31 May 2009
Newspapers are extraordinarily bad at reporting on scientific findings. A number of factors conspire to make this so. Reduced budgets force reporters to rely on press releases and flacks for story ideas and information—which leads to the printing of sensationalist claims. Reporters don’t have the time (or the scientific chops) to fully understand the implications of a story. Editors want a “hook” that makes a story interesting, even if that hook isn’t real. Journalists don’t do research—they do interviews; which results in the stories being based on arguments from authority, considered to be one of the weakest forms of argument. And newspapers have a bias for stories that contain conflict; they often inflate (or even manufacture) opposing views to create this conflict.
Reporting on linguistics is no different, except NY Times reporter Nicolas Wade has a penchant for getting stories about advances in linguistics horribly wrong. Getting one story wrong is one thing, but he’s been doing it for years. One would think the journalistic standards of the “paper of record” would be better, but evidently not. They continue to print his ill-informed articles on linguistics.
The latest article in question is here. It concerns the FOXP2 gene, the so-called “language gene.” The trouble is, the gene is no such thing. It is not a magical gene that enables people to speak. Yet in Wade’s article he is quoting researchers in Germany who have successfully replace mouse FOXP2 with the human version of the gene saying that their goal is get mice to speak. Wade couches the story in terms of a future where you will be able to talk to your pet. To be fair to Wade, it’s the researchers who got the ball rolling by overstating the implications of their research. They found that the mice with the human gene squeaked differently than mice without it, and conclude that the gene “might have been important for the evolution of speech and language in humans.” The significance of the research is now we have mice that can be used to explore exactly how FOXP2 may impact language development in animals and humans. It’s an instrumental step that will enable further research.
FOXP2 is not a “language gene.” It is a transcription factor that regulates the expression of other genes during an animal’s development. The effects of a malfunctioning FOXP2 gene are wide ranging, but most are centered on motor coordination. There are also some impacts on brain development that are less well understood.
The “language gene” myth got its start in 1990 when a London family was discovered that had a malfunctioning version of the gene. Those with the gene had difficulty articulating speech. The problem was largely one of motor coordination--they couldn’t manage the mouth and tongue movements needed to speak English understandably. But they could be taught sign language, demonstrating that the innate capacity for language was unaffected. Some members of the family also problems learning and using grammar—their basic language functions were impaired—but they also had significantly diminished IQ and their brain development was impaired in general. The problem was not limited to language. The changed squeaks in the mice are more likely the result of changed motor function, causing them to squeak differently, than any increased linguist function.
So sticking a human FOXP2 gene in an animal isn’t going to enable animal speech. Language is an extraordinarily complex trait and is undoubtedly governed by a host of genes. The problem is that Wade should know this. Back in 2005(!), linguist Geoff Pullam over at Language Log thoroughly debunked a similar article by Wade on FOXP2.
Maybe there’s some malfunctioning gene carried by newspaper reporters that doesn’t allow them to learn from past mistakes. Let’s hope not. And let’s hope that the next time Wade gets a bug to write a story about linguistics, he actually talks to some linguists before going to press. That’s not really too much to ask.
[Hat tip to Language Hat for alerting me to the issue.]
