Chapter Six
Polytheism and Scripture

For those who have been raised in the ambiance of one of the great monotheist religions—Judaism, Christianity, or Islam—polytheism poses far more problems than its multiplicity of gods. Both the centrality and the diversity of the rituals of polytheism, extending from the domestic environment to the grandest civic ceremonies, suggest that religious education and religious practice were essentially one and the same thing for pagans. Children were not instructed in polytheism apart from their introduction to established rites and participation in them. Worshippers turned to no sacred texts that they found either inspirational or inviolate. Stories of the gods, from Homer to the latest mythographers, were read and appreciated, but scarcely canonical. Polytheism had, in short, no scripture.

The absence of religious texts, as opposed to philosophical reflections on religion (such as we find in Plato, Aristotle, or Cicero), moved Arnaldo Momigliano to observe in an important passage in one of his later writings, "The mere fact that one had to study in order to be pious is a strange notion which made Judaism increasingly intellectual—not what cults were known for in the Graeco-Roman world.... We should recognize that, while in Athens and Rome thinking about religion usually made people less religious, among Jews the more you thought about religion, the more religious you be-

came."^[1] Had he lived longer, Momigliano would probably have addressed the interesting question of whether the increasing pagan awareness of monotheism, as mediated by Christianity in Greek texts, had some kind of resonance in pagan literature. The aretalogies and encomia of the greater Greek and Roman deities never became universal texts for the devout. These were genres for expression of individual, personal piety. As Momigliano recognized, one could hardly expect devotional texts from polytheist authors, in view of the essentially ritualistic rather than intellectual character of their religion. But some parallel to the monotheist scriptures might well be expected eventually in a world in which monotheism and polytheism were approaching a direct and momentous collision.^[2]

It was the literary critic Northrop Frye, reflecting on the nature of biblical narratives, who suggested that what became known in modern times as the romance constituted nothing less than a secular scripture. The stories of the romantic novels, with their fiery emotions, strong loves and hates, hair-raising vicissitudes, exotic settings, and memorable incarnations of great virtue and great vice bore many resemblances to the biblical narratives. Although ancient fiction was not a primary concern of Frye's, he quite properly included several works of the genre as forerunners

[1] A. Momigliano, "Religion in Athens, Rome, and Jerusalem in the First Century B.C. ," Annali della Scuola Normale Superiore di Pisa , ser. 3, 14 (1984), 873-92, citation from the final page. This article was reprinted in Ottavo contributo alla storia degli studi classici et del mondo antico (1987), pp. 279-96; in On Pagans, Jews, and Christians (1987), pp. 74-91; and (in Italian translation) in Saggi di storia della religione romana (1988), pp. 27-43.

[2] For a powerful analysis of the impact of this collision, see now G. Fowden, Empire to Commonwealth: Consequences of Monotheism in Late Antiquity (1993).

of the romance as secular scripture. He outlined his theories in a series of Charles Eliot Norton Lectures at Harvard.^[3] Coincidentally, only a few years later, again in a series of Norton lectures, Frank Kermode also turned to the Bible, and in particular to the New Testament, to develop a sophisticated analysis of novelistic elements in the Gospels.^[4] He argued that the problem of historical truth is so elusive in the Gospel narratives that those accounts are better viewed simply as fiction with a semblance of truth. The meaning and, obviously, the inspirational value of works of this kind do not depend upon their historical veracity, although apprehension of that meaning nonetheless does depend upon a provisional or temporary belief in their veracity. This is, in Kermode's words, a "benign deceit" that readers even today continue to countenance. "How far we do so," observes Kermode, "because of the saturation of our culture by the Gospels and traditional interpretations one need not try to say."^[5]

Both Frye and Kermode, from their separate literary perspectives, offer something like a solution to the dilemma posed by Momigliano, even though Momigliano wrote later than they did. The material in the Gospel narratives, as well as in the Acts of the Apostles, constituted a kind of narrative fiction in the form of history (, as Julian was to say)^[6] that was essentially new to the Graeco-Roman world. Its roots belonged in the inspirational narratives of the Hebrew Bible. But the written versions of these narratives, and presumably oral versions that preceded them, were disseminated among Greeks and

[3] N. Frye, The Secular Scripture: A Study of the Structure of Romance (1976).

[4] F. Kermode, The Genesis of Secrecy: On the Interpretation of Narrative (1979).

[5] Ibid . 122.

[6] Julian, Epist . 89b, 301b (Bidez).

Hellenized Jews. Their availability in the Greek language gave them a far greater impact in the Graeco-Roman world. Their potential for stimulating secular narratives grew with the portentous diffusion of Christian doctrines and practices as outlined by the evangelists. All this was set in motion in the generation after the crucifixion of Jesus, which, it will be recalled, occurred in the latter part of the reign of Tiberius.

Parallels in form and substance between the writings of the New Testament and the fictional production of the imperial age are too prominent to be either ignored or dismissed as coincidental. Both Celsus, in his attack on the Christians, and Origen, in his defense of them, recognized the similarities, particularly, as we have seen, where apparent miracles—such as the open tomb or resurrection of the dead—were at issue.^[7] It is, furthermore, a plain fact of chronology that the distinctive fictional forms of the Roman empire begin, on present evidence, no earlier than the reign of Nero and proliferate conspicuously soon thereafter. To be sure, antecedents of this fiction, such as the Homeric tales, Ctesias's Persian fantasies, Xenophon's Cyropaideia , Hellenistic travel literature, and the lost lubricities of the short Milesian tales, serve to identify some of the scattered elements that the imperial writers assimilated, brought together, and transformed in order to create what, on any accounting, was a wholly new phenomenon in Graeco-Roman literature.^[8] Let us remember that the invocation of sources and antecedents never provides an expla-

[7] See Chapter V above.

[8] For antecedents, see, for example, B. E. Perry, The Ancient Romances: A Literary-Historical Account of Their Origins (1967), or the more recent and more sophisticated book by B. P. Reardon, The Form of Greek Romance (1991). For possible Eastern antecedents, G. Anderson, The Novel in the Graeco-Roman World (1984).

nation of an innovation: they can only reveal, inadequately at best, some of the building blocks that were used to construct it.

If the implied contacts between polytheist fiction and Christian scripture are suggestive, and if the chronology of the two is remarkably similar, we may be emboldened to press Northrop Frye's concept of the secular scripture a little further. Achilles Tatius's novel, Leucippe and Cleitophon , provides a good point of departure. Near the beginning of the second book we are in the city of Tyre at a feast of Dionysus ("of the Vintage," "of the Harvest of Grapes").^[9] The writer declares that Dionysus is considered one of the gods of the city, and he pauses to tell the story of the origin of the feast. It is nothing less than an account of the discovery of wine: "No wine ever existed among men before the Tyrians had it," says Achilles Tatius, although he soon reveals that he is aware of the well-known, rival tradition among the Athenians that the god brought wine first to a certain Ikarios in Attica.^[10] What follows is a story that is attested nowhere else, neither in the mythographic tradition nor in any of the detailed accounts of Tyrian legends such as we find in the Dionysiaca of the poet Nonnos.

Before the arrival of Dionysus in Tyre everyone drank water; but, when the god found himself hospitably received by a shepherd, he offered as a pledge of friendship a cup with a delectable red drink. "Where, stranger," said the shepherd, "did you find this water stained with red? Where did you find such sweet blood? For nothing of the kind flows on the ground." After the shepherd offered an eloquent paean to the bouquet of the wine, Dionysus proclaimed solemnly, "This is the water of early autumn, this is the blood of the grape

[9] Achill. Tat. Leuc. and Cleit . 2.2.1.

[10] Ibid . 2.2.2-3.

." He then took the shepherd to a vine, plucked a cluster of grapes, and crushed it before him, saying, "This is the water, and this is the fountain."^[11]

No reader acquainted with the evangelists can miss the similarity to the Christian eucharist. Some have also seen a hint of the Johannine miracle at Cana, where Jesus turned water into wine. But, since the wine here is pressed directly from the grape, the eucharistic parallel is more compelling. In a footnote to his recent and brilliant translation of Achilles Tatius, the late Jack Winkler observed, "If the resemblance of Dionysus's words and gesture to the Christian eucharistic rite is not accidental, it must surely be interpreted as parody."^[12] Yet in the rich texture of Achilles Tatius's novel, parody is an element so hard to find that it would be rash to invoke it here. On the other hand, the parallelism is hardly likely to be accidental. The language is far too close, and the solemnity of the divine gift to mortal men too marked. Observe the phrasing in the Gospel of Matthew, for example: and —"This is my body," "This is my blood."^[13] The connection between Achilles Tatius in this passage and the account of the eucharist in the Gospels has, it must be said, been exploited by some exegetes of the New Testament, but in a very curious way. The novelist is assumed to provide reliable evidence about an otherwise unknown rite in Phoenicia in the vicinity of Tyre. That rite is then advanced, on this basis, as a pagan precedent for the utterance of Jesus before his

[11] Ibid . 2.2.4-6.

[12] J. Winkler in B. P. Reardon, ed., Collected Ancient Greek Novels (1989), p. 192 n. 25. Note that in his translation Winkler rearranged the text of Book z of Achilles Tatius. He placed chapters z and 3 between chapters 8 and 9.

[13] Matt. 26. 26, 28.

disciples.^[14] Jesus is seen, in other words, as reworking an old pagan tradition in the formulation of his eucharistic program. This is a reckless way to handle evidence that belongs indisputably to a time at least a century or so after the lifetime of Jesus.

The tendency of Christian interpreters to look for the pagan origins of Christian rites, utterances, and images has all too often obscured influences in the reverse direction. This is particularly true for late antiquity, but to some extent also for the earlier imperial period. The story of the eucharist had, by the time of Achilles Tatius, been available in all the canonical Gospels as well as Saint Paul's First Letter to the Corinthians. We know from the famous letter of the younger Pliny to the emperor Trajan, as well as from ecclesiastical texts, that the story of the eucharist and the rite enjoined by Jesus had led to disreputable reports, culminating in accounts of Thyestian banquets among the Christians.^[15] Partaking of divinity by eating human flesh and drinking human blood was not a familiar occurrence to the Greeks and the Romans. In the immemorial past the gods had done this once or twice among themselves (Kronos being the most notorious example, thought by many scholars to come from a pre-Hellenic tradition), and the horrors of Thyestes' table symbolized the practice. In recalling the boiled dinner that Tantalus served from the pieces of his son, Pelops, Pindar had to admit that he found it hard to call any of the blessed ones a cannibal .^[16] Such

[14] For a learned presentation of this kind of interpretation, see Morton Smith, "On the Wine God in Palestine," Salo Wittmayer Baron Jubilee Volume , American Academy for Jewish Research (1975), pp. 815-29.

a thing was plainly alien to him. Nor, for that matter, was this practice any more familiar to the Jews, for whom the drinking of blood of any kind was strictly forbidden and the eating of human flesh inconceivable.^[17]

Achilles Tatius therefore invented something new and exciting when he transferred the revelation of wine from Attica to Tyre. The most plausible source for his invention is the Gospel story. It makes far more sense to postulate a direct influence upon the Greek novelist than to suppose that the writer innocently preserved an otherwise unknown tradition of great antiquity that was the source that inspired Jesus himself. Ikarios in the old account of the discovery of wine and the Tyrian shepherd in the new one play similar roles, but the hieratic pronouncement of Dionysus, as it is given in Achilles Tatius, has absolutely no parallel in the versions of the Ikarios story from earlier times. The formulation —"This is..."—is unique in this text.

Of course, the most surprising thing about the eucharist is the miracle of the body and blood of Jesus Christ. Not even Dionysus is made to say that the wine is actually his blood. As Margaret Visser has recently observed in her book The Rituals of Dinner with reference to the eucharist, "The ultimate taboo is reverently broken as the people share out and consume his [i.e., Jesus's] body and blood, the proofs of his love."^[18] Theologians have long speculated about the origins of the eucharist. Not surprisingly they have found it a singularly intractable topic. When the Reallexikon für Antike und Christentum reached the letter E, the editors prudently postponed action by instructing the reader who looks up Eucharistie to turn instead to Kultmahl , an item that

[17] Cf. M. Smith, Jesus the Magician (1987), pp. 122-23.

[18] M. Visser, The Rituals of Dinner (1991), p. 36.

has not yet appeared. The eucharist is celebrated on the occasion of Passover, but as Gillian Feeley-Harnik has observed in her important book The Lord's Table, the Jewish Passover is an affirmation of family and nation, whereas the eucharist and Last Supper affirm the annihilation both of family ties and of national allegiance.^[19] Insofar as food serves as a language for expressing human relations, the totally unexpected and unprecedented admonition to eat flesh and drink blood symbolizes, according to Feeley-Harnik, an inversion of the Jewish Passover ceremony.^[20] Certainly there is nothing in traditional Judaism that would explain what Jesus was doing at the Last Supper. As we have already observed, drinking blood was taboo; and, as Morton Smith has trenchantly remarked, the rites of Judaism did not include cannibalism.^[21]

But if the origins of the eucharist remain controversial, the way in which nonbelievers perceived it is less so. The underlying concept of the eucharist would appear to be nothing less than cannibalism. As Margaret Visser says, it is treated as a rite of the highest solemnity and reverence, but it is nonetheless a breaking of the ultimate taboo. To the extent that anthropologists can help, some conclusions can be drawn— and have been drawn—from the actual rather than miraculous (or symbolic) practice of cannibalism. In one of its forms, notably among the Aztecs, the eating of human flesh that has been dedicated and sacrificed to a god allows the cannibalistic worshipper to become one with that god. This point is made well, and with reference to the Christian rite, by Peggy Reeves Sanday in her study Divine Hunger: Cannibalism as a Cultural

[19] G. Feeley-Harnik, The Lord's Table: Eucharist and Passover in Early Christianity (1981), chap. 5 ("The Last Supper").

[20] Ibid. 130.

[21] See n. 17 above.

System .^[22] She rightly draws attention to the occasional reappearance of this motif in some of the Christian martyr literature. For example, when the body of the dead Polycarp is borne from the fire in which his corpse was merely baked rather than burned up by the flames, masses of people rush forward to have a share in his holy flesh .^[23]

Cannibalism, like resurrection, was utterly alien to traditional Graeco-Roman life and thought. Before the first century of the Christian era it figures very rarely in texts of any kind apart from a few myths such as those of Kronos or Thyestes. Only the adventurous and ever-curious Herodotus seems to show any sympathetic interest whatever in the subject. He treats it three times, in connection with remote peoples on the edge of the inhabited world. Of the Massagetae, who were neighbors of the Scythians, he records that it was customary for relatives to cook and eat an elderly member of the family who had recently died (unless of disease).^[24] This endocannibalism, as the anthropologists call it, is well attested in other societies and should not be discredited in Herodotus. The same form of cannibalism is also reported for a tribe called the Padaeans, east of the Indians (and therefore about as far away as any people Herodotus talks about).^[25] The Issedonians, another people beyond the Scythians, likewise eat their relatives but are careful to cook the flesh together with that of a sheep, which is specially sacrificed at the time of the person's decease.^[26] The two meats are mixed together

[22] P.R. Sanday, Divine Hunger: Cannibalism as a Cultural System (1986), p. 193.

[23] Martyr. Polyc . 17.

[24] Herod. 1.216.

[25] Ibid. 3.99.

[26] Ibid. 4.26. On eating one's dead relations, see also the remarks of the Stoic Chrysippus, preserved in Sextus Emp. Pyrr. Hyp . 3. 247-48, a reference for which I am indebted to A. A. Long.

and served up indiscriminately at a banquet, in much the same way as was done in African Mali until relatively recently, according to still-unpublished reports from an anthropologist who has been working there.^[27]

Herodotus's three accounts of cannibalism are all confined to the eating of relatives or tribespeople, in a rite designed to ensure the cohesion and continuity of family or nation. There is no trace whatever of anything like the Aztec union with the divine through cannibalistic sacrifice and ceremony. And, apart from Herodotus, there is scarcely any trace of the practice at all until the Christian era. Morton Smith has suggested that cannibalism, real or metaphorical, may have played some role in magical practices of the Hellenistic and early Roman periods, but the evidence is extraordinarily tenuous. In two papyri wine is imagined to be blood in an invocation designed to induce love in the person for whom the spell is cast. For example, "Give blood of Osiris that he gave to Isis to make her feel love in her heart for him, give to cause her to feel a love for him in her heart, the love that Isis felt for Osiris."^[28] This sort of thing may, or may not, have some connection to the mystery cults of the imperial period, but in the case neither of magic nor of the mysteries do we find the clear motif of cannibalism that dominates the eucharist. The magical parallels were, of course, important for Morton Smith in trying to prove his famous and eccentric case that Jesus was actually a Palestinian magician of the period. But, even if we

[27] I have this information through the courtesy and generosity of Prof. Sarah Brett-Smith.

[28] Smith (n. 17 above), pp. 122-23, citing The Demotic Magical Papyrus of London and Leiden (1904), XV.

reject that case (as most do), there is no clear explanation to put in its place.

Although we do not know exactly what the origins of the eucharist were, we do know that nothing like it had been seen before. That is a major reason why the origins are so hard to find. Certainly no ritual cannibalism, documented, alleged, or metaphorical, had ever appeared within the societies confined by the Graeco-Roman world. It is in the Gospel of the most intellectual and subtle of the evangelists, John, that we can see clearly that the eucharistic rite posed substantial difficulties even for the people to whom Jesus brought his message. John acknowledged candidly that the Jews murmured against Jesus because he said, "I am the bread that came down from Heaven." Not unreasonably the Jews asked, "Don't we know his father and mother?" And they went on to ask, "How can this man give us his flesh to eat?"^[29] We can hardly doubt that the eucharist was a startling innovation in the Graeco-Roman world. It is hardly surprising that, as word of it spread, tales arose of Thyestian banquets and worse.

Accordingly, when we find cannibalism playing a very conspicuous role in imperial fiction, it is no more likely to be there accidentally than the proclamation of Dionysus about the blood of his grape. The Boukoloi, the "rangers" or "desperadoes" of the Nile Delta, are the chief but by no means only exhibit. In the first of the three executions of Leucippe in Achilles Tatius's novel, we read that the Boukoloi tied the heroine up and that one of them "next raised a sword and plunged it into her heart and sawed all the way down to her abdomen. Her viscera leaped out. The attendants pulled out her entrails and carried them in their hands over to the altar. When it was well done they carved the

[29] John 6.41-52.

whole lot up, and all the bandits shared the meal."^[30] Leucippe's lover, Cleitophon, who watched the whole episode, observed with a rueful and epigrammatic pun, "The burial of your entrails is now the nourishment of the bandits." Or, as Winkler has wittily rendered this in his translation, "Your insides are inside the outlaws, victuals in the vitals of bandits."^[31]

There is a comparable scene of cannibalism in the fragment of the Phoenician novel ascribed to a certain Lollianus and published a little over twenty years ago by Albert Henrichs.^[32] A boy is thrown on his back, his heart cut out and placed over the fire. Then someone removes it, cuts it in half, sprinkles it with barley, and drizzles a little olive oil over it. When it is adequately cooked, we read that the heart is distributed to the participants. The parallelism in the motif of cannibalism has led some to assume that the villains here must again be the Boukoloi or Egyptian rangers, but obviously that inference is by no means necessary.^[33] The subject seems to have acquired a certain vogue in the time of Lollianus and Achilles Tatius.

It surfaces again, for example, in a more conspicuous and better-known text, the Toxaris of Lucian, that memorable exploration of cultural relativism in the Roman empire. The allusive and well-read Lucian is probably evoking the reports in Herodotus when he makes his Greek speaker in the dialogue say to the Scythian interlocutor, "I should not have expected friendship to be so highly cherished among the Scythians..., judging

[30] Achill. Tat. Leuc. and Cleit . 3.15.4-5.

[31] Winkler (n. 12 above), p. 217.

[32] A. Henrichs, Die Phoinikika des Lollianos (1972), translated by G. N. Sandy in B. P. Reardon, ed., Collected Ancient Greek Novels (1989), pp. 810-12.

[33] Cf. Henrichs (n. 32 above), pp. 48-56 ("Die Anthropophagie der ägyptischen Bukolen").

by all that we hear about them, and especially the report that they eat their dead fathers."^[34] It is immaterial that it was not exactly the Scythians who did this, but their neighbors. What is significant is the presence of the item at all, presumably borrowed, if inaccurately, from Herodotus.

Nor was Lucian the only one who found such long-neglected information in Herodotus now of particular interest. Petronius was the earliest of the fictional writers of the imperial age, so far as we can tell, to introduce the theme of cannibalism and, in the process, to recall both Herodotus and the Christian eucharist. Unlike Achilles Tatius, Petronius was a parodist, and a brilliant one. His extraordinarily nuanced treatment of the eating of human flesh occurs in the last extant fragments of the Satyricon , where someone is reading out the testament of Eumolpus and then commenting on it.^[35] It may be that this is Eumolpus himself in anticipation of his own death, or perhaps it is one of his eager legatees after he has died, or perhaps he has died a death that is to turn out only make-believe, a Scheintod . We have no way of telling.

But the text of the testament of Eumolpus is unambiguously preserved and of exceptional interest. It reads as follows: "All those who are legatees in my will, apart from my freedmen, shall take what I have given only under the following condition—that they divide my body into parts and eat it in the presence of the assembled people."^[36] The recitation of the will continues, in our extant text, "We know that among certain tribes this practice is

[34] Lucian, Tox . 8. Cf. De Luctu 21 For Herodotus, see above in this chapter, with reference to Herod. 1.216, 3.99, 4.26.

[35] Petron. Satyr . 141.2-4.

[36] Ibid. 141.2. A century later Artemidorus mentions a dream (Oneir . 5.42) in which cannibalism portends inheritance of the property of the victim.

still maintained, that the dead should be consumed by their relatives, maintained indeed to such a degree that the sick are frequently blamed because, by living on, they are diminishing the quality of their flesh. With this in mind, I advise my friends not to refuse what I order, but to eat my body [corpus ] with the same enthusiasm with which they have cursed my spirit [spiritum ]."^[37]

The reference to complaints about the sick in a cannibalistic society seems manifestly to derive from Herodotus's account of the Padaeans, of whom he says, "If one of their number be ill, man or woman, they take the sick person, and, if he be a man, the men of his acquaintance proceed to put him to death, because, they say, his flesh would be spoiled for them if he pined and wasted away with sickness. The man protests he is not ill in the least; but his friends will not accept his denial—in spite of all he can say, they kill him, and feast themselves on his body. So also, if a woman be sick, the women who are her friends take her and do with her exactly the same as the men."^[38] Eumolpus clearly views the legacy hunters, the captatores , as the equivalent of the resentful Padaeans waiting for their sickly tribesman to die. But

[37] Petron. Satyr . 141.3-4. The text of the last sentence in K. Müller's edition is "his admoneo amicos meos ne recusent quae iubeo, sed quibus animis devoverint spiritum meum, eisdem etiam corpus consumant." In CQ 37 (1987), 529-32, G. B. Conte proposes devorarint for devoverint . The pleonasm devorarint/consumant weakens the sentence, and spiritum devorare seems an odd expression (it would appear to be unexampled, and it is certainly not the same as "devouring" words or books, for which examples can be produced). The contrast of spiritus with corpus is vital here: one cannot consider both edible.

why, we must ask, has Petronius recalled a Herodotean item of this kind? Why has he made Eumolpus compose his will in this bizarre way? There is, after all, nothing about a will in Herodotus, nor indeed in any of the three Herodotean reports of cannibalism. How did Petronius come to conceive of a testamentum like this? For a testamentum is what this is.

When Jesus spoke to his disciples at the Last Supper, he invoked a covenant according to which his disciples were to share in his body and blood by consuming bread and wine. In the Greek Gospels this is made explicit by the use of the Greek word . In Matthew, "This is my blood, of the covenant [ ]." In Mark, "This is my blood," again . In Luke, "This cup is the new covenant in my blood," and essentially the same words are used by Paul in addressing the Corinthians.^[39] The new translations of the Bible regularly use the word "covenant" to render , but it was not always so. The King James version, which had so tremendous an influence on English literature afterward, used the translation "testament." And that is why the New Testament is called the New Testament. In Greek, it is simply the new —"the new covenant."

In fact, the translators of the King James version were accurately reflecting the normal meaning of in the Greek of Jesus's own time. It was the word for "testament" or "will," whereas only in the context of Hellenistic Judaism was it a word for "covenant."^[40] It is fair to say that the sense "testament" was

[39] Matt. 26.28, Mark 14.24, Luke 22.20, I Cor. 11.25.

accordingly far more general than the sense "covenant." Greek speakers to whom Christian doctrine was unfamiliar would certainly have heard in these phrases the word for "testament" or "will," and understood in Jesus's instructions to eat and drink the testamentary stipulations of a man about to die. The word must have been current in Greek accounts of the Last Supper from the death of Jesus onward. This is guaranteed by the consistent and uniform use of the term when the various Gospel narratives were finally written down.

Since, as the Oxford English Dictionary guarantees, the word "testament" in English is just as unfamiliar in the sense "covenant" as it was in Greek, the ordinary English reader can have a good sense of how the Gospel text in Greek must have sounded to a Greek reader by simply listening to the words of the King James translators. Matthew: "For this is my blood of the New Testament." Mark: "This is my blood of the New Testament." Luke: "This cup is the New Testament in my blood." And so on. Of course, in more recent times, the very expression "New Testament" means, first and foremost, the aggregate of scriptural texts that we call the New Testament. But, if we can think away that association and hear the word "testament" in its normal sense, we shall have some understanding of what a Greek heard in .

The testament or "covenant" is called new, . This word too has particular connotations that differentiate it from the normal word for "new" in Greek, or, in the feminine, , "newness," implies and often means "innovation," Something radically new and different, even strange.^[41] The

, therefore, is something altogether new and different in the way of a testament. It may be suggested that the brilliant parody of Petronius makes play with this sense of "New Testament" in giving us the highly innovative, not to say strange, will of Eumolpus. It would be stranger still if the stipulation that his legatees eat his body were not a pointed allusion to the Gospel story of the Last Supper. The distinction he makes between his corpus and his spiritus reflects perfectly the opposition of (or ) and in the Gospels.^[42] That is to say, Petronius has given us the New Testament of Eumolpus. He then goes on through the mouth of a speaker (who may even be Eumolpus himself or someone else) to justify the terms of the will by recalling the practices of northern barbarians together with a few alleged cases of cannibalism in time of war and under the pressure of imminent starvation.

Petronius's treatment of this motif is not only the most brilliant in extant fiction but also, as we have observed, the earliest. Like so much else in the history of imperial fiction, it dates from the reign of Nero. It was a portent of the impact that the tales of the evangelists were to have on the imagination of writers and readers in the Graeco-Roman world for several centuries to come. The Gospels were themselves supplemented by the ac-

count of the Acts of the Apostles, a work that in small compass already strikingly resembles the novels of the polytheists.^[43] The structure and pace of the book of Acts would still be the envy of any novelistic writer today. It begins with an ascension, moves briskly to the grisly death of a treacherous disciple, and then turns smartly to a scene in which tongues of flame produce a babble of many languages. For invention and exotica, this is hard to beat. The narrative continues across a broad geographical canvas that includes Syria, Phoenicia, Cyprus, the Pamphylian coast, Asia Minor, Macedonia, and Greece. It includes dramatic courtroom scenes and rises to a climax in a richly described episode of shipwreck that is worthy to stand alongside the many shipwrecks in pagan fiction.

Yet, as the popularity of the novelists grew—and the papyri increasingly suggest that it did—it was perhaps not surprising for the Christians to pick up in turn and to exploit the very genre that seemed to have come into being, to some degree, as a response to stories of theirs that were now enshrined in the canonical Gospels. By the early third century at the latest, the Christians, to our knowledge, had acquired at least one major novel of their own along the lines of the polytheist ones and in evident imitation of them.^[44] In this work, known only from excerpts, some in the

[43] For a useful study of parallels, R. I. Pervo, Profit with Delight: The Literary Framework of the Acts of the Apostles (1987). Unfortunately, Pervo sees novels as an influence on Acts—which is implausible for chronological reasons. R. A. Burridge, What Are the Gospels? A Comparison with Graeco-Roman Biography (1992), p. 245, finds Pervo unconvincing from the perspective of genre. No one seems to have asked whether Acts, whatever its genre, could itself have influenced the novelists. Once again we have the familiar failure to look for the impact of Christianity on polytheist culture.

[44] For a still useful, general account of this work, see Perry (n. 8 above), pp. 285-93. See also W. Heintze, Der Clemensroman und seine griechischen Quellen (1914); and O. Cullmann, Le problème littéraire et historique du roman Pseudo-Clémentine (1930).

Greek original and some in Latin translation, a Roman matron takes her twin sons away to Athens after being warned in a dream that all three of them would perish if they remained. One son remains behind with his father, and that boy is called Clement, destined to be a saint. The father loses all contact with his wife and twin sons; he assumes that they have met their deaths. But he nonetheless determines to go off on a voyage to find out the truth. Young Clement, now on his own, makes his way to Jerusalem and there encounters Saint Peter, who edifies him by a series of at least twenty sermons. Peter and Clement, together with other disciples, then go to an island off the coast of Phoenicia to see the sights and, in doing so, encounter an old woman begging in front of a door. The old woman turns out ultimately to be Clement's mother, who reveals that her husband's brother had once made sexual advances to her, and it was really this that had decided her to leave Rome. Subsequently the twin brothers turn up at Syrian Laodicea under false names, and finally the father, as an old man, turns up as well.

There can be no doubt, nor has there been, that this is a Christian novel; closely parallel to the polytheist novels apart from the addition of Peter's homilies. The work is now known as the Clementine Recognitions , or, more pedantically and quite unnecessarily, the Pseudo-Clementine Recognitions . The title reflects various discoveries of the whole family of Clement in the course of the novel and also reflects the Latin name, Recognitiones , attached to the collection of excerpts known in Latin translation. The Clementine Recognitions may be said to constitute the meeting of Christian and secular scrip-

ture. They represent the appropriation, probably the inevitable appropriation, of a pagan and popular genre that itself owed so much to the miraculous narratives, both oral and written, of the early Christians. Fiction became antiquity's most eloquent expression of the nexus between polytheism and scripture.

The emergence of martyrologies, beginning in the second century, soon made more novels like the Clementine Recognitions superfluous. The martyr narratives were to provide the basis for an abundant production of instructive fiction in the centuries ahead, although the earliest martyr acts, based as they were on carefully maintained protocols of interrogation, had rather more historical veracity than was to be characteristic of the genre later.^[45] The fecundity of hagiographical narrative may even have inspired the latest of the pagan novelists, Heliodorus. At least it should be said that his account of Charicleia on the burning pyre could easily fit into a saint's life: "Charicleia climbed onto the pyre and positioned herself at the very heart of the fire. There she stood for some time without taking any hurt. The flames flowed around her rather than licking against her; they caused her no harm but drew back wherever she moved toward them, serving merely to encircle her in splendor."^[46]

The great novelists evidently appealed to Christians as much as pagans. They lost none of their appeal, even in late antiquity. If after the fourth century there were few (or none) to practice

[45] For an examination of the early martyr acts from this point of view, see G. W. Bowersock, Martyrdom and Rome , The Wiles Lectures (Queen's University of Belfast), forthcoming (Cambridge University Press).

[46] Heliod. Aeth . 8.9.13. Cf., e.g., Martyr. Polyc. 15 : "For the flames, bellying out like a ship's sail in the wind, formed into the shape of a vault and thus surrounded the martyr's body as with a wall. And he was within it not as burning flesh."

this craft any more, the works of Achilles Tatius, Heliodorus, and others continued to be read. Julian disliked them, but in his literary taste he was as atypical of his age as he was in most things. He had a horror of the erotic. The Christians, fortunately, did not, and the novelists, no less fortunately, seem to have avoided including material that was directly offensive to Christians (unless, of course, such material was censored in transmission—which is most unlikely). In any case, the Christians made these pagan works their own by piling fiction upon fiction. They were able to legitimize the novels and to ensure their survival until the time when the patriarch Photius could, with some surprise, peruse them. Heliodorus was declared to have been a bishop, and so was Achilles Tatius.^[47] More remarkable still was the early Byzantine appearance of a saint whose parents turned out to be none other than Leucippe and Cleitophon.^[48] Nothing could be farther from the truth than to say, as Ramsay MacMullen has, that Christianity put an end to a taste for novels.^[49] For a long period there were no new novelists, because the hagiographers took their place. But the extant novels continued to be read and prized.

It is easier to explain the gradual disappearance of a certain kind of creativity than it is to speculate on what promoted it in

[48] St. Galaktion, son of Leucippe and Cleitophon: Migne, PG 116.93. Cf. Perry (n. 8 above), p. 346, n. 4.

[49] R. MacMullen, "What Difference Did Christianity Make?" Historia 35 (1986), 322-43, esp. 342: "There were demonstrable changes in literature, too. Nothing similar to Heliodorus', Apuleius', or Petronius' novels could be published, nor poetry like Catullus' or Ovid's. There was a difference!" This article was reprinted in the author's Changes in the Roman Empire: Essays in the Ordinary (1990), pp. 142-55, with the passage cited here on pp. 154-55.

the first place. But in the course of these six chapters the connection between imperial fiction of various kinds and the Gospel narratives has grown ever stronger. The stories of Jesus inspired the polytheists to create a wholly new genre that we might call romantic scripture. And it became so popular that the Christians, in turn, borrowed it back—in the Clementine Recognitions and in the massive production of saints' lives. For the centrality of fiction in the history of the Roman empire, the debate between Celsus and Origen across nearly a century provided the first and principal clues. Celsus was, after all, a child of the second century, and he framed his questions in accordance with the thinking of that age. And it is not only Celsus's contemporary Lucian who authorizes us to generalize from the work of Celsus in this way. It is the simple fact that Origen essentially accepted the same premises as Celsus in trying to refute his arguments.

When narrative fiction made a triumphant return to the Greek world in the twelfth century, it did so by explicitly echoing the work of the earlier masters. The authors of the entry "Romance" in the new Oxford Dictionary of Byzantium say of this late Byzantine fiction, "Why the romance should reappear at this moment, after six centuries, is a question yet to be answered satisfactorily."^[50] Byzantinists of today are just beginning to address this question in earnest. If fiction can be seen not only within its historical context but as a part of the historical continuum itself, an answer should soon be at hand. One may be allowed to suspect that, in the vast and subtle theological writings of Byzantium in its final centuries, another Origen or even another Celsus may be found to point the way.

[50] E. M. and M. J. Jeffreys, Oxford Dictionary of Byzantium , ed. A. Kazhdan, vol. 3 (Oxford, 1991), p. 1804.