The preparation of music for publication, performance or study, usually by someone other than the composer. ‘The ideal edition need not have all the answers but should control all the questions so that users can feel themselves in possession of the best available knowledge about this music’, wrote Joel Sheveloff (1986) about the keyboard music of Domenico Scarlatti; his comment is readily extrapolated to the wider problem of editing art music in the Western tradition. This article attempts to outline the questions editors might ask about the music they edit, and by so doing address two goals: to unmask some of the ‘critically based assumptions and perceptions that usually go unacknowledged’ in editing (Brett, 1988), for the benefit of users of editions, and to outline a generalized theory for the editing of Western art music, most of which is closely linked with a written tradition. Musics of other cultures, especially those in which an oral tradition predominates, pose different problems for the editor. Editors in ethnomusicology have developed conventions of their own, particularly in regard to notation, that establish their work as an independent field.
2. Principles of critical editing.
JAMES GRIER
Musicology can claim an illustrious history of editorial practice. Since the formation of the Bach-Gesellschaft in 1850, for the production of a complete edition of the music of J.S. Bach, musicologists have produced an enormous quantity of distinguished editions, from the collected editions of most important composers to the monumental series and national collections. Much of this enterprise was driven by the sheer necessity of making the music accessible. But an element in the undertaking was the creation of a canon, a central core of repertory, whose texts carried the same philological weight as their counterparts in literature and political history. These editions constitute a statement, by the purveyors of the young academic discipline of music, of the seriousness and worthiness of their discipline within the academy. Even their presentation, in imposing folio volumes, reflects the gravity of their intent.
Guido Adler's discussion of editing (1919) highlights the concerns of this enterprise. Although he made stimulating comments about the role of style in evaluating variants, and the need for critical appraisal of sources, he devoted most of his attention to technical matters such as the modernization of notation and modes of indicating editorial intervention. Adler assumed that music editors employed philological methods borrowed from literary editing, and so focussed on problems with the scholarly presentation of music. Source study received much more prominence in a pamphlet by Max Friedländer (1922), who shows that a critical assessment of style provides the only guide for deciding between variant readings. Despite this promising start, no new contributions to the discourse appeared until after World War II, by which time the intellectual approach to editing had changed drastically. Musicologists were reacting to two trends in editing. The first was the production of ‘performing’ or interpretative editions, most commonly of keyboard music but also of music for solo instruments with keyboard accompaniment, and usually prepared by famous performers. Musicologists complained that the numerous performance instructions added by editors, such as tempo markings, dynamics, articulation, fingering and pedalling, obscured the original notation, and that, because little or no effort had been made to differentiate editorial marks from those in the source, users could not distinguish between them. Already in the last decade of the 19th century, the Königliche Akademie der Künste in Berlin was issuing editions that claimed to be free of such editorial intervention; their name for this type of edition was ‘Urtext’ (‘original text’).
Although that term is now largely discredited by scholars, the original conception was praiseworthy: to provide texts that allowed the composer's notation to speak for itself and to permit performers, especially students, to form their own interpretation based on that original text. The concept tended to become commercialized in the post-World War II period through the editions published by Henle and others (see Unverricht's discussion, Feder and Unverricht, 1959), of its unseemly use and the possible remedies afforded by the revision to West German copyright law of 1959.
The objections of the scholarly world to Urtext editions centre on the fact that they do not present what they purport to. One need not go beyond the two principal statements of the term's leading advocates. Günter Henle himself noted (1954) that sometimes an autograph and a first edition differ, in which case the editor must decide what to print; but such a text ceases to be the Urtext, the composer's own written text and becomes the editor's interpretation of it. Georg Feder (Feder and Unverricht, 1959) argues that Urtext editions must be critical editions, although he distinguishes them by their mode of presentation from the kinds of critical edition associated with the traditional collected editions and monumental series. Like Friedländer, he notes the necessity of source studies; and he affirms that Urtext editions are not what they aspire to be in his discussion of five common misunderstandings of the concept, when he notes that an Urtext edition superseded by subsequent scholarship is no longer an Urtext.
The second direction in post-World War II editing that provoked critical reaction centred on the new collected editions of the great composers. As research, largely enabled by the existence of the first wave of scholarly editions, contributed to a deeper knowledge of repertories and their sources, and critical appraisals of that knowledge continued, new editions were needed to keep pace with, and reflect, the latest developments. At the base of these projects lay a sharpening critical perspective. The original collected editions provided an enormous service to musical scholarship by bringing together, for the first time, the works of many important composers in a uniform edition. The post-World War II editions present substantial refinements in virtually every respect, resulting from several generations of research; these newer editions, too, already need revision and will continue to be challenged as research continues. These editions represent nodal points on the continually changing path of musical scholarship.
The initiatives of the 1950s refocussed attention on the preparation of scholarly critical editions that could also be used by performers. That approach stands in marked contrast to the attitude behind the first series of these editions, starting in the 19th century, which consisted largely of philological monuments and gave less attention to performance matters (although in many cases performing material was published in parallel). Musicologists responded to the challenges by considering the relationship between music of the past and the performer. Editors were urged to jettison some of the philological purity of their texts (old clefs, for example) in order to make editions more accessible to performers. At the same time, the need for critical intervention by the editor was recognized without, however, a detailed discussion of what that entailed.
These developments led almost incidentally to a consideration of the historical relationship between composer and performer. If that relationship affects editorial practice in the present, should editors not give weight to its nature in the past? A penetrating assessment of this issue came from Klaus Harro Hilzinger (1974), influenced by the ‘genetic’ editing of German philology. This approach emphasizes the processes through which a work comes into being instead of the reification of a particular state of the work. Hilzinger identified the promise this conception holds for scholarly editing in music by recognizing that convention occupies a central place in communication between composer and performer via the score. The reconstitution of conventions that governed music of the past requires a consideration of the work's historical context. The interpretative editions, for example, that motivated, in reaction, Urtext editions become primary sources for the reception of the work, a kind of oral history. Alongside these developments, largely confined to German-speaking authors, with a concern for editing rooted in the philological issues surrounding the preparation of the monumental editions since the mid-19th century, came another in the postwar years, in English-speaking nations, associated with the performance of early music. Its pragmatic approach took as a point of departure the creation of clear, usable editions of old music originally written in notation no longer familiar to practising musicians. These editors gave clear precedence to presentation over critical issues.
The one publication from this period to address criticism in editing is Walter Emery's pamphlet Editions and Musicians (1957). Emery begins by condemning ‘aesthetic and stylistic criticism’ and characterizes editing as ‘a quasi-science, and the more scientific it is, the better’, based on ‘palaeography and bibliography, and historical facts in general’. Some of his observations, however, arise from subjective, critical observations on musical style, rather than objective bibliographical, palaeographical or historical facts; he thereby shows that critical and aesthetic sense is essential to scholarly editing. The most important contribution to date, and the only one to consider the full range of critical issues, is Georg Feder's monograph Musikphilologie (1987). Two central aspects of Feder's treatment elicit discussion below; first, even though Feder realized that the entire editorial process required critical thought, he persisted in dividing the process into ‘lower’ (bibliographic and mechanical) and ‘higher’ (interpretative and critical) stages; secondly he implies throughout that the goal of editing is the determination of final compositional intentions, a view seen by literary critics as the ‘intentional fallacy’.
The present discussion takes the existing discourse as a point of departure for an examination of the critical aspects of editing. Because editing is critical, editions are interpretative and cannot claim to be definitive: no two editors will edit the same piece in precisely the same way. Every piece of music is created under a unique combination of cultural, social, historical and economic circumstances; an acknowledgement of those circumstances, and thus of the uniqueness of each creative product, affects the conception of all editorial projects – each piece, each source and each edition is a special case. A natural corollary is that different repertories require different editorial methods, or even that each edition calls for a unique approach. No set of guidelines could accommodate the plurality of solutions to each editorial problem. Every project generates the editorial procedures that best represent the editor's critical engagement with the subject of the edition and its sources.
There are four principles basic to the nature of editing: it is critical in nature; criticism, including editing, is based in historical inquiry; editing involves the critical evaluation of the semiotic import of the musical text, which is also a historical inquiry; and the final arbiter is the editor's conception of musical style, which again is rooted in historical understanding.
The first tenet arises from the rich tradition of textual criticism in philology. There, every editorial decision is taken in the context of the editor's understanding of the work as a whole; and that understanding can be achieved only through critical evaluation. The establishment of the text, then, far from being mechanical, forms part of the critical dialogue between scholar and work. The meaning of the work and the reading of the text are complementary and interdependent. Editing consists of a series of educated, critically informed choices, that is, the act of interpretation; it occurs at the intersection of the composer's authority and the editor's. Composers exert their authority over sources created by themselves or under their direct supervision, although it is affected and limited by the social, political and economic institutions through which the sources are produced and disseminated. It extends, at least indirectly, to sources whose production they do not directly supervise, as the act of reproduction tacitly acknowledges. When editors come to evaluate both types of source, they apply their own authority in forming judgments about what the sources transmit. In some cases, they must call into question the accuracy of a reading in a source. This is the point of interaction between the authority of the composer, as transmitted in the sources, and that of the editor in the course of evaluating and interpreting them. Editing therefore requires a balance between these authorities; the exact balance in any edition is the product of the editor's critical perspective on the piece and its sources, and that perspective is rooted in an appreciation of the piece's historical context.
The act of communicating the work to an audience is a fully integrated part of the creative process. By entering into this dialogue, artists abandon their autonomy and shape the work to accommodate and facilitate the act of communication. The context, social, cultural, political and economic, impinges on the final form and meaning of the work, which can be understood only as a social artefact. The same applies to the sources. Each source attests a particular historical state of the work; the editor assesses that evidence against the background of the larger historical context in which the piece was created, and the resulting text reflects his or her conception of the piece as it existed in its historical and social environment. Thus each source and each reading is considered as an individual piece of evidence for the work's history. Nevertheless, the piece begins from the composer's original conception, and through all its metamorphoses during the process of its socialization, it is the composer who is responsible for its shape. When textual criticism is undertaken within a historical frame of reference, it discerns the possible influences on the composer and how they are reflected in the sources.
This line of critical and historical inquiry focusses on the relationship between the text of a musical work and the work itself. A written text is not self-sufficient; text and work are not synonymous. For most of the Western art tradition, the act of creating a musical work consists of two stages, composing (usually synonymous with the inscription of the score) and performance. These two steps create a distinction between the work, which depends equally on the score and performance for its existence, and the text, either written (a score) or sounding (a performance) that defines a particular state of the work. The work thus exists in a potentially infinite number of states, whether in writing (the score) or in sound.
The written text, however, holds a central place in our understanding of the work. It is the principal concern of editing, which begins and ends with this physical entity. The editors' critical position observes the distinctions between work and text and between written and sounding texts, and many of the editors' critical decisions depend on their understanding of the work. Nevertheless, editing depends principally on the source texts, and ultimately a text is its product.
Notation also carries a distinctive type of meaning, for the musical text addresses not the listener but the performer. This individual, even if the composer, is a second intermediary between the work and its audience through the medium of performance; and the text functions, initially, as the means of communication between composer and performer. This relationship clarifies the nature of a musical text: it contains a set of instructions to the performer for the execution of the work transmitted. The instructions vary in specificity, but in the Western art tradition at least some details remain at the performer's discretion. Trained musicians can imagine the sounds indicated, but that is not equivalent to reading: it is the aural replication of a performance, and the silent score-reader must interpret the notational symbols just as a performer does to re-create the work.
The nature of these interpretative processes emerges from consideration of the manner in which notation communicates. The individual symbol carries no independent meaning: its meaning arises solely from its context, from which it derives its semiotic import. This in turn derives from two complementary factors that form the semiotic framework for the notation. The first is convention; that is, the assignment of a particular meaning to a specific symbol is arbitrary. There is nothing intrinsic about the addition of a stem that requires a minim to last half as long as a semibreve but convention dictates it. The second factor is the system within which individual signs operate: the relationships between signs, the ways in which their meanings depend on the significance of one another. But more than one morphological system exists, and different systems operate within different conventions. In the common practice period, duple subdivision is assumed and triple must be indicated with a dot. In 15th- and 16th-century notation, however, triple subdivision is indicated by the mensuration with the result that the dot is not obligatory. These two systems, morphologically similar, employ different conventions for indicating triple subdivision; again, both conventions are arbitrary.
Each musical sign, therefore, carries a significance dependent on context and convention. Composers are aware of this and fix the text of their work within a framework. Once the moment of inscription has passed, however, the particular context and conventions are subject to change, and new observers will bring their own set of conventions to the interpretation of the signs. So the interpretation of these signs, in performance (where the investigation of performing practice seeks to provide a guide to interpretation) or in criticism (of which editing is a branch), is a strictly historical issue, one equally rooted in the semiotic import of each sign. The interpreter must re-create, as far as possible, the historical context and conventions applicable to the text of the work, to understand the meaning of each symbol.
The third tenet of this approach calls for an investigation of the semiotic nature of musical notation, also a historical undertaking. In the context of a historical and semiotic investigation of a piece and its sources, editing depends on the editor's conception of the work's style. Taken together, the notational symbols and their semiotic meaning generate a piece's stylistic attributes. Because editing amounts to the fixing of those symbols for a given piece, style ultimately governs many of the final editorial choices. But style does not reside in the notation alone. Because notation permits some discretion to performers, the variability of performance can exert influence on the work's style. Alongside the text, then, the performing options engendered by each work are an equal partner in determining its style.
Style exists within a historical context, and its study is also a historical undertaking: it is influenced by function, genre, existing practice and feasibility of performance. The elements contributing to style appear in a variety of combinations, according to time, place, composer, genre and even the individual piece. Such considerations form a part of the historical investigation of the piece and govern the editor's critical evaluation of readings in its text. Ultimately, editing is a matter of the preference of one reading over another; all readings can be classified as good, possible ones or clear errors.
However, if readings are evaluated on the basis of conceptions of style that arise from the readings themselves, a ‘hereneutic circle’ exists. A point of entry, however, can be found, and it depends on the editor's critical acumen and his or her sensitivity to style. As style is defined, the position of individual readings within the developing conception continually changes. All readings are ‘good readings’ unless shown to be false on stylistic grounds. Good readings are not necessarily authorial; that distinction emerges from a consideration of all readings within the context of the work, the composer and related works and repertories.
Even apparently mechanical errors, such as a missing dot or an added beam that causes a bar to contain the wrong number of beats, are recognizable only because they violate conventional practices that in some measure define style. Because conceptions of style are constantly in flux, such judgments are rarely definitive or unequivocal. Moreover, there is no such thing as an ‘obvious error’. Some readings will be deemed impossible within a piece's stylistic boundaries. These are ‘clear’ errors because, on stylistic grounds, they cannot be reasonable readings. The difference between clear and obvious errors is more than semantic: the latter apparently require no explanation, but the former do. Moreover, all these terms are relative, and a clear error for one editor may be a good reading for another with a different stylistic conception of the piece.
The final category, reasonable competing readings – that is, within the boundaries of the work's style – is often ignored. Textual critics, eager to establish an original or authorial text, have regarded all other readings as unoriginal, non-authorial and therefore errors. The common-error method of stemmatic filiation was introduced to provide a means of eliminating at least some of them. In many cases, editors simply have to choose between them, relying on their conception of the piece and its relationship with its sources.
Starting from this conceptual framework, a generalized theory for editing can be proposed, within which each editor can develop a particular methodology for the project at hand. While each repertory, even each piece, presents special challenges, there is a common group of problems that underlies the process of editing irrespective of the repertory. (i) What are the nature and the historical situation of a work's sources? (ii) how do they relate? (iii) from the evidence of the sources, what conclusions can be reached about the nature and the historical situation of the work? (iv) how do this evidence and these conclusions shape the editorial decisions made during the establishment of a text? and (v) what is the most effective way of presenting the text? The remainder of this article addresses these stages in the editorial task, examining the ways in which critical thought affects each phase.
Most critical editions are founded on a thorough knowledge of the source materials. The recent collected editions of the works of Bach, Haydn and Mozart, among others, attest to the value of source studies and also confirm that further source research will only enhance our understanding of the music, its creators and practitioners. No edition, however – existing, projected or future – is definitive. New investigations, even of well-known sources, will continue to yield new insights into the music in proportion to the imagination and erudition of the investigators. All sources are both historical documents and repositories of readings. Each source, as a physical artefact, originated in a particular historical context, which directly affects the value and significance of the source for the history of the music it transmits. The authenticity of individual readings, however, still needs verification, regardless of the source's authenticity, in establishing a text: not every reading in a given source carries equal merit.
Any investigation is affected by two features of musical sources: almost all are practical, functional documents, and their production, manuscript or printed, requires specialized, technical knowledge of notation. Musical scores enable performance, and most sources are created for use as performing materials or to serve as an intermediate stage in the production of printed performing materials. There are exceptions, such as the presentation manuscripts prepared under the supervision of Guillaume de Machaut, which contain his collected works, or, some would say, the series of collected editions undertaken in the second half of the 19th century; but these are few. The functional nature of musical sources, as opposed to other types of books (literary, historical or philosophical, for example), is demonstrated by their impermanence. Scraps of music frequently turn up as binding material and end-papers in non-music books: when these sources outlived their usefulness, and their repertories became so outdated or expanded so much that a new book was needed, the obsolete books were destroyed and recycled.
Source research entails gathering the evidence, classifying the sources and evaluating the readings to establish the text. The first of these involves location, inspection, description and transcription. Circumstances usually dictate that the initial work of transcription is undertaken from microfilm or some other form of photographic reproduction; but photography can never reproduce all the details required by an editor and much of the detailed investigation, particularly inspection and description, must await examination of the source itself.
Modern bibliographic resources greatly facilitate the location of sources for the researcher. The editor will then need to determine which ones deserve closer consideration; the more he or she knows of the text's tradition, the better informed any judgment will be at the stage of establishing the text. Printed materials provide special problems, since copies from a press run (even apart from subsequent impressions or editions) may differ in some details: printers make stop-press corrections, engraved plates deteriorate, pieces of movable type shift or fall out, and sheets from different runs may be bound together. It is possible for significant variation to enter a printed text, even within copies produced at the same time. A full understanding of a print's value thus depends on the examination of as many copies as possible to determine the bibliographical status of each and to establish the variability of the text.
The tasks of inspection and description primarily concern the physical state of the source: such evidence may establish or confirm specific historical facts about the source, which can affect its significance. For example, watermarks and the dimensions of ruled staves can aid its dating and identification. Few such details can be checked in photographic reproduction; the bulk of the inspection must take place in the source's repository.
In the descriptions to be published as part of the edition's introduction, form follows function. The minimum required is the positive identification of each source so that users can locate it for themselves, with full identification: for manuscript sources this includes the city and repository where the source is held, with its shelfmark; printed sources too require full bibliographic citation. For printed materials before 1800, the citation should indicate exactly which copies were consulted. Beyond this minimum, the context of the edition and its prospective audience determine the exact form of the description. It may be preferable to publish a full codicological or bibliographical description separately, especially if this led to a fuller discussion of the source's historical circumstances; but most users would appreciate a succinct account of the historical position of the sources.
The principal task when primary sources are used in editing is transcription. If this is initially done from photographic reproductions, details can be confirmed with the originals during the main inspection of the source. In photographic reproductions, shadows cast by pin-pricks look like noteheads, bleed-through or offsets may merge with text on the page, and holes in the paper permit the next or previous page to be read as part of the current one. Inspection under ultra-violet light can only be done in situ, although beta-radiography facilitates the reading of erasures in photographic reproductions.
Larger problems arise in transcription. No transcription is objective; yet editors need to maintain some distance between themselves and the music they are transcribing, to enable the source to speak for itself. Scholars are apt to form and impose their interpretations as they transcribe, imputing sense, reason and logic on the notational symbols; but that, regardless of its critical value, may distort the source's evidence and make it more difficult to assess its importance in the classification of the sources and the establishment of the text. A diplomatic transcription (one that records the information in the source exactly as it appears, with as many details as possible) alleviates the problem. Transcription, after all, is part of the process of gathering the evidence that will form the editor's conception of the work and its context. As editors gain experience with sources, they become aware of new interpretations.
Stemmatic filiation can provide a powerful tool for the historical assessment of readings and sources, but it does not constitute a mechanical method for reconstructing lost archetypes. In its simplest form, the common-error method is based on the assumption that, when several witnesses agree in the same error, it is reasonable to postulate that it arose from a single common ancestor – that the error was committed once and copied into surviving witnesses, rather than made by several scribes independently. It should be emphasized that only clear scribal errors are useful for determining filiation; the sharing of good readings, no matter how rare, cannot show stemmatic relationships. Errors are most probably transmitted from the source in which they first occur (usually below the authorial original in the stemma); so shared error will signify, in most cases, common descent from that first source – a deduction central to stemmatic determinations because it can distinguish the ancestry of two (or more) sources that agree in error against others; the sources that agree in error descend from an ancestor in which the shared error was made, and was unknown to the other sources. On the basis of a few scribal errors the editor can draw up a stemma codicum, a ‘genealogical table’ of sources, with the help of which many readings, including reasonable competing readings, can be eliminated from consideration, together with entire sources that can be shown to have been copied from a surviving one; this process is called the eliminatio codicum descriptorum (the elimination of sources that are direct copies).
Not all problems are solved by a stemma, however. The sources may divide evenly between two reasonable competing readings; in such a case editors must select one or the other. Other complications affect this method, including ‘contamination’, the consultation by a scribe of more than one exemplar, and ‘conjectural emendation’, whereby scribes, unsatisfied with the reading of the exemplar, introduce one of their own invention. In such situations problematic readings, which could have illuminated stemmatic relationships, tend to be replaced by reasonable readings, which carry no stemmatic weight. Further, unlikely though it is, two scribes might make the same error independently. Any stemma based on textual evidence alone is built on assumption and probability. As an interpretative tool, it depends on interpretation itself, starting with deciding exactly what constitutes an error. Thus if a stemma does not represent absolute, objective truth, stemmatic filiation nevertheless provides a powerful tool for the textual critic.
The usual purpose of the method's application in philology is to determine, as closely as possible, the text of an authorial original. Many works fit this paradigm, and the reconstruction of the composer's text is an important task and one that stemmatic filiation can assist, particularly when the autograph does not survive; examples are Bach's cello suites and Haydn's String Quartets op.33 (Grier, 1996). For much music in the Western art tradition, however, it is impossible to restrict the definition of the work to a discrete compositional moment; composers introduce flexibility of interpretation, in the form of performance, and each performance creates a new reading. A source created under these circumstances may transmit a possible text that carries no greater or lesser authority than others.
Because the relationship between the act of composition and the transmission of the resulting piece is infinitely variable, the procedure to be followed in treating the sources and their readings will also vary. No single editorial theory can satisfactorily accommodate the multiplicity of situations that arise in editing, even though each of the discussed theories of textual criticism has value in some contexts. Stemmatic filiation provides a useful and powerful tool, especially for the elimination of some competing readings, but does not automatically generate a fully edited text; it is simply a critical aid in sorting some of the readings. It may be possible to eradicate errors with good readings from elsewhere in the stemma. This type of reconstruction has been criticized because it creates a text that never existed, and a so-called ‘eclectic text’, which combines readings from two or more sources, is a historical impossibility. Adherents to this argument, principally the French philologist Joseph Bédier, devised the ‘best-text’ method of editing, in which one source is used except where it is patently corrupt.
Where it is corrupt, however, it must be emended. This raises the question: how should the emendation be effected? A stemma, if built on stylistically defensible criteria, may provide a firm historical basis for the emendation. Nevertheless, readings that stemmatically ascend to the archetype should not necessarily displace all unique readings in the sources. Many of them preserve substantive alterations to the text that have arisen through its performance and transmission, so represent the living tradition of the piece, and at least are typical of what would have been heard performed at the time when it was in circulation, even if they do not represent a specific performance. An edition that attempted to reproduce an ‘original’ or ‘definitive’ text, however, would have to ignore such readings in favour of the reconstructed text of the archetype, and it would not reflect the idiosyncratic musical practices each repertory exhibits.
Similarly, the theory of the copy-text, a method developed principally in modern English philology, does not generate a fully independent method of editing. The most familiar form of the theory is that proposed by W.W. Greg (1950–51) to deal with editing problems in Shakespeare, where virtually all sources are printed. Greg divided the transmitted readings into their substantive and accidental components. The former carry meaning, as for example the words of a text. The latter include such matters as spelling, punctuation and capitalization, qualities that may not in themselves carry meaning. Greg reasoned that, in publication, Shakespeare did not retain absolute control over accidentals, as printers imposed their own style and otherwise altered the text. Consequently, he suggested that the editor choose one text of the work as the copy-text and follow its accidentals faithfully. Therein lies the distinction between this method and the ‘best-text’ method, in which a single source supplies all readings, accidental and substantive. The choice of copy-text is determined by the editor's critical appraisal of the sources. The treatment of substantives is more flexible. Greg favoured the creation of an eclectic text by drawing on all sources of the work directly associated with its author.
Despite the virtues of its attempt to deal with the historical circumstances of publication, however, the method does not address the difficulty of creating an unequivocal definition of substantive and accidental. The physical presentation, the bibliographic codes, of the work and text can carry significant meaning (McGann, 1983). It is therefore impossible to make a meaningful distinction between accidental and substantive. The problems are exacerbated when we try to transfer these concepts to music, because the semiotic nature of musical notation makes the distinction more difficult. Any graphic aspect of notation can convey meaning. So the idea of selecting a copy-text whose accidentals are to be incorporated into the edited text, already problematic in literature, becomes virtually meaningless in music.
J. Bédier: ‘La tradition manuscrite du Lai de l'Ombre: Réflexions sur l'art d'éditer les anciens textes’, Romania, liv (1928), 161–96, 321–56 [repr. separately, 1929, 1970]
W.W. Greg: ‘The Rationale of Copy-Text’, Studies in Bibliography, iii (1950–51), 19–36
G. Pasquali: Storia della tradizione e critica del testo (Florence, 2/1952)
P. Maas: Textkritik (Leipzig, 4/1960; Eng. trans. of 3/1957, as Textual Criticism, 1958)
J. Willis: Latin Textual Criticism (Urbana, IL and London, 1972)
M.L. West: Textual Criticism and Editorial Technique Applicable to Greek and Latin Texts (Stuttgart, 1973)
E.J. Kenney: The Classical Text: Aspects of Editing in the Age of the Printed Book (Berkeley, CA and London, 1974)
F. Bowers: ‘Greg's “Rationale of Copy-Text” Revisited’, Studies in Bibliography, xxxi (1978), 90–161
M.D. Reeve: ‘Stemmatic Method: “Qualcosa che non funziona?”’, The Role of the Book in Medieval Culture: Oxford 1982, ed. P. Ganz (Turnhout, 1986), i, 57–69
A. Blecua: Manual de crítica textual (Madrid, 1983)
J.J. McGann: A Critique of Modern Textual Criticism (Chicago and London, 1983)
J.J. McGann, ed.: Textual Criticism and Literary Interpretation (Chicago and London, 1985)
M.D. Reeve: ‘Archetypes’, Sileno, xi (1985), 193–201
J. Grier: ‘Lachmann, Bédier and the Bipartite Stemma: Towards a Responsible Application of the Common-Error Method’, Revue d'histoire des textes, xviii (1988), 263–78
M.D. Reeve: ‘Eliminatio codicum descriptorum: a Methodological Problem’, Editing Greek and Latin Texts, ed. J.N. Grant (New York, 1989), 1–35
G.T. Tanselle: A Rationale of Textual Criticism (Philadelphia, 1989)
G.T. Tanselle: Textual Criticism and Scholarly Editing (Charlottesville, VA, and London, 1990)
P. Cohen, ed.: Devils and Angels: Textual Editing and Literary Theory (Charlottesville, VA, and London, 1991)
G.T. Tanselle: ‘Editing without a Copy-Text’, Studies in Bibliography, xlvii (1994), 1–22
RiemannL12
O. Jahn: ‘Beethoven und die Ausgabe seiner Werke’, Gesammelte Aufsätze über Musik (Leipzig, 1866), 271–337
F. Chrysander: ‘Die Verwendung der Schlüssel bei der Herausgabe älterer Musikwerke’, AMZ, v (1870), 356–9
G. Adler: Methode der Musikgeschichte (Leipzig, 1919)
H. Wetzel: ‘Uber Textkritik, Analyse und Bearbeitung von Musikwerken’, ZMw, ii (1919–20), 429–36
M. Friedländer: Ueber musikalische Herausgeberarbeit (Weimar, 1922) [rev. version of ‘Über die Herausgabe musikalischer Kunstwerke’, JbMP 1907, 13–33]
E. Beninger: ‘Pianistische Herausgebertechnik’, ZMw, xii (1929–30), 280–98
W. Apel: ‘The Partial Signatures in the Sources up to 1450’, AcM, x (1938), 1–13
J.P. Larsen: Die Haydn-Überlieferung (Copenhagen, 1939)
W. Altmann: ‘Ist die Originalhandschrift oder der Erstdruck massgebend?’, AMz, lxvii (1940), 243b–4b
W. Schmieder: ‘Nochmals: Originalhandschrift oder Erstdruck?’, AMz, lxvii (1940), 258b–9a
M. Unger: ‘Grundsätzliches über Revisionsausgaben von Beethovens Werken’, Die Musik, xxxiii (1940–41), 52–8
E.E. Lowinsky: ‘The Function of Conflicting Signatures in Early Polyphonic Music’, MQ, xxxi (1945), 227–60
A.H. King: ‘Some Recent Trends in Music Publishing and Bibliography’, MMR, lxxviii (1948), 14–18
W.S. Drew: ‘On Song Editing’, MT, xc (1949), 113–14
R. von Ficker: ‘Probleme der Editionstechnik mittelalterliche Musik’, IMSCR IV: Basle 1949, 108
W. Emery: ‘New Methods in Bach Editing’, MT, xci (1950), 297–9
W. Bittinger: Studien zur musikalischen Textkritik des mittelalterlichen Liedes (Würzburg, 1953)
R.H. Hoppin: ‘Partial Signatures and Musica Ficta in some Early 15th-Century Sources’, JAMS, vi (1953), 197–215
T. Dart: The Interpretation of Music (London, 1954, 4/1967)
G. Henle: ‘Über die Herausgabe von Urtexten’, Musica, viii (1954), 377–80
C.L. Cudworth: ‘Ye Olde Spuriosity Shoppe’, Notes, xii (1954–5), 25–40, 533–53
S. Clercx: ‘Les “accidents” sous-entendus et la transcription en notation moderne’, L'Ars Nova: Wégimont II 1955, 167–95
F. Oberborbeck: ‘Original und Bearbeitung’, Festschrift Max Schneider zum 80. Geburtstag, ed. W. Vetter (Leipzig, 1955), 347–55
T.G. Georgiades: ‘Zur Lasso-Gesamtausgabe’, Musikwissenschaftlicher Kongress: Vienna 1956, 216
F. Giegling: ‘Probleme der Neuen Mozart-Ausgabe’, SMz, xcvi (1956), 41–3
U. Aarburg: ‘Muster für die Editionstechnik mittelalterlicher Liedmelodien’, Mf, x (1957), 209–17
E. and P. Badura-Skoda: Mozart-Interpretation (Vienna, 1957; Eng. trans., 1962)
W. Emery: Editions and Musicians: a Survey of the Duties of Practical Musicians and Editors Towards the Classics (London, 1957)
P. Mies: Textkritische Untersuchungen bei Beethoven (Munich, 1957)
Le luth et sa musique: Neuilly-sur-Seine 1957
‘Editionsprobleme des späten 18. Jahrhunderts’, IMSCR VII: Cologne 1958, 349–54
G. von Dadelsen: ‘Friedrich Smends Ausgabe der h-moll-Messe von J.S. Bach’, Mf, xii (1959), 315–34
A. Dürr: ‘Wissenschaftliche Neuausgaben und der Praxis’, Musik und Kirche, xxix (1959), 77–82
G. Feder and H. Unverricht: ‘Urtext und Urtextausgaben’, Mf, xii (1959), 432–54
E.E. Lowinsky: ‘Early Scores in Manuscript’, JAMS, xiii (1960), 126–73
H. Unverricht: Die Eigenschriften und die Originalausgaben von Werken Beethovens in ihrer Bedeutung für die moderne Textkritik (Kassel, 1960)
A. Carapetyan: ‘Problems of Editing and Publishing Old Music’, MD, xv (1961), 5–14
G. von Dadelsen: ‘Die “Fassung letzter Hand” in der Musik’, AcM, xxxiii (1961), 1–14
J. LaRue: ‘Watermarks and Musicology’, AcM, xxxiii (1961), 120–46
D. Stevens: ‘Problems of Editing and Publishing Old Music’, IMSCR VIII: New York 1961, 150–58
T.G. Georgiades: Musik und Schrift (Munich, 1962)
W. Osthoff: ‘Per la notazione originale nelle pubblicazioni di musiche antiche e specialmente nella nuova edizione Monteverdi’, AcM, xxxiv (1962), 101–27
T. Dart, W. Emery and C. Morris: Editing Early Music: Notes on the Preparation of Printer's Copy (London, 1963)
A. Donato: Preparing Music Manuscript (Englewood Cliffs, NJ, 1963)
M. Reimann: ‘Zur Editionstechnik von Musik des 17. Jahrhunderts’, Norddeutsche und nordeuropäische Musik: Kiel 1963, 83–91
K. Vötterle: ‘Die Stunde der Gesamtausgabe’, Haus unterm Stern: über Entstehen, Zerstörung und Wiederaufbau des Bärenreiter-Werke (Kassel, 3/1963), 287–95 [rev. version of article in Musica, x (1956), 33–6]
G. von Dadelsen: ‘Über das Wechselspiel von Musik und Notation’, Festschrift Walter Gerstenberg zum 60. Geburtstag, ed. G. von Dadelsen and A. Holschneider (Wolfenbüttel and Zürich, 1964), 17–25
E. Karkoschka: Das Schriftbild der neuen Musik: Bestandsaufnahme neuer Notationssymbole, Anleitung zu deren Deutung, Realisation und Kritik (Celle, 1966; Eng. trans., 1972, as Notation in New Music: a Critical Guide to Interpretation and Realisation)
G. von Dadelsen, ed.: Editionsrichtlinien musikalischer Denkmäler und Gesamtausgaben (Kassel, 1967)
G. Feder: ‘Gedanken über den kritischen Apparat aus der Sicht der Haydn-Gesamtausgabe’, Colloquium amicorum: Joseph Schmidt-Görg zum 70. Geburtstag, ed. S. Kross and H. Schmidt (Bonn, 1967), 73–81
E.E. Lowinsky: ‘Editorial Problems’, The Medici Codex of 1518 (Chicago and London, 1968)
T.G. Georgiades, ed.: Musikalische Edition im Wandel des historischen Bewusstseins (Kassel, 1971)
L. Hoffmann-Erbrecht: ‘Problems in the Interdependence of Josquin Sources’, Josquin des Prez: New York 1971, 285–93
A. Porter: ‘Problems in Editing the Music of Josquin des Prez: a Critique of the First Edition and Proposals for the Second Edition’, ibid., 723–54
G. von Dadelsen: ‘Über den Wert musikalischer Textkritik’, Quellenstudien zur Musik: Wolfgang Schmieder zum 70. Geburtstag, ed. K. Dorfmüller and G. von Dadelsen (Frankfurt, 1972), 41–5
C. Dahlhaus: ‘Urtextausgaben’, NZM, Jg.134 (1973), 334
A. Chiarelli: ‘L'Incoronazione di Poppea o Il Nerone: Problemi di filologia testuale’, RIM, ix (1974), 117–51
K.H. Hilzinger: ‘Über kritische Edition literarischer und musikalisher Texte’, Euphorion, lxviii (1974), 198–210
H. Bennwitz and others, eds.: Musikalisches Erbe und Gegenwart: Musiker-Gesamtausgaben in der Bundesrepublik Deutschland (Kassel, 1975)
H. Risatti: New Music Vocabulary: a Guide to Notational Signs for Contemporary Music (Urbana, IL, and London, 1975)
W. Rehm: ‘Notenschrift und Aufführung: die Rolle der Musikverlage’, Notenschrift und Aufführung: Munich 1977, 99–120 [repr. in FAM, xxv (1978), 135–41]
G. von Dadelsen: ‘Die Crux der Nebensache: editorische und praktische Bemerkungen zu Bachs Artikulation’, BJb, lxiv (1978), 95–112
C. Dahlhaus: ‘Philologie und Rezeptionsgeschichte: Bemerkungen zur Theorie der Edition’, Festschrift Georg von Dadelsen, ed. T. Kohlhase and V. Scherliess (Neuhausen-Stuttgart, 1978), 45–58
C. Dahlhaus: ‘Zur Ideengeschichte musikalischer Editionsprinzipien’, FAM, xxv (1978), 19–27
J. Froger: ‘The Critical Edition of the Roman Gradual by the Monks of Solesmes’, JPMMS, l (1978), 81–97
M. Bente, ed.: Musik–Edition–Interpretation: Gedenkschrift Günter Henle (Munich, 1980)
A. Mendel: ‘The Purposes and Desirable Characteristics of Text-Critical Editions’, Modern Musical Scholarship, ed. E. Olleson (Stocksfield, 1980), 14–27
K. Stone: Music Notation in the Twentieth Century: a Practical Guidebook (New York and London, 1980)
A. Atlas: ‘Conflicting Attributions in Italian Sources of the Franco-Netherlandish Chanson, c.1465–c.1505: a Progress Report on a New Hypothesis’, Music in Medieval and Early Modern Europe: Patronage, Sources and Texts, ed. I. Fenlon (Cambridge, 1981), 249–93
M. Bent: ‘Some Criteria for Establishing Relationships between Sources of Late-Medieval Polyphony’, ibid., 295–317
S. Boorman: ‘Limitations and Extensions of Filiation Technique’, ibid., 319–46
L. Finscher, ed.: Quellenstudien zur Musik der Renaissance, i: Formen und Probleme der Überlieferung mehrstimmiger Musik im Zeitalter Josquins Desprez (Munich, 1981)
G. Feder: ‘Textkritische Methoden: Versuch eines Überblicks mit Bezug auf die Haydn-Gesamtausgabe’, Haydn-Studien, v (1982–3), 77–109
L. Finscher, ed.: Quellenstudien zur Musik der Renaissance, ii: Datierung und Filiation von Musikhandschriften der Josquin-Zeit (Wiesbaden, 1983)
R. Rastall: The Notation of Western Music: an Introduction (London, 1983)
S. Boorman: ‘The Uses of Filiation in Early Music’, Text: Transactions of the Society for Textual Scholarship, i (1984), 167–84
J. Caldwell: Editing Early Music (Oxford, 1985, 2/1995)
G. von Dadelsen: ‘Zur Geltung der Legatobögen bei Bach: eine Studie für Artikulationsfanatiker und editoren’, Festschrift Arno Forchert, ed. G. Allroggen and D. Altenburg (Kassel, 1986), 114–22
J. Sheveloff: ‘Domenico Scarlatti: Tercentenary Frustrations (Part II)’, MQ, lxxii (1986), 90–118
G. Feder: Musikphilologie: eine Einführung in die musikalische Textkritik, Hermeneutik und Editionstechnik (Darmstadt, 1987)
G. Heussenstamm: The Norton Manual of Music Notation (New York and London, 1987)
P. Brett: ‘Text, Context, and the Early Music Editor’, Authenticity and Early Music: a Symposium, ed. N. Kenyon (Oxford and New York, 1988), 83–114
M. Querbach: ‘Der konstruierte Ursprung: zur Problematik musikalischer Urtext-Ausgaben’, NZM, Jg.149/1 (1988), 15–21
G. Feder: ‘Das Autograph als Quelle wissenschaftlicher Edition’, Musikerautographe: Vienna 1989, 115–44
E. Herttrich: ‘Autograph–Edition–Interpretation’, ibid., 165–84
M. Bente: ‘Ermittlung und Vermittlung: zum Begriff “Urtext”’, ÖMz, xlvi (1991), 528–31
D. Berke: ‘Urtext zwischen Wissenschaftsanspruch und Praxisnähe: die modernen Musiker-Gesamtausgaben’, ÖMz, xlvi (1991), 531–5
R. Broude: ‘When Accidentals are Substantive: Applying Methodologies of Textual Criticism to Scholarly Editions of Music’, Text: Transactions of the Society for Textual Scholarship, v (1991), 105–20
C. Eisen: ‘The Old and New Mozart Editions’, EMc, xix (1991), 513–32
G.W. Gruber: ‘“Doppel”-Editionen für den interpretatorischen Freiraum’, ÖMz, xlvi (1991), 710–11
P. Gülke: ‘Philologie und musikalische Praxis’, ibid., 535–9
M. Mender: Music Manuscript Preparation: a Concise Guide (Metuchen, NJ, 1991)
J. Wallnig: ‘“Produkt im Kopf” contra “Produkt im Ohr”’, ÖMz, xlvi (1991), 539–42
Musik als Text: Freiburg 1993
M. Bent: ‘Editing Music: the Dilemma of Translation’, EMc, xxii (1994), 373–92
R. Borghi and P. Zappalà, eds.: L'edizione critica tra testo musicale e testo letterario (Lucca, 1995)
M. Caraci Vela: ‘Introduzione’, La critica del testo musicale: Metodi e problemi della filologia musicale, ed. M. Caraci Vela (Lucca, 1995), 3–35
J. Grier: The Critical Editing of Music: History, Method, and Practice (Cambridge, 1996)
M. Toffetti, ed.: Edizioni moderne di musica antica: Sei letture critiche (Lucca, 1997)