A field dominated by palmer amaranth, or pigweed
A field dominated by palmer amaranth, or pigweed, one of the plants that has gained glyphosate resistance.
Delaware Agriculture

There’s a clear scientific consensus that heavy use of glyphosate — the active ingredient in Roundup and other brands of herbicide — has sped up the evolution of glyphosate-resistant weeds. And it’s reasonable to assume that crops genetically engineered to work hand in glove with glyphosate (like Roundup-resistant soy) are part of the problem, contributing to the popularity of the weed killer.

Now crops genetically engineered to work with other herbicides — such as dicamba and 2,4-D — look like they will soon come on line. The seed companies’ answer to the Roundup-resistance problem is: Let’s just fall back on older herbicides. An editorial published by the journal Nature recently criticized this plan. If we do the same thing with dicamba and 2,4-D that we did with glyphosate, the editorial argued, history is likely to repeat itself.

This got me wondering what we should do, then, so I started calling weed scientists. I ended up talking with three from around the country. They all agreed on the basic premise.

The increase in glyphosate use resulted in “way more glyphosate-resistant weeds, that’s indisputable,” said Andrew Kniss, an assistant professor at the University of Wyoming.

“Glyphosate was so effective, and so cheap, and so easy, so that’s what we did. People thought it was a miracle,” said Larry Steckel, a weed specialist at the University of Tennessee, a state where farmers have had serious problems with herbicide-resistant weeds.

They cautioned that the main problem was glyphosate itself, not the GMOs: The first glyphosate-resistant weeds popped up in Malaysia and Australia where — at the time — there were no glyphosate-tolerant GMOs, Kniss said. But they also agree that the main boom in glyphosate use really did have something to do with GE crops.

“The way the GMO herbicide-resistant crops were deployed was like the worst possible scenario for developing resistance,” said Carol Mallory-Smith, a professor of weed science at Oregon State University.

Of course, weeds would have developed resistance sooner or later even without the GMOs — evolution is inevitable that way. But the widespread use of these crops meant that many farmers went from one application of glyphosate every few years to multiple applications every year, and that increased the evolutionary selection pressure on weeds.

In other words, the problem is less the technology than how we use it — and we used it in precisely the wrong way.

The next question is, how could we change the rules so that farmers conserved these herbicides to maximize their useful life? It’s a classic tragedy-of-the-commons scenario. Every farmer is good at making the best weed-control decisions for an individual farm and choosing glyphosate, but all these individual decisions add up to a bad outcome for farmers as a whole: glyphosate-resistant weeds. One solution would be for the federal government to wade in with regulations — the Nature editorial suggests that the EPA should crack down on the use of herbicides — but none of the weed scientists I talked to thought that was a great idea. It’s tricky for a bureaucracy in Washington, D.C., to make good decisions for a multitude of farmers in different areas.

“It’s hard to imagine a herbicide resistance plan that would work for more than one farm, or even more than one field,” said Kniss. “Let alone a whole country.”

If we want a healthier environment, it doesn’t necessarily make sense to force farmers to use less herbicide, Mallory-Smith said. “Part of why we are so focused on herbicides has to do with environmental concerns,” Mallory-Smith said. We’ve asked farmers to do less plowing and to stop burning their fields, and so they have turned to herbicides.

Many people have a deep aversion to the idea of spraying chemicals on the fields, but Steckel said that herbicides are often the most environmentally friendly solution. “Herbicides to me are kind of like medicine. If used correctly they are not a danger to people or the environment,” he said. In contrast, he says, killing the weeds by plowing has proved to be an environmental disaster in his part of the world. “These soils just won’t hold it,” Steckel said. “When tillage was commonplace, all our soil was headed to New Orleans.”

That’s not to say that herbicides are harmless, but they have to be weighed against the alternatives. It would make the most sense to mix it up, varying by the dirt, the weather, and what the farmer did the previous year. Centralized control would surely force farmers to make some dumb decisions that were actually worse for the environment.

When economist Elinor Ostrom proposed her Nobel-winning solution for the tragedy of the commons she rejected both centralized control and just letting the market guide itself. Her solution has people coming together to govern themselves with rule that make sense on the ground. Kniss suggested something like this: Perhaps farmers could make their own weed-management plans and submit them to some local authority, who would make sure everyone was doing their share to slow the evolution of resistance.

Mallory-Smith had a different solution: “My suggestion to Monsanto was to take their salesmen off commission and put them on salary,” she said. That suggestion didn’t go over so well, but ideally the companies should be suggesting that farmers use products from their competitors, depending on the situation. Change things up enough and you’ll slow resistance. But companies have to do just the opposite: Advise farmers to buy only its chemicals. In the end, she’s pessimistic.

“My guess is that we’ll go down the same path again,” Mallory-Smith said.

Steckel disagrees — he doesn’t expect to see these herbicides dominate the market the way glyphosate did, because they just aren’t as good. Dicamba and 2,4-D have been around for a long time, and have never had the addictive appeal of glyphosate. Dicamba, for instance, only works if you spray weeds when they are less than five inches tall, while glyphosate can kill a full-grown weed (or at least it could). And that huge difference in effectiveness will remain if dicamba and 2,4-D tolerant GMOs are released.

What’s the best way through this tough spot? Pursue the science further and look for alternative controls, Mallory-Smith said. Look at the big picture, Kniss said. “This really is a symptom of the larger problem, of not enough diversity in our cropping systems,” he said. And we should also count our blessings: The intense use of glyphosate had led to less use of the other herbicides. We’ve sped up the development of glyphosate-resistance but slowed other forms of resistance, he said.

Meanwhile, farmers will make do, one way or another. Some have actually gone back to a primitive technology: hiring laborers to weed by hand. But, Steckel said, even that’s not foolproof. In India, workers weeded barnyard grass out of rice paddies by looking for the red stem. Eventually, the weed evolved a green stem — so it looked just like the rice. It had evolved resistance to hand weeding. Nature always finds a way.