By Alexander Dean
A critical edition is concerned with faithfulness to a source, and its authenticity, along with the probity of the editors involved, is bound to an understanding that the source material has been adequately and conscientiously accounted for. But authenticity is not an absolute virtue; if pursued blindly or dogmatically, the “best” can become the enemy of the “good.”
To begin with, some examples in the spirit of reductio ad absurdum: Recent Researches volumes, being modern editions in modern notation, will not reproduce the typeface of an original title page, nor will they mimic the size and shape of noteheads, the length of stems, stains from spilled wine, ragged edges of pages, etc. (Although any of these phenomena can be potentially interesting—even a wine stain can impart information! That discussion, however, belongs in the prose, and perhaps a plate, rather than in the score.)
But any source will present elements that fall into a gray area, not at once sliding into their place in even the most carefully constructed pre-transcription editorial methodology. Prime among these are layout-specific elements: those numbers, directives, and graphical notations that serve in manuscripts and early music prints to guide readers and performers safely from one page to the next. Since, in the translation to a modern edition, the layout will necessarily change, one might be tempted to dismiss any such marking out of hand, along with the source’s page numbers and other obvious candidates for tacit removal. While this is not a bad rule of thumb, at least to start out, each type of notation will need to be evaluated on its own.
First, consider that in making a transcription we are translating the layout, not discarding it. A modern book, whether it be musical notation or prose, follows a standard format from cover to cover.[1] Our familiarity with these conventions, their ubiquity, and the clarity of presentation allowed by modern print technology combine to provide us with a ready means of guidance, one not available in earlier written artifacts. Some textual and symbolic notation in early sources falls into this category: just as typeface and musical notation is translated into the commonly understood modern forms, these elements are translated into the conventions of modern layout. In these cases, little editorial fanfare is involved, since the source notation has not disappeared, nor been substantively changed. It is there, in a different form, doing the same job. In other cases, however, the process of translation leaves source notation hanging in the wind, so to speak, its job in the source material no longer needed but its role in the modern edition neither convincingly obviated nor clearly replaced. To discuss these situations, we will need to turn to some specific cases.
Page Numbers
Page numbers are clearly specific to the source in hand, with the modern edition using its own; the source page numbers are given as part of the standard bibliographic information, especially noting any discrepancies, errors, or conflicting numbering systems in the source, so that future scholars can save themselves the time of figuring such things out all over again each time. Related to the page numbering are page turn directives in a source: “verte” (turn), “verte cito” (turn quickly), and such. Again, these are obviously not going to be reproduced in a source that does not reproduce the same page breaks, and even if they were to fall at the same spot, modern clarity of presentation renders this unnecessary—we know that we should turn the page to get to the rest of the piece, if it’s not over yet (if you still read newspapers or magazines in hard copy, however, you can see similar directives to this day: “cont. on p. 2,” etc.). And given that the function of these directives, which is to guide the reader through each piece successfully, has been taken over by the table of contents, the page numbering, and the presentation of the modern book, they do not need to be described in the critical notes. They fall into the category of translation, rather than omission. Of course, if there is anything erroneous, confusing, or interesting about them, that puts them into a different category, one deserving of critical comment.
Other Page Layout Notations
But what about the directive “segue” and its cousins? Appearing between two apparently independent movements in a source, it can impart performance information that is not otherwise obvious: unlike “verte,” which usually only means “this portion of the piece is not over yet,” the meaning of “segue” can be as specific as “don’t wait too long to start the next piece,” “don’t put your bow down,” “keep looking at the conductor,” etc. Does your modern transcription already convey this information? If not, keep the “segue,” in the spirit of “if it works, don’t fix it.” And consider whether the two pieces should be separated by a double barline rather than a final barline, whether the measure numbering should start anew, and whether a full complement of title information should appear above it, since your translation of the source “segue” also involves all of these modern layout decisions.
System and staff-line breaks are also part of music layout and breed their own layout-specific notations. Custodes (also called directs) are the obvious examples, appearing throughout centuries of notated music at the ends of staves, an aid to the eye in finding one’s place in the following system. As with page numbers, there are few instances where the custos would need replication or even comment in a modern edition. But other system or line break symbols work together with repeat notation or other types of direction in ways that are not as clearly translated to the modern staff. For example, a small figure of a pointing finger, known as an “index” or “manicule” (or even “pointing finger symbol,” if you want to be blunt about it), can signify a point of return in a song’s repeat structure. Usually reserved for text, it can appear in music when the text is the main notational element, for example in the “alfabeto-text” guitar song notation. These songs tend to follow any number of traditionally understood repeat structures (such as the estribillo-copla format of the villancico), which may yet present ambiguities for our interpretation (see figure 1). In case of ambiguity, of course, one needs critical commentary, and an informed decision in the score, translating the manicule into modern repeat notation.
Figure 1. Manicule symbol in a seventeenth-century manuscript songbook, showing the point of repeat in the villancico “Dezia la moza al cura” (before the fourth line, “era sordo y daba en ella"). Florence, Biblioteca Riccardiana, MS 2793.
The manicule makes its appearance in other contexts as well. Antonella D’Ovidio’s recent edition of Lelio Colista’s trio sonatas reproduces some unusual source notation for improvisational sections within the trio sonatas. In the sources, stemless noteheads giving the continuo harmonies are given at the bottom of the parts, and notation within the staff indicates the spot where the performer is to improvise over the given harmonies, each in turn (see figure 2). In this case Dr. D’Ovidio chose to notate the improvisatory sections within the score (so there was no need to retain the manicule in the edition), but the critical notes for each piece describe the source notation, and she also includes a discussion of these sections in her introduction.
Figure 2. Lelio Colista, trio sonata no. 4, first violin part. Münster Diözesanbibliothek, Santini Sammlung, Hs. 1152.
“Solo,” “Tutti,” “Ripieno,” and Other Scoring Directives
At first glance these would seem to be performance indications, pure and simple, to be included in an edition no matter how different the page layout between source and transcription. But things are not always this simple. For example, it is not at all unusual to have both score and parts kept together in a library holding, and scores do not always provide the scoring information that is given by the partbooks; that is, the score may merely give the complete lines for the instruments, and only in looking at the titles of each partbook does one see “concerto” and “ripieno” labels. In such a case, assuming that the scoring suggested by the partbooks is chosen for the edition, the directives must be added to the score, best done through a statement in the editorial methods.
The converse may be true when transcribing from source partbooks alone into a score, especially in the basso continuo. The continuo partbook in seventeenth- and eighteenth-century sources often includes scoring indications along with the figures. In the continuo, “solo” and “tutti” may not refer to the continuo group at all but instead be warning the continuo player of what the other musicians are up to. And some partbooks can get even more specific, labeling entrances by specific instruments or giving text incipits for sections of vocal music. But if your continuo line is part of a score, none of these are necessary, since the player (or reader) can simply look up at the page and see what everyone else is doing. This was the case in Luca della Libera’s edition of Alessandro Melani, Music for the Pauline Chapel of Santa Maria Maggiore (see figure 3).
Figure 3. Alessandro Melani, “Litanie per la Beata Vergine,” continuo part, excerpt, showing “segue,” “solo,” “tutti,” and text incipit markings. Rome, Archivio della Basilica di Santa Maria Maggiore, 74/9.
And the different practical priorities at work between the past and present will require different approaches to layout as well, with their own specific notation. Everybody likes to save paper, but saving time and effort (and therefore cost) depends on the technology in use. A copyist will often save effort in a manuscript score when parts double each other by using directives such as “col basso” in lieu of writing out the doubled pitches. Nowadays it does not cost us much to cut and paste from one part to another, so duplicate parts can be realized with a general editorial methods statement rather than critical notes for each instance—unless there is some ambiguity in the source that needs discussion.
Modern software also allows paper-saving on our part by making shared staves easier to engrave; this requires the addition of the directives “1,” “2,” and “a 2” in the modern edition, for shared staves, as a representation of the source partbooks and/or score. Since these are necessary to give the same information that the source gives in a different form, and since there is no reason why the source, which uses a different layout, would have them, there is no need for individual critical notes for these directives either. Instead, the editorial methods should briefly describe the overall situation and note that directives have been added tacitly. In both these cases the critical notes can be reserved for situations where the modern transcription differs from the source reading in content rather than in the means of presenting that content.
Cuts, Inserts, and Discontinuities
Anyone who has worked with manuscript opera scores will be familiar with this situation: a score that has long passages marked out, instructions in the margins pointing to different pages, or new pages pasted in atop the initial notation. In the simplest of cases, for example when a score and partbooks survive in a single holding, and the notation, handwriting, materials, and historical context all combine to imply that all corrections to the music were made in a restricted timeframe by a restricted group of people (such as the composer and the music director of a single performance), then a modern transcription need only follow the map laid out by those corrections in the source.
But historical artifacts are rarely this obliging, so usually it takes a certain amount of musicological detective work to honor the existing materials, and critical commentary is part of the detective work. In David J. Buch’s edition of Die verdeckten Sachen (a Schikaneder production staged in Vienna in 1789) for instance, the main source is a collection of manuscript instrumental parts along with a promptbook giving the vocal lines. One of the arias, however, “Au wed’l, au wed’l, der Drache is mein,” appears in this source only as an extra page with partial lyrics. In order to fill out the music, Dr. Buch referred to other manuscript sources (which transmitted variant lyrics); all this had to be carefully documented in the critical report.
And in some cases there are multiple versions of a piece imminent to greater or lesser degrees in the surviving materials. For example, if original notation can still be read under strikethroughs, or if parts were altered after being copied from a score, or if a print, having been produced as part of the promotion for an opera, gives a reading that predates the corrections made during rehearsals and production for the performance. These cases can be a test of the modern editor’s mettle: do we reproduce all the notation simply because it is there? It certainly is tempting to choose to reproduce absolutely everything that survives, since all of that music represents the hard-won rewards of doing the research that made the edition possible in the first place. But sometimes the surviving notation is the tip of an iceberg, with different versions of a piece barely visible below the surface; in such cases you will need to make tough decisions.
If there is an earlier layer of notation discernable beneath a cut, does the cut passage make sense in the context of the other changes? Does the early layer represent a version that can be recreated in its entirety using the rest of the source? If not, then the cut passage should probably be separated out, for example in an appendix to the main edition. Or, conversely, if the original notation is more integral to the rest of the surviving notation, then the added material is better placed on its own in the appendix. In Andrew Levin’s edition of Joseph Schubert’s (no relation to Franz) Viola Concerto in E-flat Major, his source material contained violin parts in three different scribal hands. The set of parts bore a title page with a scoring much reduced from the total of parts within the set; that is, although the parts included clarinets, bassoons, and trumpet, the title page listed only flutes, oboes, and horns (along with the strings). The string parts copied first (as deduced by Levin), by scribe A, supported a version of the concerto with lighter scoring: they included cue notes for the winds in spots, and those cue notes sometimes gave “oboe” for clarinet entrances. Apparently, the clarinets were added after the first string parts were copied. But there are string parts in two other hands (B and C), and these contain revisions and newly composed passages, some of which are also added as corrections to the A parts. The “new” wind parts, the clarinets, trumpets, and bassoons, also match the “new” B and C string parts. The solo part for viola is in hand A, and like the A string parts, it transmits the “early” version of the concerto, with many corrections, including pasted-in and crossed-out passages reflecting the new material (see figure 4).
Figure 4. Joseph Schubert, Viola Concerto in E-flat Major, viola solo part, excerpt, showing added passages. Dresden, Sächsische Landesbibliothek – Staats- und Universitätsbibliothek, Mus. 3938-O-1.
So, with a bit of detective work, why not pull the earlier, more lightly orchestrated version of the concerto out of these materials? Maybe both versions could be presented as in the critical edition, the “early” and the “final.” As Levin points out, the change in orchestration in a piece performed in 1800 reflects the changes in the Viennese model of concerto performance overall, and therefore such an attempt is potentially of scholarly value. But there are problems with this idea. For one, all the surviving wind parts reflect the later version of the concerto; they transmit the corrections to the A string parts. This means that the “early” oboe and flute parts, the ones indicated by the cues in the A string parts, do not exist. They would need to be reconstructed from the later versions. Also, and more telling, is the issue that my scare quotes have been hinting at: the apparent chronology from “early” to “late.” Although the wind parts were clearly copied after scribe A copied the string parts, and after Schubert revised the score with the new passages, they were also copied by scribe A; in addition, scribe A also made corrections to the B and C string parts. In other words, the whole process took place in a restricted environment, both chronologically and in terms of the people involved. As Levin puts it, “It seems unlikely that the revisions indicate either a large chronological gap or a challenge to the composer’s intentions. In addition, it is unclear to what extent any of the intermediate stages represent a performed, performable, of complete version of the piece” (xii).
These kinds of decisions are not limited to manuscripts. Michael Noone and Graeme Skinner’s edition of Sebastián de Vivanco’s Liber magnificarum (1607) worked from a single printed source. However, there are differences between the surviving exemplars of this print. Some are minor typographical adjustments presumably made during the print run, but the most substantial is a complete reworking of one major section of polyphony: the “Deposuit potentes” verse of the first of the eighteen Magnificat settings. The reworked version of this verse comprises two folios that survive in a single copy of the print at Toledo Cathedral. Two pages from these folios were damaged and replaced, at some point near the end of the eighteenth century, with manuscript copies in the hand of the Toledo Cathdral scribe. All of this is detailed in Noone and Skinner’s critical report, and the version of “Deposuit potentes,” as revised by Vivanco, copied by the cathedral scribe, and reconstructed by Noone and Skinner, appears in full as a musical example within the critical notes (269–70).
What Else?
Every project brings its own particular challenges; a transcription necessarily involves decisions about layout, and there are many more than I have listed here. What have you found in yours? Anything surprising? How did you deal with it? Is there anything you have not decided on yet? Leave a comment and let us know!
Alexander Dean is the Managing Editor for Recent Researches at A-R Editions.
[1] See “The Parts of a Book,” in The Chicago Manual of Style, 17th ed. (Chicago: The University of Chicago Press, 2017), 4–38.
