An Ancient Crisis in Music Bibliography: The Need for Incipits

Beginning of article

To book-lovers, the name of Gutenberg is sacred. To music bibliographers, however, his magnificent achievement may contain an element of dissonance. Owing to sheer numbers, the methods of conservation for books have established conventions for music as well. Yet the unthinking application of book handling to music manuscripts (and probably to some music prints as well) has raised a host of problems. Through the centuries, the domination of "book-thinking" has instituted many practices unsuitable to manuscript material, including methods of binding, trimming, repairing, and, perhaps worst of all, misguided modernization. One would think it eminently desirable to replace old coversheets or boards with strong, clean, attactive folders--who could object to that? Unfortunately, the ragged covers so eagerly discarded often contain irreplaceable information (names of composers, copyists, and owners, often with dates; institutions that have collected the material, with their seals, leatherwork, library classification marks or numbers; performance locations, with dates and personnel involved; watermarks and other paper characteristics). A neglected class of evidence that might be called "analyzable dirt" may tell a small but important tale. For example, does the sand used for blotting ink found in many Venetian manuscripts come from the Lido or the mainland? With the development of neutron activation techniques this list could be greatly extended.

Returning to the ancient and continuing crisis in music bibliography: entries for printed books are typically arranged by author or title. If we apply this thinking to music manuscripts and prints, the result too often is a many-faceted question mark: what if the composer's name is missing; what if it is the wrong composer; what if the genre is incorrect; what if the key is misinterpreted? By comparison, the relatively happy uniformity of book title pages makes possible a genre of title-dependent bibliography utterly unsuitable for materials in which each item is unique, a rare situation in books, but relatively common in music manuscripts.

Music-bibliographical entries that use titles like "Symphony in D" as the identifying principle obviously create a quagmire of possibilities: which symphony in D is involved? We must often examine several items to find out.(1) What if a work in D-minor has mistakenly crept into the list? And what do we do if the item is temporarily misplaced, circulated to a reader, or on vacation at the bindery? We may never find out which is which. In great works of bibliography such as Eitner's Quellen-Lexikon, many entries contain the same title listed both as print and manuscript.(2) Is it actually the same work? We must examine both versions to know the truth. The British Union Catalogue of Early Music, a highly sophisticated author/title bibliography, still leaves us nearly helpless in areas such as collections, duplicate titles, and doubtful works.(3) Patient and affluent researchers in other countries may attempt to penetrate this thicket of dysinformation by sending for the relevant microfilm, only to discover that the photographers themselves sometimes cannot find the right work. The same conventional title may have been used for several works, or worse still, applied to a different work in the same key. One often feels a schizophrenia of admiration for the enormous effort that lies behind a large work of bibliography, at the same time combined with despair that its title identifications often do not suffice to locate a particular musical work. (Even the principle of organization by title can break down: the most nerve-wracking word in German bibliography is "Sammelwerke." And in Italian catalogues, all kinds of vocal works--even arias--may sometimes be found under the frustratingly general title, "Inni," the word itself an easy source of misunderstanding, since it sounds and looks so different from "hymns.")

Experiences of this sort did not deter RISM from issuing volumes of titles without sufficient further musical identification. The waste of effort is monumental, first for the compiler (particularly at the stage of proof-reading), and later for the researcher. In 1965 Hans-Rudolf Durrenmatt, Murray Gould, and I designed a data card that we hoped could be coded by RISM fieldworkers and later collected and decoded at RISM-Zentrum by mark-sensing devices directly connected to a computer database.(4) Friedrich Blume, the head of RISM at that time, told me that a committee had decided this method was "too complicated for librarians," even though it would have eliminated the most difficult and error-producing step in the compilation process, at the same time adding a positive musical identification by incipit. This curious decision proved ultimately to be correct, but not because data cards were too difficult for librarians. Instead, the rapid advance of technology quickly outmoded data cards of all sorts. At present only one functioning card reader is said to survive in New York City--in a private museum.

Music librarians of the eighteenth century seem to have understood the need for a distinctive musical identification, and collections of that period frequently contained thematic catalogues, using a universally accepted convention of the opening bars of the first violin part as the operative "title" of a work. This incipit often appeared also on the folders or coversheets containing the orchestral and vocal parts, a powerful reason why such covers should never be discarded. Thematic catalogues continue to surface: in 1983 Alexander Weinmann and Michael Ladenburger discovered a valuable but long neglected cache of manuscripts in the monastery of Vorau (Styria).(5) Most important, the collection included a perfectly preserved thematic catalogue, enabling us to reconstruct the repertory even though not all the music has survived.

A final need for incipits occurs in dealing with eighteenth-century publishers, whose egos demanded new opus numbers for reissues of works published elsewhere. Who would suspect from a series of printed titles that a Vanhal symphony printed by Andre as Op. 6, No. 1 is the same work as Hummel's Op. 9, No. 1? Such discrepancies create nightmares for cataloguers when they work from source-titles without incipits.

But if we can agree on the need for incipits, we must still decide what kind of incipits. For various reasons, particularly cost and practicality of usage, I decided to identify symphonies by an alphabetic incipit composed of the note-names beginning the first violin part, omitting rhythm and register. This has frustrated many readers who naturally would prefer the music. But the main purpose of the Catalogue had always been to untangle the numerous embranglements of identification. For this purpose, the alphabetic "name" suffices. Furthermore, in discussing disputed attributions and other endemic diseases of symphonic research, the alphabetic form requires no special software or spacing.

With the proliferation and constant improvement of music software, many researchers seem to assume that full staff notation provides the best solution. But for two reasons this may not be true. First, the relatively large amount of space required by music staves (I cannot resist the oxymoronic adjective: "space-intensive"). Hence, as I approached the publication of the …