Yesterday we had an ESPM (Environmental Science and Policy Management) guy crash PMB beer hour. It was an interesting experience, but one thing really worried me. In a discussion of “so what do you guys do in PMB?” the crasher mentioned “well I know about the people studying corn.” I’m sure you can just picture my face lighting up as I get ready to talk about the fascinating research of people like Sarah Hake and Damon Lisch, or worst case, to defend studying corn a person who might be radically anti-GMO. But of course I couldn’t be that lucky. When he said studying corn, he didn’t mean plant development, or transposons, or paramutation, or any of the other cool things corn can teach us. He meant the research of Ignacio Chapela.
First of all let me say Dr. Chapela is not a member of the department of Plant and Microbial Biology. He belongs to the same department as our party crasher, Environmental Science and Policy Management. In 2001 he published a paper, in Nature, asserting the discovery of transgenes in Mexican landrace* corn. This was big news because Mexico had banned the cultivation of transgenic maize. The implications of the paper were that #1 pollen was drifting hundreds of miles across the border from America #2 Transgenes were surviving from generation to generation in local maize populations. Unfortunately, his analysis was called into question, and other labs were unable to replicate his results from the paper.
In debates with anti-technology types Chapela’s paper will often come up, and my usual response has been to cite the 2005 paper in PNAS that, after looking at 870 plants from 125 fields and 18 locations (153,746 seeds!) in the same region as Chapela collected his sample, was unable to find any containing transgenic traits**, and then I’ll point out that Nature, possibly for the first time ever without the agreement of the author, retracted his paper saying:
Nature has concluded that the evidence available is not sufficient to justify the publication of the original paper.
For a scientist, the first, inability to replicate the results with more samples in other labs, is the more damning argument.*** But some people find arguments from authority more convincing. For example: “One of the two most prestigious scientific journals in the world said their evidence wasn’t good enough.”
Either way, this was the first time I’ve heard a comeback to the effect of “oh no, they de-retracted it.” Something I can’t find any evidence of online.
When people can’t even agree on facts, it’s impossible to expect us to agree on policy. The same can be seen in politics. The best example I can think of are whether WMDs were ever found in Iraq after the invasion. And in the same way, there can be compromise over policy (look at the health care bills being created in congress as we speak) but it’s impossible to compromise on facts. I remember being lectured for using the expression “these two genes are 60% homologous.”
“Homology is like pregnancy,” said my (excellent) lecturer, “you either are or you aren’t.”
In the same way, a fact is either true or false (or unproven). Every so often I wish for a cross between snopes and mythbusters devoted to everything from “Where Obama was born” to “Are we eating GM potatoes” (we aren’t), but the real solution to ignorance will certainly not nearly so simple.
*Landraces are really cool regional verities of corn. With incredibly diversity, they’re the product of hundreds or thousands of years of selective breeding by farmers living in different regions of central America. And they look really cool too. While some people get caught in the trap of thinking of these lines as ancient, they’re actually constantly changing as farmers search their fields for new desirable traits and trade seed between neighboring villages.
**It’s impossible to prove a negative (that no corn anywhere in central America contains the genes), but it’s possible to demonstrate that the trait couldn’t be present at a level of above say 1%
***Independent replication of results is the gold standard in science. One lab can be mistaken, or commit fraud, or even be subject to weird local variables (contamination in a tube of some chemical used by everyone in the lab). The way to disprove all those possible sources of error is to have another lab can do the same things you describe and have the same results.