Skip Navigation
Language from Scratch

Language from Scratch

According to Derek Bickerton of the U of Hawaii, the convergent evolution of creole languages permits us a window into the "default settings" of human speech.

Kevin Matthews Email KevinMatthews

It's conceivable to think of an intelligent species whose language was programmed down to the wire...as a thought experiment.

In fact, drunks are the world's most underrated language teaching resource.

Never write "research" on an immigration form. No matter what you're doing, just write "tourist."

THE NUGGETS of advice are enough reason to consult Bastard Tongues: A Trailblazing Linguist Finds Clues to Our Common Humanity in the World’s Lowliest Languages (2008), the career memoir of Derek Bickerton, well-traveled emeritus professor of linguistics at the University of Hawaii. But it's the theory that's exciting for newcomers. The theory puts "lowly" creole languages right at the center of understanding the human capacity for language.

In Bickerton's use of the term, creoles are languages that develop out of "strongly reduced input." Most typically this happens when a pidgin tongue used by plantation laborers who are enslaved or otherwise thrown together in a new geography is about all that children have to work with in developing their own speech. Wired to do more with words, the children won't accept the pidgin, a language of last resort with few words and chaotic syntax, as a finished mother tongue.

Instead, as Bickerton explained on April 11, 2008, at a book tour stop cosponsored by the UCLA Department of Linguistics and the Center for World Languages, they develop the rule-bound grammars of creole languages. Most curiously, according to Bickerton, they develop a lot of the same rules from Suriname to Equatorial Guinea and back—"a consistent core of grammar" that lays itself bare precisely where linguistic input is most restricted. Bickerton believes that this set of common "default settings" for grammar is too much for coincidence and must have an innate source, which he refers to as the "bioprogram" of human language.

He told the UCLA audience of about 60 that creole languages "give us insights into the nature of the human language faculty that we don't necessarily get from other languages." The shift from pidgin to creole differs from the regular, sometimes glacial processes of language change because creole languages, even if completely isolated from one another, converge on a set of features.

Or they tend to, and according to Bickerton. He first presented a version of his language bioprogram theory in 1975 at a conference in Hawaii for creole specialists, drawing on his own data and on Noam Chomsky's revolutionary ideas about language acquisition. The attendees promptly squashed the idea and have not stopped, while he has stood by it, he writes.

Bickerton sounds particularly baffled by a misinterpretation among some of his critics that the bioprogram is not only a blueprint for building grammar but also a photograph of the completed structure. Objections in the form of counterexample—this acknowledged creole lacks stacks of "serial verbs," those two model their interrogative sentences on French, and so on—are not sensitive to the effects of environment on any biological process, he suggested at the event.

Whatever the shortcomings of the bioprogram model may be, Bickerton made clear that it is not a code containing grammar's laws. As a set of grammatical defaults, the bioprogram resides well within the frontiers of linguistic variation that Chomsky and followers have discussed under the rubric of "universal grammar." (Universal grammar refers to a larger "envelope" containing divergent rules, Bickerton said.)

When an audience member asked why divergent grammars arise at all, why in other words we aren't bound more tightly to the bioprogram, Bickerton said that people are simply too creative for that. At this lecture, he did not offer any fuller account of linguistic divergence.

"It's conceivable to think of an intelligent species whose language was programmed down to the wire…as a thought experiment," he explained. "That is not the way we're constituted."

Whether he turns out to be right about the innate features of language use, Bickerton in his memoir points towards fascinating questions on the environmental side. For example, he reports that while reading transcripts of conversations he participated in with pidgin speakers, he finds himself unable to grasp utterances that he must have understood in the moment. He explains: "so much depends, where pidgin is concerned, on where you are, who's there, what you've been talking about, what things were visible in your surroundings, and I don't know what else, though I'd like to exclude weather and the time of day."

Linguists: may we have further research regarding "what else" and "I'd like"?

To print this page, select "Print" from the File menu of your browser.