There has been a proliferation of studies on the Greco-Roman background of Jewish and early Christian festive meals since Gordon J. Bahr wrote his influential article, “The Seder of Passover and the Eucharistic Words.”1 Many aspects of contemporary Greco-Roman dining practices, however, remain frustratingly obscure due to the nature of our sources, which are not only fragmentary and elitist but cover a very wide geographical and temporal range. Much of our most detailed data is further complicated by its location in satirical sketches in which exaggeration plays an essential role.2 We must therefore struggle against a desire to neaten the rough edges of our knowledge and fabricate a picture more complete than accurate. Salient differences separate not only the social classes but also those living in the eastern from those in the western part of the Roman Empire. The considerable time span covered by our disparate sources must also be borne in mind. True, the deeply conservative nature of meal habits3 sometimes justifies the use of data from earlier or later centuries to fill in lacunae in the intervening period—in our case, the first century B.C.E. and C.E.—but as Bahr himself remarked, a casual disregard for date remains a deficiency in many works on this subject. Our focus on the main meal of the day4 also bids us be alert to the ways in which family dinners overlap with communal banquets and secular meals merge with sacred ones. For as Ramsay MacMullen has reminded us, in this period there existed no formal social life that was entirely secular, except among the very wealthy.5
Despite such necessary cautions, however, there is much we can know about Greco-Roman meal customs. Let us then ready the room, set the table, and watch who comes and how they behave.
Festive meals usually took place late in the day, three to four o’clock being perhaps the most usual time.6 Typical dining rooms were furnished with an uneven number of stone or wooden couches made comfortable by cushions and draperies. Our literary sources speak most often of three couches, which in turn give the dining room its technical name of the triclinium, or “three-couch-room.” Dining rooms, however, could in fact be of any size. The layout of eastern dining rooms shows a preference for rooms of seven or eleven couches, each of which was equipped with its own small table.7 These rooms are therefore easily recognizable by their slightly off-center door, designed to accommodate the length of a couch on one side and the foot of a couch on the other. In the Roman world, the couches tended to be arranged around three sides of a central table; the one side left open facilitated table service as well as an unobstructed view of the after-supper entertainment. Each of these couches could hold up to three diners in comfort. It was one of Cicero’s accusations against Piso that at his dinner parties he packed in his guests five to a couch, but sprawled in solitary splendor on his own couch.8 Depending upon the configuration of individual dining rooms and the number of guests invited, these patterns varied; for example, when the guests were many, couches tended to be clustered into sub-groupings to encourage conversation.9
At festive meals, men reclined to eat.10 Lying parallel to the table upon which the food rested, the diner propped himself on his left elbow, leaving his right hand free to convey food and drink.11 One end of the couch was raised for support, but if there was more than one diner to a couch, the others had to make do with cushions. At ordinary meals, however, Athenaeus (fl. c. 200 C.E.) suggests that men customarily ate in a seated position—as they would in taverns or other informal circumstances.12 The great majority of people always ate in a seated position; reclining was a sign of elite dining.13 At certain ritual or political meals, therefore, men ate seated on simple chairs in order to demonstrate their proletarian sympathies.14 If women, children, and other inferior persons were present at festive meals, they would sit on the end of the men’s couches, or on chairs or low benches.15 In the Greek East, Plutarch (46–120 C.E.) alone mentions women lying down to eat (and that only once),16 but in Imperial Rome, high-status women customarily reclined with the men. It was a habit that shocked the Greek world.17
These differences in posture reflect a fundamental tension within festive meals. While celebrating friendliness and commonality, they also served as public proclamations of status.18 Not only was reclining itself a sign of social superiority,19 but the couches themselves, especially in the West, were carefully ranked so that one could tell at a glance who was the most honored guest—and who the least.20 The most honorable couch was the first to the right upon entering the door, and the host was easy to spot on the last couch to the left of the doorway; if he was joined by others, he was the one in the best position on that lowest couch. By tradition cherishing the ideal of equality, the Greeks resisted such sharp gradations at table. In Plato’s Symposium (early fourth century B.C.E.), guests sit according to friendliness rather than honor, and Plutarch, at the end of the first century of our era, raises the question of whether a host should seat his guests at all, or simply allow them to seat themselves.21 The fact that these sources discuss this issue suggests, of course, that precedence in seating was routinely observed. Dinner parties in the East, however, tended to be more homogeneous. Perhaps precedence was less at issue precisely because the gatherings were of the same sex, family, class, city, or age groups.22 There was, nevertheless, always the question of where to sit, and upon examination we discover that their concept of equality in these matters was less absolute than our own. Philo (30 B.C.E.–45 C.E.), for example, praises the “equality” presiding over the meals of the Egyptian Therapeutae, an ascetic Jewish group in the first century of our era. As evidence, he points to the presence of women reclining at table and the absence of slaves. To our eyes, however, such equality is compromised, if not altogether abrogated, by the separation of the women’s couches from the men’s, and the rigorous ranking of both sets of couches according to seniority.23
Provisions for dining outside, in gardens or vineyards, was a common feature in luxurious private houses.24 These couches might be permanent fixtures of stone, or made of wood and moveable. Such garden triclinia of various sizes were also available for hire in inns and taverns. Similar public facilities were also commonly found in temple complexes around the Mediterranean basin.25 Plutarch makes reference to the custom of holding private parties at a shrine, as does Paul in his letter to the Corinthian Christians (c. 54 C.E.).26 Archaeological excavations at Corinth have revealed three public dining rooms dating to the Hellenistic period (c. 323–31 B.C.E.). Cheery with red painted walls, they were furnished with eleven couches which were repainted many times in red and yellow, testifying to their regular use throughout the Roman period (c. 31 B.C.E.–476 C.E.).27 Aelius Aristides, an educated man of the second century of our era, spent long sojourns at many of the great healing shrines. From him we know the meaning implicit in this architectural feature: at meals held in temple dining rooms, the god was considered to be the party’s true host.28 Dinners honoring a particular god, however, need not be held at that god’s temple. For example, one papyrus invitation informs its recipients that “Dionysius asks you to dine on the twenty-first at the couch of Helios, Great Sarapis, at the ninth hour, in the house of his father.” Such flexibility of locale, as Ramsay MacMullen suggests, probably helped to ease the crush of would-be diners on prominent feast days.29
By the high empire (second to third centuries C.E.), couches of stone had become rare. We must surmise the presence and layout of these dining rooms from mosaic flooring and the cuttings made in it to secure the couches made of wood or other, often precious, materials that have not survived. Mosaics can also suggest the arrangement of couches; many feature scenes oriented in several directions, presumably so that the diners, no matter what the angle of their seat, could all contemplate something pleasing.30 Two mosaic patterns dominate. One is a simple U pattern, in which a central area is flanked on three sides by a continuous mosaic pattern. The other combines this arrangement with an inset T pattern, wherein the central area is extended over either side of the open legs of the U, providing more ample space for entertainment and table service. Excavations of houses in Pompeii, Africa, and Antioch attest the popularity of this U + T pattern across the Roman Empire.31
In our period, the trend was towards largeness. Once the ideal party was no longer limited to seven, nine, or eleven guests, the size and fittings of dining rooms were set free to advertise the economic resources of the host. In Roman North Africa, excavations have revealed a huge dining room of 13.2 meters long and 10.3 meters wide.32 Such a space could accommodate three to four times the number of guests recommended by earlier writers. With this increase in size, the method of serving also shifted. Even in the Roman world, food and drink were now set before the diners on individual tables or on low ledges running along the front of the couches. Recently, Pauline Donceel-Voûte has advanced the hypothesis that the room filled with benches and small tables, excavated at Qumran and previously thought to be a library or scriptorium, may in fact have been a large communal dining room similar to the one discovered in North Africa.33
In addition to the typical triclinia layout, literary and figurative sources confirm a different arrangement, namely, the use of a large semi-circular couch, or stibadium, set before a round or D-shaped table.34 Divided into wedge-shaped segments, these couches could comfortably hold five to nine guests, with the usual number being seven. Rooms built for stibadium dining included an apse into which the couch would fit while still leaving plenty of room for table service and entertainment. Given this distinctive shape, many of these dining rooms have been mistakenly identified as Christian worship spaces; once the perishable couch has disappeared, the remains resemble a small stone altar dividing an open rectangular space, often with attractive mosaic flooring, from an apsidal recess or chamber.35 This was, however, a common arrangement for secular dining rooms. Martial twice refers to it, as does the younger Pliny (both writing at the end of the first century C.E.).36 While a stibadium couch could not hold as many guests as the large triclinia looked at above, wealthy patrons who wanted to host extensive parties of this design could construct dining rooms with more than one apse; three were not uncommon (the tri-conch). The largest excavated hall of this design has seven apses and could probably accommodate up to fifty guests.37
The stibadium couch may have originated from arrangements for informal outdoor dining. Early figural representations of picnic meals show cushions arranged in this same semicircular pattern.38 The attraction of the stibadia design may have lain in precisely this connotation of a relaxed setting in which precedence in seating would not be observed. The semi-circular couch, however, soon developed its own hierarchy of seating. Initially, the middle position was the most honorable, but by the late empire (fourth-fifth century C.E.) the guest of honor could be seen reclining in the right-hand corner facing the host.39
The typical Greco-Roman “food event,” to borrow Mary Douglas’s terminology, had three parts: the hors d’oeuvres, the main course, and the dessert.40 Written-out menus were known in ancient Greece and Rome, but it was more common for the host or his appointed slave to introduce the different dishes, announcing their ingredients, mode of preparation, and any unusual attractions.41 Athenaeus gives us the most detailed information on the types of food that might be served—so detailed, in fact, that only a selection can be given here. Hors d’oeuvres were usually served at table.42 They might include olives, eggs, a variety of salad-stuffs like celery, herbs, and lettuce, as well as small song-birds.43 When Lucian, writing in the middle of the second century C.E, complains on behalf of a guest that he “alone did not get an egg,” he is lamenting being passed over even in the serving of the hors d’oeuvres.44 The main course was typically presented as a series of dishes. A common assortment was three entrées, or casseroles, and two roasts—of fish, poultry, or meat—accompanied by a variety of “relishes” such as sauces or vegetable dishes. Dessert marked a break in the meal. It was called “the second table,” because at that point the old table would be removed and a new table brought in, or, minimally, the old one wiped clean. Unlike the elaborate main dishes and appetizers, however, dessert featured relatively simple fare. We do hear of round cheese cakes, sweet biscuits, and legume mixtures, but more often of fruit such as dates, figs, apples, and especially nuts. For this reason, dessert in the East often went under the common name of “munchies” (tragemata).45 A “Complete Dinner” therefore, according to Lucian, contains: a bird, boar’s meat, hare, fried fish, and sesame and honey cakes.46 Looking at these menus, we are impressed by what strikes us as a confusion of sweet and savory tastes. But our division of these categories is peculiar to modern (and western) sensibilities.
Depending on the resources of the host, platters and bowls might be made of precious metals or figured glass.47 Toward the end of the first century of our era, a new custom was introduced in which diners ate from their own private bowls rather than a communal dish.48 Guests were also provided with perfume and wreaths for their heads.49 In high society, handsome vessels made of bronze, gold, or alabaster were thoughtfully provided for guests who needed to vomit or urinate; in attending to these needs, however, it was polite to withdraw slightly from the table.50
Romans might come to supper already equipped with utensils; if not, their host would provide them with knives, an assortment of spoons—possibly including one with a prong or two on the handle for winkling out shellfish—and toothpicks.51 In the East, scooped-out bread might take the place of a spoon. Since utensils were intended only to augment the use of fingers, guests always washed their hands before and after eating.52 Sometimes simple rinsing could evolve into elaborate purification rituals, as with the Essenes.53 Bahr maintains that a blessing might be said before the meal by one of the assembled guests, although he grants that references to this custom are rare;54 Athenaeus cites Homeric precedent approvingly that a portion of food should be offered to the gods.55 But we are much better informed about the prayers preceding drinking (on which more below); behind this custom lay the conviction that the first portion of the central foodstuff of each segment of the meal belonged to the gods. Such offerings were not only pious but civilized. Semonides therefore described the essence of uncivilized behavior as “gobbling up portions reserved for the gods before they are offered.”56 Towels protected the cushions and draperies of the couches, and tablecloths were introduced in the time of the emperor Domitian (81–96 C.E.).57 Each diner, however, was expected to bring his own napkin, which was also useful for taking home leftovers.58 Lucian can satirize a social parvenu by pointing to the lack of stains on his napkin.59 Hands or knives could also be cleaned by wiping them on a chunk of bread, which could then be tossed to the dogs.60
After the meal was over, removable tables were taken away or tucked underneath the couches, diners washed their hands, and servants swept the floor and distributed perfume and wreaths: all in preparation for the drinking which was to follow.61 Women customarily left with the tables unless they contributed directly to the entertainment by being dancers, musicians, or sexually available. Greco-Roman banquets can therefore be broken down into two main components: the supper (deipnon), which was primarily the food event, and the drinking party (symposium) which followed.62 The emperor Tiberius (14–37 C.E.), however, somewhat mitigated the division between eating and drinking by introducing the custom of taking an aperitif of wine.63
There are numerous studies on the symposium as a cultural as well as literary form. Oswyn Murray has suggested, in fact, that the symposium, as an institution, forms the basic structure of Greek society.64 Clearly, it was far more than a drinking party. Before addressing the social mechanics of the event, therefore, the ideology of the symposium, as it emerges from the literature generated by this social custom, deserves attention. First, although these groups were not based upon family ties, participants used kinship language to describe their bondedness, often designating themselves as common descendents of a particular forebear.65 A second and related concept insists on the educational role of these gatherings—that they were as much about conversation, teaching, and group initiation as about wine, merriment, and entertainment.66 The presence of male children was therefore appropriate.67 A third element of literary symposia (in careful distinction from actual social practice) is the anticipated death of its main character.68 Not only is this true of Socrates in Xenophon’s Symposion, Ulpian in Athenaeus’s Deipnosophistae, the Lapiths in Lucian’s Symposium, Praetextatus in Macrobius’s Saturnalia, Jesus in the Gospel of John, and Thecla in Methodius’s Symposium,69 but even of the buffoonish Trimalchio in Petronius’s Satyricon, who has placed in his dining room a clock and a herald to announce the hours “so that he might continually be apprised of how much of his life had been lost.”70
Before the drinking could get underway, the wine had first to be strained from the cask into a large vessel and diluted by the addition of either warm or cold water. The exact proportion was determined by an appointed guest, the symposiarch; common mixtures, however, were five parts water to two parts wine, or three parts water to one part wine.71 Every bowl mixed at a drinking party was named in honor of a god; its correct preparation, therefore, included a libation and short prayer.72 Libations consisted of spilling a shallow saucer of wine; the gesture was usually communal, with each diner pouring wine onto the floor in front of his couch. Three was a customary—or prudent—number, for as Eubulus (fourth century B.C.E.) says, “For the right thinking, I mix only three communal bowls (kraters): one to health, which they drink first, the second to love and pleasure, and the third to sleep. When these have been drunk, people with a reputation for sense go home. The fourth is no longer ours, but belongs to bad judgment (hubris); the fifth to shouting; the sixth to licence; the seventh to black eyes; the eighth to legal action; the ninth to disgrace; the tenth to madness and collapse.”73 Plutarch’s more religious formulation suggests that three bowls be drunk: to the Olympian deities, to the heroes, and to Zeus the savior. In the year 30 B.C.E., it was decreed that a libation be offered to the emperor Augustus at all banquets, public and private.74 Petronius may be parodying this custom when his startled diners exclaim, “Blessed be Augustus, father of our country.”75 Aside from this oblique reference, however, we have few examples of actual prayers. Presumably, this is because they tended to be short and formulaic: a simple request for the gods to be present and favorable.76 Athenaeus repeats a slightly more elaborate prayer taken from one of the comic plays of Menander (late fourth century B.C.E.), as though it were still typical. Addressed to the Olympian gods, it runs, “For this [offering], grant us safety, health, and many blessings; and for all of us here, enjoyment in the good things before us.”77 Appropriate hymns, such as those addressed to Dionysus, god of wine, were sung.78 Toasts to the company might follow, in which a common cup was passed around the guests from left to right.79 In addition to this vessel shared in common, servants would fetch individual small cups for each symposiast, as the dinner guests were then technically named.80 These cups varied greatly in terms of shape, size, material, and workmanship: they might be made out of glass or precious metals, studded with gems, embossed with scenes from mythology or inscribed with verse, turned out of wood, or made out of special clays mixed with spices.81 They were stored in a particular cupboard in the dining room.82 An ancient Homeric hymn celebrating Hestia, the goddess of the home, as the one to whom banqueters pour out sweet wine “both first and last,” suggests that libations, hymns, and prayers were also customary at the end of the meal event; Cicero and Cornutus repeat this sentiment.83
With the impediment posed by eating removed, conversation at the symposium could be enjoyed to the full. This was the time for philosophical speculation or the recitation of poetry or mythical stories.84 Recitation might yield to drama: the minutes of the society of Iobacchi (c. 178 C.E.) describe the orderly assignation of roles. As many of the dramas recapitulated legends about the gods, we must abandon our tendency to distinguish too sharply between entertainment and liturgy.85 It was on this part of the festive meal that ancient authors lavished their attention, insisting that the purpose of eating together was friendship and mutual edification rather than any mere supplying of bodily needs. As Plutarch put it, “We invite each other not for the sake of eating and drinking, but for drinking together and eating together.”86 Our extant reports provide actual conversations as well as lists of appropriate topics for those in need of a little coaching. Some of these strike us as funny, for example, Plutarch’s suggested philosophical gambit: “Which came first, the chicken or the egg?”87 In more genial company, after-dinner entertainment might feature the telling of jokes or riddles or the playing of communal games.88 For the socially awkward, Plutarch spells out the delicate art of telling a suitable joke.89 In still looser gatherings, a variety of erotic experiences was available.90
At the end of Athenaeus’s party, incense was burned for ritual purification followed by more prayers, another libation, and a final song to the goddess of health. This was a traditional hymn, which, while written at the beginning of the fourth century B.C.E., was apparently still in use at the end of the second century of our era: “Health, most honored of the blessed ones, may I dwell with you for the rest of my life, and may you be gracious to me and live with me. For if there is any joy in wealth or children or kingly power that makes humans godlike, or in the passions we pursue with Aphrodite’s secret nets, or if any other divine delight or surcease of toil has been revealed to humankind, it is by you, blessed Health, that everything thrives and shines in converse with the graces. Apart from you, no one is blessed.”91 After a final washing of hands and (perhaps) exchange of fond embraces, the company would depart.92
Table service was performed preferably by male slaves.93 To use women servants in front of guests was distinctly déclassé.94 Slaves served a variety of functions, especially in large households. These may be divided, as John D’Arms suggests, into three main categories: supervisory (which would include gate-keeping and guest control); various food services; and the duties of the wine staff.95
While the host certainly had some say in the matter, the actual issuance of invitations, assignation of seating, and bestowal of food and gifts fell to the designated slave.96 This slave was to keep a sharp eye on the behavior of those present; if they were greedy or intemperate in either their consumption of food and drink or their talk at table, they might find themselves henceforth disinvited.97 Uninvited guests, however, often showed up in distressingly large numbers.98 Lucian also warns that looking too attentively at the master’s wife or sons could get a guest into trouble.99
Extending the snobbery of their masters, slaves reinforced the social hierarchy evident at table. Guests who were socially passé or otherwise demoted would find themselves given inferior portions.100 Lucian lists typical grounds for complaint: your bird, unlike your neighbor’s, is tiny and stringy; your wine sour and laden with dregs—that is, if you’re lucky enough to get any at all.101 Slaves might quickly move platters of food by unfavored guests, or indeed pass over them altogether, while lingering at the couches of others.102 This sort of discrimination sounds similar to the scene of common banqueting described by Paul (1 Cor. 11).
One slave may have been designated to oversee all the tasks of the dining-room staff.103 Early in the morning, he would see that the dining room was swept, the cushions laid out, and everything generally put to rights.104 When the guests arrived, he ensured that slaves were on hand to remove their shoes and wash their feet before they reclined.105 At his signal, slaves would carry in the meal’s several courses, which, as we have seen, were at least three, but rarely more than six. Cooks, bakers, and carvers who had mastered specialized tasks, occupied prestigious positions and were probably exempt from the multiple tasks expected of slaves with less training.106 At Trimalchio’s banquet, slaves pour out snow-cooled water, spread coverlets, sprinkle the room with saffron and vermilion sawdust, carry around bread, towels, hot water, and anoint the diners’ feet with ointment and wind garlands around their ankles, all the while singing, clapping, or dancing.107 While the lavishness of these materials and their presentation are probably exaggerated, the tasks themselves seem typical. Servants also fetched and trimmed the lamps.108 When not called upon to perform, they were to maintain strict silence.109
Leftovers belonged to the slaves.110 To them also fell the punishment for any breakage, spillage, lassitude, or pilfering.111This liability was, of course, only a particular instance of the general accessibility of the body of the slave to the owner’s whims, which included his unrestricted sexual appetites as well as more casual infringements of human dignity, as when Trimalchio used a slave’s head to wipe his wet hands dry.112
The wine servers were the prestige corps, appropriately accompanying the most important part of the meal, namely, the entertainment over drinking that followed the dinner proper.113 They were ideally expected to be young, long-haired, sexually attractive, and if possible, to have Greek names.114
Manners are essentially rituals handed down from the elders. Encoding powerful messages about order, continuity, and lastingness, they are about making the past present.115 Much to the regret of liturgists, therefore, the topic of ritual behavior at meals, i.e., etiquette, is discussed by our ancient sources only in passing.116 Precisely because of the impressive continuity and iron-clad enforcement of their social encoding, meal customs tend to be passed over as too “obvious” or “natural” to need comment. When comment is aroused, it is seldom favorable. As Erving Goffman notes, “Infractions make news.”117 Athenaeus, for example, tells us that it is bad manners to be seen relishing gristle, spines, and cartilage; he also condemns indelicate grabbing, slopping, splattering, or messing with one’s food.118 When diners start a wine and food fight in Lucian’s Symposium, there are several failed attempts at intervention before the lampstands are overturned and the party disbanded.119 Organizations like the Iobacchi appointed special officers to maintain decorum; these were termed orderlies (eukosmoi), and were perhaps the ancient equivalent of our maître d’s. If the interventions of these men proved ineffective, one could then call on the services of the “horses,” who functioned, apparently, rather like our bouncers: they would “pick up” and “put outside the front door” anyone acting in a disorderly way or responsible for creating a disturbance.120
Ancient satire provides us with a fuller, if inevitably exaggerated, outline of good banqueting behavior. The polite guest in the Greco-Roman world would bathe and put on fresh clothing before appearing at his host’s.121 In the West, sartorial preference dictated that the brightly colored synthesis, originally a female garment, be worn by both men and women going to parties.122 A guest was concerned that he not be the first to arrive at dinner, but to be last was also gauche: everyone tried to come in the middle.123 As no one wanted to be thought a newcomer to polite society, it was important to know that politeness dictated that a guest should compliment the host on his furnishings; but excessive gushing was bad form.124 Reclining in a slovenly manner indicated laxity of morals as well as muscle tone.125 Dishes were to be eaten in a specific order; if the diner didn’t know the sequence, he could sneak glances at his table-companions, but would suffer shame if they noticed.126 It was always rude to appear too enthralled in the business of ingestion: to drink too eagerly, to hunch over one’s food, or to spill sauce on one’s cloak. Loading up one’s slave with goodies for the road (especially if they were someone else’s leftovers) was distinctly déclassé, but even worse was to make off with the tableware.127
When the drinking began, a new guest might not know that it was customary to reply to a toast: not to do so was boorish.128 Plutarch and other social theorists are adamant that polite conversation is that in which all can, and should, participate. The goal of polite conversation was to minimize differences of status or other bases for invidious comparison. It was, therefore, rude for any one person to assume an authoritative, inflexible, or tiresomely pedantic tone.129 Whatever the conversation or the lateness of the hour, the polite guest stayed until the talk was over. Enthusiastic applause was the correct response to the entertainment, even if it was largely invisible from one’s couch. An untutored guest might suffer additional agonies, hesitating either to ask for something to drink, although parched with thirst, lest the slave think him a drunkard, or from having drunk too much and not knowing how to excuse himself.130 Worse yet, was to vomit all over the dining room or fall dead asleep on the couches.131
Our sources are clear that women in the West often accompanied their husbands to dinner parties. This was the custom not only among the elite, as the famous love triangles of Roman elegy so abundantly attest, but also among the nonelite, as a first-century graffito from Pompeii attests, echoing the concern that guests keep their eyes off other men’s wives.132
The presence of women at festive meals with men in the East, however, continues to be debated. This is not because of any unclarity in ideology, which insists on the absence of women. Anthropologists have helped us to see how place and time are divided along gender lines in traditional societies, among which of course belongs the eastern Mediterranean world of late antiquity.133 In such societies, public spaces and the middle of the day belong to men. Women, on the other hand, occupy the interior domestic space and venture into the “male” world of the public eye only in the margins of the day. Transgression of this division threatens both sexes with shame. This ideology of division extended also into the household. Greek houses had a special room, called simply the andron, or “men’s room.” It was appropriately located farthest from the female quarters and closest to the street; this was where dinner parties were held. Here, any women present were part of the entertainment rather than the company of those entertained. Given such evidence of sustained and consistent seclusion of women, why is there any lingering dubiety about their presence at dinner parties?
Doubt arises when we try to move from an assessment of ideological prescription to that of actual social practice. It is always an open question whether our literary sources are indulging in wishful thinking rather than in accurate description of social reality. We are invited to consider this possibility by other evidence attesting to women’s participation in roles that would, according to this dominant ideology, seem exclusively male. Women in our period, for example, served as patrons of clubs, as well as leaders of synagogues and churches.134 This model of segregation is also vulnerable to the more damaging objection of elitism. For while it is notoriously difficult to speak with any degree of certainty about the quality of life or freedom of movement of nonelite women in the empire, we gather that for the majority of women seclusion was simply not an option. Both the freeborn poor and the freedwomen of late antiquity had to work for a living, and therefore were accustomed to moving in the public spaces of the city in the middle of the day.135
Several of these objections are especially germane to dinners connected with religious celebrations, since these tended to blur the distinction between private and public. Wedding feasts, for example, were both family matters at which women took their rightful place and also communal, public events to which non-kin males would be invited.136 Greco-Roman dining clubs, moreover, which combined religious and social aims, catered not only to freedmen but also to women.137 The women who held office in these societies would, as a matter of course, be present at cult meals.138 Furthermore, had commensality between men and women been indeed so unusual in the Greco-Roman world, early Christians would surely have had to defend themselves far more pointedly on this score. Matthew, however, acknowledges without apology the presence of women (and children) eating bread and fish along with four or five thousand men (Matt. 14:21, 15:38). And Luke, the most consciously hellenized of the gospel writers, presents Mary as “sitting at the feet of Jesus” in the company of the other disciples, while Martha is busy serving (Luke 10:39–40). While this expression can be taken as a technical reference to her role as a disciple, in this meal context it might also indicate that she sat at the bottom of Jesus’ couch; eating and education were, as we have seen, not at all mutually exclusive. In the early second century, moreover, Pliny, a governor in Asia Minor, wrote to the Emperor Trajan describing the behavior of some Christians in his district. He knows that men and women are gathering to share a meal, but his only concern is that the food they eat be “ordinary.”139
I suggest, therefore, that our perception of festive meals in the East, as in the West, should be one in which women may indeed figure, as perhaps do children. The evidence, to be sure, is scanty but nevertheless insistent.140 Its sparsity probably reflects not the rarity of women at dinner parties but their social invisibility. They were there, but not in a way that was noteworthy to our elite authors.141 When women come to the attention of our nonelite authors, they mention their presence without excessive comment or particular apology.
Cultural anthropologists like Mary Douglas have taught us that patterns surrounding the shared consumption of food often express social relations. Greco-Roman meals were no exception. They too encoded social messages “about different degrees of hierarchy, inclusion and exclusion, boundaries and transactions across boundaries.”142 While Plutarch knew that the purpose of gathering together for meals was the promotion of group bonding, this goal did not preclude the other important task of commensality, namely, the articulation of social stratification and control.143 Posture, placement at table, as well as the quality and quantity of food and drink served, all indicated status.144 Apart from these central social functions, however, meal customs seem to have varied enormously depending on whether the dinner took place in the East or West, in high or low society, or in public or private space; the experience of festive banquets was also extremely variable depending upon personal placement and perspective—whether one attended as a man, a woman, a child, or a slave. Any attempt to reconstruct the precise shape of early Christian and Jewish communal meals must bear in mind this rich assortment both of continuities and of variables.145
1. Novum Testamentum 12 (1970): 181–202.
2. Petronius’s satirical sketch of the freedman Trimalchio has been compared to Horace’s remarks on Nasidienus (second half of first century B.C.E.) (Serm. 2.8). Its realism has attracted attention; the classic study remains that of Erich Auerbach, Mimesis: The Representation of Reality in Western Literature, trans. Willard R. Trask (Princeton, 1953), pp. 24–33.
3. Jack Goody, Cooking, Cuisine and Class: A Study in Comparative Sociology (Cambridge, 1982), pp. 37, 44–48.
4. I omit discussion of other daily meals and their customs. For a brief introduction to these and some bibliography, see Jérôme Carcopino, Daily Life in Ancient Rome (New Haven and London, 1940), pp. 263–65. For the Greek East, see Plutarch, Quaestionum Convivialium 8.6.725F–727A.
5. Ramsay MacMullen, Paganism in the Roman Empire (New Haven and London, 1981), p. 40. J-P. Jean-Pierre Vernant, “Religion grecque, religions antiques,” in idem, Religions, histoires, raisons (Paris, 1979), pp. 11, 24. Diagrammed by Dennis E. Smith and Hal E. Taussig, Many Tables: The Eucharist in the New Testament and Liturgy Today (London and Philadelphia, 1990), pp. 21–22. Pauline Schmitt-Patel, “Sacrificial Meal and Symposion: Two Models of Civic Institutions in the Archaic City?” in Oswyn Murray, ed., Sympotica: A Symposium on the Symposion (Oxford, 1990), p. 24; Murray, “Affair of the Mysteries,” in ibid., p. 157.
6. Philodemus (first century B.C.E.) invites Piso for dinner at the ninth hour (33 Gow-Page = Anth. Graec. 11.44). Pliny the Younger suggests dinner be taken at the end of the eighth hour in the winter and at the end of the ninth in summer (Ep. 3.1.8–9 [end of the first century-beginning of the second century C.E.]; cf. Martial, Epigrams 11.52, 10.48 [end of the first century-beginning of the second century C.E.]). Lucian complains that grand dinners were held at an unseasonably late hour: in the middle of the night (De Merc. Conduct. 26 [mid-second century C.E.]). According to Athenaeus, the usual time for dinner was when the shadow from the gnomon (dial) extended ten feet (Deipnosophistae. 1.8C, Aristophanes, Eccles. 652); early dinners began at seven (Deipn. 11.502B); one guest, invited when the sundial cast a twelve-foot shadow, came by mistake at dawn instead of in the evening (Deipn. 6.243A); children may have eaten earlier, ibid., 4.156B (c. 200 C.E.). A six-foot shadow is appointed as the time for bathing during the Saturnalia (Lucian, Saturnalia 17). David E. Aune, however, says that the Greek deipnon or Roman cena “began at the ninth or tenth hour of the day and could last three or more hours (well into the night).” (Aune, “Septem Sapientium Convivium (Moralia 146B–164D),” in Hans Dieter Betz, ed., Plutarch’s Ethical Writings and Early Christian Literature [Leiden, 1978], p. 71.)
7. Bergitta Bergquist, “Sympotic Space: A Functional Aspect of Greek Dining-Rooms,” in Murray, Sympotica, p. 37. Typically, these tables were three-legged; slots cut into the floor held them stable.
8. In Pisonem 67 (first century B.C.E.). Later (invidious distinctions?) repeated by Martial 1.20, 43; 2.43; 3.60, 82; 6.11; 9.2; Juvenal, Sat. 5; Pliny, Ep. 2.6; cf. Horace, Serm. 1.4.83 ff. Also the elder Pliny on wine distribution, N.H. 14.91.
9. For sizes and arrangement of these rooms, see Bergquist, “Sympotic Space,” in Murray, Sympotica, pp. 37–65.
10. If the prophet Amos refers to this custom (6:4–7), Jews would have practiced this as early as the eighth century B.C.E. Based on funerary reliefs, the Greeks, adopting the custom from the Assyrians, began to recline at table sometime in the seventh century B.C.E. In this custom, the Romans followed the Greeks as in many other areas of table etiquette (Jean-Marie Dentzer, Le motif du banquet couché dans le proche-orient et le monde grec du Ville au IVe siècle avant J.-C. (Rome, 1982), esp. pp. 444–52. Ovid records that sitting used to be customary (Fasti 6.301 ff).
11. Seneca, De ira 3.37.4 (first half of the first century C.E.). Plutarch, Quaest. Conviv. 5.6.679E–680B.
12. For sitting in taverns: Martial, Epig. 5.70; with wives on feast days, Tertullian, Ad Uxorem 2.6 (c. 207 C.E.). Tönnes Kleberg, Hôtels, restaurants et cabarets dans l’antiquité romaine (Uppsala, 1957), pp. 114–15; J. Packer, “Inns at Pompeii,” Cronache Pompeiane 4 (1978): 49 with refs. Athenaeus provides evidence that at family dinners, an Athenian might recline to eat with his children standing beside him (Deipn. 11.460A; cf. 1.11F, 1.17F, 10.428B). As a gesture of mourning, Cato took a vow to sit at meals for the rest of his life (Plutarch, Cato Minor 56.7, 67.1).
13. Aristophanes, Vespae 1208–20, Ecclesiazusae 834–76; Athenaeus associates the introduction of the dining couch with luxury (Deipn. 1.18B, 10.427B, 11.460A); and notes where sitting is customary among Greeks and non-Greeks (Deipn. 1.11F, 1.17F–18A, 4.143E, 4.148F–149D, 4.151A–152B, 5.191F, 10.428B, 10.443A; cf. Xenophon, Anabasis 7.3.21, Posidonius, 87.15). Elio Pellizer argues that vase painting shows a wider diffusion of this custom among the “merchantile, artisan, or peasant classes” (“Outlines of a Morphology of Sympotic Entertainment,” in Murray, Sympotica, p. 181).
14. Frederick Cooper and Sarah Morris, “Dining in Round Buildings,” in Murray, Sympotica, pp. 67–85. They suggest therefore that early Christian meals would have conformed to the urban proletariat’s custom of dinning in a seated position. Laws pertaining to the mysteries of Apollo, Karneios, and Demeter in the early first century B.C.E. attest that in ritual settings, dining was not segregated (Nany Bookidis, “Ritual Dining in the Sanctuary of Demeter and Kore at Corinth: Some Questions,” in Murray, Sympotica, p. 91).
15. Dio Chrysostom, Or. 7.75–76, shows a rural household meal at which a wife and daughter sit, although there is a stranger present. Lucian suggests that even at private meals, men’s posture differed from that of women’s (Lucius 2; cf. Apuleius, Metamorphoses 1.22, 1.26 [middle to late second century]). Valerius Maximus (early first century C.E.), Factorum et Dictorum Memorabilium 2.1.2. Plutarch (end of the first century C.E.) has Melissa reclining at table but Eumetis sits, probably because of her youth (Sept. Sap. Conviv. 150B, 148C); Eumetis does not speak (ibid., 154B); both women withdraw before the serious drinking begins (ibid., 155E; cf. Plutarch, Quaest. Conviv. 1.1.612F–613A; 1.3.619D). Achilles Tatius describes several family meals where women recline (Leucippe and Cleitophon 1.5, 2.3, 2.9 [Alexandria, late second century C.E.]). Petronius, Cena Trimalchionis, 37, 54, 65, 67, 69, 70 (second half of the first century C.E.). For courtesans reclining, Lucian, Dialogi Meretricii 1.3.284, 12.1.311; Athenaeus, Deipn. 4.149C, 14.641D; Lucian, Symposium 13. There is scant evidence for children, Xenophon, Symp. 1.8; Suetonius, De vita Caesarum, Aug. 64.3; Claud. 32 (published c. 121 C.E.); Horace, Carm. 4.15.25–28; Tacitus, Annales 13.16 (late first century C.E.). When a boy received the toga virilis in his late teens, he would begin to recline (Polybius, 31.25.4). Autolycus sits next to his reclining father, even at a party given in his honor (Xenophon, Symposium 1.8 [late fifth century B.C.E.]); Athenaeus, Deipn. 4.149C, 6.255E. Athenaeus does cite festivals of inversion in both the Greek and Roman world in which slaves reclined to eat and were served by their masters and children, suggesting that ordinarily the slaves would also wait upon the children (ibid., 14.639B–C); cf. Plutarch, Quaest. Conviv. 1.1.612F–613A; idem, Quaestiones Romanae 33. Aesop, reportedly an ex-slave, sits on a stool (Plutarch, Sept. Sap. Conviv. 150A); cf. Athenaeus, Deipn. 6.257A. For archaeological analysis of benches for sitting in triclinia, see K. Dunbabin for bibliography on the arrangement in the Casa del Criptoportico (late republic, early empire) (“Triclinium,” in Murray, Sympotica, p. 123). A few provincial funerary reliefs of the third and fourth centuries of our era show women seated beside men reclining at banquets (Dentzer, Le motif du banquet couché, pp. 286–87).
16. Plutarch, Quaest. Conviv. 7.8.702E–F. Cornelius Nepos, Praef. 7.
17. (See note above.) It was a mark of “barbarity”; exhibited by the Illyrians (Athenaeus, Deipn. 10.443A; quoting Theopompus [fourth century B.C.E.], Philippica, Fragmenta Historicorum Graecorum, ed. C Müller, 1.284); Cicero, Verr. 2.1.26.66; Isaeus (420–350 B.C.E.), Disc. 3.14; Dio Chrysostom (40–112 C.E.), Or. 7.57–76. Oswyn Murray, “Symposium and Genre in the Poetry of Horace,” Journal of Roman Studies 75 (1985): 40, 48–49; cf. idem, “Symposion and Männerbund,” in Concilium Eirene 16 (Prague, 1982), pp. 50–51, comparing sympotic customs east and west.
18. John H. D’Arms, “Control, Companionship, and Clientela: Some Social Functions of the Roman Communal Meal,” Echos du monde classique/Classical Views, n.s., 3 (1984): 327–48; idem, “The Roman Convivium and the Idea of Equality,” in Murray, Sympotica, pp. 308–20.
19. Athenaeus tells us that among the Macedonians, reclining at table was a mark of adult initiation. No male could recline until he had speared a wild boar without a net. Cassander was embarrassed by still sitting at banquets at the age of thirty-five (Deipn. 1.18a). Lucian defends the art of the parasite (i.e., the uninvited dinner guest), by claiming it is “the most royal” of trades since he “plies his trade, like a king: lying down” (De Parasito 23). For changes of posture during dinner see Plutarch, Quaest. Conviv. 5.6.679E–680B; on being assigned a bad seat, Plutarch, Sept. Sap. Conviv. 148F. Petronius, Cena 38, 57, 65. See also Elias Canetti, Crowds and Power, trans. Carol Stewart (New York, 1963), pp. 387–93.
20. Plutarch, Quaest. Conviv. 1.2.615D–619A; 3.619B–F. Its inversion in the Saturnalia (Lucian, Sat. 17). Horace, Serm. 2.8.11; 19–41. Sidonius Apollinaris, Ep. 1.11.10 (mid-fifth century C.E.). Trimalchio gets it wrong, and sits in the place reserved for the most honored guest (Petronius, Cena 31.8).
21. Plato, Symp. 1.2; Plutarch, Sept. Sap. Conviv. 615C–619A; Plutarch notes that the place of honor varies among peoples, Quaest. Conviv. 1.3.619A–B; the symposium should be a democratic institution, ibid., 1.2.615C–619A.
22. Cooper and Morris, “Dining,” pp. 79, 81. Oswyn Murray, “The Affair of the Mysteries: Democracy and the Drinking Group,” in idem, Sympotica, pp. 150–51.
23. Philo, De Vit. Contemp. 67–72.
24. For Pompeii, P Soprano, “I triclini all’aperto de Pompei,” in Pompeiana (Naples, 1950), pp. 288–310; W. Jashemski, The Gardens of Pompeii (New Rochelle, 1979), pp. 9–97, 215–16, 230–31, 243–44, 247, 253.
25. See Ramsay MacMullen for relevant inscriptional and artifactual evidence and further specialized bibliography on the presence and form of kitchens and dining rooms attached to sanctuaries in Italy, Greece, Asia Minor, Syria, and Egypt (Paganism, pp. 36–39, nn. 14–27).
26. 1 Corinthians 8:10; Plutarch, Sept. Sap. Conviv. 146D, although this account is fictional. The Pseudo-Clementine letters seem to presuppose that Christians are dining with pagans at the “tables of the gods” (early third century), Hom. 9.9; 9.15 (PG 2.248, 252).
27. C. A. Roebuck, The Asklepion and Lerna . . . (Corinth: Results of Excavations Conducted by the American School of Classical Studies at Athens 14; Princeton, 1951), pp. 51–55, quoted in MacMullen, Paganism, pp. 161–62, n. 16.
28. Regarding Sarapis (delivered in Smyrna in 142) (Or. 8.54.1, Dindorf 1.93, Behr Or. 45.27, 266–67).
29. The Palmyrene invitation, translated by MacMullen, is P.Yale 85 (Paganism, p. 38).
30. See the House of the Calender, in Doro Levi, Antioch Mosaic Pavements (Princeton, 1947), 1:36–39, fig. 12; 2:pl. 5; R. Stillwell, “Houses of Antioch,” Dumbarton Oaks Papers 15 (1961): 47–57.
31. Levi, Antioch Mosaic Pavements, 1:15–25, figs. 1–2; 2:pls. 1–2; 1:91–100, figs. 36–38. The Atrium house with its pattern of mosaics suggesting the horseshoe and crossbar design is regarded as one of the earliest houses in Antioch, perhaps constructed before the earthquake in 115. By this date, therefore, Roman fashions of the layout of a dining room had been adopted in the Greek world.
32. Jean Pierre Foucher, Découvertes archéologiques à Thysdrus en 1961 (Notes et Documents, n.s. 5, Insitut d’Archéologie, Tunis, n.d.), p. 27, pl. 36. Dunbabin dates the black and white mosaic to the first half of the second century C.E., “Triclinium,” p. 142, n. 45.
33. “ ‘Coenaculum’: La salle à l’étage du locus 30 à Khirbet Qumrân sur la Mer Morte,” in Res Orientales 4, Banquets d’Orient, ed. R. Gyselen (L’Étude de la Civilisation du Moyen-Orient: Burres-sur-Yvette, 1992), pp. 61–84. This hypothesis, however, leaves unexplained the ink-wells found in room 30.
34. Also called the sigma-couch (Dunbabin, “Triclinium,” pp. 128–29 with refs.).
35. See Dunbabin for references, “Triclinium,” pp. 129, 143, nn. 54–56.
36. Martial, Epig. 10.48, 14.87; Pliny, Ep. 5.6.36.
37. J. Lassus, “La salle à sept absides des Djemila-Cuicul,” Antiquités africaines 5 (1971): 193–207. This building is not earlier than the mid-fifth century C.E. See Dunbabin for further references, “Triclinium,” p. 143, n. 62. Petronius presents Trimalchio as having four dining rooms (Cena 77).
38. “Stibadium” is derived from the Greek word for pillow (stibadion) (Dunbabin, “Triclinium,” pp. 132–33 with refs.). Christian catacomb paintings of banquet scenes place them in an outdoor setting (E. Rosenthal, The Illuminations of the Vergilius Romanus [Zurich, 1972], pp. 54–58, pl. 9: fol. 100v).
39. J. Engemann, “Der Ehrenplatz beim antiken Sigmamahl,” in Jenseitvorstellungen in Antike unde Christentum: Gedenkschrift für Alfred Stuiber, JAC Ergbd. 9 (Münster, 1982), pp. 239–50.
40. Martial, Epig. 11.31.4–7 (for an entire meal constructed out of gourds). Horace, Serm. 2.4. A house in Antioch has a mosaic floor, dated by Levi to the late second-early third centuries, which illustrates the vessels and foodstuffs featured in banquets, as well as the order in which they would appear. Dishes of silver hold appetizers of eggs, artichokes, and pig feet; a fish, a ham, two types of bird; and a cake. Interspersed among these dishes are garlands, loaves of bread, and drinking vessels. Where the central table would be located is a mosaic of Gannymede, the cup-bearer to Zeus. When the table was removed after dinner before the drinking party began, this apt motif would be revealed (House of the Buffet Supper, in Levi, Antioch Mosaic Pavements, 1:127–37, 2:pls. 23–26).
41. Athenaeus, Deipn. 2.49D. Horace, Serm. 2.8.25–43.
42. In the fifth century, Sidonius Apollinaris tells us of guests gathering in a library of a country house or villa before lunch. The sexes were segregated, and while some sat on chairs to converse, others entertained themselves with a variety of games (Sidonius, Ep. 2.9.4). Macrobius describes a similar predinner gathering (Saturnalia 1.6.1; written in the fifth century but set in the 380s, according to Alan Cameron [Journal of Roman Studies 56 (1966): 25–38]), which now has archaeological support (see Jeremy Rossiter, “Convivium and Villa in Late Antiquity,” in Dining in a Classical Context, ed. William J. Slater [Ann Arbor, 1991], esp. pp. 201–2). Although the attestation is late, this evidence may provide the Greco-Roman parallel that Bahr was unable to locate for the rabbinic custom of serving hors d’oeuvres in an antechamber (M. Nid. 2.5, “The Seder of Passover,” p. 188).
43. Athenaeus, Deipn. 2.58B-60B. Athenaeus notes that some people therefore call snails “dinner-delayers” (Deipn. 2.63D).
44. Lucian, De Merc. Conduct. 26. Eggs were also common dessert fare (Athenaeus, Deipn. 14.641F): “small” food, in any case.
45. Athenaeus, Deipn. 3.75E, 14.639B; for tragemata, 14.640B ff. Petronius, Cena 66.
46. The context is a wedding (Conviv. 38).
47. Lucian, De Merc. Conduct. 26, Symp. 14, Sat. 33; Petronius, Cena 50. Apuleius, Metam. 2.19. Athenaeus, Deipn. 6.228C–234C.
48. Plutarch, Quest. Conviv. 2.10.642F.
49. Lucian, De Merc. Conduct. 28. Athenaeus, Deipn. 3.101; 9.409E. Plutarch, Quaest. Conviv. 1.1.615B; 3.1.645D–648A; idem, Sept. Sap. Conviv. 150C-D; 159D.
50. Petronius, Cena 27, 41, 47; Lucian, Symp. 35. Athenaeus, Deipn. 1.4D, 1.17C, E. Clement of Alexandria, Paedagogus 2.3.39 (second century C.E.). The poet Euphorion (third century B.C.E.), boorishly, made water into one of the drinking cups (Athenaeus, Deipn. 11.477E). The Sybarites were the first to bring their own chamber pots to parties (Athenaeus, Deipn. 12.519E); forbidden at Naucratite dinners (ibid., 4.150A). Drinking honey was one early cure for hangover (Athenaeus, Deipn. 11.784B). For an artistic depiction of symposiastic vomiting, see Michael Vickers, Greek Symposia (London, n.d.), pp. 20–21, fig. 24.
51. For a guest who brought his own knife with a bone handle, Juvenal, Sat. 11.133 (late first century C.E.); for toothpicks, Martial, Epig. 14.22; for spoons, Martial, Epig. 14.120; Petronius, Cena 33; for snail or egg-picks, Martial, Epig. 14.121, 8.71.9–10, 8.33.23–24; for implements to induce vomiting, Martial, Epig. 3.82; Petronius, Cena 33. For pictures of first-century tableware and dining practice, Paul Veyne, “The Roman Empire,” in his A History of Private Life: From Pagan Rome to Byzantium, ed. Philippe Aries and Georges Duby, trans. Arthur Goldhammer (Cambridge, Mass., 1987), 1:187–92.
52. Vergil, Georgics 4.376–377 (late first century B.C.E.); Ovid, Ars Amatoria 3.355 (early first century C.E.); Petronius, Cena 31, 34, 47; Sidonius, Ep. 1.11.14. Athenaeus, Deipn. 1.18F, 2.60A, 7.292E, 9.408B–410C. Hand washing is elevated to religious ritual in M. Hag. 2.5; it becomes a point of dispute in Mark 7:3. Dining rooms may sometimes be identified by the presence of little basins for washing, or by evidence of drains which allowed the floor to be washed down after a meal.
53. Josephus, Bellum Judaicum 2.138–139.
54. Quintilian, Declamationes 301 (second half of first century C.E.); Vergil, Georgics 4.380–385; Athenaeus, Deipn. 4.149D–E; 15.675B–C.
55. Athenaeus, Deipn. 5.179.B–C.
56. Poetae Lyrici Graeci frag. 7.56 = Athenaeus, Deipn. 5.179D. Semonides’ dates are uncertain; he is usually dated to the seventh century B.C.E., but may have lived as early as the eighth century.
57. Horace, Serm. 2.8.10; Petronius, Cena 32.
58. Lucian, Symp. 42. Petronius, Cena 66; presents for the guests to take home, ibid., 40, 56, 60. Athenaeus, Deipn. 1.13A. Martial, Epig. 2.37.7; 7.20.13.
59. De Merc. Conduct. 15.
60. Athenaeus, Deipn. 4.149C, 9.409D; cf. Mark 7:28.
61. Athenaeus, Deipn. 9.408E–410C, 11.462C–D, 14.641D, 14.642C–F, 15.665B–C; for wreaths, ibid., 15.674B–685C; for perfumes, ibid., 15.686C–692E; for brooms, Martial, Epig. 14.82; for mosaics of an “unswept room,” Pliny, Naturalis Historia 36.60.184 (a work published two years before his death in 79 C.E.); cf. H. Meyer, “Zu neueren Deutengen von Asaorotos Oikos und kapitolinischem Taubenmosaik,” Archäologischer Anzeiger (1977): 104–10, with earlier references. For pictures, see Jashemski, Gardens of Pompeii, p. 95.
62. A form reflected in the traditions of the Lord’s Supper in Luke 22.20; 1 Cor. 11:25.
63. Pliny, N.H. 14.28.143. Athenaeus, Deipn. 2.58B–C.
64. Oswyn Murray, “The Greek Symposion in History,” in Tria Corda: Scritti in Onore di Arnaldo Momigliano, ed. E. Gabba (Como, 1983), p. 263; idem, “The Symposion as Social Organisation,” in The Greek Renaissance of the Eighth Century B.C.: Tradition and Innovation, ed. Robin Hägg (Stockholm, 1983), pp. 195–200. Cf. William J. Slater, “Peace Symposium and the Poet,” Illinois Classical Studies 6 (1981): 206.
65. Murray, “The Greek Symposion in History,” pp. 259, 266–67, 271–72; idem, “Symposion and Männerbund,” 47–52. People “met together at the symposion precisely because they had common aims and interests outside it. Thus the feast was the natural place in which to define the position of the group, not least by looking back to the past” (Wolfgang Rosier, “Mnemosyne in the Symposion,” in Murray, Sympotica, p. 233).
66. The equation of food and learning is especially strong in Athenaeus, who repeats the remark of the comic poet Euphron that “nothing distinguishes the cook from the poet: for each invention is their skill” (Athenaeus, Deipn. 1.7F; cf. Murray, “Greek Symposion,” pp. 270–71). His running joke is whether the food will ever be eaten or just endlessly discussed (Athenaeus, Deipn. 6.222A, 223D–E; Joel C. Relihan et al., “Rethinking the History of the Literary Symposium,” Illinois Classical Studies 17 (1992): 220–24, 227, 233–34). Trimalchio appropriates the custom but misunderstands the point, decreeing pretentiously that “it is imperative to display erudition while dining” (Petronius, Cena 39.4).
67. Jan Bremer assesses the evidence that the symposium had an initiatory function (“Adolescents, Symposion, and Pederasty,” in Murray, Sympotica, pp. 135–48).
68. Relihan, “Rethinking,” pp. 215–16. Symposion couches develop from the design for biers (John Boardman, “Symposion Furniture,” in Murray, Sympotica, pp. 122–24).
69. Relihan, “Rethinking,” pp. 241–42.
70. Cena 26.9. See Niall W. Slater, Reading Petronius (Baltimore and London, 1990), pp. 50–86, esp. 54–55.
71. Plutarch, Quaest. Conviv. 6.7.692B–693E. Athenaeus, Deipn. 4.129E–F, 10.426D, 10.430D–431E, 11.782A–B. Petronius, Cena 64, 65, 68. Plutarch, Quaest. Conviv. 3.9.657B–E. Diodorus Siculus, Historia, 4.3 (mid-first century B.C.E.). Greek Anthology 11.49 (first century C.E.).
72. Plato, Symp. 176A (mid-fourth century B.C.E.); Petronius, Cena 60; Athenaeus, Deipn. 4.149E, 5.179D, 10.427C–D, 11.462E–463B, 11.486F–487B. Athenaeus recalls how Plato invited the diners in the Philebus to pray “to Dionysus or Hephaestus, or whichever god has been assigned the honor of the mixing [of the wine]” (Deipn. 10.423A–B, citing Philebus 61 B,C). For a list of favorite deities toasted in banquets, see Athenaeus, Deipn. 15.692F–693F. J. Rudhardt, Notions fondamentales de la pensée religieuse et actes constitutifs de culte dans la Grèce classique (Geneva, 1958), pp. 240–45, esp. 241–42. Dentzer, Le motif du banquet couché, pp. 503–27.
73. Athenaeus, Deipn. 2.36B–C. This progression can be documented by vase paintings; see Vickers, Greek Symposia, pp. 16–20.
74. Dio of Prusa, 51.19.7. Cf. Horace, Od. 4.5.31–40.
75. “Augusto, patri patriae, feliciter” (Cena 60.7).
76. Athenaeus, Deipn. 11.500B.
77. “Blessings” translates simply as “good things.” Athenaeus, Deipn. 14.659E, quoting a cook in The Flatterer (Kock 3.82). He also tells us that Theophrastus (late fourth century B.C.E.) understood this ritual gesture as a petition “not to behave badly or drink immoderately, but to gain benefit and nobility” (Athenaeus, Deipn. 15.693D).
78. Athenaeus, Deipn. 1.16B; 2.36B–C, 10.441C, 14.628A–B, 14.630F, 15.685A; for examples of popular songs, see ibid., 15.694C–696A. Plutarch, Quaest. Conviv. 5.5.679. See Plutarch’s comparison of Greek and Jewish feasts, ibid., 4.6.671D–672C. For additional bibliography, see Aune, “Septem Sapientium Convivium,” p. 71.
79. Proposis, Athenaeus, Deipn. 10.432D, 11.498C–D; from left to right, ibid., 4.152D, 11.464A; toasts offered by Illyrian women, ibid., 10.443A.
80. Athenaeus, Deipn. 4.152C–D, 10.423B.
81. Athenaeus, Deipn. 11.460A–503F. Petronius, Cena 50, 52. Since metal can give wine an unpleasant taste, Clement of Alexandria spoke in favor of simple crockery (Paedagogus 2.3.35).
82. Athenaeus, Deipn. 11.460D–F
83. Homeric Hymn to Hestia 1.5; cf. Deipn. 5.192B–C, 5.149B–C, 5.179. Cornutus, Theol. Graec. 53 (mid-first century C.E.); Cicero, De Natura Deorum 2.67. For further bibliography, see The Homeric Hymns, ed. T. W. Allen, W. R. Halliday, and E. E. Spikes (Amsterdam, 1980), pp. 88, 428–29.
84. Plato, Symp. 176E; the Therapeutae discuss scripture after dinner (Philo, De Vita Contemp. 75–79). For lighter entertainment, see Plutarch, Quaest. Conviv. 7.8.711A–713F; 9.15.747A–748D. Athenaeus, Deipn. 14.613D ff. Burkhard Fehr, “Entertainers at the Symposion: The Akletoi in the Archaic Period,” in Murray, Sympotica, pp. 185–95, esp. pp. 187–92.
85. Lines 124–25, trans. Frederick W. Danker, Benefactor: Epigraphic Study of a Graeco-Roman and New Testament Semantic Field (St. Louis, 1982), p. 156.
86. Quaest. Conviv. 2.10.642F–643A; cf. 4.660B. For this reason the question of numbers was important: if the gatherings “become so large so that the guests can no longer talk to each other or enjoy the hospitality together or even know one another, then it ceases to be a party at all” (Quaest. Conviv. 5.5.678). George Paul, “Symposia and Deipna in Plutarch’s Lives and in Other Historical Writings,” in Slater, Dining in a Classical Context, p. 1569.
87. Plutarch, Quaest. Conviv. 2.3.635E–638A; Macrobius, Sat. 7.16.1; why truffles are thought to be produced by thunder (Quaest. Conviv. 4.2.664B–665A; Juvenal, Sat. 5.16–18).
88. Plutarch, Quaest. Conviv. 2.1.629E–634F, 5.673A; for riddles see Athenaeus, Deipn. 10.448B–453C; Plutarch, Sept. Sap. Conviv. 148D, 154B; Petronius, Cena 58. See also Christopher P. Jones, “Dinner Theater,” in Slater, Dining in a Classical Context, pp. 185–98. In classical times, the game of kottabos was particularly popular, in which dinners would flick drops of wine from their cups at a designated target. See B. A. Sparkes, “Kottabos, an Athenian After-Dinner Game,” Archeology 13 (1960): 202–7.
89. Quaest. Conviv. 2.1.629E–634F.
90. For an extensive section on the doings and witty sayings of prostitutes at symposia, see Athenaeus, Deipn. 13.555A–604D. See also Elio Pellizer, “Outlines of a Morphology of Sympotic Entertainment,” pp. 177–95; Chester Starr, “An Evening with the Flute-Girls,” La Parola del Passato 33 (1978): 401–10.
91. Athenaeus, Deipn. 15.702A–B.
92. Ibid., 15.702B–C. Text is fragmentary for this final washing: only the verb “to wipe” remains.
93. Suetonius, Tib. 42.2. Lee E. Klosinski, “Meals in Mark” (Ph.D. dissertation, Claremont, 1988), p. 212; Veyne, “Roman Empire,” in History of Private Life, 1:79. The younger free men of the Therapeutae wait on table (Philo, De Vit. Contemp. 71–72).
94. R. P. Saller, “Slavery and the Roman Family,” Slavery and Abolition 8 (1987): 71. But the slave girl Fotis serves Milo, his wife, and guest at table (Apuleius, Metam. 2.11).
95. John H. D’Arms, “Slaves at Roman Convivia,” in Slater, Dining in a Classical Context, p. 172.
96. In the Roman west: Suetonius, Calig. 39.2; Seneca, De Ira 3.37.4, Ep. 19.11; Martial, Epig. 7.86; Petronius, Cena 40, 56, 60; cf. CIL 6.3975. Sidonius, Ep. 2.9.6. For the Greek east, Matt. 22:3, 4, 9; Luke 14:17, 21, 23. Lucian, De Merc. Conduct. 15. Athenaeus, Deipn. 4.128D, 4.129E–F.
97. Seneca, Ep. 47.8.
98. Athenaeus, Deipn. 12.510A, 5.177B–178B; Plato allows guests to invite other people at the extraordinary symposium (Symp. 7.6; cf. Xenophon, Symp. 1.13). See Fehr, “Entertainers at the Symposion,” pp. 185–95. Does this perhaps common scenario lie behind Matthew’s version of the wedding feast, from which a man is expelled for not wearing an appropriate garment (Matt. 22:11–14)?
99. Lucian, De Merc. Conduct. 15, 29, Sat. 38.
100. Pliny, N.H. 14.14.91; Pliny Ep. 2.6.2–5; Martial, Epig. 9.2; Juvenal, Sat. 5.125–127; Lucian, De Merc. Conduct. 14, 26, 27; cf. D’Arms, “The Roman Convivium and the Idea of Equality,” in Murray, Sympotica, p. 318.
101. Lucian, De Merc. Conduct. 26; cf. idem, Sat. 17, 22.
102. Lucian, Sat. 17, 22.
103. Apuleius, Metam. 1.15–16; cf. CIL 6.9083. Fortunata played this role in Petronius (Cena 67).
104. Lucian, Dial. 4.
105. Plato, Symp., 175A; Petronius, Cena 31; Luke 7:44, John 13:1–11.
106. Seneca, Ep. 95.24; Petronius, Cena 50, 68; Apuleius, Metam. 10.13–14; Lucian, De Merc. Conduct. 26. Justinian, Digest 33.7.12.32. See the remarks on social status being conveyed by the specialized tasks of slaves in Saller, “Slavery and the Roman Family,” p. 71. Martial, Epig. 7.20.17; Seneca, Ep. 47.5.
107. Cena 31, 34–36, 40, 68, 70.
108. Sidonius, Ep. 1.11.10; Ammianus Marcellinus, 16.8.9 (mid-fourth century C.E.).
109. Seneca, Ep. 47.3.
110. Petronius, Cena 67, 70, 74; Horace, Serm. 2.6.66–67; Plutarch, Quaest. Conviv. 7.4.703D–E; Apuleius, Metam. 10.13–14; Lucian, Symp. 42. Athenaeus, Deipn. 6.262D.
111. As John D’Arms remarks, the brutality of the punishments seems all the more grotesque given the convivial setting. For examples, see Seneca, De Ira 3.40.2, De Clem. 1.18.2; Dio of Prusia, 54.23.1–4; Pliny, N.H. 9.39.77; Suetonius, Calig. 32.2, Ner. 5.1; Apuleius, Metam. 8.31; Lucian, Lucius 39; Petronius, Cena 34.6, 47, 49; Juvenal, Sat. 9.5, 14.15–22; Ammianus Marcellinus, 28.4.16; Martial, Epig. 3.13, 3.94, 8.23; Lucian, Sat. 4.
112. See M. I. Finley, Ancient Slavery and Modern Ideology (New York, 1980), pp. 95 ff; Keith M. Hopkins, Conquerors and Slaves, Sociological Studies in Roman History 1 (Cambridge, 1978), pp. 118–23; K. R. Bradley, Slaves and Masters in the Roman Empire: A Study in Social Control (Oxford, 1987), pp. 113–37. Petronius, Cena 27, 57. For the possible rabbinic influence on this gesture, see William M. Clarke, “Jewish Table Manners in the Cena Trimalchionis,” Classical Journal 87 (1992): 257–63, with bibliography.
113. An important text on this topic is Plutarch’s Quaest. Conviv. 7.8.711A. See also Christopher P Jones, “Dinner Theater,” in Slater, Dining in a Classical Context, pp. 185–98.
114. Petronius, Cena 41; Lucian, Sat. 24, Symp. 15, De Merc. Conduct. 16; Juvenal, Sat. 11.145–160.
115. See the theoretical studies of Pierre Bourdieu on the habitus (Outline of a Theory of Practice, trans. Richard Nice [Cambridge, 1977], pp. 72–95; idem, The Logic of Practice, trans. Richard Nice [Stanford, 1990], pp. 69–79); cf. my article, “Clement of Alexandria on the Importance of Table Etiquette,” Journal of Early Christian Studies 3 (1995): 123–41. See Margaret Visser, The Rituals of Dinner: The Origins, Evolution, Eccentricities, and Meaning of Table Manners (New York and London, 1991).
116. For conduct expected at temple dining facilities, see J. T. Milik, Recherches d’épigraphie proche-orientale, vol. 1, Dédicaces faites par des dieux (Palmyre, Hatra, Tyr) et des thiases sémitiques à la époque romaine (Paris, 1972), pp. 188–92, 287. For Latin inscriptional evidence for the custom of incising basic codes of manners on temple dining room walls, see MacMullen, Paganism, p. 39, n. 24. For the Greek equivalent, see, F. Poland, Geschichte des griechishen Vereinswesens (Leipzig, 1909), pp. 454 ff.
117. Erving Goffman, Interaction Ritual: Essays on Face-To-Face Behavior (New York, 1967), p. 51.
118. Athenaeus, Deipn. 8.347D–E; not talking as an excuse to get more eating done was reprehensible (ibid., 10.421D).
119. Lucian, Symp. 33–44, 46.
120. Statutes of the Iobacchi, 89–91. Gate-crashing was apparently a problem (Polybius, 26.1.4; Diodorus, 29.32; Athenaeus, Deipn. 10.439A).
121. Lucian, De Merc. Conduct. 14, 26; Plutarch, Sept. Sap. Conviv. 148B–C; Athenaeus, Deipn. 5.178E. Cf. the parable told by Yohanan ben Zakkai, Talmud Babli Shabbath 153a; Matt. 22:11.
122. Martial, Epig. 4.66.3; 5.79 (changed eleven times during dinner). Favored colors included green, purple, or even variegated tints; Petronius, Cena 30, 67. Ethel H. Brewster, “The Synthesis of the Romans,” Transactions and Proceedings of the American Philological Association 49 (1918): 131–43; Walton B. McDaniel, “Roman Dinner Garments,” Classical Philology 20 (1925): 268–70.
123. Lucian, De Merc. Conduct. 14.
124. Athenaeus, Deipn. 5.178F–179B; Petronius, Cena 34; Plutarch, Quaest. Conviv. 6.7.692. Theophrastus describes the flatterer as being of all the guests, the first to praise the wine, and to say to his host “How delicately you eat!” and to pick up one of the dishes from the table, commenting, “This is very fine, is it not!” (Charac. 2.10).
125. Plutarch, Pompeius 40.7; Lucian, Dial. Meret. 6.3.294; Clement of Alexandria, Paed. 2.7.54–55.
126. Lucian, De Merc. Conduct. 15.
127. Plutarch, Quaest. Conviv. 2.10.643F–644A. Ovid, Amator. 3.355–358; Lucian, De Merc. Conduct. 17, Sat. 8, Symp. 11, 36, 42; Clement of Alexandria, Paed. 2.2.31; theft of a bowl, Lucian, Symp. 46; theft of a cup, Tacitus, Hist. 1.48; Plutarch, Galba 12.2–3; Suetonius, Claud. 32.
128. From the word for “rustic” (Lucian, De Merc. Conduct. 17).
129. Plutarch, Quaest. Conviv. 1.1.614E; 1.2.616C–F; 3.644F–645C; cf. Plato, Symp. 1771A1–1778A1; Lucian, Symp. 34–35. Murray, “The Greek Symposium in History,” in Gabba, Tria Corda, p. 260. Relihan, “Rethinking,” pp. 216–18.
130. Lucian, De Merc. Conduct. 18, Sat. 18.
131. Lucian, Sat. 38, Symp. 47; Petronius, Cena 78.
132. Ovid, Amator. 3.751–752, 759–760; Horace, Odes 3.6.25–32; Pliny, N.H. 14.141; Juvenal, Sat. 1.55–57; Suetonius, Aug. 69.1, Calig. 25.1; Achilles Tatius, Leucippe 2.9. John C. Yardley, “The Symposium in Roman Elegy,” in Slater, Dining in a Classical Context, pp. 149–55. Grafitto in Pompeii, CIL 4.7698; cf. J. V. P. D. Balsdon, Life and Leisure in Ancient Rome (London, 1969), pp. 50–53.
133. Michelle Z. Rosaldo and Louise Lamphere, Women, Culture and Society (Stanford, 1974); cf. Rosaldo, “The Use and Abuse of Anthropology: Reflections on Feminism in Cross-Cultural Understandings,” Signs 5 (1980): 389–417; Jean Bethke Elshtain, Public Man, Private Woman: Women in Social and Political Thought (Princeton, 1981). Judith Romney Wegner, Chattel or Person? The Status of Women in the Mishnah (New York and Oxford, 1988), esp. pp. 145–67. And most recently, Kathleen E. Corley, Private Women, Public Meals: Social Conflict in the Synoptic Tradition (Peabody, Mass., 1993).
134. Bernadette Brooten, “Early Christian Women and Their Cultural Context: Issues and Method in Historical Reconstruction,” in Feminist Perspectives on Biblical Scholarship, ed. Adela Yarbro Collins (Chico, Calif., 1985), pp. 65–91; idem, “Jewish Women’s History in the Roman Period: A Task for Christian Theology,” HTR 79 (1986): 22–30; idem, Women Leaders in the Ancient Synagogue: Inscriptional Evidence and Background Issues, Brown Judaic Studies 36 (Chico, Calif., 1982). For a temple kitchen (culina) given by four freedwomen, see L’Année Épigraphique 1975, no. 197; cf. CIL 10.5166. A mosaic of city life from fourth-century Antioch shows a woman of high status in the market place among the men at work or play (Levi, Antioch Mosaic Pavements, 1:326–37, 2:pls. 79–80).
135. For information on the lives of lower status women, see Mima Maxey, “The Occupations of the Lower Classes in Roman Society,” in Two Studies on the Roman Lower Classes, ed. M. F. Park and M. Maxey (New York, 1975); Beryl Rawson, “Family Life Among the Lower Classes at Rome in the First Two Centuries of the Empire,” Classical Philology 61 (1966): 71–83; Susan Treggiari, “Jobs for Women,” American Journal of Ancient History 1 (1976): 76–104. Freedwomen, like their male counterparts, were particularly allied with trade, and therefore would not only be very much a part of social interchange, but might be sufficiently wealthy to threaten the traditional social hierarchy, which was based on status ascribed at birth, buttressed by the display of a certain lavishness of lifestyle. See A. M. Duff, Freedmen in the Early Empire (New York, 1958); Susan Treggiari, Roman Freedmen during the Late Republic (Oxford, 1969); Richard I. Pervo, “Wisdom and Power: Petronius’ Satyricon and the Social World of Early Christianity,” Anglican Theological Review 67 (1985): 307–25; Dale B. Martin, Slavery as Salvation: The Metaphor of Slavery in Pauline Christianity (New Haven and London, 1990); Ramsay MacMullen, “Women in Public in the Roman Empire,” Historia 29 (1980): 208–18.
136. For evidence of women at weddings and other “family” occasions, see Lucian, Symp. 8–9, 35, 44; Athenaeus, Deipn. 2.71E–F, 4.145D, 6.245A–C, 9.377B, 9.408E, 14.644D–E (four tables are set up for the women and six for the men), 14.646A (Sosibius [third century B.C.E.] mentions dinners held for women in Sparta honoring a bride; an ancient equivalent of the bridal shower?); Cornelius Nepos, Praefatio 6–7 (beginning of the first century C.E.); Ovid, Met. 12.210–220. Sometimes the women present at a wedding would have separate dining facilities (Lucian, Symp. 8–9; Petronius, Cena 67–69; Athenaeus, Deipn. 14.644D). Cf. Aune, “Septem Sapientum Convivium,” p. 72, esp. n. 45. Cooper and Morris suggest that at these “mixed” banquets, the men may have sat in deference to the custom of the women (“Dining,” p. 80 n. 46). Such deference also dictated that in depictions of their banquets all the gods sat. Dentzer cites both wedding and funeral feasts as occasions when women might be present, Le motif du banquet couché, pp. 431–32, 446–47.
137. Petronius, Cena 71. Athenaeus, Deipn. 7.202C.
138. Poseidippos summarized in Athenaeus, Deipn. 9.377B; cf. ibid., 5.185.
139. Ep. 10.96.
140. See note 15.
141. This conclusion would solve the logical inconsistencies found even in excellent articles such as that of Aune, who cannot, finally, explain how women who were “ordinarily not participants in the sacred meals of religious societies” were drawn in such large numbers to the Corinthian Christian assembly (“Septem Sapientium Convivium,” p. 77). On the other hand, Greek and Latin writers often indict women dining with men as “foreign” behavior: Livy, 1.57 (Etruscans) (end of the first century B.C.E.–beginning of the first century C.E.); Herodotus, 5.18 (Persians) (fifth century B.C.E.); Quintus Curtius 5.1.37–39 (Persians) (mid-first century C.E.); Athenaeus, Deipn. 1.23 (Tyrrhenians); Tacitus, Germania 18–19 (Germans).
142. “Deciphering a Meal,” Daedalus 101 (1972): 61.
143. Quaest. Conviv. 1.612D; 2.10.642F–643A. Factions at table were frowned upon: Statutes of the Iobacchi, 13; 1 Cor. 11:17–34. D’Arms, “Control, Companionship and Clientela,” p. 339. Athenaeus, Deipn. 5.177A–C, 4.149A–D.
144. Even among the Essenes, Rule of the Congregation [1QSa] 2.11–22; for the Corinthians, see Gerd Theissen, The Social Setting of Pauline Christianity: Essays on Corinth (Philadelphia, 1982), esp. pp. 147–74.
145. Paul F. Bradshaw, The Search for the Origins of Christian Worship (New York and Oxford, 1992), pp. 47–55. Thus, although Corley accepts that “early Christians gathered together for standardized meals,” she provides no evidence for any “standardization,” nor do I see that there is any (Private Women, p. 17). For a rethinking of the “house” church, see Robert Jewett, “New Testament Churches and Communal Meals in the Early Church: The Implications of a Form-Critical Analysis of 2 Thessalonians 3:10,” Biblical Research 38 (1993): 23–43.