In the Captives of Plautus the audience knows that Hegio has his own son before him in chains, and notices that, not recognizing his son, he causes him much suffering. Throughout the play the attentive spectator will watch for the very effective incongruities that arise from the father’s ignorance. In the Rudens, Daemones, not knowing that the girl who is trying to escape from shipwreck and the slave-dealer is his own daughter, at first seems to the informed spectator extremely insensitive to her suffering. The father’s sympathies are aroused only indirectly by his religious respect for a suppliant at the altar, then by the accident of being called in to arbitrate regarding her basket of birth tokens. Only when he has established her civil status by this accidental judgment does he learn the truth. In a more farcical form comic irony is freely used in the plays of self-deception, for example in the case of a bragging coward like the Miles Gloriosus, or in plays depending upon mistaken identity or some similar delusion, as in the Menaechmi and the Amphitruo. And in very nearly all the plays of Plautus, if it be not the chief mainstay of the plot, it appears at least here and there.

The new fragments of Menander prove that Menander had frequently constructed his plays with this effect in mind. Indeed it is the decisive factor in all that are extensive enough to permit of analysis. In the Arbitrants Smicrines all unconsciously arbitrates against his own child. In the Samia the old man is misled by a chance remark into the belief that his son has betrayed him, and the resulting irony runs through the central part of the play. And even when the facts are disclosed the son immediately sets going another series of misunderstandings (disclosed beforehand to the audience, line 432) by threatening to go into exile. The Perikeiromene is built about the same device. Two men are in love with the same maiden. One is her unrecognized brother, the other is jealous of her attentions to the former. The girl knows that the former is her brother but may not reveal the fact. However, the deity, Agnoia, has informed the spectators of the relationship so that they are in a position to view the intricate play at cross-purposes, but there is little of what we should call suspense because they have also been informed that a recognition scene will end the play satisfactorily.

In the Hero the expository prologue is lost, but we know that the prologue was the omniscient Heros, who adequately prepared the audience for what was to follow.[8] Here a husband (unrecognized), his wife, their “exposed daughter,” not yet known as such, and two lovers of this daughter, one a slave, the other a rich neighbor, all enter a tangle of delusion to which the audience has the key. In the Georgos a man expresses to a woman his desire to marry her daughter. He does not know that he is the girl’s father. As the woman—bound to secrecy—stands wringing her hands in despair, the audience—apparently informed of the secret—experiences a situation as poignant as that of the Oedipus. And finally in the Petrograd fragment of the Phasma we find a mother’s furtive visits to her daughter, born out of wedlock, and an entangled love affair, a situation which again involves the use of irony, since a fragment from the prologue shows us that the audience has been informed in advance.

It would be hazardous to say that Menander always lets the spectator into the secret beforehand so as to make use of dramatic irony, but it is striking that he does so in every instance where we possess enough of his plot in the original to test his methods. It is apparently his usual method of procedure. This is also in accord with his well-known predilection for Tyche, the counterpart in comedy of tragic fate. We need not suppose, as has often been done, that his constant reference to Tyche springs from his own philosophical doctrine. Such well-known passages as “Chance holds the helm; mortal forethought is but a delusion” (Frag., Kock, 482), etc., are, of course, comments of characters in the play. They need not be expressions of the dramatist’s own creed. But such comments would naturally come frequently in plays built on the conventional tragic form which required that the players grope their obscure way through the action in front of an audience which knew the end.

Now these observations are not meant as an attempt to rehabilitate Leo’s doctrine that the New Comedy merely borrowed all its devices of prologue, fate, recognition, and the rest from tragedy. Prescott’s incisive criticism[9] of that view must stand, with its insistence that we take Sicilian antecedents, Aristophanes, and environment into account. The new comedy was hardly as helplessly unoriginal as Leo held. The problem we have raised should rather be approached from the viewpoint of what the spectator expected and desired. It did not necessarily arise in the construction of plotless farces, in the presentation of ludicrous situations, buffoonery, and scenes centering about comical and preposterous characters. When, however, the plot was involved and a long consistent story was to be unraveled, the spectators, who knew nothing of the story, desired to be put at the same point of vantage early in the play that they naturally enjoyed when an Oedipus, a Medea, or an Orestes was presented.

When we turn to the Roman stage we seem to discover an attempt to break away from this convention, if not in Plautus[10] at least in Terence. We do not find conclusive evidence that Plautus seriously changed the construction of the Greek plots which he used except to remove the choral interludes and turn the plays into musical comedies, though it is likely that he usually avoided plays that had very intricate plots, and chose freely from those that contained laughter-producing situations. There is no evidence that he sought for suspense, or revamped any of his originals in order to attain it.

Terence, however, despite his fondness for the Greek originals and his outspoken claim of fidelity to them, seems consciously to have striven for a suspended dénouement. He does not entirely suppress dramatic irony, but he reduces its scope, he eliminates the expository prologue completely, he is chary about giving information to the spectator, preferring to keep him under tension for a part if not for the whole of the play.

A brief reference to the Adelphoe, his last play, will best reveal his procedure. Here two brothers employ different methods in bringing up their sons. Micio, who has adopted one of Demea’s sons, is indulgent, Demea is severe. Both boys enjoy themselves, Micio’s confessedly, Demea’s secretly. In fact the latter throws the burden of his escapades on Micio’s son. Hence when the action begins (1. 182) Demea is found scolding Micio because Micio’s son is setting a bad example to his supposedly virtuous brother. This is completely in Menander’s ironic style, for, as we shall see, Menander in the original had a prologue informing the audience that Demea’s son was the rascal of the two boys.

In Terence’s version, however, there is no expository prologue; the audience does not yet know the secret that Demea’s son is the source of the mischief. The irony is not wholly lost to those who have a good enough memory to recall half an hour later how misplaced Demea’s rebukes actually were. Terence is accumulating effects by suspending the revelation which Menander gave at once. But he goes even further in increasing tension. The prologue of the original had explained the bold deed that started the action, namely, that Micio’s son, in order to aid his brother, had forcibly taken from the slave-dealer the girl whom his brother loved but had not the money to buy or the courage to steal. That fact had to be presented somehow, so Terence, according to Donatus, inserts a scene in Act II which conveys the desired impression. Characteristically Terence still withholds the crucial fact that the boy is committing this crime not for himself but for his brother. Perhaps the shrewder spectators would suspect the truth. In Menander’s play they knew it from the first and laughed at Demea’s misplaced boasts. In Terence’s adaptation, however, they continue in doubt. It is not till a fourth of the play is over that Terence solves this mystery. He holds it back so long indeed that there is danger that the spectator may go too far on a mistaken clue. After the revelation, however, the audience, acquainted with a situation that Demea still fails to comprehend, can proceed for several scenes to enjoy the dramatic irony involved in this circumstance.

But Terence has one more surprise in store at the very end, to which Donatus again supplies the clue for us. At the end of the play Menander had suggested a partial conversion of Demea, while Micio went smiling to the final scene. Not so Terence. Writing for a more puritanic Roman audience, he felt the need of giving an appreciable rebuke to Micio for his lack of principle, and hence compelled him by way of consistency in his easy generosity to marry an unattractive widow.[11] In other words, with a minimum of changes in his paraphrase, Terence, without greatly reducing the dramatic irony inherent in the separate scenes, has so adapted a standard Menandrian plot based upon self-delusion (for which the spectator is prepared) that the elements of suspense and surprise have become vital factors. This seems to me to be Terence’s favorite procedure.