Divas and Scholars

by Philip Gossett

  Modern instrumentalists who employ historical instruments (and the conductors who lead them) cannot be indifferent to the limitations of many of these instruments. Some of these limitations are superable, whereas others must simply be accepted, just as they were accepted by composers who worked during the first half of the nineteenth century (and before). Without an awareness of these instrumental limitations, many details of contemporary musical notation are incomprehensible. Although each instrument poses its own characteristic problems, it is worth considering a few of them from different families as representative of the kinds of problems that arise when editing and performing Italian opera: double basses, oboes, timpani, and horns.

  Double Basses

  Italian double bass players normally employed an instrument with three strings, tuned as A–d–g (all notes for the double bass, by convention, are written an octave higher than they sound). Instrumentalists in France, instead, would have been using an instrument with four strings, with the additional lowest string tuned as E (sounding E1).15 In those many passages in operas written for Italy in which the violoncellos and the double basses play together (either with both parts written out in full or with the cellos told to play col basso—“with the bass”), they normally produce sounds an octave apart. But when the cellos play in their lowest register (from C through A), the three-stringed double bass tuned in the normal fashion cannot follow them, and hence must play the notes an octave higher, so that the sounding notes in that lowest register of the cellos are actually in unison with the notes in the double basses. No one had to be told this in the notation: everyone understood the convention, and double bass players made the adjustment automatically, using their musical sense to determine the point at which it would be best to change from one register to the other. But it would never have occurred to Rossini, when he revived for Paris operas originally written for Italian theaters, to require French double bass players to adopt an Italian three-string double bass. Rather, he must have been delighted to have the double basses at the Parisian Théâtre Italien remain in the lower register, an octave below the cello, as long as possible.16

  Still, if one examines Rossini’s 1824 Duetto for violoncello and double bass, written for the famous double bassist Domenico Dragonetti, at no point is the performer asked to play a note lower than written A, the lowest normal pitch on the instrument.17 That is not quite the case with the Sei sonate a quattro, written by Rossini in 1804 for two violins, violoncello, and double bass; but there is still no reliable edition of either of these pieces, so we cannot be certain what the composer actually wrote.18 In the modern edition, however, there are a few places where the double bass is asked to descend below A, suggesting either a scordatura (retuning of the lowest string) or the presence of notes Rossini wrote for musical reasons but expected to have changed by a performer using the standard three-string double bass. In the second sonata, for example, in the Andantino, there is a wayward G (sixteenth note) at m. 3, a passage in which all instruments are playing together; it was probably omitted in performance. In the third sonata, in C major, there are several Gs, suggesting a possible scordatura, with the low A tuned down a tone.19

  These practical problems of range and notation can have implications for written scores. In Verdi’s Stiffelio (1850), for example, in the climactic duet of the opera (N. 9), Lina, having accepted the divorce Stiffelio has demanded as the consequence of her adultery, turns on him in an E major passage to the text “Non allo sposo volgomi, ma all’uomo del Vangelo” (I am turning not to my husband, but to a man of the Gospel). This is the passage that concludes with the infamous parole sceniche “Ministro, confessatemi” (Minister, hear my confession).20 Throughout this passage (mm. 170–188), Verdi wrote the violoncellos and double basses on the same staff, since the normal cello staff is occupied by a part for a single solo cello. The cellos play in the narrow range between A and e. The double basses are notated below the cellos, beginning on the notated pitch E, but descending as low as B1 before concluding on F. All of these notes are below the range of the standard, three-string Italian double bass of the period, and many are below the compass of the standard, four-string French double basses. In these measures it seems certain that Verdi—most unusually—has notated the pitches for the double basses as they sound, rather than an octave above the sounding pitches. In other words, where he wrote E, the double bass must actually play the pitch normally written as e, and so forth. Immediately after (from m. 189), Verdi’s notation returns to its normal form, and the double bass part is written an octave above the sounding pitches.21

  The problems with the “period” double bass could not only affect the notation, but could also require retuning the lowest string of the instrument to extend the downward range, what I have referred to above as scordatura. In Il trovatore (1853), Verdi expected the Italian double bass players for whom he composed the opera to be using instruments with three strings. When he wanted a sounding A1 during the Miserere, he wrote in his autograph manuscript: “The double basses will lower the A string by half a tone,” a scordatura that would make no sense to performers using a four-string double bass.22 The critical edition therefore omits the indication in the score, consigning it instead to a footnote and discussing the problem in the commentary.

  Oboes

  Rossini was cognizant both of the expressive potential of the oboes with which he worked, and of their technical limitations: indeed, he took those limitations into account when notating his scores. The first half of the nineteenth century was a period in which the construction of oboes was undergoing many significant changes, with the proliferation of keys replacing the basic system of the late eighteenth-century oboe: finger-stopped holes, with only a small number of keys.23 On some occasions, the composer was therefore constrained to make musical choices that are little short of ludicrous. A perfect example occurs in the overture to La pietra del paragone of 1812, an overture that subsequently passed without change into the 1813 Tancredi. This overture already has a fully developed Rossini “crescendo,” an obligatory part of all the composer’s mature overtures, present in both the exposition (in the key of the dominant) and in the recapitulation (in the key of the tonic).24 As always, a Rossini crescendo involves more than a simple increase in dynamic level. It is a structured manipulation of dynamics, instrumentation, register, and articulation to create a powerful drive toward the final cadences of both the exposition and the recapitulation of the overture. Usually a four-measure phrase is played three times, and sometimes a shortened version of this phrase continues the effect for a few additional measures.

  In Tancredi, the oboes enter only after the first four-measure phrase of the crescendo in the dominant key, joining other winds and the upper strings in the second statement of the phrase. For the third statement, Rossini tightens the instrumental notch one degree further and takes both the flutes and the oboes into a higher register. Were he to have done this literally, however, it would have been necessary for the first oboe to ascend to a high e (e’”), a note beyond the normal compass of Italian oboists in the 1810s.25 Thus, Rossini modified the part, substituting a b” for the impractical e’”. In the last two measures of the crescendo, which further intensify the harmonic rhythm, he actually wrote an eighth rest where the e’” should fall (example 12.1). Not even the most faithful believer in period instruments could fail to appreciate that some advantage would be gained by using an oboe that can play reliably a high e’”.

  Timpani

  Describing early nineteenth-century timpani in the introduction to his recordings of the Beethoven symphonies, Roger Norrington writes, “Small and beaten with hard sticks, [they] sound as if they have come straight from the field of Waterloo.”26 But the situation with “period” timpani is more complicated than Norrington’s enthusiasm would lead us to believe. Every modern performer of Italian opera, for example, knows that through the first six or seven decades of the century, Italian composers expected to find only two kettledrums in the orchestr
a, usually tuned to the tonic and dominant of the prevailing tonality.27 They were considerably smaller than modern instruments: the drumheads of surviving examples from the first half of the nineteenth century studied by Renato Meucci are all less than 55 centimeters in diameter, as opposed to the dimensions specified in Grove’s for a pair of modern timpani (71 cm. and 63–65 cm.).28 In the absence of mechanisms to permit a rapid modification in the tension of the drumhead, it was more difficult for the early nineteenth-century timpanist to change the pitch of his instruments—using hand screws—than is the case with modern instruments, and therefore changes of tuning required considerable time.29 Although three timpani were regularly available at the Paris Opéra by the 1830s, in Italy the practice of introducing more than two timpani did not take hold until much later in the century.30 By contrast, the timpanist in a modern pit orchestra regularly has available four instruments. Finally, the choice of materials for the construction of timpani remains a controversial matter: some modern players prefer traditional calfskin for the drumhead, others a more reliable synthetic material made of plastic. Although all agree that the resulting sound is sensibly different, some think the advantages of plastic outweigh the disadvantages, while others do not.31

  EXAMPLE 12.1. GIOACHINO ROSSINI, TANCREDI (LA PIETRA DEL PARAGONE), SINFONIA, OBOE PARTS, MM. 108—111.

  When Rossini wanted to employ timpani for an overture in D major, such as that of Tancredi (La pietra del paragone), he had available to him only two instruments, tuned to the tonic (d) and the dominant (A). When the music modulates to the dominant key during the course of the exposition, however, A major becomes the tonic. Yet the Italian timpanist had no third drum to play the dominant pitch (e) of A major, and there was insufficient time for the timpanist to modify the pitch of one of the two instruments he did have. Rossini’s compositional decisions reveal his sensitivity to the problems in various ways. Realizing that the pitches available to the timpanist were insufficient to maintain a coherent part throughout passages where the sudden disappearance of the timpani might seem jarring, Rossini did not allow the timpani to play alone during ensemble passages of this overture. Instead, he added the Banda Turca (an assortment of other percussion instruments), effectively masking the sudden disappearances of the timpani.32

  There are times, though, when he really wanted the sound of the timpani as part of the orchestral fabric, and hence was willing to compromise with respect to its pitch. At the conclusion of the exposition, for example, both at the end of the crescendo passage and in the final cadences, the music is fully settled in A major. The composer could call on the timpani to play A, the temporary tonic, but not its dominant, e. Still, it would be ugly indeed to have the timpani omit an occasional note when the chord is the dominant in the new key. Instead, Rossini asked the timpani to adopt the pitch d, now understood as the seventh degree of the dominant harmony (E–G–B–D); see example 12.2). You are unlikely ever to have heard that effect in any theater performing Tancredi with modern instruments, for no self-respecting timpanist would allow such a pitch to stand unchanged. Whether you believe the cautious statement in The New Grove that “the question of amendment is controversial” or the bolder assertion in Grove Music Online that “many timpanists add notes to their parts at will, ignoring earlier performing practices,” performers in a normal pit orchestra simply supply the e at this point of the Tancredi sinfonia by employing a third instrument.33

  EXAMPLE 12.2. GIOACHINO ROSSINI, TANCREDI (LA PIETRA DEL PARAGONE), SINFONIA, MM. 112–116.

  While Rossini generally used timpani only in larger ensembles, for Verdi the timpani are commonly employed extensively throughout much of an opera, in passages that are loud and even in passages that are soft. Thus, in Verdi situations arise regularly where the timpani play notes that are not in the prevailing chords or are not desirable notes in those chords. The concluding ensemble of the third-act finale of Ernani, for example, the triumphant appearance of the newly elected Charles V, is in F major, and the two timpani are tuned to the tonic and dominant of that key. At a crucial point, the music modulates temporarily to A major, without allowing time for the timpani to retune. As the music grows from pianissimo to fortissimo, however, Verdi is intent on including timpani in the ensemble at the climax for their percussive qualities, even if the pitches are false. He therefore continues to write the pitches c and f, although the former represents the third of the new tonic and the latter can only be understood as the ninth of an E-dominant seventh chord. It is surely true, as Fabrizio Della Seta remarks in his edition of La traviata, that “because of their approximate intonation, the dissonant sounds produced by the timpani were pulled into the sphere of the adjacent tones, with respect to which they produced an effect more like a dull beating than a true dissonance. The result would have been enhanced by the smaller dimensions of the instruments and their proportionately reduced sonority.”34 Modern tuning mechanisms allow a timpanist to move quickly between the tonic and dominant of F major and those of A major. If one is using modern instruments, nothing would be gained in performance by forbidding such an adjustment.35 But Norrington’s Waterloo metaphor captures nicely the bracing effect older timpani can have on the sound of Verdi’s more Risorgimentotinted operas.

  Horns

  The horns employed by Italian composers in the first half of the nineteenth century were natural horns, without valves, instruments that could play easily only a fundamental tone (which depended on the total length of the tube) and its natural harmonic series. To move from one such series to another, the player inserted additional tubing in the instrument, called “crooks,” thereby changing its length and producing a different fundamental with its corresponding harmonic series. Techniques were also developed in the late eighteenth and early nineteenth centuries to permit the natural horn to play chromatic notes lacking in its basic harmonic series. By placing a hand in the opening through which sound emerges, known as the bell, and manipulating the size of that opening (a technique called “hand stopping”), the player could produce a wider variety of pitches. But the resulting sounds were very unequal, and only virtuosos developed sufficient technique to overcome many of these problems. Some notes were simple to produce, however, and composers regularly entrusted orchestral musicians with these hand-stopped notes.

  The ambiguities inherent in the choice between historical and modern horns are neatly encapsulated in Rossini’s Otello. Rather than introduce Desdemona in a solo composition, the composer presents her together with her attendant, Emilia, in accompanied recitative, followed by a duettino for the two women. The recitative begins with a lengthy orchestral ritornello, which features the most elaborate horn solo in any Rossini opera.36 But it was certainly not the first. As early as La pietra del paragone (1812) and L’Italiana in Algeri (1813), Rossini adopted the horn as his solo instrument of choice for the orchestra solos introducing the cavatinas of Clarice and of Lindoro (example 12.3).37 Each begins with a lyrical four-measure phrase (moving harmonically from the tonic to the dominant), using a similar melodic gesture: an ascending sixth from the fifth degree of the scale (g) to the third (e, a sixth above). The second phrases, which duplicate the ascending melodic sixths of the first, again begin on the tonic harmony, but now close with tonic cadences. The cadential patterns of the two melodies are cut from the same cloth. Scales, with a touch of chromaticism, ascend to the high g; similar scales, descending, lead to arpeggiated tonic triads; until trills on the second degree of the scale (d), supported by dominant chords in the orchestra, resolve to the tonic. Neither in range nor in chromaticism are these melodies beyond the capabilities of a modest player of the natural horn, employing simple hand stopping.

  EXAMPLE 12.3. A) GIOACHINO ROSSINI, LA PIETRA DEL PARAGONE, CAVATINA CLARICE (N. 3), MM. 5–14. B) GIOACHINO ROSSINI, L’ITALIANA IN ALGERI, CAVATINA LINDORO (N. 2), MM. 2–9.

  The horn solo in Otello shares many of these characteristics: the lyrical opening phrase, with faint touches of chromaticism; the scales, ar
peggio, and trill at the final cadence. But the expression is everywhere intensified, and with it the demands on the player.38 Example 12.4 shows the complete horn solo as Rossini originally conceived it. The lyrical six-measure melody (with two three-measure phrases, the first leading to the dominant, the second returning to the tonic) is very much in the style of the earlier examples. It is followed, however, by four measures (14 –17), whose large leaps into the lower register make notable demands for hand-stopping on the performer. As the music approaches the final cadence, Rossini lets loose. Three measures of ascending arpeggios (24 –26) lead to a descending chromatic scale across a full octave from high g to the g above middle c, while further arpeggios bring the line down yet another octave. A two-octave leap (28) introduces a precipitous climb to the highest note in the line (high c), after which a gradual descent, with the expected trill on d, leads to a cadence on the tonic.

  None of this is beyond what horn tutors of the early nineteenth century considered feasible for an experienced player,39 but the level of virtuosity demanded—the extended range, leaps, downward arpeggios, chromaticism—all place it well beyond the level of an average orchestral musician in Italian opera houses of the time. Nowhere else in his operas does Rossini make such extraordinary demands on a horn player, and even at the Teatro San Carlo of Naples the part proved too difficult. In Rossini’s autograph the penultimate and antepenultimate measures (27–28), although fully orchestrated, were heavily crossed out, presumably by the composer himself. This cut must have occurred early in the history of the opera, perhaps even before the first performance, for the measures are found in no other contemporary source.

  Still later, either Rossini or another musician realized that the three measures of arpeggios (24 –26) lose much of their point in the absence of the two culminating measures. As a result, the first two of these three measures were marked “si passa” (cut these), and the third “si suona” (play this). This accounts for the version of the horn solo found in most contemporary printed editions of Otello, although in rendering the part for keyboard the editions often transformed the horn arpeggio into more pianistic scales. Contemporary manuscripts, on the other hand, normally preserve all three measures of horn arpeggios. The resulting proportions are all wrong: the labor of the arpeggios brings forth nothing more than the fabled mouse (in this case, the horn trill).