Notes on the Libretto

The text of The Lucky Star is a nightmare rivaling the bibliographic intricacies of Jane Annie and Mirette. I have been able to locate, through my own research and through much appreciated work by Selwyn Tillett, four texts which form the basis of the following discussion. All these differ from one another, and there is no reason to doubt that there exist other versions of the opera, both published and unpublished, that have not yet been uncovered.

To investigate the libretto as originally conceived, a natural jumping-off point is the licence copy forwarded to the Lord Chamberlain. In the case of The Lucky Star, this version offers no help. Under normal circumstances, the license copy dates from a few days before the first performance of a play, but the license for The Lucky Star was granted on February 15, more that a month after the premiere. This is an extraordinary circumstance since plays required licensing before they could be performed. That no plays at all were apparently licensed in the month of January 1899 helps explain the later date of license, though it may be that the script was re-submitted after revisions. This would also be an extraordinary circumstance, for usually the initial license was sufficient however drastic revisions were in the course of a piece's run. The many changes made to the text of The Lucky Star were not so extreme as to warrant re-submitting to the Lord Chamberlain.

I have been able to examine two versions of the libretto, one (libretto A) from the New York Public Library and the other (libretto B) supplied by Selwyn Tillett. (These terms are used for the sake of convenience and have no bibliographic significance.) Libretto A is definitely the earlier of the two; libretto B dates from mid-March 1899. The musical numbers of each differ from the existing vocal scores, but neither libretto substantially alters the story. In general, libretto B serves to significantly increase the part of Siroco, strengthen the romance between Aloës and Tapioca, and introduce the part of the dancing girl Adza.

Libretto A has eighteen musical numbers while libretto B has twenty-one. Neither includes lyrics for the finale of act three and both include a chorus "Ho, hale him hence" in act two that is unnumbered. (The meter of this chorus is identical to the martial chorus "In courtly train" which precedes it, and was probably sung to repeated music.) Two numbers from libretto B are unnumbered, these being an additional song each for Siroco and Aloës. The duet for the King and Siroco "When away I slink" (libretto B) finishes with a note indicating that other verses would be added to it. If they were, evidence is lacking; the vocal score includes only this short version. The introductory song for the King is shorter in libretto B than in libretto A; the opening chorus of act two is reduced to a short chorus from a chorus with solos; and there are many changes in dialogue.

The best evidence for what musical numbers were actually performed onstage is the vocal score. This was available for sale to the public and meant to be used for rehearsing interested amateurs as well as D'Oyly Carte's touring companies. This solid evidence is complicated by the fact that there exist two vocal scores varying greatly in detail. The earlier, which according to Tillett is nearest in agreement with the licensed libretto, has twenty musical numbers and is 239 pages long. The later one has more music (twenty-two numbers) and fewer pages (231). Seventeen numbers and the overture are common to both, though the introductory song for the King is shorter in the second version.

There are two songs from these four sources that are unique to only one document. The earlier vocal score includes a serenade for Lazuli similar to serenades in the French and American versions of the opera. It also includes a song for the King with chorus entitled "The Ostrich" which corresponds to a five-verse ditty in the American version called "The Omniscient Ostrich; or, the Bird Who Knows It All" (a song which gained some fame, according to Gerald Bordman). The second vocal score also includes four verses for Siroco's "Horoscope"; libretto B has only two verses. The King's introductory song is longer (eleven pages) in the first vocal score than it is in the second (six pages). Laoula's ballad "When I was a child of three" has two verses in both libretti but three in both vocal scores, the additional verse being the second (there are three verses in The Merry Monarch, the American version upon which The Lucky Star is based). The coon song is expanded to twice its length in the vocal scores than the libretti, doubtless a reflection of its popularity.

One conspicuous result of the sequence from libretto A (the earliest examined source) to the second vocal score (the latest) - besides those mentioned above - is the watering down of the love story of Lazuli and Laoula. The overtly sexual "tickling trio" with Aloës and the "cooing" duet in act three have been jettisoned. The latter, which as a lyric is nearly a failure, is not missed. Surprisingly, Lytton's part was not augmented, though the Era (March 18, 1899) notes a recitative and song for Tabasco "If I didn't have to keep my temper in control." The critic didn't like the number, saying it "is not likely to take the town." Since it doesn't appear in either vocal score, he was probably correct.