In the Washington Post, for example, Marion Nestle, the Paulette Goddard professor in the Department of Nutrition, Food Studies and Public Health at New York University who supports labeling of genetically modified foods on a national scale, nonetheless told Tim Carman, “It’s weirdly complicated and unclear on key issues: what the controls were fed, relative rates of tumors, why no dose relationship, what the mechanism might be. I can’t think of a biological reason why GMO corn should do this.”
Indeed, biology gets thrown out the window when it comes to fallacious beliefs; the genetic modification in the maize to help them resist Roundup is a protein that is naturally present in all plants that we eat every day. But the study was backed by the Committee for Research and Independent Information on Genetic Engineering, which exists to oppose GM foods. Who is the founder of that group? That same Gilles-Eric Séralini. Note that no anti-GMO group has claimed that the funding source tainted the results in this instance. With mainstream science studies there are claims of black helicopters hovering over labs and lobbyists controlling government scientists but here the agenda seems to be regarded as pure. The agenda is not pure, it is the latest in a series of agenda-based efforts by a researcher out to scare people about genetic modification.
Henry Miller and Bruce Chassy compare Séralini to Duke ESP researcher J.B. Rhine in the 1950s, because Rhine also selectively counted data in his experiments, omitting results because he conspiratorially believed they could not happen by chance. Chassy and Miller are just jumping on the bandwagon to pump up pageviews for Forbes about this study, right? No, they wrote that about Séralini's last paper, in which he claimed Bacillus thuringiensis (Bt) toxin, which has been used for almost 100 years and is still sprayed by organic farmers today, caused 'cell death' when administered in incredibly high levels to defenseless cells. Well, what doesn't cause cell death in a petri dish at high levels? The line between medicine and poison varies, but it is always there. Regardless, Séralini was lauded by the anti-science contingent as a lone voice standing strong against Evil Business while biologists muttered that his experiments are "poorly designed, irrelevant, uninterpretable (but over-interpreted)".
That ESP researcher comparison is apt once again because the paper states "All data cannot be shown in one report and the most relevant are described here". This was science by press release.
Other problems were evident right away; namely the line of rats they chose and the duration of the study. The Sprague-Dawley rats they chose for this study was not by chance, especially considering that Séralini wanted to prove harm and said 90-day studies were the issue in why no harm has been found - so they went out of their way to pick rats bound to get cancer and kept them eating well beyond the 90-day adulthood period for rats. 2 years is a long time for rats but for the specific line of rats they chose, two years is a really long time, especially for rats known to get this cancer, especially when the food is not limited and that is known to be a risk factor for tumors. Such a long time that "for humane reasons this study would not have been given approval in the UK".
Professor Andrew Kniss, biologist at the University of Wyoming, noted that oddity also, and went a step further and took the known cancer rates of this line of rats and made the control group 10,000 instead of the bizarre 10. The simulation "found that 71.4% of them will develop tumors by the end of a 2 year study. That’s pretty close to 72%. But here is where sample size becomes so critically important. If we only select 10 female rats, the chances of finding exactly 72% of them with tumors is much less." Right, if the control group is 10 you could get no rats or 1 or 10, a statistical fluke result. Every biologist noted this same flaw yet anti-science activists insist it makes no difference.
See, GMOs are a pesticide sponge and then corporations sit there in their... in their corporation buildings, and... and, and see, they're all corporation-y... and they make money... photo: Shutterstock
Choosing a tiny control group was brilliant for producing the results they wanted to produce. So it's not a surprise that given the length of time and that species, 50 percent of male and 70 percent of female rats died early compared with only 30 percent and 20 percent in the control group. The control group was statistically silly. If we wanted to use their statistical sleight of hand, it looks like GM foods kept a whole lot of male rats from getting cancer and females broke even.
What was the standard deviation for the data, since some groups that ate GM food got less tumors and some groups got more? No one knows. The kind of undergraduate college-level requirement that would distinguish results from random variation was not done, and they did not show their data so no one else can do it either.
Kevin Folta, who writes here, was quoted in the Los Angeles Times and The New York Times,
"you are likely seeing variation of normal tumor incidence in a small population of rats"In other words, using a custom statistical analysis means they were on what Tom Sanders, head of nutritional research at King's College London, called a statistical fishing trip. Aggravated by the fact that they didn't have as many controls as test subjects. You know you have a problem when the LA Times is debunking your study. Los Angeles is a hotspot of kook beliefs. When I wrote an article for them the first comment was, "That was a refreshing and unexpected read in the Anti-Science LA Times", but they debunk this one thoroughly, with barely a hint of false equivalence. Andrew Revkin at NYT said anti-GM types are engaged in the “single-study syndrome” about dubious research. Prof. Michael Eisen (he has also guest posted here) wrote an email to Revkin that read
This particular study was so poorly designed – the highly sensitized line, the inexcusably small number of animals – that you didn’t even have to look at the ridiculous statements from the lead author (like “GMOs are a pesticide sponge”) to see that it was biased.It's not easy to make Monsanto look like the victim but with overtly biased efforts like Prop 37 and now a study written for the gullible among the anti-science community, activists are making that happen. What you see are scientists on one side and public relations people for anti-science groups on the other. Those are not equivalent and yet advocates side with the public relations groups.
The anti-science groups who invoke funding on every study showing GMOs are safe are conveniently ignoring that this study was funded by a partisan group that exists only to oppose GMOs. All of the methodological errors they would note (if they knew how studies were designed) in a study showing GMOs were safe are allowed to slide by here, and are simply considered more evidence we need warning labels.
They also ignore pesky facts about the rats themselves. As Tim Worstall at Forbes notes, the laboratory rats used likely had GMO food across the board. "Virgin albino Sprague-Dawley rats at 5 weeks of age were obtained from Harlan (Gannat, France)" the latest Séralini study says, but Harlan says "We do not exclude GM materials from rodent diets”. GM food for animals is quite common, even in France.
France is, as has been noted many times, the most anti-science country in anti-science Europe. Without even reading the study, France's Jose Bove, vice-chairman of the European Parliament's commission for agriculture and vocal opponent of any genetic modification, called for an immediate suspension of all EU cultivation and importing of any GM crops.
Dr. Anne Glover, EU Science Advisor, likely did a facepalm of her own over that statement, since she took the job and noted her top priority was to try and make Europeans less anti-science, and now an agriculture vice-chairman is calling for more supernatural beliefs.