ISSN: 1089-747X
Copyright © 1995–2012 by the Board of Trustees of the University of Illinois

Volume 13, no. 1:

Henry Purcell’s Operas: The Complete Texts. Edited by Michael Burden. Oxford: Oxford University Press, 2000. [xvi, 528 pp. ISBN 0-19-816445-9. $293.]

Reviewed by Andrew R. Walkling*

1. Introduction: Michael Burden and the Advancement of Purcell Scholarship

2. Editorial Principles

3. Consistency

4. Accuracy

5. Conclusion


1. Introduction: Michael Burden and the Advancement of Purcell Scholarship

1.1 Following the surge in interest in the compositional output of Henry Purcell promoted by the tercentenary celebrations of 1995, a substantial amount of scholarly and editorial work on Purcell has continued to appear throughout the ensuing decade, providing significant benefits to all students of seventeenth-century English music. In this endeavor, Michael Burden deserves an ample share of the credit: as editor of The Purcell Companion (1994), Purcell Remembered (1995), two special issues of Early Music on “Music in Purcell’s London” (November 1995 and February 1996), Performing the Music of Henry Purcell (1996), A Woman Scorn’d: Responses to the Dido Myth (1998), and the volume under review here—not to mention being the author of several important articles and organizer of (as well as participant in) conferences, concerts, and symposia—Burden has assuredly left his mark on Purcell studies. Those of us who have been involved in one or more of these projects will readily attest to his energy and dedication, while the scholarly world at large has become increasingly aware of the substantive contributions afforded by work done both by him and under his aegis.

1.2 Part of Burden’s genius has lain in his desire to seek out the unfilled lacunae in Purcell scholarship. As he pithily remarked in 1995, “the scholarly community has singularly failed to address the question of what studies of Purcell we actually need, rather than simply producing what will do well in the bookshops.”1 This commitment to tackling the desiderata has resulted in an enormous enrichment of our knowledge of Purcell’s music and its contexts, in terms not only of new discoveries but also of greater ease of access to research material—whether information presented in tabular form, reprints of primary documents, or editions of Purcell-related texts. With respect to the latter, it is the texts of Purcell’s operatic works, in particular those spoken-and-sung hybrids variously labeled “semi-operas,” “dramatick operas,” or “ambigues,” that have proven especially elusive. Thanks to their peculiar mixing of full-length play with irregularly-proportioned musical episodes, a difficulty compounded by the near-anonymity of many of their adaptor-librettists, few of these works have ever appeared in modern textual editions. As a result, scholars have been left to rely on contemporary imprints that, in nearly every case, exhibit a bewildering array of bibliographical problems and thus severely strain our efforts at providing reliable and consistent citations when discussing the works. Recognizing this deficiency, Burden has put together Henry Purcell’s Operas: The Complete Texts, a detailed scholarly edition of the four dramatic operas for which Purcell wrote substantial amounts of music—Dioclesian, King Arthur, The Fairy-Queen, and The Indian Queen—as well as his sole through-composed theatrical piece, Dido and Aeneas, a work that presents special textual challenges all its own. (The Dido chapter also includes an edition of Charles Gildon’s Measure for Measure adaptation of 1700, which provides another version of the Dido text.) Of course, the “British Orpheus” contributed to some fifty-six theatrical works all told, several of which—Circe, Oedipus, The Libertine, Bonduca, The Tempest, and The Comical History of Don Quixote—might similarly claim operatic status. The works included in the present volume are exceptional, however, not only in the amount of music Purcell provided for each, but in the fact of not being collaborative ventures, musically speaking. Only The Indian Queen includes a contribution from another composer: the concluding wedding masque set by Henry’s relative, Daniel Purcell.

1.3 The volume’s brief is, at least on first blush, relatively straightforward: to present each of the five operas in an accurate, usable edition that takes the bibliographical issues into account and provides some necessary context. There is no claim to offer earth-shaking interpretive insights, or even to rethink more than the occasional textual crux; instead, the idea is that a comprehensive accumulation of meticulously compiled empirical data should “encourage a properly critical response” (p. 31) to the texts and establish a solid foundation upon which to construct future edifices of Purcell scholarship. The organization of the book supports this essentialist methodology: Burden opens with an introduction in which, characteristically, he investigates some of the less explored aspects of the operas in order to fill in some gaps in our understanding; next, Andrew Pinnock, a frequent Burden collaborator, supplies an extensive essay on the bibliographical pitfalls to be found among the Purcellian operatic playbooks, laying out in the process the editorial premises upon which the volume is based. These two introductory essays are followed by the five textual editions, each prepared and introduced by a different author, and an appendix that presents (in a format different from that found in the main editions) solely the sung texts from each of the operas, as taken from the surviving musical scores. The volume concludes with a useful classified bibliography of relevant scholarship and, in the middle of the book, there are seventeen black-and-white plates illustrating various aspects of the performance history of the operas from the seventeenth to the twentieth century.

2. Editorial Principles

2.1 Essential to the premise of the entire endeavor is a set of editorial principles, presumably established by Burden in consultation with Pinnock, that are meant to guide the contributors and thus provide the volume’s procedural unity. This is by no means a simple matter: while three of the operas (Dioclesian, King Arthur, and The Fairy-Queen) were printed around the time of their performances as standard “playbooks”—albeit each with its own set of bibliographical complexities—the operatic version of The Indian Queen exists only in manuscript form and therefore must be approached differently. Moreover, Dido and Aeneas presents an entirely distinct set of problems: through-composed, carelessly printed in its earliest surviving textual source, reconfigured for the public theatre a decade later, and badly mangled in its musical sources, this more economical masque-like work demands yet another editorial method. Keeping in mind the difficulties that inhere in the materials, the volume aims to steer clear of any attempt to create putatively perfect versions through intrusive editorial emendation; instead, the objective is to select a single “ideal copy” (p. 88) of each work and transcribe that copy precisely, reproducing all of its typographical features and idiosyncrasies, including layout, font size, spelling and capitalization, and any errors (only the long “s,” italicized punctuation, and “VV” for “W” are silently altered). At the same time, modern aids are added, such as line-numbering, occasional explanatory remarks, and cross-references to significant variants found in other printed versions of each text. Only where “readings very irregular or doubtful” crop up are the individual editors authorized to supply emendations, with the “original (rejected or queried) readings [recorded] in footnotes” (p. 92).

2.2 As Pinnock explains, these rules are intended to follow as closely as practicable those set down in the early twentieth century by the Malone Society for its long-running series of “near-diplomatic transcriptions”—what might best be described as typeset facsimiles with editorial apparatus—of early modern dramatic texts. Obviously, the threshold for determining “readings very irregular or doubtful” is necessarily subjective, but one can certainly imagine that a carefully prepared, reliable edition would only occasionally have to alter the received text for purposes of essential clarity. Modern readers of seventeenth-century texts are generally comfortable with the age’s loose rules pertaining to orthography and punctuation and thus, under the editorial regimen established here, there is little need to interfere with lines such as “But here me, Oh Renown’d, Oh worthy Fiend, / The Favourite of our Cheif” (King Arthur, lines 747–8), or even the slightly garbled “How, Royal Fair, shall I impart? / The Gods decree, and tell you we must part” (Measure for Measure, lines 1109–10). The volume’s practice contrasts sharply with the more common “interventionist” approach normally associated with critical editions, where the editor emends at will in order to construct a presumptive replica of what the original authorial text might have looked like. In this case, however, the employment of such an uncompromisingly hands-off method seems a good idea, particularly considering the nature of the texts in question. Not only does it avoid difficulties resulting from the imposition of subjective editorial decisions on bibliographically difficult texts (a process from which only Dido and Aeneas, among these five works, would be likely to benefit), but, in configuring the editorial endeavor as chiefly a transcriptive act, it offers the potential for a measurable degree of consistency from edition to edition across the volume.

3. Consistency

3.1 It is surprising, then, to discover that such consistency does not in fact materialize. Several of the individual editors appear, indeed, to have set their own policies without consulting fellow contributors. The edition of King Arthur, prepared by H. Neville Davies, is a case in point. In a book purporting to focus on bibliographical issues, this chapter stands strangely apart, not only for the fact that, uniquely in the volume, it fails to inform us (p. 254) which exemplar (and therefore which variant state) of the 1691 edition is being presented, but, more significantly, on account of its extensive explanatory footnotes, many of which are superfluous or even, in a few cases (Prologue, lines 36–7; main text, lines 1131, 1386, 1619–20), simply wrong. One gets the impression that Davies failed to read Andrew Pinnock’s discussion of the publication’s bibliographical complexities (pp. 62–70) and, almost certainly, did not see the editions put together by his peers, among which only Roger Savage’s edition of The Fairy-Queen includes anything beyond a few scattered explanatory notes. Similar problems of consistency arise over the inclusion of textual material not found in the main playbooks. Julia Wood’s edition of The Indian Queen adds an appendix (p. 469) providing the text of “They Tell Us That You Mighty Powers Above,” while, in editing Dioclesian, Julia Muller neglects to transcribe either “When First I Saw the Bright Aurelia’s Eyes” or “Since from My Dear Astrea’s Sight,” even as she announces their absence in footnotes (pp. 208, 243).2 Fortunately, the two missing songs do appear in Timothy Morris’s appendix, “The Song Texts with Music by Purcell” (p. 488), but this chapter has problems of consistency all its own, providing extraordinarily detailed editorial apparatuses for Dido and Aeneas and King Arthur, somewhat more limited information for Dioclesian and The Fairy-Queen (where the printed libretto is inexplicably not taken into account), but a version of The Indian Queen that is nothing more than a transcription of the text from the printed Purcell Society score.

3.2 Of greater concern is the extent to which several of the contributors transgress Pinnock’s Malone-derived prohibition against editorial intrusion (described by Roger Savage on p. 343 as a “resolutely non-interventionist” approach). Muller’s edition of The Prophetess: or, The History of Dioclesian is a primary offender here. Altering the original text more than thirty times, Muller goes beyond simply correcting such mildly confusing misprints as “Squardrons” to “Squadrons” (line 1833) and “Vew” to “View” (line 994), moving into the realm of the genuinely intrusive: emending (to give just a few examples) “O” to “Oh” (line 390), “Sorows” to “Sorrows” (line 1510), “Pedestalls” to “Pedestals” (line 1703), “Courtisie” to “Courtesie” (line 2152), and “injoy” to “enjoy” (line 2413). Even more frustratingly, Muller maintains no obvious consistency in her interventions, emending “ne’re” to “ne’er” (line 863) but doing nothing with “’ore” (line 2614), and correcting the idiosyncratically singular verb of line 2641 (“Quiet, Content, and true Love, breeds more Stories”) while leaving the stage direction “At the same time Enters Silvanus, Bacchus,…” (line 2486) untouched. In line 795, “That are peculiar only to the Cæsar’s,” the offending apostrophe is expunged as grammatically incorrect, yet corresponding changes are not made to lines 844, 1323, 1770, and 2614, in each of which apostrophes are also inappropriately present or absent.

3.3 The failure of a coherent system to emerge in Muller’s edition perhaps underscores the wisdom behind the original hands-off approach, an approach that, unfortunately, the volume’s editors seem not to have subsequently enforced. This is particularly evident in Irena Cholij’s double edition of Dido and Aeneas and Measure for Measure. While the parameters of this text are admittedly more complex, Cholij falls prey to inexplicable inconsistencies, as when, in Dido, she prints the faulty line “Fate forbids what you Ensue” (line 145) as is, but then interposes herself just a few lines later to correct “A Hero fall, and Troy once more Empire” (line 155, changing the final word to the more familiar “Expire”). A well-executed, thorough, interventionist critical edition of the text of Dido and Aeneas is still badly needed, and Cholij might have made a compelling argument for dispensation from the volume’s prevailing rules in this case. Instead, she proceeds with a Malone-inspired typeset facsimile, nonetheless stepping in inappropriately on numerous occasions. These intrusions extend beyond the merely incidental (for example lines 94, 217, 242, 244/255, 272/274, and 308; see also Measure for Measure, line 303), and substantive (as in lines 99 and 213, and Measure for Measure, lines 303 and 1088—the latter in contrast to the untouched line 1077); in their zealous efforts at correction and regularization, they even manage to obscure important typographical evidence that should have remained unaltered for the benefit of the book’s scholarly users.3

3.4 This is not, regrettably, the only example of a chapter editor compounding an already unnecessary intervention with questionable editorial judgment. Emendations in both Measure for Measure (line 740) and Dioclesian (line χ2557) actually serve to undercut more logical readings in the original texts. In King Arthur, Davies is at pains to indicate places where the nineteenth-century Scott-Saintsbury edition offered “unwarranted” alterations; but he himself puts forward several suggestions for emendation (lines 985–8, 1291, 1382) that are of questionable validity. The fact that Davies finished his work too early to take into account the edition of King Arthur in the “California Dryden” series (volume 16, published 1996) represents a further impediment to this chapter’s usefulness.

4. Accuracy

4.1 In a volume whose chief editorial principle, known to bibliographers as “diplomatic,” encompasses not merely typographical accuracy but also the accurate reproduction of typographical anomalies in the originals, one is further surprised to find a significant number of errors: minor slips like “suppported” (Dioclesian, line 1701), “will” for “with” (The Fairy-Queen, line 810), and missing italicization (Measure for Measure, footnote to line 1914; The Fairy-Queen, line 1121; The Indian Queen, footnote to line 467); errors in the original that appear to have been silently, perhaps inadvertently, corrected (“variou[r]s” in The Fairy-Queen, line 1544); and even more obvious mistakes, such as inaccurate justification (The Fairy-Queen, lines 935–46 and 1280–4), explanatory notes placed at the bottom of the wrong page (The Fairy-Queen, lines 929 and 1527–32; The Indian Queen, Epilogue, line 12), and stray or incorrect line numbers in the apparatus (Measure for Measure, line 1969; Dioclesian, line “2496”; King Arthur, lines 897, “1523”). In both Dido and Aeneas and the song-text-only edition of King Arthur, the apparatus’s line numbers go seriously awry at the end and, in several instances (including Measure for Measure, line 720, and Dioclesian, line 588), the information supplied in the apparatus is simply faulty.

4.2 Given the scale and difficulty of the task at hand, it is entirely understandable that a few errors and irregularities should have crept into the five hundred-odd pages of transcription and even now, with the possible exception of Muller’s inconsistent emendations, the editions of Purcell’s four dramatic operas could be made serviceable with the application of a reasonable number of corrections. Sadly, the same cannot be said of Cholij’s Dido and Aeneas / Measure for Measure edition: the essential first section, in particular, so egregiously transgresses not only Andrew Pinnock’s Malone-inspired guidelines, but also other key principles of editorial consistency, accuracy of transcription, and even basic proofreading, as to render itself functionally useless. In both the original “Priest libretto” and the Measure for Measure version, the typography of Dido is notoriously flawed, while the relationship of the printed texts to the later musical scores opens up a further set of thorny problems, both textual and interpretive. Even so, in what can only be described as a disappointingly slipshod edition, Cholij makes matters worse by introducing, under the guise of an “accurate” transcription, a host of new flaws and inconsistencies. Some of these have already been noted above, but there are plenty of others—as at lines 44 (“Nereids” for “Nerieds”); 80–81 (incorrect leading between these two lines); 112 (“Dido.” for “Dido” with no accompanying textual note, a problem that recurs in lines 279, 291, 299, 305, 312, and 318); 160 (defective indentation); 191 (“Saylors, a” for “Saylors,a”); 223 (“likeness” for “likness”); 230 (“Precious” for “precious”); 255 (“.” lost in the process of emendation); and 318 (unspaced em-dash for spaced hyphen; cf. line 243, where the transcription is correct). Similar errors are to be found in Measure for Measure, lines 1088 (“next” for “Next”), 1460 (“That” for “That’s”), 1505 (“if” for “of”), 1805 (“do not” for “do”), and 1972 (“condenm’d” for “condemn’d”). In two sections of Dido and Aeneas, Cholij’s textual notes provide the wrong punctuation information for the (emended) original: the “∼.” in notes 72, 80, 81, 83, 85, 86, 87, 88, 115, 119, and 125 should all really be “∼,” —this quite apart from the fact that, as I have suggested in ref. 3 above, these emendations are unnecessary and misleading in the first place. In a text as dense and brief as Dido’s, to say nothing of its high profile in the musicological world, such mistakes are unacceptable.4

4.3 Less problematic for scholars, but equally flawed in its execution, is Timothy Morris’s appendix, “The Song Texts with Music by Purcell.” I have already noted the curious inconsistency in the sources consulted for each work and the line-numbering errors in the King Arthur apparatus, but there are numerous other problems as well. Like the other contributors, Morris finds himself uncomfortably caught between the official mandate for unadulterated transcription and the desire to annotate and emend at will. The result is a largely unusable hybrid of the two incommensurable procedures in which problems abound: substantive variants in the sources are arbitrarily noted or ignored; sung repetitions of text are printed or omitted with no apparent rationale; crucial (sung) words are missing entirely without explanation; capitalization, punctuation, and italicization appear not to have been considered in the apparatus; and even the line numbering is affected by perplexing anomalies (e.g., pp. 486–7, lines 190–1 and 209–10). One has to wonder whether Morris had a particular copy-text in mind for each section. We never learn, however, as there is no introductory essay, and no coherent editorial principles can be distilled from the capriciously assembled notes. In this 46-page chapter alone, I marked more than 130 queries, questionable editorial judgments, and outright errors, none of which needs to be belabored here.

5. Conclusion

5.1 It should be emphasized that the disappointment expressed in this review is in large measure a function of my own fervent wish for a scholarly resource that delivers precisely what the present work promises. I remember very well a convivial conversation I had with Roger Savage on the Oxford–Paddington train in late September 1993 (following a stimulating conference organized by Michael Burden) in which I first learned of plans for a volume bringing together the literary texts of Purcell’s major operatic works. Since that time, I have eagerly anticipated the opportunity to read, digest, and make good use of the result. However, after a careful and thorough perusal of the present offering I must regretfully conclude not only that it does not meet the standard necessary to become an indispensible resource for Purcell scholarship, but that it does not even adequately live up to its own self-proclaimed ideals. This is unfortunate, to say the least, since the book’s premises are essentially sound and the project as a whole has elicited not a little effort on the part of each of its contributors. Burden’s own introduction to the volume is, in fact, a worthy piece of scholarship. Andrew Pinnock’s erudite, if at times needlessly patronizing, technical survey also does a good job of explaining the minutiae of bibliographical principles and concepts, ranging from seventeenth-century printing-house practices to modern scientific methods of bibliographical description as codified by Fredson Bowers and others. Nevertheless, the volume’s numerous missteps, considered in the aggregate, must ultimately engender serious doubts about the work’s trustworthiness and, hence, its utility as a foundation for Purcell scholarship. In the final analysis, it has to be said that Henry Purcell’s Operas: The Complete Texts has the potential to do more harm than good, particularly if it is adopted unquestioningly as a definitive and reliable resource and its manifold flaws are thereby allowed to pass unscrutinized into the literature. I deeply regret not being able to offer a more positive endorsement of this book, which held out the possibility of so many good uses. At best, we might now still hope that Michael Burden will be able to convince Oxford University Press to support a revised edition, with the most problematic chapters completely redone. After all, there is still a tremendous need for such a thing, properly executed, and, in light of his many substantial contributions to British musicology, Burden might justifiably present a case that the present venture is no more than an anomalous lapse in an otherwise exemplary scholarly corpus.


* Andrew R. Walkling ( is Dean’s Assistant Professor of Early Modern Studies at the State University of New York, Binghamton. He is a historian of seventeenth-century England with an interest in the royal court, music, and theater in the reigns of Charles II and James II.

1 Editorial introduction to “Music in Purcell’s London I”, Early Music 23 (1995): 549.

2 Muller’s edition also regrettably omits the preface to Purcell’s published score for Dioclesian (1691), ghostwritten by John Dryden.

3 In particular, Cholij appears not to have recognized the generally consistent pattern employed by the typesetter with respect to the punctuation of speech prefixes. If we regard the single instance of a close-parenthesis on p. 1 of the libretto (Cholij’s line 30) as an error for a comma, probably attributable to careless type distribution, it emerges that the compositor nearly always supplies a period after an abbreviated name—

—whereas names written out in full are typically followed either by commas—

2 Women,

—or in the case of “Dido” by no punctuation at all (this latter appears once on p. 4 [Cholij’s line 112] and otherwise only on the final page, where it occurs at all 6 of Dido’s lines). Out of a total of 85 speech prefixes, the libretto presents a mere 4 exceptions to this rule: 3 periods that should be commas and only one vice versa, amounting to slightly more than 4.7% of all occurrences. Such a small number of exceptions falls easily within the generally high margin of error that distinguishes the libretto’s typography.

Elsewhere, Cholij’s textual note to line 269 is formatted in such a way as to obliterate the potentially significant typographical fact of an absent space between “ho, ho, ho,” and “next Motion” in the original print.

4 For a serviceable photographic facsimile of the unique exemplar of the libretto (albeit lacking line numbering), see the “Works II” Purcell Society score of Dido (ed. Margaret Laurie and Thurston Dart [Sevenoaks: Novello, 1979]). The “Works I” Purcell Society score of Dido (ed. William H. Cummings [London: Novello, Ewer, 1889]), prints a typeset facsimile that, while presenting a few errors of italicization and punctuation and at least one faulty transcription, is also to be preferred to Cholij’s effort.

How to cite an article in JSCM

Copyright © 2008 by the Board of Trustees of the University of Illinois. All rights reserved.
This document and all portions thereof are protected by U.S. and international copyright laws. For further information on redistributing items in JSCM, see the full Copyright Statement.