[In the following excerpt, Abbott searches in other genres—including the epic, the serious heroic romance, the mime, and the prologue of comedy-for elements that could have influenced Petronius's composition of the first known realistic romance.]

One of the most fascinating and tantalizing problems of literary history concerns the origin of prose fiction among the Romans. We can trace the growth of the epic from its infancy in the third century before Christ as it develops in strength in the poems of Naevius, Ennius, and Cicero until it reaches its full stature in the Aeneid, and then we can see the decline of its vigor in the Pharsalia, the Punica, the Thebais, and Achilleis, until it practically dies a natural death in the mythological and historical poems of Claudian. The way also in which tragedy, comedy, lyric poetry, history, biography, and the other types of literature in prose and verse came into existence and developed among the Romans can be followed with reasonable success. But the origin and early history of the novel is involved in obscurity. The great realistic romance of Petronius of the first century of our era is without a legally recognized ancestor, and has no direct descendant. The situation is the more surprising when we recall its probable size in its original form. Of course only a part of it has come down to us, some one hundred and ten pages in all. Its great size probably proved fatal to its preservation in its complete form, or at least contributed to that end, for it has been estimated that it ran from six hundred to nine hundred pages, being longer therefore than the average novel of Dickens and Scott. Consequently we are not dealing with a bit of ephemeral literature, but with an elaborate composition of a high degree of excellence, behind which we should expect to find a long line of development. We are puzzled not so much by the utter absence of anything in the way of prose fiction before the time of Petronius as by the difficulty of establishing any satisfactory logical connection between earlier forms of literature and the romance of Petronius. We are bewildered, in fact, by the various possibilities which the situation presents. The work shows points of similarity with several antecedent forms of composition, but the gaps which lie in any assumed line of descent are so great as to make us question its correctness.

If we call to mind the present condition of this romance and those characteristic features of it which are pertinent to the question at issue, the nature of the problem and its difficulty also will be apparent at once. Out of the original work, in a rather fragmentary form, only four or five main episodes are extant, one of which is the brilliant story of the Dinner of Trimalchio. The action takes place for the most part in southern Italy, and the principal characters are freedmen who have made their fortunes and degenerate freemen who are picking up a precarious living by their wits. The freemen, who are the central figures in the novel, are involved in a great variety of experiences, most of them of a disgraceful sort, and the story is a story of low life. Women play an important rôle in the narrative, more important perhaps than they do in any other kind of ancient literature—at least their individuality is more marked. The efficient motif is erotic. I say the efficient, because the conventional motif which seems to account for all the misadventures of the anti-hero Encolpius is the wrath of an offended deity. A great part of the book has an atmosphere of satire about it which piques our curiosity and baffles us at the same time, because it is hard to say how much of this element is inherent in the subject itself and how much of it lies in the intention of the author. It is the characteristic of parvenu society to imitate smart society to the best of its ability, and its social functions are a parody of the like events in the upper set. The story of a dinner party, for instance, given by such a nouveau riche as Trimalchio would constantly remind us by its likeness and its unlikeness, by its sins of omission and commission, of a similar event in correct society. In other words it would be a parody on a proper dinner, even if the man who described the event knew nothing about the usages of good society, and with no ulterior motive in mind set down accurately the doings of his upstart characters. For instance, when Trimalchio's chef has three white pigs driven into the dining-room for the ostensible purpose of allowing the guests to pick one out for the next course, with the memory of our own monkey breakfasts and horseback dinners in mind, we may feel that this is a not improbable attempt on the part of a Roman parvenu to imitate his betters in giving a dinner somewhat out of the ordinary. Members of the smart set at Rome try to impress their guests by the value and weight of their silver plate. Why shouldn't the host of our story adopt the more direct and effective way of accomplishing the same object by having the weight of silver engraved on each article? He does so. It is a very natural thing for him to do. In good society they talk of literature and art. Why isn't it natural for Trimalchio to turn the conversation into the same channels, even if he does make Hannibaltake Troy and does confuse the epic heroes and some late champions of the gladiatorial ring? In other words, much of that which is satirical in the Satirae of Petronius is so only because we are setting up in our minds a comparison between the doings of these rich freedmen and the requirements of good taste and moderation. But it seems possible to detect a satirical or a cynical purpose on the part of the author carried father than is involved in the choice of his subject and the realistic presentation of his characters. Petronius seems to delight in putting his most admirable sentiments in the mouths of contemptible characters. Some of the best literary criticism we have of the period he presents through the medium of the parasite rhetorician Agamemnon. That happy phrase characterizing Horace's style, "curiosa felicitas," which has perhaps never been equaled in its brevity and appositeness, is coined by the incorrigible poetaster Eumolpus. It is he too who composes and recites the two rather brilliant epic poems incorporated into the Satirae, one of which is received with a shower of stones by the bystanders. The impassioned eulogy of the careers of Democritus, Chrysippus, Lysippus, and Myron, who had endured hunger, pain, and weariness of body and mind for the sake of science, art, and the good of their fellow-men, and the diatribe against the pursuit of material comforts and pleasure which characterized the people of his own time are put in the mouth of the same roué Eumolpus. These situations have the true Horatian humor about them. The most serious and systematic discourse which Horace has given us, in his Satires, on the art of living comes from the crack-brained Damasippus, who has made a failure of his own life. In another of his poems, after having set forth at great length the weaknesses of his fellow-mortals, Horace himself is convicted of being inconsistent, a slave to his passions, and a victim of hot temper by his own slave Davus. We are reminded again of the literary method of Horace in his Satires when we read the dramatic description of the shipwreck in Petronius. The blackness of night descends upon the water; the little bark which contains the hero and his friends is at the mercy of the sea; Lichas, the master of the vessel, is swept from the deck by a wave; Encolpius and his comrade Giton prepare to die in each other's embrace, but the tragic scene ends with a ridiculous picture of Eumolpus bellowing out above the roar of the storm a new poem which he is setting down upon a huge piece of parchment. Evidently Petronius has the same dread of being taken too seriously which Horace shows so often in his Satires. The cynical, or at least unmoral, attitude of Petronius is brought out in a still more marked way at the close of this same passage. Of those upon the ill-fated ship the degenerates Encolpius, Giton, and Eumolpus, who have wronged Lichas irreparably, escape, while the pious Lichas meets a horrible death. All this seems to make it clear that not only does the subject which Petronius has treated inevitably involve a satire upon contemporary society, but that the author takes a satirical or cynical attitude toward life.

Another characteristic of the story is its realism. There are no marvelous adventures, and in fact no improbable incidents in it. The author never obtrudes his own personality upon us, as his successor Apuleius sometimes does, or as Thackeray has done. We know what the people in the story are like, not from the author's description of them, but from their actions, from the subjects about which they talk, and from the way in which they talk. Agamemnon converses as a rhetorician might talk, Habinnas like a millionaire stone-cutter, and Echion like a rag-dealer, and their language and style is what we should expect from men of their standing in society and of their occupations. The conversations of Trimalchio and his freedmen guests are not witty, and their jests are not clever. This adherence to the true principles of realism is the more noteworthy in the case of so brilliant a writer as Petronius, and those of us who recall some of the preternaturally clever conversations in the pages of Henry James and other contemporary novelists may feel that in this respect he is a truer artist than they are.

One other characteristic of the novel of Petronius must be noted in this connection. It is cast in the prosepoetic form, that is, passages in verse are inserted here and there in the narrative. In a few cases they are quoted, but for the most part they are the original compositions of the novelist. They range in length from couplets to poems of three hundred lines. Sometimes they form an integral part of the narrative, or again they illustrate a point, elaborate an idea in poetry, or are exercises in verse.

We have tried to bring out the characteristic features of this romance in order that we may see what the essential elements are of the problem which faces one in attempting to explain the origin of the type of literature represented by the work of Petronius. What was there in antecedent literature which will help us to understand the appearance on Italian soil in the first century of our era of a long erotic story of adventure, dealing in a realistic way with everyday life, marked by a satirical tone, and with a leaning toward the form of the prosimetrum? This is the question raised by the analysis of the characteristics of the story which we have made above. We have no ambitious hope of solving it, yet the mere statement of a puzzling but interesting problem is stimulating to the imagination and the intellect, and I am tempted to take up the subject because the discovery of certain papyri in Egypt within recent years has led to the formulation of a new theory of the origin of the romance of perilous adventure, and may, therefore, throw some light on the source of our realistic novel of everyday life. My purpose, then, is to speak briefly of the different genres of literature of the earlier period with which the story of Petronius may stand in some direct relation, or from which the suggestion may have come to Petronius for his work. Several of these lines of possible descent have been skilfully traced by others. In their views here and there I have made some modifications, and I have called attention to one or two types of literature, belonging to the earlier period and heretofore unnoticed in this connection, which may help us to understand the appearance of the realistic novel.

It seems a far cry from this story of sordid motives and vulgar action to the heroic episodes of epic poetry, and yet the Satirae contain not a few more or less direct suggestions of epic situations and characters. The conventional motif of the story of Petronius is the wrath of an offended deity. The narrative in the Odyssey and the Aeneid rests on the same basis. The ship of their enemy Lichas on which Encolpius and his companions are cooped up reminds them of the cave of the Cyclops; Giton hiding from the town crier under a mattress is compared to Ulysses underneath the sheep and clinging to its wool to escape the eye of the Cyclops, while the woman whose charms engage the attention of Encolpius at Croton bears the name of Circe. It seems to be clear from these reminiscences that Petronius had the epic in mind when he wrote his story, and his novel may well be a direct or an indirect parody of an epic narrative. Rohde in his analysis of the serious Greek romance of the centuries subsequent to Petronius has postulated the following development for that form of story: Travelers returning from remote parts of the world told remarkable stories of their experiences. Some of these stories took a literary form in the Odyssey and the Tales of the Argonauts. They appeared in prose too in narratives like the story of Sinbad the Sailor of a much later date. A more definite plot and a greater dramatic intensity were given to these tales of adventure by the addition of an erotic element which often took the form of two separated lovers. Some use is made of this element, for instance, in the relations of Odysseus and Penelope, perhaps in the episode of Aeneas and Dido, and in the story of Jason and Medea. The intrusion of the love motif into the stories told of demi-gods and heroes, so that the whole narrative turns upon it, is illustrated by such tales in the Metamorphoses of Ovid as those of Pyramus and Thisbe, Pluto and Proserpina, or Meleager and Atalanta. The love element, which may have been developed in this way out of its slight use in the epic, and the element of adventure form the basis of the serious Greek romances of Antonius Diogenes, Achilles Tatius, and the other writers of the centuries which follow Petronius.

Before trying to connect the Satirae with a serious romance of the type just mentioned, let us follow another line of descent which leads us to the same objective point, viz., the appearance of the serious story in prose. We have been led to consider the possible connection of this kind of prose fiction with the epic by the presence in both of them of the love element and that of adventure. But the Greek novel has another rather marked feature. It is rhetorical, and this quality has suggested that it may have come, not from the epic, but from the rhetorical exercise. Support has been given to this theory within recent years by the discovery in Egypt of two fragments of the Ninos romance. The first of these fragments reveals Ninos, the hero, pleading with his aunt Derkeia, the mother of his sweetheart, for permission to marry his cousin. All the arguments in support of his plea and against it are put forward and balanced one against the other in a very systematic way. He wins over Derkeia. Later in the same fragment the girl pleads in a somewhat similar fashion with Thambe, the mother of Ninos. The second fragment is mainly concerned with the campaigns of Ninos. Here we have the two lovers, probably separated by the departure of Ninos for the wars, while the hero, at least, is exposed to the dangers of the campaign. It was pointed out after the discovery of this find that the large part taken in the story by the carefully balanced arguments indicated that the story grew out of exercises in argumentation in the rhetorical schools.¹ The elder Seneca has preserved for us in his Controversiae specimens of the themes which were set for students in these schools. The student was asked to imagine himself in a supposed dilemma and then to discuss the considerations which would lead him to adopt the one or the other line of conduct. Some of these situations suggest excellent dramatic possibilities, conditions of life, for instance, where suicide seemed justifiable, misadventures with pirates, or a turn of affairs which threatened a woman's virtue. Before the student reached the point of arguing the case, the story must be told, and out of these narratives of adventure, told at the outset to develop the dilemma, may have grown the romance of adventure, written for its own sake. The story of Ninos has a peculiar interest in connection with this theory, because it was probably very short, and consequently may give us the connecting link between the rhetorical exercise and the long novel of the later period, and because it is the earliest known serious romance. On the back of the papyrus which contains it are some farm accounts of the year 101 A.D. Evidently by that time the roll had become waste paper, and the story itself may have been composed a century or even two centuries earlier. So far as this second theory is concerned we may raise the question in passing whether we have any other instance of a genre of literature growing out of a schoolboy exercise. Usually the teacher adapts to his purpose some form of creative literature already in existence.

Leaving this objection out of account for the moment, the romance of love and perilous adventure may possibly be then a lineal descendant either of the epic or of the rhetorical exercise. Whichever of these two views is the correct one, the discovery of the Ninos romance fills in a gap in one theory of the origin of the realistic romance of Petronius, and with that we are here concerned. Before the story of Ninos was found, no serious romance and no title of such a romance anterior to the time of Petronius was known. This story, as we have seen, may well go back to the first century before Christ, or at least to the beginning of our era. It is conceivable that stories like it, but now lost, existed even at an earlier date. Now in the century, more or less, which elapsed between the assumed date of the appearance of these Greek narratives and the time of Petronius, the extraordinary commercial development of Rome had created a new aristocracy—the aristocracy of wealth. In harmony with this social change the military chieftain and the political leader who had been the heroes of the old fiction gave way to the substantial man of affairs of the new, just as Thaddeus of Warsaw has yielded his place in our present day novels to Silas Lapham, and the bourgeois erotic story of adventure resulted, as we find it in the extant Greek novels of the second and third centuries of our era. If we can assume that this stage of development was reached before the time of Petronius we can think of his novel as a parody of such a romance. If, however, the bourgeois romance had not appeared before 50 A.D., then, if we regard his story as a parody of a prose narrative, it must be a parody of such a heroic romance as that of Ninos, or a parody of the longer heroic romances which developed out of the rhetorical narrative. If excavations in Egypt or at Herculaneum should bring to light a serious bourgeois story of adventure, they would furnish us the missing link. Until, or unless, such a discovery is made the chain of evidence is incomplete.

The two theories of the realistic romance which we have been discussing assume that it is a parody of some anterior form of literature, and that this fact accounts for the appearance of the satirical or cynical element in it. Other students of literary history, however, think that this characteristic was brought over directly from the Milesian tale² or the Menippean satire.³ To how many different kinds of stories the term "Milesian tale" was applied by the ancients is a matter of dispute, but the existence of the short story before the time of Petronius is beyond question. Indeed we find specimens of it. In its commonest form it presented a single episode of everyday life. It brought out some human weakness or foible. Very often it was a story of illicit love. Its philosophy of life was: No man's honesty and no woman's virtue are unassailable. In all these respects save in the fact that it presents one episode only, it resembles the Satirae of Petronius. At least two stories of this type are to be found in the extant fragments of the novel of Petronius. One of them is related as a well-known tale by the poet Eumolpus, and the other is told by him as a personal experience. More than a dozen of them are inbedded in the novel of Apuleius, the Metamorphoses, and modern specimens of them are to be seen in Boccaccio and in Chaucer. In fact they are popular from the twelfth century down to the eighteenth. Long before the time of Petronius they occur sporadically in literature. A good specimen, for instance, is found in a letter commonly attributed to Aeschines in the fourth century B.C. As early as the first century before Christ collections of them had been made and translated into Latin. This development suggests an interesting possible origin of the realistic romance. In such a collection as those just mentioned of the first century B.C., the central figures were different in the different stories, as is the case, for instance, in the Canterbury Tales. So original a writer as Petronius may well have thought of connecting these different episodes together by making them the experiences of a single individual. The Encolpius of Petronius would in that case be in a way an ancient Don Juan. If we compare the Arabian Nights with one of the groups of stories found in the Romances of the Round Table we can see what this step forward would mean. The tales which bear the title of the Arabian Nights all have the same general setting and the same general treatment, and they are put in the mouth of the same story-teller. The Lancelot group of Round Table stories, however, shows a nearer approach to unity since the stories in it concern the same person and have a common ultimate purpose, even if it is vague. When this point had been reached the realistic romance would have made its appearance. We have been thinking of the realistic novel as being made up of a series of Milesian tales. We may conceive of it, however, as an expanded Milesian tale, just as scholars are coming to think of the epic as growing out of a single hero-song rather than as resulting from the union of several such songs.

To pass to another possibility, it is very tempting to see a connection between the Satirae of Petronius and the prologue of comedy. Plautus thought it necessary to prefix to many of his plays an account of the incidents which preceded the action of the play. In some cases he went so far as to outline in the prologue the action of the play itself in order that the spectators might follow it intelligently. This introductory narrative runs up to seventy-six lines in the Menaechmi, to eighty-two in the Rudens, and to one hundred and fifty-two in the Amphitruo. In this way it becomes a short realistic story of everyday people involving frequently a love intrigue and told in the iambic senarius, the simplest form of verse. Following it is the more extended narrative of the comedy itself with its incidents and dialogue. This combination of the condensed narrative in the story form, presented usually as a monologue in simple verse, and the expanded narrative in the dramatic form, with its conversational element, may well have suggested the writing of a realistic novel in prose. A slight, though not a fatal, objection to this theory lies in the fact that the prologues to comedy subsequent to Plautus changed in their character, and contain little narrative. This is not a serious objection, for the plays of Plautus were still known to the cultivated in the later period.

The mine gives us still more numerous points of contact with the work of Petronius than comedy does.⁴ It is unfortunate both for our understanding of Roman life and for our solution of the question before us that only fragments of this form of dramatic composition have come down to us. Even from them, however, it is clear that the mime dealt with everyday life in a very frank, realistic way. The new comedy has its conventions in the matter of situations and language. The matron, for instance, must not be presented in a questionable light, and the language is the conversational speech of the better classes. The mime recognizes no such restrictions in its portrayal of life. The married woman, her stupid husband, and her lover are common figures in this form of the drama, and if we may draw an inference from the lately discovered fragments of Greek mimes, the speech was that of the common people. Again, the new comedy has its limited list of stock characters—the old man, the tricky slave, the parasite, and the others which we know so well in Plautus and Terence, but as for the mime, any figure to be seen on the street may find a place in it—the rhetorician, the soldier, the legacy hunter, the innkeeper, or the town crier. The doings of kings and heroes were parodied. We are even told that a comic Hector and Achilles were put on the stage, and the gods did not come off unscathed. All of these characteristic features of the mine remind us in a striking way of the novel of Petronius. His work, like the mime, is a realistic picture of low life which presents a great variety of characters, and shows no regard for conventional morals. It is especially interesting to notice the element of parody, which we have already observed in Petronius, in both kinds of literary productions. The theory that Petronius may have had the composition of his Satirae suggested to him by plays of this type is greatly strengthened by the fact that the mime reached its highest point of popularity at the court in the time of Nero, in whose reign Petronius lived. In point of fact Petronius refers to the mime frequently. One of these passages is of peculiar significance in this connection. Encolpius and his comrades are entering the town of Croton and are considering what device they shall adopt so as to live without working. At last a happy idea occurs to Eumolpus and he says "why don't we construct a mime?" and the mime is played, with Eumolpus as a fabulously rich man at the point of death, and the others as his attendants. The role makes a great hit, and all the vagabonds in the company play their assumed parts in their daily life at Croton with such skill that the legacy hunters of the place load them with attentions and shower them with presents. This whole episode in fact may be thought of as a mime cast in the narrative form, and the same conception may be applied with great plausibility to the entire story of Encolpius.

We have thus far been attacking the question with which we are concerned from the side of the subject-matter and tone of the story of Petronius. Another method of approach is suggested by the Menippean Satire,⁵ the best specimens of which have come down to us in the fragments of Varro, one of Cicero's contemporaries. These satires are an olla podrida, dealing with all sorts of subjects in a satirical manner, sometimes put in the dialogue form, and cast in a melange of prose and verse. It is this last characteristic which is of special interest to us in this connection, because in the prose of Petronius verses are freely used. Occasionally the verses in the Satirae are quoted from another source, but usually they are the compositions of Petronius himself. If it were not aside from our immediate purpose it would be interesting to follow the history of this prose-poetical form from the time of Petronius on. After him it does not seem to have been used very much until the third and fourth centuries of our era. However, Martial in the first century prefixed a prose prologue to five books of his Epigrams, and one of these prologues ends with a poem of four lines. The several books of the Silvae of Statius are also preceded by prose letters of dedication. That strange imitation of the Aulularia of Plautus of the fourth century, the Querolus, is in a form half prose and half verse. A sentence begins in prose and runs off into verse, as some of the epitaphs also do. The Epistles of Ausonius of the same century are compounded of prose and a great variety of verse. By the fifth and sixth centuries a melange of verse or a combination of prose and verse is very common, as one can see in the writings of Martianus Capella, Sidonius Apollinaris, Ennodius, and Boethius. It recurs again in modern times, for instance, in Dante's La Vita Nuova, in Boccaccio, Aucassin et Nicollette, the Heptameron, the Celtic Ballads, the Arabian Nights, and in Alice in Wonderland. A little thought suggests that it is not an unnatural medium of expression. A change from prose to verse or from one form of verse to another suggests a change in the emotional condition of a speaker or writer. We see that clearly enough illustrated in tragedy or comedy. In the thrilling scene in the Captives of Plautus, for example, where Tyndarus is in mortal terror lest the trick which he has played on his master, Hegio, may be discovered and he be consigned to work in chains in the quarries, the verse is the trochaic septenarius. As soon as the suspense is over it drops to the iambic senarius. If we should arrange the commoner Latin verses in a sequence according to the emotional effects which they produced, at the bottom of the series would stand the iambic senarius. Above that would come trochaic verse, and we should rise to higher planes of exaltation as we read the anapaestic, or cretic, or bacchiac. The greater part of life is commonplace. Consequently the common medium for conversation or for the narrative in a composition like comedy made up entirely of verse is the senarius. Now this form of verse in its simple, almost natural, quantitative arrangement is very close to prose, and it would be a short step to substitute prose for it as the basis of the story, interspersing verse here and there to secure variety, or when the emotions were called into play, just as lyric verses are interpolated in the iambic narrative. In this way the combination of different kinds of verse in the drama, and the prosimetrum of the Menippean Satire and of Petronius may be explained, and we see a possible line of descent from comedy and this form of satire to the Satirae.

These various theories of the origin of the romance of Petronius—that it may be related to the epic, to the serious heroic romance, to the bourgeois story of adventure developed out of the rhetorical exercise, to the Milesian tale, to the prologue of comedy, to the versemelange of comedy or the mime, or to the prose-poetical Menippean satire—are not, of necessity, it seems to me, mutually exclusive. His novel may well be thought of as a parody of the serious romance, with frequent reminiscences of the epic, a parody suggested to him by comedy and its prologue, by the mime, or by the short cynical Milesian tale, and cast in the form of the Menippean satire, or, so far as subject-matter and realistic treatment are concerned, the suggestion may have come directly from the mime, and if we can accept the theory of some scholars, who have lately studied the mime, that it sometimes contained both prose and verse, we may be inclined to regard that type of literature as the immediate progenitor of the novel even in the matter of external form, and leave the Menippean satire out of the line of descent. Whether the one or the other of these explanations of its origin recommends itself to us as probable or not, it is interesting to note, as we leave the subject, that, so far as our present information goes, the realistic romance seems to have been the invention of Petronius.

Notes

¹ Cf. Schmid, "Der griechische Roman," Neue Jahrb., Bd. XIII (1904), 465-85; Wilcken, in Hermes XXVIII, 161 ff., and in Archiv f Papyrusforschung I, 255 ff.; Grenfell-Hunt Fayam Towns and Their Papyri (1900) 75 ff. and Rivista di Filologia XXIII, 1 ff.

² Some of the important late discussions of the Milesian tale are by Burger Hermes (1892) 351 ff.; Norden Die antike Kunstprosa II, 602, 604 n.; Rohde Kleine Schriften II, 25 ff.; Burger Studien zur Geschichte d. griech. Romans I (Programm von Blankenburg a. H. 1902); W. Schmid Neue Jahrb. f d. klass. Alt. (1904) 474 ff.; Lucas, "Zu den Milesiaca des Aristides," Philologus 61 (1907), 16 ff.

³ On the origin of the prosimetrum cf. Hirzel Der Dialog 381 ff.; Norden Die Antike Kunstprosa 755.

⁴ Cf. Rosenbluth Beitrage zur Quellenkunde von Petrons Satiren (Berlin, 1909).

⁵ This theory in the main is suggested by Rohde Der griechische Roman² 267 (Leipzig, 1900) and by Ribbeck Geschichte d. rom. Dichtung² III, 150.