A group of heterodox cultural anthropologists have begun examining human cultural change through the lens of Darwinian evolution. Alex Mesoudi and colleagues offer this encouraging preliminary assessment of the field:
Cultural traits go extinct as a result of competition, as occurred with the gun in Japan (Perrin 1979) and with bone tools in Tasmania (Diamond 1978). Basalla (1988) amassed extensive historical evidence for the gradual accumulation of modifications through time, such as Joseph Henry’s 1831 electric motor, which borrowed many features from the steam engine, or Eli Whitney’s 1793 cotton gin, which was based on a long line of Indian devices…
Culture exhibits the convergent evolution of similar forms in unrelated lineages, such as the tendency for both teddy bears (Hinde & Barden 1985) and cartoon characters (Gould 1980) to become increasingly neotenous over the course of time. Finally, cultural traits change in function or become vestigial, as documented by Basalla (1988) for numerous technological artifacts, such as the no longer optimal QWERTY keyboard layout…
Other evolutionary archaeologists have adapted neutral drift models from evolutionary biology (e.g., Crow & Kimura 1970) to account for “stylistic variation” in artifacts. For example, Neiman (1995) demonstrated that changes in decorative styles of Illinois Woodland ceramics can be predicted by a model incorporating the selectively neutral but opposing forces of drift and innovation…
Howe et al. (2001) describe how different manuscript versions of the same text can be used to reconstruct the evolution of that text. This was demonstrated by Barbrook et al. (1998), who used cladistic methods to reconstruct the historical relationships between 58 different manuscripts of Chaucer’s Canterbury Tales, improving on previous non-phylogenetic reconstructions. Bentley et al. (2004), meanwhile, found that the frequencies of first names and patent applications in twentieth-century United States both conform to a simple model of random copying originally developed in evolutionary biology…
For example, McKone and Halpern (2003) developed a population genetic model of androgenesis, a rare phenomenon seen in freshwater clams, Saharan cypress trees, and stick insects, where the offspring acquire nuclear DNA from the male parent only. The model predicts that mutations causing androgenesis will often spread rapidly to fixation in an initially non-androgenetic population, and in some cases cause extinction of that population because of the loss of females, perhaps explaining the rarity of such mutations.
Feldman and Cavalli-Sforza (1989) modelled the coevolution of genes for lactose absorption and the cultural trait of dairy farming, finding that the allele for lactose absorption will spread provided there is a high probability that the offspring of dairy farmers themselves become dairy farmers, but not otherwise, even with a significant viability advantage. Another case study examined the evolution of handedness (Laland et al. 1995b), proposing a model that gave a better fit to patterns of handedness in families and among twins than do leading purely genetic models…
Boyd and Richerson (1985), meanwhile, have developed a model of runaway cultural selection similar to runaway sexual selection, which they argue can account for a range of cultural traits, from oversized yams in Ponapae to extensive tattooing in Polynesia (paralleling elaborate sexually selected biological traits such as the peacock’s tail)…
…research in sociology on the diffusion of innovations (Rogers 1995) examines how new ideas and technologies are transmitted through naturally occurring populations. Typically, questionnaires or interviews are employed to assess the past and present use of the innovation by the respondent, and used to compile a picture of diffusion through the population. Classic studies have examined the diffusion of new types of seed among farmers (Ryan & Gross 1943) and antibiotic among doctors (Coleman et al. 1966). A recurring finding from more than 3,000 diffusion studies is an S-shaped cumulative adoption curve (Rogers 1995), which indicates a slow initial uptake, followed by a rapid increase in adoption, and finally another slow period as the population reaches saturation (similar sigmoidal dynamics characterise the diffusion of alleles)…
Many of these diffusion studies, however, can be criticised for not clearly identifying a priori the putative selection pressure responsible for the diffusion and then testing this prediction in natural populations, as is commonly done in evolutionary biology (Endler’s first method)… One recent study that did specify a priori a hypothesised selection pressure is Bangerter and Heath’s (2004) study of the “Mozart effect,” the idea that exposure to classical music enhances intelligence, especially during childhood. Although it has very weak scientific support, this idea has gained wide currency in the United States mass media, which Bangerter and Heath (2004) hypothesised was because it offers a cheap and easy way of supposedly enhancing one’s child’s development. This hypothesis predicts that the Mozart effect should be more prevalent in the mass media of states where there is poor academic performance and low spending on education, which Bangerter and Heath showed to be the case…
[C]ultural traits may adapt to physical features of the environment such as temperature or rainfall. Cultural traits may also compete with and adapt to other cultural knowledge, equivalent to the biotic environment. Finally, cultural traits may adapt to biologically evolved or implicit features of human cognition. This has no exact equivalent in the biological world, although perhaps there is a loose parallel in genetic or developmental constraints on adaptation, or in the coevolution of symbionts and hosts (Dennett 2001; 2002)…
The practical methods of human behavioural ecology… involve observing and recording behaviour in natural environments… guided by the predictions of formal mathematical models… An example is the occurrence in Tibet of polyandry, which has been shown to be functionally adaptive under the particularly harsh environmental conditions of the region (Crook & Crook 1988)…
[B]ecause cultural traits rely predominantly on human minds for their storage and transmission, there is the possibility of adaptation to biologically evolved or developmentally acquired cognitive features of those minds… Evolutionary psychology (e.g., Barkow et al. 1992) also provides a rich theoretical and empirical body of research on biologically evolved features of human cognition that might be predicted to bias cultural transmission in particular directions. Sperber and Hirschfeld (2004) similarly argue that the diversity of some cultural traits, such as religious beliefs or classifications of animals and plants, is the result of adaptation to biologically evolved domain-specific cognitive capacities (e.g., folk biology; Atran 1998). There is also evidence that biomechanical properties of the human vocal apparatus significantly constrain the form of words (MacNeilage & Davis 2000). The infant vocal tract, for example, favours simple consonant-vowel alternations such as “dada” and “mama” (MacNeilage & Davis 2000), which may explain why such word forms are used in many languages to denote parents…
It may also be fruitful to study the adaptation of cultural traits to alternative transmission media such as printed documents or the Internet, and to examine whether such media are merely direct extensions of cognitive capacities or whether they generate their own novel transmission constraints (see Donald 1991)…
In order to illustrate the universality and substrate-neutrality of his replicator-centred theory of evolution, Dawkins (1976) coined the term meme to describe a cultural replicator, or a unit of cultural transmission. Memetics has been developed further by Hull (1982), Dennett (1995), Blackmore (1999), and Aunger (2000b; 2002), amongst others…
A more detailed picture of the mechanisms of cultural transmission requires an understanding of how the brain processes relevant information…
Aunger (2002) has recently attempted to integrate memetics with neuroscience, arguing that a robust conceptualisation of the “meme” must specify its material basis in the brain. He proposes that memes should be seen as electrochemical states of multiple neurons, and offers a definition of a “neuromeme” as “a configuration in one node of a neuronal network that is able to induce the replication of its state in other nodes” (p. 197). As Aunger acknowledges, however, any attempt to provide a more detailed description and theory of a neuromeme is severely limited by the current lack of understanding within neuroscience concerning the precise neural and molecular basis of how learned information is stored in the brain…
[T]here are several reports of category-specific naming impairments of human patients with brain damage who have been found to recognise and correctly name all items except those in a specific category, such as fruits and vegetables or country names (Crosson et al. 1997). Such studies suggest that at least some learned knowledge stored in human brains is organised into separable semantic categories. There is also recent evidence that single neurons in the human medial temporal lobe respond to the higher-order abstract representation of a specific person or building (Quiroga et al. 2005)…
[T]here is evidence from a number of species of behavioural traditions not obviously attributable to genetic or ecological differences, and hence thought to constitute socially learned cultural patterns. For example, Whiten et al. (1999; 2001) documented 39 putative cultural traits in chimpanzees from various regions of Africa, such as tool usage and grooming behaviour. Similar regional differences inferred to be cultural in origin have been observed in orangutans (van Schaik et al. 2003) and capuchins (Fragaszy & Perry 2003; Perry et al. 2003a), as well as in the vocalisations of birds (Catchpole & Slater 1995) and mammals (especially cetaceans: Janik & Slater 1997), and behavioural traditions in fish (Helfman & Schultz 1984; Warner 1990). There are obvious parallels here with the databases compiled by cultural anthropologists documenting worldwide geographical variation in human culture…
[A] number of studies have tracked the diffusion of innovations within nonhuman communities, paralleling the research tradition of the same name for human technology (Rogers 1995). The most famous case is the diffusion of potato-washing in a community of Japanese macaques (Kawai 1965). Lefebvre (1995) found that 16 of 21 reported cases of the diffusion of foraging innovations in primates exhibit a rapid accelerating pattern of adoption characteristic of cultural transmission (an example of which is the S-shaped distribution reported by Rogers 1995)…
[P]opulation genetic modelling has been used to analyse patterns of nonhuman culture, specifically birdsong. Lynch and Baker (1993) found that the geographical distribution of chaffinch songs can be accounted for by a neutral model in which mutation, migration, and drift are at equilibrium. Lachlan and Slater (1999) adopted gene-culture coevolution methods to find that vocal learning can be maintained in a “cultural trap” formed by the interaction between genes (which specify the constraints on songs) and culture (the songs themselves). Gene-culture coevolutionary methods have also been used to explore how song learning might affect speciation (Beltman et al. 2004) and the evolution of brood parasitism (Beltman et al. 2003).