The Bow Arrived Late, Spread Fast, and Changed Everything (Almost Everywhere)

Around 1,400 years ago, hunters across an enormous stretch of western North America picked up the bow and arrow for the first time. Within a few generations, across the deserts and canyons from northern Mexico to California and the American Southwest, the older weapon system (the atlatl, a hand-held spear-throwing lever) had all but vanished. Further north, in the glacial highlands of the Yukon and Alaska, the story played out differently: hunters kept both weapons in their kit for another thousand years.

That dual picture comes from a new study published in PNAS Nexus this March, led by Briggs Buchanan of the University of Tulsa, alongside Marcus Hamilton (University of Texas at San Antonio and the Santa Fe Institute), Metin Eren (Kent State University and the McDonald Institute for Archaeological Research, Cambridge), and Robert Walker (University of Missouri). The team compiled 140 radiocarbon dates from 136 well-preserved organic weapons, the actual wooden shafts and frames of atlatls, darts, bows, and arrows, recovered from sites spanning roughly 10,000 years.

Why Organic Weapons Matter

Most of what archaeologists find from prehistoric weapon systems are stone or bone tips. The wooden shafts, bowstaves, and atlatl boards rot away in ordinary soil conditions. That means researchers are usually forced to guess which delivery system a stone point belonged to, based on weight, shape, and size. It works, but it introduces uncertainty.

The weapons in this study survived because they were recovered from environments that act like natural time capsules: glacial ice patches in Canada and Alaska, and dry caves and rock shelters across the southwestern deserts. When you can radiocarbon-date the actual bow or atlatl rather than just its stone tip, you know, with very little ambiguity, what weapon was in use and when.

A Simultaneous Debut, Two Very Different Outcomes

Using statistical techniques including optimal linear estimation and Bayesian logistic regression, the team found that the bow first appears at roughly the same moment (around 1,400 calibrated years before present) in both the northern and southern portions of western North America. That near-simultaneous timing, Buchanan and colleagues argue, is best explained by a single invention event followed by rapid cultural transmission, rather than by people in different regions independently figuring out how a bow works.

What happened next, though, diverged sharply by region. In the south (spanning from northern Mexico through the Great Basin and Southwest to California), the atlatl disappears from the archaeological record almost immediately after the first bows show up. The logistic regression curve for this region is practically a vertical line: one technology in, the other out. The authors describe it as a case of disruptive innovation, where the new tool's advantages in accuracy, range, rate of fire, and versatility were so obvious that people abandoned the older system wholesale.

In the north (the Yukon, Northwest Territories, British Columbia, and Alaska), the curve is far more gradual. Atlatl darts continue to appear in the record alongside arrows for more than a millennium after the bow's introduction.

Why Keep the Old Weapon?

The northern pattern fits a well-documented trend in how hunter-gatherer societies at higher latitudes manage ecological risk. According to multiple comparative studies cited in the paper, foragers in environments with extreme seasonality, unpredictable prey, and short growing seasons tend to maintain wider, more diverse toolkits. A failed hunt in a subarctic winter carries higher stakes than one in an arid but more temperate landscape, and carrying a backup weapon system is a form of insurance. The atlatl may have retained advantages in certain conditions (hunting in deep cold, targeting specific game) even as the bow proved superior in others.

In the milder, more predictable environments of the south, that kind of redundancy carried fewer benefits. Societies there could afford to specialise, converging on the single most effective solution. The bow won, and the atlatl was gone.

Correcting Earlier Timelines

The findings push back against several earlier proposals for the timing of bow technology in North America. Some researchers had placed the bow's arrival in the far north as early as 12,000 years ago, or suggested late Pleistocene bow use in the Southwest. The radiocarbon evidence from preserved organic weapons does not support those claims, according to Buchanan and his co-authors. If bows were used that early, those episodes appear to have been independent, prior occurrences with no direct connection to the wave of adoption that swept the west around 1,400 years ago.

The study also challenges John Blitz's influential model of a slow north-to-south diffusion gradient, which proposed that the bow arrived in the far north around 5,000 years ago and took millennia to reach the Southwest. The new data suggest the opposite: the transition was late and nearly simultaneous across a vast area.

From the Ground to the Collection

Atlatl hooks and dart foreshafts carved from bone or antler survive more commonly than their wooden counterparts, and we see examples from both North and South American contexts pass through our cataloguing room from time to time. Prehistoric projectile points, whether intended for dart or arrow, remain among the most widely collected categories of ancient artefact, and smaller stone points from the late prehistoric period (the era when the bow was in active use across the west) are still accessible to collectors at a range of price levels. Make sure to check out our future auction catalogues for stone tools and weapons.