When a 2012 study came out suggesting that a certain type of genetically modified corn caused cancer in rats, many were skeptical. Since then, one scientific group after another has said that the study doesn’t tell us anything new. So on one level it was no huge surprise when the journal that had published this paper, Food and Chemical Toxicology, retracted it on Thanksgiving Day. But it was surprising, or at least illuminating, on another level: Retractions are usually reserved for deliberate deception or major mistakes; in this case, the reason for retraction was simply insignificance.
In a statement the journal publishers wrote: “Ultimately, the results presented (while not incorrect) are inconclusive, and therefore do not reach the threshold of publication for Food and Chemical Toxicology.”
What does it mean that a “not incorrect” but “inconclusive” paper fails to “reach the threshold”?
First, a quick review of the study and the blowback. A team led by Gilles-Eric Séralini fed GM and conventional corn to rats (and spiked the water of other rats with the Round Up herbicide glyphosate) for two years. At the end of that period grotesque tumors bulged from many of the rats. There were more tumors and deaths in certain groups of rats fed on GM corn and glyphosate.
The critiques came from all over. You can read those yourself if you want to dig into the details of the controversy, but basically it boils down to this: The rats used in the experiment get lots of tumors if you let them live long enough. As much as 87 percent of female rats develop tumors under normal conditions. Enough tumors and deaths occurred among the 10 control rats eating conventional corn that the higher levels of disease seen in the other rats could have occurred by chance.
Amid all the uproar, the editor of Food and Chemical Toxicology asked Séralini for the raw data and began a thorough review. Here’s what happened, according to the publisher’s statement:
Unequivocally, the Editor-in-Chief found no evidence of fraud or intentional misrepresentation of the data. However, there is a legitimate cause for concern regarding both the number of animals in each study group and the particular strain selected … A more in-depth look at the raw data revealed that no definitive conclusions can be reached with this small sample size.
In other words: The study was just not good enough.
Which brings us back to the question: Should “just not good enough” be grounds for a retraction? Ivan Oransky, who monitors scientific retractions at the blog Retraction Watch, wrote:
[Many] scientists say that retraction should be reserved for fraud and serious error. Does that hold for a paper that many criticized as deeply flawed — and which challenged GMOs, whose use is supported by many scientists?
I suspect this decision has a lot more to do with messy public relations than with science. From a strictly scientific perspective, the Séralini study isn’t a big deal: It’s just one of the many safety trials that researchers have performed on GM foods; it amounts to little when considered as part of the whole picture. The way scientists usually deal with this sort of finding is to ignore it. As Séralini’s team pointed out in a response to the retraction [doc], many other studies might be disqualified if they were held to the same standard.
From a public-relations perspective, on the other hand, the Séralini study is a huge deal. To people who aren’t familiar with the larger body of research, and who mistrust GMOs, it looks like proof that GM food kills. It produced disturbing images of tumorous rats that continue bouncing around social media to this day. The retraction provides an easy rejoinder to that sort of thing. As in, “You know that study was retracted, right?”
In this case, the retraction was not “reserved for fraud and serious error,” as Oransky puts it. Instead, it was deployed when an insignificant study got outsized attention. Perhaps the most interesting thing to observe here is that the cold hard rules of the scientific world bent in response to the messy realities of the real world.
P.S.: One final note on a rumor flying around about this. Séralini’s team alleged that the retraction was triggered when the journal appointed the scientist Richard Goodman, who once worked for Monsanto, to its editorial board. Goodman said he had nothing to do with it and is only involved with the journal from afar. Nature talked to both sides.