligence and enthusiasm had won the younger man around completely.15 In June 1959, a report on their joint field observations of baboons was presented to a conference memorialized two years later in The social life of early man (1961), edited by Washburn. Washburn and DeVore’s paper in the volume included what became a kind of emblem of the Washburnian approach: a chart comparing baboons and ‘pre-agricultural humans’ along eleven variables, including group size, diet, play, sexual behaviour, dominance and use of sounds and gestures.16

Fig. 1. Sherwood Washburn in 1963 (photograph from the Papers of S. L. Washburn, Series 7, Carton 6, Folder 31, Bancroft Library, University of California Berkeley; used with the permission of the Bancroft Library).

By 1961, Washburn had moved to Berkeley, taking the new physical anthropology with him, DeVore and all. Soon Peter Marler, based in the Berkeley zoology department, found himself teaching, examining and generally getting to know a growing number of anthropologists interested in animal behaviour. Marler had arrived at Berkeley in 1957 as an emissary from the still-largely European science of ethology, dedicated to describing and explaining the behaviour patterns of animals under natural conditions. He had studied at Cambridge, where two ethological enthusiasts, the former entomologist William Thorpe and his assistant Robert Hinde, had set up an ornithological field station in the village of Madingley. Like Thorpe, Marler had become fascinated by the fact that the songs of certain birds are not everywhere the same, but differ in stable ways from one region to another. Together the two men began experiments on song acquisition to tease out the mixture of instinct and learning that, as they and others surmised, accounted for this dialectical variation. By the time Marler left Cambridge for Berkeley, he had initiated, in addition to work on song learning (which would become the major thrust of his Berkeley laboratory), two other, related, highly original lines of investigation. One looked at how the acoustic structures of bird vocalizations — analyzed by a new device, the sound spectrograph — adapted these signals to their biological functions. The other considered the varieties of information conveyed in bird vocalizations and, increasingly, other animal signals. Up to the early 1960s, Marler’s professional work as an observer and experimentalist of animal communicative behaviour had thus dealt exclusively with birds, indeed local birds. At Berkeley, the chaffinches of Madingley wood gave way to the white-crowned sparrows of the Bay Area. When the time came for this intellectually adventurous ethologist to extend his first-hand knowledge in a new direction, it would have been natural enough to turn to the sorts of species studied by his non-birdsong students at Berkeley: the insects, fish or small mammals of California. That Marler turned instead to the monkeys and apes of Africa was due largely to the influence of Washburn.17

Ethological zoology and Washburnian anthropology were not, to be sure, wholly congruent. As we shall see, the different weightings they assigned to anatomical experiments in part led, a few years later, to Marler changing his mind about whether to engage the question of animal calls as symbolic. Still, there was overlap enough for fruitful collaboration. Along with the Berkeley comparative psychologist Frank Beach, Marler and Washburn created the University’s first animal behaviour station. It became home to a primate colony where anthropologists trained before heading out to the field. Marler thought Washburn himself enormously impressive—a kind of latter-day Cuvier, seemingly able to conjure vanished forms out of single bones.18 For his part, Washburn encouraged Marler’s primatological interests, asking him, for instance, to review a new book on field studies of the mountain gorilla by a young American zoologist, George Schaller, for Science.19 But the senior man was far too busy for the kinds of informal discussion that can transform curiosity into commitment. It was his students, and especially the gregarious DeVore, who did the most to tempt Marler towards primate behaviour.20 No doubt they were largely responsible for Marler’s agreeing, at rather short notice, to survey recent advances in the study of primate communication when the invitation came, around New Year 1963. As he set to work educating himself, primates were still no more than a side interest. But by the end, they had become rivals with the birds for his attentions.21

The invitation was from the ‘Primate Project’, a year-long symposium organized in part by Washburn, and taking place at the nearby Center for Advanced Studies in the Behavioral Sciences at Stanford in 1962–1963. As DeVore explained in his editor’s preface to the culminating volume, Primate behavior: Field studies of monkeys and apes(1965), the aim ‘was to bring together a group of people who had just completed long-term field studies of monkeys and apes, and by providing them with sufficient opportunities for discussion over many months, to arrive at mutual understandings concerning the description and interpretation of primate behavior’.22 Just why, at that moment, did so many people find themselves studying monkeys and apes in the field, or engaged in comparably serious studies of captive or semi-captive primates? We have looked at the considerations that led an American physical anthropologist to channel his students that way. But rather different considerations led others — and the bodies funding them — in the same direction.23 In England, for instance, there was now a colony of rhesus monkeys under observation at Madingley, thanks to Robert Hinde’s collaborations with the psychologist John Bowlby and their mutual interests in using primates to model the effects of parent–child separation in humans. As Hinde’s expertise with monkeys developed, he attracted the attention of the Anglo-Kenyan hominid palaeontologist Louis Leakey, who, independently of Washburn, had decided that living apes offer a window onto the pre-human past. Leakey had offered the job of ape watcher to his then secretary, Jane Goodall, got her started observing chimpanzees at the Gombe Stream Chimpanzee Reserve in Tanganyika, and then arranged a crash course for her in behavioural zoology with Hinde.24 Distinctive yet again was Stuart Altmann’s path to the primates. As a Harvard graduate student in biology in the mid-1950s, he had set out to revive and update the pre-war tradition of Carpenter, observing monkey societies in the field but theorizing about them as cybernetic systems. The project was regarded as so idiosyncratic that the only faculty member prepared to supervise was a young professor of ant behaviour, Edward O. Wilson.25

In the early 1960s, then, the primatologists — as they were not yet called — were a diverse group, with much to talk about. Marler was not of their number. But he had a matchless knowledge of animal signalling and a demonstrated ability to synthesize disparate materials. When the Primate Project fieldworkers turned to him, about midway through their year together, to distil some generalizations from their data on communication, he found the pickings extremely rich. A table assembled for his published paper, comparing the sizes of the vocal repertoires of various species, drew almost wholly on recent work. For the gorilla, there was Schaller’s field study of mountain gorillas in Congo. For the chimpanzee, there was, in addition to Goodall’s work (represented elsewhere in the volume, though not here), the field study of the London-based Vernon and Frances Reynolds in the Budongo Forest in Uganda. For the baboon, there were the field studies of DeVore (in Kenya) and the Bristol zoologist Ron Hall (in southern Africa). For the rhesus, there were Altmann’s field studies on Cayo Santiago, and the more recent work of Hinde and Thelma Rowell with the Madingley rhesus colony. For the Japanese macaque, there was data from Hiroki Mizuhara, a fellow at the Center and representative of a thriving Japanese primatological tradition. For the langur, there was the field study of a former Washburn student, Phyllis Jay, in north India. For the howler monkey, there was another of Altmann’s field studies, as well as Carpenter’s. And for the brown lemur, there was the work of Richard Andrew, a graduate (like Rowell) of Cambridge ethology, now based at Yale and absorbed in investigating vocalizing primates in his lab and at the Bronx Zoo.26

What did Marler make of it all? For the most part, he was struck by how different primate communication systems — and those of the higher primates especially — were from the bird systems he knew so well. On his sifting of the data, three major contrasts emerged. One was that, where the vocal or visual signals of birds tended to be complete unto themselves, the signals of monkeys and apes tended to be composite. Interacting at close range, primates seemed to express themselves with face and body and voice all at once. By and large, a signal in one sensory modality could thus be understood only in the communicative round.27 A second, related difference was that, where bird vocalizations were often highly discrete in structure, and could be quite melodious, primate vocalizations tended to fall within graded, rather unlovely continua. Graded signalling made life much more difficult for the investigator, for it was all too easy to identify as two or more signals what were in reality varieties of a single signal. Rowell, indeed, had reported tumbling into just this trap in her rhesus studies, realizing only gradually that what she had picked out as nine separate sounds in fact made up a unitary system.28 The same properties that put human students of primate communication at a disadvantage were, however, advantageous to the primates themselves, according to Marler. Composite, continuous, close-range signals had much richer potential than their opposites, in his view, to communicate ‘information about the slightest changes in the nature and intensity of moods in the signalling animal’.29Furthermore, he argued, the high proportion of such signals among some primates made sense on selectional grounds. Rhesus monkeys, baboons, gorillas and chimpanzees rarely cohabited with creatures whose signals might become confused with their own, so evolutionary pressure for signal distinctiveness tended to be low. At the same time, these animals had such complex social lives, soaking up so much communicative energy, that there was a premium on the use of signals capable of conveying graded information—hence the abundance of graded, composite signals.30

The near-monopoly of social matters among signalling primates marked the third major contrast that Marler found with birds. Where birds in their vocalizations could sometimes convey fairly precise information about their environments, primates seemed to converse about their environments hardly at all; and what they did say was unimpressively vague:

By far the greatest part of the whole system of communication seems to be devoted to the organization of social behaviour of the group, to dominance and subordination, the maintenance of peace and cohesion of the group, reproduction, and care for the young. Inter-individual relationships are complex enough in monkeys and apes to require a communication system of this high order of complexity. But there is little application of the communication system to events outside the group, beyond the existence of signals signifying potential danger. There are no calls associated with, say, water or food, such as are known in some birds … . Environmental information, present or past, figures very little in the communication systems of these animals, and a major revolution in information content is still required before the development of a variety of signals signifying certain objects in the environment and a system of grammar to discourse about them can be visualized.31

Musing, in closing, on this designative deficiency among the non-human primates, Marler ranged widely over the ontogeny and phylogeny of language. He noted that human children use graded vocalizations before learning the local, discretely-organized language. All around us, in other words, graded communication systems mature into discrete ones—and in primates no less. What, asked Marler, could have motivated a parallel change in the past: the shift in a single hominid lineage from a graded, genetically inherited communication system to a discrete, culturally inherited one? Natural selection was the obvious short answer. To go further, one needed to know under what conditions, precisely, it had become advantageous to acquire language through vocal learning. The answer to that question, Marler concluded, with a Washburnian flourish, could be reached only with further study of communicating primates in nature.32

3. The primatological field studies of Marler and Struhsaker

Unprecedented in quantity and quality, the new field data, in Marler’s view, nevertheless left much to be desired. The fieldworkers’ main interest was social life; and as Jane Lancaster, a Primate Project contributor, later noted, they had tended to treat communication as ‘a means of entrance into the workings of the social system … not an object of study in itself’.33 At the start of his survey, Marler sketched what he considered a better way. Investigations should proceed, he advised, in two steps: first, the assembly of a trustworthy inventory of signals and the responses these typically evoke; second, the testing of generalizations with field experiments, of the sort J. Bruce Falls was pioneering with birds.34 In his most recent paper, Falls had shown that playback of artificially modified recordings of song — altering pitch, length, sequencing of elements and so on — could be used to identify the song properties clueing birds to whether a songster belongs to their species.35 Falls’s new method provided a brilliant riposte to the complaint that bio-acoustic studies ignored the question of what animals attend to in their own sounds. Synthesis and playback of birdsongs were more easily accomplished, Marler conceded, than the same with primate vocalizations, given the often graded and composite character of the latter. Nevertheless, field experimentation would, in his estimate, be a major task for the primate communication students of the future.36

How hard it was even to reach that stage was something Marler would shortly understand more vividly. In 1964–1965 — the year student unrest boiled over in Berkeley — he took sabbatical leave in Uganda in order to study the vocalizations of its richly varied apes and, especially, monkeys. Nothing in his field experiences with birds in Europe and America had prepared him for the trials of studying forest monkeys in Africa. They proved very wary of humans, so that observation and recording often had to take place from the road. Worse, they lived high up in the dense canopy, making them hard to see and even harder to track. In the end, Marler concentrated his efforts on the vocal communication of two kinds of colobus and two kinds of guenon. He recalls doing a few playbacks; none, however, made it into the published reports, which stuck instead to description, spectrographic analysis and the probing of structure–function relationships along familiar lines.37 A couple of years later, the Marlers returned to Africa, for a summer with the chimpanzees at Gombe Stream in Tanzania (as it now was) as the guests of Jane van Lawick-Goodall (as she now was). Over the years, Goodall had developed a system of baiting — little cement cubicles loaded with bananas and topped with lids opened from a distance — which brought lots of chimps right up to the camp. For six weeks, as Goodall’s husband, the acclaimed wildlife photographer Baron Hugo van Lawick, filmed the chimps in full expressive flight, Marler recorded their vocalizations and other sounds on tape, while Goodall or one of her assistants described what was observed into a dictaphone. ‘The vocalization study is going EXCELLENTLY’, Goodall wrote to Leakey that August. Besides the film — a unique, and uniquely useful, document — the summer project yielded the first evidence that individual primates have individualized calls. There were not, however, experimental playbacks, since Goodall forbade these, fearful that the wrong recording played to the wrong animal might disturb the social structure catastrophically.38

By now, Marler was professor at the new Institute for Research in Animal Behavior, a joint venture of the New York Zoological Society — sponsors of the Bronx Zoo — and the Rockefeller University, also in New York City. A kind of research Mecca for the sciences, the Rockefeller gave its professors freedom from undergraduate teaching and also what amounted to personal departments. Working alongside Marler as junior faculty were two recent PhDs from his Berkeley laboratory, representing his two main research interests: Fernando Nottebohm and Thomas Struhsaker.39 Although Nottebohm dabbled in field research on monkey vocalizations, his forte was, and would remain, laboratory studies of birdsong.40Struhsaker, however, had become a bona-fide field primatologist, indeed one of the few from Berkeley who had arrived via zoology rather than anthropology. Beginning in late spring 1963, after six months’ searching for the right site and the right subjects, he had spent around a year at the Masai–Amboseli Game Reserve in Kenya, studying the behaviour and ecology of another guenon, the vervet monkey (Cercopithecus aethiops). The circumstances proved ideal. With prodigiously varied fauna and Mount Kilimanjaro as a backdrop, Amboseli was picture book Africa—just the sort of setting that had attracted Struhsaker to primate research (his first published paper was on flight in bats). The site also offered crucial practical advantages, notably its short grass, making for easy tracking and observation of the often ground-dwelling monkeys. Just as importantly, the vervets turned out to tolerate human company well. There was even scientific expertise on hand, in the form of Stuart Altmann and his wife Jeanne, a mathematician. They had been in search of a field site when they met Struhsaker in Nairobi, about a month after he had settled at Amboseli. The rapport was instant. Struhsaker invited the Altmanns to join him at Amboseli, which they promptly did, soon launching what became a long-term project on baboon social behaviour. For the next year, the three dined and discussed together nearly every evening. Effectively Stuart Altmann became Struhsaker’s field supervisor; and when the time came to write up the vervet data, he did it back at Altmann’s academic base, the University of Alberta.41

Fig. 2. Thomas Struhsaker with vervets in Amboseli, Kenya, ca. 1963. Photograph taken by Stuart Altmann. From the private collection of T. T. Struhsaker, reproduced by permission of Dr Struhsaker.

Vervets, Struhsaker found, do a lot of gesturing and vocalizing. Given the volume and variety of these activities, a serious interest in vervet communicative behaviour was always on the cards, even without Marler and Altmann as teachers. As part of Struhsaker’s comprehensive study of vervet behaviour and ecology, he indeed carried out a detailed observational and spectrographic survey of the vervets’ vocalizations. In December 1964, at a conference Altmann organized in Montreal on primate social communication, Strushaker presented his findings. On his count, there were twenty-one different ‘stimulus situations’ evoking vocalizations in vervets, and upwards of thirty-six different vocalizations. The phrase ‘stimulus situation’ had been Carpenter’s; and the centerpiece of Struhsaker’s paper was a table clearly modeled on Carpenter’s vocalization tables. Stretching over four pages in its published form, the vervet version listed, for each stimulus situation, the sounds elicited, the age and sex of the animals making the sounds, the probable ‘message’ communicated (where Carpenter’s tables had ‘probable function’), and the responses observed in surrounding vervets and other animals. 42 Situations 14 through 19 dealt with non-vervet threats in the environment, including predators.

In light of the later fame of vervet alarm calls for python, eagle and leopard, a couple of points are worth noting. First, as the numbering scheme might suggest, the alarm calls were given no special prominence in the paper. There was no flagging up of the alarming situations as somehow especially important, for questions of animal semantics or anything else.43 Second, with the exception of what Struhsaker called the ‘snake chutter’, the famous trio of predator calls are not easy to locate in the six alarming situations.44 Two of the situations involve eagles—numbers 18, ‘initial perception of major avian predator’, eliciting a ’Rraup’, and 19, ‘proximity of major predator (mammalian and avian)’, eliciting both a ‘threat-alarm-bark’ and a ‘chirp’. Among the occasionally proximate mammalian predators the vervets met were leopards. Like eagles, leopards elicited ‘chirps’, as did cheetahs, serval cats and lions. Furthermore, according to Struhsaker, vervets ‘Uh!’-ed at spotted hyenas and Masai tribesmen, ‘Nyow!’-ed at baboons and bush cats, and chuttered in non-snake-chuttery ways at human observers like himself. Moreover, the ‘chirp’ and other vocalizations functioned, in his view, not just to alarm the preyed-upon, but to threaten the predators.45

It was Marler, not Struhsaker, who picked out just three alarm calls from this rich deposit and described them as intriguingly semantic, representational, and symbolic:

Field study of the vocalizations of the vervet monkey reveals at least six sounds that seem to be provoked by the presence of predators [citing Struhsaker]. The call given correlates with the identity of the predator, so that a snake elicits a different call than does an eagle or a leopard. The behavior of the predator is also significant. Flying and perched eagles elicit different calls, for example. There is some evidence that the signals represent different environmental situations to respondents. The initial responses that they elicit are in some cases different. The ‘snake chutter’ of the vervets elicits approach and examination of the snake from a distance. When they hear a ‘chirp’ call, given in response to a leopard, the monkeys run to trees and climb to the topmost branches. In response to a ‘rraup’ call, given for an eagle, they run from open areas into thickets and descend from treetops. Direct perception of these three types of predator may elicit the same three patterns of response. Although the relationships are not specific enough for us to think of the three signals as names for the three types of predator mentioned, in principle, these signals begin to approach the phenomenon of object naming.46

The occasion was a review paper on ‘Animal communication signals’, published in Science in August 1967. It had started life two years earlier as Marler’s contribution to a Wenner–Gren symposium on animal communication.47 His brief was to tour the research horizon from an ethologist’s perspective. Fresh from his year in the Ugandan forests, and mindful, no doubt, of the psychologists and linguists present, Marler announced at the start that any non-anthropocentric study of animal communication had to deal in animal responses to animal signalling. The rest of the paper advised attention to such matters as how responses orient animals spatially; the different kinds of behaviour stimulated by signalling; correlations between signal variation and response variation; the way the different parts of a signal function (and how useful Fallsian experimental playbacks with artificial signals can be here); the range of sensory modalities in play in animal communication; and the range of situations stimulating particular signals. It was under this last heading that Marler introduced the vervet alarm calls to the wider world.In a key 1961 paper, ‘The logical analysis of animal communication’, he had dismissed semantics as irrelevant for understanding animal communication.48Revisiting the subject now, he showed plenty of the old scepticism. He noted with disapproval, for instance, the way that semantic considerations forced distinctions with little biological or logical warrant, as between alarm signals and sexual signals. Both kinds of signal, he pointed out, serve the Darwinian ends of survival and reproduction. Sexual signals, moreover, could be construed as representing or symbolizing inner states. Yet only alarm calls, supposedly representing or symbolizing things in the external world, receive the honorific ‘semantic’. Restricting attention to such apparently externally-referring signals, there were, in Marler’s view, a number of interpretative difficulties. Consider that an animal responding to a signal enters what is arguably a new stimulus situation. Is the behaviour ensuing a response to this new situation, or to the former, signal-evoking one? Marler illustrated by way of the Cornell linguist Charles Hockett’s assertion that a call eliciting feeding behaviour in gibbons symbolizes food. Marler distinguished two possibilities. It might be that, on hearing the call, a gibbon begins to salivate, and to respond in other ways preparatory to feeding. But it might also be that the call merely stimulates a hungry animal to approach, with feeding behaviour arising only after the respondent sees or smells food. ‘In the latter case’, wrote Marler, ‘it is less easy to decide whether the food call really represents food to the respondent, for other stimuli will also elicit approach even in a hungry animal’. Even the impressively semantic vervet calls had their share of complications. He noted that the calls in themselves do not always suffice to determine the pattern of response. Vervets in a thicket when they hear the call for ‘eagle’ tend to crouch down, whereas vervets that hear the call when out in the open tend to run for the nearest thicket. ‘In such cases’, he concluded, ‘the role of the context must be taken into account in a consideration of whether a semantic system is involved’.49

In what sense, then, were chutterrraup and chirp the vervets’ names for, respectively, python, eagle and leopard? Marler offered no clear verdict, and his lingering discomfort with the question was manifest. Why raise it, then? Undoubtedly one reason was sheer surprise at what Struhsaker had found. As we have seen, Marler came away from the Primate Project persuaded that the vocalizations of monkeys and apes convey hardly any information about their environments. The vervet calls seemed to fill precisely the gap Marler had described; and on that account alone, they must have struck him as deserving wider notice. Less obvious, however, are his reasons for publicizing them under the banner of semanticity, symbolization, object naming—all concepts he had previously kept at arm’s length. We turn now to the other developments behind Marler’s semantic turn.

4. Neuroscience, emotions and vocalizing primates

Like all good mysteries, this one has a red herring: the work of W. John Smith. Six years younger than Marler, Smith was a Canadian-born, Harvard-trained zoologist with interests closely paralleling Marler’s own. In a 1963 paper on ‘Vocal communication of information in birds’, and a 1965 paper on ‘Message, meaning, and context in ethology’, Smith had argued that animal signals in themselves convey only general messages, which acquire specific meanings in context.50 With Marler present, Smith had presented his case at the 1965 symposium on animal communication.51 When Marler wrote of the need to take context into account in deciding the semanticity of the vervet predator calls, he cited Smith. More than ten years later, when Dorothy Cheney and Robert Seyfarth returned from doing playbacks with vervets in Amboseli, they presented their results, to funding agencies and then the world at large, as overturning a Smithean consensus.52 But Smith’s thesis could hardly have provoked Marler all that much. For one thing, Smith had credited Marler as a pioneer in bringing signal context to the ethological fore (and thanked him for helpful discussion).53 For another, as we have seen, Marler regarded primate signals as overwhelmingly composite in nature. At most, Smith had promoted to a rule what Marler was prepared to accept as an often-correct generalization, especially as applied to apes and monkeys. Indeed, what Marler conceded about a vervet’s location determining its response to a ‘rraup’ call was all Smith required for the call’s meaning to count as contextual. The vervet calls were thus Smithean whether or not — to pose the most famous contextual question later settled by the playback experiments — the calls evoke predator-specific responses (looking up at the eagle call, down at the python call, and so on), or merely prompt general looking around, with adaptive responses following the sighting of a predator.54

The one clear disagreement between Marler and Smith turned on a point neither original to Smith nor, at first, put with much vehemence. In Smith’s view, animal signals in themselves referred not to the world, but to the inner state of the signaller—to, as he put it in his 1965 article, ‘some aspect(s) of the state of the central nervous system (CNS) of that individual’, including a state of readiness to act in a certain way. He did not present this argument as especially revisionist or sceptical or placing a barrier between animals and environmentally-referring humans. On the contrary, he gave the impression that his argument was orthodox communication theory and applied equally to humans and non-humans.55 He did not comment in print on the vervet alarm calls until his 1977 synthetic book The behavior of communicating: An ethological approach. There, true to form, he presented the calls as conveying escape messages, albeit unusually narrow ones. ‘Rraup’ referred, for Smith, not to eagles, but to a readiness to escape into the thickets; ‘chirp’ referred not to leopards, but to a readiness to stay in cover or go find some up a tree. Looking next at birds, he argued that claims about alarm calls which ‘effectively name a single class of predator’ have always evaporated on closer inspection, as lots of different things have invariably been found to elicit the call. ‘The conservative interpretation at present’, Smith concluded, ‘is that alarm displays have no referents external to the communicators’.56 Here a gauntlet was laid down.57But even in 1977, the theme of environmental reference in animal signalling occupied a small part of Smith’s attention. Back in 1965, it barely registered at all.

The anthropologists, however, were another matter, as Marler soon learned. His symposium talk on vervet calls and other animal signals took place in June 1965. A few months later, he was again in Austria, again for a Wenner–Gren symposium, this time on primate social behaviour. Another participant was one of Washburn’s Ph.D. students at Berkeley, Jane Lancaster. As noted above, she had been part of the Primate Project. Now she presented an interim report from her dissertation research on — to use the title of her published paper (also of her dissertation) — ‘primate communication systems and the emergence of human language’. Her thesis did not comport well at all with the possibility that vervets refer in their calls to pythons, eagles and leopards. For Lancaster, humans alone had the ability to name objects in the environment. Indeed, she saw the evolution of this ability as the true beginning of language, since language, in her view, was first and foremost a system of names, organized in traditional ways. The communication systems of the other primate species, she wrote, ‘have little relationship with human language, but much with the ways human beings express emotions through gesture, facial expression, and tone of voice’. In support, she appealed to three classes of evidence: communication in monkeys and apes, so overwhelmingly social in character; the acquisition of language in children, whose first words, she reported, tend to name objects, and who do not express emotions with words until quite late (not because children recapitulate the evolutionary past, but because they develop in ways shaped by that past); and third, anatomical and physiological studies of the brains of monkeys, apes and humans. According to Lancaster, object naming had been shown to have a separate anatomical basis from emotional expression—and a uniquely human one.58

She took her neuroscience straight from the Boston neurologist Norman Geschwind’s long, two-part review of the scientific literature, ‘Disconnexion syndromes in animals and man’, published that same year. The disconnections concerned were between different parts of the brain, as in aphasia, Geschwind’s specialty. In outline, the Geschwindian picture — a composite of old and new facts, principles, conjectures, theories — was as follows. In monkey, ape and human brains, different regions of the cortex have different functions. Some of the regions receive sensations; one region receives visual sensations, another auditory sensations, and so on. Each sensory region connects with an adjacent ‘association area’, which in turn connects with other parts of the brain, enabling associations between sensations and movements, or between sensations and emotions. In non-human animals especially, there are abundant connections with the limbic system, lying below the cortex, and widely associated with the emotions. In no animals are there abundant connections between the sensory association areas themselves; so, by that route, there can be only weak associations between one kind of sensation and another kind. In human brains, however, there was a further association area, found in the inferior parietal lobule, in a region containing a structure called the ‘angular gyrus’, barely present or absent from monkey and ape brains. Connected to each of the different sensory association areas, this structure enabled humans easily to form ‘intermodal’ or ‘cross-modal’ sensory associations, such as those required for naming (as when the sight of a python comes to be associated with the sound ‘python’). On this picture, the limbic system and the angular gyrus make for sharp contrasts. The limbic system is animal, ancient, impassioned, inflexible, submerged. The angular gyrus is human, recent, dispassionate, flexible, surface. Language emerged with the evolution of the angular gyrus and the intermodal associations it made possible.59

Fig. 3. The left hemisphere of the human brain according to Geschwind. The parts identified are: the frontal, temporal, parietal and occipital lobes; the primary visual, auditory and somesthetic regions of the cortex; and the angular gyrus (diagram from Geschwind, 1964, p. 159; used with the permission of Georgetown University Press).

The consequences for interpreting vervet calls were straightforward. However much they look like names for pythons, eagles and leopards, this cannot be the case, since vervets lack the cerebral equipment needed for such precise environmental reference. As Lancaster put it:

The rare examples of non-human primates’ communicating some information about the environment in their high-intensity alarm calls are not relevant to the evolution of a system of object-naming. The similarity between these abilities is only a minor, superficial one, and the underlying mechanisms are entirely different … [I]t is apparent that the ability of the vervet monkey to refer to the environment, a design feature which is superficially similar to man’s object-naming but which is in fact based on quite different underlying neurological mechanisms, cannot be suggested as representing a possible step toward language. Object-naming is unique to man because the anatomical basis of the ability is also unique to man.60

She did not here cast doubt on whether different kinds of predator evoke acoustically distinct calls from vervets, or whether those calls elicit adaptive responses in surrounding vervets. She took all that to be true. What she doubted was that the vervets were representative. For her, they were a special case, intelligible but not to be generalized from. They were, on her presentation, rather like apes that had learned a few words after extensive training by humans. What was striking about apes, Lancaster argued, was not that a little language could be drummed into them, but that it could be done only with extraordinary difficulty. Similarly, the vervet case should teach us not that animals can name objects in their environments, but that, when species have come under sufficient pressure from a number of quite distinct dangers, natural selection may have differentiated their alarm calls in ways that mimic environmental reference. But there was all the difference between as-if naming and the real thing.61On the face of it, this slamming down of a language barrier between humans and animals was the last thing to be expected from the new, baboons-and-bushmen physical anthropology. But in fact Lancaster’s argument pressed a couple of Washburnian buttons. It fit well, as she showed, with one of Washburn’s major themes, the importance of tool use in the making of modern humans. Like tool use, object naming would have greatly expanded the power of our hominid forebears to exploit the environment in ways that gave them a selective advantage. Even better, she had reached this conclusion in fine Washburnian fashion, combining the results of field studies of primate behaviour and classy research in functional anatomy.62 And the research came no classier than Geschwind’s, whose 1965 paper was bringing an end to decades of disenchantment with localization studies. There had, to be sure, been much progress since the days of Ferrier — the sensory association areas were identified in the 1890s and 1900s, the limbic system in the 1930s and 1940s — but it was felt that no useful general principles linking brain and behaviour had emerged.63 This was what Geschwind’s paper supplied. It was as much backward-looking as forward-looking—an attempt to restore a classical but neglected neurological heritage. To anyone familiar with the 1860s, indeed, its conclusions about language have a strongly déjà vu quality: the angular gyrus as the hippocampus minor come again; the distinction between neocortical object-naming and limbic emotional expression an updating of the Jacksonian doctrine distinguishing higher, propositional, humans-only language from lower, emotional, animal expression.64

There was more to the new comparative neuroscience of language than old ideas, of course. Some innovative experimental work was on view at the Montreal meeting where Struhsaker talked of vervet alarm calls. Bryan Robinson, from the Yerkes Primate Regional Center in Atlanta, announced the development of a ‘tele-stimulation’ unit that would enable a new mixing of lab and field: while monkeys with electrodes embedded in their brains, and solar-powered stimulation and transmission units on their heads, interacted with their social group, a researcher hidden out of sight would, by remote control, selectively stimulate different parts of different individuals’ brains.65 Detlev Ploog, at the Max Planck Institute of Psychiatry in Munich, reported work moving in a similar direction, integrating studies of brains, vocal and other behaviours, and social dynamics—in his case, with laboratory colonies of squirrel monkeys.66 But it was Geschwind’s synthesis that gave such investigations moment, showing how they bore on the big questions about mind, brain, language and evolution.67 When Bill Thorpe, at a November 1965 meeting on language and the brain, argued for parallels between bird song and human language, he got a Geschwindian rejoinder from Horace Magoun, a pioneering student of the cerebral bases of emotional expression. According to Magoun, animal experiments and anatomical studies all ‘seem to oppose the view that man’s capacity for speech evolved from the abilities for emotional vocalization present in lower animals’:

On the contrary, man’s communication by symbols, both vocal and written, appears to represent an entirely novel functional increment related to acquisition of associational cortex … One can conclude that there are two unrelated central neural mechanisms for vocal expression in vertebrates: one for non-verbal affective communication, widely present in the animal brain stem, and a second for verbal communication, present only in the lateral neocortex of the brain of man.68

Two years later, Robinson, citing Geschwind on the human brain — and Marler on how primates vocalize mainly ‘in situations bearing some emotional valence’ — echoed the point:

[I]t appears that human speech and primate vocalization depend on two different neural systems. The one is neocortical; the other, limbic. This suggests that human speech did not develop ‘out of’ primate vocalization, but arose from new tissue which permitted it the necessary detachment from immediate, emotional situations. The neurological evidence suggests that human language arose in parallel with primate vocalization, surpassed it, and relegated it to a subordinate role.69

All of this reached a wider audience at the end of the decade with the publication in 1970 of a popular paperback, New horizons in linguistics, where the psycholinguist John Marshall, in a chapter on the biology of communication, explained why the vervet alarm calls were not names, but merely expressions of emotions — vocal components of a wider, fixed physiological response to danger — before introducing the Geschwindian picture:

Even the alarm calls of the vervet monkey which seem, superficially, to be ‘naming’ the type of predator are more plausibly regarded as expressing no more than the relative intensity of the fearful and aggressive emotions aroused by the various predators … An emotive interpretation of primate calls is also suggested by the observation that vocalizations can be obtained by electrically stimulating sites in the limbic system of monkeys. The brain area is known to be crucially implicated in the mediation of emotional behaviour … In order for vervet monkey calls to qualify as ‘names’, a particular call would have to be associated with a particular object irrespective of the emotion aroused by the perception of that object. In the human case, the expression ‘leopard’ is appropriately used in referring to members of the class of leopards. Whether one is attracted or repelled by leopards is irrelevant.70

5. Conclusions

The discovery of the vervet predator calls brought otherwise submerged disciplinary allegiances to the surface. For all that field studies of primate behaviour benefited from Washburn’s patronage, and for all that they now stand for the Washburnian synthesis as a whole, field primatology was in fact a lesser partner in that synthesis, well below functional anatomy in stature. In a 1973 paper, ‘The promise of primatology’, Washburn made plain his sense of the scientific pecking order, in the course of a discussion that touched on the vervet controversy. How instructive it was, he wrote, that descriptive studies of human language and animal signalling over several decades had failed to do what experimental studies of the brain had done in just four years: disclose the radically different nature of the two systems. Between 1966 and 1970, Ploog and Robinson had shown that limbically-implanted electrodes elicit normal vocalizations in monkeys (but do not elicit words in humans); Ronald Myers had demonstrated that monkeys with much of their cerebral cortex removed vocalize normally (again in sharp contrast with humans); and Michael Gazzaniga and others engaged in the surgical splitting of the corpus callosum for therapeutic purposes in humans had found that the sound code of language is not globally distributed, but localized in evolutionarily recent areas of the cortex. (‘The cortex on the dominant side of the brain controls the code’, Washburn summarized, ‘and with lesions centering in the area of the angular gyrus, the larger the lesion, the greater the linguistic deficit’.) Were field primatologists to take the new neuroscience to heart, wrote Washburn, they would refrain from attributing distinctive ‘meanings’ to different vocalizations in abundance, as Struhsaker had done in his vervet study. Instead the sounds would be assimilated to a small number of multimodal signals involved in primitive behaviours: feeding, mating, responding to danger and so on.71

Marler, unsurprisingly, saw matters rather differently. The idea that there were two utterly different kinds of vocalization, the human kind (voluntary and rational) and the animal kind (involuntary and emotional), looked from his point of view like just the sort of subjective, anthropocentric distinction that ethology was founded to repudiate.72 Nor did the distinction’s coming with anatomical credentials make it any more impressive. There was no tradition among ethologists of automatic deference to the conclusions of anatomists. On the contrary, ethologists had long regarded the comparative study of patterns of behaviour as a means of supplementing and even correcting the evolutionary genealogies arrived at through comparative anatomy. Furthermore, Marler was not at all accustomed to thinking of studies of natural behaviour the way Washburn did, as generally indifferent to function and innocent of experimental method. Marler’s best known contributions as an ethologist had been heavily functional and experimental. Indeed, when, in the early and mid-1970s, he began exploring the interpretative issues surrounding the vervet predator calls and their status as affective or symbolic, it was precisely on Darwinian, functional grounds that he cast doubt on the purely affective reading.73

To Marler, it was a mistake to suppose that vocalizations had to be either purely affective or purely symbolic. Rather, he argued, one should view the pure cases as endpoints on a continuum, with natural selection determining the precise mix of affective and symbolic information in a species’ signals, according to its distinctive survival needs. If partly symbolic signals are adaptive for vervets under natural conditions, then we should expect to find vervets using such signals.74 We should also expect to be able to tell one way or the other without having to crack open vervet skulls. The experimental playbacks of Marler, Cheney and Seyfarth indeed translated the affect-vs.-symbol problem into an ethologically tractable question, or rather several such questions.75 The chief question was this: when the alarm calls are played back in the absence of predators, do the calls elicit predator-specific adaptive responses? If so, then, it was held, the calls themselves must convey information about the exact nature of an environmental threat. Therefore the calls cannot be purely affective. A second, more involved question was: when python and eagle calls are played back at the same length and volume as the typically longer, louder leopard calls, do the vervets respond as if to news about pythons and eagles? If not, then, it was held, the calls can be regarded as purely affective, since vervets must be attending only to call features that scale with level of arousal (the more frightened you are, the louder and longer you call). Otherwise, the calls can be regarded as at least partly symbolic, since, irrespective of the caller’s state of arousal, acoustic structure cues adaptive behaviour.76

While, for Marler, Darwinian reasoning legitimated an experimental detour around the primate brain, for Washburn, it did nothing of the kind. He regarded the uniqueness of human language as an anatomical fact which an adaptationist, Darwinian origins theory had to accommodate. ‘Obviously, in the evolutionary process’, he wrote in ‘The promise of primatology’, ‘the human system must have been derived from the non-human kind, but in spite of many efforts, there is no direct evidence for intermediate stages’. He argued that the human sound code first took shape thanks to chance variations in the brains of now-extinct hominids, who were thus endowed with unprecedented abilities for communicating knowledge. The adaptive advantages that flowed from this anatomical innovation rebounded on brain size and organization, articulatory powers and the sound code itself. ‘This point of view’, Washburn continued, ‘stresses the factors of the brain, the phonetic code, the adaptive value of the system, and the rapid communication of knowledge’. Nothing could be more retrograde, to Washburn, than a behavioural science that turned its back on the experimental–anatomical results and started treating the brain once again as, in his phrase, a ‘black box’.77 When Cheney and Seyfarth gave talks on their Amboseli research at Berkeley in 1979, they faced a visibly unimpressed Washburn.78

Just as striking as this disciplinary fissure was the failure of Darwinism to bridge it. Darwinian theory has long been a rallying point for those who dream of a unified science.79 The philosopher Daniel Dennett has described natural selection as a ‘universal acid’, capable of dissolving all the barriers to a coherent, scientific view of nature and the place of humans within it.80 (Not coincidentally, Dennett has been an influential advocate of the vervet alarm playbacks as exemplary of Darwinian cognitive science.)81 Related to the ideal of unity under Darwinism is the suspicion that a denial of continuity between humans and non-human animals must be motivated by a reluctance to go all the way with Darwin. In Cheney and Seyfarth’s own version of their history — the one they shared with me when I interviewed them recently — Washburn and his students came across as Darwinians who, alas, fell at the final hurdle of language. They had sought behavioural continuities across the board; but when it came to language, the urge to preserve human uniqueness overwhelmed scientific sense. Presented thus, the vervet playback experiments look like the completion of the Washburnian synthesis, filling the gap that Washburn himself could not bear to fill, though his own program demanded it. But Washburn was no crypto-creationist. His Darwinism was as thoroughgoing as Marler’s. As Washburn listened to the results of the Amboseli field experiments, it was not the loss to Darwinism of human linguistic uniqueness that bothered him, but the fact that it was being proclaimed without a single new neuroscientific fact to gainsay the accumulated evidence in support of uniqueness.82 From Washburn’s point of view, the vervet alarm playbacks took the cause of a Darwinian primatology backwards, not forwards. The case of the vervet predator calls and the Washburnian synthesis thus stands as an instance of disunity under Darwinism, when the detergent powers of natural selection theory failed to scour away the distinctive commitments of independent disciplines. Whatever unity the behavioural sciences eventually achieve, it will not be Darwinism alone that deserves the credit.


This paper represents a substantial expansion of part of the paper I presented at the workshop on ‘Fielding the Question: Primatological Research in Historical Perspective’, held at the University of Leeds, December 2003. I am grateful for helpful discussion to all who participated on that occasion, but especially to my fellow presenters Jonathan Burt, Mandy Rees and Marion Thomas, to our commentator, Graeme Gooday, and to Jonathan Hodge. The research and writing herein were completed during a sabbatical year funded partly by the School of Philosophy, University of Leeds, and partly by the Leverhulme Trust. For their generosity in interview and correspondence, and for permission to draw on unpublished sources, I am also grateful to Dorothy Cheney, Irven DeVore, Marc Hauser, Phyllis Lee, Peter Marler, Robert Seyfarth and Thomas Struhsaker.


1 See for example Marshall, 1970, Seyfarth et al., 1980a, Bickerton, 1990, Hauser, 2000.
2 For a summary of the vervet predator calls as acoustic and behavioural phenomena, see Cheney and Seyfarth (1990), esp. Chs. 4–6. Examples can be heard online at the website of Marc Hauser’s Harvard laboratory: (accessed 20 January 2005).
3 On Von Frisch’s research and the controversies surrounding it, see Munz (2005).
4 For a recent examination of the vervet calls as clues to language’s animal origins, see Anderson (2004).
5 Struhsaker (1967a).
6 Seyfarth et al., 1980a, Seyfarth et al., 1980b.
7 On the long-run history of the primate playback experiment, emphasizing a heyday around 1890 associated with the American naturalist Richard Garner and a second heyday around 1980 associated with Cheney, Seyfarth and Marler: see Radick (2005).
8 The phrase is from Seyfarth, Cheney, & Marler (1980a), p. 803.
9 The vervet studies as a whole are summarized in Cheney & Seyfarth (1990). For the use of ‘habituation–dishabituation’ playback experiments to test whether vervets attend to meaning or merely to acoustic structure, see pp. 151–158.
10 Washburn (1951), quotation on p. 155. A table summarizing Washburn’s views on the differences between the two anthropologies can be found in Washburn (1952), p. 716. On the 1950 symposium, see Bowler (1986), pp. 240–244. For ‘racial quality’ as discussed in the most popular work of Washburn’s Harvard teacher, Earnest Hooton, see Hooton (1947), pp. 658–661. On Washburn’s life and work, see Washburn, 1983, Haraway, 1989; and esp. Strum, Lindburg, & Hamburg (1999), which includes reprintings of several key papers, testimonials from students and colleagues, and a bibliography of his writings.
11 Washburn (1952), p. 719.
12 Ibid., pp. 723–724. In later life, Washburn credited his conversion from anthropometrical anthropology to two sets of influences: his mentors in medical (hence functional) anatomy, W. T. Dempster at Michigan and S. R. Detwiler at Columbia; and his reading of three books, Percy Bridgman’s The logic of modern physics (1927), which ‘gave clear guides to the relations of technique to analysis’, and thus revealed biometry as designed to describe not man, but bones; C. K. Odgen and I. A. Richards’ The meaning of meaning (1923), which ‘helped one to see the nature of words and avoid at least some of the common errors’; and Bronislaw Malinowski’s Argonauts of the Western Pacific(1922), which ‘read like a great novel’, and so taught by example that ‘[h]uman behavior cannot be reduced to trait lists’. Washburn (1983), pp. 3, 6–7, quotations on p. 6.
13 Ibid., pp. 2–6. Washburn noted (p. 4) that Carpenter had set up camp a few miles from where Schultz was based. The separation was more than just geographic. According to DeVore, Washburn tried and failed repeatedly to interest the more senior men in each others’ work, with Schultz complaining to him at one point: ‘Sherry, that man calls a tree limb a “behavioural platform”. I have no patience with him!’ Interview with I. DeVore, September 2002. En route to Asia, Washburn worked as a lab assistant to the era’s other well-know student of primate behaviour, Solly Zuckerman. On Zuckerman, see Ribnick (1982), p. 59, and Jonathan Burt’s paper in this issue.
14 See Washburn (1983), pp. 16–17.
15 Interview with I. DeVore, September 2002. See also Haraway (1989), pp. 219–220.
16 Washburn & DeVore (1961), p. 102.
17 On Marler’s life and work, see Marler, 1989, Radick, 2005. On the history of ethology, see Burkhardt (2005).
18 On Marler’s interactions and collaborations with Washburn, see Marler (1989), pp. 328, 330. On the primate colony at the station, see Washburn, 1983, Ribnick, 1982. For many ethologists, including Marler, Frank Beach represented the acceptable face of comparative psychology. For Beach’s classic critique of the anthropocentrism of his own discipline, see Beach (1950). Beach, Washburn, Hinde and the ethologist Niko Tinbergen had been among the participants in important meetings in the mid-Fifties on behaviour and evolution; see Roe & Simpson (1958).
19 Marler (1963).
20 In a 1966 letter of recommendation for DeVore, Marler wrote that DeVore ‘is more responsible than anyone else for my present enthusiasm for primate behavior’. Letter from Peter Marler to Douglas Oliver, 27 January 1966, in the Papers of Peter Marler, University of California Davis.
21 On the invitation and its consequences, see Marler (1989), pp. 330–331.
22 DeVore (1965), p. ix.
23 Observers then and later have noted the stimulus that increasing use of primates in postwar medical research gave to field studies of these animals. ‘The combined efforts of anthropologists, psychologists, and zoologists, which stems in part from demands made by the health-oriented sciences for better information, has instigated a wide range of field studies of primates’, wrote Marler in his 1963 review of Schaller’s gorilla study (Marler, 1963, p. 1081). Funding for the Primate Project indeed came from the National Institutes of Health (DeVore, 1965, p. viii). On the founding in these years of seven regional primate centers in the USA by NIH and other bodies, see Dukelow (1995), and also the useful review by Weidman (1996). Medicine played a less direct role as well, through the development of antibiotics and other drugs making fieldwork in tropical climates less dangerous; see Ribnick (1982), p. 55. For historical surveys of primatological fieldwork in this and other periods, see Ribnick, 1982, Haraway, 1989; and Amanda Rees’s paper in this issue.
24 On the early days of Madingley primatology, see Hinde (2000), pp. 106–107. On Goodall and Leakey, see, e.g., Haraway, 1989, Goodall, 2001.
25 On Altmann’s early work, see Haraway, 1989, Wilson, 1994. Wilson here credits the first stirrings of sociobiology to his 1956 visit with Altmann to the rhesus colony on Cayo Santiago, near Puerto Rico. (The colony was NIH’s—see note 23 above.) Altmann’s melding of animal communication and information theory was independent of Marler’s; see Haraway (1989), p. 401 n. 11, on the ‘two streams of communications analysis’, one in the USA, the other in the UK and Europe. Indeed there were still more—a French stream, for instance (Moles, 1963), and a Russian one too (Zhinkin, 1963).
26 Marler (1965), p. 558. Schaller, Goodall, the Reynolds, DeVore & Hall and Jay all have papers in the same volume, DeVore (1965). For the relevant papers by the others, see the references in Marler’s table. On postwar Japanese anthropology, see for example Ribnick (1982), pp. 55–57. For vivid — and typically quantitative — testimony to the explosion of primatological field studies since the mid-Fifties, see the graph in Altmann (1967a), p. xi. On a study of the vocalizations of semi-captive baboons in the Soviet Union of the era, see Zhinkin (1963), pp. 150 passim.
27 Marler (1965), p. 583. DeVore stressed in interview that his impression watching baboons — and Washburn’s too — was that vocalizations are mere ‘exclamation points’ to what is expressed by face and gesture. Interview with I. DeVore, September 2002.
28 Marler (1965), pp. 558–565, discussing, on pp. 559–561, Rowell (1962).
29 Marler (1965), p. 583.
30 Ibid., pp. 565–566.
31 Ibid., p. 584. See also his remarks on p. 567 and in Marler (1963), p. 1082. Of mountain gorillas: ‘there is no evidence that certain sounds are associated with particular objects in the environment’.
32 Marler (1965), p. 584. For rather different reflections on the same set of Primate Project signalling data, see Bastian (1965).
33 Lancaster (1968), p. 441.
34 Marler (1965), pp. 545–546.
35 Falls (1963). The most famous experimental studies in this genre from this era are those of the French scientist J.-C. Bremond, credited with discovering the ‘syntactical rules’ governing the construction of the song of the European robin. See Catchpole and Slater, 1995, Terborgh, 1996.
36 Marler (1965), p. 569.
37 The colobines studied were the red colobus (Marler, 1970) and the black-and-white colobus (Marler, 1969a, Marler, 1972). The guenons — the genus including the vervets — were the red-tailed monkey and the blue monkey (Marler, 1973). On the difficult field conditions, see Marler, 1972, Marler, 1973. On the year in Uganda, see Marler (1989), pp. 331–332. The fieldwork with monkeys — and perhaps especially that reported in Marler (1969a) — also generated a kind of Primate Project postscript, surveying the roles of primate signals in increasing, maintaining or decreasing space between groups and within them; see Marler (1968). It is a strikingly classical paper, deliberately modelled on Tinbergen’s herring-gull studies (p. 423), and dealing not in ‘information’ but in ‘functions’.
38 On Marler’s time with Goodall, see Marler, 1969b, Marler and Hobbett, 1975, Marler, 1989, Goodall, 2001. To another correspondent she reported (p. 54): ‘Flint [a chimp] has a thing about Peter Marler. He runs up and hits him at the slightest provocation! Doesn’t like his beard!’ Goodall described the baiting system in Lawick-Goodall (1968), p. 317, with a photo on p. 357. I learned of the prohibition on playbacks from interviews with Marc Hauser, Phyllis Lee and Peter Marler in 2002 and 2003.
39 On the Institute in its early days, see Marler, 1989, Griffin, 1989.
40 In his 1969 Science paper on the black-and-white colobus, Marler compared spectrograms of the roar of adult male black-and-white colobus monkeys from Uganda with Nottebohm-recorded roars of adult male red howler monkeys from Trinidad. So similar were the roars in physical structure, wrote Marler, that ‘it is tempting to think of it as an example of convergent evolution in behavior’. Marler (1969a), p. 95.
41 For much of the biographical detail here, I am indebted to Thomas Struhsaker (pers. comm., Autumn 2004). On Struhsaker’s route to Amboseli, and his research there, see Marler, 1989, Struhsaker, 1967a. On the Amboseli vervets’ ecology, habits and rapid habituation to humans, see Cheney & Seyfarth (1990), pp. 19–24.
42 Struhsaker (1967a), table on pp. 314–317, with discussion on pp. 304–312. For Carpenter’s tables, from his 1934 howler monkey paper and his 1940 gibbon paper, see Carpenter (1964), pp. 76–77 and 248, respectively. For discussion, see Montgomery (2005).
43 Struhsaker himself identified the paper’s major contribution as its demonstration that — contrary to the views of Marler and Altmann — the functional contexts of calls can and should be taken into account from the beginning in devising call taxonomies; see Struhsaker (1967a), pp. 318–319, 323. Altmann, however, did make a connection between the vervet alarm calls and questions of semanticity. By now in collaboration with the linguist Charles Hockett, with whom Altmann shared interests in information-theoretic approaches to animal communication, Altmann reviewed Hockett’s list of language’s ‘design features’ and their applicability to primate communication systems. Under ‘semanticity’, Altmann wrote that alarm calls and food calls were ‘instances of semantic messages’, and that Struhsaker had ‘described a variety of alarm calls in vervet monkeys’. See Altmann (1967b), p. 338, also p. 339.
44 The primatologist Alison Jolly, who was not present but had a paper read, got the impression that Strushaker had identified five alarm calls; see Jolly (2000), p. 77.
45 The picture is even more complex than indicated here. For full details, see Struhsaker (1967a), pp. 304–312. For his conclusions about gestural communication, social structure and ecology among the vervets and other cercopithecines, see respectively Struhsaker, 1967b, Struhsaker, 1967c, Struhsaker, 1967d.
46 Marler (1967), pp. 771–772. In fact, Struhsaker reported that whether they were flying or perched, eagles evoked the ‘rraup’ call; see Struhsaker (1967a), p. 308. On his published account, it was not eagle behaviour, but the duration of an encounter with an eagle, that led to a change in the call elicited, from ‘rraup’ to ‘chirp’. Furthermore, it was only adult females and juveniles who rraup-ed on initial perception of eagles and chirped subsequently. Sub-adult and adult males gave ‘threat-alarm-barks’ in response to eagles from beginning to end. See p. 311 for Struhsaker’s summary.
47 On the symposium, see note 51 below.
48 Marler (1961), p. 299.
49 Marler (1967), pp. 771–772. Again, Struhsaker’s own report made little fuss about how the vervets’ contexts shaped their responses to ‘rraup’. He wrote merely that ‘Immediately subsequent to the Rraup call, the vervets ran or dropped into the dense thickets from the tree branches and the open, short-grass areas’. Struhsaker (1967a), p. 308.
50 Smith, 1963, Smith, 1965. For biographical background, see the contributor’s note in Sebeok (1968), p. xv.
51 Hockett was also present. On the symposium date and participants, see Sebeok (1968), p. v. For a report on the symposium, see Ramsay (1966). For Smith’s paper, see Smith (1968). The symposium organizer was Thomas Sebeok, a linguist based at Indiana University, in the process of launching himself with great energy into what he called ‘zoosemiotics’. Sebeok was to become a central figure in organizing sceptical reactions to the ape language projects of the late 1960s and 1970s. Apart from Marler’s paper, the symposium papers, and others besides, all appeared in two volumes: Sebeok, 1968, Sebeok and Ramsay, 1969.
52 See, e.g., R. M. Seyfarth and D. C. Seyfarth, Proposal to the National Science Foundation, December 1979, pp. 3–5, in the Marler Papers, U. C. Davis; Seyfarth, Cheney, & Marler (1980b), pp. 1090–1092 .
53 See Smith (1965), pp. 406 and 408.
54 Following the London communications theorist Colin Cherry (as, previously, had Marler), Smith argued ‘that meaning be identified as the response selected by the recipient from all of the responses open to it’; Smith (1965), p. 406. But what is ‘the response’? On Smith’s interpretation, a ‘rraup’ call that elicits crouching has a different meaning from a ‘rraup’ call that elicits running. It follows for the Smithean that ‘rraup’ has no determinate meaning independent of context. For the post-playback Cheney, Seyfarth and Marler, by contrast, what mattered was not the ultimate escape behaviour but the immediate response. On their interpretation, ‘rraup’ means ‘eagle’ since the call consistently evokes looking up, irrespective of the listener’s situation. Context enters only in determining a vervet’s course of action once it has been put on the alert for an eagle, thanks to a call representing ‘eagle’. See Seyfarth, Cheney, & Marler (1980b), p. 1092.
55 Smith (1965), quotation on p. 405.
56 Smith (1977), p. 181.
57 For discussion of Smith’s contextualism and its impact, see Hauser (1996), pp. 60–62.
58 Lancaster (1968), quotation on p. 446. On her dissertation, and the date of the second Wenner–Gren meeting (September 1965), see her online CV, at the website of the Department of Anthropology, University of New Mexico: (accessed 20 January 2005).
59 Lancaster, 1968, Geschwind, 1965; but see also the more schematic, language-focussed Geschwind (1964). Geschwind called the angular gyrus the ‘association area of association areas’; see Geschwind (1965), p. 641. For a lucid discussion of the Geschwindian picture and its sources, see Rosenfield (1985), pp. 49–51. For Lancaster’s debt to discussions with Geschwind, see Lancaster (1968), p. 439.
60 Lancaster (1968), pp. 453–454.
61 On the vervets, see, ibid., pp. 444–446; on ‘the difficulty with which a monkey or ape learns any rudiment of what might conceivably be called language’, see p. 453 n. 5; also p. 451. When Lancaster spoke, the most recent attempt to teach human language to an ape was that of Keith and Catherine Hayes, who, around 1950, tried — and largely failed — to teach an infant chimpanzee named Viki to speak English. See Hayes (1952). For discussion of the Hayes’ experiment in the context of similar efforts, see Candland (1993), esp. pp. 286–289.
62 On object naming, tool use and human evolution, see Lancaster (1968), p. 453 n. 5; pp. 456–457. For Washburn on the same themes, see Washburn (1959), esp. pp. 27–29, where he drew on Wilder Penfield’s experimental research on the proportions of motor cortex devoted to different parts of the body — the best-known localization study of the previous generation — to argue that the ‘reason that a chimpanzee cannot learn to talk is simply that the large amounts of brain necessary for speech are not there’ (p. 28). For Penfield’s research, conducted on patients undergoing brain surgery under a local anaesthetic, see Penfield & Rasmussen (1950); for discussion, see Fancher (1996), pp. 101–105. Washburn and Lancaster later published a brief commentary together on the evolution and origin of language; see Washburn & Lancaster (1971).
63 On Paul Flechsig’s turn-of-the-century studies of association cortex, see Finger (1994), pp. 308–309. On the identification of the limbic system, see ibid., pp. 286–290, and esp. Durant (1985). On the disenchantment with localization studies in the pre-Geschwind era, see Rosenfield (1985), p. 49; see also Robinson, 1967a, Geschwind, 1974.
64 On the hippocampus minor controversy, see, e.g., L. G. Wilson (1996). On the Victorian debate over language, brain function and human evolution more generally, and the contributions of the English experts on aphasia Frederic Bateman and John Hughlings-Jackson in particular, see Radick (2000). On the archival dimension of Geschwind’s 1965 paper, see the quotation from Geschwind in the obituary by Galaburda (1997), p. 296. See also Rosenfield (1985), pp. 49–51. For Geschwind’s papers on the history of studies of aphasia, see Geschwind (1974). Another historically-minded aphasiologist of the period with an interest in animal communication and the origin of language was MacDonald Critchley. See Critchley, 1958, Critchley, 1960, Critchley, 1970. His archly discursive papers had little impact on the wider discussion, however.
65 Robinson (1967a).
66 Ploog (1967). Ploog and his colleagues went on to do a great deal of research along these lines.
67 For many testimonials to the impact of Geschwind and his work on the rising generation, see Schachter & Devinsky (1997). Note too the titles of two journals begun in 1966: Experimental Brain Research and Physiology and Behavior.
68 Magoun in Darley (1967), p. 18. Geschwind was present at the meeting.
69 Robinson (1967b), p. 353. He wrote that in his experimental study of rhesus monkeys, ‘several hundred neocortical sites were stimulated without the production of any type of sound’ (p. 353).
70 Marshall (1970), pp. 234–236, with a further, approving summary on p. 240 of Geschwind’s picture of the language-capable human brain, where the angular gyrus region supports the cross-modal connections needed for object naming, and the language-incapable brains of the ‘sub-human’ primates, who lack the angular gyrus, and enjoy only limbically-mediated sensations.
71 Washburn (1973), pp. 45–47, quotation on p. 46. Washburn’s indictment of descriptive field primatology — as a science forever treating as separate what, experimentally considered, belongs together — echoed his criticisms of the biometrical physical anthropology of his youth. For an explicit, and unflattering, yoking together of the old anatomy and the new primatology, see ibid., p. 45. In an earlier discussion of the vervet findings, he was less dismissive, but just as doubtful about whether vervets name specific kinds of thing in the environment, and for the same reasons. See Washburn (1968), pp. 39–41, esp. p. 40.
72 For remarks along these lines, see Marler (1961), pp. 296–297.
73 See esp. Marler (1977).
74 Ibid., pp. 223–224, 227–228.
75 Marler’s ability to translate the affective/symbolic problem into experimental terms went hand in glove with his adaptationist confidence that things are as they are for functional reasons. Consider that natural selection can preserve and accumulate only those differences influencing an animal’s survival and reproduction. If partly symbolic alarm calls never promote survival better than wholly affective calls, selection can never favour symbolic communication. Yet — so this line reasoning goes — ours is a Darwinian world, and it contains symbolic communication. Therefore selection must have favoured partly symbolic alarm calls at least once in the past; and to have thus been favoured, those calls must have promoted survival better than non-symbolic alternatives.
76 It was with these latter experiments (see Seyfarth, Cheney, & Marler, 1980b, pp. 1088–1090) that Marler’s continuum interpretation came into its own, since the longer, louder python and eagle calls did seem to produce more intense python and eagle escape responses. In other words, call features that scale with arousal (length and loudness) seem to cue vervet listeners to the relative fearsomeness of a threat, while a call feature that does not scale with arousal (acoustic structure) informs vervets about the nature of the threat. Thus do affect and symbolization work together.
77 Washburn (1973), quotations on pp. 46 and 47. See also Washburn & Lancaster (1971).
78 Interview with Robert Seyfarth and Dorothy Cheney, March 2004.
79 For a recent, much-publicized call for knowledge united under Darwinism, see E. O. Wilson (1998). Wilson recalls his first encounter with modern Darwinism as follows (p. 4): ‘I was enthralled, couldn’t stop thinking about the implications evolution has for classification and for the rest of biology. And for philosophy. And for just about everything’.
80 Dennett (1995).
81 See esp. Dennett (1983).
82 In their papers on the vervet playbacks, Seyfarth, Cheney and Marler did offer observations showing that, in their words, ‘the link between alarm call and escape behaviour, while real, can be severed, as might be expected if animals were capable of operating in a “symbolic” mode’. Seyfarth et al., 1980a, Seyfarth et al., 1980b. As we have seen, the limbic system was associated with behaviours that were automatic and inflexible as well as affective. For a strong restatement of Washburn’s established views on animals, brains and language in the same year the playback papers were published, see Washburn & Moore (1980), pp. 172–178.


Altmann, S. A. (Ed.).  (1967a).  Social communication among primates.  Chicago: University of Chicago Press.

Altmann, S. A.  (1967b).  The structure of primate social communication.  In Altmann (1967a, pp. 325–362).

Anderson, S. R.  (2004).  Doctor Dolittle’s delusion: Animals and the uniqueness of human language.  New Haven and London: Yale University Press.

Bastian, J. R.  (1965).  Primate signaling systems and human languages.  In DeVore (1965, pp. 585–606).

Beach, F. A.  (1950).  The snark was a boojum.  American Psychologist, 5, 115–124.

Bickerton, D.  (1990).  Language and species.  Chicago: University of Chicago Press.

Bowler, P. J.  (1986).  Theories of human evolution: A century of debate, 1844–1944. Baltimore and London: Johns Hopkins University Press.

Burkhardt, Jr., R. W.  (2005).  Patterns of behavior: Konrad Lorenz, Niko Tinbergen, and the founding of ethology.  Chicago: University of Chicago Press.

Busnel, R.–G. (Ed.).  (1963).  Acoustic behaviour of animals.  Amsterdam and London: Elsevier.

Candland, D. K.  (1993).  Feral children and clever animals: Reflections on human nature.  Oxford: Oxford University Press.

Carpenter, C. R.  (1964).  Naturalistic behavior of nonhuman primates.  University Park, Pennsylvania: Pennsylvania State University Press. 

Catchpole, C. K., & Slater, P. J. B.  (1995).  Bird song: Biological themes and variations. Cambridge: Cambridge University Press.

Cheney, D. L., & Seyfarth, R. M.  (1990).  How monkeys see the world: Inside the mind of another species.  Chicago: University of Chicago Press.

Critchley, M.  (1958).  Animal communication.  Transactions of the Hunterian Society, 16 (1957-8), 89–112.

Critchley, M.  (1960).  The evolution of man’s capacity for language.  In S. Tax (Ed.), Evolution after Darwin, 3 vols., vol. 2: The evolution of man (pp. 289–308).  Chicago: University of Chicago Press.

Critchley, M.  (1970).  The nature of animal communication and its relation to language in man.  In M. Critchley, Aphasiology and other aspects of language (pp. 126–143). London: Edward Arnold.

Darley, F. L. (Ed.).  (1967).  Brain mechanisms underlying speech and language.  New York and London: Grune and Stratton.

Dennett, D. C.  (1983).  Intentional systems in cognitive ethology: The ‘Panglossian paradigm’ defended.  Behavioral and Brain Sciences, 6, 343–390 (includes extensive peer commentary and Dennett’s responses).

Dennett, D. C.  (1995).  Darwin’s dangerous idea: Evolution and the meanings of life.  New York: Simon and Schuster.

DeVore, I. (Ed.).  (1965).  Primate behavior: Field studies of monkeys and apes.  New York: Holt, Rinehart and Winston.

Dewsbury, D. A.  (Ed.).  (1989).  Studying animal behavior: Autobiographies of the founders. Chicago: University of Chicago Press.  First published as Leaders in the study of animal behavior.  London: Associated University Presses, 1985.

Dukelow, W. R.  (1995).  The alpha males: An early history of the regional primate  research centers.  Lanham, Maryland: University Press of America.

Durant, J. R.  (1985).  The science of sentiment: The problem of the cerebral localization of emotion.  In P. P. G. Bateson, & P. H. Klopfer (Eds.), Perspectives in ethology, 6: Mechanisms, 1–31.

Fancher, R. E.  (1996).  Pioneers of psychology (3rd  ed.).  New York and London: W. W. Norton.

Falls, J. B.  (1963).  Properties of bird song eliciting responses from territorial males. Proceedings of the XIII International Ornithological Congress, 259–271.

Finger, S.  (1994).  Origins of neuroscience: A history of explorations into brain function. Oxford: Oxford University Press.

Galaburda, A. M.  (1997).  Norman Geschwind (1926-1984).  In Schachter & Devinsky (1997, pp. 295–297).

Geschwind, N.  (1964).  The development of the brain and the evolution of language.  In C. I. J. M. Stuart (Ed.), Monograph Series on Languages and Linguistics, no. 17 (pp. 155-169).  Washington, D.C.: Georgetown University Press.  Reprinted in Geschwind (1974, pp. 86–104).

Geschwind, N.  (1965).  Disconnexion syndromes in animals and man, parts I and II.  Brain88, 237–294; 585-644.  Reprinted in Geschwind (1974, pp. 105–236).

Geschwind, N.  (1974).  Selected papers on language and the brain.  R. S. Cohen, & M. W. Wartofksy (Eds.), Boston Studies in the Philosophy of Science, 16.  Boston: D. Reidel.

Goodall, J.  (2001).  Beyond innocence: An autobiography in letters – the later years.  Ed. D. Peterson.  New York: Houghton Mifflin.

Griffin, D. R.  (1989).  Recollections of an experimental naturalist.  In Dewsbury (1989, pp. 121–142).

Haraway, D. J.  (1989).  Primate visions: Gender, race, and nature in the world of modern science.  London: Routledge, Chapman and Hall.

Hauser, M. D.  (1996).  The evolution of communication.  Cambridge, MA: MIT Press. Hauser, M. D.  (2000).  Wild minds: What animals really think.  London: Allen Lane/ Penguin Press.

Hayes, C.  (1952).  The ape in our house.  London: Victor Gollancz.  First published in 1951, New York: Harper & Row. 

Hinde, R.  (2000).  Some reflections on primatology at Cambridge and the science studies debate.  In Strum & Fedigan (2000, pp. 104–115).

Hooton, E. A.  (1947).  Up from the ape (rev. ed.)  New York: Macmillan.

Jay, P. C. (Ed.).  (1968).  Primates: Studies in adaptation and variability.  New York: Holt, Rinehart and Winston.

Jolly, A.  (2000).  The bad old days of primatology?  In Strum & Fedigan (2000, pp. 71–84).   

Lancaster, J. B.  (1968).  Primate communication systems and the emergence of human language.  In Jay (1968, pp. 439–457).

Lawick-Goodall, J. V.  (1968).  A preliminary report on expressive movements and communication in the Gombe Stream chimpanzees.  In Jay (1968, pp. 313–374). 

Marler, P.  (1961).  The logical analysis of animal communication.  Journal of Theoretical Biology, 1, 295–317.  Reprinted in facsimile in L. D. Houck, & L. C. Drickamer (Eds.), Foundations of animal behavior: Classic papers with commentary (pp. 649–671).  Chicago: University of Chicago Press, 1996.

Marler, P.  (1963).  The mountain gorilla: ecology and behavior.  [Review of G. Schaller’s 1963 book with that title.]  Science, 140 (7 June), 1081–1082.

Marler, P.  (1965).  Communication in monkeys and apes.  In DeVore (1965, pp. 544–584).

Marler, P.  (1967).  Animal communication signals.  Science, 157 (18 August), 769–774.

Marler, P.  (1968).  Aggregation and dispersal: Two functions in primate communication.  In Jay (1968, pp. 420–438).

Marler, P.  (1969a).  Colobus guereza: Territoriality and group compostion.  Science, 163 (3 January), 93–95.

Marler, P.  (1969b).  Vocalizations of wild chimpanzees: An introduction.  Recent Advances in Primatology, 1, 94–100.

Marler, P.  (1970).  Vocalizations of East African monkeys: I. Red colobus.”  Folia Primatologica, 13, 81-91.

Marler, P.  (1972).  Vocalizations of East African monkeys II: Black and white colobus. Behaviour, 42, 175–197.

Marler, P.  (1973).  A comparison of vocalizations of red-tailed monkeys and blue monkeys,  Cercopithecus ascanius and C. mitis, in Uganda.  Zeitschrift für Tierpsychologie, 33, 223-247.

Marler, P.  (1977).  Primate vocalization: Affective or symbolic?  Reprinted in T. A. Sebeok, & J. Umiker-Sebeok (Eds.), Speaking of apes: A critical anthology of two-way communication with man (pp. 221–229).  New York and London: Plenum Press.  First published in G. H. Bourne (Ed.), Progress in Ape Research (pp. 85–96).  New York: Academic Press. 

Marler, P.  (1989).  Hark ye to the birds: Autobiographical marginalia.  In Dewsbury (1989, pp. 315–345).

Marler, P., & Hobbett, L.  (1975).  Individuality in a long-range vocalization of wild chimpanzees.  Zeitschrift für Tierpsychologie, 38, 97–109.

Marshall, J. C.  (1970).  The biology of communication in man and animals.  In J. Lyons (Ed.), New horizons in linguistics (pp. 229–241).  Harmondsworth: Penguin.

Moles, A.  (1963).  Animal language and information theory.  In Busnel (1963, pp. 112–131).

Munz., T.  (Forthcoming).  The bee battles: Karl von Frisch, Adrian Wenner and the honeybee dance language controversy.  Journal of the History of Biology, special issue on animal communication.

Penfield, W., & Rasmussen, T.  (1950).  The cerebral cortex of man: A clinical  study of localization of function.  New York: Macmillan.

Ploog, D. W.  (1967).  The behavior of squirrel monkeys (Saimiri sciureus) as revealed by sociometry, bioacoustics, and brain stimulation.  In Altmann (1967a, pp. 149–184).

Radick, G.  (2000).  Language, brain function, and human origins in the Victorian debates on evolution.  Studies in History and Philosophy of Biological and Biomedical Sciences, 31C, 55–75.

Radick, G.  (Forthcoming.)  Primate language and the playback experiment, in 1890 and 1980.  Journal of the History of Biology, special issue on animal communication.

Ramsay, A.  (1966).  Animal communication, report of a conference.  Current Anthropology, 7, 251–253.

Ribnick, R.  (1982).  A short history of primate field studies: Old World monkeys and apes.  In F. Spencer (Ed.), A history of American physical anthropology, 1930–1980 (pp. 49–73). New York: Academic Press.

Robinson, B. W.  (1967a).  Neurological aspects of evoked vocalizations.  In Altmann (1967a, pp. 135–147).

Robinson, B. W.  (1967b).  Vocalization evoked from forebrain in Macaca mulatta. Physiology and Behavior, 2, 345–354. 

Roe, A., & Simpson, G. G. (Eds.).  (1958).  Behavior and evolution.  New Haven: Yale University Press.

Rosenfield, I.  (1985).  A hero of the brain.  New York Review of Books (21 November), 49–55.

Rowell, T. E.  (1962).  Agonistic noises of the Rhesus monkey (Macaca mulatto).  Symposia of the Zoological Society of London, 8, 91–96. 

Schachter, S. C., & Devinksy, O. (Eds.).  (1997).  Behavioral neurology and the legacy of Norman Geschwind.  Philadelphia and New York: Lippincott–Raven. 

Sebeok, T. A. (Ed.).  1968.  Animal communication: Techniques of study and results of research.  Bloomington, Indiana: Indiana University Press.

Sebeok, T. A., & Ramsay, A. (Eds.).  (1969).  Approaches to animal communication.  The Hague: Mouton.

Seyfarth, R. M., Cheney, D. L., & Marler, P.  (1980a).  Monkey responses to three different alarm calls: Evidence of predator classification and semantic communication.  Science, 210 (14 November),801–803.

Seyfarth, R. M., Cheney, D. L., & Marler, P.  (1980b).  Vervet monkey alarm calls: Semantic communication in a free-ranging primate.  Animal Behaviour, 28, 1070–1094.

Smith, W. J.  (1963).  Vocal communication in birds.  American Naturalist, 97, 117–125.

Smith, W. J.  (1965).  Message, meaning, and context in ethology.  American Naturalist, 99, 405–409.

———.  (1968).  Message–meaning analyses.  In Sebeok (1968, pp. 44–60).

———.  (1977).  The behavior of communicating: An ethological approach.  Cambridge, MA: Harvard University Press.      

Struhsaker, T. T.  (1967a).  Auditory communication among vervet monkeys (Cercopithecus aethiops).  In Altmann (1967a, pp. 281–324).

Struhsaker, T. T.  (1967b).  Behavior of vervet monkeys and other cercopithecines.  Science, 156, 1197–1203.

Struhsaker, T. T.  (1967c).  Social structure among vervet monkeys (Cercopithecus aethiops). Behaviour, 29, 83–121.

Struhsaker, T. T.  (1967d).  Ecology of vervet monkeys (Cercopithecus aethiops) in the Masai–Amboseli Game Reserve, Kenya.  Ecology, 48, 891–904.

Strum, S. C., & Fedigan, L. M. (Eds.)  (2000).  Primate encounters: Models of science, gender, and society.  Chicago: University of Chicago Press.

Strum, S. C., Lindburg, D. G., & Hamburg, D. (Eds.).  (1999).  The new physical anthropology: Science, humanism, and critical reflection.  Upper Saddle River, NJ: Prentice–Hall.

Terborgh, J.  (1996).  Cracking the bird code.  New York Review of Books, 11 January, 40–44.

Washburn, S. L.  (1951).  The analysis of primate evolution with particular reference to the origin of man.  Reprinted in W. Howells (Ed.), Ideas on human evolution: Selected essays, 1949–1961 (pp. 154–171).  New York: Atheneum, 1967.  First published in Cold Spring Harbor Symposia on Quantitative Biology, 15, 67–78.

Washburn, S. L.  (1952).  The strategy of physical anthropology.  In A. L. Kroeber (Ed.), Anthropology today: An encyclopedic inventory (pp. 714–727).  Chicago: University of Chicago Press.

Washburn, S. L.  (1959).  Speculations on the interrelations of the history of tools and biological evolution.  In J. N. Spuhler (Arr.), The evolution of man’s capacity for culture (pp. 32–39).  Detroit: Wayne State University Press.

Washburn, S. L.  (1968).  The study of human evolution.  Eugene, Oregon: Oregon State System of Higher Education.

Washburn, S. L.  (1973).  The promise of primatology.  Reprinted in Strum, Lindburg & Hamburg (1999, pp. 43–48).  First published in the American Journal of Physical Anthropology, 38, 177–182. 

Washburn, S. L.  (1983).  Evolution of a teacher.  Annual Review of Anthropology, 12, 1–24.

Washburn, S. L., & DeVore, I.  (1961).  Social behavior of baboons and early man.  In S. L. Washburn (Ed.), Social life of early man (pp. 91–105).  London: Methuen.

Washburn, S. L., & Lancaster, J. B.  (1971).  On evolution and the origin of language.  [With reply by L. Carini.]  Current Anthropology, 12, 384–386.

Washburn, S. L., & Moore, R.  (1980).  Apes into man: A study of human evolution (2nd ed.).  Boston: Little, Brown.

Weidman, N.  (1996).  [Review of Dukelow (1995).]  Isis, 87 (4), 756–757.

Wilson, E. O.  (1994).  Naturalist.  Washington, D. C.: Island Press.

Wilson, E. O.  (1998).  Consilience: The unity of knowledge.  New York: Knopf.

Wilson, L. G.  (1996).  The gorilla and the question of human origins: The brain controversy.  Journal of the History of Medicine and Allied Sciences, 51, 184–207.

Zhinkin, N. I.  (1963).  An application of the theory of algorithms to the study of animal   speech: Methods of vocal intercommunication between monkeys.  In Busnel (1963, pp. 132–180).