Article: Music and Structure in Roman Comedy

Well over a century ago, Friedrich Ritschl and Theodor Bergk independently reached the same conclusion regarding the markings of DV and C in some of the manuscripts of Plautus: the initials stand for diverbium and canticum; and their association, respectively, with scenes in iambic senarii and scenes in other meters implies that in Roman comedy passages in iambic senarii were unaccompanied, whereas passages in all other meters were cantica, accompanied by the tibiae (Ritschl 1871–72, Bergk 1872).¹ Ritschl’s and Bergk’s conclusions provide the best explanation not only of the rubrics in the manuscripts, but also of descriptions of performance in the grammarians and other ancient authors, and several allusions to the tibicen in the plays of Plautus. It is therefore not surprising that while there have been a few dissenters, most scholars have agreed with Ritschl’s and Bergk’s basic distinction between iambic senarii and other meters based on accompaniment.² The implications of that distinction for the structure of Roman comedy, however, have remained unappreciated.

Not that meter has been ignored by those proposing structural rules for Roman comedy. Soon after Ritschl and Bergk presented their discoveries, Andreas Spengel (1877) argued that arrangement of lyric passages, accompanied stichic verse, and iambic senarii contributed to a five-act structure in all the plays of Plautus. Much more recently, Adolf Primmer has contended that all Roman comedies can be divided into three or more acts comprised of combinations of accompanied and unaccompanied meters (1984, 16–20). Although both Spengel and Primmer offer some valuable insights into the metrical arrangement of the plays, two factors vitiate their theories. First, in their determination to find universal structural principles, both scholars try to force the metrical patterns of many plays into procrustean schemata, ignoring passages which do not fit the molds they have created.

Second, both scholars assume regular division of the plays of Roman comedy into acts divided by musical interludes. There is indeed one passage where a musical interlude certainly occurred (Pseud. 573a; see below), and in other passages the exit and reentrance of a character would make such an interlude likely (Cistell. 630, Trin. 601). Spengel, Primmer, and others who have assumed regular musical interludes have postulated that these occurred regularly when the text suggests that the stage was empty, and temporal verisimilitude would demand that an interval should pass before one of the characters reenters.³ Given the freedom with which Plautus and Terence apparently ignore the constraints of temporal verisimilitude even when the stage is not left empty, however, any attempt to prove the existence of interludes based on how much time must pass between events on stage divided by an empty stage is doomed to failure; and as Conrad, Questa and others have shown, there is much evidence to suggest that with a very few exceptions Roman comedies were performed continuously, without act breaks.⁴

In her study of Plautus’ polymetric cantica Helen Hull Law approaches musical structure from another perspective. She observes that a great number of Plautus’ plays feature regular patterns of iambic senarii, followed by polymetric cantica, followed by trochaic septenarii; and that on many occasions polymetric cantica occur at the beginning of new units of action (1922, 103–5). Law’s observations are invaluable for an appreciation of the musical structure of Plautus’ plays; nevertheless, they are limited by her emphasis on the polymetric cantica.

It is not at all clear how the polymetric cantica were performed. Most scholars have assumed that polymetric passages were sung, whereas the performance of stichic verse other than iambic senarii was either “melodramatic,” with actors speaking independently of the accompaniment, or “recitative,” in which actors chanted in time with the tibiae.⁵ There is, however, no evidence that the polymetric cantica were “sung” in our sense of the word, or that the method in which the polymetrics were performed differed from the performance of other passages labeled cantica in the manuscripts.⁶ The most important musical distinction therefore appears to have been not that between the polymetric cantica and stichic meters, but rather that between unaccompanied iambic senarii...