What Romans Actually Ate
Roman food is one of the most misrepresented topics in popular history. The standard image — wealthy Romans reclining at banquets, eating dormice and vomiting between courses to make room for more — is accurate for a narrow slice of Roman society at a specific moment in imperial history and almost entirely wrong for everyone else. Most Romans ate simply, cheaply, and without couches.
The staple of the Roman diet was grain. Bread and porridge — puls, a thick wheat or spelt mash — were the foundation of what the majority of the population ate every day. Grain was so central to Roman social stability that the state organized its supply directly: the annona, the grain dole, eventually provided free or subsidized grain to several hundred thousand residents of the city of Rome. This was not charity in the modern sense. It was political infrastructure. A city that could not feed its population was a city that would riot, and Rome had learned this lesson repeatedly.
Legumes were the protein of the poor. Lentils, chickpeas, and beans supplemented the grain base. Vegetables — cabbage, turnips, onions, leeks — were grown in garden plots and purchased from street vendors. Olive oil was the fat, used for cooking, lighting, and personal hygiene. Fish sauce — garum — was the universal condiment, present in virtually every Roman recipe that survives, from street food to banquet dishes, in quantities that suggest the Roman palate was oriented toward fermented umami in a way that modern Western cooking is not.
Meat was expensive and infrequent for ordinary Romans. Pork was the most common when it appeared; beef was rarely eaten because cattle were more valuable as draft animals than as food. Fish was more accessible in coastal cities and was preserved via salting and fermentation for inland distribution. The Roman street food economy was substantial: most urban Romans did not have kitchen facilities in their apartments, and a network of thermopolia — essentially ancient fast food counters with heated containers built into stone countertops — served hot food throughout the day. Pompeii’s archaeological record has identified over eighty thermopolia in a city of perhaps twenty thousand people, which gives some sense of how central this infrastructure was.
The banquet culture that dominates the popular imagination was real but socially confined. The convivium was a formal dinner party among the elite, and its excesses were real enough to generate considerable moralizing literature from Roman writers who found them offensive. Apicius, the first-century cookbook that survives as the primary source for Roman elite cuisine, describes dishes of considerable complexity and expense — flamingo, sea urchin, elaborate sauces involving dozens of ingredients. These dishes were not Roman food. They were the food of a particular class at a particular moment, and they were already considered excessive by many Romans who observed them.
The reclining posture at dinner was not about excess — it was a status marker. Reclining at table indicated that one was a free adult citizen of sufficient standing to eat in the Greek manner. Children, slaves, and women of lower status typically sat. The physical arrangement of a Roman dining room — the triclinium, with three couches arranged around a central table — was designed to display social hierarchy as clearly as it facilitated eating.
The vomiting story is mostly false, or at least grossly exaggerated. The vomitorium, contrary to its name, was an architectural term for the exit passages in an amphitheater — the routes through which crowds could rapidly exit, not a room for purging. Some Roman sources do mention induced vomiting at banquets, but as a medical practice rather than a systematic method of extending eating capacity. The image of Romans routinely purging between courses is a later distortion that says more about how subsequent cultures imagined Roman decadence than about Roman practice.
What Roman food actually reveals is a society stratified enough that the gap between elite and ordinary diet was vast, integrated enough that certain ingredients — garum, olive oil, bread — crossed all class lines, and urban enough to have developed a sophisticated street food economy well before the modern world thought it was inventing one. The Romans ate well or badly depending entirely on where they sat in the social structure. The dormice were real. They were also beside the point.