84ⁱⁱⁱ. Gianni Schicchi

According to the known documents, Puccini mentioned Gianni Schicchi for the first time on 1 March 1917 (in I-Mr, Carteggio Puccini-Clausetti, no. 38), when he informed Ricordi that Forzano was gathering information for this opera in Florence. Forzano himself maintained, however, that he proposed the idea when Puccini returned from futile journeys to London and Paris in search of a comic subject (in April, May, and June 1913). This time frame agrees with the date of Puccini’s and Forzano’s first meeting (see NOTES to 84ⁱⁱ).Note: *See Forzano 1957, 14-15. Forzano’s account is supported by a quotation in Adami 1935, 151 (possibly originating from 1913 in spite of Adami’s certainly wrong date of 1917, if it is authentic at all).  So it is conceivable that Puccini first considered the Gianni Schicchi episode from Dante’s Divina commedia (Inferno, Canto XXX) for a one-act opera already in summer 1913. However, that was realized only at the beginning of 1917, long after Il Tabarro was finished and the decision had just been made to compose Suor Angelica.
There was obviously insufficient source material in Dante’s text for Forzano to construct an entire libretto of Gianni Schicchi based on the few lines pertaining to that episode in the Divina Commedia. Research has shown that Forzano made extensive use of a Divina Commedia commentary by an anonymous 14th-century Florentine writer, published in 1866.Note: * Commento alla Divina Commedia d’Anonimo fiorentino del secolo XIV, ora per la prima volta stampato a cura di Pietro Fanfani, Bologna, 1866; it is quoted extensively in Girardi 1995, 416-17.  However, it is uncertain whether Forzano also knew Gildo Passini’s three-act comedy Gianni Schicchi (first performed on 21 September 1918 at the Teatro Olimpia in Milan).Note: *As asserted by Salemi 1994-95, in which there is also a facsimile of that drama from the periodical Comoedia of 20 February 1922. Since neither Forzano and Puccini nor Passini later made any legal claims against each other, it is likely that both Forzano’s libretto and Passini’s play were based on the same older source.
Shortly after 1 March 1917, Forzano provided a first draft of the libretto.Note: *See I-Mr, Carteggio Puccini-Clausetti, no. 39. Two documents (published in Fraccaroli 1957, 210 and 211), which are difficult to date, have led to the assertion in writings about Puccini that he interrupted the composition of Suor Angelica to work on Gianni Schicchi. However, the actual dates of composition do not support this assumption. As usual, after the initial euphoria, Puccini repeatedly doubted the suitability of Gianni Schicchi as a subject for one of his operas (see, for example, Epistolario 1982, no. 141, probably written in June 1917). To enhance the medieval authenticity of the libretto, Puccini and Forzano consulted the eminent Florentine Dante scholar, Orazio Bacci.Note: *See Marchetti 1973, no. 448, which — contrary to the editor’s opinion — probably dates from September 1917.
Only immediately after finishing the composition and instrumentation of Suor Angelica (which, according to 84ⁱⁱ.B.1, occurred on 14 September 1917) did Puccini begin composing Gianni Schicchi (see Carteggi 1958, no. 713, of 17 October 1917) — astonishingly hoping to finish it within little more than one month, but of course nothing came of that, for Puccini also had many other obligations. He only began the notation of the full score on 3 February 1918 (see 84ⁱⁱⁱ.B.1), for which he asked Ricordi (on 15 February) to provide special music paper with many staves for the voices.Note: *See I-Mr, Carteggio Puccini-Clausetti, no. 54. In the first half of March there suddenly was a short interruption of Puccini’s usually very purposeful work.Note: *See I-Mr, Carteggio Puccini-Clausetti, nos. 59 and 61. That interruption is also detectable from Puccini’s autograph double leaf numbering of 84ⁱⁱⁱ.B.1. Shortly before that, he had apparently already delivered the first part of the full score (up to fig. 59) to the publisher.
On 1 April 1918 Puccini informed Ricordi that he had particular difficulties with the finale.Note: *See I-Mr, Carteggio Puccini-Clausetti, no. 65, and Puccini 1981, no. 42. However, on 19 April he finally reported that the opera was finished.Note: *See I-Mr, Carteggio Puccini-Clausetti, no. 67; the autograph full score 84ⁱⁱⁱ.B.1 has 20 April 1918 as the final date. Shortly thereafter Puccini sent the last part to Carignani to complete his work on the piano-vocal score.Note: *See I-Mr, Carteggio Puccini-Clausetti, no. 71 of 30 April 1918.
In the months before the first performance Puccini was occupied with corrections of the copyist’s full score which was apparently rather faulty.Note: *See I-Mr, Carteggio Puccini-Clausetti, no. 75 of 26 May 1918, no. 76 of 27 May, no. 78 of 8 June, no. 82 of 27 June, and no. 83 of 29 June. On 22 November 1918, Puccini proposed to include the relevant lines from Dante in the printed libretto, so the background of the opera would be more comprehensible.Note: *See I-Mr, Carteggio Puccini-Clausetti, no. 107 However, that idea was never implemented.
The few changes (far less than in any other of his operas) made by Puccini immediately before and/or after the first Italian performance on 11 January 1919 were published in 84ⁱⁱⁱ.E.2, which represents the definitive version of the score. That certainly was the version presented in the second Italian performance in Florence on 10 May 1919.