Studies of animal culture have not normally included a considerationof cetaceans. However, with several long-term field studies nowmaturing, this situation should change. Animal culture is generallystudied by either investigating transmission mechanismsexperimentally, or observing patterns of behavioural variation inwild populations that cannot be explained by either genetic orenvironmental factors. Taking this second, ethnographic, approach,there is good evidence for cultural transmission in several cetaceanspecies. However, only the bottlenose dolphin(Tursiops) has been shown experimentally topossess sophisticated social learning abilities, including vocal andmotor imitation; other species have not been studied. There isobservational evidence for imitation and teaching in killer whales.For cetaceans and other large, wide-ranging animals, excessivereliance on experimental data for evidence of culture is notproductive; we favour the ethnographic approach. The complex andstable vocal and behavioural cultures of sympatric groups of killerwhales (Orcinus orca) appear to have no paralleloutside humans, and represent an independent evolution of culturalfaculties. The wide movements of cetaceans, the greater variabilityof the marine environment over large temporal scales relative tothat on land, and the stable matrilineal social groups of somespecies are potentially important factors in the evolution ofcetacean culture. There have been suggestions of gene-culturecoevolution in cetaceans, and culture may be implicated in someunusual behavioural and life-history traits of whales and dolphins.We hope to stimulate discussion and research on culture in theseanimals.

Human hyperculture appears to have been produced by the amplificationof the kind of normal culture shared by cetaceans and other animalsand presumably by our ancestors. Is there any possibility thatcetaceans could be subject to these amplifying processes, which mayinclude: sexual selection; within-group moral behavior; culling oflow- cultural-capacity individuals through predation orself-predation; and reciprocal positive feedback between culture andthe capacity for culture.

Rendell and Whitehead's thorough review dispels notions that cultureis an exclusive faculty of humans and higher primates. We applaudthe authors, but differ with them regarding the evolution ofcetacean culture, which we argue resulted from the availability ofabundant but spatially and temporally patchy prey such as schoolingfish. We propose two examples of gene-culture coevolution: (1)acoustic abilities and acoustic traditions, and (2) transmission ofenvironmental information and longevity.

Rendell and Whitehead note the necessary, complementary relationshipbetween field and laboratory studies in other species, but concludetheir article by de-emphasizing the role of laboratory findings incetacean research. The ambiguity in field studies of cetaceansshould argue for greater reliance on the laboratory, which hasprovided much of the available research supporting the hypothesis ofcetacean culture.

Social learning in the strict sense is learning by observation orinstruction. Learning by observation appears to be an analogueprocess while learning by instruction is digital. In evolutionarybiology this distinction is currently thought to have implicationsfor the extent to which mechanisms can function successfully as aninheritance system in an evolutionary process.

The sacrilegious proposition of the existence of cultures in whalesand dolphins should open the discussion of cultures in otheranimals, allowing us to find what is unique in human cultures. Theethnographic approach used by all anthropologists is the key in thisinvestigation and revealed that cultural differences are present inanimals and could result from different learning mechanisms.

Genomic conflicts are potentially involved in the evolution andmaintenance of culture. Maternal genes contributing to neocorticaldevelopment could influence biases in the acquisition ofinformation. Specifically, relatedness asymmetries due to multiplepaternity are expected to lead to an increased reliability andreceptivity of matrilineally-transmitted information. This viewcomplements the gene-culture coevolutionary model adopted by Rendelland Whitehead.

Rendell and Whitehead argue persuasively that individual foragingspecializations, if socially learned, are examples of cetaceanculture. However, they discount ecological variation experienced byindividuals within a population as a factor in such behavior. Isuggest that ecological variation may play an important role inindividual foraging specializations and describe several ecologicalparameters that may help us understand the high frequency of thisinteresting behavior in the marine habitat.

The evidence of high cognitive abilities in cetaceans does not standup to close scrutiny under the standards established by laboratoryresearchers. This is likely to lead to a sterile debate betweenlaboratory and field researchers unless fresh ways of taking thedebate forward are found. A few suggestions as to how to do this areproposed.

Twisting definitions of a concept to serve one's purposes is rarely afruitful exercise and the demonstration of culture is, necessarily,a two-step procedure: (1) documenting a behavioral differencebetween populations, (2) demonstrating that this difference hasspread through the group by means of social learning. This willavoid putting the cart before the horse.

While the evidence that cetaceans exhibit behaviours that are everybit as cultural as those recognised in chimpanzees is unequivocal, Iargue that it is unlikely that either taxon has the social cognitivemechanisms required to underpin the more advanced forms of culturecharacteristic of humans (namely those that depend on sharedmeaning).

Demonstrating cetacean communicative cultures requires documentingvocal differences among conspecific groups that are socially learnedand stable across generations. Evidence to date does not providestrong scientific support for culture in cetacean vocal systems.Further, functional analyses with playbacks are needed to determinewhether observed group differences in vocalizations are meaningfulto the animals themselves.

Culture among cetaeceans has important philosophical implications.Three receive attention here. First, these animals are more likehumans than we had previously thought. Even so, we must affirm andrespect their otherness. Second, only a fresh approach to researchmakes this kind of information available. Third, whales and dolphinsshould now be included with us in an extended moral community.

Vocal imitation does not imply an ability to imitate nonvocal motorpatterns. Exponential growth in frequency of a behaviour in apopulation does not imply diffusion by social learning.Distinguishing analogues from homologues of human culture will avoidconfusion in discussion of evolution of culture. Original sources donot demonstrate social learning, imitation, or teaching ofintentional beaching or lobtail feeding in cetaceans.

Human beings grow into cultural knowledge, within a social andenvironmental context, rather than receiving it ready made. Thisseems also to be true of cetaceans. Rendell and Whitehead invoke anotion of culture long since rejected by anthropologists, andfundamentally misunderstand the nature of ethnography. A properlyethnographic study of cetaceans would directly subvert theirpositivist methodology and reductionist assumptions.

The mechanisms for cultural transmission remain disputable anddifficult to validate through observational field studies alone. Ifcontrolled experimental laboratory investigation reveals that aputative mechanism is demonstrable in the species under study, theninferences that the same mechanism is operating in the fieldobservation are strengthened.

Studies on captive dolphins have shown that they are capable ofsocial learning. However, ethnographic data are less conclusive andmany examples given for social learning can be explained in otherways. Before we can claim that cetacean culture is unique we needmore rigorous studies which are fortunately not as difficult asRendell and Whitehead seem to think.

Dichotomous definitions of culture and language do not generateproductive questions. Instead, more progress can be made byidentifying components of each that other animals might plausiblypossess. The evolutionary, ecological approach advocated by Rendelland Whitehead holds great promise for helping us to understand theconditions under which natural selection can favor similarcapacities in differently organized brains.

Cetacean cultural transmission is associated with lengthenedpostmenopausal life histories and relatively stable matrilinealsocial structures. Although Homo erectus was not marine adapted,broadly comparable selection pressures, life history profiles, andsocial structures can be inferred.

Cetaceans are likely candidates for social learning and culture.Meager experimental evidence suggests that some cetaceans possessthe requisite cognitive skills for social learning. Equally sparseethnographic data provide clues about possible outcomes of sociallearning. Although the available evidence is consistent with thenotion of culture in cetaceans, caution is warranted due to the manygaps in the data.