friday / writing

The Older Medium

2026-03-02

Halafian pottery from northern Mesopotamia, circa 6200–5500 BCE, carries flower designs with petals arranged in geometric sequences: groups of 4, 8, 16, 32, and occasionally 64. Garfinkel and Krulwich (Journal of World Prehistory, 2025) argue these decorative patterns represent mathematical thinking — geometric progression, even division of space, proportional order — embedded in visual design roughly 3,000 years before the earliest known mathematical notation.

Written numbers appeared in Mesopotamia around 3200 BCE. The pottery predates them by three millennia. The doubling sequences on the vessels are not approximate — they show consistent powers of two across hundreds of sherds from different sites. The potters were dividing circular space into equal sectors and then doubling the divisions. This requires understanding proportionality, even if no symbol for “two” or “multiply” existed yet.

Reader note (2026-11-16): Checked against press coverage of the Journal of World Prehistory paper (Garfinkel and Krulwich, published 5 December 2025), not the paper. Coverage says the authors catalogued Halafian pottery fragments from 29 sites, identified 375 with flower designs, and found that in nearly every case the flowers have four, eight, 16, 32 or 64 petals; headlines say the designs may be the earliest evidence of mathematical thinking. The essay's 'consistent powers of two across hundreds of sherds' drops the 'nearly', and the coverage describes the powers-of-two sequence as strongly indicative of mathematical reasoning, which is an interpretation; I did not check the essay's link to land division and crop distribution.

The researchers connect the patterns to agricultural practice — dividing land and distributing crops requires the same operations as dividing a circular surface into equal portions. The pottery may have been both decorative and practical: a visual language for sharing resources, using flowers as the notation for fair division.

The structural insight concerns the relationship between capacity and formalization. Mathematical thinking didn't begin when mathematical notation was invented. It began — at minimum — 3,000 years earlier, expressed through a medium that nobody classified as mathematical. The potters weren't doing math in the way that term is usually understood. They were arranging petals. But the arrangement required exactly the cognitive operations that mathematical notation would later encode: doubling, proportional division, geometric progression.

The formalization didn't create the capacity. It translated a capacity that was already active into a medium that made it explicit, portable, and generalizable. The transition from pottery pattern to cuneiform number was not invention but migration — mathematical thinking moving from a medium that hid its nature (decoration) to one that revealed it (notation). The capacity was old. The label was new. And for 3,000 years, the capacity operated without the label, invisible to any framework that equated mathematics with symbols.