These are exciting times for the research in language evolution! Never before have we been so close to solving this evolutionary puzzle, and you can also be part of this exciting moment in science history!

It has been suggested that many human-specific traits, including language, evolved as humans found new ways of controlling their environment. Examples include developing collective ambushing strategies to hunt for stronger animals; showing altruistic and inclusive behaviors towards the elderly, children, and disabled individuals, minimizing the fitness consequences of their vulnerability; and taking part in traditional learning and division of labor, which allowed for the development and optimization of agriculture, tool-making, and hunting, among other fundamental human activities.

Our ability to manipulate the environment to our own advantage would have developed as we evolved a more gregarious lifestyle, ultimately avoiding many sources of natural selection, such as predation and foraging efforts. In this context, human sociality would have entered a feedback loop, reducing common environmental selective pressures, and making itself the main source of selection in our species. Individuals showing more tolerance to social stress, and more cooperative instead of aggressive behaviors toward each other would have been granted selective advantages, and a number of the unique traits that define our species would have evolved as direct or indirect effects of these selection pressures for peaceful and cooperative living.

This process has been termed “self-domestication” because, in addition to the selection for traits that breeders desire their organisms to evolve, one of the salient features of domestication is the replacement of commonly found environmental sources of selection by sources of selection that are imposed by socialization with individuals of a same or different species.

The relaxation of natural selection allowed the domesticated species to evolve a so-called “domesticated phenotype”, which is characterized by a suite of traits mostly linked to the retention in adulthood of traits typically found at younger ages in the ancestor species, a phenomenon called “neoteny”. For example, domesticated animals tend to have smaller teeth, jaws, and faces, and are less aggressive than their wild ancestors - all of which are juvenile traits. Interestingly enough, these are all traits that humans show when compared to reconstitutions of hominid ancestors.

Most subscribers of the self-domestication hypothesis agree that relaxation of natural selection would have freed us from constraints limiting our behavioral repertoire, and allowed for the evolution of greater behavioral flexibility and complexity. This means that instead of having specific behaviors innately “wired in”, humans could learn complex new behaviors from personal experience, and by observing others. One of these complex learned behaviors is the human language. If analyzed within the framework of the self-domestication hypothesis, the human language would have evolved following a relaxation of environmental pressures that demanded stereotyped vocalizations, possibly as a result of one or more of the following factors: cooperation in order to cope with environmental sources of selection (for example, predation), dependency on another form of communication (for example, gestural communication), and increased intelligence.

Unfortunately, language itself does not fossilize, and to study its evolution, we have to investigate what we know about today’s human language and other similar communicative systems in nature. Here at the Brain Evolution Laboratory of the Department of Anthropology at UC Berkeley, we chose to investigate the evolution of human language evolution by turning to comparative animal research. While the study of communicative systems of other primates, such as marmosets, gibbons, and chimpanzees, has proven to be a valuable tool in the study of human language evolution, these animals are challenging to study for ethical as well as practical reasons. In contrast, songbirds have provided a highly feasible animal model in the field.

[caption id="attachment\_12986" align="aligncenter" width="624"] Madza Y Farias-Virgens (right) and Yevgeniya Sosnovskaya (left) - Brain Evo Lab – Department of Anthropology – UCBerkeley[/caption]

Sing-song bird

Even though birdsong is a simpler communication system than human language, obvious parallels exist between birdsong and human language at many levels. Like a child learning to speak, a songbird must first hear the vocal sounds of adults, and then hear its own voice when learning to imitate those sounds. In the beginning of this learning phase, baby birds sing a very faint, unstructured song, akin to babbling in human infants. This immature chirping gradually develops into a more stable birdsong during adulthood, in the same way that babbling develops into coherent words. There are also remarkable parallels between how bird brains process song and how human brains process language. A consistent set of around 50 genes show similar patterns of activity in the respective brain regions responsible for human language and birdsong. This means that songbird model systems can be developed for studying human language evolution, and we believe to have found the ideal species for the job. Our model system is composed of two strains of the same songbird species: the domesticated Bengalese finch and its wild ancestor, the white-backed munia.

Although never bred for its singing abilities, the Bengalese finch has evolved a more flexible song than its wild relative. White-backed munias have a simple song due to pressures to avoid confusion with other cohabiting finch species and efforts to decrease predation risk, as simpler songs are more easily recognized by other birds of the same species, without attracting the attention of predators of a different species. It has been proposed that the freedom from these constrains has left the female preference for more complex songs as the main selective force in captivity, leading the Bengalese finch to develop a more complex song than its wild ancestor.

White-backed munias also have higher levels of the stress hormone, leading to more vigilant behavior when compared with the Bengalese finches. This has been proposed to limit opportunities for learning complex songs. Without the danger of being preyed upon or the effort of having to look for food, Bengalese finches had the freedom to vary their melodies, which could have given more opportunity for greater song complexity to emerge. The breeder’s direct or indirect selection for more relaxed birds that would have a higher tolerance to confinement and social stress is another item to factor into this equation. This selection would have led to the selection for more juvenile individuals, which could learn more variable and complex songs.

In our work, we propose to test the hypothesis that the consistent relaxation of selection resulting from the Bengalese finch domestication has played an essential role in allowing this bird to evolve a more flexible and complex song. To test our hypothesis and its predictions, we intend to investigate how gene function changes in the brains of these two bird strains, and examine the birds’ genome sequences in attempts to find signals of selection or lack thereof. Once these patterns of gene activity and evolution are known for this songbird system, the same patterns can be investigated in studies comparing humans to other primates. This work would also allow us to better examine the self-domestication hypothesis, which has been investigated mainly by searching for common traits between humans and domesticated species (e.g. dogs, birds, foxes, etc), and by analyzing the evolutionary forces and trade-offs underlying the occurrence of those traits.

Darwin was among the first people to argue for parallels between birdsong and human language. In his book The Descent of Man, he claims that “the sounds uttered by birds offer in several respects the nearest analogy to language". Was he right? So far, the evidence points to “yes”- but to which degree can the evolution of birdsong teach us about the evolution of language in our own species?

Help us discover the answers at the Brain Evo Lab!

Madza Y Farias-Virgens and Yevgeniya Sosnovskaya - Brain Evo Lab – Department of Anthropology – UCBerkeley

Title image courtesy Lip Kee Yap, CC License 2.0