Continuo in Mozart Operas

I am wondering about the use of the continuo in Mozart Operas or other works. I’d have little doubts about his juvenile works such as Ascanio, Lucio Silla, Re Pastore, still I usually can’t hear a continuo realization in the recent recordings of those operas.
And what about Clemenza di Tito, such a late Opera but written in a style which would allow continuo? About the same for Idomeneo.

Eroic operas of the same era by Hasse or other composers are more frequently heard having a continuo realization, maybe Mozart still carries the weigth of the XX century practice while those others composers are approached more freely?

Not an Opera, but the Requiem autograph bears numbers for continuo on the bass line, still I don’t usually hear any organ or harpsichord in performances.

And of course there is to wonder about the three comedies Nozze di Figaro, Così fan tutte and Don Giovanni. One can be lead to think they are too modern in style and language to bear a continuo realization. Still I once attended to a (moderately-boring) performance of Così fan Tutte at the Teatro San Carlo, Naples, Italy. Of course there was no continuo realization throughout the opera, excepted in the trio “Soave sia il vento”. It was a magic moment, the harpsichord truly added something to the beauty of the music.
So I am a bit puzzled.

It seems to me that it would be expected that, at least in those operas with secco recitativo, the continuo harpsichord (or fortepiano) would be employed in the arias and ensembles. Conducting as we know it was not as highly developed in the later 18th century as from Weber onwards. Although directing from the violin was also common, Haydn played fortepiano in London for the performance of his symphonies, although none is specifically called for in the scores.

BUT, as those responsible for the balance in recordings are often unaware, the effect of the instrument would, of course, be more evident to the players than to the audience, to the point where the sound would be submerged in the orchestral tone in the more vigorous pieces. Even in Messiah, I have found that it is more useful to control the ensemble by conducting many of the choruses and maintaining eye contact while seated at the harpsichord than to actually play, striking the keys only during the numbers for solo voices.


Thanks David.

Your opinion leads to another related issue. Operas with secco recitativo were composed well into the XIX century: Rossini (both buffo and serio) and come buffo Donizetti, for example. Would you employ the harpsichord/fortepiano in those as well?

BTW, yesterday I attended to another Così fan Tutte production here in Naples. Boring and confusing and not-so-subtle staging, but very fine musical rendering. A small orchestra, conducted at the fortepiano (used only in recitativos, if I didn’t miss anything) played by the conductor. It has been a real lesson in recitativo-accompanying. Outstanding.

The conductor was Dan Ettinger, who I didn’t know of before yesterday. Just minutes ago I learned the Neaples Opera Theatre, San Carlo, has appointed him as musical director, starting from January 2023. I am happy for this choice.

As a side note, Ferrando was interpreted by Maxim Mironov. As he is a Russian, I was scared he would have been booed by the public. As it turned out, the theatre didn’t cancel his contract for being Russian, the public has been warm with him as with the other interpreters, and this has been a small but significant reason of satisfaction for me.

Minutes after the ending, we were in the Pizzeria in front of the San Carlo when Mironov went in with other two guys. I wouldn’t never have the courage but my wife greeted him and asked to sign our two copies of the program. He was very nice, smiled, asked her if we had enjoyed the production, and signed the programs.

We in Italy, in Milan, had a university course on Dostoevsky canceled because he was Russian, and in Genua a Russian-Italian film festival was canceled as well…

domenico.statuto Domenico Statuto
March 28

Thanks David.

Your opinion leads to another related issue. Operas with secco recitativo were composed well into the XIX century: Rossini (both buffo and serio) and come buffo Donizetti, for example. Would you employ the harpsichord/fortepiano in those as well?

I can offer a very brief summary (from memory, drawing on press reviews, theatre accounts, etc.) of the situation at the King’s Theatre (the “Italian Opera”), Haymarket, London, where performances of Mozart’s operas did not start until 1806, with La clemenza di Tito.
The accompanying instrument for recitative was the harpsichord until 1805, in which year the piano-forte took its place. Although a few reviews commented unfavourably on the sound of the harpsichord, the change was not universally welcomed, and they reverted, apparently very briefly, to the harpsichord.The main harpsichordist at the time was Giacomo Gotifredo Ferrari (1763 - 1842), and the Acting Manager (effectively the director) was Michael Kelly (1762 – 1826) who, as Mozart’s original stuttering Don Curzio and Don Basilio in Le nozze di Figaro (Vienna, 1786). would have been familiar with Mozart’s practice.

By 1817, bowed instruments apparently had been substituted for the piano-forte in the prevailing accompaniment of recitative, and it is unclear whether the latter continued to play at other times: a duo consisting of Domenico Dragonetti (1763 – 1846), double bass, and Robert Lindley (1776 – 1855), violoncello, played together, sharing the same stand. Lindley played chordally, and although we might assume that Dragonetti, with only three strings, played just the written bass, one review commenting on the vocal intonation of Giuseppe Naldi says that “Mr Dragonetti gave the chord”. As Rossini didn’t arrive in London until 1823 it is likely that this practice continued then.


I am not very well up in Rossini research. Until recently there were no really authoritative scores, and I havent seen them. Secco recitativo is secco recitativo, isnt it? And doesnt it need a keyboard? Remember, Stravinsky used harpsichord in The Rake’s Progress.

Personally I am not a fan of any kind of piano in recitative: it sounds out of place, but that may just be me.

My Mozart scores are mostly all pre Urtext era and I have never seen a holograph manuscript of a Mozart opera, though they are all available in excellent facsimile now. IMSLP has a facsimile of a contemporary copyist’s acore of Figaro and the curious thing is that, like the printed Peters score I have, there are no figurings. Of course, the harmonies of the recits are all very simple, but that is not the case for all the arias and ensembles. It is clear that sometimes the harpsichord should play the first chord of the next number to complete the cadence, but otherwise my suggestion above is not borne out by the written evidence. Be that as it is, in opera, as in many live productions, not everything needs to be written down if everyone knows it and it is uncontroversial.

It is the case that Mozart’s operas (and many of the symphonies and concerti) were not much performed by the end of the 19th century. Mahler was out on a limb there in performing them, without a tradition to help him. He played the recits himself on a spinet, but I dont think he did more than that. (He did play continuo in Bach in New York with a piano prepared “to sound like a harpsichord”, but loud enough for Carnegie Hall.) The Italian operas were sung in German and the public was not very enthusiatic about them, according to Bruno Walter.

Che pazzi!


Personally I am not a fan of any kind of piano in recitative: it sounds out of place, but that may just be me.

Same for me, never heard a fortepiano in recitativos which I could find agreable. However, who knows why, this was not the case yesterday night.

My experience with FP on recits, both playing and listening, is the same - unsatisfying.

One notable exception however is using it in Haydn oratorios. It feels right, and seems probable.

As a frequent player of both in opera houses and festivals, my experience has been that of practicality above all else. As a continuo player, our purpose is first and foremost to accompany and support the singers; the seasoning with frequent roulades and witticisms are secondary to this. The fortepiano does lend itself to more of this, because the dynamic contrasts allow a subtlety there, but the harpsichord can outstay its welcome.
I don’t think players in the 1770s and 80s were so fussy and played the pianoforte/fortepiano largely with the same technique anyway. The fortepiano’s sound does sometimes get eaten in the theatre and requires amplification/foldback, which in turn create imbalances. I read that in Vienna, the direction of a concert from the harpsichord by one of the older Italian Kapellmeisters at the time of Mozart was considered a curiosity and worthy of remark by a reviewer. The Van Swieten Concerts had Süssmayr playing harpsichord for Mozart’s Davide Penitente and Handel reworkings, but these were works in an ancient style. I have often played in big tuttis but left the delicately scored music to the players, for continuo playing can “lock” their sense of phrasing and rubato.

In London, the change from harpsichord to pianoforte seems to have taken place fifteen or so years later in the opera theatre than in the concert room.

Charles Burney’s account of Haydn’s first London concert (1791) is well known: “Haydn himself presided at the piano-forte; and the sight of that renowned composer so electrified the audience, as to excite an attention and a pleasure superior to any that had ever been caused by instrumental music in England.” If “presided” leaves doubt as to what Haydn did at the keyboard, he himself gives an account (fourth London Notebook) of a performance on 10 April 1795: “I was invited to a musical soirée at the Prince of Wales’ in Carlton House. An old Symphony was played, which I accompanied on the pianoforte; then a Quartet; and afterwards I had to sing some German and English songs. The Princess sang with me, too; she played a Concerto on the pianoforte quite nicely.” Two weeks earlier he had observed Clementi (Haydn’s third London Notebook, 24 March 1795): “Mara, having returned from Bath, gave her Benefit-Music in Hannovers Room [the Hanover Square Rooms]. There were not more than 60 persons in the audience. It is said that she never sang better than at that time. Janiowick conducted. Mr. Clementi sat at the pianoforte, and conducted his new grand Symphony, without success.”

At the King’s Theatre, Haymarket, where Mozart’s operas were only gradually introduced from 1806 onwards, the change from harpsichord to pianoforte accompaniment, which was commented upon in the press, took place in 1805, with a very brief reversion to the harpsichord because the pianoforte wasn’t universally preferred at first. The principal ‘cembalo’ player (the term used in the theatre accounts) at the time was Giacomo Gotifredo Ferrari (1763 – 1842). Mozart’s own practice, for what it was worth, would have been known there; the ‘acting manager’ in the theatre at the time was Michael Kelly (1762 – 1826), who had created the roles of Don Basilio and Don Curzio in the first performance of Mozart’s Le Nozze di Figaro in 1786. By 1817 – perhaps a little earlier – the pianoforte is no longer mentioned; recitative was accompanied there by the celebrated duo of Domenico Dragonetti (double bass) and Robert Lindley (violoncello), the latter, at least, playing chords; one press account of Giuseppe Naldi’s singing reported that 'Mr Dragonetti gave the chord [sic]’. This suggests that within a decade or so, the sound of bowed strings was preferred for this purpose to that of either keyboard instrument. And when, from 1806 onwards, Naldi frequently appeared in role of the eponymous musicomane in Il fanatico per la musica (a London version of Simone Mayer’s farsa per musica, Che originali! (1798)), he played the pianoforte, not the harpsichord, on stage.