who belongs in the "culture club"
ARTICLES
Do Animals Have Culture? KEVIN N. LALAND AND WILLIAM HOPPITT
WHY IS CULTURE RARE IN ANIMALS? OR IS IT?
The concept of culture has become a quagmire for the social sciences. In- tuitively, we all know what it is and feel it to be important, yet it has proven virtually impossible for social scientists to derive a consensual defi- nition or to find a satisfactory means of operationalizing it.1,2 Indeed, after much painstaking self-analysis and soul searching, many social anthro-
pologists seem to have given up on the notion of culture altogether.3,4 Ironi- cally, and perhaps benefiting from blissful ignorance of the problems, bi- ologists and students of animal behav- ior are currently giving unprece- dented attention to the “culture” of other animals.5–7 This community has witnessed an explosion of interest in animal social learning and purported animal culture6,8 exemplified by a pro- liferation of books, conferences, and papers dedicated to the topic.5,9 –11
Whether or not the culture concept has utility for our own species, an in- creasing number of researchers are using it to explain both diversity and transgenerational continuity of be- havior in animals.
In the absence of a satisfactory and universally accepted definition of culture, most answers to the ques- tion, “Do animals have culture?” take on a hackneyed appearance. De- fine culture one way and it is the exclusive province of human beings; another way, and a multitude of spe- cies are deemed worthy of the acco- lade. At one pole of the spectrum, many social anthropologists con- ceive of culture as so deeply shrouded in language, politics, mo- rality and institutions, that it would be grossly devalued were it attrib- uted to dolphins or apes. At the other extreme, biologists Charles Lums- den and Edward Wilson12 concluded that in excess of 10,000 species, in- cluding some bacteria, manifest
some form of extra-genetic inheri- tance sufficient to be deemed cul- tural. Thus, central to the issue of whether animals exhibit culture is the problem of whether it is possible to come up with a nonarbitrary def- inition of culture; one that will prove useful to a substantive proportion of interested researchers. We begin by seeking such a definition, and in do- ing so are guided by the twin criteria of consensus and utility. In simple terms, we are looking for the kind of definition of culture that our profes- sional colleagues will recognize as something with which they can work.
The first step is to ask whether such a definition would be more use- ful if it were broad and minimalist, reflecting the continuity between hu- mans and other animals, or whether a narrow and exacting definition that stresses human uniqueness might have greater utility. We sus- pect that to most biologically minded researchers the answer to this question is immediately appar- ent. A narrow definition, say some- thing like, “Cultures are systems of linguistically encoded conceptual phenomena that are learned through teaching and imitation, socially transmitted within populations, and characteristic of groups of people,” is not likely to prove particularly useful, at least not to the issues ad- dressed in this article. This is not only because the answer to the ques- tion of whether or not animals have culture is a fait accompli; by defini- tion, they do not. But in addition, this denotation would act as a bar- rier to understanding of the evolu- tionary roots of culture. It would en- courage the misguided view that humans are (uniquely) unique, which is only trivially true since all
Kevin Laland is a Royal Society University Research Fellow and Reader in Biology at the University of St. Andrews, U.K. His research encompasses a range of topics related to animal behavior and evolution, particularly social learning, cultural evolu- tion, and niche construction. He is the co-author (with Gillian Brown) of Sense and Nonsense. Evolutionary Perspectives on Human Behavior (2002) published by Oxford University Press and (with John Odling-Smee and Marc Feldman) of Niche Construction. The Neglected Process in Evolution (2003) by Princeton University Press. E-mail: Knl1@st-and.ac.uk Will Hoppitt is a BBSRC-funded PhD stu- dent in the Department of Zoology at the University of Cambridge, U.K. His re- search includes empirical work on social learning in birds and mathematical mod- eling of social learning processes.
Key words: social learning, innovation, teaching, imitation, niche construction
Evolutionary Anthropology 12:150 –159 (2003) DOI 10.1002/evan.10111 Published online in Wiley InterScience (www.interscience.wiley.com).
Culture is probably not rare in animals, although hard experimental evidence is lacking. The strongest case for culture is found in the species most amenable to experimental manipulation, rather than in nonhuman primates. Human culture is much more likely to be cumulative than animal culture, but the reasons for this are not well established. At this point, there is no reason to assume that cumulative culture depends critically on teaching, imitation, language, or perspective-taking. Currently, animals are being judged according to stricter criteria than humans.
150 Evolutionary Anthropology
species are unique. No light would be shed on how culture came into existence, nor on human’s place in nature. More fundamentally, it would act as a barrier to the integra- tion of the biological and social sci- ences. Premature, over-exacting dis- tinctions jeopardize the ability to see relationships between culture-like phenomena in diverse taxa.
Given the primitive state of knowl- edge of animal cultures and the cur- rent malaise in human culture stud- ies, we suggest that a broader definition is likely to be more stimu- lating. It may prove valuable to de- bate the criteria by which cultural and acultural species are distin- guished and perhaps, in time, even to make distinctions between differ- ent classes of culture. But such dis- tinctions, if they are going to be in- formed, can be based only on empirical data. A broad definition encourages such data to be collect- ed; a narrow definition does not. A broad definition has the additional advantages that it provides a com- parative framework with which to investigate the evolutionary history of culture and that it encourages the cross-fertilization of information and methods between the biological and social sciences. Naturally, the definition could also be too all-en- compassing, so that “culture” be- comes unrecognizable and research- ers become alienated. An overly broad definition, like an overly nar- row one, would have neither utility nor consensus.
By what means can we settle on a broad and useful definition of cul- ture? One approach is to synthesize the array of disparate definitions of human culture and isolate the crite- ria that are deemed essential by all, or at least most, parties.1,7 What then, when we strip it down to its barest bones, is the essence of cul- ture? One central criterion is that culture is built upon socially learned and socially transmitted informa- tion. The term “culture” does not ap- ply to inherited genetic information or the knowledge and skills that in- dividuals acquire on their own. A second is that this socially transmit- ted information can underpin group- typical behavior patterns, which
may vary from one population to the next. Thus, culture helps to explain both continuity within groups and diversity between groups.
These considerations lead us to the following working definition of culture, toward which we hope a substantial proportion of research- ers interested in cultural evolution would be sympathetic. Cultures are those group-typical behavior pat- terns shared by members of a com- munity that rely on socially learned and transmitted information. (Later we will discuss addition characteris- tics that some researchers regard as essential to human culture.)
Which Animals Have Culture?
According to the preceding defini- tion, which animals have culture? There are two kinds of answers to this question. The first kind is based exclu- sively on hard experimental evidence. That is, for which species do we have reliable scientific evidence of natural communities that share group-typical behavior patterns that are dependent on socially learned and transmitted information? The answer, which will surprise many, is humans plus a handful of species of birds, one or two whales, and two species of fish.
No doubt many readers will find this conclusion disturbing, while primatolo- gists will probably be up in arms. How can we attribute culture status to fish and not chimpanzees? A full explana- tion will follow, but the short answer is that for chimpanzees, as for other non- human primates, the hard evidence that their “cultures” are socially learned is not yet there.13–15 Neither evidence of group-typical behavior patterns nor a demonstration that the species is capa- ble of social learning is in itself strong evidence for culture.16
The second kind of answer to the question of which animals have cul- ture is a best guess. If we were to say which animals we believe have cul- ture, based on our knowledge of ani- mal social learning, observations of natural behavior of animals, intuition, and the laws of probability, we would say that many hundreds of species of vertebrate have culture. After all, many vertebrates have been shown to be capable of social learning of a sort that theoretically could maintain cul-
ture.5,10 While our particular beliefs are of little intrinsic value, we suspect that many researchers that study ani- mal social learning would agree with us on this point.
To return then to our question, cul- ture is as rare or as common among animals as it is defined to be. On the assumption that broad definitions lead to greater utility and consensus than do narrow ones, we have defined culture according to its necessary conditions. This definition leads us to the view that culture is not at all rare, although hard experimental evidence of cultures among animals, including nonhuman primates, is currently lacking.
WHAT IS UNIQUE ABOUT PRIMATE CULTURE?
A “Brainist” Bias?
A slightly (but only slightly) face- tious answer to the question of what is unique about primate culture is that primates uniquely benefit from the bi- ased anthropocentric perspective of investigators who themselves, without exception, are primates. Nonhuman primates are probably studied more than most other species; the findings of these studies perhaps receive greater attention than those of other species; and arguably lower standards of evidence are required for primates than other species to be awarded some sophisticated psychological or social attribute, including culture. This bias, we suspect, stems in part from a need dating back to Darwin for researchers to demonstrate continuity of mental abilities between humans and other animals. The easiest and most convenient means by which such continuity can be demonstrated is to find the most sophisticated cog- nition in our closest relatives, the apes, slightly less impressive abilities in monkeys, and increasing levels of stupidity as the degree of relatedness with ourselves diminishes. Reinforced by the observation that brain size crudely fits this pattern (if we ignore some inconvenient outliers such as cetaceans), many of us begin our re- search expecting to find that nonhu- man primates will be smarter than other animals. From an evolutionary perspective, this bias is unjustifiable, since convergent evolution could
ARTICLES Evolutionary Anthropology 151
manifestly have favored intelligence, sophisticated cognition, or culture in quite distinct lineages.
Indeed, the compelling assumption that species with larger volumes of the relevant brain structures will show greater cognitive sophistication is still disputed17 and remains largely un- tested. There is evidence among non- human primates for a link between brain size and behavioral flexibility, as measured by frequency of reports of innovation, social learning, and tool use.18 However, this finding demon- strates that big-brained species ex- hibit more social learning than small- brained species, and not that the former employ more sophisticated psychological processes than do the latter.
In sum, it is not clear that a big brain is necessary for culture as de- fined earlier. If some fish exhibit cul- ture, as we will argue, then presum- ably a large brain is not a requisite feature. Given its expense to produce and maintain, all of the neural tissue that primates have in abundance must have some important function. Yet the fact remains that it may have little to do with culture. Most people, our- selves included, believe that nonhu- man primates have culture in spite of the scientific evidence, rather than be- cause of it.
Why There Is Better Evidence for Culture in Fish Than in Primates
What evidence is needed to demon- strate that a species exhibits culture, according to our definition? Research- ers would need to establish that the animal concerned exhibits diverse group-typical behavior patterns in natural populations and, in addition, that these behavior patterns are un- derpinned by social learning. The lat- ter criterion requires ruling out alter- native explanations for the behavioral diversity, including that the popula- tions exhibit genetic differences or that variant ecological conditions have resulted in different patterns of asocial learning.
In principle, ruling out these alter- natives through experimentation is straightforward. Two experimental manipulations are necessary. In the
first experiment, a sample of individ- uals from population A are introduced into population B, and vice versa, ide- ally at a formative age. The observa- tion that the introduced animals adopt the behavior exhibited by mem- bers of the host population is incon- sistent with an explanation in terms of genetic differences between popula- tions and consistent with an explana- tion reliant on some form of learning. In the second experiment, population A is collectively removed from its en- vironment and replaced in the envi- ronment of population B, which in turn is removed and replaced in the
environment of population A. If the introduced populations come to ex- hibit the same behavior as the former residents, this would suggest that the behavior results from shaping to di- vergent ecological conditions, and asocial learning could not be ruled out. However, the observation that the introduced population exhibits group- typical behavior that differs signifi- cantly from that exhibited by the former residents would be inconsis- tent with an explanation based on eco- logical differences between environ- ments and consistent with culture.
Ideally, both experiments would be carried out on the same populations, which would potentially provide the most compelling evidence for culture. The first experiment, in combination with alternative evidence that the be- havior in question cannot be learned asocially, or the second experiment, in combination with alternative evi- dence that there are no relevant ge- netic differences between popula- tions, would also provide satisfactory evidence for culture.
To our knowledge these experi- ments have not been carried out in a systematic and rigorous manner in any nonhuman primates. What is more, in many cases, including for chimpanzees, it is difficult to envisage that they will. Not only would such experiments be extraordinarily expen- sive, and present enormous logistical challenges, but many people would re- gard such manipulations as unethical. Consequently, the claim that nonhu- man primate populations exhibit cul- ture rests exclusively on observations of natural populations in situ.
That there are population differ- ences in the behavioral repertoires of many primates is now well docu- mented. Best known are the distinct tool-using and other group-typical be- havior patterns of chimpanzee popu- lations.19 –22 Similar behavioral diver- sity consistent with a cultural explanation is reported in several other primates, notably orangutans23
and capuchins.11,24
Differences in the behavioral reper- toires of chimpanzee populations have become, for many, the jewel in the crown of nonhuman primate cul- tural traditions and the prime exem- plar of animal culture. Yet, in the ab- sence of the kind of experimental evidence described, the case for chim- panzee culture rests entirely on a plausibility argument. There is clear experimental evidence that chimpan- zees are capable of social learning.25
Indeed, the evidence for seemingly more complex forms of social learn- ing, such as imitation, is as good in this species as in any other, humans aside.26,27 Many of the group-typical behavior patterns, such as termiting, appear quite complex, and research- ers envisage that they would be diffi- cult to learn alone. Some behaviors,
Given its expense to produce and maintain, all of the neural tissue that primates have in abundance must have some important function. Yet the fact remains that it may have little to do with culture. Most people, ourselves included, believe that nonhuman primates have culture in spite of the scientific evidence, rather than because of it.
152 Evolutionary Anthropology ARTICLES
such as hand-clasp grooming,28 ap- pear arbitrary, even functionless, and consequently seem to many to be un- likely candidates for behavior shaped by ecological contingencies (see also the primate social conventions de- scribed by Perry and Mason24). In their seminal paper on chimpanzee cultures, Whiten and coworkers20
claim to have found “39 different be- havior patterns, including tool usage, grooming and courtship behaviors, [that] are customary or habitual in some communities but are absent in others where ecological explanations have been discounted.” For most in- terested researchers, the probability is that at least some of the group-typical behavior patterns reported for chim- panzees are learned socially. Yet to our knowledge, in not one instance is there irrefutable evidence that a natu- ral chimpanzee behavior is socially learned.
Indeed, there are serious grounds for being suspicious of this plausibil- ity argument, and senior figures in the animal social-learning community re- main unconvinced.13–15 The main problem is that it is difficult to rule out ecological explanations for the ob- served variation in behavior. Whiten and coworkers20 consider the possibil- ity of false-positive reports of cultural variation resulting from an “ecologi- cal explanation, for which absence is explicable because of a local ecologi- cal feature.” However, only three of sixty-five behavior patterns were actu- ally removed because an ecological explanation could be given. Skeptics might be forgiven for questioning how seriously this alternative was consid- ered. For instance, in the case of a population reported as not nut-ham- mering, a skeptic would want to know not only that there are suitable nuts and hammers present, but also that there are not alternative and more readily accessed alternative sources of foods with equivalent nutritive and energetic properties that render nuts a comparatively less profitable food source. Similarly, the skeptic would want to know whether populations re- ported to not fish for ants or termites are exposed to exactly the same spe- cies of ants and termites as those that do, and that the ants and termites be- have the same way in each case.
This problem is highlighted by a re- cent analysis of chimpanzee ant dip- ping by Humle and Matsuzawa.29
Whiten and coworkers20 singled out ant dipping as a particularly compel- ling case of chimpanzee cultural vari- ation where the differences between communities are not related solely to whether the behavior is present or ab- sent, but where two different methods of ant dipping were employed. In the first or “pull through” method, re- ported as customary at Gombe, a long wand is held in one hand and a ball of ants is wiped off with the other. In the second “direct mouthing” method, customary at Tai, a short stick is held in one hand and used to collect a smaller number of ants, which are transferred directly to the mouth. Like Boesch and Boesch30 and McGrew19
before them, Whiten and colleagues20
were satisfied that ecological differ- ences between the Gombe and Tai sites could not explain this variation. Furthermore, they concluded “it is difficult to see how such behavior pat- terns could be perpetuated by social learning processes simpler than imita- tion.” Humle and Matsuzawa29 inves- tigated ant dipping among chimpan- zees of Bossou, in southeastern Guinea, a site at which both forms of ant dipping are observed. They found that both the length of the tools and the technique employed were strongly influenced by the nature of the prey. Several species of ants were preyed on, and these differed in their density, aggressiveness, behavior, and in whether they were migratory or at nest sites. In situations in which ants were abundant, aggressive, or had se- vere bites, the first method, with its long wand, was employed, and was probably associated with fewer bites than the second method, while in other circumstances the direct mouthing method was used. Skeptics would find it easy to envisage that chimpanzees could individually be shaped by biting insects to use the strategy that resulted in the fewest bites.
Humle and Matsuzawa’s29 findings suggest that an ecological explanation for ant dipping cannot be ruled out. If confidence in this flagship example of cultural variation in chimpanzees must be suspended, then how many
other so-called chimpanzee cultural behaviors will eventually be ques- tioned, when careful scrutiny reveals that variation coincides with ecologi- cal differences? Will all thirty-nine of Whiten and coworkers’20 behaviors eventually be whittled away? Humle and Matsuzawa’s29 analysis implies that, had the experimental manipula- tions described been employed by translocating Gombe and Tai chimps, we would have found Gombe-born chimps using the direct mouthing method in Tai and Tai-born chimps using the pull-through method at Gombe (that is, chimpanzees would have failed the culture test), although we will not know for sure unless this experiment is actually done. Let us be clear: Neither Humle and Matsuza- wa’s29 analysis nor the hypothetical translocation experiment would rule out a cultural explanation for ant dip- ping. However, while the alternative ecological explanation remains plau- sible there will always be a reason to doubt the claim of culture. The trans- location experiment method is supe- rior because, where such experiments have been carried out and behavior consistent with cultural variation has been observed, we are not left doubt- ing whether unconsidered ecological factors might explain the variation.
While primatologists are forced to resort to circumstantial evidence for culture (illustrated by the plausibility arguments put forward by Boesch,21
Whiten,22 and Perry and Mason24), other researchers are free to employ more direct experimental methods. Helfman and Shultz31 translocated French grunts (Haemulon flavolinea- tum) between populations and found that while those fish placed into estab- lished populations adopted the same schooling sites and migration routes as the residents, control fish intro- duced into regions from where the residents had been removed did not adopt the behavior of former resi- dents. Helfman and Shultz’s study combines the procedures for experi- ment 1 and 2 described earlier in one of the most elegant demonstrations of animal culture to date. Their study is not quite the same as the symmetrical experiment suggested because only one population was introduced into the environment of the other. None-
ARTICLES Evolutionary Anthropology 153
theless, this is strong evidence for cul- ture. Warner32 showed that translo- cated populations of Bluehead wrasse (Thalassoma bifasciatum), a Carib- bean coral-reef fish, did not adopt the same mating sites as the populations they replaced, but rather maintained distinct sites long after the manipula- tions, consistent with a cultural expla- nation. This finding is a direct appli- cation of the experiment 2 method. Combined with the observation that high levels of mixing during the early life of this species mean that reef pop- ulations are not subject to significant genetic differentiation, this finding provides acceptable evidence of cul- tural variation.
In the light of a burgeoning litera- ture on learning and cognition in fish, in which it is quite apparent that the abilities and complexity of social be- havior of this group have been under- estimated,33 we should hardly be sur- prised that two species of fish should provide some of the best evidence for animal cultures. A recent review of fish cognition33 reveals a rich experi- mentally derived databank reporting social intelligence (social strategies, Machiavellian intelligence, coopera- tive hunting) as well as sophisticated foraging skills, tool use, cognitive maps, and the construction of com- plex artifacts. Fish have compara- tively small brains and are not re- nowned for their intelligence or learning abilities. Yet there is now un- equivocal evidence from laboratory and field studies that a variety of dif- ferent species of fish are capable of social learning, including learning how to find food, which foods to eat, how to recognize predators, how to avoid predators, and perhaps even with which fish to mate (see Brown and Laland34 for a review). If fish can do all this with small brains, then maybe simple mechanisms could ac- count for primate behavior.
When it comes to hard experimen- tal evidence for culture, the only tax- onomic group that comes close to fish is birds, for which the cultural creden- tials of song variants and mating pref- erences are reasonably well-estab- lished in several species.35–37 It has been observed that female quail (Coturnix japonica) at lekking sites prefer to mate with particular
males.38 Furthermore, in the labora- tory it has been demonstrated that preferences for particular male traits can be culturally transmitted between female birds.37 However, to our knowledge it has never been demon- strated that differences in female pref- erence at lekking sites in the wild are cultural. Freeberg39,40 has gone one step further toward demonstrating culture in cowbirds (Molothrus ater). Male cowbirds in two distinct popula- tions exhibit different types of court- ing behavior and the females in each population prefer to mate with males who court in the local fashion. Free- berg found that juvenile birds of a common genetic background, when housed with adult birds from one or other of the populations, adopted the courting behavior and mating prefer- ences of the population they were housed with (see also Freeberg and coworkers41). This case introduces the possibility of an unusual transmission mechanism in which juvenile males’ courting behavior is shaped by adult females and juvenile females’ prefer- ences are shaped by the behavior of adult males. However, we cannot en- tirely rule out an ecological explana- tion, since during the experiment each population was housed at a different location. However, this would seem to be unlikely to account for the ob- served differences in behavior. In- deed, there is a large body of evidence that the vocal signals of many avian species may be cultures (reviewed by Freeberg36). However, as far as we are aware, there is no case where an eco- logical explanation for population dif- ferences in vocal repertoire can be ruled out entirely.
There is also some suggestive evi- dence for culture in several species of cetaceans. Killer whales (Orcinus orca) show a number of pod-specific behavior patterns, including foraging specializations, migration patterns, and vocal dialects. Rendell and White- head42 argue that ecological explana- tions can be ruled out, since sympat- ric pods develop different pod-specific behaviors. They also consider a ge- netic explanation unlikely since pods are matrilineal, so that inheritance of behavior would have to be principally from the mother. However, sex-linked genetic traits or genomic imprinting
(the inactivation of an allele depen- dent on the parent it was inherited from) could account for this pattern. Likewise, an ecological explanation could be viable if sympatric pods use the same habitat in different ways. Of course, such behavior itself could be cultural but, alternatively, different pods could have genetic predisposi- tions to use a habitat in a particular way. Rendell and Whitehead42 pro- vide what we feel is stronger evidence for culture in Humpback whales (Megaptera novaeangliae). During the breeding season all males in a popu- lation sing, at any one time, “nearly the same song,” but this song evolves over time, changing even within a breeding season. The rapidity of change would seem to rule out a ge- netic explanation, although demo- graphic changes or common external influences might account for song evolution; however, neither would seem likely to account for the ob- served synchronized change in song between many individuals. This evi- dence that can be mustered for birds and cetaceans is at least as strong, and in some cases stronger, than that for nonhuman primates.
The primate-centric brainist bias is well illustrated by a reconsideration of one of the most well-known cases of animal social learning, that of sweet- potato washing by Japanese ma- caques.43 For a novel cultural behav- ior pattern to spread through an animal population two processes are necessary: There must be the initial inception of the behavior, which we refer to as innovation, and its spread to others by social learning, which is diffusion. Imo, the famous originator of potato- and wheat-washing, has be- come celebrated as a great innovator, and therefore as an initiator of ma- caque “culture.” For instance, E.O. Wilson44 in Sociobiology and John Bonner45 in The Evolution of Culture in Animals both refer to Imo as “a genius,” while Kummer and Goodall46
describe her as “gifted.” In the 1960s, food washing seemed an unlikely nat- ural behavior for monkeys. Hence, these behavior patterns were regarded as novel and intelligent. When the food-washing behavior spread it was described as “preculture” or “pro- toculture,” the clear implication being
154 Evolutionary Anthropology ARTICLES
that it could be regarded as an analog, or perhaps even a homolog, of human culture.
With the benefit of hindsight, and in the light of accumulated scientific knowledge, Imo’s achievements and those of her troop members appear more modest. It seems that food washing is a stable feature of ma- caque behavior, and Imo’s innovation involved the application of an estab- lished behavior pattern to novel foods.47 Experimental studies have re- vealed that food washing is learned relatively easily by monkeys and can become common in a troop through processes other than the imitation of a rare “creative genius.”47 There is no evidence that food washing spread through Imo’s troop by imitation, teaching, or any unusually sophisti- cated form of social learning. Indeed, there are grounds for concern that it may even have been an artifact of human provisioning.14 The lauding of Imo as a genius and the use of the terms “protoculture” or “precultural” to describe this case of social trans- mission were widely accepted only be- cause the behavior was reported for a primate. Would the scientific commu- nity have accepted the same terminol- ogy to describe social learning in rats or budgerigars?
The members of our laboratory have carried out a detailed analysis of the factors that underpin innovation in fish, birds, and monkeys, and found that state-dependent factors (sex, size, social rank, hunger level) account for much of the variation. These and a host of other findings48 –52 lead us to the view that the adage, “Necessity is the mother of invention” probably ap- plies to most animal innovation. Cer- tainly when it comes to foraging inno- vation, more often than not the innovators are those individuals with the greatest motivation to find food. To our knowledge, no consideration has been given to the possibility that Imo was the first individual to apply food washing to sweet potatoes merely because she was among the hungriest individuals in the troop, yet this would seem a real possibility. Our experiments have found evidence for personality differences in the prob- lem-solving ability of guppies, a small freshwater fish. Laland and Reader48
found that some fish consistently solved foraging problems, not because they were hungrier, faster swimmers, more active, or healthier, but because of some stable personality character- istic. To our knowledge there is no equivalent experimental evidence for personality differences in problem- solving ability in any species of non- human primate. Many primate stud- ies report differences in performance between individuals, but none have ruled out alternative explanations for individual differences (for example, motivational factors). Hence, the vari- ation cannot reliably be interpreted as reflecting differences in personality. The one study of which we are aware that sought such evidence, Fragaszy
and Visalberghi’s53 investigation of in- novation in capuchin monkeys, failed to find it.
Let us give the single fish that con- sistently solved our maze problems faster than others the name of Bertha. We do not yet know whether it is ap- propriate to describe the personality trait that Bertha exhibits as intelli- gence or creativity as opposed to, say, boldness, perseverance, or nonconfor- mity but at least we know that there is some feature of Bertha’s make-up that makes her an effective problem solver. Gifted highbrow intellectual qualities, as opposed to less salubrious person- ality traits, have been attributed to Imo in an essentially arbitrary man-
ner and, indeed, in the absence of any evidence that it was her personality to which her success can be attributed. Under such circumstances, Bertha seems to have greater claims to the “genius” plaudit than does Imo.
To return to the question that heads this section, it remains conceivable (indeed, we view it as quite likely) that nonhuman primates exhibit a greater richness of cultural variation com- pared with other taxa. What is striking about the repertoires of chimpanzees is the multifaceted nature of their pur- ported culture, in comparison with which the cultural repertoires of fish and birds appear impoverished (see Boesch21 and Whiten22). Yet we see no reason to value weak evidence for pri- mate cultures over strong evidence for culture in birds and fish. The best ev- idence for culture is found in the spe- cies that are most amenable to exper- imental manipulation.
We would not wish primatologists to forsake the study of population variation in behavior just because they cannot prove that the differences are cultural. How can primatologists get around their inability to carry out translocation experiments? We have two suggestions. First, appropriate data collection would allow the feasi- bility of genetic and asocial learning explanations to be assessed and re- jected if the probability that they can account for the data is unrealistically small. Laland and Kendal54 describe simple statistical methods that would allow alternatives to the social-learn- ing explanation to be dismissed if re- searchers have access to good pedi- gree and diffusion data (see also Fragaszy16). For instance, experimen- tal data from laboratory or captive studies estimating the probability of asocial learning could also be used to assess the probability that a particular pattern of diffusion or level of inci- dence is explicable in terms of asocial processes.
Second, researchers could carry out a careful analysis of the small number of cases in which individuals have dis- persed, transferred between popula- tions, or reintroduced into established populations, or in which populations have been introduced into new areas (see Boesch21 for examples). A careful analysis of these natural translocation
There is no evidence that food washing spread through Imo’s troop by imitation, teaching, or any unusually sophisticated form of social learning. Indeed, there are grounds for concern that it may even have been an artifact of human provisioning.
ARTICLES Evolutionary Anthropology 155
experiments may eventually shed light on primate cultures. This, after all, is one reason that we can be confident that humans have culture.
WHAT (IF ANYTHING) IS UNIQUE ABOUT HUMAN
CULTURE?
If we are to identify unique factors of human culture, then we must com- pare its quantifiable characteristics with purported animal culture on a level playing field. On what grounds can human and animal cultures be distinguished?
Intuitively, the sheer complexity of human behavior seems to divide us from the rest. Driving cars, building houses, and writing articles on culture would be orders of magnitude more complex than the most intricate chim- panzee tool-using behavior. Yet to suggest that what renders human cul- ture unique is its complexity merely begs the question of why human cul- ture should be more complex. Com- plexity is a consequence of the fea- tures of humanity that render our culture unique, and not itself such a feature.
Galef14,15 has suggested that the similarities between animal and hu- man culture may be more superficial than real because animal and human culture rest on different processes. Galef places emphasis on teaching and imitation, which he regards as underpinning human culture but un- important to animal social learning. Conversely, Galef regards purported animal cultures as heavily dependent on local and stimulus enhancement, and other comparatively simple pro- cesses that mediate social learning. Boyd and Richerson55 make a similar argument. While these researchers have not specified precisely how teaching and imitation are deemed to render human culture more stable, complex, or cumulative than pur- ported animal cultures, to many the hypothesis is both intuitive and com- pelling.
Tomasello13 identified three impor- tant characteristics of human culture: universality, the observation that some behavior patterns are exhibited by vir- tually everyone in the society; unifor- mity, or widespread conformity to con-
ventions; and history, manifest in a stable, transgenerational, cumulative culture. He suggested that while chim- panzee behavioral traditions exhibit strong evidence for universality, there was only weak evidence for uniformity or history. Tomasello hypothesized that, compared with chimpanzee “cul- tures,” the greater universality, unifor- mity, and history of human culture are a manifestation of higher fidelity of in- formation transmission among hu- mans, reflecting differences in the psy- chological mechanisms employed. In accord with Galef,14 Tomasello13 ar- gues that human culture alone is reliant on teaching and imitation, together with language and perspective taking.
In reconsidering these hypotheses, we begin with a focus on teaching. We
know of no single widely accepted case in any nonhuman animal that satisfies contemporary definitions of teaching,56 while human children are taught vast amounts of cultural knowledge. At first sight, it is difficult to conceive of a clearer dichotomy in the intellectual abilities of animals and humans. Yet this dichotomy is vulnerable to criticism from both sides. Purported instances of animal teaching are evaluated according to strict criteria. For example, Caro and Hauser56 defined teaching as fol- lows:
An individual actor A can be said to teach if it modifies its behavior only in the presence of a naı̈ve ob-
server, B, at some cost or at least without obtaining an immediate benefit for itself. A’s behavior thereby encourages or punishes B’s behavior, or provides B with experi- ence, or sets an example for B. As a result, B acquires knowledge or learns a skill earlier in life or more efficiently than it might otherwise do, or that it would not learn at all.
The use of this rigorous definition has lead to the view that information transmission in animals is inadvertent and not actively facilitated by the demonstrator. Yet there are other means by which animals actively ad- just their behavior in a manner that facilitates learning in another animal, which do not satisfy definitions of teaching. The opportunity teaching of felids provides one example;57 the food calls of adult callitrichid mon- keys,58,59 the food and alarm calls of mother hens,60 and the stranding of whale calves42 may be others (see also the “stimulating” and “facilitating” of infants by chimpanzee mothers de- scribed by Boesch21). Irrespective of their goal or knowledge of the learn- er’s abilities, the demonstrators ap- pear to function as teachers in these examples. Indeed, leading researchers have called for teaching to be rede- fined in functional terms.61 There seems to be a gradation of teaching- like phenomena, yet this has received little attention from students of ani- mal social learning. As a result, there is no satisfactory classification scheme or terminology to describe an- imal teaching. Just as the reputations of animals have suffered from the treatment of teaching as an all-or- nothing phenomenon, so our own species has benefited. Human teach- ing is not required to satisfy the strict criteria applied to animals, and “teaching” in common parlance when applied to humans is a much broader concept altogether. It would be inter- esting to know what proportion of hu- man teaching actually satisfies Caro and Hauser’s56 definition. Teachers recognize the value for children of op- portunity teaching, and many believe that allowing children to discover and learn for themselves is frequently more productive than more classical forms of pedagogical instruction.
. . . we see no reason to value weak evidence for primate cultures over strong evidence for culture in birds and fish. The best evidence for culture is found in the species that are most amenable to experimental manipulation.
156 Evolutionary Anthropology ARTICLES
(Boesch21 and Whiten22 make similar arguments.)
The same point can be made with regard to imitation, where once again animals are being judged according to a stricter definition than is applied to humans. The fact that humans are ca- pable of imitation is no reason to re- gard it as the sole or even principle process of information transmission underlying human culture. Perhaps children learn how to get around their neighborhood, where the grocery store is, and which locations are dan- gerous by following parents around, like reef fish. Perhaps humans acquire many skills through emulation, like other primates.62 Even when they “imitate,” do humans really copy each other’s exact motor patterns? We sus- pect that “program-level imitation,”63
where the underlying organization of behavior is copied, rather than the surface details, is more common. While we are open to the idea that teaching and imitation may be more important to human culture than to animal social learning, we would like to reserve judgment until humans and other animals are evaluated according to the same standards. It is conceiv- able to us that a careful analysis of teaching and imitation in humans would leave our species looking con- siderably less distinct.
For Tomasello,62 teaching and imi- tation are significant because they are deemed to generate a higher degree of fidelity of transmission among hu- mans than is found for animals. We see two problems with this argument. First, we know of no evidence that there is greater fidelity associated with human than animal culture. Cav- alli-Sforza and coworkers’64 survey of patterns of cultural transmission re- ported average parent-offspring corre- lations for habits, entertainments, and sports to be as low as 0.07, 0.16 and 0.13, respectively. Boyd and Richer- son55 reported parent-offspring corre- lations ranging from 0.34 for attitudes towards feminism to 0.8 – 0.94 for po- litical party affiliation, with values of around 0.5– 0.6 being common. Can we really be sure that humans have higher cultural fidelity than animals? Second, to our knowledge, there is no experimental evidence that imitation results in higher fidelity transmission
than do other forms of social learning. Heyes65 argues compellingly that the fidelity of cultural transmission is in- dependent of the learning processes that individuals employ to acquire a behavior; rather, it depends on the patterns of reinforcement that indi- viduals subsequently experience. As we argued earlier, much of what in common parlance is described as hu- man imitation is more appropriately termed program-level imitation, yet this will not necessarily have high fi- delity.
Moreover, universality and unifor- mity do not seem to be features on which human culture and animal so- cial learning can be distinguished. Field evidence for purported animal culture is provided either by different behavior in two or more populations of the same species (group-typical be- havior) or by an increase in the fre- quency of a novel behavior in a single population (diffusion). In the former case, unless the differences between the populations’ behavioral reper- toires are substantive they are likely either to be unnoticed or to be attrib- uted to random fluctuations, observer error, or ecological differences. It is when behavioral variants are custom- ary in one population and not in oth- ers that a claim of cultural variation is most likely to be put forward. Simi- larly, diffusions typically are identi- fied only when a significant propor- tion of the population has acquired the novel variant. Not all diffusions reach universality in animal popula- tions, but not all human traditions are universal either. The claim for animal culture usually rests on the recording of behavioral variants that are univer- sal, near universal, or at least com- mon within a population.
Mathematical analyses have found that in most circumstances where nat- ural selection favors reliance on social learning, conformity is also fa- vored.55,66 These theoretical consider- ations lead us to expect most animal social learning to be conformist, with individuals commonly adopting the behavior of the majority. This predic- tion has empirical support across a broad array of species, and social learning in which the probability of adopting a pattern of behavior in- creases with the proportion of demon-
strators is reported in animals as di- verse as guppies, pigeons, and rats.67
Conformity may be a characteristic feature of animal social learning, as it is for humans. If, as Tomasello13
notes, chimpanzee populations are characterized by individuality rather than uniformity, then it is chimpan- zees that are “unique.”
A stronger case can be made for history: Much human culture is char- acterized by the “rachet effect,”13 with an increase in the complexity or effi- ciency of technology over time (see also Alvard4). There is little direct ev- idence within animal populations of additive, incremental improvement in behavior or technological advance. Evidence for cumulative evolution in tool manufacture has recently been reported in New Caladonian crows (Hunt and Gray, personal communi- cation). There are one or two primate behavior patterns, such as leaf clip- ping and nut cracking by chimpan- zees, and sweet-potato washing by Japanese macaques, for which a case can be made for racheting,21,68,69 but they remain speculative and conten- tious.
However, before we congratulate ourselves for isolating a distinctive and unique feature of human culture, it is worth reflecting on two points. First, one can look at millions of years of our ancestors’ lithic technologies and many ethnographic studies of tra- ditional peoples without seeing evi- dence for racheting. Oldowan and Acheulean hand-axe technologies lasted without significant changes for over one million years each.70 Cul- tural evolution in other primates could be cumulative at rates equiva- lent to those in early Homo or even preindustrial societies, and we might not notice. Second, the assertion that animals do not exhibit racheting is tantamount to the argument that all nonhuman group-typical behavior patterns could have been invented in their final form by a single innovative individual. Many researchers, partic- ularly those familiar with the complex behavior of apes, find this difficult to believe. Yet it matters little whether nonhuman animals exhibit no rachet- ing, slow ratcheting, or very little ra- cheting: There clearly is a fundamen- tal quantitative difference in the
ARTICLES Evolutionary Anthropology 157
degree to which contemporary human and animal cultures are cumulative.
If purported animal cultures do not exhibit cumulative evolution, then why not? Tomasello’s and Galef’s em- phasis on human’s capacity for teach- ing and imitation provides one sort of explanation, but from the preceding discussion it is apparent that this has only limited support. What is more, explanations that posit the historical emergence of a special character (teaching, imitation, language) as crit- ical to the evolution of human culture do little to stimulate empirical re- search. These explanations betray an underlying assumption that only a unique human characteristic can ex- plain unique human culture. We do not accept this logic. Not that we think capabilities such as teaching, imitation, and language are unimpor- tant to the evolution of human cul- ture, but rather that an emphasis on them puts the cart before the horse. When it comes to the issue of what is unique about human culture, we see two key questions: Why do so few an- imal innovations spread and why is so much human culture stable and trans- generational? While we are not yet able to provide clear answers to these questions, we see no barriers to their empirical investigation, and believe such research would contribute sig- nificantly to understanding of what makes human cultures unique.
Reader (personal communication) found that of 606 cases of learned be- havioral innovation in nonhuman pri- mates only 16% spread to a second individual, and an even smaller pro- portion spread throughout the group. Parallel observations, with many in- novations but comparatively few dif- fusions, have been made in bird pop- ulations.71 This seems to contrast with human innovation, where many, al- though still by no means all, advanta- geous innovations spread.72 What are the barriers to the spread of innova- tion in animals?
First, we suspect that one reason cumulative culture appears rare in an- imals is that there are comparatively few natural circumstances where it would pay an individual to adopt a more advanced means of doing some- thing when a simpler solution works well. A prime question seems to be,
are humans less content to satisfice (that is, make do with a functional but sub-optimal solution to a problem) than other animals? If so, why?
Second, in hierarchical animal soci- eties a disproportionate amount of in- novation appears to be carried out by low-status individuals.50,51 Low-rank- ers often occupy the periphery of the social group, are attended to less than high-ranking individuals, have less in- fluence on group behavior, tend to move away or are behaviorally inhib- ited when approached by higher- ranking group members, and are more vulnerable to scrounging than are high-ranking individuals. All these characteristics would mitigate against the diffusion of innovations generated by low-rankers. Conversely, in more egalitarian societies social learning appears commonly to exhibit confor- mity, and experimental studies have found that conformity makes it less likely that animals will adopt novel behavior.67 Human societies obvi- ously are not more egalitarian or less conformist than animal societies. (In- deed, Henrich and McElreath66 de- scribe conformity and prestige learn- ing biases as characteristic of humanity.) So how are these inhibit- ing effects circumvented? Does lan- guage allow innovators to advertise their inventions? Does trade give them an incentive to do so? Do social con- tracts reduce the likelihood of their exploitation? Again, there are clear opportunities for empirical investiga- tion.
Why is so much human culture sta- ble and transgenerational? Purported animal cultures are largely reliant on short-lived horizontal transmis- sion,73,74 while humans appear to ac- quire large amounts of information from the parental generation.64,75,76
These findings suggest that the lineage leading to Homo sapiens has been se- lected for increasing reliance on verti- cal and oblique cultural transmission. Theoretical analyses imply that a shift toward increased transgenerational culture reflects greater constancy in the environment over time,55 yet there is no evidence that ecological environ- ments have become more constant over the last few million years. Cul- ture may be adaptive for a larger range of conditions in humans than
other animals because humans ex- hibit a greater capacity to construct and regulate those conditions, that is, for “niche construction.”16,77,78 Regu- latory behavior such as tracking game, storing food, and building shel- ters damps out temporal variation in conditions and resources to ensure that cultural behaviors maintain their utility from one generation to the next.77,78 While the capacity for niche construction is universal to living creatures, human niche construction is extraordinarily powerful, in part as a consequence of our culture. Perhaps what is unique about human culture is that, through niche construction, cultural transmission has become self-reinforcing, with transgenera- tional culture modifying the environ- ment in a manner than favors ever more culture.
ACKNOWLEDGMENTS
Research supported by a Royal So- ciety University Research Fellowship to KNL and a BBSRC studentship to WH.
REFERENCES
1 Durham WH. 1991. Coevolution: genes, cul- ture and human diversity. Stanford University Press. 2 Kroeber AL, Kluckman C. 1952. Culture: A
critical review of concepts and definitions. Cam- bridge, MA. 3 Kuper A. 2000. If memes are the answer, what
is the question? In: Aunger R, editor. Darwinising culture: the status of memetics as a science. Ox- ford: Oxford University Press. 4 Alvard M. 2003. The adaptive nature of cul-
ture. Evolutionary Anthropology 12:3:136 –149. 5 Heyes CM, Galef BG. 1996. Social learning in
animals: the roots of culture. London: Academic Press. 6 Shettleworth SJ. 2001. Animal cognition and
animal behavior. Anim Behav 61:277–286. 7 de Waal F. 2001. The ape and the sushi mas-
ter. London: Penguin. 8 Galef BG Jr, Giraldeau L-A. 2001. Social in-
fluences on foraging in vertebrates: causal mech- anisms and adaptive functions. Anim Behav 61: 3–15. 9 Zentall TR, Galef BG Jr. 1988. Social learning:
psychological and biological perspectives. Hills- dale, NJ: Erlbraum. 10 Box HO, Gibson KR, editors. 1999. Mamma- lian social learning: comparative and ecological perspectives. Cambridge: Cambridge University Press. 11 Fragaszy D, Perry S, editors. n.d. Traditions in nonhuman primates: models and evidence. Chicago: University of Chicago Press. In press. 12 Lumsden CJ, Wilson EO. 1981. Genes, mind and culture. Cambridge: Harvard University Press.
158 Evolutionary Anthropology ARTICLES
13 Tomasello M. 1994. The question of chimpan- zee culture. In: Wrangham R, McGrew W, de Waal F, Heltne P, editors. Chimpanzee cultures. Cambridge: Harvard University Press. p 301–317. 14 Galef BG Jr. 1992. The question of animal culture. Hum Nat 3:157–178. 15 Galef BG Jr. n.d. Social learning: promoter or inhibitor of innovation? In: Reader SM, Laland KN, editors. Animal innovation. Oxford: Oxford University Press. In press. 16 Fragaszy D. 2003. Making space for tradi- tions. Evol Anthropol 12:61–70. 17 Macphail EM. 1982. Brain and intelligence in vertebrates. Oxford: Clarendon Press. 18 Reader SM, Laland KN. 2002. Social intelli- gence, innovation and enhanced brain size in primates. Proc Natl Acad Sci USA 99:4436 –4441. 19 McGrew WC. 1992. Chimpanzee material cul- ture: implications for human evolution. Cam- bridge: Cambridge University Press. 20 Whiten A, Goodall J, McGrew WC, Nishida T, Reynolds V, Sugiyama Y, Tutin CEG, Wrangham RW, Boesch C. 1999. Cultures in chimpanzees. Nature 399:682–685. 21 Boesch C. 2003. Is culture a golden barrier between human and chimpanzee? Evol An- thropol 12:82–91. 22 Whiten A, Horner V, Marshall-Pescini S. 2003. Cultural panthropology. Evol Anthropol 12:106 –108. 23 van Shaik C. n.d. In: Fragaszy D, Perry S, editors. Traditions in nonhuman primates: mod- els and evidence. Chicago: Chicago University Press. In press. 24 Perry S, Manson J. 2003. Traditions in mon- keys. Evol Anthropol 12:71– 81. 25 Whiten A, Custance DM. 1996. Studies of im- itation in chimpanzees and children. In: Galef BG Jr, Heyes CM, editors. Social learning in an- imals: the roots of culture. New York: Academic Press. p 291–318. 26 Whiten A, Custance DM, Gomez J, Teixidor P, Bard K. 1996. Imitative learning of artificial fruit processing in children (Homo sapiens) and chim- panzees (Pan troglodytes) J Comp Psychol 110:3– 14. 27 Whiten A. 1998. Imitation of the sequential structure of actions by chimpanzees (Pan troglo- dytes) J Comp Psychol 112:270 –281. 28 McGrew WC, Tutin CEG. 1978. Evidence for a social custom in wild chimpanzees. Man 13:234 – 251. 29 Humle T, Matsuzawa T. n.d. Ant dipping among the chimpanzees of Bossou, Guinea and some comparisons with other sites. Am J Prima- tol. In press. 30 Boesch C, Boesch H. 1990. Tool use and tool making in wild chimpanzees. Folia Primatol 54: 86 –99. 31 Helfman GS, Schultz ET. 1984. Social tradi- tion of behavioral traditions in a coral reef fish. Anim Behav 32:379 –384. 32 Warner RR. 1988. Traditionality of mating- site preferences in a coral reef fish. Nature 335: 719 –721. 33 Bshary R, Wickler W, Fricke H. 2002. Fish cognition: a primate eye’s view. Anim Cogn 5:1– 13. 34 Brown C, Laland KN. 2001. Suboski and Templeton revisited: social learning and life skills training for hatchery reared fish. J Fish Biol 59:471–493. 35 Catchpole CK, Slater PJB. 1995. Bird song:
biological themes and variations. Cambridge: Cambridge University Press. 36 Freeberg TM. 2000. Culture and courtship in vertebrates: a review of social learning and trans- mission of courtship systems and mating pat- terns. Behav Processes 51:177–192. 37 White DJ, Galef BG. 2000. “Culture” in quail: social influences on mate choices of female Coturnix japonica. Anim Behav 59:975–979. 38 Kirkpatrick M. 1987. Sexual selection by fe- male choice in polygynous animals. Ann Rev Ecol Syst 18:43–70. 39 Freeberg TM. 1996. Assortative mating in captive cowbirds is predicted by social experi- ence. Anim Behav 52:1129 –1142. 40 Freeberg TM. 1998. The cultural transmission of courtship patterns in cowbirds, Molothrus ater. Anim Behav 56:1063–1073. 41 Freeberg TM, Duncan SD, Kast TL, Enstrom DA. 1999. Cultural influences on female mate choice: an experimental test in cowbirds (Molo- thrus ater). Anim Behav 57:421–426. 42 Rendell L, Whitehead H. 2001. Culture in whales and dolphins. Brain Behav Sci 24:309 – 382. 43 Kawai M. 1965. Newly-acquired pre-cultural behavior of the natural troop of Japanese mon- keys on Koshima islet. Primates 6:1–30. 44 Wilson EO. 1975. Sociobiology. Cambridge: Harvard University Press. 45 Bonner JT. 1980. The evolution of culture in animals. Princeton: Princeton University Press. 46 Kummer H, Goodall J. 1985. Conditions of innovative behavior in primates. Philos Trans R Soc London, Series B 308:203–214. 47 Visalberghi E, Fragaszy D. 1990. Do monkeys ape? In: Parker ST, Gibson KR, editors. Lan- guage and intelligence in monkeys and apes. Cambridge: Cambridge University Press. p 247– 273. 48 Laland KN, Reader S. 1999. Foraging innova- tion in the guppy. Anim Behav 52:331–340. 49 Laland KN, Reader SM. 1999. Foraging inno- vation is inversely related to competitive ability in male but not in female guppies. Behav Ecol 10:270 –274. 50 Reader SM, Laland KN. 2001. Primate inno- vation: sex, age and social rank differences. Int J Primatol 22:787–805. 51 Reader SM, Laland KN. n.d. Animal innova- tion: an introduction. In: Reader SM, Laland KN, editors. Animal innovation. Oxford: Oxford Uni- versity Press. In press. 52 Day RL, Coe RL, Kendal JR, Laland KN. n.d. Neophilia, innovation and social learning: a study of intergeneric differences in callitrichid monkeys. Anim Behav. In press. 53 Fragaszy DM, Visalberghi E. 1990. Social pro- cesses affecting the appearance of innovative be- haviors in capuchin monkeys. Folia Primatol 54: 155–165. 54 Laland KN, Kendal JR. n.d. What the models say about animal social learning. In: Fragaszy D, Perry S, editors. Traditions in nonhuman pri- mates: models and evidence. Chicago: Chicago University Press. In press. 55 Boyd R, Richerson PJ. 1985. Culture and the evolutionary process. Chicago: The University of Chicago Press. 56 Caro TM, Hauser MD. 1992. Is there teaching in nonhuman animals? Q Rev Biol 67:151–174. 57 Caro TM. 1980. Predatory behavior in domes- tic cat mothers. Behavior 74:128 –148.
58 Roush RS, Snowdon CT. 2000. Quality, quan- tity, distribution and audience effects on food calling in cotton-top tamarins. Ethology 106:673– 690. 59 Roush RS, Snowdon CT. 2001. Food transfer and development of feeding behavior and food- associated vocalizations in cotton-top tamarins. Ethology 107:415–429. 60 Nicol CJ, Pope SJ. 1996. The maternal feeding display of domestic hens is sensitive to perceived chick error. Anim Behav 52:767–774. 61 Shettleworth SJ. 1998. Cognition, evolution and behavior. New York: Oxford University Press. 62 Tomasello M. 1990. Cultural transmission in the tool use and communicatory signalling of chimpanzees? In: Parker S, Gibson K, editors. Chimpanzee cultures. Cambridge: Harvard Uni- versity Press. 63 Byrne RW. 1998. Learning by imitation: a hierarchical approach. Behav Brain Sci 21:667– 721. 64 Cavalli-Sforza LL, Feldman MW, Chen KH, Dornbusch SM. 1982. Theory and observation in cultural transmission. Science 218:19 –27. 65 Heyes CM. 1993. Imitation, culture and cog- nition. Anim Behav 46:999 –1010. 66 Henrich and McElreath, 2003. Evolutionary Anthropology 12: 67 Day RL, Macdonald T, Brown C, Laland KN, Reader SM. 2001. Interactions between shoal size and conformity in guppy social foraging. Anim Behav 62:917–925. 68 Boesch C. 1993. Toward a new image of cul- ture in wild chimpanzees? Behav Brain Sci 16: 514. 69 McGrew WC. n.d. Twenty lessons learned from labors in cultural primatology. In: Joulian F, editor. How the chimpanzees stole culture. Culture and meaning in apes, ancient humans and modern humans. Kluver/Plenum Press. 70 Foley R, Lahr MM. 2003. On stony ground: lithic technology, human evolution and the emer- gence of culture. Evolutionary Anthropology 12: 71 Lefebvre L. 2000. Feeding innovations and their cultural transmission in bird populations. In: Heyes C, Huber L, editors. The evolution of cognition. Cambridge: MIT Press. p 311–328. 72 Rogers EM. 1995. Diffusion of innovations, 4th ed. New York: Free Press. 73 Galef BG Jr. 1988. Imitation in animals: his- tory, definition, and interpretation of data from the psychological laboratory. In: Zentall TR, Galef BG Jr, editors. Social learning: psycholog- ical and biological perspectives. Erlbaum. p 3–28. 74 Laland KN, Richerson PJ, Boyd R. 1993. An- imal social learning: towards a new theoretical approach. Perspect Ethol 10:249 –277. 75 Hewlett BS, Cavalli-Sforza LL. 1986. Cultural transmission among Aka pygmies. Am Anthropol 88:922–934. 76 Aunger R. 2000. The life history of culture learning in a face-to-face society. Ethos 28:1–38. 77 Laland KN, Odling-Smee J, Feldman MW. 2000. Niche construction, biological evolution, and cultural change. Behav Brain Sci 23:131– 175. 78 Odling-Smee J, Laland KN, Feldman MW. 2003. Niche construction. The neglected process in evolution. Princeton, NJ: Princeton University Press.
© 2003 Wiley-Liss, Inc.
ARTICLES Evolutionary Anthropology 159