Ogmius Exchange

Science, Genetically Modified Foods, and the Rumsfeld Doctrine

As they have for the last ten years, next spring U.S. farmers will plant more genetically-modified (GM) corn, cotton and soybean seeds.  GM crops have been a huge success with farmers: today, more than 80% of the cotton and soybeans grown in the U.S., and over half the corn, are GM varieties.  Biotech crops have probably been the most rapidly-adopted agricultural technology in history.

But the biotech revolution has stalled.  Today, nearly ten years after the first crops came to market, the same four major GM crops are being grown, with the same two GM traits – pest-resistance and herbicide-tolerance. GM varieties of sugar beets and potatoes have been approved by regulators, but they’re not on the market because no one wanted to buy them.  Recently, Monsanto announced after years of development that it was shelving GM wheat – in part because wheat farmers weren’t sure they wanted it.

What happened?

Biotech’s big problem is a nagging concern about the safety of GM foods, especially in the foreign markets that are the big buyers of U.S. grain and seed products.  Marketers are concerned that consumers will simply reject biotech foods and foods with ingredients derived from GM crops.  As a result, lucrative export markets in Europe, Japan, and other nations are effectively closed to new GM varieties.

The debate about the safety of biotech foods has been going on unabated for at least 15 years.  Biotech advocates argue that companies have done extensive safety testing on GM products.  They point to the numerous reports of independent scientific experts, including the National Academy of Sciences, that testify to the safety of biotech foods.  They point to the nearly ten years that U.S. consumers have been eating foods derived from GM crops without a “single, solitary example of a sniffle or headache,” as one industry spokesman puts it.  The lingering fear about the safety of biotech foods drives the biotech industry crazy.

Maybe they should talk to Defense Secretary Donald Rumsfeld.

Rumsfeld famously stated, “[T]here are known knowns; there are the things we know we know.  We also know there are known unknowns; that is to say there are some things we do not know.  But there are also unknown unknowns – the ones we don’t know we don’t know.”

Biotech’s safety perception problem comes from the “unknown unknowns.”  Company scientists know about the new protein that is intended to be expressed in the engineered plant, and they can test it for toxicity and allergenicity.  And they know how to look for the “known unknowns”: the unintended genetic changes that can change the known nutritional composition of a food.   But the trick is the “unknown unknowns”: how do you begin to look for subtle changes in the thousands of proteins in a typical whole food when you have no idea of what to look for or where to look for it?

It’s the “what if?” question:  what if scientists discover something bad twenty years from now?  It’s not an altogether unreasonable concern, given the well-publicized stories of once-promising technologies whose flaws became apparent only after widespread use.  

The latest GM food report by the National Research Council, Safety of Genetically Engineered Foods: Approaches to Assessing Unintended Health Effects, wrestled with this exact question.   The NRC panel re-confirmed that there’s nothing uniquely risky about genetic engineering: all breeding techniques carry some risk of creating unintended genetic changes.   But after reviewing existing and possible future technologies to screen genetic changes in new foods, the NRC panel concluded that safety has to be assessed on a case-by-case basis and cannot be 100% assured in advance.  Among other things, the NRC recommended a better monitoring program to detect any problems that might be caused by the introduction of new foods.

Did this latest attempt to use science to reassure skittish consumers succeed?   The biotechnology industry hoped so: one spokesman said, “[T]his report should lay to rest the few naysayers who continue to question the safety of these crops.”   Apparently, the naysayers didn’t read the same script.  A spokesperson for one group, the Center for Food Safety, said instead that the “NAS report tells us that genetic engineering may cause harmful unexpected changes in our food.” 

One of the reasons this debate has proven so intractable is because it assumes that safety is solely a scientific issue.  That assumption is misplaced.   Science can tell us about risks, but people bring their own set of values to bear in determining whether those risks are acceptable.  We willingly take risks every day that would be totally unacceptable in a different context.   We don’t think twice about driving to work despite the quantifiable risk of death or injury.  The same level of risk from a nearby toxic waste dump or from contaminated food would be intolerable.

Part of the judgment that people make about risks also depends on their perceptions of benefits.  People understand, for example, that drugs may have side effects, but they are willing to accept that risk as a trade-off for the drug’s benefits.  But when there are no perceived consumer benefits – as is the case for the biotech foods currently on the market – consumers are likely to be more risk-averse, particularly when they have alternatives they believe are safer.

Obviously, we don’t consciously perform quantitative risk-benefit calculations on the myriad daily uncertainties of life.  If we tried, we’d never do anything.   As a practical matter, we exercise some simple heuristics.   Chief among those is prior experience: if I drove to work safely the last 100 times, chances are pretty good I can do so again today.  (Not that this is always a good heuristic: continuing to smoke because it hasn’t killed you yet is probably not the optimal answer from a health perspective.)   In public opinion polls on biotech food, for example, public concerns about safety are reduced when people learn that they’ve already been eating it for ten years.

We also rely on trusted proxies.  They could be neighbors, media, public interest groups or the family doctor – sources that have proven in the past to provide trustworthy advice.  On food safety issues, the food industry isn’t highly credible: consumers recognize that companies want to sell their products.    In the U.S., the most trusted source of information about food safety turns out to be the government, particularly the FDA.   In contrast, European consumers have little trust in their governments in the wake of the “mad cow” debacle, where governments initially tried to reassure consumers that there were no risks in eating meat – and then later had to admit there were.

The lack of prior experience with biotech foods, combined with the perceived lack of benefit and the absence of any trusted proxy on the safety issue, has led to the current skepticism about safety and hostility toward biotech foods in Europe and other parts of the world.   More assurances from scientists that such concerns are misplaced are unlikely to change the dynamic.   Fears about the “unknown unknowns” can be overcome only through experience and trust, neither of which can be earned overnight.