During the work on my biography of Giacomo Puccini, published in 1989, I became aware of the regrettable defect that so many substantial facts in the literature pertaining to Puccini’s life and works are incorrect, distorted, incomplete, unreported, and often undocumented. Therefore, I endeavoured to portray the composer’s life by making full use of the primary source material accessible at that time, and to be more correct than was previously possible, even in the more trustworthy accounts of the past.
In the meantime, further research and a myriad of other relevant discoveries have revealed new information that has enabled me to improve and correct many details of my earlier work. It became increasingly clear that a new and thorough overview of the authentic sources would be an indispensable basis for any serious examination of Puccini’s compositions and working method. That led me to the decision to compile this catalogue of his works by registering and describing insofar as possible all of the printed and unprinted sources in public and private collections throughout the world.
As anticipated, this was no easy undertaking, for it required that virtually every detail of information in the literature be verified against the original sources (if accessible), supplemented with the findings of my own extensive research. However, in view of the inherent difficulties, certain errors and shortcomings are inevitable, and future revisions of some elements of this catalogue may be necessary and indispensable. In any event, with the help of my colleagues and collaborators, I hereby present the current state of scholarly research in Puccini studies.Note: Particularly Giorgio Magri of Pietrasanta, Italy, bountifully contributed to the current state of Puccini research. For many years, and parallel to my own studies, Magri worked on a book about Puccini’s non-operatic compositions, entitled Il musicista Puccini. Weary of the difficulties involved in the publication of his book, Magri placed his manuscript and its wealth of discoveries at my disposal. Since I often make use of Magri’s work, I duly mention it in the respective entries of this catalogue. Unfortunately, for the time being, Magri’s manuscript can only be obtained directly from the author or as a copy of the typescript in I-Lgp.
Giacomo Puccini did not leave behind an extraordinarily large number of compositions. Therefore, there is no need for a catalogue to organize an enormous body of music, as was the case in the past with, for example, the catalogues for Bach, Haydn, Mozart, Beethoven, and Schubert. On the contrary, the clarity of Puccini’s achievement with its core of twelve more or less famous operas has moulded the general consensus that nothing of note in Puccini studies remains to be elucidated.
Cecil Hopkinson’s Bibliography of the Works of Giacomo Puccini, published in 1968 (Hopkinson 1968), was invaluable for revealing the error of that consensus. Hopkinson’s publication made it clear that Puccini’s operas partly underwent considerable changes after their first performances, and that the current printed editions do not always make Puccini's revision process apparent. Hopkinson, a bookseller and collector with a limited knowledge of music, took commendable care in identifying and classifying the printed editions of Puccini’s operas, but he neglected Puccini’s other works and almost completely refrained from describing the autographs and other manuscript sources. Wherever possible, Hopkinson’s cursoriness, omissions, and series of mistakes have been systematically revised and corrected in this catalogue.
CHOICE OF LANGUAGES
To achieve the broadest possible international readership, this catalogue (originally prepared and written in German) is published in English. Translations of Italian quotations have been either set in square brackets within the actual quotation or provided in footnotes. However, titles of compositions and their various published editions are generally given in the original language and not translated into English, even if anglicized forms for them are available (e.g. Il Tabarro and not The Cloak).
In this context, the question of capitalization in titles was problematic. The prevailing rules are quite clear in English, but not at all in Italian. In Italy there is a general tendency to use lowercase letters for all words after the initial capitalized word of the title (except, of course for places and proper names). That system, however, lacks consistency. Even the modern Ricordi editions have Le Villi, La Bohème, La Fanciulla del West, and Il Tabarro, while the same publisher’s catalogue spells “fanciulla” and “tabarro” all in lowercase typography but capitalizes “Villi”, “Bohème”, and “Trittico.” Furthermore, all of the Sonzogno editions (from the first to the most recent) read: La Rondine and not La rondine. As for Puccini, he did not hesitate to combine upper and lowercase letters, and he often (but not always) used lowercase initial letters even for the names of cities. Ultimately, we decided to retain the capitalization of initial letters in all those cases, also in accordance with contemporary English (and German) usage.
RUBRICS FOR EACH ENTRY
The same system of rubrics has been applied to every composition listed in this catalogue.
For the operas, the TIME OF ORIGIN refers to the entire period of Puccini’s involvement with the composition, in most cases starting with his intensive work on the libretto.
Apart from the premiere PERFORMANCES, other productions are mentioned only if they represent an important stage in the work’s subsequent history. Wherever possible, CASTS were ascertained from contemporary sources. Some errors and omissions are possible, for example when the first names of lesser-known singers were either cited in different ways or not mentioned at all.
Incipits are not given for the operas or for works that are easily accessible in modern printed editions. In the case of other compositions, the incipits generally represent the beginning or other relevant parts of the work as faithfully as possible, but without indicating every detail in the manuscripts.
SOURCES are always subdivided into the following sections:
A. Sketches and Drafts
B. Autographs
C. Copies
D. Facsimiles
E. Printed Editions
Throughout this catalogue, all SOURCES are identified in the same way and in the same sequence (in bold if it is not a reference within the same number of the catalogue): the number assigned to the work in this catalogue, the letter of the relevant section, and an ordinal number (generally an arabic numeral). Thus, “74.E.3” signifies the third printed edition of Madama Butterfly.
The Sketches and Drafts (section A) are not sorted according to their chronological origin, which is often doubtful, but according to their location in the respective work. The term “sketches” refers to the notation of isolated passages, while “drafts” are passages of a continuous composition (the difference is explained below). With the operas, only larger holdings in certain collections have been assigned the usual arabic numerals; separated sketches and drafts, however, are identified by a roman numeral for the act,Note: Numerals for acts are, of course, not applicable to the one-act operas of the Trittico (84ⁱ, 84ⁱⁱ, and 84ⁱⁱⁱ); in Le Villi (60) they are replaced by the numbers of the individual pieces. an arabic numeral for the nearest rehearsal number, and a lowercase letter. Thus, “62.A.I.24.b” signifies the second known sketch for a passage closest to rehearsal number (fig.) 24 in Act 1 of Edgar. This system allows for later additions in the appropriate locations throughout the catalogue.
Autographs (section B) denote Puccini’s autograph manuscripts in full score and other sources in which his autograph revisions, modifications, emendations, and other indications were recorded after his completion of the full score.
Copies (section C) signify non-autograph manuscripts in a copyist’s hand; they are listed in chronological order.
Facsimiles (section D) are arranged according to the order in the previous sections A–C.
Unless stated otherwise, I have examined all the documents in sections A–D in the original or in photocopies.
Printed Editions (section E) are discussed below.
Paper formats are generally given only as 8°, 4°, or 2°; dimensions in centimeters (if determinable) are only given if they allow us to draw certain conclusions concerning the historical context. However, some caution is advisable, for many leaves in the manuscripts are damaged at the edges or torn off, thereby casting doubt on their exact size.
LITERATURE: The information in this rubric is restricted to the origin and publication of the respective work; it does not include interpretative analyses, however interesting such analyses may sometimes be. References to earlier literature pertaining to Puccini, despite its historical value, have been omitted if superseded by more recent research.
Generally, the information in the NOTES is based on the literature cited, supplemented by findings from my own extensive research, which frequently corrects the literature without explicitly indicating the discrepancies.
PUCCINI’S WORKING METHOD AND THE AUTOGRAPH SOURCES
It is remarkable that such varying quantities of sketches and drafts (section A) exist for the operas. In some cases they are extensive, but for the most part only a few leaves of sketches and drafts have been preserved, and nothing at all has come to light for La Rondine. For a better understanding of this situation, one must keep in mind Puccini’s working method, which he maintained virtually unaltered throughout his life.
Puccini began composing an opera by writing down sketches in different degrees of thoroughness. Often they consist of harmonies and short rhythmic or melodic phrases. Sometimes he annotated his sketches with written remarks pertaining to their tonality, instrumentation, or future utilization. Puccini usually started sketching (in ways that are not always easy to decipher) as soon as the very detailed discussions with his librettists were underway. Sometimes a leaf will contain sketches for unrelated passages, suggesting that Puccini did not always know in advance or have a precise idea of how the music he was notating would ultimately be used, if it would be used at all, or what modifications might be required.
After the preliminary sketches, he started writing out composition drafts based on the libretto. These drafts contain all the voice parts and a rudimentary piano accompaniment (sometimes with indications for the instrumentation) representing what would eventually become the orchestral score. The drafts are not always continuous, but treat the material and passages which were clear to Puccini at the time. When Puccini told somebody that he had started a composition, he generally meant that such drafts were in progress, whereas he sometimes referred to his preliminary jottings as “preluding” (“preludiare” in Italian). As a rule, he only began work on the instrumentation after an act had been largely completed in extensive composition drafts.
The different phases of sketching, drafting, and orchestrating overlapped in many ways. As a result, earlier sketches were often reworked while the composition draft was being written out, and drafts were being revised even while Puccini was occupied with the final orchestration. Sometimes large fully orchestrated sections of certain works were delivered to the publisher before the rest of the composition had been completely orchestrated so that the materials required for the first performance could be prepared and put into production. Those materials generally consisted of handwritten (later printed) copies of the full score and the chorus and orchestra parts, which the publisher provided to the theatres on a rental basis, and a printed piano-vocal score for singers learning their parts, for other theatre personnel, and for sale to the public.
It is certain that Puccini must have written down more or less extensive sketches as well as complete composition drafts for all of his operas before committing the autograph full score to paper. However, surprisingly few of those sketches still exist. Judging from those known to survive, given that Puccini evidently did not systematically destroy all his manuscripts from the preliminary stages of a composition, there should actually be thousands of pages of sketches and drafts awaiting discovery. Puccini often gave away single leaves from his working material, and it is obvious that the recipients of those pages (i.e. Puccini’s heirs, employees of Ricordi, friends of the composer, and others) later sold them or put them up for international auction. For example, Puccini’s daughter-in-law Rita sold many autograph manuscripts and other Pucciniana on the open market (see Zeileis 1992, 134-35). Since 1976, the Puccini material offered for sale by the music antiquarian Hans Schneider of Tutzing, Germany, most probably stems from Rita Puccini. After her death in 1979, and during the following years of legal disputes over the inheritance and the legitimacy of the composer’s heirs, the autograph holdings of Puccini’s family visibly and considerably decreased.
Moreover, there are probably many more leaves of sketches and drafts in private collections than are listed in the “A” sections of this catalogue. Often the owners do not want to reveal their identity; it is also possible that they do not even recognize the importance of the manuscripts in their possession. Therefore, the “A” sections of this catalogue are very preliminary in nature, and considerable additions can be expected in the future.
In some cases, I was permitted to specify the owners of autograph material, but in several instances the owners preferred to remain anonymous. In those cases, the holdings are listed as “Private collection in ….”
PRINTED EDITIONS (SECTION E)
Printed editions of Puccini’s non-operatic works are, in general, rare. Since my objective has been to present as complete a survey of those works as possible, I have tried to be very thorough in cataloguing the printed editions, even when there are no differences between them.
A special commentary is required here regarding the printed editions of the operas and their relation to the autograph full scores. For the most part, the autograph scores are preserved in the archives of Puccini’s principal publisher, Ricordi in Milan (I-Mr); the autograph full score of La Rondine, published by Sonzogno, is missing. Puccini’s autographs generally represent the original version of their respective opera at the time it was first delivered to the publisher. However, Puccini occasionally reused the original autographs later to make considerable changes and revisions (particularly so in the case of Le Villi (60), Edgar (62), and Madama Butterfly (74)). Often, additions were inserted or written into the original score; cuts were marked by crossing out autograph material; and, in some cases, single or several leaves that were no longer valid or required were torn out of the manuscripts. Far from being definitive, Puccini’s autograph full scores were in effect working copies of the music. They also served as the master scores for the publisher’s copyists and engravers, who often inserted remarks of their own. Wherever possible, I have attempted to describe the actual form of each autograph full score in a manner that identifies the status of the version it preserves, as compared to the subsequent states, alterations, modifications, and revisions in the various printings.
As a rule, the piano-vocal scores of the operas were prepared simultaneously with the final handwritten version of the full score. From Edgar (62) to Gianni Schicchi (84ⁱⁱⁱ) , the piano-vocal scores of Puccini’s operas were produced from his autographs by Puccini’s friend, Carlo Carignani in very close collaboration with the composer. When Puccini made changes after the first performance (as he always did), the printed piano-vocal scores were first in line to reflect those alterations, documenting the current state of the respective opera and serving as the basis for subsequent performances. As far as possible, the publishers made use of the existing plates for those changes; indeed, many of the currently available piano-vocal scores of Puccini’s operas are still largely printed from the original plates.
In contrast, for economic reasons, the performance materials, most of which were not engraved but duplicated from copyists’ scores and parts, were changed by the publisher less frequently. It was only in 1915 (with Manon Lescaut, 64) that Ricordi began to publish full scores of Puccini operas in 8° format for public sale. Prior to that (and even for a long time thereafter), changes were entered by hand into the rental copies of the conductor’s scores and orchestra parts. Sometimes the modifications were entered directly in the rental material, not by the publisher, but by various conductors, based on the published piano-vocal scores they and their singers used in rehearsal. The few extant early conducting scores are filled with handwritten changes made on the occasion of particular performances (in accordance with more recent printed piano-vocal scores).
In several instances, changes in the published piano-vocal scores can be reconstructed from existing autograph sources. In many other cases, there is no known evidence in Puccini’s handwriting to explain the different versions of the piano-vocal scores. Nevertheless, one has to assume that the piano-vocal scores published in Puccini’s lifetime for the most part represent authentic sources. They were intended for the broadest possible circulation, and Puccini would not have condoned large-scale departures from his actual intentions. Of course, this does not answer the question of Puccini’s authentic text in particular instances. It is striking, for example, that (quite apart from obvious publisher’s errors) the full scores often have layers of corrections that are inconsistent with the versions of the same passages in the piano-vocal scores. Only an extensive comparison of the autographs and the contemporary printed material, in conjunction with a wealth of supplementary documentation (especially Puccini’s correspondence and the letters of his collaborators), may eventually yield editions that are philologically and musicologically valid. In spite of the prevailing uncertainties, it is very likely that such editions would result in considerable changes in the standard texts.Note: In the 1950s, there was a fierce controversy concerning the supposed or real differences between the printed editions and the autographs; see Vaughan and (for an opposing point of view) Gavazzeni. This controversy is definitely still of crucial importance for a scholarly critical edition of Puccini’s operas, even though Vaughan was very exaggerated in his hair-splitting and misguided in his absolutely high regard of the autographs, while the skilled conductor Gavazzeni too often disregarded the actual sense of Puccini’s autograph writing. In this catalogue, however, such details generally had to be ignored. The problems of a critical edition of Puccini’s operas are also discussed in Roger Parker. “Historical Introduction” and “Critical Commentary” in Giacomo Puccini, Tosca. Opera completa per canto e pianoforte (edited by Roger Parker, Milan: Ricordi, 1995), written when that editor was occupied with the preparation of a new edition of Tosca, which, though one of the easier cases, even then raised some unanswerable questions. Fundamental considerations pertaining to this problem can also be found in Fairtile 1996, 380-425.
This catalogue certainly cannot achieve such a goal, but it can prepare the way by describing the changes Puccini made or permitted in his operas during his lifetime. In this respect Hopkinson spoke of different “versions”, a term applicable at best to Edgar and La Rondine, and to a lesser extent perhaps to Le Villi and Madama Butterfly. Apart from these operas, however, the changes mostly involve cuts or additions at particular points and on various occasions, although Puccini sometimes revoked them at a later date.
In view of these considerations, I decided to catalogue only those printed editions that present distinctly different musical states of the work in question Note: Information about the differences between editions must be restricted to the most important points. The most detailed, if somewhat confused and not always accurate account is found in Fairtile 1996.That is why certain printings have been omitted if they only vary the wording on the title page (e.g. changes in the publisher’s address), but do not differ musically from earlier editions. Interested collectors should refer to Hopkinson 1968, who endeavoured to list all such particulars, even though there are several lacunae in his somewhat incomplete documentation of existing printings. On the other hand, such variants are sometimes useful for dating the editions, and of course have been employed for that purpose in the present volume.
In principle, however, caution is advisable, for it was common practice (not only with Ricordi) to leave the engraved plates unchanged as long as possible and to bring the front matter commercially up to date in terms of addresses, prices, etc., with every new printing. That is why it is impossible to deduce the precise date of the music from the date on the title page. In several cases, there are even blatant discrepancies.Note: See, for example, the absolutely incorrect title page in a copy of the third version of La Rondine (83.E.3). Moreover, one must be prepared to encounter publisher’s errors that occurred during the binding of the gatherings.Note: In I-Mr, there exists a piano-vocal score of Suor Angelica (84ⁱⁱ) in which the first pages of music have a French text while all the other pages are in Italian. This must certainly be treated as a publisher’s error rather than catalogued as a legitimate variant.
This is also the reason why I have refrained from presenting a bibliographically exact rendering of the printed title pages, as is frequently found in Hopkinson. Instead, I only reproduce the information that essentially characterizes the print or is sufficient to identify it and distinguish it from other printings. For the same reason, the total number of pages generally refers to the main body of the music, which most often starts with page “1”, and excludes the pages of front matter, which in most cases were left unnumbered by the publisher.Note: Occasionally, I have indicated the total number of pages by giving the front matter pages in roman and the music pages in arabic numerals.This is the easiest way to identify most of the various editions.
Concerning the order of the printed editions of the operas, the first problem is how to evaluate the relationship between the full scores and the piano-vocal scores. It is true that piano-vocal scores are secondary derivations from full scores, but in the case of Puccini’s operas they mirror the publication history of the work and serve to document many variants resulting from Puccini’s habitual process of making revisions. In general, however, the piano-vocal scores were issued earlier than the duplicated full scores, and they were much more frequently modified in new editions to accomodate Puccini’s changes. Despite some misgivings, I finally decided to retain Hopkinson’s method of treating full scores and piano-vocal scores on an equal basis and to identify them with the same catalogue number to signify that they are mutually related. Therefore, I have listed all copies of full scores known to me, regardless of how they were reproduced (i.e. by lithography from copyist’s scores or by engraving). I also opted to refrain from listing the various chorus and instrumental parts available as rental material, for they are of no particular historical importance.
The order of the printed editions is not rigidly chronological. Editions that represent virtually the same state of the respective opera, but are worth mentioning for various reasons, have been catalogued under the same arabic numeral, with a lowercase letter added for the piano-vocal scores and an uppercase letter for full scores (e.g. 74.E.7a and 74.E.7B). Later editions, and especially the posthumous ones, often contain numerous minor variants and discrepancies concerning tempo, articulation, or metronome markings. Occasionally those discrepancies were the result of carelessness on the part of the publisher, but sometimes they are merely baffling. As a result, none of the editions listed is truly identical to any of the others. Several editions were sold over many decades, as proven by Ricordi’s blind stamps.Note: The blind stamps varied in shape over the decades. In this catalogue, they have been recorded independently of their actual shape: by month and year, or by year only (e.g. “10.93” means October 1893). Blind stamps were embossed on the first pages of printed editions when the copies were bound or brought to market. They have nothing to do directly with the printing process and still less with the engraving of the edition in terms of its musical contents. Ricordi abandoned the practice of systematic blind stamping in the early 1940s, although occasionally one finds copies bearing blind stamps dating as late as the 1950s. Also, there are many earlier copies that lack blind stamps. Generally, they seem to be library copies that were deposited for registration purposes and not intended for the market. Of course, it is also possible that blind stamps were simply omitted by mistake in the manufacturing process. The conductor’s scores, which were likewise not intended for the market, lack blind stamps, and so, of course, do the editions from other publishers. Sometimes changes in the engraving can only be discovered by chance, there being no indication that the prints are not identical. Nor is it always possible to establish the authenticity of such changes.
Furthermore, in historically or editorially well-founded exceptional cases, I have listed editions issued by other publishers (e.g. in the United States, the former German Democratic Republic, Russia, or Japan) after the copyright had expired, or in blatant violation of it. Many of them are nothing more than reproductions of earlier Ricordi editions; their occasional differences are rarely of editorial importance and their editors certainly had no access to the autographs. On the other hand, Ricordi’s own foreign-language editions have been included, especially since some of them contain evidence of Puccini’s authentic revisions.
Differences between printed editions of the operas are generally cited by referring to the present-day rehearsal numbers, most of which match their earlier counterparts (that is why the current editions lack certain rehearsal numbers which were skipped by the publisher in passages that Puccini cut in the process of revision). The number of bars before or after a rehearsal number is indicated by the signs “+” or “-”; thus, “fig. 10+5” means 5 bars after rehearsal number 10. A series of consecutive bars is indicated by the symbol “/”; thus, “fig. 10+5/7” means bars 5 to 7 after rehearsal number 10, and “fig. 10-7/5” means bars 7 to 5 before rehearsal number 10. Since most of the rehearsal numbers are placed directly above a bar-line, for greater accuracy the bars immediately after, before, or at the rehearsal number are marked “+1” or “-1”, respectively. If for some reason a rehearsal number appears above the middle of a bar, that bar is likewise indicated by a + (or -)1.
Printed editions of separate numbers from the operas have been excluded from this catalogue, since including them would have resulted in an over-abundance of titles lacking in bibliographical interest. I also refrained from mentioning the multitude of arrangements for the most diverse combinations of instruments, as well as the editions for solo piano with many of the lyrics above the music and the vocal parts incorporated in the piano part (at one time a very popular way for people to enjoy playing operas at home). It also would have been disturbing and confusing to include such editions, since Ricordi rarely altered them to reflect changes in the respective opera and they frequently (many years later) conveyed versions that were long obsolete.Note: I-Lgp preserves a separate edition in German of Butterfly’s aria “Che tua madre”, corresponding to the version in74.E.7a with the copyright year 1910 [!]. Here the underlying Italian text is still in the completely different original version, which had already been extensively modified in 1906. Another striking example is the piano solo version of Turandot, which for years after the first performance was still sold with Alfano’s finale in the version of 91.E.1 (already changed before the première).
An important source for the identification of many editions are Ricordi’s so-called LibronNote: Literally: “large books.” (MRL; see Libroni). These volumes comprise a continuous handwritten list of Ricordi’s plate numbers from the beginnings of the firm to the present day, containing the composers’ names and the titles of their works. It was here that Ricordi’s plate numbers (some of which were determined very early) were and still are assigned to planned publications. Frequently, the names of the copyists in charge as well as dates, prices, and numbers of pages were added later. Unfortunately, all of that information refers only to the respective plate numbers; changes within the same plate number were documented at best by erasing the page numbers and prices and entering new ones in their place. The Libroni are thus helpful in dating first editions, foreign-language editions, and later editions with new plate numbers, but they are not useful where it concerns the frequent changes in Puccini’s operas, especially since many of the earlier plate numbers were retained throughout the work’s publication history. In recent years, Ricordi has taken the opposite approach and assigned different plate numbers to editions that differ only in their binding.
In this respect, it is important to understand what the dates in the Libroni actually signify. Until the 1930s, they were not always in accord with the sequence of plate numbers, i.e. they do not necessarily identify the date when the plate number was assigned (as is often asserted in the literature pertaining to Puccini and other Ricordi composers). Instead, they frequently indicate the date when technical production of the edition actually began. Some excellent examples of how the publisher proceeded are provided by Puccini’s highly popular operas, which for Ricordi also meant financial success. Starting with La Bohème (67), Ricordi chose round plate numbers for the piano-vocal scores at a rather early stage (the lower – and thereby skipped – plate numbers were subsequently assigned to works by other composers) and the actual date when work on the edition began was recorded later. Only from the 1940s do the dates vary for the rubrics “Catalogazione”, “Ordine di stampa [printing order]”, and “Pubblicazione,” with the latter two categories frequently left blank. Therefore, in all cases it is essential to observe these subtle distinctions of the dates in the Libroni while also taking into consideration all other relevant sources.
An example of such a source is Ricordi’s handwritten card file (MRK), which records the dates of printing and the quantities printed for particular plate numbers. This card file obviously refers only to copies produced for public sale and does not report printings issued in advance for copyright purposes (see below), for Puccini’s own use, or for other interested persons. Unfortunately, only part of this card file, which was in use until about 1950, has been preserved intact.
In many cases, the time frame for the identification of editions is simplified by Ricordi’s practice of depositing copies in various locations in accordance with national and international copyright regulations (see Fuld 1966, 20-21). In Italy, to obtain legal protection, printed works had to be deposited with the prefect of the province (in Puccini’s case, Milan), with additional mandatory copies of operas to be deposited at the Naples Conservatory and the Biblioteca Braidense in Milan. Such deposits had to be made within three months after the first publication or performance. Many of the prefect’s deposit copies were passed on to the Conservatorio di Santa Cecilia in Rome and to the Biblioteca Nazionale Centrale in Florence; the copies from the Braidense library were transferred to the Milan Conservatory.
Similar regulations stipulated that copies be deposited in France at the Bibliothèque nationale in Paris, and in the United Kingdom at the British Museum (now the British Library) in London (deposits were also required at several other libraries in the UK upon demand). As of 1891, the copyright regulations in the United States were particularly strict. There, copies had to be deposited in the Library of Congress in Washington, DC, by the day of publication at the very latest (this explains why US-Wc frequently possesses very early copies). No obligatory copyright deposits were required in Germany and Austria, but since those countries were of high commercial interest, Ricordi provided copies of their new publications to the most important libraries there as well, very early on just to be on the safe side. All of those deposit copies were obviously printed earlier than the copies listed in MRK.
This catalogue records all of the holdings of Puccini’s works in the following copyright depository libraries:
I-Fn
I-Mc
I-Nc
I-Rsc
A-Wn
D-B
D-Mbs
F-Pn Note: The Département de la musique of the Bibliothèque nationale de France now additionally preserves the former holdings of the Bibliothèque-musée de l’Opéra, the library of the Opéra-Comique, and portions of the library of the Conservatoire (although the physical location of the material is shared between the Bibliothèque nationale and the Bibliothèque-musée de l’Opéra). The catalogues of these two places are not identical and still use the old systems of shelf marks. Since it is uncertain whether this situation will change in the library’s new premises, all copies from those collections included in this catalogue have been listed as holdings of the Bibliothèque nationale de France (F-Pn).
GB-Lbl
US-Wc
Copies in other libraries are only recorded if they bear any distinguishing characteristics (such as an early blind stamp), if they are among the few extant copies of the respective edition, or if there is no copy of the edition in the copyright depository libraries of the relevant country. However, no claim to completeness is implied in such cases, since I could only provide first-hand accounts of the holdings in the libraries I actually visited.
For the United States, the extensive Puccini holdings in US-NYp have been completely recorded as an easily accessible alternative to US-Wc. Furthermore, the most important copies of the Puccini collection in US-R (Ruth T. Watanabe Special Collections) are listed as well. The need to include those two collections was reinforced by the catastrophic physical condition of the older copies in US-Wc, which are normally only accessible on microfilm (however, I was generously permitted to see the originals). In addition, for ease in locating other holdings in the United States and Canada, I have also mentioned the printed copies listed in the National Union Catalogue (NUC), even though it is unfortunately frequently unreliable. Wherever possible, I confirmed the existence of the material reported to be preserved in the respective libraries.
I have also listed the copies in my own substantial collection, which includes, among other things, more than two-thirds of the piano-vocal scores listed in this catalogue. At some future time, my collection will be accessible for research at a suitable facility.
In general, library holdings of printed editions published after Puccini’s death have been excluded from this catalogue; exceptions have only been made if the respective editions were the first posthumous scores (such as the 8° full scores), or if their identification is facilitated by including them.
I have actually seen most of the copies mentioned in this volume; all other cases, i.e. where the bibliographical references have been taken from library catalogues or other secondary sources, are indicated by an asterisk (*) after the shelf mark. This is not to say, however, that all of the copies thus identified still exist. Many of the libraries in the various conservatories and other teaching institutions allowed (and sometimes still allow) historic piano-vocal scores to be used by students and faculty members for private study (singing lessons, etc.). Consequently, many of those scores have been destroyed by abuse. In other instances, copies have been misplaced or can no longer be traced by the library employees, or have been stolen, or (as in Berlin) have been missing since the last war. When I visited I-Nc in 1996, nearly one-third of the Puccini piano-vocal scores listed in its catalogue could not be found; in I-Mc, approximately one-sixth of the catalogued holdings are missing. Copies that could not be located for these or similar reasons have been marked “(-)”; other information concerning those copies (shelf marks, blind stamps, dates of registration) has been gathered from catalogues or other sources.
THE LIBRETTI
I have declined to provide a systematic description of the numerous libretti, some of which exist in partly handwritten early versions while others were frequently published for particular performances. Of course, such sources are germane to the history of an opera’s origins and revisions, and in that respect they have been taken into consideration as far as necessary. However, published libretti are not primary sources because their publication followed other considerations and reasons of copyright. Thus, many of Puccini’s opera libretti, some of which remained in print for years (or are still in print today), were disseminated by the publishers and their agents in their original form — even though portions of the texts had been significantly altered by changes made in the music. This particularly applies to libretti in foreign languages.Note: A German libretto of Manon Lescaut (64) (in I-Lgp) with the blind stamp of “10.28” (four years after Puccini’s death) still has the original text of 64.E.2b dating from the summer of 1893 and does not incorporate any of the many and diverse changes made in the Italian versions of the opera. Similarly, the extended text of the final duet of Turandot (91) was retained in the published German libretto for a very long time (see the copy with the blind stamp of “10.40” in I-Lgp). In principle, the libretti do not allow us to draw conclusions about the musical state of the relevant opera.
Dieter Schickling