The Hopeful Monster of Human Language

A sleeping zebra finch (image courtesy of Margoliash lab)

A sleeping zebra finch (courtesy of Margoliash lab)

One of the repeated themes of the Darwin/Chicago 2009 meeting two weeks ago was the history of the anti-evolution movement, a resistance that has actually changed form, even *cough* evolved, quite a bit since The Origin of the Species. At the opening night event in Rockefeller Chapel, science historian Ronald Numbers talked about differences between the anti-Darwinists led by William Jennings Bryan in the 1920’s (immortalized in the Scopes Monkey Trial and Inherit the Wind) and today’s intelligent design supporters and creationists. Surprisingly, Bryan and his followers were considerably less extreme than today’s anti-evolutionists, as Numbers explained that most who railed against Darwinism in the early 20th century were fine with the evolution of animals over billions of years, they merely could not abide that humans also evolved.

The evolution vs. creation debate has obviously become a lot more complicated since then, but Bryan’s primary objection has lingered – the core of most people’s opposition to evolution is the idea that humans must be somehow separate and different from the rest of the natural world. One “proof” of this uniqueness is the complexity of human language, a form of communication that, to the casual observer, appears in an entirely different league from the songs, gestures, or simple noises that animals use to share information. The assumption that the more complex forms of human language are unique is even held by some in the field of linguistics and psychology, including the legendary Noam Chomsky, who argued as much in a 2002 Science paper with cognitive psychologist (and Darwin/Chicago speaker) Marc Hauser.

That assumption is a handicap to the study of language, argue University of Chicago’s Daniel Margoliash and Howard Nusbaum in a recent issue of the journal Trends in Cognitive Science. The idea that human language is biologically unique, and thus the kind of “hopeful monster” geneticist Richard Goldschmidt coined to describe the sudden appearance of a new feature in evolutionary history, walls off language from the world of biology. Perceiving human language in its proper evolutionary context, and thus exposing it to the tools of comparative biology, will allow scientists to fully understand how language works and where it originated, Margoliash and Nusbaum conclude.

Organismal biology can use the shared ancestry of species to answer questions about humans by conducting experiments in animals. Obviously, this is not a groundbreaking concept – research on lab animals such as rats and mice have been part of science and medicine for centuries. But while it’s generally accepted to study cancer in a rat, studying the syntax of language in a bird and extending those results to humans is a reach for linguists who believe human language to be unique or recently evolved.

“Dan and I have been amazed that even those people who readily acknowledge the importance of evolution in understanding language will restrict the evolution of language to the last few million years,” said Nusbaum, professor and chair of psychology at the University of Chicago. “In essence, by moving the developmental language past apes and saying there are no extant species besides humans who have that capability, that limits the comparative study of language to one species – which makes comparison impossible.”

But the more scientists look at animals that use vocal communication, the fewer items remain on the “human language is unique” list. One example from Margoliash and Nusbaum’s paper is “creolization,” the ability of children to create a new hybrid language from two different languages that still retains normal structure. This phenomenon is used as evidence for Chomsky’s universal grammar, the idea that humans (and only humans) share an innate template for language that is filled in by whatever language they grow up hearing.

But earlier this year, Olga Feher at the City University of New York found that zebra finches raised in isolation – who develop their own unique birdsong – are nonetheless able to tutor younger finches toward a “normal” birdsong. That’s evidence for an innate language template in birds, perhaps similar to the one that helps human children construct a language with normal structure out of erroneous input. Such similarities throw open the doors to further experiments that test the “universal grammar” of humans in a model species where more dramatic manipulations are possible.

“I wouldn’t say that birds have universal grammar,” Nusbaum said. “But we now have a model system to do very fundamental neurophysiological experiments.”

Jumping between species can push the science forward in many different ways, Margoliash and Nusbaum point out from personal experience. In 2000, Margoliash’s lab found that sleep helps zebra finches consolidate song learning from the previous day. That discovery led to a collaboration with Nusbaum’s lab, and the pair published a paper in 2003 with Kimberly Fenn that found sleep helped college students retain perceptual learning of computer generated speech, much like learning to understand a foreign accent.

“It gives us an enormous amount of power be able to look in a different kind of system and understand how sleep plays a neurophsysiological role compared to humans,” Nusbaum said. “Biology seems to find similar solutions to similar problems even when they occur in different species, and that’s information we can use in scientifically understanding¬† something like language.”

About Rob Mitchum (518 Articles)
Rob Mitchum is communications manager at the Computation Institute, a joint initiative between The University of Chicago and Argonne National Laboratory.
%d bloggers like this: