How not to debunk anti-GMO propaganda

Graham Brookes and Bruce Chassy wrote a detailed rebuttal of a paper by G.F. Botta et al. in the American Journal of Plant Science, which questioned the value of GMO soybeans in Argentina. On the basis of their analysis, Brookes and Chassy conclude that:

The major deficiencies identified in this paper lead us to question the thoroughness of the review process undertaken by the American Journal of Plant Sciences, as it is our professional assessment that this paper should not have been accepted for publication in any reputable peer review journal.

I cannot dispute that. I can, however, point out that in seeking to orchestrate an invincible attack, Brookes and Chassy are perhaps guilty of stressing a single, not very helpful view of the nature of diversity, which is also entirely unnecessary to discredit the Botta et al. paper.

The main thrust of their rebuttal is to challenge the statements surrounding the use of the weedkiller glyphosate, and as an outsider that convinced me that the original paper was indeed deeply flawed. The two experts then go on to address another claim in the original paper.

‘Up until not recently agrarian diversity had always been increasing. However, in industrialized countries, plant and animal genetic engineers, trading houses and governments themselves combined forces to supply new varieties and uniform breeds that would replace the tremendous heterogeneity already existing’.1

That’s easy enough to skewer, and Brookes and Chassy do so well enough.

From the very beginning of agriculture biodiversity of crops has, in fact, been continually contracting as farmers and breeders selected the most desirable crops (see for example, L.T. Evans. Feeding the 10 Billion, Cambridge University Press).

Sloppy to confuse crops and varieties, but hardly criminal. Unfortunately, from my perspective, instead of leaving it at that, they go on to help readers to do something that they say Botta et al. have not done, viz.

to comprehend the nature of breeding modern crop varieties. While a single or limited number of varieties may be grown in one region at a particular time, this does not mean biodiversity is being lost. Quite the contrary, since modern varieties have complex combinations of parents this results in the incorporation of numerous diverse traits from many ancestors.

Yes. And, what is your point?

This argument deliberately or ignorantly confuses the differences among genetic diversity in a pedigree, genetic diversity within a population, and genetic diversity among populations planted at a given time in a given place. Brookes and Chassy then go on to show how crop failures that the rest of us associate with “reductions in ‘genetic variety’” are in fact no such thing. Forget their view of the Irish potato famine. Consider instead their account of Southern Corn Leaf blight, which halved maize yields in the US in 1970. Here is how Brooks and Chassy describe the cause of the epidemic:

In 1970, all US corn had, what is known as N cytoplasm (a designation of the genotype of the mitochondria).  Around this time a second cytoplasm called T was introduced which facilitated male sterility for hybrid seed production.  Ironically, it was the new cytoplasm that was susceptible to blight.

So far, so true. As a result of the widespread adoption of Texas cytoplasm to confer male sterility – because it makes the production of F1 hybrids easier and cheaper – almost all the maize planted in the US shared a single type of mitochondrial genome that rendered them all susceptible to Southern Corn Leaf blight. As far as that susceptibility is concerned, there was genetic uniformity in the crop, no genetic diversity at all. So how do Brookes and Chassy describe that?

So, in this case, it was an increase in diversity—not a decrease—that caused the problem.

How can they possibly say that?

Before the widespread adoption of Texas cytoplasm, there was a range of genetically-based susceptibilities among maize varieties. Afterwards, all varieties were susceptible. Because they all shared one set of genes.

Here is how a panel of the US National Academy of Sciences described what happened:

The corn crop fell victim to the epidemic because of a quirk in the technology that had redesigned the corn plants of America until, in one sense, they had become as alike as identical twins. Whatever made one plant susceptible made them all susceptible.
Uniformity is the key word–the plants were uniform in that special sense, and uniformity in a crop is an essential prerequisite to genetic vulnerability.

There are other aspects of the Brookes and Chassy rebuttal that I am unhappy with2 but I’m sufficiently self aware not to pursue them here.

Instead, I’ll just conclude that genetic diversity in a pedigree is not the same as genetic diversity in a population, a species, or an ecosystem. Especially not as far as resilience is concerned.

Just the facts, ma’am
Friends will, I know, be wondering why on earth I’m bothering. Indeed, after a little digging, I wondered why Brookes and Chassy bothered too. The American Journal of Plant Sciences is published by Scientific Research Publishing, which has been described as a scam and which “borrowed” papers published elsewhere and called them its own. It may be peer-reviewed, in the strict sense, but it is also clearly a joke. So why bother debunking a joke? Equally, Academics Review, the site where Brookes and Chassy (who is one of the founders) published their debunk, seems to be withering on the vine. The debunk itself isn’t dated, although on 20 March David Tribe (another founder of the site) gave notice that Chassy and Barfoot have just put out a dissection of Botta et al.3 Few of the pages at Tribe and Chassy’s site seem to have a date, and there hasn’t been any “news” since May 2010. It’s all a bit sad, because a repository of countervailing facts would be useful, especially if not sullied by internal errors.

  1. Let’s leave aside entirely the very strange wording of this paragraph from Botta et al., and indeed the very title of their paper, despite the obvious clues they contain as to the worth of the journal they chose as the happy recipient of their research. []
  2. Ex-situ conservation is “much more effective at preserving biodiversity”. Why “farmers all over the world are willing to buy improved seeds”. []
  3. I’ve no idea who this Barfoot cove might be. []

12 Replies to “How not to debunk anti-GMO propaganda”

  1. Some of my colleagues at IRRI in the early 90s used to argue that by combining lots of different alleles into rice varieties, and disseminating them through INGER, diversity was being increased. What was the diversity of a crop of IR64, I asked for example, grown across millions of hectares? Since all plants were (assumed to be) identical, zero diversity between plants. Could never get them to undertsnad that aspect.

  2. I am frustrated with the lack of a reasonable definition of agricultural biodiversity both within and across species. Yes, there may be genetic material from more sources in any given field due to breeding, but if all of the plants are the same, it’s meaningless for any real purposes like disease resistance. I understand that uniformity is one of the ways that we’ve gotten to high yields but maybe we can sacrifice that a little bit and plant >1 varieties that are close to uniform. In my thesis I toyed with the idea of evolutionary breeding à la Suneson, and sometimes I really do wonder if that would be a be a better way to go than inbreeding and hybridization, at least in low-input situations. Like that would ever happen.

    1. There are definitions — or rather, measurements — around the place. A couple that are gaining ground for use on farms are Richness and Evenness (which can be measured for different levels, like field, farm and community.

      Planting variety mixtures is a good option under many circumstances, and there is good evidence that it protects short term against some diseases and pests. One would also expect it to delay the emergence of resistance. It is being researched, but as long as externalities dominate the economics to the extent that they do is unlikely to get much traction. Alas.

      1. Jeremy, would you share more details about measuring Richness and Evenness? Is R diversity at crop/field/social level, and E a measurement of intraspecies allelic diversity in relation to other variables (population size, potential # of allellic combinations)? Do either measure (or predict) functional resilience?

        1. Ha. Funny you should say that. I’ve got a paper here that I need to do a Research Blogging for, which is based on those measures. So far, E is not allelic, but only based on variety names. But in principle you could do that. Essentially, R is the number of distinct varieties (or alleles etc) and E is the area (as a fraction of the whole) occupied by each.

          See A global perspective of the richness and evenness of traditional crop-variety diversity maintained by farming communities.

  3. There is some interesting discussion here about biodiversity, but I would like to make the point that in terms of protecting crops against infections by pests , diversity of different protection traits within a crop increases the resilience of the protection systems at least to the identified pests corresponding to the traits.

    It seems to me that the colleagues of Mike Jackson at IRRI may have been thinking about the relative functional diversity of protection present in any purposefully bred crops relative to the case if those traits had not been deployed. This is the kind of counterfactual argument about functional biodiversity. I know this is not the same type of concept as degree of variation of genetic components between different plants in the field but it does have some connection with a significant component of crop protection against disease, which is the reason mainly we are talking about biodiversity in this situation,

    Another type of diversity of we might talk about is patches of different types of crops in different fields which would have some increased protection against disease and this could be achieved by rotation of different varieties and different types of crops from season to season in some kind of randomised fashion across the local region. Definitely encouragement of sensible crop rotations of some sorts is in a sense increasing the biodiversity, but in any one season all plants within one field would be the same.

    1. Point made. In a bred variety, lots of different alleles are combined into a single entity from a multiplicity of sources, and in a sense that individual can be considered to be more diverse than varieties it replaced. But within the crop there is zero diversity between plants (in the case of a self fertilizing crop like rice), and this is something several of my colleagues just couldn’t accept.

  4. The above comments are dealing with environmental selection (pests and diseases) on various types of genetic mixtures. Farmers also can be involved in selection. When I was a collector for the Kenya/FAO dryland farming project in Katumani my samples of sorghum were rapidly screened by agronomists and breeders, put in some kind of composite mix (I don’t know the technical term), seed multiplied and distributed direct to farmers, all within two years of collection. Advice was given to farmers for them to manage the mixture as they wished, for example, by selecting for earliness or apparent drought tolerance. I left before the results were out but the technique seemed in no way anything unusual to the project staff. This combines environmental selection with farmer selection in multiple locations (monitoring results must have been a problem). From memory, Harlan Sr. and Martini did something similar for environmental selection around 75 years ago in barley bulk populations.

    1. It sounds to me like they were creating landraces. Landraces are a subject which i am extremley interested in. Mainly my interest comes from the idea that landraces with lots of gentic diversity have lots of genetics which can help them to adapt to a wide range of biotic and abiotic factors like disease, pests, climate adaption, etc.

      I am currently experimenting with this idea at home. I am currently working on creating my own watermelon landrace, sunflower landrace, and possibly others so i can adapt my own varieties easily to my growing conditions.

  5. I’m glad to see that Jeremy’s commentary on our “Botta critique” has prompted the above discussion. I think Jeremy makes a number of excellent points. And I am appreciative that he noticed the column we posted and thought it merited a response.

    One point of total agreement with Jeremy’s observations is that we have not been able to grow as fast as we had hoped. David and I both still have day jobs and funders have not been generous in supporting the concept. Nonetheless we have just launched the Spanish language edition at and hope in the near future to add several more topics. Clearly bad science is accumulating and misinformation is being spread faster than any of us can clean it up. Maybe this should a higher priority for those of us who are interested in helping society get the science right.

    As to why we would even comment on a paper in a journal of such low impact and quality as the American Journal of Plant Sciences I can only say that the purpose of such articles published in lack-luster quasi peer-reviewed journals is not to impress scientific colleagues. I believe it stems from a carefully planned strategy and a deliberate effort to create peer-reviewed references that can be cited not to the community of scientists but to the unsuspecting non-scientist, journalist and even policymaker. These kinds of papers are cited by reviewers and in the media. They are literally creating a new reality and substituting biased worldviews for the legitimate scientific literature. If the scientific community doesn’t ferret our frauds and fabrications they will become truths. That’s why we singled out Botta et al. To be honest, it was a random shot and a spit in the ocean. We could have chosen from dozens of other bad papers.

    I have asked the webfolks to delete the comments on biodiversity to which Jeremy took exception. Not because I think he is right and we were wrong but simply because they are not necessary to the points we were making. They were a parenthetic comment that we almost didn’t include in the first place. That said, it is obvious from the foregoing discussion, and from the reaction to our comments about biodiversity, that one must be very careful to spell out very clearly what one means when venturing an opinion on biodiversity. We apparently did not so a clarification follows.

    Crops are not natural, they are man-made. They should not be viewed as some sort of germplasm bank that preserves biodiversity. That is not their role in the big scheme of things. But plant breeders do combine as many different useful traits into new varieties as is possible; much of what plant breeders do is cross or introgress multiple disease resistance traits into crops. In that sense a single so-called “monoculture” of corn represents many diverse genetic backgrounds–some wild, some of landrace origin, and some from modern varieties. And in the example discussed in the Botta review, it was exactly that kind of attempt to diversify corn germplasm that went astray. There’s no denying corn is a monoculture but it one that over the years has been vested with many diverse traits.

    Biodiversity has taken on an almost spiritual meaning. I do not mean to in any way diminish the importance of biodiversity but sometimes when people start believing they stop thinking. Biodiversity doesn’t necessarily provide insurance of survival and lack of biodiversity isn’t necessary a death warrant. A couple of examples follow.

    The American Chestnut comes to mind as species that was diverse and widespread through diverse ecosystems in the US. Three billion trees, 25-30% of all trees in the Eastern US, were nonetheless wiped out by a single disease, Asian bark fungus (Cryphonectria parasitica) in a few short decades. Just a few scattered individuals remain of this once common tree. Biodiversity did not save them.

    On the other hand, the Quaking Aspen (Populus tremuloides) is propagated by root and shoots. A grove of these Aspen is essentially a giant monoclone; the specie itself appears to have very little biodiversity. They have survived for perhaps 100s of millions of years and have been called the largest living organisms since all the trees in a grove are connected by roots and can be thought of as part of one distributed organism. Lack of diversity appears not to have hurt this specie.

    A number of common crops are propagated clonally as well. Naval oranges, bananas, potatoes, pineapples, apples are examples. Why doesn’t biodiversity matter to these crops? Because they were produced by plant breeders and are propagated commercially before being planted by farmers. As noted above, crops aren’t a repository in which we bank biodiversity, they are just human constructs we use to feed ourselves and our animals, and to provide fiber. Now that I have thought about it for a while I think we should have made these more clearly and should have taken Botta et al. to task for conflating agriculture and biodiversity. I thank Jeremy for pointing out that we had not thought this through. We will edit our critique accordingly. That’s one of the wonderful things about writing on the web….

    1. You really plan to delete the sections I was disagreeing with? If you think being able to cover your tracks is “one of the wonderful things about writing on the web,” then I fear for your objectivity in anything.

      I continue to disagree with your view of biodiversity.

Leave a Reply

Your email address will not be published. Required fields are marked *