April 22, 2014
Program notes by Martin Pearlman
Tonight’s opera is one of only three by Monteverdi that have come down to us. L’Orfeo from 1607, his very first, is generally acknowledged to be the first great opera in the repertoire. Then, after a gap of 33 years, during which Monteverdi wrote many operas that are tragically now lost, we have two great works from near the end of his life: Il ritorno d’Ulisse (1640) and L’incoronazione di Poppea (1642).
Il ritorno d’Ulisse is based on the last part of Homer’s Odyssey, which tells of the homecoming of Ulysses after an absence of 20 years and of his slaying of his wife’s suitors, who have taken over his palace. Giacomo Badoaro’s libretto, set to music by the 73-year-old Monteverdi, was premiered to great acclaim in 1640 during the carnival season in Venice, and, because of its success, it was revived in the following season, an unusual distinction for an opera of the time. The performances took place at one of Venice’s newly opened public opera houses. Not only were production budgets severely limited in these public institutions, but writing for a broader public affected the kinds of stories that were set to music. In this case, the story was one familiar to the audience, with plenty of blood and gore, a far cry from the nymphs and shepherds of earlier court productions.
The opera then dropped from view until the late 19th century, when a manuscript was rediscovered in Vienna, perhaps from a later production that may have taken place there. But it was not until the mid-20th century when further documents were found, that it was proven to be a genuine “lost” opera of Monteverdi.
Our performing version
Of Monteverdi’s three surviving operas, Il ritorno d’Ulisse is certainly the least well-known and least performed. One reason for this undoubtedly has to do with the nature of the surviving material. The music for this opera survives in only one manuscript score, although a number of copies of the libretto have been found. We have nothing of this opera in the composer’s own hand, only the one score hastily and sometimes carelessly written out by a copyist, probably after the composer’s death. It is lacking many details (some of which may have been explained to the performers in rehearsals), it is incomplete in places, and it has a number of small errors. It is clearly a working copy designed for a particular production. Several scenes from the libretto are missing in the score, having either been lost, cut from that production, or perhaps never set to music in the first place. A performance therefore requires many decisions to fill in the holes in what the manuscript tells us.
For Boston Baroque’s production, I have checked every note and word of the surviving manuscript, a process which led to countless small adjustments — plus a few major ones– to the existing modern editions. In many places, where the manuscript is incomplete or unclear, it required a variety of interpretive decisions. In the middle of Act I, for example, there is a written instruction that the orchestra should play a brief sinfonia while the sleeping Ulysses is carried in. We are instructed that, “so as not to wake him,” the sinfonia should be played quietly and be limited to only one chord. However, there is no music provided — only a bass “C” to tell us what the unchanging harmony should be; it is left to us to create a brief interlude. In other places, only a bass line is given for an instrumental piece, and one must add the other parts.
One important revelation in the manuscript occurs at the point where Ulysses slays the suitors. Just as the orchestra begins to build momentum for that climactic moment, most editions and performances have Ulysses interrupt the rhythmically driving music to insert a line of recitative as a prayer to Minerva. I have always found it dramatically weak to stop the forward motion, in order to insert the one line prayer and then resume the rhythmic music. Looking at the manuscript, one sees that this extra line of music (for which the words are not in the libretto) appears to have been added later as a footnote at the bottom of the page, suggesting that it is probably not original. Our performance therefore omits this insert, allowing the rhythmic momentum to build to the end of the scene.
Instruments in our version
The manuscript score does not specify the instruments that should be used. The five-part ritornelli, or musical interludes, would almost certainly be for strings, with some other instruments possibly added at times for color. For most of the opera, though, the score gives only two staves of music: a vocal line and an instrumental bass line below it. It is left to the performers to decide how to harmonize the bass line and to decide which instruments should play it. We use a variety of continuo instruments for this purpose — two harpsichords, an organ, two theorbos, cello, dulcian (early bassoon) — to vary the sonority according to the dramatic situation.
Probably the biggest difference among performing versions of Ulisse is in the matter of orchestral accompaniments. In the original score, the orchestra plays very little. It has mostly extremely short instrumental interludes (some as brief as 10 seconds). Beyond that, it accompanies singers only in a few limited spots, most notably at the moment when Ulysses slays the suitors and in Penelope’s song of joy in the final scene of the opera. All of this comes to less than 15 minutes of playing out of a full length opera. The rest of the score is written for singers without the ensemble, accompanied only by a continuo bass line.
The question then is whether the manuscript score is complete, or whether instruments were meant to accompany singers in places where there is no music specified for them. Every production must address this issue. Some composers (e.g. Dallapiccola and Henze) have orchestrated the work throughout, giving it something closer to a 19th-century operatic sound. Not only does this change the basic character of Monteverdi’s work, but it also makes it impossible for the singers to be rhythmically free in declaiming their text — and it restricts the ability of the continuo players to improvise and react to the singers as they are meant to do in this music. At the other extreme, there are performances that limit themselves strictly to the notes that are written down, so that the orchestra plays very little and almost never accompanies singers. To me, this last choice seems unnecessarily austere, of questionable authenticity, and even somewhat timid. To have the ensemble sit silent for over 90% of the opera would have been as artistically and financially wasteful in the 17th century as it would be in the 21st. There are, of course, other performances that fall somewhere between these two extremes.
My version for Boston Baroque is in this middle ground. I have composed orchestral parts to accompany the singers at certain key moments of heightened drama, when a character breaks out of recitative into song. For the most part, these are simple accompaniments, designed not to interfere with the singers. The sonority of the ensemble varies according to the dramatic moment, sometimes using only low strings, sometimes solo violins, sometimes the full five-part string ensemble, and occasionally adding recorders or cornetti.
There are plenty of hints to support this approach. In this manuscript, there are places where a few interpolated notes appear to be cues for instruments that are meant to be playing, even though there is no music written for them. There are models for this in some other operas of the time, where we do see written-out parts for the instruments to accompany singers. There are even some operas that give instructions for an aria to be played “with violins” or “with all the instruments,” even though no instrumental parts are shown in the score.
Orchestral accompaniments like these can heighten moments of true song. But the core of this music is still in the freer speech patterns of recitative accompanied by a continuo section that can follow the singers with an improvised accompaniment.
The libretto vs. the musical score
There are many places in this opera, where the libretto differs from the surviving musical score. Sometimes a few words are different; sometimes an entire scene in the libretto is not found in the musical manuscript. For those scenes, have we lost music, or did Monteverdi never set them to music in the first place?
Wherever possible, I have chosen to follow the musical score, rather than the libretto, since librettos of that time did not always reflect the finished product. Librettists often considered their work to be independent poems and sometimes retained their original material, even after a composer may have altered or omitted some of it in his opera.
Monteverdi’s letters show that he was quite demanding of his librettists and often required changes. His primary interest, he said, was to portray the range of human emotions. The “missing” mythological scenes may well never have been set to music, because he wished to keep the opera focussed on the human drama. Similarly, “missing” choruses of nereids and sirens or of underworld shades were, in all likelihood, never set to music. Not only would they have shifted the focus of the opera away from the strong emotions of the human characters and even of the gods, but choruses were also not a common feature in the cash-strapped public opera houses of Venice at the time.