78. La Fanciulla del West

The first performance of David Belasco’s drama The Girl of the Golden West was presented in 1905 in Pittsburgh, prior to being performed in New York the same year.Note: *The text has been published e.g. in the collection Representative American Dramas, edited by Jonas Moses Montrose (Boston: Little, Brown, and Company 1933).  Puccini probably attended a performance of the play there in late January 1907.Note: *See Winter 1918 vol. 1, 75, and vol. 2, 213-14. 
Despite years of searching for a new libretto, by no means had he decided at the beginning of 1907 to choose Belasco’s play as the basis for his next opera, although before returning to Italy he asked Belasco for a copy of the play so it could be translated into Italian for further study (see Winter 1918, vol. 2, 214). However, Puccini sketched to Illica a scenario for a possible opera, thematically connected with the California gold rush but with a completely different plot.Note: *The document, similar to a letter, is preserved in US-Wc and published in Valente 1971, 67-69; contrary to 
Valente’s assumption, Puccini probably wrote it in the first months of 1907.

Then, however, Puccini intensively reverted to his project to compose an opera about Marie Antoinette (Maria Antonietta), with Luigi Illica as the librettist, which had already been in development for quite some time. It was only in May 1907 that Belasco's consent was given (Randall and Gray Davis 2005, 181 no. 1), and Puccini solidified his intention to compose the “Girl”, as he was fond of calling this new opera prior to finalizing the definitive title.Note: *See Puccini’s unpublished telegram to Sybil Seligman of 15 June 1907 in MEL, as well as Seligman 1938, 135 ff. At the same time, the librettist had already been selected: Carlo Zangarini (1874-1943), a young writer who until then had only worked on a few modest projects for Ricordi, but had never written a complete libretto of his own. Therefore, Puccini took pains to draft the dramatization of the libretto and its substantial differences from Belasco’s original.Note: *See Seligman 1938, 139, and Epistolario 1982 No. 97.
In the middle of August 1907, Zangarini visited the composer at his vacation house in Boscolungo Abetone.Note: *See Marchetti 1973, No. 339, and Epistolario 1982  No. 99, certainly written on 25 August 1907.  Zangarini returned one month later to see Puccini at his other summer residence of Chiatri (see Carteggi 1958 No. 526) and in October at Torre del Lago (see Seligman, p. 147). There were the usual difficulties with the libretto drafts,Note: *See the excerpts from Puccini’s letters in Zangarini 1930. leading to a complete libretto only at the end of January 1908 (see Seligman 1938, 152). Nevertheless, Puccini apparently began with composition sketches for the opera as early as November 1907 (see Carteggi 1958 No. 530). A much later letterNote: *Cecchini 1980-1993 No. 2800, according to the original in US-NH (Koch 707.123) dated 8 February 1911. indicates that Puccini had also used or at least knew “musica indiana e americana”. An employee of the American embassy in Rome helped him search for “original” music.Note: *For a correct enumeration and evaluation of the sources previously often erroneously identified, see Atlas 1991b.
In the spring of 1908, an open conflict arose between Puccini and Zangarini (see Carteggi 1958 No. 543, and Seligman 1938, 153), and Puccini enforced the procedure that had proven effective since La Bohème of summoning a co-author (see Epistolario 1982 No. 101): Guelfo Civinini (1873-1954), an author as young and relatively inexperienced as Zangarini, but a brillant and adventurous journalist. From the middle of April 1908, Civinini reworked Zangarini’s libretto together with Puccini.Note: *See Puccini’s unpublished letter to Sybil Seligman of 14 April 1908, MEL No. 98; MRC 1907/08 - 14/421 of 22
April 1908; Epistolario 1982 No. 104; Ross and Berra 1979, 859-60; Seligman 1938, 153-54.
Meanwhile, by the end of February 1908, Puccini had already begun to notate the continuous composition draft of Act 1 (see 78.A.1), and for a long time he corresponded with Civinini about changes for this act.Note: *See Epistolario 1982 Nos. 102 and 109, the latter certainly written on 11 July 1908. In the middle of June Puccini had played through the beginning of the first act for Toscanini and Gatti-Casazza, the future team to head New York’s Metropolitan Opera, and who wanted the première of Puccini’s newest opera for their calling-card performance (see Gatti-Casazza 1973, 157-58). At that time, the title of the opera was still uncertain: Puccini shrank from the word “fanciulla” and preferred “La Figlia del West” or “L’Occidente d’Oro."Note: *See Epistolario 1982 No. 102 and Seligman 1938, 157, of 30 August 1908.
Puccini’s serious marital crisis, beginning in early October 1908 and culminating in the suicide of the maid, Doria Manfredi, interrupted his work for several months.Note: *See Seligman 1938, 166, of 4 October 1908, and then again 176-77, of 16 April 1909. So the composition draft of Act 1 was finished only on 28 April 1909 (see 78.A.1). Also afterwards, the turbulence of his private life prevented Puccini from continuing the composition. Apparently he returned to it near the end of July 1909 (see Seligman 1938, 180-81). Most of the work on Act 2 was done in August and September 1909 in Bagni di Lucca (see also the dates in 78.A.1), and on 30 September Puccini communicated that he had “nearly finished the second act” (Seligman 1938, 185), specifically mentioning the love duet. In October at the latest, the definitive title was accepted. Puccini inscribed it on a photograph dedicated to Giulio Ricordi that month as: “La Fanciulla del West” (see Ferrando 1984, 286).
At the beginning of November 1909, Puccini thought to be nearly ready with the composition of Act 3Note: *See Seligman 1938, 187, of 9 November 1909. but probably did not complete it yet (see below). Instead, already on 22 November he turned to the instrumentation of the opera. The autograph numbering in 78.B.1 indicates that he began with the entrance of the miners (at fig. 5). The beginning was obviously entered in the autograph full score in January, after the completion of the act on 21 January 1910, and probably still underwent some subsequent condensing (see 78.B.1). Immediately afterwards, Puccini orchestrated Act 2. At times, Carignani was with him to prepare the piano-vocal scoreNote: *See Marchetti 1973, No. 374, of 25 February 1910. the engraving of which Ricordi indeed began on 14 March (see 78.E.1). The act was finished on 7 April (see 78.B.1), and Acts 1 and 2 were sent to the publisher at the end of March or the beginning of April.Note: *See MRC 1909/10 - 12/424 and 13/146. One month later, Ricordi began engraving the full score (see 78.E.1a). At the beginning of July, Puccini expected to receive the full score proof sheets of Act 2.Note: *See Puccini’s unpublished letter to Giuseppe Albinati, Casa Ricordi, of 6 July 1910, last documented in Stargardt’s auction of 26/27 November 1985.
The unfinished third act proved problematic, so that the completion of the composition and the instrumentation were virtually simultaneous.Note: *See e.g. Puccini’s poem to Giulio Ricordi of 11 July 1910: “E dico che lavoro / ma ancora non strumento [And I say that I work, but do not yet orchestrate]” in Sartori 1958, 353.  Puccini demanded changes in the text of the finale from Zangarini, their differences having been reconciled in the meantime.Note: *See Zangarini 1930, second issue, 10, of 21 July 1910 It can be seen from the autograph full score that separate passages were later insertions (see 78.B.1). The composition of the final scene and probably also its arrangement in piano-vocal score were finished on 27 July 1910 (see Carteggi 1958 No. 567), followed by the instrumentation on 6 August. But the definitive distribution of the text was obviously not yet completely finalized (see 78.B.1 and 78.E.1). On 11 August, Ricordi confirmed the receipt of the ending of the full score (MRC 1910/11 - 2/318). In that month, in Viareggio, Puccini and Toscanini discussed the details for the first performance, particularly the appropriate tempi (see Giovannetti 1958, 133-36); so that conductor exerted an influence on details of the first edition of the full score (78.E.1a) and the first piano-vocal score (78.E.1). This is also documented by the surviving proofs bearing Toscanini’s handwritten corrections (see 78.B.3 and 78.B.4).
Rehearsals in New York began in October (see Carteggi 1958 No. 569), with the performance version based on 78.E.1a/78.E.1. The première, however, incorporated the changes made by Toscanini and Puccini before and during the rehearsals (see the notes to 78.B.3 and 78.B.4). For the most part, those modifications pertained to matters of expression and instrumentation; nearly all of them became established in the second edition of the full score (78.E.3a) published by Ricordi in June 1911 at the latest. 78.E.3a also contains other corrections, previously documented in the piano-vocal scores 78.E.2 and 78.E.3. Puccini partially described these changes in a letter to Toscanini of 1 February 1911 (Carteggi 1958 No. 577).
After the world première, Puccini implemented two further substantial changes comprising extensive cuts he made shortly before the first Italian performance of the opera in Rome. Since he described them to Toscanini in a letter of 1 June 1911 (Carteggi 1958 No. 582), they may also have been executed in the first London performance three days before. In any event, for the most part the Roman première represented the version of 78.E.4a/78.E.4, even though these newly revised editions of the full score and the piano-vocal score were only published almost one year later. The numerous performancesNote: *In Brescia, Lucca, Liverpool, Turin, Naples, Genoa, Budapest, and Monte Carlo. attended and supervised by Puccini between the Rome production and the publication of the revised editions certainly followed that version as well, although the performance material in print was still 78.E.3a and 78.E.3.
It is probable, but not definite, that the changes in Act 1 made by Puccini at the end of 1912 for the performance at La Scala should be considered as the definitive corrections (see 78.B.6). However, they were still not included in the octavo format of the full score (78.E.5) begun in the last year of Puccini’s life. Nevertheless, in 1919 Puccini mentioned a “progetto di tagli per il 1° atto [project for the cuts in Act 1]” (Carteggi 1958 No. 762), perhaps referring to the passages deleted from 78.B.6.
Some years later, during the hiatus in Puccini’s work on Turandot in autumn 1921, necessitated by the new concept for the libretto of that opera, he considered revising the end of the love duet in the second act of La fanciulla del West Note: *Intimated in an unpublished letter to Guido Zuccoli of 5 November 1921 in US-NH, Beineke Rare Books and Manuscripts Library. and finally provided a more effective version expanded by 16 new bars. Puccini wanted this revision to be implemented in performance for the first time in Rome (27 December 1922),Note: *See Carteggi 1958 Nos. 843 and 845 of 11 and 17 October 1922, respectively. but the singers on that occasion refused to accept this addition.Note: *See Puccini 1981, Nos. 117 and 118; Marchetti 1976; Restagno 1974, 211. Consequently, the honor of presenting the first performance of the opera in Puccini’s final version unexpectedly fell to a certainly very modest provincial performance in Viareggio (see Carteggi 1958 No. 870), which therefore garnered explicit praise from Puccini (see Puccini’s letters to Maria Bianca Ginori-Lisci in Critica 1976, No. 55).
The version with this late addition was first published in the sales edition of the full score (78.E.5), issued after Puccini’s death. It is remarkable that the much more widely disseminated piano-vocal score adopted this definitive form (78.E.5a) only after 1940. Since the opera was so rarely performed after Puccini’s death, it is obvious that a new edition was of no economic interest to the publisher.