Although food science is not my area of specialization, looking into Genetically Modified Organisms (GMOs) has been one of my recent pet topics. When first introduced to the controversy surrounding their use and safety, I was highly sympathetic to the naturalistic claims which warned against unintended consequences from the relatively rapid development of these technologies. In the wake of the 2008 recession which was largely caused by greedy corporations and banks, it seemed entirely plausible that corporations like Monsanto were putting profits ahead of safety.
However, as I began to actually read the research that the naturalist crowd used to support their position, I noticed it was filled with methodological flaws and dubious claims. While it is difficult to chase every thread to verify or refute them, it became quickly apparent that the naturalist position was incredibly weak exclusively relying on disproven studies (often in sham journals). Without these studies to ground their position, there was no foundation for the claims and the remainder of their position thereby degenerates into logical fallacies and conspiracy theories. As such, the anti-GMO movement is yet another anti-science movement.
This is not to imply that GMOs should get a free pass. Each one should be carefully analyzed and examined for safety on numerous grounds (toxicity, allergenicity, etc...). Fortunately and despite claims to the contrary, there is a large body of work supporting the safety of GMOs.
Below I've collected some of the most common claims regarding GMOs and their related technologies as well as my conclusions that have lead me to this position.
Claim - Roundup Ready crops are dangerous because it allows farmers to use more Roundup on crops that eventually end up in our food and if itís bad for weeds, it must be bad for humans.
Response - Roundup works by targeting specific pathways in plants. Not surprisingly, there has been no conclusive demonstration of any negative effects on humans or any other animal. Those opposed to GMOs cite a number of studies to claim that there are negative effects (although their claims and the sources they cite often fail to distinguish between the GMO itself and the effect of Roundup), but every study Iíve come across that claims to support this position has had massive methodological flaws and cannot be considered as valid evidence (see individual claims for more information).
Claim - Since GM crops are more resistant to pesticides like Roundup, this means that more is used.
Response - While overall pesticide use is up, it is not due to GM crops. It is due to increase acreage. Acre for acre, the amount of pesticide use has actually decreased thanks to GM crops.
Claim - A 2013 study linked Roundup to a range of health problems and diseases, such as obesity, diabetes, heart disease, depression, autism, infertility, cancer and Alzheimerís disease.
Response - The study cited makes claims that are not quite as represented by those opposed to GMOs. Instead of definitively linking these problems with GMOs to humans, the authors of the papers look at other species, cherry picking potentially negative effects that are far from conclusive, and then postulating that there might be similar effects in humans. However, even the authors of the paper stay well away from making any claims of certainty on these issues. As such, the claims are vastly overstated for even what the paper says.
But what of the study itself? The authors attempt to hide is that there are studies that have directly tested the effects they claim in humans. In fact, they even cite one of those studies in their paper but fail to address the conclusion that GMOs have been found not to have any effect. Instead of paying attention to the studies that have been done and confirm a lack of toxicity, the authors instead choose to cite as their main evidence a thoroughly discredited study by Seralini.
Lastly, the journal in which the paper was published, Entropy, is not a respected journal within the scientific community and appears to be a haven for cranks who can pay to have their writings published there. This is precisely what seems to be the case given the unsubstantiated claims and the lack of authors expertise in the subject area. The lead author is an "independent scientists and consultant" with no apparent training and numerous publications that make it clear he has an ax to grind. The second author is an MIT computer scientist who has published other papers that have been thoroughly debunked.
Claim - Monsanto is so overzealous in their patent infringement that they sue farmers whose fields are accidentally pollinated by pollen from the GM crops.
Response - While this is a frequent claim Iíve heard, Iíve never been shown an instance of this happening. Most people that make this claim point to the recent Supreme Court case of Bowman v. Monsanto. However, this case is completely unrelated to the claim of accidental pollination. Rather, the case revolves around a farmer who seeded his fields with seed from a granary meant for feed for livestock. This seed was likely to contain seeds patented by Monsanto with the Roundup resistance. Bowman then grew the plants, used Roundup on them which killed the ones which werenít the resistant GM variety, and Bowman then proceeded to use those seeds for future generations.
Since Bowman knowingly sorted plants based on their Roundup resistance, the court ruled that Bowman had intentionally violated the patent by not paying to use that intellectual property. Bowmanís defense was that once the seeds passed into the granary, Monsanto could no longer enforce the patent. This claim was rejected by the court. However, in relation to the original claim, Bowmanís defense was never that he was ignorant of the presence of the GMO seed or that the fields were seeded accidentally.
Response - The study in question can be found here. As noted in my previous response about Roundup, this is a highly cited study that fails to pass scientific muster. The study has severe methodological problems. In particular, the authors examined the intestines of pigs raised on GM and non-GM corn and rated them on the amount of inflammation being nil, mild, moderate, and severe. However, this is a highly subjective breakdown. To guard against bias studies that have such breakdowns should be double blind. However, there is no indication that this one was.
Further, the non-GM group was significantly higher for mild and moderate inflammation. This does not follow the proposed trend and in fact counts against the claim. But given the aforementioned ambiguity in categorization, it would be statistically stronger for the authors to have broken the categorization into two groups, say low and high. In this case, the difference between the two groups vanishes entirely. When resizing your categories to make the groupings larger causes the significance to disappear, this is an indication that what was found was nothing more than statistical noise. Proper research should crunch the numbers and attempt to discern what the probability of a false positive would be but this paper makes no such attempt.
Complicating factors were also downplayed. The researchers state that several of the pigs developed pneumonia. Based on rates of typical infection in animal rearing, the pigs in the study were 3-5 times more likely to be sick. Yet the authors tried to play this off as typical.
Aside from the shoddy science, it also fails common sense. If these effects were real and significant it is something that should have been readily noticed by farmers themselves.
None of this comes across as surprising given the study was published in the dubious Journal of Organic Systems. While scientific sounding, this is not a journal that the scientific community recognizes. While slicker in presentation, it seems to be akin to the dismal Journal of Cosmology which is a ďjournalĒ set up by and for pseudoscientists. The Journal of Organic Systems is not featured in PubMed (a database of scholarly publications relating to the medical field). One way at looking at the credibility of a journal is its Impact Factor. Major journals have a factor in the high 30ís. The impact factor for the JoOS is so low that it could not be measured.
More reading on this claim:
Claim - A study found that GM crops and/or Roundup caused tumors in rats.
Response - As with most other studies claiming to find issues with GM crops, this study fails to meet a reasonable standard for evidence. As with the pig study cited above, the data itself was highly conflicting. Some test groups of rats that had Roundup added to their water had higher rates of cancer, others did not. No mathematical analysis was performed to determine how many standard deviations more or less this may have been for the test vs. control groups. Instead, this study cherry picked the cases that had higher rates of cancer and used this to claim that Roundup was more dangerous.
However, this is likely to again be statistical noise. The researchers inexplicably chose rats that are already genetically prone to cancerous tumors. This made no methodological sense since other breeds were available that are much less prone and would thereby be a better way to highlight any potential increase.
Also, if the hypothesis was correct, then increasing doses should increase the tumor rate. But the authors presented no such correlation, which again reinforces that likelihood that their conclusions came from statistical noise. Independent analyses of it have supported this.
The study has been considered so flawed that six French scientific academies (a country known for being highly skeptical of GMOs) have condemned the study citing problems with the experimental design, statistical analysis and animals used, and inadequate data. It has also been rejected by the European Food Safety Authority. Due to the flawed nature of the study, the paper in which the article was published as retracted it.
Claim - A study found BT caused leukemia in rats.
This study is extremely bizarre in its methodological design from the beginning. Instead of using BT as it is actually used in farming practices, the researchers used the whole bacteria in spore form that the BT gene was extracted from before that gene was inserted into crops. This introduces an indeterminate number of variables that are not constrained and does not reflect how BT would be found when associated with GMO crops. Additionally, the concentrations were exceptionally high (ten to a hundred million times higher concentrations than humans would ingest, proportionally). Again, this is not an accurate representation of how BT would be ingested. The experimental design is also weak in that it uses only 60% of the number of animals that is recommended as a minimum for a reliable test.
As with the pig study and the rat tumor study, this study did not reflect a trend as should be expected with increased dosages leading to increase problems. In this study, the rats with the highest rate of leukemia were the ones with the lowest dosage. Again, this would be expected for statistical noise.
The authors also cite discredited work, misrepresent another, and publish in a journal that is not respected in the scientific community. In fact, its parent company (OMICS) is on the list of potentially predatory journals. Incidentally, the paper had originally been published in a different journal which is well respected (Food Chem Toxicology), but was withdrawn for unknown reasons.
Claim - GM wheat disables genes in humans which is fatal and can be passed down to future generations.
Response - The heart of this claim lies in a particular strain of GM wheat which creates a type of RNA to disable two other genes in the wheat. His question was whether or not that RNA (which could potentially be long lived) could be ingested and find complimentary genes in humans, disabling them and interfering with our health. The researcher found similarities between one of the wheat genes and a human gene that control part of the human production of carbohydrates and would damage the liver over time if weakened.
However, for that to happen, the amount of the silencing RNA entering the portions of the body in charge of this would have to be sufficient to cause harm. To do that, the RNA would have to go through digestion intact. Studies have shown that similar types of RNA can get through, but only in minuscule amounts which makes it extremely unlikely that this effect would be harmful.
And that's only assuming that the genes in question really are as comparable as the author thinks. This too is unlikely since, when the author made the comparisons, he did not know exactly what RNA would be used and how it would interact with the genes. As such, other scientists have pointed out that his conclusions boil down to nothing more than speculation. Others have performed similar analyses and concluded that there isNo significant similarity found.
While that is the extent of the original work, the derivatives from the naturalistic community have gone on to make even more bizarre claims. While it is true that any genetic changes could be inherited, for that to happen, the changes would have to be made in the gametes (eggs and sperm). But since the genes are not being expressed in those cells the RNA could not act on them, making it impossible for the changes to occur. Furthermore, if ingestion was fatal, then reproduction would be impossible anyway, again, making it impossible for the damage to be hereditary.
Claim - BT crops are killing bees and/or butterflies. These are important pollinators for all plant life on Earth.
Response - BT is short for Bacillus thuringiensis, a naturally occurring bacterium that forms crystals that are effective insecticides. Genes from this bacteria have been extracted and inserted into some GMOs causing the plants themselves to be insect resistant. However, the insect must actually consume the plant.
In 1999 a study was released suggesting that butterflies were being harmed by ingestion of the pollen from the plants. While this initially passed the sniff test since the pollen is a part of the plant, this study was quickly criticized since the pollen was expected to contain significantly lower amounts of the BT toxin. Followup studies were unable to reproduce the effect and it was eventually decided that the effect the study found was due to improper collection procedures that accidentally mixed pollen with other parts of the plants which did contain the toxin and was then fed to the butterflies.
Similarly, when the mysterious Colony Collapse Disorder (CCD) of bees was discovered in 2007, fears were revived that BT corn was responsible. However, a study that same year effectively ruled that out. There have been recent claims that Russia made threats to go to war with America over the support of these crops, but this claim is false.
More reading on this claim:
Response - While this claim is again, rather sweeping, several articles are often cited as examples of lower nutritional content. Below I've examined three particular articles cited to support this position.
Lappe et al.
The first is a 1999 study from Lappe et al. The naturalist advocates claim that the study shows that "GM soy had 12Ė14% lower levels of cancer-fighting isoflavones than non-GM soy". While it is true that this particular study did find a decrease in isoflavones but this decrease is still within natural variations of soy. As such, it cannot be firmly established that the GM version is inherently responsible for the decrease.
What is not entirely correct is the claim that isoflavones are "cancer-fighting" since this claim paints a picture with a broad brush. The details are somewhat different. In particular, isoflavones have only been implicated in preventing breast cancer and even then only in Asian populations. In contrast, other studies have indicated that isoflavones may be cancer causing.
Thus, it's a weak claim that this decrease represents a significant difference in "nutritional" value when the actual nutritional benefit is so limited and still contested.
Shewmaker et al.
Citing the 1999 study from Shewmaker et al. natural advocates claim "Canola (oilseed rape) engineered to contain vitamin A in its oil had much reduced vitamin E and an altered oil-fat composition, compared with the non-GM control."
Indeed, the article does claim that the tocopherol (vitamin E) content was reduced, but the full article is behind a paywall and it's impossible to say how much. Unlike the isoflavones which are of dubious nutritional significance, Vitamin E is certainly an important compound and the amount of variation would be nice to know. However, a more recent study has conflicting results with the Sewmaker et al. noting the levels of tocopherol were "1.5-fold higher in the 35S-PAC seed, compared with the non-transgenic control" (emphasis added). Thus, the claim that vitamin E is inherently reduced in GM canola is not well established.
Additionally, it should be noted that the main thrust of the Shewmaker et al. article is not about tocopherol. It concerns the expression of the beta carotene (vitamin A) which it, among other studies have noted was indeed increased. While this is an aside to the claim that vitamin E levels were decreased, it does contradict the broader claim that "GM crops do not have increased nutritional value and/or are lower in nutritional content."
Jiao et al.
A third claim cites a 2010 study by Jiao et al and makes a rather large claim that:
Experimental GM rice varieties had unintended major nutritional disturbances compared with non-GM counterparts, although they were grown side-by-side in the same conditions. The structure and texture of the GM rice grain was affected and its nutritional content and value were dramatically altered. The variation ranged from 20 to 74% for amino acids, from 19 to 38% for fatty acids, from 25 to 57% for vitamins, from 20 to 50% for nutritionally important trace elements, and 25% for protein. GM rice varieties variously showed markedly decreased levels of vitamin E, protein, and amino acids. The authors said that their findings ďprovided alarming information with regard to the nutritional value of transgenic riceĒ and showed that the GM rice was not substantially equivalent to non-GM.
As with many other articles, this one appears behind a paywall, making it hard to determine the full truth of the claim. However, the bulk of the quote (the portion regarding the variation in nutritional content) comes directly from the abstract. Notably, while the quote demonstrates these values were "altered" it does not say they were lowered. Rather, these just look to be variations around a mean. Indeed, another article citing this one notes that Jiao and all found "increases and decreases, inconsistent between lines". As such, it is incorrect for the claim to paint this as a systematic decrease. Furthermore, the out-of-context quote from Jiao et al does not disclose the amount of natural variation within rice.
Thus, the claim is flawed because it does not represent an actual decrease, as the organic advocates claim, nor does it demonstrate that the variations were inherent to the GM process as opposed to natural variation. The referencing article only notes that the Jaio article found "a 25% reduction in protein content was observed for one antifungal GE line, which was therefore considered by the authors to be less nutritious." In short, Jiao et al aren't even claiming that the majority variations are what the organic advocates claim. That claim does not come from the cited paper and seems to be entirely fabricated with the exception of the protein.
Overall, the reality is that the insertion of genes in GM products can disrupt other genes which produce important nutrients or other unintended consequences. However, the likelihood of this happening is relatively small given how much of most genomes are non-coding. It remains an important topic to investigate what the unintended consequences may be before engaging in human consumption. However, numerous studies are conducted and few of them have revealed significant differences. The vast majority have demonstrated no statistical difference in the unintended changes.
While there certainly are some unintended consequences, they are few and far between. Taken together, the articles cited to not paint a picture that supports the overall claim that GM crops are diminished in nutritious content. While there are some compounds that certainly are less, the naturalistic claim that GMOs are always less is demonstrably false. Indeed, many sources have shown distinct increases, which are dishonestly ignored by the organic proponents, even when it is in the very paper they cite. Instead, the organic lobby cherry picks counterexamples, grasps at straws to pick compounds that are not well established as necessary nutrition, and misrepresents their own sources.
More Reading On This Claim:
Response - Simple answer on this one: The table doesn't show what organic advocates claim it does. It doesn't reflect the characteristics of the corn, but rather the soil and is likely fabricated at that.
Response - The study in question can be found here. Typically, I find the Union of Concerned Scientists a very respectable organization. I have agreed with them on many issues and my family even donates to them. Ultimately, I cannot find anything inherently wrong with their review on this issue.
That being said, when evaluating such issues, one should consider other large scale meta-studies. While the UCS found that GM crops have lower yields, studies from other well respected organizations, published in highly respected journals, have found the exact opposite. In particular, one from Nature found that GM crops on average have a 25% greater yield. Given that many studies support that GM crops fare better and there only seems to be one large scale review that says otherwise, the outlier is the one that should be considered suspect.
Furthermore, it also begs the question that, if GM crops did not fare better, why farmers whose very livelihoods are dependent on yield, would continue to use GM crops. So while I cannot pin down a specific issue with the UCS review, the ancillary evidence places it in doubt.
Response - While cross pollination can certainly occur, it begs the question of why it is a problem? Unless those opposing GM plants demonstrate that there is harm, then this should not be any more of a concern than with cross pollination of other plants. Unfortunately for them, I have come to the conclusion that their claims of potential harm are false as shown by this page.
Response - Monocultures are large populations of crops that are not genetically diverse putting them at risk for a single parasite to eradicate the entire system. If there are not sufficient variety, a food supply could be in jeopardy. This has happened several times in the past. The most known example is the Irish potato famine, but mass destructions have also occurred with certain banana strains and may happen again to the current strain of bananas. So there is certainly historical legitimacy to this fear.
That being said, these examples point out the fallacy of the argument from those opposed to GM crops: None of the crops I've just listed were genetically modified through modern means. They were and are, entirely organic. Thus, the argument is not about GM crops, but rather about farming practices. So long as GM crops keep being introduced with sufficient diversity, there is little difference on this front unless farmers limit themselves to very narrow stocks.
Response - The claim can be traced primarily suicides in the late 1990's and early part of this century but has seen recent surges again in 2011-2012. The claim stems from rashes of suicides among farmers who have gotten themselves in deep debt buying seeds for crops that ultimately fail. Since many farmers were using GM cotton, many drew an association between the two. However, correlation is not the same as causation.
While many GM crops did fail, so too did other crops. A review of the issue in 2005 found many other causes, primarily "government apathy, the absence of a safety net for farmers, and lack of access to information related to agriculture". Indeed, GM cotton was not introduced to India until 2002 and suicides had well predated this.
Response - This claim seems to come primarily from a 2012 paper, which demonstrates that organic farming does indeed pull more carbon dioxide out of the atmosphere.
However, this is only a small part of the story. What it leaves out is that large scale industrial farming involving GMOs reduces carbon dioxide emissions considerably. In particular, organic farming is generally small in scale and does not have efficient methods of planting, harvesting, or transport. Similarly, organic crops tend to need more fertilizer which increases methane emissions, another greenhouse gas.
As such, this makes fields that are more likely to include GMOs more environmentally friendly.
Response - I liken the attempt to label GMOs to the Selman v Cobb County School District court case in which anti-evolutionists attempted to require labels on science they found objectionable by putting stickers in textbooks warning that evolution was just a theory in an attempt to demonize it. The same is true here. As shown above, the science doesnít support that GMOs are harmful so there is no impetuous to label GMOs for safety reasons. The consequence would be that the demonization of GMOs would be perpetuated and this unfairly endangers the businesses that create them.
I agree with the argument that consumers should be made aware of what theyíre purchasing and thereby protecting consumer choice. However, I feel it would be best to counteract fearmongering via labeling by working to have informative labels on products. A good outline for a way to do this has been presented at the BioFortified blog.