Nassim Taleb, author of The Black Swan and NYU statistician extraordinaire, has ducked out of a debate with me that had been arranged by the Journal of Markets and Morality. The topic of the debate was to be, "Do GMOs [genetically modified organisms] present cause for moral concern?"
"The goal of this controversy," the editor explained, "is to assist our readership (economists, political scientists, theologians, moral philosophers, ethicists) in developing a more informed understanding of the issues at stake in the current state of the GMO debate, addressing concerns of fact, morality, and policy." The plan had been to publish two rounds of back-and-forth between us.
Taleb was invited to participate because he and several colleagues had earlier published, at his fooledbyrandomness website, a very anti-GMO working paper, "The Precautionary Principle (with Application to the Genetic Modification of Organisms." In that paper, Taleb and his colleagues claimed that "GMOs represent a public risk of global harm," suggesting that GMOs could result in global "ecocide" and even perhaps "the extinction of human beings or all life on the planet." Taleb and another colleague were afforded the opportunity to express their alarm in The New York Times, where they declared that "the risk of G.M.O.s are more severe than those of finance." Human extinction is certainly worse than even a global financial meltdown. The upshot is that Taleb wants "prescribe severe limits on GMOs."
After Taleb had agreed to contribute to the journal's GMO debate, the managing editor contacted me and asked if I would like to participate. Citing my "previous scholarly and popular work and experience," he asked me to write the first essay in the debate series. I duly submitted my essay, in which I debunk the many claims made by Taleb and his colleagues about the dangers allegedly posed by modern biotech crops. I conclude that it was fallacious arguments against developing and growing modern biotech crops that are cause for great moral concern.
I waited for Taleb's response. It never arrived. The editor told me last week that Taleb, for reasons unclear, had withdrawn from the debate.
But since my essay responds pretty directly to the claims made in Taleb's anti-GMO working paper, let's go ahead and debate anyway.
Do GMOs Present Cause for Moral Concern?
Banning biotech crops under the pretense of implementing a "non-naive" version of the precautionary principle would be a great moral wrong. Such a ban would deny access to the significant known benefits that modern biotechnology is already providing to human beings and the natural world, all based on wholly unjustified assertions that these crops one day will somehow produce catastrophic "ruin."
First, let's review the extensive benefits offered by the current versions of biotech crops. Next, let's evaluate what recent research has found with regard to the human health and ecological safety concerns associated with modern biotech crops. We'll end by considering the argument that the absence of evidence of harm is not evidence of absence of an inevitable GMO doom.
So far, biotech crops have chiefly been enhanced to resist pests and herbicide applications, although other traits—including resistance to disease, drought, and salt—are now being made available to farmers too. Pest resistance has generally been instilled by adding versions of a gene from the soil microbe Bacillus thuringiensis (Bt) for a crystal protein that kills insect pests when it is activated in their alkaline guts. Bt has been used widely in organic farming. Decades of research have shown that it is safe for people and other vertebrates to eat. Herbicide resistance has been conferred on most modern crops by adding a gene for the EPSPS protein obtained from the soil microbe Agrobacterium sp. strain CP4. Again, research has shown that the amount of the EPSPS protein regularly consumed by people is safe to eat.
In 2014, a group of Italian biologists did a comprehensive review of the last 10 years of research on biotech crops that encompassed 1,783 different scientific studies. These studies dealt with such concerns as the crops' impacts on natural biodiversity, the possibility that they'll exchange genes with wild relatives, and their effects on the health of people and other animals. In the review, the biologists concluded that "the scientific research conducted so far has not detected any significant hazard directly connected with the use of GM crops."
So most scientific evidence finds that biotech crops are safe for people and the environment. What then are the benefits? In a 2014 meta-analysis of 147 studies, a team of German researchers reports that the global adoption of genetically modified crops has reduced chemical pesticide use by 37 percent, increased crop yields by 22 percent, and increased farmer profits by 68 percent. They conclude that there is "robust evidence of GM crop benefits for farmers in developed and developing countries." Therefore it is no surprise that farmers around the world have (when regulators permit it) embraced these enhanced crop varieties. The global extent of biotech crops has increased more than 100-fold from 4.2 million acres in 1996 to about 450 million acres in 2014. Eighteen million farmers in 28 countries planted them in 2014.
Ideological opposition to biotech crops is actually killing people and harming the natural world. Consider the case of Golden Rice, in which non-profit Swiss researchers used genetic engineering to boost the production of the vitamin A precursor beta-carotene. Researchers estimate that vitamin A deficiency is responsible for 1.9 to 2.8 million preventable deaths annually, mostly of children under five years old and women. Some 125 million children under age five suffer from vitamin A deficiency, of which 250,000 to 500,000 go blind each year. Half of these children die within a year of becoming blind. Vitamin A deficiency also weakens immune responses, greatly increasing the risk of severe illness. According to one 2014 estimate, the 10-year delay in getting Golden Rice to poor farming families has resulted in the loss of 1.4 million life-years in India alone.
Meanwhile, biotech crops increase yields. Increased yields mean that farmers can grow more food, feed, and fiber on less acreage, thus sparing more land for nature. Biotech crops are partially responsible for the imminent arrival of peak farmland. If global crop yields had remained stuck at 1960 levels, farmers around the world would have needed about 3 billion more hectares to grow enough food for the world's current population. That's about the size of the USA, Canada, and China combined.
Instead, the amount of land farmed for crops increased from 1,371 million hectares in 1961 to 1,533 million hectares in 2009. In other words, the amount of land plowed increased by only about 10 percent as yields have tripled. Recent research estimates that continued agricultural intensification, including the further deployment of genetically enhanced crops, could return 150 million hectares of land to nature by 2060. The amount of farmland restored to nature could rise to as much as 400 million hectares if ethanol subsidies were eliminated.
In addition, shifting wood and pulp production to plantations of faster-growing genetically enhanced trees would help speed the trend toward restoring natural forests and thus aid in the protection of many endangered species. In addition, plant breeders are making strides toward dramatically lowering the amount of nitrogen that fertilizer crops need, strengthening disease resistance, incorporating salt and drought tolerance, and boosting photosynthetic efficiency. Much progress is being made using molecular techniques that enable researchers to rapidly identify and crossbreed crop plants that express promising traits like drought tolerance, disease resistance, or higher yields.
The advent of the fantastically versatile CRISPR genome-editing technology gives plant breeders "molecular scissors" that can cut and paste genetic information, producing stable and heritable genomic changes quickly and easily without introducing foreign DNA. Heretofore, once plant breeders had identified a useful gene—say, one for disease resistance—in a landrace, it would take years of crossbreeding to transfer it to a high-yielding cultivar of the same crop species. With CRISPR, if breeders identify a natural gene variant conferring natural fungus resistance in a less productive landrace, they can simply edit the corresponding gene variant to match in the more productive strain, thus conferring the same natural fungus resistance on it. Using CRISPR means that plant breeders can dramatically speed up the process of getting useful genes into high-yielding crop varieties.
Given these clear health and environmental benefits, why do Taleb and his colleagues so fear modern biotech crops that they want to "prescribe severe limits on GMOs"?
Biotech Crops and the Alleged Risk of Human Extinction
Taleb and his colleagues want to impose their supposedly non-naive version of the precautionary principle to forestall activities when "consequences can involve total irreversible ruin, such as the extinction of human beings or all life on the planet." And GMOs, they feel, could result in "irreversible environmental and health damage" or cause "an irreversible termination of life at some scale, which could be planetwide." They Biotech crops, they claim, pose a systemic risk of global ecocide.
It is a trivially true statement that if some activity will eventually lead to total ruin, then total ruin, even if it takes a long time, will eventually follow that activity. Taleb and his colleagues just assume that producing and growing modern biotech crops is such an activity, then trivially predict a GMO apocalypse. There is a lot of hand-waving about the dangers of global connectivity and dose response relationships that may be relevant to the workings of financial markets, but they provide no justification for their assumption of biotech disaster. Unwarranted dire assumptions in; unjustified devastating consequences out.
First, as University of Milan researcher Giovanni Tagliabue has pointed out, Taleb and his colleagues lump modern biotech crops and the technologies used to produce them into a "nonsensical GMO pseudo-category." Tagliabue argues that it is "theoretically and practically impossible to precisely specify the supposed common denominator" for crops modified using many disparate techniques for resistance to disease, drought, salt, herbicides, or pests or for agronomic characteristics such better fertilizer response, reductions in allergens or carcinogens, improved fruit quality, and so forth.
Despite the lack of an adequate definition of what they fear, Taleb and company assert that the precautionary principle applies to GMOs "because their risk is systemic. There are two aspects of systemic risk, the widespread impact on the ecosystem and the widespread impact on health." They more or less repeat themselves by additionally claiming that "the systemic global impacts of GMOs arise from a combination of (1) engineered genetic modifications, (2) monoculture—the use of single crops over large areas." Evidently the alleged systemic risk to ecosystems stems from their assertion that "GMOs have the propensity to spread uncontrollably, and thus their risks cannot be localized." With regard to purported health risks posed by GMOs, they claim that "foods derived from GMOs are not tested in humans before they are marketed." Another odd claim is that "biologists and agronomists are adopting top down engineering strategies and taking global systemic risks in introducing organisms into the wild."
Let's take these one by one.
Do modern biotech crops have a propensity to spread uncontrollably? No. No knowledgeable person worries that modern cultivars of corn, soybeans, cotton, rice, wheat, potatoes, and so forth will invade and take over forests, prairies and wetlands. The fact is that artificial selection for characteristics that increase yields and make harvest easier have made crops unfit to compete in the wild. "No GMO plant, or any other vegetable for that matter, is capable of spreading uncontrollably across the planet," notes Tagliabue.
Over the centuries of domestication, farmers and crop breeders made vast changes in crops by unknowingly targeting genomic regions containing both structural and regulatory genes. Farmers were selecting novel and improved strains generated by chance mutation and variation. Conceptually, it's the same process today, just using better and less random methods.
Perhaps by "spread uncontrollably" they are expressing badly articulated worries that genes from biotech crops will somehow invade and devastate the natural world? It is well established that over the centuries domesticated crops have regularly exchanged genes with their wild relatives. Such hybridizations can have deleterious ecological effects such as increasing weediness or reducing evolutionary fitness. Genes from biotech crops have also been outcrossed into wild relatives, but their effects are similar to those that occur when conventional cultivars and wild relatives exchange genes. Producing hardier weeds or occasionally causing a rare plant species to go extinct does not amount to anything like global ecocide.
What about monoculture? In a sense, agriculture is monoculture. After all, it was invented thousands of years ago when farmers figured out that they could get more food by segregating plants that they wanted to grow (crops) into fields that excluded competing plants that they didn't want (weeds). It is true that large swathes of arable land are devoted to growing crops like wheat, corn, rice, soybeans, and potatoes. But monoculture is not particular to biotechnology and does not mean genetic homogeneity. For example, a 2010 survey of just U.S. seed catalogs reported that farmers had more the 7,000, 3,800 and 100 different varieties of corn, soybean, and cotton seeds respectively available to them. In addition, crop breeders using either conventional or modern biotech methods have successfully maintained the overall genetic diversity of the cultivars they release to farmers.
What about the claim that foods made from biotech crops are not tested in people? In the United States, the Food and Drug Administration is responsible for overseeing the safety of foods. The agency states, "Foods from GE [genetically engineered] plants must meet the same food safety requirements as foods derived from traditionally bred plants." Except for pharmaceuticals, potentially toxic substances are not generally tested in people. Even in that case, pharmaceutical regulators have arguably proven to be far too cautious when assessing the safety of new drugs.
To ascertain the safety of non-drug substances, researchers use standard tests in animals and human cells. Companies that want to sell new biotech crop varieties must submit safety-testing results to the FDA. The agency will not approve the new cultivar until it is satisfied that it safe for humans to eat. As usual, Taleb and company deploy their inapt rhetorical strategy and simply assert that the tests the FDA relies upon are inadequate. Never mind that for nearly 20 years human beings and livestock have eaten trillions of servings of food and feed made from biotech crops with no scientifically discernible harm.
Taleb and his colleagues claim that "human experience over generations has chosen the biological organisms that are relatively safe for consumption." But that is simply not true. For nearly a century now, plant breeders have been massively rearranging the genomes of crop plants by blasting them with radiation or exposing them to caustic chemicals. The Food and Agriculture Organization's Mutant Varieties Database lists over 3,200 officially released mutant varieties from 214 different plant species grown in more than 60 countries throughout the world. Over 1,000 of the listed mutants are cultivars of major staple crops that farmers plant on tens of millions of hectares. If these varieties are safe for people and the environment (and they are), then it stretches the imagination beyond the breaking point to believe that the comparatively minor changes made using more precise modern biotechnological techniques could result in planetwide devastation and human extinction.
Finding no smoking gun, Taleb and his colleagues must hyperbolically conjure one and liken growing biotech crops to playing Russian roulette. They have it backwards. By pushing to ban biotech crops, Taleb and company are demanding that poor people continue to spin those metaphoric cylinders whose chambers are already fully loaded with real disease, hunger, and back-breaking labor. Modern biotechnology can empty a few of those chambers, and thus reduce the chances that when the trigger is pulled disaster will ensue.
"For [genetically modified] crops to be part of the solution, biosafety assessments should not be overly politically-driven or a burdensome impedance to delivering this technology broadly," the ecologist Peter Raven has cogently argued. "Biosafety scientists and policy makers need to recognize the undeniable truth that inappropriate actions resulting in indecision also have negative consequences. It is no longer acceptable to delay the use of any strategy that is safe and will help us achieve the ability to feed the world's people." Fallacious arguments against developing and growing modern biotech crops is cause for great moral concern.