Few animals turn up on more American surfaces than the wild horse. It gallops across car hoods and beer labels and gymnasium floors, across the letterhead of banks that have never seen a rangeland. We have made the mustang into a national shorthand — for freedom, for the frontier, for whatever it is we suspect we have lost. That is an extraordinary amount of symbolic labor to ask of an animal. And symbols, it turns out, are considerably easier to protect than animals are.
On August 20, The New York Times published an investigation into the Bureau of Land Management's Wild Horse and Burro Sale Authority program, reporting that the agency has been selling federally protected mustangs — some for as little as $25 a head — to buyers who resell them within days into an export pipeline that ends at slaughterhouses across the Mexican and Canadian borders. An Ohio livestock trader named in the reporting is described as having taken hundreds of animals out of the program — 496 in a single year, by one sanctuary's count, under a bulk purchase contract with shipping at public expense. More than 3,700 horses moved through the sale program after the current administration took office, roughly double the previous pace (Culliton, 2026); Skydog Sanctuary, which tracks the sales, puts the nineteen-month total above 6,300. Killing a wild horse with federal money is illegal. So is selling one into that pipeline. Congress put both prohibitions in the same sentence.
The law being worked around here is the Wild Free-Roaming Horses and Burros Act of 1971, and its opening sentence is the tell. Congress declared that wild horses and burros "are living symbols of the historic and pioneer spirit of the West," that "they contribute to the diversity of life forms within the Nation and enrich the lives of the American people" (Wild Free-Roaming Horses and Burros Act, 1971). Read that again with an eye for what it does not say. It does not say these animals can suffer. It does not say they have interests of their own, or lives that go better and worse from the inside. The statute protects them roughly the way we protect a monument: for what they mean to us. Fifty-five years on, that framing is precisely where the structure gives way — because a symbol, whatever else is true of it, cannot be hurt.
Well — the rest is plumbing, and the plumbing is worth following, because the word loophole flatters everyone involved. A 2004 rider known as the Burns amendment rewrote the 1971 Act to say that qualifying excess animals — those over ten years old, or passed over for adoption three times — shall be made available for sale "without limitation." Congress, alarmed at what that invited, has attached language to every Interior spending bill since fiscal 2010 barring appropriated funds from "the destruction of healthy, unadopted, wild horses and burros" or from "the sale of wild horses and burros that results in their destruction for processing into commercial products." Both halves are in there. The BLM's own sale contracts bind buyers not to resell for slaughter. So the pipeline does not run through a hole in the statute. It runs through the distance between a rule and its enforcement — a bill of sale nobody follows up on, a buyer nobody screens, a four-animal purchase cap the Secretary is free to waive. In March 2025 a federal judge in Colorado set aside the BLM's $1,000-per-head Adoption Incentive Program, which advocates had long called a pipeline to slaughter. The traffic did not stop. It moved one program over.
This is not really a story about a clever lawyer finding a gap in a statute; there is no gap. It is close to a textbook demonstration of what Albert Bandura called moral disengagement — the ordinary cognitive maneuvers by which decent people take part in harm without ever experiencing themselves as harming anyone (Bandura, 1999). Bandura catalogued the moves, and the sale program runs most of them at once. Displacement of responsibility: the agency did not kill the horse; the buyer did. Diffusion of responsibility: the chain is long enough that no single link feels load-bearing. Euphemistic labeling: "sale authority," "excess animals," "good home," and — in the Heritage Foundation's Project 2025 blueprint for Interior — the recommendation that Congress permit the BLM "to dispose humanely of these animals" (Pendley, 2023). Disposal is a word for objects. It arrives in the sentence with the question already settled.
Indeed, the whole arrangement holds because of what we already believe about horses. In American moral life the horse sits in the pet drawer, alongside the dogs and cats, the animals that get names. We closed our own slaughter plants in 2007; we do not, as a culture, eat them. And yet tens of thousands of American horses cross a border every year to be eaten by people who do. Hal Herzog's phrase for this is "some we love, some we hate, some we eat" (Herzog, 2010), and his point was never that we are hypocrites. It is that the categories are unstable, largely unexamined, and assigned by culture long before reason shows up — a sorting stable enough within a person that researchers can measure it, and now simply call it speciesism (Caviola et al., 2019). Here the sorting is done by geography. The horse is a pet on this side of the line and livestock on the other, and the animal does not change at the checkpoint. Only our view of it does.
There is a second mechanism running underneath. Research on the meat paradox finds that when people are reminded an animal is going to be eaten, they quietly revise downward their estimates of that animal's capacity to think and to feel (Bastian et al., 2012). The denial is not a lie anyone tells; it is an adjustment made fast and without noticing, in whichever direction permits comfort. And it travels with the animal. A mustang standing on a ridge in Nevada is a mind — that is the entire basis of the mythology. The same mustang in a holding pen with a freeze brand and a lot number is inventory. Nothing about the horse moved. The category did.
Then the numbers do their own quiet work. The BLM's own figures put 85,466 horses and burros on the range as of March 2026, against an appropriate management level of 25,592, and 58,025 more standing in off-range corrals and pastures that May; holding them consumed roughly $101 million in fiscal 2024, about two-thirds of the program's budget (Bureau of Land Management, 2026). Those figures are accurate and they are also anesthetic. Decades of work on the identifiable victim effect show that sympathy attaches to the one and drains away from the many — that people give more, and feel more, for a single named individual than for a statistical multitude, and that inviting them to think analytically about the totals suppresses the feeling further (Small et al., 2007). A herd is a management problem. A gray mare who has stood in a pen for six years is not.
So — the strongest version of the other side, which deserves stating plainly rather than waved past. The range genuinely is over capacity, by better than three to one. Horses in a drought year die badly, of thirst and starvation, and nobody who has watched it calls that freedom. Holding fifty-eight thousand animals for life is neither humane nor affordable, and sanctuary capacity — however heroic the sanctuaries are — will never absorb that number. Advocates who oppose every removal, every fertility program, and every disposition pathway do own a share of the impasse. That argument is real, and indignation does not answer it.
But notice what it presupposes: that the horses are fundamentally a supply problem, and that the only live question is which outflow to choose. And notice what it does not license. Whatever one concludes about roundups or contraception, no version of the overpopulation argument requires the federal government to sell a protected animal for twenty-five dollars to someone it has ample reason to suspect will resell it into a slaughter pipeline within a week, and then treat the outcome as none of its concern. That is not management. That is laundering, and the euphemism is the giveaway.
It is also not new, which is the part worth sitting with. Between 2008 and 2012, under an entirely different administration, a Colorado rancher named Tom Davis bought 1,794 wild horses from the BLM and later acknowledged sending very nearly all of them to Mexico. The Interior Department's Inspector General reported in 2015 that the agency had not followed the law, and had not followed its own policy either (U.S. Department of the Interior, Office of Inspector General, 2015). Neither federal nor state prosecutors brought a case. A four-animals-per-buyer cap adopted the year before is still on the books — and, as this month's reporting shows, still routinely set aside. The rule survived. It just stopped applying. A gap that survives two decades and both parties is not really a loophole. It is the design doing what it was built to do, and the design was set in 1971, the moment Congress reached for the word symbol.
Which is why the repair cannot only be procedural. Researchers who study the moral circle have shown that its width is a real and measurable thing about a person — that how far someone extends moral concern is stable, varies enormously between individuals, and tracks closely with empathy, with a felt connection to the natural world, and with what they will actually give up for a creature outside the usual boundary (Crimston et al., 2016). The circle is not fixed. It is drawn, and it can be drawn wider. We have been looking at the wild horse for fifty-five years and seeing an idea: the West, the frontier, ourselves. It is a beautiful idea, and it has protected almost nothing. The mare in the pen in Nevada does not know she is a living symbol of the historic and pioneer spirit of anything. She knows the sound of the truck. Whatever we owe her, we owe her for that — and not for the myth she was drafted into.