The missing context in the “Mathematical” argument against GMOs
Our new guest writer Stuart Hayashi examines a basic flaw with the idea that math coupled with the use of the precautionary principle argues for the danger of GMOs.
Stuart is a freelance writer based in Hawaii. He is the author of The Freedom of Peaceful Action: On the Origin of Individual Rights and Life in the Market Ecosystem. He has also spent time as an analyst and aide at the Hawaii State Capitol in both legislative houses.
Are you interested in writing about a science related topic? Then check out our information on becoming a guest writer.
The Important Context Omitted from the “Mathematical” Argument Supposedly Proving the Unacceptable Danger of GMOs
By Stuart K. Hayashi
Some public commentaries proclaim that, on account of a philosophic notion called the Precautionary Principle, a moratorium or ban should be placed on the use of genetically modified organisms (GMOs). Among such commentaries, we sometimes find the claim that, when you look at the math, you should concede that GMOs are too dangerous for human use and that genetic engineering technology should be curtailed. The argument is as follows. If we talk about the risk of a GMO doing damage on any one particular day, it seems that that risk is minuscule. But what is the statistical risk of a GMO inflicting harm one day . . . eventually? As time advances, that risk of a GMO eventually causing turmoil increases exponentially just as, with every passing minute, your risk of dying increases. Therefore, the argument concludes that as long as transgenic technology is employed, it is inevitable that one day, something devastating concerning GMOs will occur. Therefore, the one method whereby we can guard ourselves against this otherwise-impending harm is to avoid usage of genetic engineering altogether.
A serious flaw detracts from that argument. The argument is akin to an experiment employing an experimental sample but no control sample. The experimental sample in question is the risk to human life and the ecosystem posed by the use of transgenic technology. However, it is ultimately meaningless to speculate about such a risk in a vacuum. In real life, there are myriad other risks. For the supposed calculation of the risk of employing GMOs to provide any contextual meaning, that risk must be compared against other risks—known risks. The risk of GMO usage is not put into perspective until it is compared against the alternative. There are risks within agriculture—both within the organic and conventional methods—that can be mitigated by GMO technology. For example, if GMO technology can reduce the quantity and amount of natural-resource inputs per unit of food produced, that can mollify some environmental risks that for decades have been associated with farming prior to the development of agricultural GMOs.
The “mathematical argument” against GMOs therefore omits important context. It fails to compare the risk of transgenic technologies against the risk of using currently existing technologies in the absence of transgenics and other GMO-related technologies. In a controlled experiment, the experimental sample would measure the risks associated with GMOs and the control sample would provide data on the risks associated with not using GMOs.
Moreover, many of the hypothetical risks presumed to arise strictly from the advent of GMO technology actually have already been present since farmers first engaged in traditional selective breeding. One might say that there is the chance that a GMO will run loose and become an “invasive species” that out-competes other organisms in the wild. But consider that traditional selective breeding has always brought forth this very same risk. Something selectively bred in captivity can also escape into the wild and theoretically become an “invasive species.” Yet organisms bred in captivity have seldom caused catastrophe when released in the wild, because such organisms are normally dependent on human beings to survive and propagate. On account of their being bred primarily to exhibit traits that are useful to humans, such organisms created for human use seldom express traits that make them well-suited to propagating in the wild.
Those circumstances are unlikely to change with GMOs. When scientists alter an isolated gene of some species—say, corn—for agricultural applications, that change is not radical but incremental and targeted. The scientists change a very specific protein—a very specific gene. By contrast, when you engage in traditional selective breeding, thousands more genes are sorted randomly. Even if you alter twenty specific genetic sequences simultaneously through transgenic technology, those alternations will, overall, still be less drastic, on a chemical level, than the alterations made through selective breeding. Therefore, when you engineer corn through old-fashioned crossbreeding, that crossbreeding imposes a much more radical change in the gene pool of corn plants than does the recombination of an isolated sequence of DNA.
It is possible that, one day, someone might exploit genetic engineering to weaponize a disease such as anthrax or smallpox. However, that is a matter very different from scientists altering corn and rice for agricultural purposes. GMOs are a very broad category of life forms and products, and the “mathematical argument” is too vague and ill-defined in its concept of what “GMO” entails. Given the broadness of the GMO category, the possibility that someone might try to weaponize anthrax, for instance, fails to provide evidence that it is unacceptably dangerous to employ transgenics for the purpose of allowing corn to produce Bt to protect the crop from predatory insects.
Image Credit: Lynn Friedman | CC
I grew up on an organic grain farm in Saskatchewan, and worked for five years as a USDA-contract organic inspector. But I left the organic movement when I realized it was all just a bureaucratic scam designed to propel a political agenda.
If the precautionary principle was applied to the organic industry it would have been banned long ago.
I’m glad that someone else has had serious doubts about the “mathematical” model. One cannot apply the Precautionary Principle unless there is potential for ruin, which is a fact even stated by Taleb himself (http://www.fooledbyrandomness.com/pp2.pdf). In section 10.3, the reason for why GMOs should have the Precautionary Principle applied to them is stated. The reason that Taleb states is that genetic modification does not mix with monoculture. He then goes on to that in nature mutations happen, and plants with harmful mutations are selected out (die). This happens slowly over time, whereas in genetic modification, the change happens all at once. He also states that any one mutation is unlikely to ever become a majority of the population. However, I do not believe he is correct about any of these points.
The first is monoculture. How is this a “GMO problem” when many conventional and organic farms use monoculture as well? I don’t see how he continues to get away with mentioning that as a reason to apply PP to GMOs.
The second is that the information provided on basic plant evolution is flawed. The argument used in this section is an appeal to nature in disguise. Natural mutations = good because they take a long time. GMOs = bad because the transgenes take less time to be put in the genome. This argument doesn’t take into account the history of polyploidy in plant evolution. Allopolyploidy (having one set of chromosomes from two different species) events have taken place throughout plant evolution. All plants alive today are the products of at least one allopolyploid event sometime in the past. An allopolyploid event happens in one generation (not how Taleb describes plant evolution) and changes the expression of many, many genes (not just 1-3 like in GMOs). Furthermore, this plant likely will not be able to breed with either of the parent plants, meaning it will form a new species if it can breed with itself and is suitable for its environment. I would also like to mention that horizontal gene transfer, chromosomal rearrangements, autoployploidy, and aneuoploidy are all genetic events that can take place in just one generation as well.
The third is that his reasons for calling GMOs “systemic” are unfounded. Could the pollen from GMOs really spread around the world? Let’s think about this for a bit. The reproductive, temporal, and geographical boundaries that are in place here are great. However, assuming those are somehow not enough on their own, I still don’t think it would be possible. GMOs have no further ability to reproduce than their non-GMO counterparts. Pollen from all of the conventional, organic, and heirloom varieties of our crops all have the ability to pollinate wild relatives (and vis versa). If the pollen from these plants hasn’t been accused of “contaminating” the gene pool of plants around then world, then how can we justify naming this as a possibility for GMOs? It just doesn’t make sense.
Since we’re speaking mathematically:
GMO’s are not equal (or substantially equivalent) to there non GMO counterparts
If they were indeed equal ~ GM detection tests would have nothing to detect
It’s hard to tell if you’re being facetious or not.
If you are being facetious, then disregard this comment. However if you are not being facetious, the point of the article is that speaking mathematically about the safety of GMOs is an unsound position to take.
GM detection kits are nothing more than protein or DNA detection kits. You can use these kits to detect non-gmo strains equally as well. It’s only a matter as to what proteins and/or DNA the kit is tuned for.
[…] (Lanham, Maryland, 2014–15). His articles on philosophy and current events have been published in Skepti-Forum, Ideas in Action TV, Scout & Engineer, and the Honolulu Advertiser, and his commentaries on […]