Challenging The New York Times: Is FOXP2 really a "speech gene"?


One of the things I hope to do in this space is facilitate communication between scientists and science writers about how to best describe complex scientific research to the public. After hearing some concern from University of Iowa neuroscientist and Behavioral Neuroscience Editor-in-Chief Mark Blumberg about Nicholas Wade's recent New York Times story, "Speech gene shows his bossy nature," I invited Blumberg to submit an open letter to Wade. Here it is, along with Wade's response:

Dear Nicholas Wade, 

I'm very sympathetic to the terrific challenges you face in making new scientific research appealing and digestible to the public. But I have some specific concerns about your latest report on the FOXP2 gene, beginning with the headline, "Speech gene shows his bossy nature."

Can we really still call FOXP2 a "speech gene"? As you know, this FOXP2 mutation was originally identified in a London family, many of whose members exhibited profound language impairments. From that single observation, it became known as the "speech gene." But there was always the distinct possibility that the mutation influenced a myriad of other brain and body functions that, in turn, affected speech. Indeed, given all that we know about how genes work - as well as our sad history with grandiose claims about single-gene effects on behavior - wouldn't it be wise for all of us to be more cautious when communicating these findings to the public? As for people with FOXP2 mutations, a well-informed colleague has told me that they do indeed exhibit a variety of problems beyond those related to speech, just as we would expect. I fear that these other problems have not been adequately studied precisely because they detract from the preferred "speech gene" narrative.

As to its "bossy" nature, you write that FOXP2 "does not do a single thing but rather controls the activity of at least 116 other genes." That's true, but let's put it in context. As you know, such distributed effects are nothing new; genes are always part of complex networks and therefore are hardly ever expected to do a single thing. Thus, FOXP2 is part of a large, complex network of genetic and non-genetic factors that, under the right developmental conditions, appear to contribute to the human faculty for language - and a lot of other things as well. In fact, many studies have now shown conclusively that mice with the FOXP2 mutation exhibit changes in a myriad of organ systems, including the lung and brain. And yet FOXP2 is called a "speech gene" rather than a "lung gene" or a "brain gene."

I suggest that the better alternative is to describe FOXP2 in less dramatic terms - which you do very nicely when you write that "the whole network of [language-related] genes has evolved together in making language and speech a human faculty." It's frustrating, then, to read references in the same article to a simplistic and outmoded view of gene action - for instance, when you write of "genes under FOXP2's thumb" and FOXP2 as "a maestro of the genome." The new Nature findings actually portray a more sober view of FOXP2's powers. 

This is not the first time that you have written about FOXP2 in the Times. Last May, you wrote an article entitled "A human language gene changes the sound of mouse squeaks." The subject of your piece was another scientific article, this one published in Cell, that reported on changes to brain and behavior in mice engineered to express the human version of the FOXP2 gene. One of the authors of that paper is quoted by you as promising that "We will speak to the mouse." An extraordinary promise coming from a scientist, don't you think?

It was the link to human language that garnered that mouse study so much acclaim. And what did they find: that the "humanized" infant mice emit vocalizations of a slightly lower pitch than typical infant mice. Having researched similar vocalizations in rats for many years, I knew before reading the Cell paper that the findings almost certainly had nothing to do with human language. In fact, any manipulation that alters the body size or respiratory system or larynx or a host of other factors in these animals could account for the small change in pitch of the mouse vocalizations. Given FOXP2's influence on so many organ systems, it would have been astonishing if their vocalizations had not been affected.

Trumping up FOXP2 as yet another star gene in a series of star genes (the "god" gene, the "depression" gene, the "schizophrenia" gene, etc.) not only sets FOXP2 up for a fall; it also misses an opportunity to educate the public about how complex behavior - including the capacity for language - develops and evolves.


Mark S. Blumberg, Ph.D.
F. Wendell Miller Professor, Department of Psychology, University of Iowa                      
Editor-in-Chief, Behavioral Neuroscience



Dear Mark,
I'm a little puzzled by your complaint, which seems to me to ignore the special dietary needs of a newspaper's readers and to assume they can be served indigestible fare similar to that in academic journals. 
You question whether FOXP2 can be called a speech gene and you suggest it could equally well be called it a lung gene.  But language is more interesting to most people, scientists included, than is lung function. It's because of FOXP2's connection with language that so many labs are working on it. So I cannot see any reasonable objection to calling it "a gene that underlies the faculty of human speech."
The role of this article was to update readers on a new finding, not to review the history of ideas about FOXP2. So there's no space to go into the argument about the gene's precise involvement with speech and language, much of which we have covered in earlier articles.
I won't comment on the headline on the article - reporters don't write headlines and are generally not consulted about them.
I don't see what's wrong in calling FOXP2 a "maestro of the genome," a phrase that would apply to many transcription factors. Yes indeed the gene is expressed in several other tissues besides the brain. But I had 550 words in which to set the story up in non-technical language, explain why it was interesting, and give general readers a flavor of what the researchers had found. There was simply no space for the qualifications you mention and they are not essential to the story.
You cite an earlier article about the mouse which Svante Paabo genetically engineered to carry a human FOXP2 gene. Then you ask if I didn't realize that Paabo was making "an extraordinary promise" in saying "We will speak to the mouse." Well, no, I didn't - I thought it was obvious he was making a joke. He's surely implying the mouse is rather unlikely to speak to him.
Your view is that Paabo's paper on his FOXP2 mouse was of little interest, and it's true that he and Wolfgang Enard only found a large number of rather subtle changes, including slightly different isolation whistles. But I think most people would say the experiment was important and needed to be done, even if we don't really understand yet what all the changes mean. That's why I thought it was worth writing up.
I don't understand your complaint that FOXP2 is being given star treatment. It's in the limelight because it's a really interesting gene that may provide the entryway to a major human faculty. If it fails to do so, we'll write that up too. Are you suggesting we should tell our readers nothing about FOXP2 for the next 10 years until we have a definitive answer? - That's the role for encyclopedias and review articles.
As for missing an opportunity to educate the public, that, with respect, is your job, not mine.  Education is the business of schools and universities. The business of newspapers is news.
Nicholas Wade 
Reporter, The New York Times
Author, The Faith Instinct