The Endangered Species Act always had a hole in it. It was intended to protect ecosystems as well as individual species—it says so right in the original 1973 text—but it has no provisions to do so directly. For decades, conservationists successfully plugged that hole by arguing in court that the ESA’s prohibition of harm to individual species includes destroying a species’ habitat. Now the Trump administration wants to negate that argument by asserting that to harm an endangered species means only to injure or kill it directly: to rip it out by the roots or blow it away with a shotgun.

Habitat destruction has been the most common threat to endangered species in the U.S. since 1975. If the administration succeeds in redefining harm to exclude it, the Endangered Species Act won’t be able to effectively protect most endangered species.

That much of the act’s power can be destroyed by tweaking its definition of one phrase reveals its central weakness. Preserving old-growth forest for a single owl species (to give a classic example) means the forest—and everything living there—suddenly loses protection if that owl goes extinct anyway (as the northern spotted owl very well could). And the law requires that the government undertake heroic and expensive measures to save the most imperiled species, rather than using habitat protection to shore up populations before they truly crash. “The act has no concept of preventive medicine,” the conservation advocate and author Suzanne Winckler wrote in these pages in 1992. “On the contrary, it attempts to save the hardest cases, the equivalent of the terminally ill and the brain-dead.”

Conservationists haven’t really wanted to talk about this, though, on the theory that opening debate about the law would risk losing it all. The ESA passed during a unique moment in the early 1970s, when a Republican president could talk about the nation’s “environmental awakening,” and for all its flaws, the act is still considered one of the strongest and most effective biodiversity-protection laws in the world. But the Trump administration has now opened that debate—forcing a conversation about how we protect species and ecosystems that some conservationists say is long overdue.

Many conservationists have a long-standing dream solution to the ESA’s circuitous mechanism for protecting places: What if we just protected ecosystems directly? Forty-one percent of terrestrial American ecosystems are at risk of collapse, according to a 2023 report by NatureServe, a nonprofit that collects and analyzes data on biodiversity. Most of them are largely unprotected.

Jay Odenbaugh, an environmental philosopher at Lewis & Clark College, in Portland, Oregon, told me that shifting to protecting ecosystems would obviate the need to “chase down every last little species.” It would be more efficient. “We can’t save everything,” Odenbaugh said. “What we are trying to do is protect larger structural features.”

Reed Noss, a conservationist based at the University of Florida and the Southeastern Grasslands Institute, does still want to try to save every species. But he argues that only a few—large carnivores that face persecution and orchids collected for illegal trading, for example—need special, individual protections. Meanwhile, Noss estimates that 85 percent of species could be saved by simply protecting a sufficiently large chunk of each type of American ecosystem. He has therefore been one of the most vocal advocates for what he calls a “native ecosystem–protection act” to supplement the ESA since the 1990s.

The U.S. already has multiple systems that categorize lands and fresh water into ecosystem types. The U.S. National Vegetation Classification, for instance, describes natural systems at a series of scales from very broad types, such as “Forest & Woodland,” to hyper-specific descriptors, such as “Eastern White Pine-Eastern Hemlock Lower New England-Northern Piedmont Forest.” An ecosystem-protection act would direct the government to choose (or develop) one such classification system, then ensure that each type of ecosystem had sufficient area protected.

Making that decision would surely involve ecologists arguing over how to categorize ecosystems. Philosophers might argue about whether ecosystems even exist—if they are more than the sum of the organisms that comprise them. But, for the purposes of policy, more important than arriving at essential truths would be creating categories that make sense to the public and describe the things the public cares about: old-growth forest, tallgrass prairie, the Everglades, Great Basin sagebrush steppe, the deciduous forests of the Northeast, and so on. Something like this was tried with Pacific Northwest old-growth forest in the 1990s; known as the Northwest Forest Plan, it is meant to protect not just the owl but old growth more broadly—but the plan, which is still in use, covers only one ecosystem type.

Part of the appeal of a system that directly protects ecosystems is that it recognizes that they’re dynamic. Species have always moved and evolved, shifting the composition and relationships within systems through time. And today, climate change is prompting many species to move. But Odenbaugh and Noss see ecosystems as entities that will remain coherent enough to protect. Florida, for instance, has sandhill ecosystems (sandy hills that support longleaf pine and oaks with wire grass) and wet flatwoods (which are seasonally inundated)—and “a sandhill and a flatwoods are going to remain a sandhill and a flatwoods even if their species composition changes due to climate change,” Noss told me. A robust network of many different kinds of ecosystems—especially one well connected by corridors so species can move—would support and protect most of America’s species without the government having to develop a separate plan for each flower and bee.

Many who fight on conservation’s front lines still hesitate to advocate for such a law. The Environmental Species Act, as it is, achieves similar purposes, they argue—and it could be pushed in the opposite direction that the Trump administration wants to pull it.

When I spoke with Kierán Suckling, executive director for the Center for Biological Diversity, which is dedicated to forcing the federal government to abide by its own environmental laws, he described his vision of a conservation-minded president who could, like Donald Trump, use executive power quickly and aggressively, only to conserve nature. “The secretary of the interior and the head of Fish and Wildlife, they have, already, the power under the ESA to do basically anything they want, as long as it is supported by the best available science,” he said. So, in theory, they could translocate species to help them survive climate change, or broaden the boundaries of “critical habitat,” which is protected from destruction by actions taken, permitted, or funded by the federal government (unless exceptions are granted).

Daniel Rohlf, a law professor at Lewis & Clark College who has studied the ESA for more than three decades, agrees that decisive leadership could do more to protect ecosystems by skillfully wielding the current ESA: “Critical habitat” could be treated as sacrosanct. Federal actions could be assessed not just for direct harm to species but for the harm they would cause via greenhouse-gas emissions. The “range” of a species could be defined as its historic or possible range, not just the scraps of territory it clings to in the present. “You could do all that tomorrow under the current version of the act,” Rohlf told me. And he believes that, unlike many of the actions Trump is taking, a lot of these stronger interpretations would likely hold up in court.

The political prospects for an entirely new ecosystem-protection act are low, even in a Democratic administration: Although 60 percent of Americans tell pollsters that “stricter environmental laws and regulations are worth the cost,” these days politicians of all stripes seem to want to cut red tape and build stuff. And Suckling believes that his organization and others like it will be able to block or undo Trump’s proposed changes to the ESA’s definition of harm. “We overturned all his first-term ESA regulation changes and are confident we’ll overturn this one as well,” he said. The U.S. may well just keep conserving the way we have been, through the ESA, and often in court.

But an ecosystem-protection act could also be a unifying cause. Love for American landscapes is bipartisan, and protecting ecosystems would not necessarily mean outlawing all human use inside them. Ranching and recreation are compatible with many ecosystems. Tribal management could protect biodiversity and support traditional use. Caring for these ecosystems takes work, and that means jobs—physical, outdoor jobs, many of which can be filled by people without college degrees. Farmers and ranchers can also be compensated for tending to ecosystems in addition to growing food, buffering their income from the vagaries of extreme weather and trade wars.

The United States is an idea, but it is also a place, a beautiful quilt of ecosystems that are not valuable just because they contain “biodiversity” or even because they filter our water, produce fish and game, and store carbon. Our forests, prairies, mountains, coastlines, and swamps are knit into our sense of who we are, both individually and as a people. We love them, and we have the power to protect them, if we choose to.