It’s been quite a few years since I’ve opened my copy of Quantz’s On Playing the Flute (Reilly translation), but there was a time when it was a veritable bible to me.
Yes, of course, Johann Joachim Quantz (1697–1773) can come across as a stuffy old man, documenting a tradition of performance that was well-established in 1752 and may even then have been erring in the direction of old-fashioned. All the same, he was one of the very few to take on not only the basic mechanics of an instrument – which most of the tutors seemed to do – but the fundamental principles of performance practice and musicianship. The extent to which Quantz’s observations and recommendations can be applied to even slightly later music is arguable, but in the case of concerto cadenzas much of what he says is echoed and, if anything, amplified by later writers. First, when to include them?
“The abuse of cadenzas is apparent…if in instrumental music they are introduced in pieces in which they are not at all suitable; for example in gay and quick pieces in two-four, three-four, three-eight, twelve-eight, and six-eight time. They are permissible only in pathetic and slow pieces, or in serious quick ones.”
Of course, the “gay and quick” movements, especially if they are in rondo form as was traditional for finales, tend to contain their own variant on the cadenza: the fermata or Eingang, sometimes referred to in English as a “lead-in”. But an Eingang is a mere flourish (leading the soloist into a statement of the rondo theme), not a full-fledged cadenza, which has a particular task to play:
“The object of the cadenza is simply [in my copy I at some point inserted a little ‘!’ above the word ‘simply’] to surprise the listener unexpectedly once more at the end of the piece, and to leave behind a special impression in his heart.”
After acknowledging the near impossibility of establishing rules for something that is by definition extemporaneous and informal, Quantz then attempts to set out some prescriptions. He’s that kind of guy. These include:
“Cadenzas are of either one or two parts. Those of one part [i.e. for a single instrument] are chiefly extempore… They must be short and fresh, and surprise the listeners, like a bon mot. Thus they must sound as if they have been improvised spontaneously at the moment of playing.* Hence you must not be too extravagant, but must proceed economically [Ger: “like a good innkeeper”], especially if you often have the same listeners before you.”
*In writing “as if” JJQ is already acknowledging that mid-18th-century cadenzas, although “chiefly extempore” might very well be prepared in advance.
Then there’s the part that, I must confess, always made me feel resentful:
“Vocal cadenzas or cadenzas for a wind instrument must be so constituted that they can be performed in one breath. A string player can make them as long as he likes, if he is rich enough in inventiveness. Reasonable brevity, however, if more advantageous than vexing length.”
Ok, so his concluding sentence tempers things, but I did always think it unfair that inventive string players be encouraged to ramble at will while us flute players et al were constrained by our breath capacity. After all, from a purely musical perspective, there’s no reason why a wind cadenza (if the player is rich enough in inventiveness!) cannot be longer than a single breath. And a string cadenza in which the player’s phrasing didn’t “breathe” would be pretty awful. (It rankles even more when you consider that an oboist’s breath, being rationed via a tiny reed, goes much further than a flutist’s.)
Still, Quantz’s overall point is a sound one and his translator (1966/85) amplifies it in a footnote: “Observance of this rule would be of inestimable benefit to many modern performances of 18th-century flute concertos.”
More important, the matter of keeping a cadenza short and pithy, epigrammatic and surprising, is one aspect of performance practice that evidently lasted well past the mid-18th-century, since Johann George Tromlitz’s Detailed and Thorough Tutor for Playing the Flute repeats the sentiment in 1791, the year of Mozart’s death:
“Just as one has more freedom to branch out here [in cadenzas rather than fermatas], more care is required. However, this freedom is abused these days; especially on stringed and keyboard instruments [the cadenza] is often made as long as the entire movement up to that point, and one gets so far away from the subject that when the ritornello returns one has to recollect whether or not it belongs to this movement; so ideas are heaped on ideas, or to be more accurate, notes upon notes, which do not manifest the slightest feeling of the main subject; just as long as the cadenza is very long-winded, people cry: bravissimo!”
By the sounds of it (and there’s plenty more!), the “abuses” that Quantz touched on had become worse. There is a palpable wistfulness when Tromlitz adds:
“I do not mean to say by all this that a beautiful cadenza suitable to the material cannot have its good effect; if it is proper to the whole, is of an appropriate length, and is fashioned in good taste and with understanding, then it is indeed a beautiful thing.”