Humans killed mammoths using a combination of stone-tipped weapons, coordinated group tactics, and landscape traps, with evidence stretching back nearly 14,000 years in North America alone. No single method dominated. Instead, early humans adapted their strategies to terrain, group size, and the behavior of these massive animals, which could weigh over six tons.
The Earliest Direct Evidence
The oldest confirmed proof of humans hunting a mammoth-like animal in North America comes from the Manis site in Washington state. There, a bone projectile point was found embedded in the rib of a mastodon, dated to roughly 13,900 years ago. The point was made from mastodon bone itself, shaped thin and broad with sharp lateral edges. This predates the famous Clovis culture by about 900 years, showing that people were already crafting specialized weapons and targeting megafauna well before the toolkit most associated with mammoth hunting even existed.
By 13,000 years ago, Clovis hunters across North America were using distinctively fluted stone points. Sites like Naco and Lehner in Arizona, Miami in Texas, and Blackwater Draw in New Mexico all contain Clovis points found alongside mammoth remains, sometimes still embedded in bone. At Naco, multiple complete points were found with an unbutchered mammoth, suggesting the animal died from its wounds and walked away from the hunters before collapsing.
Weapons: Thrusting Spears and Braced Pikes
For years, researchers debated whether Clovis points were thrown at mammoths, thrust into them at close range, or simply used to butcher already-dead animals. A 2024 study in PLOS One proposed a compelling answer: many Clovis weapons may have functioned as braced pikes. In this scenario, a hunter would plant the butt of a long spear against the ground and let the charging or driven animal impale itself on the stone point, similar to how medieval soldiers braced pikes against cavalry.
The evidence for this comes partly from the points themselves. Clovis kill sites show a surprisingly low rate of the impact fractures you’d expect from thrown projectiles. Some researchers initially argued this meant the points weren’t used for hunting at all. But the pike hypothesis offers another explanation: the points were mounted in collapsible foreshafts designed so that on impact, the wooden shaft would split and the stone point would slide inward rather than snapping. This clever engineering reduced breakage, a serious problem with thrusting spears, where points shattered frequently on contact with thick hide and bone.
At Blackwater Draw, artifacts from a 1936 excavation appear to represent two such collapsing pike assemblies whose foreshafts buckled and points receded during impalement, exactly as the model predicts.
Pit Traps: Driving Mammoths to Their Deaths
Not all mammoth hunting relied on direct combat. In 2019, excavations at Tultepec, Mexico, revealed two enormous pits dating to 15,000 years ago that were unmistakably human-made traps. Each pit measured about 5.5 feet deep and 82 feet long. Inside them, researchers recovered 824 mammoth bones, including eight skulls, five jawbones, a hundred vertebrae, and 179 ribs.
This was a landmark discovery. Previous mammoth “megasites” in Eurasia and North America had always left open the question of whether humans were actively killing the animals or just scavenging carcasses. The Tultepec pits provided, for the first time, clear evidence of direct, premeditated attacks. Researchers believe groups of 20 to 30 hunters would isolate a single mammoth from its herd and drive it toward the pits using torches and branches. Once the animal fell in, it would be killed. The pits may have been part of a chain of traps, increasing the odds that a fleeing mammoth would stumble into one.
What About Driving Them Off Cliffs?
One popular idea held that Neanderthals and early humans drove mammoths off cliff edges, similar to the bison jumps used by Native Americans on the Great Plains. The most famous candidate for this was La Cotte de St Brelade on Jersey, one of the most spectacular Neanderthal sites in Europe, where heaps of mammoth and woolly rhinoceros bones were found piled at the base of a cliff.
A detailed reinvestigation, however, all but ruled this out. The plateau above the cliff was so rocky and uneven that mammoths would never have walked up there voluntarily. Even if they had, hunters would have needed to chase them down a steep dip and back up the other side before the animals reached the edge. As archaeologist Beccy Scott of the British Museum put it, there was simply no plausible way Neanderthals could have forced mammoths along that route. The bone accumulations likely resulted from other causes, possibly natural deaths or butchery of animals killed elsewhere and processed at the cliff base.
After the Kill: Butchering a Mammoth
Killing the animal was only the beginning. Processing a mammoth carcass was an enormous task, and stone tool cut marks on bones confirm that humans did it systematically. At La Cotte, cut marks found at the base of woolly mammoth tusks, in areas that would have been covered by skin, show that hunters were slicing through hide and connective tissue to detach the tusks from the skull. At Ambrona in Spain, scanning electron microscope analysis verified cut marks on an elephant skull and long bones, though they were few in number. A site in northern Italy yielded a single mammoth carcass with lithic artifacts and direct butchery evidence.
Confirmed cut marks on mammoth bones are actually quite rare across the archaeological record. This doesn’t mean butchery was uncommon. Stone tools often leave marks only on bone surfaces where flesh was thin, and most cuts happen through muscle and fat that leave no trace on the skeleton. The bones that survive tens of thousands of years in the ground represent a tiny fraction of the evidence that once existed.
How Much Did Hunting Contribute to Extinction?
The question of whether humans hunted mammoths to extinction remains one of the most debated topics in paleontology. A 2025 systematic review in Frontiers in Mammal Science tallied the scientific literature and found the field nearly evenly split: 23% of studies cited humans as the primary driver of late Quaternary megafauna extinctions, 23% cited climate change, and 20% proposed a combination of both. But the review noted a clear trend. Studies that actually model the extinction process, rather than just discussing it, overwhelmingly find little support for climate-only explanations and strong support for human-driven ones. This has made anthropogenic models increasingly dominant over the past decade.
The picture that emerges is not one of reckless slaughter but of a slow demographic squeeze. Mammoths reproduced slowly, with long pregnancies and years between calves. Even modest hunting pressure, sustained over centuries, could have tipped populations into decline, especially when combined with a warming climate that was already shrinking their preferred grassland habitat. Humans didn’t need to kill mammoths often. They just needed to kill them slightly faster than the animals could replace themselves.