Scientific publishing is in crisis. A number of empirical studies have shown that 80 to 90 percent of the claims coming from scientific studies in major journals fail to replicate. This is alarming, and the problem is only likely to become worse with the proliferation of “predatory publishers” of many open-access journals (which anyone can read online without a subscription fee). According to Gina Kolata writing in the New York Times in 2013, the journals published by some of the worst offenders are nothing more than cash-generating machines that eagerly, uncritically accept virtually any submitted paper as long as the authors pay a hefty publication fee.
Another trend, related and equally worrisome, is the increasing frequency of publication of the results of flawed “advocacy research” that is designed to give a false result that supports a certain cause or position and can be cited by activists long after the findings have been discredited. The articles are often found in the predatory open-access journals.
The application of modern molecular-genetic-engineering techniques to agriculture has provided fertile ground for fraudulent scientists — most often funded by self-interested parties such as the makers and purveyors of organic products — whose intention is to cast doubt about the safety and/or effectiveness of the products of the new techniques and products. In fact, with the exception of wild berries, wild game, wild mushrooms, and most fish and shellfish, everything in North American diets is derived from organisms that have undergone some purposeful genetic modification.
Genetic modification by means of selection and hybridization has been with us for many centuries, and the techniques employed along the way, including the most recent ones, are part of a seamless continuum. Breeders routinely use radiation or chemical mutagens on seeds to scramble a plant’s DNA to generate new traits, and more than half a century of “wide cross” hybridizations, which involve the movement of genes from one species or one genus to another, have given rise to plants that do not exist in nature; they include the varieties of corn, oats, pumpkin, wheat, rice, tomatoes, and potatoes that we buy and consume routinely.
Genetic modification by means of selection and hybridization has been with us for many centuries, and the techniques employed along the way, including the most recent ones, are part of a seamless continuum.
What’s new is not the genetic engineering of food but the techniques for accomplishing it. And the newest methods — recombinant-DNA technology and new gene-editing techniques such as CRISPR/Cas9 — are far more precise and predictable than their predecessors. For more than a quarter century, there has been a consensus among scientists that the molecular techniques for genetic modification are extensions, or refinements, of more-primitive techniques.
The foregoing is important because much advocacy research denies the continuum of genetic modification and depicts the organisms that result from the superior molecular techniques as a new, unique “category” that poses novel risks. That depiction is conceptually flawed. Advocacy research often tries to “prove” things that are unprovable and incorrect, not unlike trying to disprove the laws of thermodynamics by inventing a perpetual-motion machine.
We consider below some of the more egregious examples of flawed advocacy research on genetic engineering.
In 1998, two years after the first commercial introduction of a crop genetically engineered (GE) with molecular techniques, Árpád Pusztai at the Rowett Research Institute in Scotland published a paper describing the effect of feeding genetically engineered and non-engineered potatoes to laboratory rats. Pusztai claimed to show that feeding them potatoes that were genetically engineered to express a protein known to be toxic to certain insects caused damage to the rats’ immune system and stimulated abnormal cell division in the digestive tract of the animals.
However, many research groups, including some of Pusztai’s own collaborators, concluded that his research methodology was fundamentally flawed, that he misinterpreted his own data, and that no conclusions about the safety of genetically engineered foods could be drawn from his data — or, indeed, from his experimental design. The experiments have been criticized for the small number of animals, the use of inappropriate statistical procedures, and the fact that a diet consisting only of raw potatoes is inadequate and even harmful.
Most significant, Pusztai failed to use appropriate controls, the aspect of an experiment that allows the researcher to examine the consequences of making only the change or intervention under study. Specifically, the control potatoes in his experiment had a different history from that of the genetically engineered potatoes — which was the likely explanation for any differences that he might have observed. We say might have observed, because — this is the most damning flaw of all — trained independent experts could see no differences between the tissue preparations from the control and the experimental groups.
By 2007, bad science had a new poster boy. A group led by Gilles-Éric Séralini in France published a reevaluation of 90-day GE-feeding-trial data that the European Food Safety Authority (EFSA) had examined in 2004. Using questionable statistics, Séralini claimed that there was evidence of harm to rats fed GE crops. The publication of the paper came with the typical media hype and fear-mongering. The EFSA convened an expert panel, which concluded that the minor differences found by Séralini were consistent with changes found in standard feeding trials and did not represent evidence of harm.
Undaunted, Séralini in his next published paper (2009) claimed that three different GE crops caused liver and kidney damage to rats. Again the EFSA examined the “new evidence” of harm and, as before, found no adverse effects and issued a stinging rebuke.
Undeterred, the Séralini group doubled down and performed a two-year study of rats fed a diet of genetically engineered feed. They published their results in 2012, again accompanied by a well-orchestrated global media blitz that featured lurid images of huge tumors in the rats fed a diet based on GE crops.
The paper was, in a word, atrocious. For one thing, it failed to mention that the breed of rats used in the study spontaneously generate tumors as the animals age. Scientists with pre-publication access to the article were outraged, and when it was published, the journal ran twelve critical letters that outlined why the paper should not have passed peer review.
The powerful images of rats with huge tumors saturated the airwaves for weeks, generating fear and consternation among non-scientists. When food-safety experts around the world examined the paper, however, it became obvious that these experiments were more junk science, and eventually the pressure from scientific bodies around the world persuaded the journal to retract the paper.
There is so much wrong with the experimental design and interpretation of data in all of Séralini’s published work that the conclusion is inescapable that he and his colleagues intended to get a spurious, preordained result. Why? Séralini is president of the scientific board of a self-described anti-genetic-engineering NGO that apparently is hosted by his laboratory; he has a long and sordid history of anti-genetic-engineering and anti-agricultural-chemicals activism; and his research has been funded by two large “GE-free” French supermarket chains that sell organic and homeopathic products. (Organic agriculture bans products made with molecular-genetic-engineering techniques.)
A large EU-funded study assessed the usefulness of 90-day and one-year rat-feeding trials. It found that one-year feeding studies did not provide useful additional data and, more important, that feeding studies lacking prospectively chosen, plausible endpoints can lead to “randomly generated significant differences between animals fed with the [genetically engineered] test material and animals fed with a controlled diet.” And “such results do not provide useful information for risk assessment,” it concluded.
In other words, studies such as those of Séralini are ill-conceived and, at best, represent data-mining that unearths random associations.
In Italy, University of Naples researcher Federico Infascelli published a series of papers in which he claimed to have found problems in animals fed feed from genetically engineered crops. After an Italian senator (who is also a scientist) found questionable figures in three of his papers, an investigation was launched by the university.
We must be skeptical ofclaims that genetically engineered crops are fundamentally different in any way from plants that have been modified over time with an increasingly sophisticated array of techniques.
A forensic audit of data in the eight Infascelli papers revealed evidence of data duplication, deletion of control data, splicing in of control data, reusing data and misrepresenting their origins, and even complete fabrication of one data set (available here). The head of the university investigation confirmed to the journal Nature that “the committee found that the papers contain intentional data manipulation.”
The small but zealous anti-genetic-engineering community touted Infascelli’s articles as proof of the dangers of genetically engineered crops. But as is par for the course for the anti-genetic-engineering cabal, his findings are contradicted by voluminous legitimate research by food and drug and health agencies, agribusiness companies, and independent academic investigators worldwide.
As with experimental attempts to prove that perpetual-motion machines are the wave of the future, we must be skeptical of supposed demonstrations of systematic dangers from genetically engineered crops — or, for that matter, of claims that such crops are fundamentally different in any way from plants that have been modified over time with an increasingly sophisticated array of techniques.
Why, then, do these claims continue to be made? Well, scientists cheat and lie for many of the same reasons that people commit espionage and betray their country — money, ideology, disillusionment, delusion.
Money is the key to much of the misconduct in research on genetically engineered organisms. There exists in Europe and North America, in particular, a vast, well-established, highly professional protest industry fueled by special-interest groups — especially the makers and sellers of organic and “natural” products (see this and this) — seeking to line their own pockets while compromising the public interest. A review of tax returns of the “non-profit” activist organizations that oppose agricultural biotechnology and other modern production methods and fund much of the spurious research reveals that more than $2.5 billion is being spent annually, mostly on the production of propaganda.
Whatever the reasons for misconduct by scientists and journals, we would invoke the admonition of DeWitt Stetten Jr., the former deputy director of the U.S. National Institutes of Health: “Science cannot tolerate the man who takes lightly his moral obligation to report strictly what is true.”
— Henry I. Miller, a physician and molecular biologist, is the Robert Wesson Fellow in Scientific Philosophy and Public Policy at Stanford University’s Hoover Institution. He was the founding director of the Office of Biotechnology at the Food and Drug Administration. Robert Wager is a faculty member in the biology department at Vancouver Island University in Nanaimo, British Columbia, Canada.