Preferred Citation: Rosand, Ellen. Opera in Seventeenth-Century Venice: The Creation of a Genre. Berkeley:  University of California Press,  c1991 1991.

2—Dramma per musica : The Question of Genre

Division into Acts

The division of their dramas into acts seems to have been a much simpler proposition for the librettists than adherence to the unities. To begin with, they had only two choices: three acts or five. Classical precedent as articulated by Horace was unambiguous: five-act division was the accepted norm for ancient drama.[30] But strong competition was readily available from the living tradition of the cornmedia dell'arte as well as from Spanish drama, with their three-act structures.[31] By its very simplicity, in fact, the choice between five or three acts focused the librettists' dilemma with particular clarity. It demanded a commitment, one that could not be hedged. Whatever the decision, it was an immediate confession that classical precedent either was or was not being followed. Given librettists' alleged discomfort with the necessity of abusing the unities, it is surprising how few of them chose the five-act format.

We have considered the defense by the librettist of Le nozze d'Enea of his five-act format on the basis of verisimilitude: five acts offer twice as many intermissions as three, hence twice as many opportunities for the audience to imagine the passage of time. It is interesting that this author does not even attempt to cite classical authority for his choice, but somewhat sheepishly acknowledges the modern preference for three acts. Busenello, on the other hand, justifies the five acts of his Giutio Cesare (1646) with a perfunctory bow to classical precedent: "If the acts are five and not three, remember that all ancient dramas, and especially the tragedies of Seneca, are divided into five acts" (Appendix I.12a). But Giulio Cesare happens to be Busenello's only five-act libretto.[32] And, as we have seen, he shows no compunction in dismissing classical precedent in the prefaces to his other librettos—and even in the preface to

[30] Horace, Ars poetica (The Complete Works of Horace , ed. Casper J. Kraemer, Jr. [New York, 1936]), 403: "A play which is to be in demand and, after production, to be revived, should consist of five acts—no more, no less."

[31] All of the scenarios in Flaminio Scala's Teatro delle favole rappresentative (Venice, 1611), for example, are in three acts. Significantly, however, most of the literary dramas by the same and other authors, which were designed to be read rather than staged, are in five acts. See, for example, Flaminio Scala, Il finto marito (1618) in Commedie dei comici dell'arte , ed. Laura Falavolti (Turin, 1982), 215-365. A number of early Roman librettos, themselves rather strongly influenced by the Spanish tradition, were also in three acts. These include La rappresentatione di Anima e di Corpo (1600) and Eumelio (1606), as well as Giulio Rospigliosi's Erminia sul Giordano (1633) and Stefano Landi's Sant'Alessio (1634). The "rules" of Spanish drama in the seventeenth century are articulated in Lope de Vega, Arte nuevo de hater comedias en este tiempo (Madrid, 1609). Lope himself abandoned the "rules" of tragedy, including five-act division and the "unities," in his plays. In the letter to Cicognini mentioned in n. 27 above, for example, he recommended the ordering of "actions . . . to cover the space not just of a day, but of many months, even years."

[32] And despite contrary evidence in the libretto itself (allusions in the preface and prologue), it may also be the only one never set to music. It was certainly not performed in 1646, the season for which it was intended, since all of the Venetian opera houses were closed in that year. See Walker, "Errori," 16.


Giulio Cesare with respect to other issues. Busenello reveals his true nature here, and the true functioning of his education. For him and his fellow academicians, the entire arsenal of precedent existed to be used or discarded as needed. And the needs changed frequently. What passed for legitimate one day failed the next. In any case, five-act librettos were exceedingly rare in Venice, even in the early years of generic insecurity.[33]

Three acts may have been the accepted norm for modern opera librettos (as they were for Spanish drama and cornmedia dell'arte ), but at least one of our early authors was moved to justify his choice. In the preface to his first libretto, Delia (1639), Giulio Strozzi claims three-act division as natural: "I have divided the work quite deliberately into three acts, a division common to all things: beginning, middle, and end"; and he defends it against the silent proponents of the "classical" five acts by distinguishing between ancient drama and his own work: "The ancients made five in theirs, because they interspersed them with singing [i.e., choruses]. This work, being wholly sung, has no need of so many pauses" (Appendix I.15e).[34]

After the early 1640s, the three-act division, a clear bow to modern taste, became completely conventional for dramma per musica ; five acts, however, remained the norm for spoken dramas.[35] The issue did not arise again until the end of the century, when a few of the most radical neoclassicizing librettists, especially Frigimelica Roberti, but also Zeno, used five-act division as an emblem of their orthodoxy.[36]

[33] Among the few were Paolo Vendramin's Adone (1640), the anonymous Le nozze d'Enea (1641), and Badoaro's Il ritorno d'Ulisse in patria (1640) and L'Ulisse errante (1644)—but the score of Ritorno is in three acts and the other operas may also have been altered; we cannot know since their scores are lost. For a brief discussion of the differences between the manuscript librettos of Il ritorno d'Ulisse and the single extant score, see Wolfgang Osthoff, "Zu den Quellen von Monteverdis Ritorno di Ulisse in patria," Studien zur Musikwissenschaft 23 (1956 ): 67-78.

[34] In defending his division into acts, Strozzi implicitly and incidentally acknowledged his belief that ancient drama was only partly sung, and therefore not a model for him. His choice of the term azzione rather than the more common atto , however, smacks of self-conscious classicism that is undermined by the tripartite division. Such classicizing or pseudo-classicizing word choices were typical of Strozzi. In a similarly oblique acknowledgment of classical precedent, he ostentatiously gave the three acts of his next two librettos, La finta pazza (1641) and La finta savia (1643), Greek labels: "Protasi," "Epitasi," and "Catastrofe," thereby implying an awareness of ancient practice that would be belied by the three-act division itself. (These terms appear in Donatus's fourth-century commentary on Terence, i, 22, 27; see W. Beare, The Roman Stage [London, 1964], 217 and ch. 25, "The Roman Origin of the Law of Five Acts.") In another instance, Strozzi dignified the un-unified place of his Proserpina rapira with a Greek term, anatopismo , defined in the libretto as "un error di luoghi havendo qui per vaghezza il Poeta congiunto insieme il Lago di Pergusa, il monte d'Etna, il promontorio Pachino, etc."

[35] All of the plays written by Giacomo Castoreo for the Teatro ai Saloni were in five acts (with musical intermedi). Castoreo's librettos, however, were in three acts.

[36] Most of Frigimelica Roberti's librettos have five acts. They and other five-act librettos by such poets as Zeno and Pariati were a specialty of the Teatro S. Giovanni Grisostomo; see ch. 13 below.


2—Dramma per musica : The Question of Genre

Preferred Citation: Rosand, Ellen. Opera in Seventeenth-Century Venice: The Creation of a Genre. Berkeley:  University of California Press,  c1991 1991.