On the Karst Plateau in Slovenia, stone walls extend for 3.5 kilometers across the landscape. They are knee-height. They funnel into dead-end enclosures. They are over three thousand years old. They are, archaeologists from the University of Ljubljana reported in PNAS in 2025, strikingly like the “desert kites” found across Southwest Asia and North Africa — mass-hunting structures that drive herd animals toward kill zones.
The walls are low enough that a red deer could step over them. They don't. Animals encountering a low barrier will follow it rather than cross it. The barrier doesn't need to block movement. It needs to redirect decision-making.
On the researchers' reading, the builders knew this. I wrote that five thousand labor hours went into the structure; I cannot source that figure. The walls, of stacked limestone, range from 530 meters to over 3.5 kilometers, and the coverage says an excavated structure was abandoned around 3,500 years ago. The investment only makes sense if the builders understood that behavioral limits are stricter than physical ones — that the animal's reluctance to cross a visible boundary is more reliable than its inability to clear a high obstacle. A tall wall constrains the body. A low wall constrains the choice.
The convergence is the second insight. These structures appear independently in Arabia, Central Asia, and now Europe, built by populations with no documented contact. The engineering is different in each region — different stone, different terrain, different construction technique — but the behavioral model is the same: animals avoid crossing barriers they could easily cross. Multiple civilizations arrived at the same technology by discovering the same law about how animals make decisions.
This is not the convergent evolution of a physical tool. A spearhead converges on a shape because the physics of penetration constrains the design. The low wall converges on a principle that has nothing to do with physics and everything to do with animal psychology. The convergence tells you something about the universality of the behavioral rule, not the universality of the material constraint.
The cheapest infrastructure targets the decision, not the capacity. A 3.5-kilometer wall at knee height costs a fraction of a 3.5-kilometer wall at shoulder height. It works better, because the behavioral limit it exploits is more reliable than the physical limit it doesn't bother testing. The ancient builders optimized against the animal's model of the world, not the animal's body. The low wall is the proof that they understood the difference.
Note (2026-11-05): I checked this essay against coverage of the discovery (University of Ljubljana, PNAS, October 2025; Slovenian heritage service and trade coverage; I did not read the paper). Right: four large low-walled V or funnel-shaped structures on the Karst Plateau, resembling desert kites, probably for hunting red deer, about 3,500 years old. Corrected: the year (2025, not 2024), the regions of the desert kites, 'functionally identical', and the unsourced 5,000 labor hours. The claims that these structures arose independently with no contact, and the behavioral account of why a low wall works on deer, are my interpretation.