User:Groupuscule/GMO

This is not a Wikipedia article: This is a workpage, a collection of material and work in progress that may or may not be incorporated into Genetically modified food controversies. It should not necessarily be considered factual or authoritative.

This page grew out of a discussion at Talk:Genetically modified food controversies. Its goal is to assess the following claim which is currently made on Wikipedia:

“	There is broad scientific consensus that food on the market derived from genetically modified crops poses no greater risk than conventional food.	”
— Genetically modified food, Genetically modified food controversies

Here, you will find analysis & research suggesting (1) that the sources presented to support this claim are inadequate and (2) that many alternative sources contest this claim. The work presented here is wide-ranging but not comprehensive, and should not be interpreted as a polemical argument that genetic modification is dangerous. This text should be understood primarily as a response to the stalled and disorganized discussion taking place on article talk pages. For more explanation of why this page exists, please read this section and the section immediately following.

I am sure that I have made mistakes, small and large, while searching, reading, compiling, and annotating. If you find errors in this document please let me know at the talk page here or at my user page. Please do not edit this page directly; I have organized it carefully in the hopes that it will be somewhat legible for those new to the debate. Please feel free to fork any part of this page if you wish to discuss its contents line by line.

The question of safety in genetically modified food is obviously important to the world. More immediately, it is important to the Wikipedia community. Multiple Wikipedia articles currently make a claim about the "safety" of consuming an enormous range of new organisms— in terms not used even for well-known substances such as aspirin or aloe vera. The community must address this claim, even if only to verify it more adequately.

“Broad scientific consensus” is phrase that has appears in a number of recent secondary reports about genetic modification. When certain authors see fit to report on a "scientific consensus", “broad” seems to be the modifier of choice—to the point where you can search for “scientific consensus” on genetic engineering and see that it's almost always described as “broad”. This claim is also made prominently in several Wikipedia articles.

A great deal of energy has gone into asserting the existence of this consensus. But what evidence has been provided?

We will demonstrate that many of the sources provided are used inappropriately, and that they themselves misrepresent the factual basis for their claims. Our analysis will not rely on original research, but on assessment of the quality of the sources—and of how they relate to the claim. Thus, we will limit ourselves to examination of the sources themselves and to secondary literature that comments on these sources directly.

The biggest repeaters of this phrase are Forbes, Discover, and Scientific American. The claim is sometimes sourced to Pamela Ronald, a professor of genetic engineering at UC Davis.

Ronald was using the claim as early as 2008, as you can see here at her blog. She uses this wording in several places. Ronald's first defense of the claim, that I have found, occurred in a 2010 “debate” hosted by The Economist:

There is broad scientific consensus that GE crops currently on the market are safe to eat. The National Research Council (NRC), a non-profit institution that provides science, technology and health policy advice to the US Congress, reports that the process of genetic engineering poses a similar risk of unintended consequences as conventional approaches of genetic alteration. After 14 years of cultivation and a cumulative total of 2 billion acres planted, GE crops have not caused a single instance of harm to human health or the environment. The NRC findings have been confirmed by leading scientific agencies around the world. For instance, the Joint Research Centre, the European Union's scientific and technical research laboratory and an integral part of the European Commission, recently concluded that there is a comprehensive body of knowledge that adequately addresses the food safety issue of GE crops and that the crops currently on the market have not caused any known health effects. In contrast, every year there are thousands of reported pesticide poisonings (around 1,200 each year in California alone; 300,000 deaths globally). [Emphasis added here and below.]

Google Scholar searches suggest that the phrase appears in her published work with a 2011 article in Genetics. Here, Ronald expands further, but cites the same two studies:

There is broad scientific consensus that genetically engineered crops currently on the market are safe to eat. After 14 years of cultivation and a cumulative total of 2 billion acres planted, no adverse health or environmental effects have resulted from commercialization of genetically engineered crops (Board on Agriculture and Natural Resources, Committee on Environmental Impacts Associated with Commercialization of Transgenic Plants, National Research Council and Division on Earth and Life Studies 2002). Both the U.S. National Research Council and the Joint Research Centre (the European Union's scientific and technical research laboratory and an integral part of the European Commission) have concluded that there is a comprehensive body of knowledge that adequately addresses the food safety issue of genetically engineered crops (Committee on Identifying and Assessing Unintended Effects of Genetically Engineered Foods on Human Health and National Research Council 2004; European Commission Joint Research Centre 2008). These and other recent reports conclude that the processes of genetic engineering and conventional breeding are no different in terms of unintended consequences to human health and the environment (European Commission Directorate-General for Research and Innovation 2010).

This is not to say that every new variety will be as benign as the crops currently on the market. This is because each new plant variety (whether it is developed through genetic engineering or conventional approaches of genetic modification) carries a risk of unintended consequences. Whereas each new genetically engineered crop variety is assessed on a case-by-case basis by three governmental agencies, conventional crops are not regulated by these agencies. Still, to date, compounds with harmful effects on humans or animals have been documented only in foods developed through conventional breeding approaches. For example, conventional breeders selected a celery variety with relatively high amounts of psoralens to deter insect predators that damage the plant. Some farm workers who harvested such celery developed a severe skin rash—an unintended consequence of this breeding strategy (Committee on Identifying and Assessing Unintended Effects of Genetically Engineered Foods on Human Health and National Research Council 2004).

(The rest of the article does not address the topic of safety, focusing instead on the virtues of genetic modification for farmers. Ronald's article at the Scientific American blog is substantially identical.)

Ronald does not review the literature herself, but relies on two sources for her claim. The first is a report published in 2004: “Safety of Genetically Engineered Foods: Approaches to Assessing Unintended Health Effects”.

This report also does not perform a literature review. Its main purpose seems to be establishing a rhetorical equivalence between foods that are “genetically engineered” and other foods—produced through "conventional" agriculture (i.e., cross-breeding)—which are “genetically modified”. The usefulness of this report from a public relations standpoint is clear: it blurs the line between biotech engineering and conventional agriculture.

Even so, the report very clearly does not support Ronald's claim. It states that there is not yet enough information available to determine the "biological relevance" of changes made through genetic engineering:

Although compositional changes can be detected readily in food, and the power of profiling techniques is rapidly increasing our ability to identify compositional differences between GE food products and their conventional counterparts, methods for determining the biological relevance of these changes and predicting unintended adverse health effects are understudied. As discussed in this report, further advances in analytical technologies and their interpretation are needed to address these limitations. (p. 177)

One of the main features of the report is this chart, which ranks the levels of risk to be expected from different models of genetic modification. It is quite obvious from this chart that genetic engineering bears greater risks of unintended genetic effects than does conventional agricultural breeding.

Addendum 20:45, 14 June 2013 (UTC) Before this page even got started, Historyday01 identified a part of the NRC report (pp. 118–119) which suggests that genetically engineered foods are inherently more likely than conventional counterparts to trigger allergic reactions and adverse effects. If we are to understand that the NRC's report reflects scientific consensus—and instead of attempting to 'win the argument' on this point, we must insist that it does not necessarily do so)—it would seem to suggest a consensus that genetic engineering poses more severe risks for human health:

A large number of compositional changes in foods may potentially arise from any method of genetic modification of food. Furthermore, genetic engineering, as previously discussed, has a higher probability of producing unanticipated changes than some genetic modification methods, such as narrow crosses, and a lower probability than others, such as radiation mutagenesis. Therefore, the nature of the compositional change merits greater consideration than the method used to achieve the change, for example, the magnitude of additions or deletions of specific constituents and modifications that may result in an unintended adverse effect, such as enhanced allergenic potential. Constituents whose levels are increased could well include some of the “natural” toxins present in food, thereby enhancing the potential for adverse effects to occur with consumption of that food. Examples of deletions of specific constituents that merit consideration are those intended to enhance nutrient bioavailability by reducing barriers to absorption.

Modifications intended to enhance uptake of essential nutrients (e.g., reduction of phytic acid to improve iron or zinc bioavailability, and thus decrease the risk of iron or zinc deficiency) are particularly attractive. Paradoxically, the more effective such modifications are, the likelier are unintended effects on the bioavailability of other dietary constituents, that is, changes that increase uptake of essential trace elements also may increase the bioavailability of unwanted contaminants, such as toxic heavy metals.

Hazards that may be of concern after this type of general evaluation are toxicities, allergies, nutrient deficiencies and imbalances, risks related to nutrient displacement, and risks related to endocrine activity and diet-related chronic diseases. These categories are not exclusive. For example, although idiopathic (without known origin) reactions also are distinct possibilities, they are not discussed because, by their very nature, they are presently impossible to predict. Since many idiopathic reactions are likely genetically determined, they may be predictable in the future as genetic polymorphisms are better understood. The International Life Sciences Institute has reviewed the safety of DNA in foods (ILSI, 2002b) and has published a monograph on Genetic Modification Technology and Food: Consumer Health and Safety (ILSI, 2002a).

Ronald cites a 2008 report from Europe titled “Scientific and Technical Contribution to the Development of an Overall Health Strategy in the Area of GMOs ”. This report takes a stance that much more aggressively favors genetically modified crops—and opposes regulations.

As it happens, this report explains directly why Ronald's equivocation (repeated constantly by Team GMO on Wikipedia) about conventional food is wrong:

Following the comparative safety assessment approach, the safety of a GMO is established relative to a conventional counterpart, which implicitly presumes the safety of the latter. This is based on the fact that whilst conventional foods usually have not been tested for safety, their history of safe use indicates that a positive balance has been found between the potentially negative and positive effects of the many substances present within these foods.

It also mentions at least one pretty good idea for improving tests of genetically modified products:

... these authors recommended including an additional non-GM diet that has been spiked with the transgenic protein, so that effects due to this protein and other components of the GM diet can be distinguished.

But this document does not affirm the existence of a scientific consensus on GMO safety. Like the previous report, it is more of a rhetorical advisory to regulators than an attempt to comprehensively review the available literature. The introduction (p. 5) states: “It is important to note that the analyses and discussions which have led to the present report have concentrated on the current approaches to assess the potential health effects of GM food and feed products and not on the nature of those effects themselves.”

Indeed it acknowledges some forms of GMO risk. For example: “particular consideration should be paid in the environmental risk assessment to GMOs containing antibiotic resistance genes in order to phase out any antibiotic resistance genes that may have an adverse effect on human health and/ or the environment.” But the paper does not dwell on negative consequences of these risks, precisely because it envisions a best-case scenario in which genetically modified foods are perfectly well-regulated.

The report does mention one test on humans: A 1999 study in which 11 people were fed genetically modified tilapia for five days. Their blood did not contrast significantly with samples taken from a control group.

In a section on animal testing, the report describes ambivalent results. Some studies have not found health effects of genetically modified foods. Others have. In Wainwright et al. (2003): “The results show some differences between the GM canola oil and borage oil groups, including decreased body weight and altered brain lipid composition.”

In another study:

A group of researchers has also published various studies on the ultrastructure of cells of various organs (liver, spleen, testes) of mice fed glyphosate resistant soybean for up to eight months [Vecchio et al., 2004, and references therein]. Whilst these authors note that the nucleus and other organelles may show changes depending on the diet, the cause of these changes has not been established. In addition, the origin of the GM soybean is not specified in detail and the model employed is not routinely used in toxicity testing.

Finally, the report notes an accidental release of genetically modified food known to provoke allergies:

Cry9C also has elicited an allergic serum reaction in Brown Norway rats, which are known to be IgE-hyperresponders, whilst also another protein without known allergenic properties tested positive in the same test. Based on these considerations, Starlink™ has previously only been allowed onto the market for feed use. Despite this, it accidentally has become commingled with human food products derived from maize, such as taco shells. This has instigated a major recall action and a request to consumers to report any allergic reactions that might have been related to the consumption of Starlink™-containing products.

Based on the two reports described above Pamela Ronald made this statement: “There is broad scientific consensus that genetically engineered crops currently on the market are safe to eat.”

And that statement has been repeated multiple times by “science journalists”, smugly contemptuous of those who disagree.

“Broad scientific consensus”, Ronald's phrase has also been heavily promoted on Wikipedia, appearing at Genetically modified food, Genetically modified food controversies, and Regulation of the release of genetically modified organisms. An acrimonious edit conflict erupted at March Against Monsanto over whether the “broad scientific consensus” must be mentioned as a counterweight to activists' claims.

Dozens of other related pages defer safety concerns to the “controversy” and “regulation” pages—with the result that “broad scientific consensus” is portrayed as the final word on a wide range of topics.

What citations are invoked in support of these claims?

As of 9 June, Regulation of the release of genetically modified organisms still says: “there is now broad scientific consensus that GE crops on the market are safe to eat.” Two citations are given. One is a report from 2000 by the Organisation for Economic Co-operation and Development (OECD); the other is Ronald's 2011 article at Scientific American.

We have seen that Ronald's claim is based on smoke and mirrors. What of the OECD? One might easily say that the OECD is biased towards Western industry. One might also argue that the report is outdated, given how many new products have entered the market since the year 2000.

But first: what is the basis for the claim being made? The footnote on Wikipedia helpfully provides this quotation:

Much experience has been gained in the safety assessment of the first generation of foods derived through modern biotechnology, and those countries that have conducted assessments are confident that those GM foods they have approved are as safe as other foods.

This quotation comes from item 4 in the executive summary. The next lines of this item read:

Nevertheless, some have raised concerns about the adequacy of existing test methods. For example, more standardised procedures to establish substantial equivalence are needed, as well as improved methods to assess the allergenicity of proteins new to the diet (together with their digestibility and toxicity) taking regional differences in diet into account.

Immediately, we can conclude that although “those countries that have conducted assessments” may be satisfied with the products they have approved, noteworthy actors that are not countries would beg to differ. Quite clearly, the statement applies to certain governments, not to scientists.

The report goes on to describe (pp. 26–27) a number of very serious objections to the current process for approving GMOs as safe. The report attributes these objections not to dissident countries but to “some countries” and to the World Health Organization.

The report is explicit about the lack of consensus:

If a protein is shown to be resistant to typical digestive fluids, there may be added exposure to the intact protein or to large pieces of the protein. This digestive resistance would lead to a different analysis than if the protein were broken down as expected. However, there is still no consensus on resistant proteins being a significantly different risk if none of the other toxicity tests yields adverse results. (p. 27)

In short, the sources currently used at Regulation of the release of genetically modified organisms to demonstrate a “broad scientific consensus” roundly suggest the opposite.

Moving on to the Genetically modified food controversies, we find a more moderate claim and a pile of citations. (The claim was moderated and the new citations added after critical discussion on the talk page; however, no serious effort was made to re-examine the “consensus” claim based on the presentation of new sources. The situation is similar at Genetically modified food and the citations are identical.)

The six hyperlinked footnotes are already intimidating. In the footnotes, we see that even more studies are actually linked, since footnote six poetically contains six unique citations. These footnotes get a lot of use throughout the article.

Do these 13 studies support the claim for which they are cited? Let's dive in.

Footnote 1 is a short press release issued by the American Association for the Advancement of Science.

There is actually some secondary literature on this press release, which contains little analysis of its own. Dated 20 October 2012, the statement addresses a then-upcoming vote by Californians on labeling of genetically modified food. It urges them to reject labeling

Michele Simon wrote an article about this statement, asking “Is AAAS Serving Science or Monsanto?” With this article, Simon calls attention to several broad and unsupported claims made by the AAAS Board—and questions the Board's motivations, given chair Nina Fedoroff's ties to the chemical industry.

The Council for Responsible Genetics also objected to the AAAS statement.

We are deeply concerned that a scientific body such as the AAAS would take such an action without giving a complete review of the science behind its statement.

As scientists, they should know that citing a few studies in favor of their position can no longer be considered a compelling argument. Indeed, the AAAS Board did not conduct a thorough analysis of the literature, nor did they include studies that could cast doubt upon their conclusions. The truth is we do not know conclusively what the long-term effects of growing and consuming GM crops will be.

There have been very few systematic and independent animal studies testing the safety of GM crops. Since 1992 the FDA policy considers the insertion of foreign genes into the plant genomes of crops as the equivalent of hybrid crops - crosses within the same species - and therefore exempt from the regulations on food additives.

Yet we know enough to have valid concerns. The plant genome is not like a Lego set; it is more like an ecosystem. You simply cannot predict the safety of gene inserts unless you do the testing.

Most GM food studies have been generated by industry and it is the industry itself with sole access to so much of the data. There is little funding of independent studies on the effects of GM foods, and those few scientists who have engaged in such studies and reported concerns are discounted. Their concerns cannot be resolved without serious and independent scientific study.

We are particularly concerned that at a time when conflicts of interest have become a major concern in science that the AAAS Board would not openly divulge that some in the AAAS leadership appear to have longstanding ties to the biotech industry. Since these ties have not been transparently disclosed, it is unclear whether there could also be ties to industrial concerns that might influence decision making of the AAAS leadership. Surely any reader of their position is entitled to such facts in considering their position. We advocate for full disclosure of all such ties by AAAS leaders.

The fact that no deaths have been attributed to GM crops does not mean they are safe. We do not see deaths associated with bisphenol A (BPA) and yet there are hundreds of studies pointing to risks. Risks that consumers have carefully considered when choosing whether or not to buy products containing BPA.

We hope to explore these concerns in greater depth with future reports. For now: both of these quotations make clear that the AAAS position does not reflect a consensus of scientists, but in fact a pre-election political position taken by a small number of people.

Now, gentle reader, you may already feel the onset of Press Release Fatigue—that frustration you experience, when reading through dozens of angry bloated “statements”, that makes you want to throw your hands up and let someone else deal with the problem.

But just look at this report for a minute. Does it seem honest? Does it seem like a good source for Wikipedia's article on controversies about genetically modified food? Should it be the most frequently used source for this article? (It is cited six times in all.) Why is currently the most frequently used source?

What of the sources on which the AAAS calls? They state:

The EU, for example, has invested more than €300 million in research on the biosafety of GMOs. Its recent report states: “The main conclusion to be drawn from the efforts of more than 130 research projects, covering a period of more than 25 years of research and involving more than 500 independent research groups, is that biotechnology, and in particular GMOs, are not per se more risky than e.g. conventional plant breeding technologies.”

Could this be true? Did the European Union, notorious for resisting the influx of GMO food from the USA, finally conclude that genetic modification is safe?

The document cited is actually a compilation of essays, only one of which pertains to “GMOs and Food Safety”. The author is Harry A. Kuiper, who reiterates the argument that the European testing regime is strong and will be able to detect and block potentially harmful organisms. He writes:“Extensive research on GMOs, co-funded by the European Commission over the last two decades, has significantly contributed to being able to identify and characterize possible risks associated with foods/feed derived from GMOs.”

Recall that the AAAS promised an equivalence between genetically modified and conventional foods. Kuiper does write: “These activities provide at least equal assurance of the safety of these foods compared to conventional counterparts, provided these GM products have been approved by the EU and the national food safety evaluation procedures.”

Just as we saw in Ronald's work, above, the AAAS has distorted the claim in its reference, ignoring the role of regulators in guaranteeing safety. This is not a minor omission—it is a complete omission, because the documents being cited do not review the literature on GMO safety. They only discuss a best-case scenario for European regulators.

There is no particular reason to believe in this best-case scenario, especially because Kuiper is a high-ranking regulator—or deregulator—who has pushed lower GMO standards since 2003.

Testbiotech (a small group of researchers led by Christoph Then) and Corporate Europe Observatory describe their complaint against Kuiper's position:

Testbiotech, supported by Corporate Observatory Europe (CEO), is today filing a new complaint with the EU Ombudsman questioning the independence of the chair of the panel of experts tasked with assessing the risk of new genetically engineered plants entering the European Union. Harry Kuiper has chaired the GMO Panel at the European Food Safety Authority (EFSA) since 2003 but has also maintained strong ties with International Life Sciences Institute (ILSI) including taking part in a task force led by a Monsanto employee. ILSI is funded by the food and agrochemical industry and Kuiper's work on the task force was alongside staff from Bayer, Dow AgroSciences, Dupont and Syngenta, all of which produce genetically engineered plants.

Testbiotech research has shown that the work of this ILSI task force has directly influenced Kuiper’s work at EFSA. Kuiper is expected to leave the GMO panel within the next few months as he comes to the end of his term. Christoph Then of Testbiotech said: “We urgently need more clarity. Harry Kuiper has been involved in each and every case of risk assessment of genetically engineered plants since the start of EFSA. The public has a right to know if consumers and the environment were really protected in the best possible way.”

Nina Holland from Corporate Europe Observatory (CEO) added: "Harry Kuiper's position as a chair of the GMO Panel is a clear case of conflict of interest. This raises important questions about the decisions made while he was chair and we want the Ombudsman to investigate this. "

Again, even if Kuiper were completely correct, his arguments fall far short of demonstrating a consensus on GMO safety. We have here another case of one person misleadingly portrayed as thousands of scientists.

Footnote 2 cites a political resolution passed in December 2012 by the American Medical Association.

This resolution, too, came in advance of the California vote on GMO labeling. The document is much more comprehensive than the AAAS statement (and even compared to the EU report cited therein).

The resolution says that no links have been found between genetic engineering and health problems, and that although some risks exist they are small. It argues that there is low risk to human health from certain scenarios, including horizontal gene transfer to humans, toxic transgenes, and allergens. Its statement on allergens is the most definitive: "To date, no evidence has supported an increased degree of allergenicity of bioengineered foods compared to their non-bioengineered counterparts.” It does not evaluate the possible effects of glyphosate (the main ingredient of Monsanto's Roundup), or of various other identified risks of genetic engineering.

The resolution makes reference to a 1987 position by the National Academy of Sciences (home of the National Research Council) which suggested that there was no inherent reason to scrutinize genetically engineered crops more than conventional crops. This organization has itself been forced to revise this assessment, as we saw in the 2004 report discussed above.

However: The AMA resolution does not say that genetically engineered foods are safe, or that there is a broad scientific consensus on the matter.

The conclusion of this statement opposes labeling but says that stronger regulation is necessary: “Council believes that pre-market safety assessment should shift from a voluntary notification process to a mandatory requirement.”

Several people noted that the AMA's recommendations seemed strangely contradictory. Marion Nestle wrote: “Here's what surprises me: in recommending premarket safety testing, which is not now required, the AMA appears to be raising serious questions about the safety of GM foods. If such doubts exist, shouldn't GM foods be labeled so the public has a choice?”

Ronnie Cummins of the Organic Consumers Association (a group which did also have a financial stake in the GMO labeling vote) said:

We are disappointed, and frankly confused. On the one hand, the AMA is telling consumers that GMOs should be tested for any potential health hazards, before food manufacturers are allowed to sell them to the public. On the other hand, they’re effectively saying that it’s OK that products containing GMOs are not labeled. It makes no sense to acknowledge enough doubt about the safety of genetically engineered ingredients to recommend pre-market testing, but disagree that consumers should have the right to know which foods contain GMOs. Shouldn’t consumers be able to avoid GMOs unless they have first been proven safe?

In fact, the AMA's argument against labeling is this: “... the FDA cannot require labeling based solely on differences in the production process if the resulting products are not materially different or do not pose a safety risk”. It would, in this view, be unfair to label all genetically modified foods when only some of them actually constitute a risk. However, the current voluntary regime of regulations does not, according to the AMA, do a good job of determining which GMOs are risky. Therefore, goes the logic, labeling is wrong but more pre-market regulation is necessary.

An interesting position with debatable merits—but not support for the “broad scientific consensus” theory.

Footnote 3 is “20 questions on genetically modified foods” at the World Health Organization website—specific authorship unknown. Two excerpts:

Q8. Are GM foods safe?

Different GM organisms include different genes inserted in different ways. This means that individual GM foods and their safety should be assessed on a case-by-case basis and that it is not possible to make general statements on the safety of all GM foods.

GM foods currently available on the international market have passed risk assessments and are not likely to present risks for human health. In addition, no effects on human health have been shown as a result of the consumption of such foods by the general population in the countries where they have been approved. Continuous use of risk assessments based on the Codex principles and, where appropriate, including post market monitoring, should form the basis for evaluating the safety of GM foods.

Q15. What is the state of public debate on GM foods in other regions of the world? The release of GMOs into the environment and the marketing of GM foods have resulted in a public debate in many parts of the world. This debate is likely to continue, probably in the broader context of other uses of biotechnology (e.g. in human medicine) and their consequences for human societies. Even though the issues under debate are usually very similar (costs and benefits, safety issues), the outcome of the debate differs from country to country. On issues such as labelling and traceability of GM foods as a way to address consumer concerns, there is no consensus to date. This has become apparent during discussions within the Codex Alimentarius Commission over the past few years.

There's nothing here we haven't already discussed. There are no sources, author, or date. I don't know why this source is included. By the way, a substantial report located in the same region of the WHO website states:

Conflicting assessments and incomplete substantiation of the benefits, risks and limitations of GM food organisms by various scientific, commercial, consumer and public organizations have resulted in national and international controversy regarding their safe use as food and safe release into the environment. An example is the debate on food aid that contained GM material offered to countries in southern Africa in 2002, after 13 million people faced famine following failed harvests. This international debate highlighted several important issues, such as health, safety, development, ownership and international trade in GMOs.

Such controversies have not only highlighted the wide range of opinions within and between Member States', but also the existing diversity in regulatory frame works and principles for assessing benefits and risks of GMOs.

Footnote 4 is the 2004 National Research Council report discussed above. Remember this chart?

A comment in the footnote reads: “See pp11ff on need for better standards and tools to evaluate GM food.” Presumably this message assists readers who find this footnote after reading the claim: “There is a view from many of the scientists and regulators who support GM food that there is a continuing need for improved testing technologies and protocols to identify and manage risk better .” That's just fine.

But nothing in this study supports the claim of safety equivalence between GM and conventional food—let alone the claim of “broad scientific consensus”.

We should be troubled by the fact that this article is misused in the same way by Ronald and on Wikipedia article.

Footnote 5 is A decade of EU-funded GMO research (2001-2010) We've already discussed how this reference takes us to Harry Kuiper and the world of European regulatory agencies. Again, we should be dismayed by the pattern of misuse—this time mimicking the AAAS statement.

Footnote 6 brings us another six studies.

& First is “Safety of Genetically Engineered Food”, a fact sheet coming out of UC Davis in 2006.

The paper mentions several incidents of toxicity resulting from genetically engineered foods, concluding in each case that the genetic engineering itself was not the problematic element. (It discusses the “Starlink” incident descirbed above, as well as a faulty batch of genetically engineered L-tryptophan from Japan.)

This document is three pages long. It does conclude: “While genetic engineering of foods continues to generate concern and controversy for some consumers, evidence to date has not indicated that any foods developed for human consumption using genetic engineering techniques pose risks greater than foods produced using traditional methods.”

Simple enough. This is probably the best time to use the saying “absence of evidence does not imply evidence of absence”. It is quite a short paper, doesn't do much of a literature review, and doesn't address the issue of consensus. Still: of all the sources so far, this one is neither greatly abusive nor greatly abused.

Next is Pamela Ronald's article in Genetics. We've already talked about that article and delved into the sources it misuses.

Henry I. Miller's “A golden opportunity, squandered” uses the phrase “broad scientific consensus”! Look!

There is absolutely nothing about Golden Rice that should require endless case-by-case reviews and bureaucratic dithering. As the journal Nature editorialized in 1992, a broad scientific consensus holds that ‘the same physical and biological laws govern the response of organisms modified by modern molecular and cellular methods and those produced by classical methods....[Therefore] no conceptual distinction exists between genetic modification of plants and microorganisms by classical methods or by molecular techniques that modify DNA and transfer genes.’ [2]

1992 is kind of a throwback for a 2009 article, but maybe Nature really had a point. Even if we take Miller's claim at face value, however, he is not making a blanket statement about the safety of genetically modified foods. He is claiming that the process of genetic engineering incurs no additional risk. But let's dig a little deeper.

When I do a search for the editorial I find that it is cited often and almost only by Henry I. Miller. The half-page editorial [sorry, paywall; check your local library] itself is an endorsement of the George H.W. Bush administration's new policy on biotechnology: the “substantial equivalence” rule that has exempted genetically engineered foods from government testing. The editorial (following a warning against painting the US as a “black sheep” on climate change or forcing an “uneconomic use of resources”) reads:

The Administration of US President George Bush has just issued a policy on the regulation of biotechnology that is utterly in keeping with good science. Perhaps it should not be surprising that this is so, but for the past two decades biotechnology has gained such a reputation as a boeygman of science that it is refreshing to see that clear thinking prevailed at the White House, where a biotechnology policy has been in the works for more than a year.

The policy, which is meant to inform the way individual regulatory agencies handle biotechnology products, states that “the same physical and biological laws govern the response of organisms modified by modern molecular and cellular methods... [Therefore] no conceptual distinction exists between genetic modification of plants and microorganisms by classical methods or by molecular techniques that modify DNA and transfer genes.”

Yes, you read that correctly. The words quoted by Miller are not attributed to a “consensus” of any sort; they are not even those of the Nature editorial board—they are a quotation from a draft version of a permissive government policy.

Actually Nature also mis-attributes the quotation, which they have spliced together in the wrong order from a 1989 handbook (p. 15 and p. 14) published by the National Research Council! You'll recall that from above that the National Research Council had by 2004 completely reversed its position on this very particular issue. (Don't forget that chart!)

The Nature editorial continues in praising this policy, discounting the fears of skeptics, and repeating the argument that conventional foods and pesticides are also dangerous. Here is the closest it comes to discussing the safety of genetic modification:

It is not surprising that biotechnology products, particularly those released into the environment by the agricultural and chemical industries, have elicited strong negative reactions from environmental groups, as well as from ordinary citizens. After all, the very scientists who developed recombinant DNA technology were the ones who alerted the public to its potential hazards, particularly if gene-spliced organisms were to multiply out of control. But 20 years of real-life experimental and commercial science has shown those fears to be largely baseless, while the benefits of the technology (creating herbicide-resistance crops, for instance) are easy to identify.

It also says: “The US policy also acknowledges one of the important scientific truths about modern biotechnology products. They may be safer than their conventionally derived counterparts, largely because their characteristics—often down to the level of DNA sequences—are so thoroughly known.” New evidence (much of it published in Nature) has since shown this article of faith to be wrong in several ways.

But, to recap: in 2009, Henry I. Miller made a false claim about a 1992 Nature editorial, attributing to a “broad consensus of scientists” text which had been questionably reproduced from a 1989 NRC report. And then someone on Wikipedia added the article to support a claim that not even Miller had made.

Moving along.

Bett, Ouma & De Groote paper: “Perspectives of gatekeepers in the Kenyan food industry toward genetically modified food” (2010). [Paywall!]

If you just looked at the title of this paper, you might think it's about attitudes of “gatekeepers in the Kenyan food industry” and not about scientists.

OK, that's true. The paper is about convincing Kenyans that GMOs are good for them. But it does contain this sentence: “Empirical evidence shows the high potential of the technology, and there is now a scientific consensus that the currently available transgenic crops and the derived foods are safe for consumption (FAO, 2004).” Boom! What's “(FAO, 2004)”? (And isn't that a funny use of “is now”, in 2010?)

Well, FAO 2004 is this report from the UN Food and Agriculture Organization: “]http://www.fao.org/docrep/006/Y5160E/Y5160E00.htm The State of Food and Agriculture 2003 – 2004: Agricultural Biotechnology: Meeting the needs of the poor?]”.

It's a big report, so let's skip to the section on “Health and environmental impacts of transgenic crops”. They did used some materials from an industry funded biotech research center. Let's see what they have to say.

Currently available transgenic crops and foods derived from them have been judged safe to eat and the methods used to test their safety have been deemed appropriate. These conclusions represent the consensus of the scientific evidence surveyed by the ICSU (2003) and they are consistent with the views of the World Health Organization (WHO, 2002). These foods have been assessed for increased risks to human health by several national regulatory authorities (inter alia, Argentina, Brazil, Canada, China, the United Kingdom and the United States) using their national food safety procedures (ICSU). To date no verifiable untoward toxic or nutritionally deleterious effects resulting from the consumption of foods derived from genetically modified crops have been discovered anywhere in the world (GM Science Review Panel). Many millions of people have consumed foods derived from GM plants - mainly maize, soybean and oilseed rape - without any observed adverse effects (ICSU).

So they claim there's a consensus on safety... but as usual they're not the ones who did the literature review. We'll check out the report that they cite shortly. We might also be a little suspicious of the claim that safety testing is working for the US, since due to the “substantial equivalence” rule, there is no safety testing. But check out what the report says next:

The lack of evidence of negative effects, however, does not mean that new transgenic foods are without risk (ICSU, GM Science Review Panel). Scientists acknowledge that not enough is known about the long-term effects of transgenic (and most traditional) foods. It will be difficult to detect long-term effects because of many confounding factors such as the underlying genetic variability in foods and problems in assessing the impacts of whole foods. Furthermore, newer, more complex genetically transformed foods may be more difficult to assess and may increase the possibility of unintended effects. New profiling or “fingerprinting” tools may be useful in testing whole foods for unintended changes in composition (ICSU).

What? Seriously? So genetically modified foods are “safe to eat” even though the long-term health effects are unknown? What do they even mean by “safe to eat”? It's almost as if someone came in and rewrote the first part of the report to make genetic modification sound safer than it really is.

Does the ICSU report support FAO's conclusion? This is a 2003 report written by “G. J. Persley, The Doyle Foundation for The International Council for Science”. G. J. Persley is the head of the Doyle Foundation. (The Doyle Foundation is named for late World Bank consultant Jack Doyle; G. J. Persley was his spouse. The “ICSU report” seems to be their most recent publication.)

Persley writes:

Currently available genetically modified foods are safe to eat. Food safety assessments by national regulatory agencies in several countries have deemed currently available GM foods to be as safe to eat as their conventional counterparts and suitable for human consumption. This view is shared by several intergovernmental agencies, including the FAO/WHO Codex Alimentarius Commission on food safety.

So Perley's argument is once again not about scientific consensus but about government regulators. Government regulators say GMOs must be safe, so they must be safe—even though the US government does not regulate GMOs, based on rules passed against the objection of its own scientists. And as usual, Persely doesn't conduct a literature review but instead cites the OECD, whose report from 2000 we've already discussed.

Consumers International said this about the FAO report:

 Reported successes are not the result of GM. The FAO report champions the successful use of molecular markers for pearl millet in India, tissue culture for virus-free planting stocks of bananas in Kenya and the eradication of rinderpest. However, the report fails to emphasise that these processes do not involve any genetic engineering or genetic modification.

• A biased, outmoded and unilateral assessment. The report does not seriously consider any downside to agriculture biotechnology in general and transgenic technology in particular. It suggests genetic engineering alone can feed the poor and clean up the environment and pays little attention to the widely acknowledged alternatives.
• The report contradicts the FAO’s own findings. The chapter: ’From the Green Revolution to the Gene Revolution’ pays no attention to the environmental and social downsides of the initiative. This is despite the FAO regional office in Asia’s long-held acknowledgement of the downsides of the Green Revolution and the need to look for more ecologically and socially rational forms of agriculture.
• A betrayal of its own endeavour. The FAO has previously put out some excellent papers on the ethics of biotechnology and its recent consultations on GM have been well balanced. Its latest drive to promote SARD (sustainable agriculture and rural development) should also be commended. This report falls well short of FAO standards and is a stain on its reputation.
• The absence of a meaningful consultation process has left many NGO’s feeling angered. This comes after the FAO agreed to a multi-stakeholder process to develop the concept of food sovereignty, including consultation with NGOs and rural people’s organisations.

Dominic Glover explained in 2010:

The period around the turn of the twenty-first century was punctuated by the release of a succession of weighty reports by major international organisations and august scientific institutions, which encouraged the development and commercialisation of genetically modified (GM, transgenic) crops to improve developing-country agriculture (FAO 2004; IFAD 2001; IFPRI 1999; Nuffield Council on Bioethics 1999; Royal Society of London et al. 2000; UNDP 2001). Although they were sprinkled with qualifications about careful safety assessment and socio-economic factors, these documents nevertheless appeared to represent an emerging scientific and policy consensus that GM crop technology would be ‘pro-poor’.

That optimistic consensus depended on a number of key, unacknowledged and often questionable assumptions about the ways in which the technology would be developed and its likely impacts on poverty, hunger and the livelihoods of the poor (Levidow 2001; Scoones 2002a, 2007).

Logical and tolerant as we are, we might not dismiss these reports simply because they are based on an effort to market genetically modified products to the Third World. However, we have seen that the reports themselves do not provide evidence of a scientific consensus about anything; they simply attribute this finding of consensus to other reports. This consensus is about as real as Orbis Tertius—but more dangerous.

Quan, McCluskey, & Wahl, “Effects of Information on Consumers' Willingness to Pay for GM-Corn-Fed Beef”, 2004

Another one of many studies on how to market GMO products to regular people—by telling those people that GMO products are safe. “There is general agreement among scientists on the safety of meats from animals fed on GM feeds.” This is quite a different thing than the safety of genetically modified foods themselves.

As it happens, the source cited is a press release from the group “Internutrition” criticizing the government of Switzerland for its moratorium on genetically modified products. As far as I can tell, this press release only says that some scientists (and others) expressed “their doubts” (“leurs doutes ”) during the political process. The press release cites no reports or other documents whatsoever.

Internutrition is an biotech consulting group.

Let's take this moment to consider the arguments made by a swarm of sources in 2000 – 2004 that GM products “currently on the market” are safe because they have been approved by government regulators. Should we infer that if regulators have not approved the products as safe, then they might not be safe? Can something be “safe” for Americans but not for the Swiss? Epistemologically modified food for thought.

Our last source might also be the most interesting. First, because it's the only literature review in the whole list of 13 citations. Second, because of the source.

On Wikipedia, this citation gives a date of 2011. This date is obviously wrong, as none of the references listed were published after 2004.

We have known since 2002 that AgBioWorld is secretly hosted by The Bivings Group, a consulting firm employed by Monsanto to spread disinformation over the internet. (More on The Bivings Group later. Interestingly, 2011 is also the year that The Bivings Group shut down, to be regrouped into “The Brick Factory”. Maybe they were too busy to do an update.)

The post is written Christopher Preston, Senior Lecturer in Weed Management at the University of Adelaide.

“We noted the excellent contribution by Dr. Chris Preston, 'Peer Reviewed Publications on the Safety of GM Foods,' AgBioView, Dec. 3, 2004”, wrote two Monsanto representatives on the same website.

So that gives you some idea of what you're looking at … and answers the date question.

Arpad Pusztai had this to say:

The next comment is that they looked at academic as opposed to production studies. These latter have very little scientific value; we used to call them at the Rowett Institute: "feed them and weigh them".

The next point is that my Nutrition and Health (2002) paper is not a review but, rather interestingly, Dr Preston did not mention our 2003 review that was published just at the same time as the Pryme & Lembcke review (even though I gave the reference to it in my previous comments on his assertions that he must have received because he did publicly respond to one of the points) but which also included more papers with analytical comparisons between GM and parent lines.

I think Dr Preston's list is quite revealing in terms of his scientific approach to this topic, particularly as regards the failure to distinguish between a scientific study and an animal production exercise. When I was asked by Professor Mosenthin to write my next review (to be published next year) he emphatically asked me to leave out all production studies from the review as these may be of some value to commercial animal production but have limited scientific value.

And by the way, says Pusztai:

Actually, Dr Preston missed two Malatesta papers, perhaps because they both show bad effects on the liver and the pancreas of mice fed RR soya, and quite a few others, but for these he will have to read my new review next year.

Key points so far:

• None of them provide evidence of a “broad scientific consensus” that “food on the market derived from GM crops pose no greater risk than conventional food.”

• A number of these sources, like Pamela Ronald's article, make completely false claims based on egregious misrepresentations of the sources they cite.

• Others do not make false statements—but neither do they support the claim made on Wikipedia that “There is broad scientific consensus that food on the market derived from GM crops pose no greater risk than conventional food.”

• Some of these reports play a shell game, referring back to other reports, which refer to still others, and so on.

• For a surprisingly large number of these documents, the ultimate source for their claim of comparative GMO safety is the fact that government regulators have approved genetically modified food for sale on the market. Within this rationale, the claim that "food on the market derived from GM crops pose no greater risk than conventional food" is at its very best tautological, since the food is on the market because regulators allowed it onto the market.

• The rule adopted by the US government (in 1992) dictates that genetically modified foods are "substantially equivalent" to conventional foods and therefore do not require independent testing. The existence of this rule does not constitute a scientific consensus, and it is reprehensible for anyone to represent it as such. We hope to delve further into the "substantial equivalence" rule—how it was adopted, how it was used, and how it is portrayed—in future reports.

• It is imperative that Wikipedia not misrepresent the position of government regulators as the position of a "broad consensus of scientists"—yet this is precisely what has been done by some editors (in citing the WHO) and by some of the authors described above.

• These sources do not address the safety hazards associated with eating residues from Monsanto's “Roundup” herbicide—identified currently as appropriate to include within the page on "genetically modified food controversies" but not acknowledged as a potential source of risk.

• In most cases, there were clear vectors of influence that might explain why the author promoted the conclusion that they did. We do not need to consider these sources of bias as independently disqualifying factors; however, they help paint a picture that explains some of the misrepresentation we are witnessing.

• Incidentally, the claim that “No reports of ill effects have been documented in the human population from GM food” is also utterly wrong but not our focus right now.

Based on the inadequacy of these sources alone, Wikipedia should not claim that there is a “broad scientific consensus” on the comparative safety of genetically modified foods.

Thank you for reading—and perhaps responding with appropriate action. Pleased stay tuned for future reports on this issue. groupuscule (talk) 05:56, 12 June 2013 (UTC)

Added 19:37, 14 June 2013 (UTC)

[Discussion of individual users deleted, as per consensus, 18:53, 8 July 2013 (UTC)]

I think it's also important to respond to the claim—made by all three primary critics of the report—that what I have done is original research that does not belong on Wikipedia, even in userspace.

For guidance, I turn to the policy pages Wikipedia:No original research and Wikipedia:Identifying reliable sources.

It is worth mentioning that these pages are primarily guidelines for article content, not discussion content. Talk pages and user spaces are still part of Wikipedia, and there are risks involved in publishing essay-like texts in these spaces. (More on this in the next section.) I maintain, however, that the analysis provided in the above report is not only a legitimate form of discourse on Wikipedia, but mandated by our pursuit of reliable sources.

Under “Context matters”, the Wikipedia policy states: “The reliability of a source depends on context. Each source must be carefully weighed to judge whether it is reliable for the statement being made and is an appropriate source for that content. In general, the more people engaged in checking facts, analyzing legal issues, and scrutinizing the writing, the more reliable the publication. Sources should directly support the information as it is presented in an article.” [Original emphasis]]

Do the sources presented in the articles on genetic engineering “directly support” the claims being made? Are they “reliable” in context? The analysis above is precisely an effort to “carefully weigh” these very questions. Again, this type of discussion does not belong in an article, but is a necessary part of editing these articles.

I apologize for any controversial statements in the Part 1 that might have seemed like unsourced opinion. I tried to limit the discussion to the sources already presented in the article and secondary sources that commented on these directly. Part 2 contains many reliable secondary sources which confirm the assertions in Part 1—most saliently, the assertion that “consensus” often stems from the pronouncements of government regulators and not from the current of scientific opinion.

Wikipedia has an independent policy page on topics related to human health. This is Wikipedia:Identifying reliable sources (medicine). This policy enjoys special emphasis because: “Wikipedia's articles, while not intended to provide medical advice, are nonetheless an important and widely used source of health information.[1] Therefore, it is vital that the biomedical information in all types of articles be based on reliable, third-party, published sources and accurately reflect current medical knowledge.”

Claims that a certain category of food is definitely just as safe as any other food should therefore receive the highest level of possible scrutiny. Current sources do not meet these standards. The medical advice policy does caution against using individual trials as indicators of the state of scientific opinion—this is an important guideline to remember for editors raising questions about GMO safety. The policy reads:

The best evidence comes primarily from meta-analyses of randomized controlled trials (RCTs).[2] Systematic reviews of bodies of literature of overall good quality and consistency addressing the specific recommendation have less reliability when they include non-randomized studies.[3] Narrative reviews can help establish the context of evidence quality.”

None of the thirteen articles cited in support of the “safety” claim meet these criteria. (Frankly, I am surprised that this safety claim did not long ago raise red flags for the vigilant editors of WikiProject Medicine. Perhaps a claim of this sort was not expected on these pages.) Some articles that do fit these descriptions are reviewed in Part 2.

Other important elements of policy:

• “Care should be taken with journals that exist mainly to promote a particular point of view.” This concern is reasonable and raises questions about journals on both sides of the debate over GMOs.

• “Editors should also consider whether the bias makes it appropriate to use in-text attribution to the source”. This policy is reasonable and commonly followed; it highlights the exceptional status of a claim like “broad scientific consensus” expressed in Wikipedia's editorial voice. And of course it also indicates that we may wish to attribute, in text, sources who challenge the safety of GMOs.

I am not happy about having to create an independent page to discuss these issues. I am sensitive to the risk of using talk pages and user pages as platforms (i.e. “soapboxes”) for personal opinion. I certainly would not use my userspace to publish the results of my own experimentation with plants or human health. Nor would I use this space to write political essays (about, for example, the “Occupy Movement”, of which, as my user page makes clear, I am a not-uncritical supporter.) I have tried very hard to avoid polemicism and emotional appeals—for example, I deleted the image File:Blister roundup.jpg from a draft of Part 2 because even though I believe the image ought to be seen (at Glyphosate or Monsanto) I felt it would have a disproportionate emotional impact here.

Ultimately, creating this page seemed necessary in order to conduct sustained analysis of the issues at hand. Legitimate concerns, raised by multiple editors, are routinely drowned out by a chorus of voices saying “GMOs are safe”. (Or, perhaps worse, equivocating and pretending to argue about distinctions—i.e. “just as safe as conventional foods”—while ignoring serious issues.

Now we're going to go further, investigating the question of who thinks genetic modification might not be comparatively safe. We will also look at different reasons why some scientists don't think all genetically modified food on the market is comparatively safe.

We are going to move from some very specific evidence of dissensus (scientists who don't agree with the claim) into more general evidence (claims of paradigm shift in biology). In each section, we will make sure to cite reliable sources that connect the issue under discussion directly to the safety of genetically engineered food.

The American Academy of Environmental Medicine has resolved: “GM foods have not been properly tested for human consumption, and [...] there is ample evidence of probable harm “ (You can see who's on their board of directors.)

Although organization policy statements are not necessarily good indicators of consensus (as we have seen above), they do represent nice counterpoints to the AAAS statement discussed above.

Many scientists have signed this letter calling for a moratorium on the release of genetically modified organisms. It states:

The hazards of GMOs to biodiversity and human and animal health are now acknowledged by sources within the UK and US Governments. Particularly serious consequences are associated with the potential for horizontal gene transfer. These include the spread of antibiotic resistance marker genes that would render infectious diseases untreatable, the generation of new viruses and bacteria that cause diseases, and harmful mutations which may lead to cancer.

Some of the signatures on this website linked above are bogus, and the official count is therefore not meaningful. (You can find a verified list of the first 85 signatories here.) We should still note the endorsements of Ruth Hubbard, David Suzuki, Vandana Shiva, George M. Woodwell, and Mae-Wan Ho.

Many others (with some overlap) have signed a letter in support of Gilles-Eric Seralini, yet another scientist who has been attacked in the press because he found that genetically modified crops might be dangerous to animals.

Brian Wynne has specifically challenged claims of consensus on GMOs. Wynne objected so strenuously to (what he perceived to be) the falsely promoted consensus that he (and Helen Wallace) dramatically resigned from a GMO “dialogue” held by the UK Food Standards Agency.

Like organisational statements, multiple data points about individual scientists cannot themselves prove dissensus—but they do help to paint the picture.

According to a recent article in the Egypt Independent, more and more scientists are validating claims made by Putzstai and Seralini that GMOs can be toxic enough to damage the liver and other organs, and to weaken the immune system.

On an international level, some of the world’s most renowned scientists from various independent research institutes, such as the Rowett Institute in Scotland, the Russian Academy of Sciences in Moscow and the CRIIGEN in Paris, have conducted similar genetically modified food experiments on rats, mice and other animals.

Their results corroborate Kaoud’s observations: The rodents had reproductive problems, immune system issues, accelerating aging, cholesterol, organ damage and gastrointestinal problems.

A load-bearing pillar in many of the arguments for a “broad scientific consensus” on comparative GMO safety is the notion that US regulators have approved GMOs as comparatively safe. By itself, this s claim puts the cart before the horse: the rule about “comparative safety” has precluded any actual investigations. But one might at least point to the regulators who approved the policy of treating genetically modified crops just like others.

In fact, however, the FDA's own scientists do not believe that this policy is safe and objected to this policy even as it was being passed.

William Freese & David Schubert discovered this hidden disagreement and published their findings in 2004. Freese summarizes:

The U.S. regulatory agency most commonly cited as vouching for the safety of GM foods exercises the least authority in regulating them. Theoretically, the novel transgenic proteins in GM crops fall under the “food additives” provisions of the FDCA. Food additives must undergo extensive premarket safety testing, including long-term animal studies, unless they are deemed “generally recognized as safe” (GRAS). Biotech companies have successfully claimed GRAS status for all of their new GM proteins (and by extension, the GM crops that contain them). FDA has yet to revoke an industry GRAS determination and require food additive testing of any transgenic crop. This blanket GRAS exemption is based on the notion of “substantial equivalence”—the strong, a priori presumption that GM crops are essentially the same as their conventional counterparts. Interestingly, this policy allowing industry to police itself was established in the face of considerable opposition from FDA working scientists. Some called for mandatory review of every new GM crop; others for toxicology studies. Dr. Louis Pribyl, an FDA scientist, was particularly concerned about unintended effects from the haphazard introduction of foreign genetic material. Administrative superiors at FDA and the White House apparently did not heed these concerns, resulting in today’s voluntary consultation process.

Under voluntary consultation, the GM crop developer is encouraged, but not required, to consult with FDA. The company submits to FDA the conclusions of any research it may have conducted, but not complete studies. FDA never sees the methodological details of the company’s research, which is essential to identify unintentional mistakes, errors in data interpretation, or intentional deception. Even within this lax, voluntary system, companies have sometimes failed to comply with FDA’s occasional requests for additional data.

Contrary to popular belief, then, FDA has not formally approved a single GM crop as safe for human consumption. Instead, at the end of the consultation, FDA merely issues a short note summarizing the review process and a letter that conveys the crop developer’s assurances that the GM crop is substantially equivalent to its conventional counterpart. Under this voluntary system, FDA cannot fulfill its role of reviewing GM foods for the presence of toxins or allergens, alterations in nutritional content, or unintended effects of genetic engineering.

So, there are many scientists who dispute the safety of GMOs. But maybe they're outliers! There are indeed a lot of scientists. So has anyone explicitly refuted the claim of “consensus”?

Yes. Plenty of authors make explicit mention of this claim in order to refute it.

Freese & Schubert, the researchers who uncovered dissent at the FDA, directly support the claim made in Part 1 about government decisions being misrepresented as science:

A thorough understanding of how GE foods are currently regulated is essential because claims regarding the safety of these crops are based largely on assessments by government regulators, which in turn are founded mostly on unpublished studies conducted by the crop developer.

Tom Philpott (farmer, Grist and Mother Jones author) says the claim is “clearly false”. In a column unambiguously titled “Scientific Consensus on GM is an Illusion”, He writes:

Widely compared to the Intergovernmental Panel on Climate Change (IPCC), which definitively established a scientific consensus around climate change on its release in 2007, the IAASTD engaged 400 scientists from around the globe under the aegis of the World Bank and the UN’s Food and Agriculture Organization. According to the Executive Summary of the Synthesis Report, the effort was originally “stimulated by discussions at the World Bank with the private sector and nongovernmental organizations (NGOs) on the state of scientific understanding of biotechnology and more specifically transgenics.”

If transgenic-crop technology had captured the broad approval of the global agricultural-science community, here was the place to show it. But what happened? According to the Executive Summary of the Synthesis Report:

Assessment of biotechnology is lagging behind development; information can be anecdotal and contradictory, and uncertainty on benefits and harms is unavoidable. There is a wide range of perspectives on the environmental, human health and economic risks and benefits of modern biotechnology; many of these risks are as yet unknown. [...]

The application of modern biotechnology outside containment, such as the use of genetically modified (GM) crops, is much more contentious [than biotechnology within containment, e.g., industrial enzymes]. For example, data based on some years and some GM crops indicate highly variable 10 to 33 percent yield gains in some places and yield declines in others.

The report goes on to call for a whole new framework for crop-biotechnology research—an implicit rebuke to the current one:

Biotechnologies should be used to maintain local expertise and germplasm so that the capacity for further research resides within the local community. Such R&D would put much needed emphasis onto participatory breeding projects and agroecology.

Thus, whereas the IPCC revealed broad agreement among the global scientific community around climate change, the IAASTD—arguably the “IPCC of agriculture”—showed deep ambivalence among scientists over transgenic crops.

The real question becomes: How can serious publications like Seed claim that skepticism toward GMOs reflects a “scientific flip-flop”? To be sure, the illusion of a broad consensus holds sway in the United States, and the IAASTD has clearly failed to correct it. The US media greeted its release with near-complete silence—in stark contrast to its reception in the European media.

For explanation, Philpott turns to peer-reviewed research published by Don Lotter in the International Journal of the Sociology of Food and Agriculture.

Here's what Lotter says about the state of scientific consensus:

A major conflict is imminent in science. On the one side are scientists, universities and corporations who have invested nearly 25 years and tens of billions of dollars in the genetic engineering of organisms (transgenics), mostly bacteria and plants, for food, pharmaceutical, and industrial uses. On the other side is a flood of evidence that food plant transgenics – not bacterial or pharmaceutical plant transgenics – is fatally flawed and has been resting on a theoretical foundation that has crumbled away as the science of genetics reinvents itself. Adding to this side is a worldwide grass-roots movement opposed to genetically engineered foods.

Lotter's work has been questioned because it was published in the International Journal of Sociology of Agriculture and Food and was therefore peer reviewed by sociologists—not biologists. This rejection is dubious, given the sources currently used to support the claim of “consensus”. More importantly, “consensus” is inherently a social phenomenon, and the issue has sociological aspects as well as toxicological ones. If you peruse the journal's website, you will see that it encourages rigorous interdisciplinary studies of agriculture issues, discussing scientific results as well as government behavior. You will also see that “All articles published in this journal have undergone internal editorial scrutiny and external, triple-blind peer review.”

Lotter himself is highly qualified to discuss the issue. As Bonnie Powell writes:

Lotter has a Ph.D. in agroecology from the University of California, Davis, and a master of professional studies in international agricultural and rural development from Cornell University. At various times he has taught environmental science, soil science, plant science, entomology, and vegetable crop production for Santa Monica College, Imperial Valley College, and UC Davis. He is not tenure track and said in a phone interview last night that this paper certainly "wasn't going to help my chances of getting a job" (for reasons that will become clear in a bit). He was recently a visiting scholar in the department of plant pathology at Colegio Postgraduados in Chapingo, Mexico. His research on organic agriculture has been published in the Journal of Alternative Agriculture and the Journal of Sustainable Agriculture.

Global bans and limits on genetically modified crops raise an important issue for the “safety” promised by Wikipedia. The following claims do not add up:

& Foods “currently on the market” are comparatively safe. & Regulators can be trusted to identify foods that aren't safe and regulate them. & Some regulators have imposed severe limits on genetically modified foods; others are completely permissive.

Somewhere in the argument, we are playing fast and loose with the claim that regulators are ensuring safety. As John Upton recently wrote (regarding a new wave of GMO bans in response to the surprise appearance of genetically modified wheat in Oregon):

The budding global backlash is a reminder that while America is a friendly place for most GMO crops, other countries consider transgenic foods to be abhorrent. GMO wheat has not been authorized to be grown or sold anywhere in the world. Monsanto ceased efforts to market the transgenic wheat in 2005 when it became clear that America’s export-dominated market would not tolerate it.

In India:

Jairam Ramesh said he had taken note of "tremendous opposition" from state governments within India, broad public resistance and the lack of a scientific consensus. "This would be the first GM vegetable crop anywhere in the world so I have been very sensitive and I have arrived at this decision which is responsible to science and responsible to society," he said tonight.

Claims of consensus also originate mostly from the US. (Claims from international organisations dominated by the US do little to counteract this fact.) Abroad, the dissensus is still more obvious. User:Iselilja offered the following at Talk:March Against Monsanto:

In Norway, where I am from, there is no scientific consensus that GMO does not have adverse ffects on health and environment. The Norwegian law on the matter is very strict, and almost no GMO has been accepted. The politicans are guided on this issue by the The Norwegian Biotechnology Advisory Board. Here are some excerpts from a recent interview with the director of the board, Sissel Rogne, who is a professor and specialist in genetechnology and named "Academic of the Year" in Norway in 2012. The interview is headlined "Biotechnology Advisory Board calls for better GMO research":

"(Journalist): Why do Norwegian researchers disagree so much about GMOs? (Sissel Rogne): Natural sciencists are as subjective as any other, and have the ability to pick out the articles and points of view that fits with their political views. GMO field has become very politicized. There are also differences between being thorough and to be comprehensive. The treatment of GMOs is first and foremost comprehensive, and not necessarily thorough assessment of health and environmental effects. In the bottom lie assumptions that are continued without the researchers going into the material. Many articles are based on a simple and short study that is used to show that genetically modified corn is safe. I can not understand that this is good research."

"Is genetically modified food dangerous? - There is nothing to suggest that it is "dangerous". The word dangerous is seldom used in connection with GMO foods unless there is talk of allergies. The major problem of GMO food is not whether it is dangerous, but if it is safe or healthy. Food is consumed differently in different cultures and ages. For instance, the maize porridge porridge is the first staple food many babies get when they are three months, although Norwegian children do not get porridge from GMO corn. When genetically modified corn is one of the most important products on the market, one must ask whether the corn is tested out from being a food for children. Today the usual tests based on adult rats for 90 days. I'm not impressed with how scientific it is tested."

"It is alarming to observe how some are critical to critical research on GMOs. One must ask oneself whether the current tests are good enough to determine if food is safe." (End of excerpts)

According to Sheldon Krimsky, the 1975 Asilomar Conference on Recombinant DNA (a.k.a. Asilomar II) for the first time set out public guidelines on genetic modification. The Asilomar conference also laid out the “Central Dogma” model of genetics as a justification for genetic engineering. We'll have more to say about this model soon. To some degree, the Conference pacified critics of the process (including scientists) who raised the issue of unintended effects.

Susan Wright explains:

However, while pressures to move quickly in the recombinant DNA field intensified, the controversy over its hazards continued. It was not a controversy that was easily resolved, largely because of the multidimensional complexity of the issues involved. […] In the course of the controversy, dozens of possible risk scenarios were contemplated. A dearth of empirical evidence compounded the uncertainties which emerged in analysis of these possibilities.

Given the complexity of the issues, as well as the fact that the techniques of engineering were evolving rapidly, the debate on the hazards of this field might well have continued indefinitely. In fact, it was soon restricted and ultimately closed down.

Wright notes that many sociologists of science have historically analyzed scientific consensus as a consequence of developments within the scientific community. However, she argues, the apparent scientific “consensus” on genetic engineering is better understood as a campaign, organized by a series of meetings from 1976–1978, that had more to do with reassuring the public than with confronting possible risks:

The dominant image of these meetings, as portrayed in press coverage and in official reports, is one of 'scientific' meetings with 'scientific' agendas. In fact, this analysis has shown that a principal motive for the meetings was the protection of biomedical research from external regulation. At the Enteric Bacteria meeting, the most private of the three, this motive was made quite clear, and numerous comments—with no explicit dissent—suggest that all of the participants accepted it.

Even the weak guidelines established by the Asilomar conference were systemically eroded in the 1980s. Krimsky again:

Beginning in the 1980s, the US regulatory response to biotechnology moved toward guidelines, which documents a departure from the command control regulations of the 1970s. This was a response to a pro-market, anti-regulatory shift in the political culture of government. As part of this shift, a new ideology of ‘junk science’ created a false dichotomy between ‘good science’ and ‘bad science’ to derail any attempts to use the weight of circumstantial evidence and precautionary approaches to regulate biotechnology. No new laws were passed in the United States for genetically modified organisms. Instead, laws passed to regulate chemicals were stretched to apply to GMOs. This resulted in some unusual adaptations oflanguage, such as designating a ubiquitous non-GM soil organism (Pseudomonas) a pesticide. This microbe, which resides on the leaf surfaces of plants, possesses a protein that can act as an ice-nucleating particle for super-cooled water when the temperature reaches a few degrees below freezing. When the gene that codes for this protein is excised (‘ice minus’), it no longer can serve as a nucleating site for frost formation. If the natural organism (‘ice plus’) facilitates ice formation below freezing temperatures thereby causing damage to the plant then it can be designated a pest; its GM variant (‘ice minus’) can then be thought of as a pesticide since it protects the plant from frost damage.

As demonstrated by Freese and Schubert, the drive to avoid regulations continued into the 1990s, and led the FDA to adopt the “substantial equivalence” rule over the protest of its own scientists.

Numerous literature reviews have repeatedly expressed the viewpoint that there is not enough data to pronounce genetically modified foods (even comparatively) safe. Some have gone further and suggested that there is evidence of dangers.

A particularly clear example of the deficit in evidence is the lack of animal toxicity studies to last longer than 90 days. Humans will not eat genetically modified food for 90 days—indeed it seems as though soon they will be expected to eat a genetically modified diet for 90 years. This is a major shortcoming of the existing literature that is discussed inside and outside of the scientific community.

An early literature revieew by Pryme and Lembcke (“In vivo studies on possible health consequences of genetically modified food and feed--with particular regard to ingredients consisting of genetically modified plant materials”, Nutrition and Health 17, 2003) found

In conclusion we feel that much more scientific effort and investigation is necessary before we can be satisfied that eating foods containing GM material in the long term is not likely to provoke any form of health problems. It will be essential to adequately test in a transparent manner each individual GM product before its introduction into the market.

Three major literature reviews were published in late 2011. (Unfortunately, due to time limitations, I have not not been exhaustive in my search for literature reviews. I have yet to find a detailed “review of reviews”. I believe that as per the reliable source guidelines discussed in the Addendum to Part 1 above, multiple peer-reviewed literature reviews are sufficient cause for Wikipedia to report on scientific dissensus even if other reviews may contract.) These reviews all concern experiments using laboratory animals fed with genetically modified food products. As studies in humans are few and far between, these represent a major (but not exclusive) source of information about possible risks to humans. However, note that other relevant literature reviews are discussed elsewhere in this report (e.g. in the section on toxicity of Roundup itself).

A 2011 literature review published by José L. Domingo and Jordi Giné Bordonaba in Environment International provides a realistic assessment of the research on GMO toxicity. This review provides a good explanation of the research being used to support the industry's conclusions. A number of studies have fed experimental animals with genetically modified food, compared them to a control group fed with conventional food, and reported no major differences in growth and nutrition:

Although the WHO declares that the GM products that are currently on the international market have all gone through risk assessment by national authorities, the risk assessment of GM foods in general, and crops in particular for human nutrition and health, has not been systematically performed as indicated in the scientific literature (Domingo, 2007; MagañaGómez and de la Barca, 2009). Evaluations for each GM crop or trait have been conducted using different feeding periods, animal models, and parameters. The most common result is that GM and conventional sources induce similar nutritional performance and growth in animals. However, adverse microscopic and molecular effects of some GM foods in different organs or tissues have been reported to a certain extent (Magaña-Gómez and de la Barca, 2009)

Domingo had conducted literature reviews in 2000 and 2006, finding both times that for the second time, that “if data on toxicological assessment of GM foods/plants existed, these had not been reported in scientific journals, and therefore, they were not available to the general scientific judgment.”

What's new since then?

Some studies by Séralini et al detected serious toxicity in rats fed with Monsanto's genetically modified potatoes. These results were heavily criticized by industry-funded scientists, and the community has had difficulty drawing conclusions from them.

Three varieties of genetically modified maize proved toxic to rats. Different strains seemed to be toxic in different ways. The toxicity was, to some degree, “dose-dependent”—i.e., eating more of the genetically modified maize turned out more poisonous for the rats.

Recently, de Vendômois et al. (2009) performed, for the first time, a comparative analysis of blood and organ system data from trials with rats fed three main commercialized GM maize (NK 603, MON 810 and MON 863). The authors found for the 3 GMOs new side effects linked with GM maize consumption, which were sex- and often dose-dependent. Effects were mostly associated with the kidney and liver, the dietary detoxifying organs, although different between the 3 GMOs. Other effects were also observed in heart, adrenal glands, spleen and hematopoietic system. It was concluded that these data highlighted signs of hepatorenal toxicity, possibly due to the pesticides specific to each GM corn (glyphosate and AMPA in NK 603, modified Cry1Ab in MON 810 and modified Cry3Bb1 in MON 863). In addition, unintended direct or indirect metabolic consequences of the genetic modification could not be excluded. To date, and to the best of our knowledge, this study has not been scientifically questioned.

In additional to Seralini & co., Domingo & Bordonaba identify two major groups of researchers:

With respect to recent studies on safety assessment of GM soybeans, the scientific literature shows rather contradictory results. Two research groups have been especially active in relation to those investigations. One of them, headed by Dr. Delaney from Pioneer Hi- BredInternational, Inc. (Johnston, IA, USA),has reporteddatashowing that various GM soybeans were safe. In contrast, the group headed by Dr. Malatesta from the University of Verona (Verona, Italy) has shown notable concerns. A summary of recent studies is next presented.

Domingo and Bordonaba go on to describe a number of studies which found no differences, or no major differences, between rats fed for about 90 days with genetically modified food and their counterparts fed with conventional food. There really are quite a lot of these, and you can see them summarized in the table on page 739.

In their conclusion, they write that there is now an “equilibrium” between studies asserting GMO safety and those finding risks... with an important caveat:

In spite of this, the number of studies specifically focused on safety assessment of GM plants is still limited. However, it is important to remark that for the first time, a certain equilibrium in the number of research groups suggesting, on the basis of their studies, that a number of varieties of GM products (mainly maize and soybeans) are as safe and nutritious as the respective conventional non-GM plant, and those raising still serious concerns, was observed. Moreover, it is worth mentioning that most of the studies demonstrating that GM foods are as nutritional and safe as those obtained by conventional breeding, have been performed by biotechnology companies or associates, which are also responsible of commercializing these GM plants. Anyhow, this represents a notable advance in comparison with the lack of studies published in recent years in scientific journals by those companies (Domingo, 2007).

Domingo and Bordonaba are quite explicit about the level of disagreement within the field—even over GM maize, the most tested product.

And: “Similarly, scientific controversy is also present in relation to the safety of GM soybeans.” And:

Especially critical is the recent review by Dona and Arvanitoyannis (2009), who remarked that results of most studies with GM foods would indicate that they may cause some common toxic effects such as hepatic, pancreatic, renal, or reproductive effects, and might alter the hematological, biochemical, and immunologic parameters. These authors also concluded that the use of recombinant GH or its expression in animals should be re-examined since it has been shown that it increases IGF-1 which, in turn, may promote cancer. A harsh response to that review was recently published in the same journal (Rickard, 2010). This is indeed only an example on the controversial debate on GMOs, which remains completely open at all levels.

Séralini, Mesnage, Clair, Gress, de Vendômois & Cellier present “Genetically modified crops safety assessments: present limits and possible improvements”, published in 2011 with Environmental Sciences Europe.

The group obtained extensive data from relevant studies, and focused on genetically modified foods currently sold on the global market:

We have obtained, following court actions or official requests, the raw data of several 28- or 90-day-long safety tests carried out on rats. The thing we did was to thoroughly review the longest tests from both a biostatistical and a biological point of view. Such studies often analyze the biochemical blood and urine parameters of mammals eating GMOs, together with numerous organ weights and histopathology. We have focused our review on commercialized GMOs which have been cultivated in significant amounts throughout the world since 1994 (Table 1).

They reaffirm Domingo and Bordonaba's claim of a major descrepancy between industry-funded and independent research. (They also state a critique of the OECD “consensus” described above.)

Only Sakamoto's and Malatesta's studies have been more than 90 days long (104 weeks and 240 days with blood analyses in Japanese for the first one). Moreover, such tests are not obligatory yet for all GMOs. No detailed blood analysis is available for Malatesta's study, as it mostly includes histochemistry at the ultrastructural level; moreover, the latter tests have not been used to obtain the commercial release by the firm. However, this work has been performed by researchers independent from the GMO industry; it is an important element to take into account for an objective interpretation of the facts, as pointed out in the case of the risk assessments conducted by regulatory agencies with Bisphenol A. For instance in the latter case, it was observed that none of the industry-funded studies showed adverse effects of Bisphenol A, whereas 90% of government-funded studies showed hazards at various levels and various doses [8]. However, regulatory agencies still continue to refer only to industry-funded studies because they are supposed to follow OECD norms, even if such standards are not always appropriate for the detection of environmental hazards[9].

Recall that Domingo and Bordonaba also identified Malatesta's research as significant. Here's more explanation:

For one of the longest independent tests performed, a GM herbicide-tolerant soybean available on the market was used to feed mice. It caused the development of irregular hepatocyte nuclei, more nuclear pores, numerous small fibrillar centers, and abundant dense fibrillar components, indicating increased metabolic rates [17]. It was hypothesized that the herbicide residues could be responsible for that because this particular GM plant can absorb the chemicals to which it was rendered tolerant.

When analyzing data from Monsanto and the EFSA, these reviewers found evidence of dangerous effects in the kidneys:

Kidney dysfunctions are observed with mBt maize producing mutated insecticides such as in MON863. For instance, we quote the initial EFSA report: "Individual kidney weights of male rats fed with the 33% MON863 diet were statistically significantly lower compared to those of animals on control diets", "small increases in the incidences of focal inflammation and tubular regenerative changes in the kidneys of 33% MON863 males." This was confirmed by the company tests [25] and another counter analysis revealed disrupted biochemical markers typical of kidney filtration or function problems [2].

(Disturbingly, the authors of the Monsanto study themselves wrote: “This study complements extensive agronomic, compositional and farm animal feeding studies with MON 863 grain, confirming that it is as safe and nutritious as existing conventional corn varieties.”)

The group has also done a meta-analysis of their hard-won data, and found statistical evidence of major disruption, particularly in the male kidenys.

Some GMOs (Roundup tolerant and MON863) affect the body weight increase at least in one sex [2,14]. It is a parameter considered as a very good predictor of side effects in various organs. Several convergent factors appear to indicate liver and kidney problems as end points of GMO diet effects in these experiments [2,5,15,16]. This was confirmed by our meta-analysis of all in vivo studies published on this particular topic (Table 2). The kidneys are particularly affected, concentrating 42% of all parameters disrupted in males. However, other organs may be affected too, such as the heart and spleen, or blood cells [5].

This review is substantially more alarmist than the previous. Its authors insist that there is “growing evidence of concern” about toxicity of GMO foods:

We can conclude, from the regulatory tests performed today, that it is unacceptable to submit 500 million Europeans and several billions of consumers worldwide to the new pesticide GM-derived foods or feed, this being done without more controls (if any) than the only 3-month-long toxicological tests and using only one mammalian species, especially since there is growing evidence of concern (Tables 1 and 2).

Curious whether anyone had responded to this review, I did a Google Scholar search expecting to find hostile responses to these claims. None appeared. Instead, I found only dozens of sources that cited this review—some in peer-reviewed journals, some in totally unreliable sources. New studies have emerged that confirm the findings presented above. For example:

Gab-Alla, El-Shamei, Shatta, Moussa, & Rayan (2012) tested Ajeeb YG (a spin off of MON 810) on rats. They too found damage to multiple organs—more over the length of the study. Their conclusion is short: “The results of this study showed several changes in organs/body weight and serum biochemistry in the rats fed on GM corn. These findings indicate potential adverse health/toxic effects of GM corn and further investigations still needed.”

Snell, Bernheim,Bergé, Kuntz, Pascal, Paris, & Ricroch, publishing (most recently of the three; late in 2011; sometimes cited as 2012) in The Journal of Food and Chemical Toxicology, conclude unequivocally that “GM plants are nutritionally equivalent to their non-GM counterparts and can be safely used in food and feed.” This review may be the strongest source in support the claim of “broad scientific consensus” currently displayed in the Wikipedia article. (I thank ImperfectlyInformed for highlighting its importance.) First, the authors review 8 short term studies:

Eight 90-day feeding trials conducted using transgenic maize, rice and soybean were performed on rats. The five studies using maize (Hammond et al., 2004, 2006a,b; Mackenzie et al., 2007; Malley et al., 2007) found no differences between the diets containing GM material and the ones which did not. These studies concluded that the maize grain tested were as safe and nutritious as existing commercial maize hybrids.The two studies using rice (Poulsen et al., 2007; Wang et al., 2002) found statistically significant differences between the control and the GM diet groups. However, in both studies most of these observed differences were within the normal biological range and were not indicative of harm.

Then, they explain that they plan to focus on long-term (longer than 90 days) studies:

Themselves In this paper we address the following question: Do recently published studies on long-term effects of GM plants, i.e. studies significantly longer than the 90-day sub-chronic tests, as well as multigenerational studies, present new evidence indicative of some adverse effects?

See Table 2 of their report for a list of the studies reviewed. The authors do record some studies that record long-term differences, but in all cases say that the differences were not significant or that the study was not properly carried out. The authors argue that Malatesta's studies finding health effects of genetically modified soybeans are invalid because they did not specify the origin or growth environment of the modified soybean plants. On this point, they cite a [acnfp.food.gov.uk/meetings/acnfpmeet2006/acnfpjul06/acnfpminsjuly2006 the minutes] of a meeting the UK Advisory Committee on Novel Foods and Processes:

Members noted that the papers did not state the origin of the GM and non-GM soya used in the feeding studies. There were no details of whether the soya had been grown in a field or under controlled conditions and whether or not the GM and non-GM soya were grown, handled and processed under similar conditions.

In rebuttal of a more recent Malatesta study, the authors quote Williams & DeSesso (2010) arguing that the work suffered from six methodological errors:

(i) ‘‘controlling for potential litter effect’’, (ii) ‘‘using an appropriate number of experimental animals per group and acquiring a sufficiently robust sample of independent observations’’, (iii) ‘‘establishing the representativeness of observations’’, (iv) ‘‘adhering to the principles for stereologic morphometry’’, (v) ‘‘using appropriate statistical methods (for study design as well as for data analysis)’’, (vi) ‘‘controlling for potential confounding factors, including those related to differences in diet phytoestrogen contents’’.

The reviewers find that many long-term studies use flawed protocols, and write:

...this review reveals deep weaknesses shared by most longterm studies because of non-adherence to standard procedures outlined in the OECD Test (1998).

The studies reviewed here are often linked to an inadequate experimental design that has detrimental effects on statistical analysis as far as the most frequently used statistics are concerned. Internationally agreed test methods should be used for toxicity testing (EFSA, 2011).

In particular, they say that long-term studies do not use enough animals, and they reiterate that these studies do not use appropriate forms of genetically modified crops on the market. They also say that the majority of long-term sudies did not use “isogenic cultivars”, meaning that they compared genetically modified plants to traditional plants that may not have been the immediate source for the new breed. (Thus, differences observed might be attributable to a difference between the control plant and the modified plant's ancestor, rather than to the genetic modification itself.)

They suggest that some papers should never have been published:

The observations of major flaws in some papers highlight the urgent need to improve the reviewing process before publication of papers addressing this subject. This would avoid spreading confusion in the general press, which may not be able to judge the real scientific quality of publications.

They acknowledge that public researchers may have difficult gaining access to the correct plants:

Because of recurrent lack of compliance with international standards of many studies, a critical situation has arisen where the private sector may not want to provide plant material for studies. Unfortunately, without such collaboration from the private sector, public laboratories may not always be able to conduct studies using appropriate plants lines. In this context, more rigorous statistical prerequisite and sound toxicological interpretations of the results would encourage a virtuous scientific collaboration between public laboratories and private firms, particularly to access to the different isogenic lines that are true comparator of GM lines.

The reviewers affirm:

Overall, the available long-term studies do not yield new safety concerns and confirm that the studied GM varieties (most of them are major commercial products) are nutritionally equivalent to their non-GM conventional counterparts.

They stress that the studies reviewed were publicly funded:

In the present review, most of the studies mentioned were not conducted as part of a regulatory safety assessment process but were exploratory studies performed by public research laboratories. Ten out of the 12 long-term studies examined in this review were all performed within the public sector using public funding.

Again, this review makes an extremely strong statement in definitive terms about the comparative safety of genetically modified food, and is absolutely worth citing and discussing in pages about this topic.

Snell, Bernheim,Bergé, Kuntz, Pascal, Paris, & Ricroch do not discuss long-term studies conducted by Séralini, de Vendômois, and Clair, (though it may possibly implicate them with its critique of methodology). Séralini, Mesnage, Defarge, Gress, Hennequin, Clair, Malatesta, Joël Spiroux, & de Vendômois, in a response to critics also published in the Journal of Food and Chemical Toxicology are highly critical of the review, arguing that a double standard has been used to include studies that find genetically modified foods safe, and exclude others.

In addition, one of the criteria for biological relevance employed by Monsanto and other critics of our study is the linearity or lack thereof in response to the dose. Such a dose–response relationship cannot be claimed from a trial using only 2 doses of test material as employed in the initial NK603 assessment (Hammond et al., 2004). We therefore find it surprising that the relevance of Monsanto’s and the agencies’ conclusion of safety was not challenged due to such protocol insufficiencies. A recent review of the literature is often cited as a proof of the safety of GMO consumption on a longterm basis (Snell et al., 2012). However, of the 24 studies they evaluated, only 2 are long-term on rodents, since a 2 year feeding period with pigs or cows do not constitute a life-long experiments. The 2 rodent studies quoted by Snell and colleagues are from Sakamoto et al. (2008)where not all rats fed transgenic soy were analyzed, and Malatesta et al. (2008a)in mice fed again GM soy, which showed at an electronic microscopy level effects of this product on hepatic function. Moreover, of the 24 studies cited, 16 did not mention the use of the closest isogenic non-GM line as a control, many did not describe the methods in detail, and contained additional deficiencies (Snell et al., 2012). However, all these studies were accepted as proof of safety regardless of the inadequacies highlighted here. It would appear that conclusions of safety seem to need fewer requirements than conclusions of toxicity. However, scientifically it is easier to conclude an outcome of toxicity than safety. This was not the first time regulatory agencies used such double standards to minimize independent research findings in regard to industry findings (Hilbeck et al., 2012; Myers et al., 2009a).

Christoph Then of Testbiotech (an advocacy group discussed in Part 1) issued similar criticisms of the study. These criticisms were not published in a peer-reviewed journal and do not carry the same weight as the above literature reviews. Neither, however, are they invalid, and they do shed more light on this ongoing dispute.

A quite comprehensive (but in its conclusions scientifically flawed) overview by Snell et al. (2011) lists 12 chronic studies and 12 generational studies with genetically engineered plants. But looking at the genetic alterations (called events) that currently have market authorisations in the EU for food and feed (47 events), only three are included in the review by Snell: Bt11, MON810 and Soybean 40-3-2, known as Roundup Ready soybean.

The review by Snell has been used by EFSA to support its position that there are no safety concerns about the genetically engineered plants it has already assessed. Monsanto has included Snell’s review in dossiers it has forwarded to EU authorities. And the review by Snell was used by national authorities, such as the Netherlands NVWA, in their response to Séralini et al’s. research findings.

Nine feeding studies reviewed by Snell et al. used glyphosate tolerant soybeans such as known as Roundup Ready crops. Five of these studies (conducted by a group of Italian researchers) showed signals of negative health impacts.

Interestingly, out of the studies included in Snell’s review, it is those reporting possible negative health impacts that have been criticized and rejected by EFSA on the grounds of problems with the methods used (EFSA 2010). In contrast, studies that did not show any adverse health effects have been accepted by EFSA. At least four of the feeding studies highlighted by Snell et al. were assessed by EFSA during their risk assessments of genetically engineered plants. These studies were accepted without any criticism, even when there appeared to be flaws in their design and execution.

Then also claims that Snell accepted studies in violation of OECD protocols.

But reading the review by Snell et al. more carefully, it becomes evident that none of the studies included meet OECD standards for chronic toxicity or carcinogenicity studies (OECD Guidelines 451, 452, 453). Deficiencies include:

• too few animals per test group,

• too few groups for comparison,

• insufficient details on the tested material and not using the correct comparison materials in control groups.

• in several trials animal species are used that are not common for investigations of health risks.

Several of the studies as listed by Snell et al (2011) are even not meant to be toxicological studies - they are nutritional studies, so few conclusions can be drawn on health risks. Thus on the basis of Snell et al (2011) - which also includes many inaccuracies about the feeding studies reviewed – we do not believe it is appropriate to conclude that genetically engineered plants do not need to be subjected to chronic or generational feeding studies.

It is—at minimum—too soon to tell whether the findings of this literature review will be borne out.

Roundup is a poison that kills almost every plant that hasn't been genetically modified to resist. A very obvious and common objection to Roundup is that residue on plants may hurt the animals that eat the plants. In many cases, these animals are humans.

This is a major issue with “genetically modified food”. Furthermore, issues with Roundup are currently accepted as topical at the genetically modified food controversies page. But instead of being discussed as a legitimate concern, it is hidden—for example, among “individual studies”, where it is dismissed on the grounds of... an individual study.

Anna Lenzer reports in Mother Jones on a wide range of scientific studies questioning the safety of Roundup. She writes: “Roundup is used so much that scientists around the world are reporting with alarm the extent to which glyphosate is turning up in the food, water, and even the air around us.”

Monsanto has been convicted of false advertising because it misrepresented Roundup as safer than it is.

If Roundup is sprayed selectively on crops that are genetically modified to resist Roundup; and genetically modified crops are therefore more likely to contain Roundup; and Roundup is dangerous—then this is an issue with the comparative safety of genetically modified crops. As Mesnage, Gress, Defarge & Séralini write: “Agricultural genetically modified (GM) plants are essentially plants which contain pesticides, because they were designed to tolerate or produce pesticides.”

From the Séralini, Mesnage, Clair, Gress, de Vendômois & Cellier literature review described above:

Most Roundup ready plants have been modified thanks to the insertion of a mutated EPSPS gene coding for a mutated enzyme, which is not inhibited by glyphosate. Therefore, GM plants exposed to glyphosate-based herbicides such as Roundup do not specifically degrade glyphosate. They can even accumulate Roundup residues throughout their life, even if they excrete most of such residues. Glyphosate and its main metabolite AMPA (with its own toxicity) are found in GMOs on a regular and regulatory basis [10,11]. Therefore, such residues are absorbed by people eating most GM plants (as around 80% of these plants are Roundup tolerant).

Benachour, N., and G. E. Seralini found in 2009 that Roundup was particularly lethal to human cells:

In conclusion, mixtures called “formulations” change cell permeability, toxicity, and pathways of xenobiotics: In all cases, cell death is induced more by R than by AMPA or G, and the latter provokes apoptosis (from 50 ppm in HUVEC cells) without membrane damage. By contrast, G mixed with adjuvants in R formulations disrupts cell and mitochondrial membranes and promotes necrosis. It becomes obvious that the “threshold” level of action of the herbicide should take into account the period and length of exposure, the presence of adjuvants, in particular POEA, metabolism, and bioaccumulation or time- delayed effects. All of the above effects are demonstrated below the recommended herbicide agricultural dilutions (from 104 ppm). This clearly confirms that the adjuvants in Roundup formulations are not inert. Moreover, the proprietary mixtures available on the market could cause cell damage and even death around residual levels to be expected, especially in food and feed derived from R formulation-treated crops.

Reviewing the literature in 2013, Mesnage, Gress, Defarge & Séralini write that Roundup is deadly for human cells:

Roundup (R) was highly toxic on human cells, from 10-20 ppm, far below agricultural dilutions. This occurred on hepatic (HepG2, Hep3B and embryonic (HEK293) as well on placental (JEG3) cell lines, but also on human placental extracts, primary umbilical cord cells (HUVEC) and freshly isolated testicular cells (Richard et al.2005; Benachour et al. 2007; Benachour & Seralini 2009; Gasnier et al. 2010; Clair et al. 2012). All formulations cause total cell death within 24h, through an inhibition of the mitochondrial succinate dehydrogenase activity, and necrosis, through the release of cytosolic adenylate kinase measuring membrane damage.

In “Cytotoxic and DNA-damaging properties of glyphosate and Roundup in human-derived buccal epithelial cells” (2012), Koller, Furhacker, Nersesyan, Misık, Eisenbauer, Knasmueller summarize the research:

The assumption that the herbicide and its preservations do not cause DNA damage and cancer in humans is based on the results of long-term carcinogenicity studies with rodents in which the herbicide was fed or injected i.p. to the animals and also on findings from genotoxicity studies (for details see ‘‘Discussion’’ section).

However, in the last decade, a number of studies have been published, which indicate that occupational exposure of humans to the pesticide is associated with increased cancer risks (Bolognesi et al.2009; De Roos et al. 2005, 2003; Eriksson et al. 2008; Hardell and Eriksson 1999; Hardell et al.2002; McDuffie et al. 2001). Furthermore, it was found in field studies that contact of agricultural workers due to spraying leads to induction of micronuclei (MNi) that are formed as a consequence of structural and numerical chromosomal aberrations and to comet formation in peripheral lymphocytes (Bolognesi et al. 2009; Paz-y-Min˜o et al.2007).

In their research, they found that Roundup was toxic to human cells, and that glyphosate and Roundup caused genetic damage in human cells:

The most relevant finding of the present study is the observation of a significant, dose-dependent induction of two types of nuclear anomalies that reflect genomic damage by G and its formulation. As shown in Fig.3a–c, the former markers (MNi, BN-MNi, and NB) were significantly increased after exposure of the cells. The most sensitive endpoint was MNi induction; treatment of the cells with highest dose of R (20 mg/l) caused a threefold increase over the background, and with the corresponding concentration of G, a weaker effect was seen (Fig.3a, b). NPB was a less responsive endpoint, and the only significant effect was obtained with the highest R dose (20 mg/l). [...]

Our findings support the assumption of a possible association between G exposure and increased cancer risks and underline the importance of further human studies for example of Mni experiments with exfoliated buccal cells from agricultural workers and laborers that are exposed in factories. At present, no such data are available and estimates of inhalative exposure due to spraying, which are mentioned in the review of Williams et al. (2000), are vague and have not been published in scientific journals. It is known that more than 90% of all human cancers are of epithelial origin (Cairns 1975). Therefore, the present observation of induction of DNA damage under conditions that are relevant for humans in epithelial cells derived from the oral cavity should be takes as an indication for potential adverse effects.

In 2012, Antoniou, Habib, Howard, Jennings, Leifert, Nodari, Robinson, & Fagan published a literature review in the Journal of Environmental and Analytical Toxicology. Titled “Teratogenic effects of glyphosate-based herbicides: Divergence of regulatory decisions from scientific evidence”. As the title makes quite clear, the claims of regulators who assert that Roundup is safe do not match with scientific opinion on the topic. The reviewers discuss dozens of studies, connecting Roundup with a wide range of health problems, including, especially, birth defects. Their paper is quite detailed and contains many insights on topics of interest to us here. It concludes:

Studies published in the peer-reviewed scientific literature have raised major concerns regarding the potential for glyphosate and its commercial formulations to cause birth defects and other reproductive problems. In addition, a debate has emerged over the reported effects on human health of herbicide application in regions that produce GM glyphosate-tolerant crops and about the safety of food and feed produced from these crops.

Regulatory authorities and industry affiliates have defended the use of glyphosate largely by citing the industry-sponsored toxicological tests conducted for regulatory purposes, which they claimed showed no evidence of teratogenicity. However, the German authorities’ draft assessment report revealed that even these industry tests contained clear evidence of glyphosate-mediated teratogenicity and reproductive toxicity. Many of the malformations observed in these studies are of the type associated with the retinoic acid signalling pathway. Paganelli et al. [1] showed that this was the mechanism through which glyphosate and Roundup exercise their teratogenic effects.

It is noteworthy that these industry tests were commissioned by the same companies that stand to profit from regulatory authorization. Regrettably, this system possesses an inherent risk of bias and makes it especially important that the regulatory assessment is rigorous. Yet in the EU, the evidence suggests that this was not the case. The significance of clear teratogenic effects of glyphosate in rabbits and rats found in tests commissioned by industry were minimized by German regulators. A scientifically rigorous assessment was further impeded by the outdated design of the standard tests, which are not sufficiently sensitive to detect effects from realistic exposures. As a result, the German authorities suggested, and the EU adopted, an acceptable daily intake (ADI) for glyphosate that is unreliable and could potentially result in exposures that cause harm to humans.

Another relevant factor is that the industry teratogenicity tests were on glyphosate, the presumed active ingredient of the herbicide, and not on the herbicide formulations as sold and used, even though studies indicate that the formulations are more toxic for certain endpoints than glyphosate alone.

A substantial body of evidence demonstrates that glyphosate and Roundup cause teratogenic effects and other toxic effects on reproduction, as well as genotoxic effects. From an objective scientific standpoint, attempts by industry and government regulatory bodies to dismiss this research are unconvincing and work against the principle that it is the responsibility of industry to prove that its products are safe and not the responsibility of the public to prove that they are unsafe. The precautionary principle would suggest that glyphosate and its commercial formulations should undergo a new risk assessment, taking full account of the entirety of the peer-reviewed scientific literature as well as the industry-sponsored studies. Experience to date suggests that the new risk assessment should be conducted with full public transparency by scientists who are independent of industry.

Roundup kills plants by inhibiting a certain enzyme that plants use to make amino acids. The product has been claimed nontoxic for humans because we do not use this enzyme. However, we live symbiotically with gut bacteria who do use the enzyme and are harmed by Roundup. This effect has recently been subject to greater scrutiny.

Glyphosate suppresses helpful gut bacteria, increasing Clostridium botulinum infections among cattle. Krüger, Shehata, Schrödl, & Rodloff  , “Glyphosate suppresses the antagonistic effect of Enterococcus spp. on Clostridium botulinum”, Anaerobe, 2013.

Anthony Samsel and Stephanie Seneff reviewed the literature on Roundup, gut bacteria, and public health, finding:

In addition to aiding digestion, the gut microbiota synthesize vitamins, detoxify xenobiotics, and participitatein immune system homeostasis and gastrointestinal tract permeability [14]. Furthermore, dietary factors modulate the microbial composition of the gut [15]. The incidence ofinflammatory bowel diseases such as juvenile onset Crohn’s disease has increased substantially inthe last decade in Western Europe [16] and the United States [17]. It is reasonable to suspect that glyphosate’s impact on gut bacteria may be contributing to these diseases and conditions.

However, the fact that female rats are highly susceptible to mammary tumors following chronic exposure to glyphosate [9] suggests that there may be something else going on. Our systematic search of the literature has led us to the realization that many of the health problems that appear to be associated with a Western diet could be explained by biological disruptions that have already been attributed to glyphosate. These include digestive issues, obesity, autism,Alzheimer’s disease, depression, Parkinson’s disease, liver diseases, and cancer, among others. While many other environmental toxins obviously also contribute tothese diseases and conditions, we believe that glyphosate may be the most significant environmental toxin, mainly because it is pervasive and it is often handled carelessly due to its perceived nontoxicity.

Another relevant study: Manuela Malatesta, Chiara Caporaloni, Stefano Gavaudan, Marco B.L. Rocchi, Sonja Serafini, Cinzia Tiberi, and Giancarlo Gazzanelli, “Ultrastructural Morphometrical and Immunocytochemical Analyses of Hepatocyte Nuclei from Mice Fed on Genetically Modified Soybean”, Cell Structure and Function 27 173–180 (2002).

Our observations demonstrate significant modifications of some nuclear features in GM fed mice. In particular, GM fed-mice show irregularly shaped nuclei, which generally represents an index of high metabolic rate, and a higher number of nuclear pores, suggestive of intense molecular trafficking. Moreover, the roundish nucleoli of control animals change in more irregular nucleoli with numerous small fibrillar centres and abundant dense fibrillar component in GM-fed mice, modifications typical of increased metabolic rate.

Also see: Alejandra Paganelli , Victoria Gnazzo , Helena Acosta , Silvia L. López , and Andrés E. Carrasco.“Glyphosate-Based Herbicides Produce Teratogenic Effects on Vertebrates by Impairing Retinoic Acid Signaling”. Chem. Res. Toxicol., 2010, 23 (10), pp 1586–1595 DOI: 10.1021/tx1001749,

Roundup includes glyphosate, which is poison—but also has other ingredients, which are also poison. One reason you might not realize that Roundup contains other ingredients is that Wikipedia's page on Roundup was redirected to the page on Glyphosate. As discussed in the literature review on birth defects (above), industry-funded studies have also neglected to test Roundup itself.

Consuming Roundup directly is of course lethal, as is well known in India, where debt-ridden farmers have committed suicide by drinking it. And according to Song, Kim, Seok, Gil, Hong (2012), it may actually be the surfactant that kills:

In severe intoxicated cases, patients suffer severe toxic effects, such as respiratory distress necessitating positive pressure ventilation, metabolic acidosis, arrhythmia, renal failure, electrolyte imbalance, or death (3-5). We have also found that patients who ingest large volumes of glyphosate herbicides are also at risk of hypotension, mental deterioration, respiratory failure, and arrhythmia, particularly when the calculated surfactant volume in herbicide product is over 8 mL (6). Indeed, it has been proposed that the surfactant in the formula, and not the glyphosate, is the essential element of these mixtures responsible for human toxicity (7,8). [...]

In our experimental setting, the 2 surfactants manifested cytotoxicity at concentrations between 1 µM and 100 µM, whereas glyphosate was not found to be cytotoxic in that range. [...]

Cellular toxicity associated with the glyphosate–surfactant mixture may impact membrane integrity, metabolic activity, mitochondrial activity and the rate of total protein synthesis. The current results indicate that the combination of glyphosate and TN-20 result in a mixture with potentiated toxicity. The toxicity of the combination was proportional to the TN-20 concentration when the glyphosate concentration was fixed. However, this relationship was not observed when the TN-20 concentration was fixed and the glyphosate concentration was varied. Therefore, it appears that TN-20 alters the toxicodynamics of glyphosate.

Kim, Hong, Gil, Song, & Hong found in 2013:

Cellular toxicity associated with the glyphosate–surfactant mixture may impact membrane integrity, metabolic activity, mitochondrial activity and the rate of total protein synthesis. The current results indicate that the combination of glyphosate and TN-20 result in a mixture with potentiated toxicity. The toxicity of the combination was proportional to the TN-20 concentration when the glyphosate concentration was fixed. However, this relationship was not observed when the TN-20 concentration was fixed and the glyphosate concentration was varied. Therefore, it appears that TN-20 alters the toxicodynamics of glyphosate. The precise mechanism of this process cannot be determined from the present results; however, one working hypothesis is that TN-20 initially causes cell membrane injury, resulting in the disruption of the barrier to glyphosate entry into the cell. Thereafter, both glyphosate- and TN-20-mediated toxicities occur.

Clair, Linn, Travert, Amiel, Séralini, & Panoff confirmed Roundup toxicity in microorganisms that was not even proportional to the amount of glyphosate involved:

Our first observation is the relatively comparable toxicity profiles on three essential food microorganisms, in a short 24-h period (Fig.1a–c). Roundup is always more potent than glyphosate, and in all cases, toxic from levels 10–100 times below the lowest agricultural uses (10,000 ppm). G. candidumand L. cremoris are more Roundup sensitive thanL. bulgaricus. Similar impacts have not been observed for glyphosate alone (except for G. candidumat 10,000 ppm). Roundup effect was not proportional to glyphosate concentration in the Roundup formulation, since R400 is almost 10-fold more inhibitory than R450 (Fig. 1a). [...]

Moreover, like on eukaryotic cells, impact of glyphosate is not proportional to its concentration in Roundup formulations, confirming adjuvants are not inert—an observation that supports the findings of previous studies [23, 25]. Similar effects on microorganisms have been reported previously in the literature; in fact, it appeared that Ichthyophthirius multifiliis and T. thermophilatolerate glyphosate but not Roundup. The commercial formulation was then 100 times more toxic than the active ingredient (G) [44].

A common genetic modification performed on crops is the addition of genes from the bacterium Bacillus thuringiensis (Bt). Bt genes may be dangerous for reasons that do not involve genetic engineering per se. Most of the sources about “scientific consensus”, advanced by Team GMO on Wikipedia, discuss only genetic engineering per se; they do not discuss individually dangerous traits.

It has been claimed: ”There is broad scientific consensus that food on the market derived from GM crops pose no greater risk than conventional food.” If there is substantial scientific opinion that Bt genes may be dangerous, this fact would serve as a direct and serious counterexample to the claim. And in fact, there is substantial evidence that Bt in genetically modified crops has effects in humans.

Bernstein et al (1999) found evidence of skin reactions in farm workers exposed to Bt pesticides for only a few months. In their view, this finding strongly suggested that contact with Bt (though not oral ingestion) poses a serious hazard:

Although occupationally related clinical diseases were not observed in this cross-sectional survey, the fact that skin and serologic tests of immediate hypersensitivity developed in some workers indicates that adverse IgE mediated health effects could develop if repetitive exposures continue in some of these workers. Longitudinal surveillance studies will be necessary to establish whether this would occur. These results also suggest that future large-scale urban spraying of Bt pesticides may not be innocuous and may require more direct health monitoring and surveillance.

In addition to the implication that skin sensitization to Bt in pesticides could be a precursor of clinical IgE-mediated diseases, several aspects of this investigation may be relevant to other current health issues: immediate hypersensitivity induced by bacteria and transgenic foods engineered to incorporate pesticidal genes in their genomes. First, because skin sensitivity to spore and vegetative components of a nonpathogenic species of Bacillus was clearly demonstrated, future awareness about the allergenic potential of environmental bacteria should be increased, even though this phenomenon has been recognized for relatively few such organisms (e.g., Staphylococcus aureus, Streptococcus pneumoniae, and Moraxella catarrhalis) (19,20). There is presently strong evidence of a close molecular genetic relatedness between Bt subspecies and the B. cereus food pathogen that would support this call for caution (21,22). Further, in the case of the Bacillus genus, the possibility of cross-allergenic epitopes in an unrelated species such as B. subtilis should be appreciated because this organism or its products may occur in both occupational and nonoccupational environments (23,24). Conversely, results of this investigation should partially allay recent concerns about the occurrence of possible adverse health effects in consumers after exposure to transgenic foods (25,26). Because reactivity to the Btk pro-8 endotoxin was only encountered in 2 of 123 workers sensitized by the respiratory route, it is unlikely that consumers would develop allergic sensitivity after oral exposure to transgenic foods (e.g., tomatoes, potatoes) that currently contain the gene encoding this protein. However, future clinical assessment of this possibility is now feasible because of the availability of reliable Bt skin and serologic reagents developed during the course of this investigation.

Freese & Schubert found that EPA scientists considered this warning potentially series, and acknowledged the need for these tests; yet the agency renewed its approval of Bt corn and did not conduct the tests.

In an assessment of Bt crops, expert advisors to the EPA who reviewed the Bernstein study and one of Vazquez et al.’s four studies concluded that: “These two studies suggest that Bt proteins could act as antigenic and allergenic sources” (SAP Bt, 2000, p. 76). Different approaches were called for to further 10 characterize the allergenic risk of Bt proteins: “Only surveillance and clinical assessment of exposed individuals will confirm the allergenicity of Bt products...” (SAP Bt, 2000, p.76). Finally, the EPA’s experts noted that testing for potential reactions to Cry proteins in Bt spray and Bt crops could be undertaken now: “The importance of this [Bernstein’s] report is that reagents are available that could be used for reliable skin testing and serological evaluation of Bt protein exposed individuals.” Unfortunately, in 2001 the EPA re-registered Bt corn for 7 years without making use of these reagents (EPA BRAD, 2001d, p. I2). The Agency has also discounted other evidence of the potential allergenicity of Bt proteins.

Schubert & Freese directly contest the claim that plants modified to produce Bt are equally safe as “conventional” plants exposed to Bt spray:

The EPA’s chief justification for approval of Bt crops in the absence of crucial data is that Bt sprays have a history of safe use, and so Bt crops are presumed to be safe as well. This presumption is not justified for several reasons. First, it is reasonably clear that Bt sprays do cause allergic symptoms, as detailed at the beginning of this case study. Expert advisers to the EPA told the Agency that more studies are needed to determine the allergenic risk posed by Cry proteins in general – whether from Bt sprays or crops (SAP Bt, 2000). Secondly, there is likely much greater chronic exposure to Cry proteins in Bt crops than in sprays. Cry proteins in Bt sprays break down within several days to two weeks upon exposure to UV light (Ignoffo and Garcia, 1978; Behle et al., 1997), while this is obviously not the case with Bt crops, which produce the toxin internally in grains and other plant tissues. Thirdly, Bt sprays are composed primarily of endotoxins in an inactive crystalline form. They are only toxic to insects with alkaline gut conditions that permit solubilization of the crystal to the protoxin, followed by proteolytic cleavage to the active toxin (Hilbeck et al., 2000). Bt crops, on the other hand, are generally engineered to produce the Bt toxin (e.g. Bt 11), which is active without processing, or a some what larger fragment (e.g. MON810). There is also evidence indicating that Cry toxins are more immunoreactive than Cry protoxins (Freese, 2001). Finally, the trend to increased Cry protein expression fostered by the EPA’s “high-dose” strategy to slow development of pest resistance to Bt crops (EPA BRAD 2001e) may result in an increase in consumers’ dietary exposure to Bt proteins. For instance, Mycogen/Pioneer’s Herculex Cry1F corn, registered in 2001, expresses at least an order of magnitude more Cry protein in kernels than MON810 (Mendelsohn et al., 2003). Use of chloroplast transformation, while still at the experimental phase, raises Bt protein levels still higher (Kota et al., 1999). Thus, even if one ignores the evidence of allergenicity and 13 concedes that Bt sprays have a history of safe use, this is clearly not adequate grounds on which to judge Bt crops and their incorporated plant pesticides as safe.

Really, the idea of a consensus on the comparative 'safety of genetically modified foods', for human beings, is ludicrous on its face. Genetically modified foods have not been around for much time; they are a new element of the food supply. No one has eaten genetically modified food from the cradle to the grave—unless they died young.

There may be a consensus among scientists that no good evidence thus far suggests any risks in eating genetically modified food. I don't agree that this is the case, but it's a logical possibility. But no one, no matter how white their lab coat, can claim scientifically, unequivocally, that genetically modified organisms are safe to eat.

We frequently see the caveats “on the market right now” and “compared to conventional foods”. These caveats do restrict the scope of the claim, but they do not address the issue of timeframe. No human beings have consumed only engineered corn for two decades, let alone a lifetime, let alone across multiple generations.

The timeframe for observation highlights the difference between climate change skeptics and “GMO skeptics”, whom industry lobbyists would like to paint as similar. There is an enormous gap between proving that something is or is not happening—and proving that something will not happen in the future.

If you're talking with a five-year-old, you might have a much easier time showing them gravity than convincing them that eating vegetables won't turn them into a vegetable. And how would you convince them of the latter? You might tell them that you have eaten vegetables your whole life, but are demonstrably not a vegetable. In doing so, you would be appealing to evidence that spans an appropriate time frame.

Brian Merchant:

Scientists, he says, have by and large found no credible health risks to humans consuming GMO foods. But, as Tom Philpott, the Mother Jones food writer that Kloor takes to task, points out, we’re still not entirely certain what the long-term effects of growing and eating Monsanto-synthesized soy might be. It’s still a relatively young innovation. There’s some valid reason to exercise caution — especially because there’s a dearth of long-term studies on GMOs; the flawed “rat tumor” study was billed as the first. The GMO industry has blocked access to many scientists’ inquiries, making research into their health impact difficult.

Which is why the analogy to climate skeptics doesn’t work. It’s false equivalence. 97% of climatologists working in the field say that humans are causing the planet to warm. And climate science focuses on a phenomenon that’s occurring whether we like it or not; to deny that human activity is warming the planet to a dangerous degree, and to throw your lot in with the 3% that thinks otherwise, is foolhardy when a stable global climate is on the line. The best science says we need to act, and act swiftly. If we don’t, it could spell disaster.

I’s a different story with genetic engineering — you can agree that scientists have generally shown that eating GMOs won’t yield health problems, but still be skeptical that the whole enterprise is a good idea — or simply not want to eat something whose genome has been tampered with, just like you might not eat raisins. Let’s say for the sake of simplicity that 97% of scientists agree that eating GMOs won’t harm you (I doubt it’s that high). We have the luxury to explore the ideas of the dissenters because there’s no imperative to crank out as many GMOs as possible right now, unless you’re Monsanto. We can afford to be all the more careful with the genetically modified food we’re putting into our bodies; there are extant alternatives, after all. We don’t have that luxury with climate change.

Scientists have identified long-term consumption of genetically modified food as an unknown area warranting further research.

Malatesta et. al. (2008) write:

Diet is considered one of the most important environmental factors affecting lifespan. Genetically modifed (GM) crops, in which new genes have been inserted into the original genome, are nowadays distributed all over the world, thus frequently becoming part of human and animal diets (Sanvido et al. 2007). The fact that GM food may affect human or animal health is debated: actually, no consensus exists neither on the test designs nor on the criteria to be assumed for assessing the presence of possible pathological signs (Doull et al. 2007; Séralini et al. 2007). However, it cannot be ignored that some scientific reports have described structural and molecular modifcations in different organs and tissues of GM-fed animals (e.g. Ewen and Pustzai 1999; Malatesta et al. 2002a, b, 2003a, 2005, Vecchio et al. 2004; Tudisco et al. 2006; Trabalza-Marinucci et al. (2008). These observations suggest that the risk of genetically modifed crops cannot be ignored and deserves further investigations in order to identify possible long-term effects, if any, of GM food consumption that might help in the post market surveillance (Kuiper et al. 2004)

Within this context, it seems of prime importance to elucidate whether a GM-containing diet may interfere with the ageing process, since senescence is characterized by progressive changes in several cellular functions that eventually result in disease and/or the loss of the ability to successfully respond to stress and xenobiotics (Jameson 2004). In fact, to test GMO-related effects on laboratory mammals, experiments have usually been performed for some months only, thus making impossible to detect long-term consequences (Séralini et al. 2007).

(The study found that “Several proteins belonging to hepatocyte metabolism, stress response, calcium signalling and mitochondria were differentially expressed in GM-fed mice, indicating a more marked expression of senescence markers in comparison to controls.”)

“Laboratory tests are not worst-case scenarios”, write Bøhn, Primicerio, Traavik in response to one group's claim that German's ban of MON810 was “unjustified” because of how safe the product is. They continue:

There is a tendency to overlook that laboratory tests do not represent worst-case scenarios [15]. It is more precise to say that laboratory studies test single factor, short-duration, often in high dose experiments under stable conditions without concomitant stress from biotic (competitors, predators, parasites, etc.) and abiotic factors (climate, pollutants, etc.). Such environmental factors are likely to stress organisms in addition to what they feed on and what they are exposed to. A more recent study by Bøhn et al.[16] tested the impact of predator smell in addition to feeding Bt-maize (the same MON810-based feed as discussed above) in D. magna. It was shown that fitness differences increased between animals of D. magna fed Bt- versus non-Bt maize when test animals were exposed to a fish predator as an additional stress-factor. In line with this, Sih et al.[17] showed that a pesticide (carbaryl) was 46 times more toxic when a tested tadpole species was under threat of a predator in addition to being exposed to the chemical. Such results provide insights beyond additive standard tests and may illustrate limitations in existing risk assessment methods and protocols.

Katherine Van Tassel, an expert on bioethics and health ethics from whom we'll hear more later, writes that we are still operating in a “health risk information void”:

The FDA’s presumption that GM plant foods are bioequivalent to traditional food is a consequence of the remarkable growth in the development of new technologies which far outpaces the science necessary to identify the human health risks associated therewith. This scientific lag time creates a period when there is an information void with regard to risks to human health. As this information void is slowly filled through scientific experimentation, the level of uncertainty over health risks commonly progresses from ignorance (where scientists don’t know what they don’t know) to indeterminacy (where scientists know what they don’t know but can plan the scientific experiments necessary to find out) to, finally, a tipping point in the state of knowledge when classic probability analysis can be applied to predict, or quantify, risk levels to human health. This Article refers to this lag time when the state of knowledge over health risks has moved out of ignorance and into indeterminacy as ‘the space between’ or ‘the health risk information void.’ This Article points out that government regulators are operating in ‘the space between’ when it comes to genetically modified plants marketed for human consumption.

Van Tassel continues with an explanation of why harmful changes may not express themselves immediately:

Epigenic mechanisms don’t alter DNA sequences, but affects the DNA by preventing its expression through various processes including DNA methylation,106 DNA packaging,107 protein methyltransferases,108 and the location of the DNA within the nucleus.109 These mechanisms turn genes on and off during the course of an organism’s life in response to environmental factors. For example, this accounts for vernalization, the well-known process of exposing certain plants to low temperatures to trigger flowering.110 Epigenetic inheritance between organisms means that a GM food might produce generations of offspring before an environmental trigger or disease activates a DNA sequence to produce a toxic or allergenic protein. Thus, a GM plant that may be safe to consume when first produced, could cause unintended health consequences when one of its progeny is consumed at some future point in production.

David Schubert describes a scenario with possible long-term effects, and also makes an important point about long-term health:

Given that GM plants will sometimes produce different amounts of proteins, and perhaps totally new proteins, as compared with the parental species, what are the possible results? A worst-case scenario would be that an introduced bacterial toxin is modified to make it toxic to humans. Prompt toxicity might be rapidly detected once the product entered the marketplace if it caused a unique disease, and if the food were labeled for traceability, as were the GM batches of tryptophan. However, cancer or other common diseases with delayed onset would take decades to detect, and might never be traced to their cause. Conversely, plant flavonoids and related molecules have great health benefits7, and there is evidence that these can be depleted in GM crops8

Genetic engineering changes the frequency with which certain genes appear in an organism's genome. In particularly, genetically engineered organisms contain a greater proportion of meta-genes that regulate the expression of genes themselves. As described at Wikipedia's “Techniques of genetic engineering” article, “most constructs contain a promoter and terminator region as well as a selectable marker gene”.

Often, the promoter comes from the Cauliflower mosaic virus, a DNA retrovirus that, like HIV, hijacks a host genome in order to reproduce. The vector—the mechanism by which the new DNA is inserted into the target genome—is often a plasmid from Agrobacteria. (Though sometimes the new DNA is shot haphazardly into the host with a small metal gun.)

These control sequences of DNA may interact with the plant genome in unpredictable and dangerous ways. Mae-Wan Ho writes:

Several expression-cassettes are usually stacked in series, each cassette consisting of a gene with a promoter (as well as terminator). The promoter is often taken from a virus in order to make the gene over-express. The stacked cassettes are, in turn, spliced into a vector, the most widely used being the T-DNA of Agrobacterium; and it is this whole construct which is intended for integration. But unpredictable deletions, rearrangements and repeats are invariably present in the actual insert. Transgenic instability often persists after the insertion event (6). The constitutive over-expression of transgenes placed under viral promoters may be one cause of gene-silencing (7), and recombination hotspots may be another. The borders of T-DNA are recombination hotspots, as is the 3' end of the 35S promoter from the cauliflower mosaic virus (CaMV) (8), which is used in practically all transgenic plants currently released. Recombination hotspots are expected to increase the likelihood of secondary mobility and horizontal gene transfer (9). Secondary mobility within the host genome may result in rearrangements and other effects that could drastically alter the agronomic and other properties of the transgenic line.

That plant genomes, like animal genomes, harbour retrotransposons, relict retroviruses and pararetroviruses (10) is no reason for complacency. These sequences have existed for millions of years in the genome, and are probably no longer harmful, either to the plant itself or to other organisms interacting with it. However, should they recombine with the CaMV promoter (or modules thereof) in the transgenic DNA, live viruses may well be regenerated. The transgenic DNA may also be mobilized by these relict elements.

Again, these mechanisms carry great potential danger for unknown effects on the genome. All known methods for inserting genetic material are risky and uncertain. An obvious manifestation of this uncertainty is that repeated attempts to produce the “same” GMO can have different results. Biotechnicians must repeat the DNA-insertion process multiple times before they produce a plant that even exhibits the trait they expect.

Latham, Wilson, & Steinbrecher report:

Both insertion-site and genome-wide mutations may result in transgenic plants with unexpected traits. Despite the supposed precision of genetic engineering, it is common knowledge that large numbers of individual transgenic plants must be produced in order to obtain one or a few plants that express the desired trait in an otherwise normal plant. Even after selection, there are many reports of apparently normal transgenic plants exhibiting aberrant behavioural or biochemical characteristics upon further analysis. These unexpected traits range from altered nutrient or toxin levels to lower yields under certain environmental conditions, see references in [46, 56, 56]. Among others are altered interactions with soil microorganisms [58], susceptibility to pathogens [59], altered insect resistance [60], and plant reproductive characteristics [61]. Despite the paucity of publicly accessible data and lack of monitoring of commercial transgenic crop varieties, commercial (ie, approved) transgenic plants have also been observed with unintended traits. Verified examples include stem splitting and decreased yields in transgenic soybean plants [62] and a 67-fold reduction in beta-carotene content in a transgenic squash variety engineered for virus resistance (USDA Application # 95-352-01).

These examples show that unexpected transformation-induced phenotypes can affect any aspect of plant phenotype, including those of value or concern to humans. Furthermore, the incidence of unintended phenotypes in transgenic plants seems to be high, indicating that plant transformation is currently not predictable. Few unexpected phenotypes have been followed up but we propose that an important source of unpredictability is likely to be the mutational consequences of plant transformation.

The use of Agrobacterum tumefaciens as a vector for introducing new DNA has been particularly well studied. In addition to inserting the DNA desired by engineers, Agrobacterium gene insertion has side effects that are partially understood.

In particular, Agrobacterium gene transfer can trigger a sort of immune reaction in the genome of the host organism. Kuta & Tripathi write in a 2005 review article:

The relationship of Agrobacterium to host plants is unique among plant pathogens. Many aspects of the plant-Agrobacterium interaction are not yet fully understood. It was earlier reported that Agrobacterium does not induce the hypersensitive response in target plants, even though the bacterium introduces several proteins into the host cell (Robinette and Matthysse, 1990). However, there are now several reports of high necrosis and poor survival rate of target plant tissues during the process of Agrobacterium-mediated T-DNA transfer (Pu and Goodman, 1992; Deng et al., 1995; Perl et al., 1996; Mercuri et al., 2000; Chakrabarty et al., 2002; Das et al., 2002). This could be the consequence of plant’s hypersensitive reaction to Agrobacterium infection. Recently, it was demonstrated that plants can modulate their gene expression in response to Agrobacterium infection and that Agrobacterium can actually trigger the plant defense machinery (Ditt et al., 2001).

Hypersensitive reaction (HR) is known to be one of the plant defense responses and it is generally characterized by a rapid, localized cell death around the infection site and the accumulation of antimicrobial agents (HammondKosack and Jones, 1996; Richter and Ronald, 2000). It is the sequence of events during HR that subsequently lead to necrosis of the collapsed cells (Goodman and Novacky, 1994).

They conclude:

The Agrobacterium induced HR could lead to rapid and large generation of reactive oxygen radicals in target plant cells, resulting to plant cell death (necrosis), oxidative stress to the invading Agrobacterium cells, production of toxic antibacterial substances and the deleterious effects on DNA molecules, especially at the site of oxidative burst.

Kunik et al. (2000) found that Agrobacterium can target the cells of mammals, including humans, and note that certain health problems may be related:

That Agrobacterium carrying a neomycin resistance gene within its T-DNA generated stable antibiotic-resistant lines of human cells expands its potential host range from plants, yeast, and filamentous fungi to mammalian cells. Here, transformation of human cells has been observed in laboratory conditions; whether it may be relevant biologically in nature remains unknown. Interestingly, Agrobacterium or Agrobacterium-related species have been suggested to be involved in several human diseases (55, 56).

Note: Giammanco et al. (2004) found infections of Agrobacterium tumefaciens present in multiple patients at the same hospital, all with bloodstream infections related to catheters. There hasn't been enough research on this topic yet, and so I would not classify it as an obvious challenge to the “broad scientific consensus” claim (thought it may be, this connection is not sourced); however, this information does belong at the pages about Agrobacterium and Genetic engineering techniques.

“Large-scale chromosomal rearrangements were observed in both transgenic rice plants”, found Takano, Egawa, Ikeda & Wakasa in 1997.

Our observations of several events indicative of illegitimate recombination accompanying rearrangement in two transgenic rice plants suggest that rice protoplasts have a propensity for this form of recombination. If individual plant cells demonstrate variability in this capacity as is the case for Xenopus (Lehman etaL, 1994), it may be of value to isolate a cell line with a decreased recombination potential. A predominance for illegitimate recombination in cultured cells may preclude the identification of an isolated homologous recombination event necessary for gene targeting and gene replacement. Further studies of these phenomena will be helpful in this regard.

The Takano et al. study is widely cited, and it is clear that actual genetic engineers are aware of this phenomenon and keen to explore it further. (Note that they are more interested in determining what works than what is politically convenient. Note also that genetic “side effects” are not inherently bad—in many cases they may represent the host organism protecting itself from an unknown outside organism. The point is that the genetic engineering process introduces uncertainty that proponents are unwilling to admit.) Only political groups and polemicists have clung to the fiction developed in the late 1980s that engineers can insert DNA into a host organism's genome with no side effects whatsoever. Yet this fictional model of Genetic engineering is presented as truth on Wikipedia.

In 1998, Nacry, Camilleri, Courtial, Caboche & Bouchez reported in Genetics:

We now have convincing evidence that T-DNA integration can provoke profound rearrangements in plant genomes, both at the chromosomal level and at the gene level. The prevalence of large chromosomal alterations in T-DNA transformants is difficult to assess, but the actual frequency of such events could be significant and needs further examination.

In 2001, Belgian researchers studying Monsanto's Roundup Ready soybeans found unknown DNA not described in the company's official description of the new genome.

The New York Times reported:

Monsanto acknowledged that the extra DNA was there, but it said it was confident that the soybean was safe and that the unknown DNA had no effect on the plant. Dr. Jerry J. Hjelle, the company's vice president for regulatory affairs, said the DNA segment had been in the crop since the beginning as it went through testing to prove its safety.

Products made from Roundup Ready soybeans have been eaten by people and animals for five years with no reports of health problems. Still, the findings could cause some embarrassment for Monsanto and the agricultural biotech industry because they raise questions about how well the molecular makeup of the products is characterized.

Roundup Ready soybeans contain a gene from a bacterium that allows the plants to withstand Monsanto's Roundup herbicide. Farmers can thus spray their fields with Roundup throughout the growing season to kill weeds without harming the crop. More than half the soybeans grown in the United States are now Roundup Ready. In Europe and Japan the beans are approved for use but not for planting.

This is the second time that scientists have found something in Roundup Ready soybeans that Monsanto did not seem to know was there and had not cited at the time of the product's approval.

Last year the Belgian scientists and Monsanto, working independently, found that the soybeans contained not only one complete copy of the bacterial gene, as intended, but two fragments of that gene. Monsanto filed reports with regulators around the world offering data to show that the fragments were not active genes and had no effect on the plant.

More studies found weird DNA in genetically modified organisms:

• Rang, A.; Linke, B.; Jansen, B., 2005: “Detection of RNA variants transcribed from the transgene in Roundup Ready soybean”. European Food Research and Technology 220(3/4): 438-443

This study finds new DNA substantial enough to be transcribed, and finds that “expression” of the spliced sequence does not always work as Monsanto expected—resulting in several different RNA expressions.

Lim Li Ching and Maria von Weizsäcker explain these results:

Monsanto declared in reaction to these reports that the additional DNA segments are not functional, meaning that they are not transcribed to RNA, not translated into proteins and therefore, have no effects whatsoever on the soybean.

Inconsistent with Monsanto’s statement, a new report published in December 2004 found RNA variants in RR soya, arising from read-through of the nos terminator in RR soya. There was at least partial transcription of the unintentionally inserted segment of DNA (a 250-bp fragment of the epsps gene). The read-through product was further processed (post-transcriptionally), resulting in four different RNA variants. The RNA variants may result in the production of (as yet unknown) proteins, which could be potentially harmful to the plant or even to the consumer. The scientists also raised the possibility that the mechanisms responsible for producing the RNA variants are related to the nos terminator. This matter is of particular interest as the nos terminator is commonly used in the production of many genetically modified organisms. Thus, RNA variants could be also expressed in these GMOs.

Furthermore, there is evidence that some small RNA molecules, best known for their role in causing gene silencing, may have effects that can be systemically spread throughout an organism, can affect more than the primary target gene (and there is no way at present to predict what these genes will be) and can be transmissible through food, causing lasting and sometimes heritable effects.

One major problem with industry-funded studies, in addition to general concerns of bias, is that they may sidestep the issue of side effects completely by using proteins from a different source.

Freese & Schubert write that companies may perform tests with “surrogate proteins” grown not in the genetically modified plant, but in well-controlled bacterial samples.

Biotechnology companies rarely test the transgenic protein actually produced in their engineered crops. Instead, for testing purposes they make use of a bacterially generated surrogate protein that may differ in important respects from the plant-produced one. The same genetic construct used to transform the plant is expressed in bacteria (usually E. coli), and the surrogate transgenic protein is then extracted from the bacteria. This surrogate protein is then employed for all subsequent testing, such as short-term animal feeding studies and allergenicity assessments. This is, however, a serious mistake in testing paradigms, since plants and bacteria are very likely to produce different proteins even when transformed with the same gene (for discussion, see Schubert, 2002).

Again, the corporate experimental model takes on faith that no genetic changes occur during the engineering process. As we are showing (and Schubert & Freese note) this assumption is known to be wrong.

Industry studies used surrogates to test MON 810 even though it was known by 1995 that the engineering process (using CaMV) was affecting DNA of the host plants:

Many Bt corn hybrids planted on millions of acres in the US are derived from Monsanto’s MON810 event, which contains the Cry1Ab insecticidal toxin discussed above. However, an unpublished molecular characterization study on MON810 reveals that the genetic construct broke apart during the transformation process, resulting in several unintended consequences (Levine et al., 1995). The following aberrant transfection events were noted: 1) An undefined portion of the E35S enhanced cauliflower mosaic virus promoter was incorporated into MON810; 2) Only a fragment (about 70%) of the intended full-length cry1Ab protoxin gene was incorporated; 3) Thus, by definition the NOS termination sequence was not integrated; instead, the cry1Ab gene fragment fused with enough DNA to code for 2 amino acids (Levine et al., 1995), DNA that apparently derives from the host plant. These unexpected transfection events create the potential for production of a fusion protein. Yet Western blots apparently did not reveal the predicted expression product of the open reading frame, a 92 kD fusion protein, but rather only a 63 kD “tryptic core” protein. Levine et al. speculate that their failure to detect the putative 92 kD fusion protein is “probably due to low expression or rapid degradation to the trypsin-resistant product during the extraction procedure.” The authors do not report any formal experiment to test either of these possibilities.

In addition, Lee et al. (1995) and Lee and Bailey (1995) report that the safety testing for MON810 and related Bt corn lines employed a bacterial surrogate Cry1Ab made in E. coli, not the fusion protein apparently produced by MON810. These two studies attempt to demonstrate equivalence between the plant-produced and bacterial surrogate Cry1Ab proteins to justify use of the latter in safety testing, yet the equivalence testing compared only the trypsin-generated cores of the plant and bacterial proteins. Results of testing with this bacterial surrogate clearly may not reflect the toxic and allergenic profile of the putative corn-produced fusion protein. Thus, the properties of the plant-expressed protein remain largely unknown (see Freese, 2001 for a fuller discussion).

Biologist & ecologist Barry Commoner has synthesized the above arguments about what is still unknown. He says that not enough information is known about genetic engineering in general, and about propriety corporate designs in particular. We have already found, though, that genetic modification can trigger surprising modifications to the genome of the host plant. He wrote in 2002:

Most alarming is the recent evidence that in a widely grown genetically modified food crop - soybeans containing an alien gene for herbicide resistance - the transgenic host plant's genome has itself been unwittingly altered. The Monsanto Company admitted in 2000 that its soybeans contained some extra fragments of the transferred gene, but nevertheless concluded that "no new proteins were expected or observed to be produced." A year later, Belgian researchers discovered that a segment of the plant's own DNA had been scrambled. The abnormal DNA was large enough to produce a new protein, a potentially harmful protein.

One way that such mystery DNA might arise is suggested by a recent study showing that in some plants carrying a bacterial gene, the plant 's enzymes that correct DNA replication errors rearrange the alien gene's nucleotide sequence. The consequences of such changes cannot be foreseen. The likelihood in genetically engineered crops of even exceedingly rare, disruptive effects of gene transfer is greatly amplified by the billions of individual transgenic plants already being grown annually in the United States.

The degree to which such disruptions do occur in genetically modified crops is not known at present, because the biotechnology industry is not required to provide even the most basic information about the actual composition of the transgenic plants to the regulatory agencies. No tests, for example, are required to show that the plant actually produces a protein with the same amino acid sequence as the original bacterial protein. Yet, this information is the only way to confirm that the transferred gene does in fact yield the theory-predicted product. Moreover, there are no required studies based on detailed analysis of the molecular structure and biochemical activity of the alien gene and its protein product in the transgenic commercial crop. Given that some unexpected effects may develop very slowly, crop plants should be monitored in successive generations as well. None of these essential tests are being performed, and billions of transgenic plants are now being grown with only the most rudimentary knowledge about the resulting changes in their composition. Without detailed, ongoing analyses of the transgenic crops, there is no way of knowing if hazardous consequences might arise. Given the failure of the central dogma, there is no assurance that they will not. The genetically engineered crops now being grown represent a massive uncontrolled experiment whose outcome is inherently unpredictable. The results could be catastrophic.

At the end of this quotation, Commoner alludes to a critique of the “Central Dogma” concept of biology. This Dogma is useful in helping biology students to understand DNA, RNA, and proteins. But of course, science is not about dogma. And now, there is a lot of new information to challenge this model of cell function.

Excessive faith in the central dogma produces images such as the one we see on many pages about genetic engineering. File:Breeding transgenesis cisgenesis.svg is a dangerously biased image because it represents the outcome of technological genetic engineering as identical to the outcome of traditional interbreeding—and this is simply not an accurate portrayal. As Commoner explains above, the process of shooting one organism's DNA directly into another organism's genome can have unintended effects on the genome itself.

Richard Strohmann, a pioneer of systems biology, had made a related argument:

The reason why Monsanto can claim scientific soundness is that they are only answering the technical question, `Can I move this gene and this characteristic from A to B?' They are not asking the questions that the current understanding of cell biology demands. You can ask the technical question and get the answer you are looking for. You can take a gene from A and put it into B. We know that. But that's the only question we can answer with certainty. We now realize that there are a whole host of other questions.

In the years after Commoner and Strohman wrote the above essays, a consortium of researchers working on the ENCODE project have released paradigm-shifting research that confirms their concerns about “Central Dogma” model. The first paper from this project was published in 2007, after four years of research.

The ENCODE consortium announced, first, that ”The human genome is pervasively transcribed, such that the majority of its bases are associated with at least one primary transcript and many transcripts link distal regions to established protein-coding loci.” A 2012 update states: “The vast majority (80.4%) of the human genome participates in at least one biochemical RNA- and/or chromatin-associated event in at least one cell type. Much of the genome lies close to a regulatory event: 95% of the genome lies within 8 kilobases (kb) of a DNA–protein interaction (as assayed by bound ChIP-seq motifs or DNase I footprints), and 99% is within 1.7 kb of at least one of the biochemical events measured by ENCODE.” This finding is, in itself, stunning, since it challenges the claim that most of the human genome is Noncoding DNA (a.k.a. “junk DNA”).

At the same time, the ENCODE researchers found that the human genome actually contains far fewer protein-coding genes than previously thought. One paper connected this research with summarizes how the results push the already-stressed “one gene one protein” model to its breaking point.

Some significant challenges to the gene were already in place. Many genes actually overlap each other within the DNA sequence. Some proteins are actually based on multiple separate DNA sequences, spliced together after smaller segments are transcribed into RNA. (Additional enzymes and RNA accomplish and regulate this splicing through channels that are not fully understood.) ENCODE research has found that these phenomena are widespread:

Overall, the ENCODE experiments have revealed a rich tapestry of transcription involving alternative splicing, covering the genome in a complex lattice of transcripts. According to traditional definitions, genes are unitary regions of DNA sequence, separated from each other. ENCODE reveals that if one attempts to define a gene on the basis of shared overlapping transcripts, then many annotated distinct gene loci coalesce into bigger genomic regions. One obvious implication of the ENCODE results is that there is less of a distinction to be made between genic and intergenic regions. Genes now appear to extend into what was once called intergenic space, with newly discovered transcripts originating from additional regulatory sites.

The ENCODE researchers also suggest that more diseases than previously thought might be attributable to single nucleotide polymorphisms—small mutations that can have large consequences.

The implications of this research are enormous and have only begun to reverberate through the scientific community. One particularly vitriolic and well-publicized criticism of the ENCODE results suggests that we must not interpret most of the genome as “functional”. This criticism does not challenge the basic findings that most of the genome is transcribed. We should predict more criticism from established labs and scientists, whose authority will be seriously undermined unless they can adapt to these new findings.

The fact is, the ENCODE findings are the opposite of “fringe” science. They're developed by a huge consortium, working with an unprecedented dataset, and they have published multiple peer-reviewed studies in Nature, Genome Biology, and Genome Research.

Don Lotter (whose work we mentioned above) makes the connection between the ENCODE results and genetic engineering. Lotter explains these results and then writes:

With the collapse of the one gene–one protein doctrine, and with the perspective of the mutational consequences of plant genetic engineering, the numerous ‘red flag’ incidents in the history of crop genetic engineering may begin to make sense. Many of these were incidents that, by themselves, should have put the scientific community on alert and put the entire process under intense scientific scrutiny. ”

Even the mainstream news reported on this connection. When the first ENCODE studies were published by Nature in 2007, Denise Caruso (a veteran biotech journalist) reported in the New York Times:

The $73.5 billion global biotech business may soon have to grapple with a discovery that calls into question the scientific principles on which it was founded. [...]

With that link now in place, the report is likely to have repercussions far beyond the laboratory. The presumption that genes operate independently has been institutionalized since 1976, when the first biotech company was founded. In fact, it is the economic and regulatory foundation on which the entire biotechnology industry is built. […]

The principle that gave rise to the biotech industry promised benefits that were equally compelling. Known as the Central Dogma of molecular biology, it stated that each gene in living organisms, from humans to bacteria, carries the information needed to construct one protein.

The scientists who invented recombinant DNA in 1973 built their innovation on this mechanistic, "one gene, one protein" principle.

Because donor genes could be associated with specific functions, with discrete properties and clear boundaries, scientists then believed that a gene from any organism could fit neatly and predictably into a larger design - one that products and companies could be built around, and that could be protected by intellectual-property laws.

This presumption, now disputed, is what one molecular biologist calls "the industrial gene." "The industrial gene is one that can be defined, owned, tracked, proven acceptably safe, proven to have uniform effect, sold and recalled," said Jack Heinemann, a professor of molecular biology in the School of Biological Sciences at the University of Canterbury in New Zealand and director of its Center for Integrated Research in Biosafety. […]

Even more important than patent laws are safety issues raised by the consortium's findings. Evidence of a networked genome shatters the scientific basis for virtually every official risk assessment of today's commercial biotech products, from genetically engineered crops to pharmaceuticals.

"The real worry for us has always been that the commercial agenda for biotech may be premature, based on what we have long known was an incomplete understanding of genetics," said Heinemann, who writes and teaches extensively on biosafety issues.

"Because gene patents and the genetic engineering process itself are both defined in terms of genes acting independently," he said, "regulators may be unaware of the potential impacts arising from these network effects." Yet to date, every attempt to challenge safety claims for biotech products has been categorically dismissed, or derided as unscientific.

In going through the sources cited as evidence of a “broad scientific consensus” we got some impression of how pro-GMO authors use very simple “Central Dogma” logic to argue a priori against scrutiny of genetic engineering. Recall how Henry I. Miller, for example, quoted a 1992 Nature editorial—to argue in 2009 that “golden rice” did not need to undergo testing. Miller maintains a tone of moral outrage as he argues that nothing is unknown about splicing certain known genes into a known organism.

Golden Rice, which, after the insertion of two genes coding for phytoene synthase (psy) and phytoenedesaturase (crt I), is able to accumulate b-carotene in the endosperm, the edible portion of the genetically altered rice grains (see http://www.goldenrice.org/Content2-How/how1_sci.html). The concept is simple: although rice plants do not normally synthesize b-carotene in the endosperm because of the absence of two necessary enzymes of the biosynthetic pathway, they do make it in the green portions of the plant. By using recombinant DNA techniques to insert the two genes that express these enzymes, the pathway becomes functional and the rice grains accumulate therapeutic amounts of b-carotene.

Miller equates the ability to describe a process using scientific vocabulary with mastery over the process. More concretely, he simply ignores the years worth of evidence that have shed light on the complex and unpredictable reality of genetic engineering. He is attributing predictable results to a technique proven to be unpredictable. But more and more scientists are taking issue with this type of logic.

Caruso mentions Miller directly:

“Both theory and experience confirm the extraordinary predictability and safety of gene-splicing technology and its products,” said Dr. Henry I. Miller, a fellow at the Hoover Institution who represented the pro-biotech position. Dr. Miller was the founding director of the Office of Biotechnology at the Food and Drug Administration, and presided over the approval of the first biotech food in 1992.

Now that the consortium’s findings have cast the validity of that theory into question, it may be time for the biotech industry to re-examine the more subtle effects of its products, and to share what it knows about them with regulators and other scientists.

Katharine Van Tassel argues in the Boston University Journal of Science & Technology Law that ENCODE has decisively overturned the assumptions behind the regime Miller promotes:

Directly contrary to the Central Dogma, in the past year numerous scientific discoveries involving the network effects of junk DNA, hybrid mRNA, SNPs and epigenetics have created a new model of a Networked Gene. Instead of viewing DNA as just a string of biological code, scientists have a new understanding that DNA is a highly complex operating system where a gene which expresses itself one way in a donor organism may not express itself the same way when dropped into an entirely different organism with its own complex operating system. In other words, scientists now know that genes operate in a highly contextual way, engaging in intricate biochemical cross- talk. Consequently, changing the context in which a gene operates can change the way the gene works. And changing how even one gene works can have a ‘butterfly effect’ on the entire organism. Critically, epigenetics and epigenetic inheritance explain that these unintended consequences can be passed on to future generations and may not manifest themselves until triggered by external environmental factors.

In the context of GM foods, a genetic modification changes the biochemical cross-talk between genes, creating genetic material that has never existed before in nature. This novel genetic material can create unintended health risks, as seen with the case of the GM peas that contained a novel and unexpected allergenic protein and primed test mice to react to other allergens.6 The bottom line is that the scientific acceptance of the existence of the networked gene establishes that the FDA’s presumption that GM plant food is bioequivalent to traditional plant food is no longer scientifically supportable and that a new system for GM plant food regulation is required.

Van Tassel highlights a crucial qualitative difference between conventional breeding and genetic engineering:

For hundreds of years, traditional breeding techniques have been used to add resistance to disease, enhance nutritional value and increase production yields of plants used for food.7 However, traditional techniques are limited to transferring genetic material between the same species, or a closely related species or genera: for example, between a wild variety of a plant and its modern crop variety.8 With the development of recombinant genetic technologies, genetic information can be transferred between different genuses such as a fish and a tomato9 or, through a process called gene stacking, between a round worm, a chicken and a pig.

She continues, discussing how the ENCODE findings explain certain disastrous results with genetic modification:

Directly contrary to the Central Dogma’s view of the gene, the ENCODE project reveals that many genes actually overlap one another and share stretches of molecular code.67 This study also overturns the long-held assumption that vast stretches of DNA that flank genes are just biologically inactive junk.68 Instead, sections of previously characterized junk DNA modulate a labyrinthine of silencing, switching and splicing operations (described below) necessary to sort out the complex messages sent by the overlapping genes.69 Adding another dimension to this complicated system, the ENCODE project demonstrated that “genes and the DNA sequences that regulate their activity are often far apart along the six-foot long strands of DNA.”70

This network effect, or biochemical cross-talk, can have a significant effect on protein expression, as was the case with the GM peas discussed in the next section. These effects are currently undetectable by the FDA as the products of gene expression are not tested as produced by the novel organism. 71 For example, under the FDA testing scheme, the fact that these GM peas had novel and unexpected allergenic properties would not have been discovered until after the GM peas had been introduced into the market and caused allergic reactions.

1. Genetically Modified Peas

An example of the impact that the network effect can have on protein expression came in 2002 when scientists at Australia’s national research organization, The Commonwealth Science and Industrial Research Organization (“CSIRO”), decided to end their 10 year-long project to bring GM peas to market.72 Green beans contain a natural protein that inhibits weevils from digesting starch which causes the weevils to starve to death.73 This protein has no history of allergenicity.74 In the CSIRO project, peas were genetically modified to contain this protein to provide the peas with the same protection against weevils that is found in green beans.75 Right before the GM peas were scheduled for market release, one of the scientists decided to perform animal testing on the protein as expressed by the GM peas. Surprisingly, not only was the protein discovered to be allergenic, it also primed the test mice to react to other allergens.76 The researchers discovered subtle differences in the way that sugars were added to the protein that were thought to be due to post-translational modification.77 This post-translational modification can’t be explained by the Central Dogma model; however, there are several possible explanations under the Networked Gene model. The first possibility is that the differences in the protein was caused by hybrid mRNA.

2. mRNA Hybrids

In addition to the discoveries regarding junk DNA, the ENCODE Project made several other significant discoveries, one of which involves mRNA. To create a protein, a cell must first transcribe a gene in DNA into messenger RNA (mRNA).78 The mRNA then drifts over to a ribosome which uses the mRNA as the instructions for assembling the amino acids into a chain that forms the protein. 79 Another basic precept of the Central Dogma is that one gene makes one copy of mRNA which makes one protein.80 The ENCODE project has demonstrated that DNA produces over twice the amount of mRNA than the Central Dogma predicts it should.81 In addition to the mRNA produced by the genes, mRNA transcripts are being produced that include both genes and their adjacent sections of junk DNA.82 Approximately eighty percent of mRNA produced by DNA is this “extra” mRNA.83 None of this extra mRNA is being used to create proteins.84 Some ENCODE scientists theorize that this extra mRNA may “fine-tune” or “modulate the activity of the genes themselves.”85

In the context of genetic modification, when a gene from one species is transplanted into the DNA of another species, mRNA is being created that is a hybrid of the transplanted gene and the host junk DNA. This is the creation of hybrid mRNA that has never existed before in nature. Whether this newly introduced genetic material is biologically active and whether this activity will alter the way in which critical proteins are produced (possibly the situation with the GM peas) and the amount of that production, have yet to be examined.

C. The Splicing Role of SNPs in Junk DNA

Another mechanism that could have produced the allergenic properties of GM peas is the splicing role of SNPs discovered by the Genome Regulators in Disease (GRID) Project published in January of 2008.86 This study revealed that very small variations in junk DNA, called SNPs (single nucleotide polymorphisms), control the natural processing of messenger RNA via a process called splicing.87 The SNP’s that are unique to an individual lead to changes in the splicing process that could be responsible for dramatic differences in the way that genes produce proteins.88 These differences are responsible for the vast variety of phenotypic differences (physical and physiological attributes) in individuals.89

“‘Regular’ splicing is the process by which long strings of nucleotides in a gene’s pre-messenger RNA (pre-mRNA) are discarded, and the remaining strings of nucleotides are spliced together into one continuous strand of messenger RNA (mRNA) that produces one unique protein.”90 However, regular splicing fails to explain how only 25,000 genes can produce all of the 100,000 proteins in the human cell.91 A new study published in December of 2008 reveals that the additional 75,000 proteins are produced by a process referred to as “alternative splicing,” defined as “a process that selectively activates alternative splicing sites along the pre-messenger RNA strand to assemble different subsets of RNA nucleotides into a variety of mRNAs. Each mRNA then produces a single protein.”92 The authors of this cutting edge study point out that the majority of genes utilize alternative splicing.93

More than one-half of all genetic diseases are caused by mistakes made in the alternative splicing process caused by mutations in DNA sequences.94 These mistakes can cause mRNA to include sequences that should have been deleted. Importantly, “small changes in a nucleotide sequence near a splice point can lead to large changes in the splice site choice and proteins produced.”95 Metaphorically, consider pre-mRNA as a long sentence. The “nucleotides and splice sites are the words of the sentence. Adding or deleting one word . . . can radically change the meaning of the sentence.”96 Taking this metaphor one step further, when scientists are genetically modifying a plant, they are making major edits to these sentences. These edits can make changes in the splicing processes, producing altered proteins that could have a ‘butterfly effect’ on the entire organism. This type of alteration may have been the cause of the newly allergenic properties of the GM peas.

Van Tassel confirms that the “substantial equivalence” model of safety has lost all validity:

Putting this new understanding of the highly contextual nature of genes together with epigenetic studies which demonstrate the myriad ways that the environment can activate or silence certain genes (allowing for billions of possible outcomes), it is easy to see how the new model of gene function challenges the simplistic assumption engendered by the Central Dogma which underlies the FDA’s regulatory scheme.

In fact, the hybrid mRNA discovery and the case of the GM peas do more than just challenge the presumption of bioequivalence; they provides direct evidence that the transferred genes and the products of their expression are not bioequivalent to their counter parts in the original organism. The ENCODE project demonstrated that eighty percent of the products of expression of a transferred gene are new. These mRNA hybrids, the product of the coupling of junk DNA from the donee, and the transferred gene from the donor, have never before existed in nature. Thus, the FDA can no longer claim that the donor product and the donee product are bioequivalent. Because they are not bioequivalent, the FDA will be hard pressed to continue in its position that common experience with the donor product can be used as proxy, or indirect, evidence that the donee product is equally safe.

• Many scientists think genetically engineered foods might not be as safe as their conventional counterparts. Particular concerns include:

• Effects of the herbicide Roundup, which people may consume along with genetically modified “Roundup Ready” plants.

• Effects of the insecticide Bt, particularly as produced endogenously by genetically modified plants.

• Unknown genetic effects resulting from the process of inserting DNA into a host organism.

• Policies and statements released by governments are represented as assurances of safety, but do not reflect scientific consensus.

• On the topic of toxicity in animal feeding studies, recent, high-quality literature reviews (and one meta-analysis) offer significantly conflicting results. There is an intense dispute in progress concerning which studies should be admitted into the data pool as evidence of toxicity.

• Industry-funded studies often do not even test the product they are declaring safe, substituting instead analogues which they misrepresent as identical.

• In the case of Roundup, industry studies have focused on glyphosate even though it is other ingredients which are lethal to humans.

• In the case of genetically modified crops, industry studies have sometimes tested lab-generated proteins rather than the crops themselves. These tests rely on the faulty assumption that genetic engineering itself has no effect on the genome of the host plant.

• Scientific consensus—particularly after ENCODE findings, first published in 2007—is trending away from the “Central Dogma” model of DNA, RNA, and protein. Because the Central Dogma model underpins claims of “substantial equivalence” advanced in the 1980s and early 1990s, safety claims made under these auspices are increasingly treated as suspect.

• Immediately remove the claim “There is broad scientific consensus that food on the market derived from GM crops pose no greater risk than conventional food” from all pages on Wikipedia where it currently appears.

• Remove the “fringe theories” tag at March Against Monsanto; and if the state of scientific opinion is to be discussed in addition to the claims of protestors, discuss both sides of the debate.

• Create a new page on “Safety of genetically modified food”, since it seems clear that the Genetically modified food controversies page is devoted to debates in the media, not in the scientific literature.

• Restore the page on Roundup in order to discuss effects beyond glyphosate (as well as marketing and commercial deployment).

• Use sources and information provided here to improve pages on ENCODE, Techniques of genetic engineering, Central dogma of molecular biology and other relevant pages.

Thank you for your continued attention. groupuscule (talk) 19:37, 14 June 2013 (UTC)