CHAPTER 10

The GRAMMAR of ROMAN DINING

Florence Dupont

The Cognitive Value of the Roman Diet

Like many other Mediterranean civilizations, ancient Greece in particular, Rome was a sacrificial culture: a domestic animal could be transformed into edible meat (i.e., slaughtered and butchered) only if ritually sacrificed. The Romans used the blood sacrifice as an opportunity to define their identity as civilized men and members of a community. They situated themselves with respect not only to the gods and the animal kingdom but also to other human beings. In addition, they confirmed the legitimacy of their ties to the soil. Eating meat in Rome was therefore intimately associated with the fundamental rite of Roman religion, the sacrifice.

To be sure, meat was not the only food eaten by Roman citizens, which is to say, by civilized men. Indeed, any people that subsisted on nothing but meat and animal by-products was by definition barbarian: the Germanic tribes, which primarily consumed milk, meat, and cheese, are a prime example.

Meat, a necessary but not sufficient part of the civilized man’s diet, was a central part of every festive Roman meal, no matter how simple. The master of the household might make a personal sacrifice, usually to the household gods, and then offer a portion of the sacrificial meat to his guests, or he might purchase his meat from a butcher shop that sold meat from public sacrifices. This meat was referred to as a caro, or share, and the banquet at which it was consumed was referred to as a cena, or place of sharing. The Roman banquet was thus, above all, a sharing of meat. In polytheist Rome, the sharing of bread had no symbolic value. More than that, if the host of a banquet had tried to serve nothing but products of the earth, his guests would have perceived his action as an insult.

Although the culture of Rome was a sacrificial culture, like that of the Greeks, it was not a banquet culture. The Romans never adopted the Greek symposion, in which the drinker, possessed by wine, welcomed into his body divinities such as Eros, Dionysus, and the Muses. In fact, Roman civilization had no tradition of religious possession, whether erotic, prophetic, or poetic, hence no place for the Dionysiac rituals and banquets of the Greeks. Indeed, the Romans did not believe in Dionysiac possession and looked on it as a form of charlatanism. Guests at Roman banquets were served meat and wine at the same time. The wine was simply a drink—a special kind of drink, perhaps, but not a sacred drug.

Rome was thus different from classical Greece in that it lacked any form of social ritual based on convivial drinking. Roman banquets were not unlike the banquets of archaic Greece: dais, the word for “banquet” in Homer, derives etymologically, like the Latin word cena, from the root for sharing sacrificed meat. One important difference, however, was that the Homeric banquet was an occasion for celebrating communal memories, during which a bard, possessed by the Muse, chanted an epic. There was nothing like this in Rome, where banquets were not ordinarily an occasion for the cultivation of language.

ROMAN FOOD SYMBOLISM The Romans took a great interest in all matters connected with eating, including the customs of others as well as themselves. Philosophers, historians, satirists, comic poets, orators, and encyclopedists obsessed with food tirelessly applied the well-known maxim: “Tell me what you eat and I will tell you who you are.” To this they added: “Tell me with whom you eat.” Transgressing the Roman dietary code classified a person as “other”—that is, outside civilized Roman norms. Among those so categorized were not only Germans and Numidians but also tyrants and would-be tyrants, professional philosophers, wealthy freedmen, and gladiators. Constant reminders of dietary norms were built into political institutions and regulative social practices. Censorial edicts and sumptuary laws regulated the luxury of banquets, and violators were threatened with social and political disgrace. People addicted to splendid dining or guilty of overindulgence became the butts of satiric poetry, graffiti, and gossip.

For a noble Roman, the concept of honor involved both personal frugality and generous hospitality. The princely ethic, by contrast, was one of gourmandism and self-indulgence. Vitellius was a monster who devoured the gods’ share of sacrifices and downed the previous day’s leftovers in the most wretched of taverns. Augustus betrayed his plebeian origins and his pettiness by his inability to participate in a banquet and his constant nibbling; when it came to food, he lacked generosity.

Like other peoples, Romans “ate symbols.” Indeed, symbolism was one of their paramount concerns. They therefore formulated countless ethical precepts concerned with eating (see table 1), just as other cultures formulated rules governing sex. Eating was one of the languages of “distinction” by means of which Romans situated themselves in time, space, and society. For them, food was “good to think with” (to borrow a phrase from Lévi-Strauss).

SYMBOLIC REASON AND UTILITARIAN REASON This symbolic aspect of Roman eating is one reason why the subject cannot be studied solely from an economic standpoint. If symbolic logic governed the Roman alimentary economy, the reverse was not true. An example will help to make this point clear.

There is general agreement that grain growing and indeed agriculture in general yielded little profit in Roman society, whereas livestock made people wealthy. From a purely economic standpoint, this makes no sense. Indeed, the growing urban population should have led to rising prices and a brisk and profitable trade in grain. This was not the case, however, because wheat was considered necessary for survival. Hence to make a profit selling grain was frowned upon: wheat was exempted from the laws of the market, unlike foods regarded as luxury items. Grain had to be sold at cost or even distributed free of charge (the annona was a free annual grain distribution to Roman citizens, paid for out of public funds). The city did not, however, impose on itself a duty to feed the indigent one by one. In fact, the citizen-farmer remained the Roman ideal: every Roman was supposed to own enough land to supply his family’s daily needs. Romans used the term paupertas, the root of our “poverty,” to refer to this idea of self-sufficiency without excess. Loss of self-sufficiency was sharply distinguished from poverty by the word inopia (misery), which was regarded as a moral failing resulting in social disgrace because it implied a failure to manage one’s patrimony in a proper manner. Misery made a person an outcast. Such an individual was seen as physically repellent (sordida). Even if not threatened with starvation, the miserable wretch ceased to be culturally a free man. In times of drought or war, when free grain was distributed to Roman peasants and unemployed city dwellers, inopia became a communal event, a pestilentia, or scourge, which lifted the burden of moral culpability from the shoulders of individuals. Citizens, it was felt, must not be allowed to endure the economic and social consequences of shortages due to a breakdown of the pax deorum.

By contrast, sacrificial meat and other banquet items, including wine, were free to vary in price. When Sulla ordered the price of meat reduced so that the people of Rome could celebrate a certain holiday, for example, he was accused of pandering to the plebs. Thus it is fair to say that two economic systems coexisted in Rome, each corresponding to a different symbolic function of food: the nourishment of free citizens and the celebration of religious holidays.

LANGUAGE AND UTTERANCE Historians have been able to trace the evolution of the Roman diet in sufficient detail to propose a periodization. The changes they have noted are essentially quantitative in nature, however. For nearly a millennium food symbolism in Rome remained stable. It must therefore be analyzed as an enduring structure: the symbolic significance of bread, meat, and vegetables did not change from Cato’s De agricultura to the Historia Augusta. No doubt this was because Roman food symbolism was inextricably intertwined with the blood sacrifice and agriculture, which were integral parts of Roman identity. Only after the major cultural revolution brought about by the advent of Christianity did food customs truly change.

This symbolism can be studied as a language in which each meal can be considered an utterance. Meals, whether solitary or communal, had both a physiological significance (the survival of the individual) and a symbolic significance (expression of affinity to Roman culture). The two were inseparable.

Typology of Roman Foods: Fruges and Pecudes

In describing a food system, one can focus on either production or consumption. The Romans did both, and in so doing they elaborated two homologous representations, classifying food according to where and how it was produced and where and how it was consumed (see table 1).

With respect to production, the basic distinction was between fruges, products of the cultivated soil, and pecudes, foods derived from animals raised for meat on uncultivated land (fallow fields and woodlands). In Roman parlance there were two broad categories of agricultural land (ager romanus). The first consisted of plowed fields (arua) together with vineyards and orchards (horti). These terms evoked a sedentary occupation of the soil (economic aspect) as well as the construction of cities and roads (political aspect). Agriculture—especially plowing, which transformed the soil from an uncultivated to a cultivated state—was thus an eminently civilized occupation in both the technical and religious senses. The second category consisted of uncultivated land in which wild animals (ferae) as well as domestic herds and flocks were free to roam; it included forests, mountains, rivers, lakes, and the sea (silua, saltus, amnes, lacus, mare). It is worth noting that the Romans applied the term pecudes to both domestic and wild animals. Roughly speaking, it meant “herd” or “flock” and was used in speaking of fish and birds as well as land animals. The Romans did not have artificial pastures, lands cultivated for use by animals. Their prata were not pastures but fallow or wasteland in which animals were allowed to graze. To the Romans, prata, covered with brambles and unappealing to the eye, were “mournful” lands.

TABLE 1. SYMBOLOGY OF THE ROMAN DIET

Oppositions within Roman Civilization
FOOD PRODUCTION
Fruges (products of the soil):	Pecudes (edible animals):
cultivated land	uncultivated land
cost-free	costly
farmer = civilized	shepherd, hunter = savage
hard	soft
cooked	rotten
inanimate	animate
digestion = blood, muscles, bone	corruption = excrement
FOOD CONSUMPTION
Prandium (snack):	Cena (banquet):
vegetables	meats
cold	hot
upright	reclining
outside	inside
war and politics	peace
effort	relaxation
restoration	pleasure
stomach (venter)	gullet (gula)
frugality	luxury
rusticity	civility
country	city
solitude	sociability
routine	holiday
greed	generosity
Oppositions between Rome and the Outside World
Rome:	Outside World:
Roman soldiers	barbarians
bread	meat and dairy products
settled	nomadic
farmers	shepherds
sacrificers	nonsacrificers
Intersection of the Two Kinds of Oppositions
Civilized:	Uncivilized:
citizens	shepherds
bread	wild plants
city	periphery

These untamed spaces were inhabited by shepherds and hunters. This type of exploitation did not, in Roman eyes, establish any relationship between man and the soil. Shepherds and hunters were nomads, not unlike the most savage barbarians. They were troublesome people, often slaves without wives or homes and all too likely to become bandits or mercenaries. They were neither citizens nor civilized. Yet these savages tended their herds not outside the territory of the city but on its fringes. They played an essential role in Roman civilization since they provided meat for sacrifices and banquets, thereby ensuring convivial relations both among Romans and between Romans and their gods.

FRUGES: FRUITS AND VEGETABLES Cultivated land was of two kinds: plowed land and gardens. Implicit in the legal and religious status of each type of land were other, symbolic differences.

Gardens (horti), including vineyards and orchards, were the more civilized type of cultivated land. Roman gardens were expected to produce vegetables all year round. Many different crops were grown using a variety of techniques for irrigation, fertilization, and cultivation.

Gardens supplied the Romans with holera, a category that included many green vegetables, colza, tubers, and edible bulbs but not beans, peas, and other pod plants, which were called legumina. The list of holera was a lengthy one: it included various kinds of cabbage, varieties of cardoon (Cynara cardunculus, a Mediterranean plant related to the artichoke and cultivated for its edible leaves and roots), salad greens, leeks, herbs, turnips, carrots, parsnips, garlic, and onions, to name a few.

Like other Mediterranean cultures, Rome lived on vegetables. Every Roman wanted a garden of his own, even in the city. The typical garden was inside the walls that surrounded the household, or domus, whence the term hortus, meaning “enclosure.” Like the household and crossroads gods, the garden gods were lares, a type of tutelary deity that symbolized permanent human occupation of a piece of land.

As far as products of the earth were concerned, garden vegetables, together with fruits and grapes, were thus the most civilized type of food and therefore edible raw or barely cooked (in salads or in the form of wine). Indeed, the term “raw” was a misnomer, since these foods were partially or totally “cooked” (cocta) by the sun. So when fruits and vegetables were harvested, they were, unlike meat, never in a raw state and subject to immediate spoilage.

FRUGES: CEREALS AND BEANS Cereals (frumentum, most notably wheat) and leguminous, or pod-bearing, plants (legumina) were quite another matter. These came from less civilized pieces of land, which each year had to be rescued by plowing from the ravages of winter. As such, they were considered to be less “cooked” by the sun than fruits and vegetables. Nevertheless, the Romans did not look upon harvested grains as “raw,” for they believed that such grains remained alive and capable of giving rise to new plants. Hard wheat was roasted whereas soft wheat was not; then the grains were milled or more commonly crushed to yield flour. This process nipped the incipient young plant in the bud, as it were, so that the resulting flour was considered to be “uncooked” and thus apt to decay, hence subject to the same taboos as dead flesh. That is why wheat was stored as grain, milled only when ready for use, and immediately baked into bread or boiled.

Legumina were like cereals in that once the seed was “killed,” it began to rot. Indeed, the Romans apparently believed that leguminous plants decayed even more rapidly than wheat. That is why the Jupiter’s priests (flamines) were forbidden to eat broad beans as well as flour and flesh.

Two things were economically significant about the fruits of the earth: they cost nothing other than manpower, and growing them did not add to a man’s wealth. And they also differed from animal herds in another respect. The land and its products were heterogeneous: one did not eat soil or amass capital in the form of grain.

PECUDES: A LUXURY OF THE WILD Edible animals were classified according to their natural habitat: air, land, or water. Land animals were of two kinds: domesticated herds, which were raised for eventual sacrifice, and wild animals, which were hunted. In Rome sacrifice and hunting were the two sources of meat.

The main victims of public sacrifices were cattle, sheep, and pigs. Sacrificial animals could be male or female, old or young, or even nurslings. Rituals were designed to provide the gods with appropriate victims. Domestic sacrifices generally used young animals: lambs, kids, piglets, and pullets. These supplied most of the meat for home banquets. Pigs were the most common sacrificial victims, so pork was a central feature of most banquet meals.

The ancients believed that every domestic species had a counterpart in the wild, and they hunted accordingly. The boar, or wild pig (porcus siluestris), was therefore a prime quarry, suitable for the most sumptuous banquets.

When it came to “heavenly flocks,” or birds, the same distinction was made between domestic and wild species. Chickens, geese, pigeons, and peacocks were raised on farms, while wild birds were trapped with birdlime.

Only the “schools of the sea,” or fish, were all wild. They were caught with lines or nets. Various shellfish were also eaten, most notably oysters.

The nice distinction between domestic and wild animals was spoiled, however, by the existence of an intermediate group of semiferae, or “half-wild animals.” Hunting and fishing were seasonal activities: there was no fishing in winter and no hunting in spring or summer. Wealthy Romans therefore kept and bred game animals on their farms. They might breed boar and raise fish in hatcheries. The resulting animals were semiferae. Yet there was nothing hybrid about the manner of their slaughter: these half-wild animals were hunted or caught and never sacrificed.

Whether taken from wild, half-wild, or domestic animals, meat was always a luxury item, although the saddle of a Lucanian boar was hardly a dish to compare with an old hen marinated in wine to soften up its flesh. To sacrifice an animal from one’s herd was to spend a portion of one’s capital to please the gods, whereas to kill an animal just to eat its meat was considered wasteful.

PECUDES: ON THE FRINGES OF THE CITY Though produced on the fringes of the city’s territory, the flesh of pecudes (especially fish and meat from adult animals) was usually sold in urban markets. Hence the Romans considered such flesh to be an urban product, or in any case a product “sold in town,” which made it even more of a luxury item.

The path that meat followed from source to market reveals certain interesting features of Roman culture. Some meat came from the wildest reaches of Roman territory to be sold at the heart of the most civilized place in the Roman world, the city of Rome itself, while other meat came from forests on the fringes of Roman estates to be sacrificed in the master’s household. This established a close relationship between two sharply contrasting parts of Roman territory or of the rural estate: the center and the periphery. The structure of the ancient city-state of Latium reflected this duality, as did the Roman estate, and it was duplicated on a larger scale in the imperium romanum, whose “periphery” included much of the known world. Gold, perfume, and exotic game (giraffe, bear, antelope, pheasant) came to Rome from India, Africa, and the forests of Germania. No doubt this explains why luxury was identified with the exotic in Rome and why the city was transformed into a vast world market, as we know from the third-century C.E. writer Aelius Aristides.

THE SYMBOLIC ECONOMY The territory of Latium and the Roman Empire thus consisted of a series of civilized and uncivilized regions. At the heart of the empire was the Urbs, Rome itself, the center of Roman political and religious life. The soil of Rome was entirely “cultivated” by urban development, but except for perhaps a few small gardens, it produced no food. Yet the city was filled with markets and shops selling luxury goods. Meat from cattle sacrificed on major occasions was sold at the forum boarium while other sacrificial meat was sold at the macellum, and fish and other seafood could be found at the forum piscarium.

Ringing central Rome was the Roman countryside, or rus, with its gardens, cultivated fields, and country houses (villae). This was the only fertile part of Roman territory. The work done here was honorable, worthy of a Roman citizen.

Beyond the rus lay the true periphery of uncultivated, hence sterile, land: forest, mountain, and swamp. Rome never developed a “naturalist” ideology or celebrated spontaneously fertile land. Wild plants (herbae) were not deemed fit for human consumption: leaves, seedlings, roots, and wild herbs were seen as fodder for animals. The only exceptions were medicinal plants such as nettles and a few luxury foods such as wild asparagus and mushrooms. Wild plants were substituted for vegetables and grains only in the diet of the miserably poor, and when this happened it was a scandal because human beings were reduced to the level of herbivorous beasts. Not surprisingly, shepherds were among those in this condition. Exploiting the wasteland on the periphery was a fit occupation only for slaves and the poorest peasants. Even hunting was a servile occupation.

FRUGES AND PECUDES: THE HARD AND THE SOFT, THE COOKED AND THE ROTTEN The dichotomy between products of the cultivated soil and edible animals was part of a general classification of animate and inanimate objects involving a distinction between the cooked and the rotten (see table 2). As we have seen, vegetables and grains were seen as partially or wholly cooked by the sun while still in the field. Hence there was little or no danger of corruption after the harvest. Cooking continued inside the human stomach and then the liver, which the Romans believed transformed digested food into blood, muscle, and bone.

By contrast, animals began to rot while still alive, for life was a process of gradual corruption that could be slowed or suspended by the presence of anima, the hot, dry, vital breath of life. With death, however, the anima departed. The body of the animal then decayed into a cold, foul-smelling liquid. To slow this process, one drained the freshly killed animal of blood and removed its viscera, which were mere concretions of blood, for the spilled blood (cruor) was the first part of the animal to be transformed into rot (sanies). This blood was offered up to the gods while still hot, along with the roasted viscera. The pecudes were then roasted and transformed into meats (carnes), but these did not undergo further cooking in the body (through digestion) and were therefore not transformed into muscle or bone. Instead, meat continued to rot in the human stomach, from which it was evacuated in a putrid form (stercus) whose religious status was the same as that of carrion. Unlike fruges, meat offered no nourishment to the civilized man.

TABLE 2. THE COOKED AND THE ROTTEN: CULTIVATED PLANTS AND EDIBLE ANIMALS

	Preparation	Transformation in the Body
Fruges (products of the soil)
planting = cooking	cooking optional	digestion by cooking
Liminal foods
broadbean = animalized vegetable	death = corruption
bacon = vegetized animal	boiling
Pecudes (edible animals)
life = slow corruption	death = rapid corruption	total corruption
	+
	cooking = slow corruption

This characteristic corruption of animate creatures was connected with their mobility: life (anima) was incapable of animating wood or stone, which were incorruptible, to be sure, but also inflexible. Romans called these “hard,” in contrast to animals, which were “soft.” These terms referred not so much to an external quality perceptible to the touch as to an internal solidity. Otherwise it would be hard to understand why a boiled turnip or lettuce would have been deemed harder than a boar joint. “Hard” meant unlikely to disintegrate whereas “soft” was applied to things likely to decompose into heterogeneous elements, like a sauce that goes bad. Culture hardened; savagery softened.

If we apply Claude Lévi-Strauss’s culinary triangle to Roman foods, we find that edible animals fall along a scale running from hard to soft, raw to rotten, bacon to oyster. The Romans classified the oyster as soft, moist, and cold. It was already so corrupt that many ate it roasted and peppered in order to prevent serious damage to the liver and stomach. Bacon, hardened by salting, drying, and smoking, was the only form of animal flesh that could be preserved and eaten boiled. The desiccation of the meat not only halted the corruption process but “cooked” it to produce incorruptible bacon or ham. Hence for the Romans the pig was not an animal but meat on four legs. Similarly, the products of the cultivated soil could be arranged on a scale from rawest to most cooked, from broad bean to grape. The best grapes were actually so full of life that the juice obtained from them continued to “cook” after being sealed in a barrel, thus yielding wine. The grape remained alive. And the wine itself, if not doctored or diluted with water, remained a living being against which the man who drank it had to struggle. By contrast, the broad bean resembled a dead animal.

We find the same thought structure in Roman medical texts on food. Celsus, for example, employs the same complementary categories, which transformed the spatial divisions of Roman territory into a dietary code: fruges build the body, carnes purge the body. Both undereating and overeating were believed to have deleterious consequences.

Roman Meals: The Location and Symbolism of Eating

Food production, then, was organized around a basic opposition between fruges and pecudes. This developed into a further series of oppositions related to consumption: there were two types of meals, the prandium and the cena; two purposes for eating, nourishment and pleasure; two parts of the body involved in the act of eating, the stomach and the “gullet”; two social forms of eating, isolation and conviviality; and two emblematic figures in satirical literature indicative of two harmful perversions, avarice and parasitism.

The Roman diet was a diet of extremes, ranging from everyday frugality for people of all social ranks to great orgies of wine and meat consumption on special occasions, when overindulgence was de rigueur. Banquets staged by the state, civic leaders, and provincial dignitaries and always associated with sacrifices served as pretexts for all sorts of prestige expenditures. Aristocrats spent lavishly, and at times ruinously, on public banquets—to the point of tossing whole sides of beef into the Tiber. The center of the civilized world was not only the world’s greatest marketplace but also the scene of vast festivals of consumption: the Roman people enjoyed the fruit of victory. Over time the lavishness of these celebrations increased, as world conquest enriched the Romans. Yet one should not make the mistake of imagining the citizens of the early Roman republic as Romans of a later period liked to imagine them—as bearded, half-naked rustics chewing vegetables in smoky huts. In the fifth and fourth centuries B.C.E., Rome was a city within the ambit of Etruscan-Greek culture, and it shared the refinements and techniques of that culture.

CENA AND PRANDIUM The cena brought together a group of men to celebrate a special occasion in a covered space (a house, portico, or garden sheltered by a velum). These men always reclined while eating (if women were present, it was traditional for them to sit upright). Usually the company consisted of a well-defined social group: a family, patron and clients, friends of roughly equal age, professional or sacerdotal colleagues, or neighbors. The number of guests in each dining room was limited to a dozen or so, but there could be many dining rooms. Sometimes the company consisted of no more than a farmer, his wife, his sons, their wives, his grandchildren, and perhaps a few farmhands, but still the cena was always a celebration, no matter how limited the fare. It was never a routine meal. A cena of substantial proportions might be called a conuiuium. If it was a ritual banquet, the word was epulum.

Romans generally ate just enough to satisfy their hunger, however, without ceremony, often alone and just about anywhere and anytime. But not just anything: they ate only “frugal,” fortifying dishes made from fruges. A meal of this sort was called a prandium.

Days were not organized around mealtimes. Romans ate whenever they felt hungry. Banquets marked breaks in the daily routine. Furthermore, the cena/prandium distinction coincided with certain other distinctions concerning time and space.

The cena, for example, belonged to what was called otium, meaning both leisure time and peacetime. Otium comprised inactivity of two kinds: the otiosus was not a soldier, and he was not active in civic life. A typical cena might take place on a winter afternoon in the dining room of a Roman house or country villa. The guests would arrive at around two or three o’clock, well scrubbed, relaxed, and clad in robes or tunics. The banquet would end at nightfall.

By contrast, the prandium was the meal of Romans engaged in war, politics, or any activity that required effort (labores). It might be a breakfast or lunch, but there were no fixed times. For bachelors and soldiers in the field, dinner was also a prandium; for families in mourning, it was the single daily meal.

Since the prandium was purely a response to individual need, it reflected each person’s character and moral qualities. Seneca was proud of taking a journey during which he ate nothing but dried figs, with or without bread. To endure a lack of food and live on virtually nothing was considered a mark of the grandeur of a man’s soul, but only if such extreme frugality in daily life was regularly compensated by a joyous banquet.

BANQUET MEATS: THE IMPERATIVES OF PLEASURE By definition, a cena was an occasion at which sacrificed meat was consumed—and not just sacrificed meat but other foods included in the term carnes (that is, foods not digested by the body, upon which they had a “softening” effect). Meat was called “rotten” yet considered delicious, soft, and tasty. Through sacrifice, men shared the pleasure of eating meat with the gods at the cena. In every respect, the cena was tailored to the pleasures of the guests: the food, the sumptuous service and table settings, the couches, and the beautiful slaves—all were designed to enhance the pleasure of the diners. The cena was to the prandium as pleasure was to effort, as superfluity was to necessity (see table 1).

The prandium nourished; the cena entertained. To be sure, banquet guests consumed some nourishing food, such as bread and vegetables. But the dinner consisted mainly of meat, which was “indigestible” and eaten only for pleasure. Even the “real” foods, those that truly nourished, were supplied in such quantity that they further added to the pleasure of the experience, as did the ingenuity of the preparation: wild asparagus and mushrooms gathered in the forest might be served along with vegetables grilled like meats. Even bread became a delicacy: the bread of Picenum, for example, was made of crushed wheat that was soaked for nine days, then kneaded with grape juice before being baked in pots.

For the enjoyment of the cena the Roman possessed a special organ whose sole function was to experience culinary pleasure: the gula, located between the throat and the beginning of the esophagus. Because pleasure was not experienced in the mouth, one could not spit out a piece of meat after tasting it. The delicious rottenness had to be ingested in order to be savored.

The cena menu consisted of three courses combining pleasure with nutrition. The first was the gustatio, which included two symbolic foods, eggs and olives, accompanied by bread and honeyed wine. These appetizers may well have been intended to appease the hunger of the guests, for the verb gustare meant “to have a bite to eat” and referred to the prandium. Only after this did the true eating for pleasure begin. A more sophisticated version of the gustatio might nevertheless include sample “meat” dishes, such as oysters and other shellfish, dormouse, and thrush. Next came the cena proper, which was organized around a sacrificial meat (caro). At the most sumptuous banquets this might be some delicacy from a pig, possibly a pregnant sow (the nipple and vulva were especially prized, along with dozens of other select parts). With this, game animals and birds were often served, as well as fish either minced or accompanied by sauce. The dinner ended with the secundae mensae (second course), for which fruit (pomum) was favored. Walnuts, dried figs, and grape preserves were common. These fruit dishes constituted the most “civilized” portion of the banquet as well as the sweetest. Again, shellfish, small birds, and other animal delicacies sometimes accompanied this final course, along with quantities of wine.

To organize a successful cena was a complicated matter, requiring considerable art. Too much pleasure provoked disgust; not enough provoked frustration. The master of the house had to adapt to the occasion and to the personality of his guests while seeking to preserve his own reputation by appearing neither stingy nor pretentious. Success called for an ability to make one’s guests feel closer to one another through the culinary pleasure they shared.

THE FARMER-SOLDIER: SYMBOLIC WHEAT Bread, olives, onions, wine, vegetables drenched in olive oil and cooked, salad, and figs—the prandium was supposed to be a cold, frugal, vegetarian meal for consumption by a single individual. A person who feasted alone on a sow’s nipple or tripe stew, whether in his own home or at an inn, could only be a social outcast: a servile gravedigger, an insolent freedman, or an impudent noble (Lucullus never dined at home).

Such a meal, offering just the bare minimum necessary for survival, was the nourishment of active citizens, of people engaged in projects that required effort. Among such people, one can make a further distinction between the soldier and, say, the farmer. In civilian life, a man who ate alone, even a worker, normally ate vegetables. He might also eat bread, but bread was not an essential part of the peasant prandium as normally represented. Curius Dentatus, who was consul in 290 B.C.E. as well as a hero of the Samnite wars and a model of ancient frugality, contented himself with barely cooked colza when Campanian ambassadors surprised him one night with a visit. The historical accuracy of this anecdote is unimportant. What matters is that a Roman noble could eat such things without descending to the level of the herbivores. In peacetime, garden vegetables—civilized foods—were the only meal that a rural Roman required.

In wartime, however, this was not the case. Soldiers were bread eaters. Even if they took their bread with olives, onions, figs, and oil, the bread was essential, and soldiers who were served nothing but legumina—or, even worse, meat—protested. If they were fed colza, they turned it into herb bread that drove them wild. The phrase “herb bread,” used by Caesar’s soldiers confined in Dyrrachium, spoke volumes. Why didn’t they cook their colza like Dentatus? Because for soldiers the only civilized food was bread. The soldier’s prandium centered on wheat, whereas the farmer’s centered on vegetables.

Bread was thus the symbolic food of the citizen-soldier. If a Roman peasant never went to war, did he need to grow wheat? In the very early days of Rome, the mobilized citizen provided not only his own equipment but also his own provisions for the duration of the campaign, or at any rate until he was in a position to take his subsistence from the enemy (a soldier consumed 1.7 to 2.2 pounds of wheat per day). Wheat was easily stored and transported in the form of grain. In Mediterranean countries wheat, along with dried fruit, was the food of travelers in remote regions. Once transformed into bread, moreover, wheat was the “hardest,” most compact of cooked foods. Over time it hardened without rotting. It made the soldier’s body, already heavily armored, hard and compact like itself. Agility was not a Roman martial virtue. The legionnaire was expected to display stamina, resistance to hardship, and imperviousness to fear. He was expected to stand firm in the face of enemy blows. When the army needed mobile soldiers, it relied on Rome’s barbarian allies.

Because a Roman peasant did not become a citizen until he was inscribed on the rolls of the army, bread symbolized citizenship despite the fact that Romans actually consumed little wheat in peacetime. Wheat was eaten when other foods were scarce, but otherwise Romans preferred vegetables. The garden was the “poor man’s farm.” Only people who had achieved a certain wealth, who belonged to a higher class in the censitary (or propertied) hierarchy, raised wheat. To do so was a mark of ease, indispensable to the citizen. Although it is true that the state, in practice, soon took it upon itself to supply soldiers with food as well as arms, wheat was still the crop that allowed a Roman to leave his garden and be more than just a peasant (rusticus).

Wheat was symbolically significant for another reason as well: it served as a buffer against famine and scarcity, allowing peasants to survive without eating wild plants. It was an honorable substitute for vegetables. Grain was distributed whenever the poorest citizens would otherwise have been forced to “graze” for acorns and nettles or eat animal fodder such as vetch. Plebeians demanded such distributions as a right of free men—the right to a civilized food, a soldier’s nourishment. In Rome famine was not so much a threat to the lives of the poorest citizens as to their civic liberty.

“Bread, cabbage, and sacrifice”—such was the Roman diet, reduced to a motto that sums up three spatial and temporal aspects of Roman civic life: war, countryside, holidays.