Baking can be hard work: it takes time, elbow grease, and patience. But the effort lockdown-bakers expend in kneading is nothing in comparison to the processing that turns grain into flour. If you wanted to make flour today, the internet advises that you use a coffee grinder; if you wanted to do it at scale in antiquity, you needed a mill. Working a mill, however, was one of the most feared and laborious tasks in the ancient world. It was a punishment for rebellious enslaved workers and a mere step up from the death-sentence of working in the mines. New archeological evidence shows that mill workers and miners may have even worked together to select materials and ease their burden.
To anyone who voluntarily became a baker during the pandemic and even ground their own grain, all of this might seem quite strange. Small hand-mills—which were used by domestic bakers in antiquity as well—aren’t too cumbersome and they are a step up from a mortar and pestle. But the larger hourglass and rotary mills used in Roman bakeries involved physical strength and were often powered by donkeys and “chained convicts” (as Pliny noted in his Naturalis Historia). The hourglass mills, which can still be seen in the bakeries at Pompeii, were comprised of a solid bell-shaped stone over which a hollow hourglass shaped stone was placed. The hourglass stone both functioned like a hopper that fed grain into the mill and was also turned around the base stone to grind the grain. The worker (human or donkey) would circle the mill turning the hourglass stone as he walked.
Working in a mill was so tough that it was a form of punishment. Messenio, a character in the playwright Plautus’ Menaechmi, lists being sent to the mill alongside being whipped or placed in fetters as a punishment for laziness. In the mill, Messenio says, the disobedient enslaved worker would find himself hungry, exhausted, and cold. As Professor Sarah Bond writes in her book Trade and Taboo, “laboring in the mill was better than being sent to the mines, [but] it was still a terrible punishment.” Mill prisons, she writes, remained a reality in the Roman empire though, and from the fourth century A.D., they were sometimes staffed by a mix of enslaved and penal workers. With much of the Roman empire being sustained by a grain-based diet, there was always a demand for workers.
Even enslaved children learned to fear the mill. A beautiful piece of graffiti etched into the wall of a school for enslaved children in second-century Rome shows a donkey turning a mill. The child who wrote the inscription—in lettering worthy of a high-quality manuscript—expresses the hope that having worked hard in school, he will, as an educated scribe or copyist, be able to escape the threat of the mill.
It’s illuminating, therefore, to look under the bonnet, as it were, of one of antiquity’s most arduous occupations. A joint team of geologists and archaeologists from the University of Texas at Austin, the University of Tokyo, and Old Dominion University worked on a study of stone implements discovered at the Roman site of Volubilis in modern-day Morocco. The results were recently published in the Journal of Archaeological Science. Previous generations of scholars had hypothesized that millstones had been imported as part of long-distance trade agreements, but the geological analysis of the various millstones found that the rock used to make them had been quarried locally.
Even more interesting is the study’s revelation that particular kinds of rock were used to make particular implements: “grain millstones,” the authors write, “are made of alkali basalts; olive mills are made from high-energy clastic and fossiliferous…limestone; and dough mixers are made from a lower energy limestone.” The stone used for the dough mixers was more impermeable, which made it “ideal for dough mixers” because dough contained water. Not only were there preferred materials for each task, but somebody also knew what they were doing.
The authors conclude that there must have been communication and cooperation between the craftsmen of the city—quarriers and bakers—to determine the preferred materials for each cumbersome task. Of course, the selection of the best material feels instinctive. As Columbia classicist Joseph Howley put it to me, “aren’t tools always made out of the best implement or material for whatever purposes? Nobody is arming Roman soldiers with bronze if there’s iron to be had.” And he’s right: this kind of technological evolution is in some ways common sense.
At the same time, while we might have assumed that stonecutters offered vats made of various stones and bakers or olive workers quickly determined which material worked best, this discovery sheds light both on who had specialized knowledge in antiquity and whom technological advances benefitted. The knowledge about which stone worked best for which kind of work isn’t found in the manuals put together by elite writers. It’s revealing, Howley said, that we think that it should be.
What the study shows is that expertise is not just found among educated elites. Non-elite craftsmen, including the enslaved and formerly enslaved, possessed mathematical, engineering, and geological knowledge that was put to use to improve not just productivity and efficiency but also working conditions. The advanced technology developed to crush and press olives, for example, has been called “one of the most mechanically advanced in antiquity.” Roman olive presses employed “the principles of rotary motion, the lever, the wedge, and the screw to make the tasks easier and more efficient.” These technologies were developed by (or at least in conjunction with) non-elite artisanal workers. Moreover, they benefitted those who operated these devices: necessity, after all, is the mother of invention.
It’s intriguing and perhaps also ironic that, despite its instinctiveness, this discovery was made by the collaboration of brilliant scholars from the separate and often siloed disciplines of geology and archaeology. Perhaps knowledge-sharing and collaboration is something we have to learn from ancient bakers.