Volume 13, no. 1:

David Schulenberg*

Recent Editions and Recordings of Froberger and Other Seventeenth-Century Composers

Deutsche Orgel- und Claviermusik des 17. Jahrhunderts: Werke in Erstausgaben, vol. 2. Edited by Siegbert Rampe. Kassel: Bärenreiter, 2004. [xxii, 106 pp. ISMN M-006-52602-4. €42.95.]

Johann Jacob Froberger: Toccaten, Suiten, Lamenti: Die Handschrift SA 4450 der Sing-Akademie zu Berlin. Facsimile and modern edition. 2nd ed. Edited by Peter Wollny and the Sing-Akademie zu Berlin. Kassel: Bärenreiter, 2006. [xxv, 55 pp. ISBN 978-3-7618-1783-4. €72.]

Vingt et une suites pour le clavecin de Johann Jacob Froberger et d’autres auteurs: Dresden, Sächsische Landesbibliothek, Ms. 1-T-595 (Strasbourg, 1675). Edited by Rudolf Rasch. Stuttgart: Carus, 2000. [xxxiv, 109 pp. LCCN 2001-380684. €66.]

Johann Jacob Froberger: A Hitherto Unrecorded Autograph Manuscript. London: Sotheby’s, 2006. [16 pp.]

Johann Jacob Froberger. Froberger ou l’intranquillité. Blandine Verlet, harpsichord. Auvidis/Naïve, 2000. [Astrée naïve E 8805.]

Johann Jacob Froberger: The Strasbourg Manuscript: Fourteen Suites. Ludger Rémy, harpsichord. Classic Produktion Osnabrück, 2000. [CPO 999 750-2.]

Johann Jacob Froberger (1616–1667): The Unknown Works, vol. 1. Siegbert Rampe, harpsichord, clavichord, and organ. Musikproduction Dabringhaus und Grimm, 2003. [MDG 341 1186-2.]

1. Introduction

2. Rampe’s Deutsche Orgel- und Claviermusik des 17. Jahrhunderts

3. New Froberger Sources: Two Apographs and an Autograph

4. The Repertory of the New Sources

5. The New Sources: Titles, Programs, and Dates

6. The New Sources: Textual Criticism

7. Froberger Recordings

8. What Next for Froberger Scholarship and Performance?

References

Appendix

Tables

1. Introduction

1.1 The late Howard Schott opened his dissertation, which was essentially an edition of Froberger’s keyboard music, by admitting in effect that it would sooner or later become obsolete: “It is not yet taken for granted in the realm of music that the works of important composers should receive a constant re-editing such as is regarded as entirely normal in the literary world.”¹ For those who came of age in the 1970s and 1980s, Schott’s edition was a belated replacement for the old Austrian Denkmäler, whose texts no longer reflected current understanding of the sources or the musical style.² Since then, much more has been learned not only about the sources and the music but also about their historical context, and the entire repertory of seventeenth-century keyboard music in Europe has been undergoing a re-examination in recent scholarly work.

1.2 The pace quickened in the late 1990s with the discovery of what would turn out to be the first of two major manuscript sources containing copies of Froberger’s keyboard music. These arrived as scholars were piecing together a chronology of the composer’s music, based in part on a reconstruction of what were presumed to be at least two missing volumes of keyboard pieces.³ At the same time, important biographical information was being added, and the extant musical sources were being subjected to greater scrutiny than ever, revealing new details about copyists, dates, and lines of transmission.⁴ The year 1993 also saw the first volume of a new edition of Froberger’s keyboard works by Siegbert Rampe;⁵ issued just as Schott’s last volume appeared, this seemingly bore out Schott’s words about “re-editing.” Although the new project is not under review here as such, other publications of its editor are considered below.

1.3 Most recently, a major new autograph has surfaced. At this writing, its whereabouts are not publicly known and its contents are inaccessible. An extensive description has been published in an auction catalogue which is republished in the present issue of this Journal (see Maguire) and reviewed below on the basis of the present author’s examination of the manuscript.⁶ But although Sotheby’s generously granted ample time to view the manuscript, copying of its contents was not permitted, making it impossible to give a satisfactory discussion of the manuscript or its contents here. Clearly this manuscript ought to have a major impact on the evaluation of all other Froberger sources, including those discussed in this article. But because it may be some time before the new source is accessible to scholars (if ever), a detailed consideration of the auction catalogue is justified.

1.4 Despite these ongoing developments, performers have tended to neglect this repertory. Nor has it been the subject of much analytical or interpretive criticism, even though for more than a century musicologists have recognized Froberger’s music, or at least his suites, for its subjective expressivity. Recordings have been relatively rare, in part because Froberger and those around him were never as flashy as their north-German contemporaries, probably also because many of their most remarkable compositions, such as Froberger’s famous laments, are preserved in texts that leave much to the player’s intelligence and imagination. With the exception of Schott, editors have tended to duplicate errors and inconsistencies present in these texts, puzzling the would-be performer. No contemporary treatises explain the performance of this repertory, which tends to be heard on instruments and with conventions applied that are more appropriate to later, eighteenth-century works.

1.5 The present contribution reviews representative examples of recent research and performance. In addition to publications based on the three “new” manuscripts, it considers several recordings as well as a sample volume from what seems to be no less than a project to publish the entire corpus of seventeenth-century Germanic keyboard music. Although Froberger’s music will be the main concern, I begin with the last-named item, which serves to put Froberger’s work in a broader musical context.

2. Rampe’s Deutsche Orgel- und Claviermusik des 17. Jahrhunderts

2.1 Since the early 1990s, the editor of the present volume has been responsible for an extraordinary outpouring of books, articles, recordings, and editions of mostly Baroque music. These include “complete” editions of the keyboard music of Sweelinck, Weckmann, Georg Muffat, and Froberger, as well as a collection of “first editions” of music by more obscure composers, of which this is the second volume. In principle, this is a project that will cheer every scholar and performer of seventeenth-century European music, for the repertory has long needed to be revisited. And in many respects the series does just that, offering hefty and apparently up-to-date scholarly apparati as well as musical texts that show new wrinkles in format and notation.

2.2 Yet in other respects Rampe remains very much in the tradition of predecessors such as Guido Adler and Max Seifert. The project is organized in terms of composers, and by genre within the works of a given composer. So much is probably unavoidable in a commercially issued edition intended for “practical” as well as scholarly use. One result, for better or worse, is that Rampe can claim to present the complete works of obscure figures such as Marcus Olter (represented by a single work). But another result is that works are removed from their original contexts, and pieces preserved together in the sources are re-sorted into categories defined by the editor (or publisher). Moreover, the sheer scale of this project, apparently undertaken largely by a single scholar, raises the question of whether any one editor, no matter how brilliant, can stay abreast of the burgeoning scholarship in this area. It is troubling to find frequent citations to promised future publications by the editor, yet few references to relevant work by scholars based outside northern Europe.⁷ Under such circumstances, a reader must be vigilant that evidence is being accurately evaluated and that valid deductions about provenance, attribution, handwriting, and are not shading into what Peter Williams has called “assertorial musicology.”⁸

2.3 I fear that the latter surfaces all too frequently in what may well be Rampe’s most important project, the Froberger edition. The first two volumes of that edition have been reviewed elsewhere,⁹ and although the present publication differs in important respects, two common and somewhat contradictory features are, on the one hand, an almost alarming accumulation of information about sources, copyists, and related matters, and on the other hand a failure to evaluate or interpret the musical texts that they preserve. In the most recent volumes of the Froberger edition, this has led to a proliferation of readings, versions, and even works; variant readings from sources of sharply differing quality are presented essentially uninterpreted, alongside movements whose attribution remains in doubt despite the editor’s assertions about their authorship. For the experienced scholar-performer, this may be a bonus, but for anyone else it is a source of confusion, as the editor provides little guidance toward understanding the status of individual versions or readings.

2.4 The present volume comprises more obscure works preserved in unique sources, yet even here an assertorial style in the volume’s preface gives way to a remarkably uncritical approach to editing the musical texts. The latter are so poorly edited and, it seems, so arbitrarily selected, as to raise questions about the competence of the editor. Problems begin with the title of the volume, which is more than a little misleading. The word “Deutsche” is interpreted liberally to include “music by seventeenth-century composers working within what were then the borders of the Holy Roman Empire” (p. xiii). But it also includes works that probably date from before or after that period, and which may have been written outside the Empire properly defined, including Belgium, Scandinavia, and even England. On the other hand, the volume excludes works for the Austrian court appearing in the editor’s Organ and Keyboard Music at the Imperial Court 1500–1700.¹⁰ Composers represented—mostly by just one or two works—include Buttstett, Cornet, Erbach, Caspar Hassler, Kuhnau, Peter Philips, and Nicolaus Adam Strunck, as well as eight more obscure figures and several pieces that remain anonymous. By no means all of the works are “first editions”; in addition to several acknowledged re-editions, Rampe edits for the second time Strunck’s ricercar on the death of his mother, the Hassler work, and the canzon by Olter (which, despite the apparent tonality of its opening entry, is best described as being in C not F minor).

2.5 As useful as it may be to have such pieces transcribed into modern notation, one wonders exactly how the repertory was selected; seven other known attributions to Strunck are left to languish in old editions. Hence the present volume will serve at best as a sort of Anhang to the existing repertory of available pieces. Many of the pieces are corrupt, incompetent, or both; at least some of those assigned to better-known composers are of doubtful attribution. For instance, the pavanes attributed to Christoph Walter and Hieronymus Brehme in a manuscript now in Sweden are at best student exercises, full of parallel fifths and other solecisms. A toccata from the Turin tablatures, attributed to “Matth. Kinigl,” is an incompetent pastiche; Rampe has recognized two passages quoted from the eighth toccata in Frescobaldi’s Toccate e partite of 1615, but the piece borrows from toccatas 9 and 10 as well.¹¹ Hence twenty-two of the piece’s thirty-three measures, and the only interesting ones, are by Frescobaldi, cobbled together with a stylistically foreign introduction and a lame transitional passage. Almost as much a pastiche is an anonymous Fantasia. Auff 2. Clavier, which quotes liberally from Sweelinck’s Echo Fantasia in C.¹² It may be of some interest that such music was copied and perhaps played. But what does it tell us except that some musicians were content to play haphazard arrangements of favorite bits from early seventeenth-century classics?

2.6 In fact very little of this music will be of interest to any but the most committed specialists, and it is hard to believe that the editor even played through certain pieces in the volume, whose nonsensical chords and inept voice leading can only be the work of bad pupils or hacks. Especially incredible is the claim that Christoph Walter “stands … among the most significant German keyboard composers of the sixteenth [sic] century (p. xx).”¹³ His music rather provides evidence for something resembling a folk tradition in provincial places, where evidently the outward forms but little of the content of elite music was understood.¹⁴ That copyists reproduced such barren and pointless music as Walter’s fantasias is a reminder that scribes could be even less competent than composers, probably incapable of hearing in their heads the music they were copying. Even the ricercar by Strunck on the death of his mother, of interest for its autobiographical title, is a disappointingly square, inexpressive exercise in chromatic melodic intervals and unusual leaps, lacking the variety and ingenuity that lends interest even to Froberger’s driest contrapuntal exercises.¹⁵

2.7 In a statement of editorial policies, Rampe accuses unnamed predecessors of over-use of “conjecture” and “reconstruction” (p. xxi). But the manifestly faulty nature of many of the present sources means that any edition based on them must make corrections if it is to be more than a diplomatic facsimile. And Rampe does sporadically emend the text or insert notes in brackets. Yet, in a “Conzon” (sic) attributed to Christian Erbach—a pleasant little piece, perhaps an intabulation from the late sixteenth century—two suggested corrections are clearly wrong, and two obvious howlers are ignored.¹⁶ Within the first sixteen measures of an anonymous “Toccata 6. toni” (no. 6), three easily corrected voice-leading errors are left to stand, and one measure is left with an unexplained extra sixteenth note.¹⁷

2.8 Arguing against an attribution of the latter work to Scheidt, Rampe, following Dirksen, speaks of “severe weaknesses in compositional execution”¹⁸—but at least in these passages the problems are evidently ones of transmission. Rampe is probably right to question Scheidt’s authorship of the work. But insofar as stylistic considerations enter into the decision, these would have to be based on a text edited to eliminate such obvious mistakes.¹⁹

2.9 The contents of the volume as a whole fall into three categories:

The attributions of two chorales attributed to Buttstett, and a little praeludium (prelude and fugue) assigned to Kuhnau, are almost plausible on a stylistic basis, if one overlooks questionable details of voice leading. The Buttstett “Choral â 3” on “Vom Himmel hoch” resembles Bach’s so-called Arnstadt chorales in alternating between brief flourishes and simple settings of each phrase of the melody.

2.10 Far more problematical are the attributions to Cornet and Philips, where Rampe attempts to add to the corpus published in recent “complete” editions of these two composers. The “Fantasia 4. toni” attributed to Cornet is a competent example of mid-seventeenth-century Netherlandish organ writing, interesting for the registration markings “Cornet in x” and “x out.” But although Rampe views the variations on “Den Lustelyiken may” (no. 12) as “stylistically consistent” with Cornet’s work (p. xv), the short-winded embellishments, which sometimes leap by sevenths and other odd intervals, constitute a failure to make sense of the tune’s asymmetrical phrasing and are hard to attribute to any competent composer.

2.11 Philips is a much more important composer, arguably second only to Byrd among English composers of around 1600. The spotty transmission of his keyboard music has prevented its importance from being recognized, and one would like to find more works comparable to those preserved in the Fitzwilliam Virginal Book. Alas, the five pieces here attributed to him are all anonymous in the manuscripts. Rampe, admitting that their authorship will be settled only “by the discovery of new sources” (p. xviii), offers no good reason for attributing them to Philips.²⁰ All, including another version of “Den Lustelyiken may,” appear to be intabulations or arrangements, a genre associated with Philips because of their prominence in the section of the Fitzwilliam Virginal Book devoted to his music. But only two, a “Che fa” after Marenzio and an “Almande d’amor,” seem to be the work of even a competent musician, and neither has anything like the richness of counterpoint and variety of figuration found in pieces attributed to Philips.

2.12 Despite so many deficiencies, the volume may prove useful if only for providing a sampling of the types of music that were being copied on the peripheries of the central tradition represented by Froberger. Facsimile pages provide examples of the diverse types of notation used for this music, and although the facsimiles should have been sharper and larger, they at least prove that many of the faulty readings have been transcribed accurately. Still, one can barely make out the ornaments in the Hassler Fantasia, which are said to be “unique for a German composition of this period” (p. xv). Whether or not these signs are original parts of the text (which seems doubtful), they support Rampe’s assertion that this repertory was “generally played with a great many ornaments, most of which were not written down” (p. xvi)—an observation that will be recalled below in evaluating the Froberger recordings.

3. New Froberger Sources: Two Apographs and an Autograph

3.1 As befits their importance, the two recently discovered apograph Froberger manuscripts have been issued in sumptuous although quite different sorts of editions. As yet awaiting similar treatment is the newly uncovered autograph, which has, however, been described in a sixteen-page color brochure published as a supplement to the much larger complete catalogue for Sotheby’s November 2006 auction (see Maguire). Any one of these manuscripts would have constituted the most important Froberger discovery for a century or more; together they will greatly enrich our understanding of the composer and his works, although not without raising further questions. In addition to the previously unknown (and unsuspected) pieces in the autograph, they provide improved texts for several famous but poorly transmitted works; they also promise to shed light on the compositional and reception histories of the music. Moreover, titles and other rubrics in all three manuscripts seem to bear not only on the programmatic significance of some pieces but on Froberger’s biography. It will take time for scholars to digest the evidence presented by the three manuscripts, which need to be considered in the context of other Froberger sources.

3.2 The brochure published by Sotheby’s includes a detailed physical description and inventory of the new autograph, in connection with which the assistance of Peter Wollny is acknowledged (p. 16). Bound in covers that show the arms of Emperor Leopold I, the manuscript comprises three sections containing, respectively, six fantasies, six caprices, and five four-movement suites followed by three one-movement laments. Entirely new are the first twelve pieces as well as one suite, a “Meditation,” and a tombeau. Also new are the titles attached to some of the previously known pieces. Facsimiles in the brochure show portions of five pieces, but only the very beginning and very end of two of the unknown works. These are nevertheless sufficient to establish the autograph character of the handwriting and the closeness of the musical texts to those of the Berlin manuscript, SA (for identifications of such short titles, see the Appendix), which nevertheless differs in the absence of some accidentals, ties, and ornament signs.

3.3 One minuscule error occurs in the transcription of the title page—the one non-autograph portion of the manuscript—where the second of the dance types contained in the manuscript is actually spelled “Chigues,” not “Gigues.” Whether this could help localize the handwriting remains to be seen. In addition, although the brochure describes the manuscript as “very faintly dated in a later hand” (p. 3), it does not make clear that this phrase refers to an entry in faded pencil in the upper right of the title page. This inscription reads (probably) “Anno 1666”. The brochure’s posited date of ca.1665–1667 is probably based not on this pencil entry but on the assumption that the manuscript was written after the latest of the works in it were composed. But the only dates that can be considered reasonably well established are those of A20²¹ and the previously unknown tombeau for Duke Leopold Friederich of Württemberg-Montbéliard, who died in 1662.

3.4 It is, perhaps, an overstatement that the “discovery changes the course of Froberger studies and, by extension, the history of seventeenth-century music” (p. 4). Nor is it clear that the newly revealed fantasies and caprices are “longer, comprising more sections, than Froberger’s known earlier examples” (p. 7). The longest of these new pieces occupies ten openings, each displaying just two systems of four staves each. This is the equivalent of just five openings (ten pages) in the larger format of the Vienna manuscripts, and most of the present pieces are shorter. Possibly more distinct in style are the new Meditation and Tombeau for Sibylla and her husband; these are perhaps less restrained, more toccata-like, than similarly named pieces known previously. But clearly, evaluation of the new music must await publication.

3.5 In any case, none of the new pieces shows gross differences in style or form from those previously known. Hence, although obviously of great importance, the manuscript is unlikely to fundamentally alter our understanding of the composer or the repertory. Its apparently late date makes it clear that it is not one of the autograph collections presumed missing from Vienna (the postulated Libri 1 and 3). Yet it does appear to have been another compilation of pieces for a member of the Habsburg family—conceivably Margarita Teresa of Spain, whose marriage to Leopold by proxy in 1666 at Madrid might have been the occasion for which Froberger visited there, as documented in the title for the new Meditation. Unfortunately the brochure offers no information about provenance.

3.6 Although the greatest interest naturally attaches to the new autograph, its discovery does not diminish the importance of the two other sources. No work is common to all three manuscripts, although four suites and one lament are shared by the autograph and one of the two recently discovered copies. The remainder of this discussion therefore focuses on the editions of the two apograph manuscripts and their musical contents. The first of these two sources to resurface has become known as both the Stuttgart and the Bulyowsky (Bulgowsky) manuscript, after its presumed place of origin and its Slovak copyist, respectively. I will refer to it as “Dl,” using the siglum for the library that now houses it (D-Dl).²² Bruce Gustafson has already reviewed the present edition,²³ yet it is worth reexamining in relationship to the Berlin manuscript, which surfaced just as scholars were beginning to digest the significance of Dl. “SA,” as I shall call the Berlin source, is one of the thousands of manuscripts returned by the Ukrainian government in 2001 to the Sing-Akademie zu Berlin, a private organization whose music archive disappeared during World War II.²⁴ The archive had never been properly catalogued, and its inclusion of an important keyboard manuscript from the seventeenth century was hardly suspected. Although the Sing-Akademie has been inconsistent in the past in making its archive available to scholars and performers, it has served the musical world well in publishing the present source, which stands out among the thousands of possibilities not only for its importance but for its size and format, making it particularly suitable for publication. A catalogue of the complete collection is now in preparation.

3.7 Five smaller manuscripts in the archive, SA 4441–4445, contain additional works attributed to Froberger: a praeludium or fantasia, and four suites (Partien). Although the style is simpler and later than Froberger’s, the playful titles attached to the suites form a curious counterpart to the more serious ones in SA. The consistent format of the suites, each comprised of prelude, allemande, courante, sarabande, double, and gigue, suggests they were planned as part of a series (the key signatures are modern: F minor, E-flat major, B-flat major, and E major). Peter Wollny identifies these copies as being in the hand of the Berlin organist Johann Peter Lehmann (d. 1772), but there is no evident connection between him and a Berlin manuscript tradition involving Froberger’s contrapuntal works.²⁵

3.8 Alexander Silbiger has already described the most important features of SA and of the present edition.²⁶ Whereas Dl has been edited in a conventional scholarly format, with detailed critical apparatus, SA is given in both facsimile and transcription, but with only cursory textual commentary.²⁷ Both volumes include useful prefaces, and although some will find the layout of the transcription in the SA volume a little cramped, in principle both editions are models of their respective kinds. The monochrome photographic facsimile of SA is extremely clear, indeed easier to read and better looking than the actual manuscript, at least in the feeble December light in which I saw it in 2004. The entire editorial apparatus of the SA volume appears in both German and a competent English translation, save for a brief list of readings from the manuscript that have been altered in the transcription.

3.9 Unfortunately, neither volume is entirely satisfactory as an edition, and because they are predicated on somewhat different editorial principles, it is exceedingly difficult to compare the texts of the two manuscripts by studying the two transcriptions. A particular frustration, which Gustafson noted, is that the edition of Dl distinguishes between original and editorial ties only in its critical commentary. Indeed, the editor of Dl, Rudolf Rasch, has generously supplemented the text of the manuscript, even including an entire movement that is preserved only in a concordance (the double of the courante in Suite 23).²⁸ In a suite attributed to Poglietti, Rasch has reconstructed faulty readings “as much as possible in the style of the second half of the seventeenth century” (p. 108);²⁹ the results are plausible, but Poglietti is so little-known and his keyboard music so poorly preserved that any conclusions about his style must be considered provisional. A more literal edition of some of the Froberger pieces in Dl has already appeared in volume 3 of Rampe’s “complete” edition, where these appear as alternate versions of movements published elsewhere in the series. But, in keeping with Rampe’s general approach, the unique readings of these versions are not adequately evaluated, nor could Rampe take into consideration the more recently discovered concordances in SA.

3.10 Whatever one may think of their methodologies, at least the editions by Rasch and Rampe appear to be accurate. Unfortunately, the transcription of SA is not as reliable as one would like. The volume, for which Peter Wolny shares responsibility with the Sing-Akademie, has now appeared in a “second revised issue” that has eliminated some but by no means all of the errors found in the original issue of 2004; only one entry has been added to the all-too-brief “Critical Report” (pp. xiv–xv). Equally regrettable is the failure of the transcription to distinguish many editorially added ties, accidentals, and even notes, which appear in normal type.³⁰

3.11 To be sure, the notation of ties and accidentals is especially problematical in this repertory. Some copyists were evidently quite slovenly in this respect—but I do not think that this provides evidence for a “standard organist’s practice in which identical pitches in the same voice were sustained rather than being struck again” (as claimed on p. xix). The distinction between tied and restruck notes is a vital expressive resource on keyboard instruments—organs as well as harpsichords and clavichords. Froberger certainly understood this, even if some copyists did not. In the allemande of Suite 17, for example, it makes a difference whether or not one restrikes tenor c' on the third beat of measure 9; restriking the note makes audible the dissonance that arises as the alto moves from e' to d' (the first edition showed an editorial tie in the tenor). On the other hand, the brisé texture in measure 14 of the same movement makes it imperative to tie the note g' rather than restriking it on the third beat, where the arpeggiation of a 6/5-chord begins with bass E; yet here the edition originally refrained from suggesting a tie.³¹ Inasmuch as the corrected transcription has improved these two readings, it is surprising to find that similar mistakes remain elsewhere; for instance, one still looks for consistency in how ties are added (or not) in two parallel passages in the Tombeau for Blancrocher (mm. 15–16, 18–19).

3.12 Gustafson questioned the wisdom of making such emendations in an edition based on a single source, especially when emendations are not clearly identified as such in the printed text.³² But although both copies are relatively reliable, emendations remain necessary if either edition is to present a musically coherent text. Especially when a facsimile is also present, little purpose is served by reprinting its text verbatim in places that show a coarsening of the composer’s rhythm, erroneous ties, or wrong notes and accidentals. The copyists of both manuscripts make certain characteristic types of errors; alert editing would have eliminated these more consistently. Those in SA appear to be run-of-the-mill copying mistakes, mostly of omission (ties, accidentals, and occasional notes in inner voices). On the other hand, Bulyowsky made frequent errors of commission, to judge from numerous superfluous accidentals in Dl. Many of the latter can be interpreted as anachronisms imposed by a later copyist who was not entirely familiar with Froberger’s harmonic language.³³ Bulyowsky was a competent composer with a special interest in exotic tonalities, as attested by the presence in Dl of his suite in B-flat minor. He knew what he was writing and might well have added accidentals where he (wrongly) thought them necessary.

3.13 As a result of these editorial failings, we still lack satisfactory modern editions for some of the pieces in these two manuscripts. This applies especially to three laments for which SA provides concordances to the very faulty texts in Min. 743. Nevertheless, Wollny, who wrote the Preface for the edition of SA, deserves profound thanks for having recognized the significance of the manuscript and for providing a wealth of detailed information about the probable copyist, the provenance and contents of the manuscript, and the background to the extraordinary programmatic titles and rubrics attached to many movements.³⁴ Froberger scholars attempting to decipher the meanings of the latter will be especially grateful for the accurate transliteration and translation of the titles and for Wollny’s research into obscure subjects rarely visited by musicologists, such as the genealogies of minor imperial nobility and local Rhine river customs and geography.

3.14 On the other hand, as Silbiger has already observed, the volume provides insufficient information about the physical appearance and structure of the manuscript itself.³⁵ Yet it may be that little additional information could be extracted from the manuscript itself; there is no fly leaf or title page, and I saw no verbal or graphic indications of its origin in my own examination. Despite the large format, “superior” paper, and “lavish” leather binding (p. xvii), the latter is now badly worn, and much of the paper is in poor condition. The first leaf (pp. 1–2) is now detached, as is the next group of pages, comprising an entire fascicle (pp. 3–10), but all of the compositions present are complete. Not evident from the facsimile is the wearing of the paper, especially in the lower right corners; this plus occasional pencil corrections raises the possibility that the manuscript was used into the nineteenth century.³⁶ More recent additions are the pencil pagination and bracketed numbering of the individual pieces, as well as the blue and purple stamps added in Ukraine.³⁷ Many corrections in ink throughout the manuscript could be the work of the copyist, but others are unlikely to be his.³⁸ The first twenty pages, bearing six toccatas, are ruled in systems of 6 + 7 staves; this changes to 5 + 5 for the remaining 56 pages, containing suites and laments. But although the change coincides with the start of a third fascicle, and various forms of F-clef occur in the course of the manuscript, all appears to be in the same hand, tentatively identified by Wollny as that of the Hamburg organist Johann Kortkamp (1643–1721).³⁹ If so, this may be significant in interpreting the unique features of SA, for Kortkamp was not only a pupil of Weckmann but also a chronicler (author of the so-called “Hamburger Organistenchronik”) for whom certain verbal rubrics in the manuscript might have carried special interest as historical documentation.

3.15 Other rubrics in SA (some paralleled by entries in the new autograph) prescribe the use of discrétion in performing many of the individual movements. In the six toccatas these indications are coordinated with symbols that evidently indicate where discrétion in this sense should cease.⁴⁰ Schott argued that the term, which could refer either to slow, deliberate tempo, or to rhythmic freedom, probably signified the latter in Froberger’s music.⁴¹ In fact, the present rubrics occur so often in pieces that are also marked lentement—allemandes and laments, in addition to the passages in the toccatas—that the two meanings may have merged for the copyist of SA. Mattheson was Schott’s source for connecting the expression with rhythmic freedom; since, as Wollny notes (p. xviii), Mattheson apparently knew these pieces, if not this very manuscript, SA (or its Vorlage) might even be the source from which Mattheson derived his understanding of performance practice in the toccata, which he discusses in Der vollkommene Capellmeister.⁴² Why, however, does this manuscript bother marking both the beginnings and ends of such passages when their boundaries will be obvious to anyone familiar with the style? The implication is that SA was written by or for someone too remote from the composer, in time or place, to be completely conversant with the tradition.

3.16 Also relating to performance is the presence, as in Dl, of somewhat idiosyncratic ornament signs, which in SA appear uniquely in the gigue of Suite 9.⁴³ The main sign in question, claimed here to be “of Viennese provenance” (p. xix), must stand for an accento or port de voix, perhaps followed by a pincé. Its presence here, alongside the signs for tremblements and pincés familiar from later music, is a further indication that this music was performed with numerous ornaments, at least by the time SA was written out. The inclusion of these ornament signs only in particular movements again points to a pedagogic intention; because the signs are not found in other copies, it is unlikely that they go back to the composer.

3.17 Perhaps the most remarkable feature of the manuscript are its many new titles and programmatic rubrics, which may add significantly to the meager information available about the composer’s biography. Two opening movements (in Suites 27 and 30) are accompanied by especially lengthy “Beschreibungen,” otherwise unknown, that are transcribed and translated in the preface, with helpful commentary. That for Suite 30 confirms and elaborates upon the laconic title found in Min. 743, previously the only known source. In Suite 27, Dl identifies the allemande as “nommée Wasserfall,” and SA puts to rest speculation as to what the latter title might have meant, showing that the “Wasserfall” was not a geographic phenomenon but a fall into the water by one of Froberger’s companions while crossing the Rhine. Twenty-six “Notenfälle,” mentioned by Mattheson in a puzzling comment,⁴⁴ turn out to be not notes of a scale but as many verbal annotations added in SA at the foot of the page giving this movement.

3.18 An immediate consequence of this discovery is to invalidate any association of the title “Wasserfall” with the anonymous suite in E-flat that was attributed to Froberger as “Suite 29” in the editions of Schott and Rampe. The Allemande of the latter suite contains Frobergerian echoes, including an opening quotation from the Lamento of Suite 12.⁴⁵ But in light of the anonymous transmission among pieces by Reinken and Böhm, a north-German origin seems a better guess.⁴⁶ The designation “Suite 29” should be retired; the A-minor suite edited by Adler under that title, although including several movements elsewhere attributed to Froberger, opens with an interesting but stylistically dubious allemande and must be a later compilation.⁴⁷

4. The Repertory of the New Sources

4.1 From “Suite 29” it is logical to turn to the repertory of these manuscripts. The twelve new contrapuntal works in the new autograph must, for the time being, be disregarded, together with its new suite, meditation, and tombeau. Each of its four remaining suites and one lament recurs in either Dl or SA but not both, nor in the autographs extant in Vienna.

4.2 Dl contains 21 suites; following an arrangement found in other seventeenth-century manuscripts, 14 were copied starting at the front of the manuscript, the others from the back. The first 13 suites, containing 58 movements, are attributed to Froberger. More precisely, the first movement in each set of pieces is headed by a roman numeral from I to XIII, Froberger’s name being included in the title of the first movement (usually in abbreviated form, e.g., “XII. | Allem. de Froberg.” for the copy of Suite 15). Any doubt that these sets of pieces were meant to serve as distinct groupings of movements is dispelled by the fact that Suites 1 and 15, both in A minor, were copied consecutively as nos. XI and XII.

4.3 This section of the manuscript ends with a fourteenth suite in A major, anonymous and unnumbered. This suite is in another hand, and there is no way of knowing whether the second copyist’s failure to maintain Bulyowsky’s labeling scheme is an indication for or against Froberger’s authorship. Rasch tentatively accepted the suite as Froberger’s, and it is included on a recording of Froberger’s works from the “Strasbourg manuscript” (see below). Unlike “Suite 29,” which at least imitates Froberger’s style, this work seems remote from it.⁴⁸ But although style analysis can justifiably raise questions about attributions, it is too subjective to settle them. We simply do not know enough about the evolution of Froberger’s style to be certain that simplistic or weak movements, such as those making up the present suite, could not be early or atypical works of his. The same goes for the three unique movements that Dl includes in Suite 23; these are considered below. To be sure, were any of these pieces to turn out indeed to be by Froberger, the fact would be of primarily historical interest; adding them to the canon of his works would not affect his significance as a composer.

4.4 By contrast, the pieces in SA are all anonymous. But all are elsewhere attributed to Froberger, whose name might have been present on the missing title page. Questions have been expressed about Froberger’s authorship of Suite 27 (including the “Wasserfall” Allemande), as well as Toccatas 13 and 14,⁴⁹ but their inclusion in Dl as well as SA establishes Froberger’s authorship beyond reasonable doubt. Moreover, the six toccatas belong to the group published in 1693 by Bourgeat and also circulating in independent manuscripts in places as scattered as England, France, Sweden, and Vienna.⁵⁰

4.5 The Froberger works in the two manuscripts are listed in Table 1. Before proceeding further it will be worth considering how best to refer to these pieces. Rampe’s edition introduced “FbWV” numbers, also used in the worklist accompanying his article on the composer in the new MGG.⁵¹ Numbering systems are never perfect, but it is unwise to replace one by another unless the advantages of the new system outweigh the confusion caused by abandoning the older one. Rampe’s system is no improvement over the century-old numbering based on the sequence of pieces in Adler’s edition.⁵² Therefore I will retain Adler’s numbers for the toccatas and suites. David Starke, in his pioneering study,⁵³ found it useful to designate movements individually, and I will refer, for instance, to the Allemande of Suite 1 as A1, courantes, sarabandes, and gigues being designated accordingly.

4.6 Rampe is neverthless right to reconsider the use of the word “suite” to refer to groupings of these pieces. The word is not documented in connection with this repertory prior to Roger’s edition (10 Suittes) of 1698. But the concept is implict in the organization of Dl and SA, not to mention Libro 2 of 1649. Rampe’s use of the term “partita” is no less arbitrary than that of “suite,” but as Froberger already used the Italian term to mean “movement” (in the general title of the 1649 autograph), it is illogical to apply it to a sequence of French dances.⁵⁴

4.7 But did Froberger compose dance movements for inclusion in specific suites? If instead he wrote them individually, did he later collect all of them into sets, or did some remain orphans, free to wander from suite to suite at the whim of copyists?⁵⁵ Musical parallelisms sometimes imply the grouping of movements, and in a few suites the courante is derived entirely by a process of variation from the allemande (as in Suite 1). More frequently one finds extended parallelisms that fall short of complete movements (as in Suites 3–5), or briefer parallelisms involving memorable turns of phrase or harmonic progressions (as in Suite 27). No parallelism can irrefutably establish that two movements were composed to form part of a group—anybody could have composed a variation at a later date—but at least it provides a musical reason for a grouping, which otherwise is based simply on a common tonality. With Froberger it is mainly the gigues that wander, in some cases in order to “complete” early suites that evidently lacked such movements.⁵⁶ Froberger may at first have regarded gigues as less essential to a suite than other movements, and perhaps he sometimes supplied them improvisationally.⁵⁷ Some gigues, moreover, exist in triple- as well as duple-time versions. But it remains unclear whether he was responsible for moving certain gigues from one suite to another, or for the triple-time versions G7, G11, G13, and G15 (see Tables 2a, b, c).⁵⁸

4.8 One way in which movements wandered is suggested by a remark of Constantijn Huygens, who sent the lutenist St. Luc “a jig by the late, great Froberger, which I have transcribed for lute. There you will find some excellent passages and a marvelous ending. I know nothing of his by heart other than this, but I have taken pains over this piece and made it my own study, playing it for no one other than myself, as it is not a taste for every palate.”⁵⁹ Huygens seems to imply that he has learned the piece in his own arrangement, perhaps writing it down only in order to pass it on to a friend. Although this may have been an exceptional case, it illustrates the freedom with which Froberger’s music could be treated, as well as a lack of consideration for the possibility that the Gigue was a fixed movement in a larger group of pieces.

4.9 The issue of wandering gigues is separate from that of their position within the suite. That Froberger did consider the suite as an integrated whole is suggested by a comment in Weckmann’s copy of Suite 20 (in the Hintze manuscript), indicating that Froberger by this time was putting the gigue in second place. But the comment in Weckmann’s manuscript did not necessarily apply to earlier suites, even though the new autograph does give all the gigues in second place (including that of Suite 20).⁶⁰ Some gigues can stand effectively in either position, but others seem better left in second place, as when a triple-time gigue concludes with a passage in duple time, the latter serving as a return to the style of the preceding allemande; if so, then Roger’s re-ordering of Suite 10 (vis-à-vis SA and the autograph) was a mistake. Suite 18 has a relatively short, slight gigue, which comes last in Roger’s print (and in the D-minor version found in two English manuscripts); the suite produces a more profound, elegiac effect when it closes with its relatively powerful sarabande. The gigue of Suite 20 is entirely in duple time, but near the end it returns to the style of the allemande, in a passage that SA marks “NB avec discretion,” effectively pairing the two movements.

4.10 Although they occasionally disagree on where to put the gigue, Dl and SA both preserve suites intact, that is with all the movements that are elsewhere documented as concordances. The only exceptions involve several “wandering” gigues in Dl, and this may be a reflection of that source’s relatively early origin, or rather its preserving what appear to be early works or early versions. Hence both manuscripts, if not particularly close to the composer, are at least products of a tradition that understood Froberger’s concept of the suite, at a time when the grouping of pieces into integral sets must have seemed unusual and was far from being universally respected by copyists. SA includes the six suites also found in the 1656 autograph (Suites 7–12); of these Dl has only Suite 11, which it gives in a distinct version.⁶¹ On the other hand, Dl contains seven of the eight suites in Roger’s 1698 print (Suites 13–20). SA also contains seven suites that appeared in that edition, but two of these were already present in the 1656 autograph.

4.11 But despite these “grouping” concordances between the “new” manuscripts and previously known sources, the ordering of suites in each case is different. Rasch commented (p. 24) on the apparently systematic choice of keys for the first six suites in Dl (c-d-e-F-g-a). But the copyist Bulyowski had a special interest in tonal relations, as witnessed by his own suite in B-flat minor and later publications, and the ordering of the first six suites could have been his own idea. SA lacks this hexachordal arrangement of tonalities, and it opens with six toccatas (as do two of the Vienna manuscripts)—but it cannot be a coincidence that the toccatas follow the same sequence of keys as the first six suites (d-G-F-e-a-g). Does this imply pairing of toccata and suite in performance? No such relationship occurs consistently in the autographs.⁶² In the new autograph, the first two groups of pieces share four tonalities, three of these occupying the same position in each set (1 = a, 2 = e, 6 = F). The fourth common tonality, described in the Maguire inventory as “B-flat major,” does not occur in any previously known Froberger works. The first suite is again in A minor, but thereafter there is no evident organizing principle other than non-repetition of a given tonality within a section of the manuscript.

4.12 Nevertheless, SA shares with the new autograph a focus on suites whose first movements bear titles; these are followed by laments. In SA, it is apparent from the programmatic titles of their opening movements that the first half-dozen suites form a group; all of these movements are described as “faite” (made) for some reason or at some place, and five are to be played with discrétion. They are separated from the next five suites by the single-movement Tombeau for Blancrocher.

4.13 The final section of SA contains all six suites from the 1656 autograph, in a seemingly random order interrupted by Suite 20 and the lament for Ferdinand III. Whether or not we include the Tombeau for Blancrocher in this group, this final portion of the manuscript bears a double association with the imperial family and with death, as signified by the presence of what we might call the “coronation” Suite 11, followed immediately by the “lamentation” Suite 12. If the aging Froberger had been contemplating the meaning of fame, fortune, and fatality, his thoughts might have found expression in the present selection of pieces. The same holds even more clearly for the new autograph, which closes with two tombeaux that are closely preceded by two meditations: contemplations of future deaths that lead to laments for actual ones.

4.14 Inasmuch as six of the last seven suites in SA belong to the autograph Libro 4, it is tempting to suppose that the manuscript’s first six suites are those of the missing Libro 3. This hypothesis would not be ruled out by the putative dating of Suites 13 and 27 to the early 1650s (see below). On the other hand, one would then need to explain the dedication of Suite 18 to the “Duchess of Wirtemberg” well before the time when Froberger is known to have worked for the Dowager Duchess Sibylla of Württemberg-Montbéliard.⁶³ The choice and ordering of suites in SA might just as possibly have been a copyist’s, reflecting local conditions or even a particular interest specifically in programmatic pieces. Certainly musicians of the next generation or two were already aware of this special aspect of Froberger’s music, to judge from references by Kuhnau and Mattheson.⁶⁴

4.15 In the absence of clear patterns common to other major Froberger sources in the grouping and ordering of suites, it is impossible to identify the present pieces as belonging to one of the autograph volumes whose loss has been hypothesized by Froberger scholars since the nineteenth century. But perhaps it has been a mistake to assume that the missing autographs had the same structure as Libri 2 and 4; after all, the unnumbered autograph of 1658 contains only contrapuntal pieces, and the new one lacks toccatas. Nor can it be taken for granted that the groupings and orderings of pieces in the imperial manuscripts had any sort of permanent or definitive status for the composer. Froberger’s teacher Frescobaldi had issued a series of publications containing somewhat similar types of pieces, but the contents of some of these books were substantially revised upon republication.⁶⁵ The present manuscripts tend to confirm what scholars had already suspected from previously known sources: that it was also Froberger’s practice to return to finished works, revising them on occasion. Hence it would not be surprising if he also revised the grouping or ordering of sets of pieces. But in any case, we do not know that the missing libri contained toccatas and suites at all, or whether manuscripts prepared for other patrons—such as the one reportedly presented to the elector of Saxony⁶⁶—contained the same repertory as those presented to the emperor, organized by the same principles. Mattheson was evidently familiar with a lost manuscript that was similar in organization. But the selection of pieces in any such manuscript could have partially overlapped that of the imperial autographs; the six toccatas in SA could be an authentic set prepared by Froberger for some purpose, even though two of the pieces also occur in the imperial Libro 2.

5. The New Sources: Titles, Programs, and Dates

5.1 It is reassuring that, while providing new and extended titles for many pieces, none of the new manuscripts directly contradicts any previously known programmatic rubrics for Froberger’s music. On the other hand, these and other sources give distinct versions for many titles, and two different pieces are described as having been “made in honor of” Duchess Sibylla (A18 in Dl and the new autograph, A17 in SA).⁶⁷ A third, the “Afligée” (first movement of the suite in F major in the new autograph), was “made at Montbéliard for” her (emphasis added). All three pieces are allemandes, and although the last cannot yet be evaluated, the relatively heterogeneous nature of the figuration in A17 and A18 might, together with the fairly extended dimensions and sophisticated style of the following movements in both suites, be taken as characteristic of Froberger’s later works. The Lamentation for Ferdinand III from Suite 12 is an allemande of the same type, presumably written shortly after his death in 1654. Even more “heterogeneous” is the Allemande of Suite 20 ( for which all three manuscript sources give the title “Meditation faite sur ma mort future”; SA adds that it was written at Paris in 1660). An initial impression of the new Meditation for Sibylla and the Tombeau for her husband is that both are also “heterogeneous” allemandes, containing agitated, even virtuosic figuration sometimes more characteristic of the free sections of a toccata than of an allemande. But this is a highly subjective way of characterizing style, and even with the new autograph we have precious few other chronological landmarks tying individual pieces to specific dates.

5.2 Rubrics in both Dl and SA associate Suite 13 with the period around 1652, when Froberger is known to have been in Paris. According to SA, the allemande was written for the Marquis de Termes (“faite pour remercier Monsieur le Marquis de Termes des faveurs et bien faits de luy recežs â Paris”), evidently the same to whom the dying Blancrocher commended his children.⁶⁸ César August de Pardaillan de Gondrin, marquis de Termes, was first gentleman of the chamber of Monsieur (Gaston, duc d’Orléans), a devious character if one can believe the account in Tallemant des Réaux’s Historiettes.⁶⁹ This Allemande is relatively restrained in style, retaining the imitative element present in older ensemble dances and evident in examples from Froberger’s Libro 2 (e.g., A3). The pieces in the latter book, dated 1649, might have been composed somewhat earlier, but in any case the use of variation technique and the absence of gigues in all but one of those suites do look like “early” traits. If so, then it is significant that Suite 13 contains a relatively lengthy imitative gigue, which according to both Dl and SA was called “La rusée mazarinique.” The title seems to refer to Cardinal Mazarin’s surreptitious return to France on Christmas Eve 1651, at the end of what historians term his first exile;⁷⁰ morever, both manuscripts have a rubric dictating that the cadenza-like passage at measure 18 be played “lentement et avec discretion comme le retour de Mr. le Cardinal Mazarin à Paris.”⁷¹ It is odd that SA attaches title and rubric to what is probably the original, duple-time, version of the gigue, whereas Dl includes both only in its second copy of the same movement, in triple time.⁷² But in any case we have here another indication that this type of gigue may indeed be a development of the period after 1649, as suggested above.

5.3 Other titles in SA flesh out the association of Suite 11 with the imperial family, which was previously evident only in the iconographic vignettes accompanying the copy in the 1656 autograph, where it precedes the lament for Ferdinand IV. The title in SA associates A11 with the election of Ferdinand IV as King of the Romans, that is, as emperor-elect to succeed his father Ferdinand III (which was not to be); this title would accord with the eagle that heads the copy in Libro 4. But SA describes the Courante of the same suite as written for the birthday of the “jeune princesse imperiale,”⁷³ and the Sarabande for the coronation of Eleonor Gonzaga as empress; do these titles actually accord with the sword and scepter that precede the respective movements in the autograph? The latter, unlike SA, places the Gigue second and heads it with an orb; SA has no title for this movement. These discrepancies suggest some imprecision, if not an evolution, in exactly what these pieces represented. Especially as Dl seems to give a distinct, earlier version of the Allemande, caution is appropriate before assuming that the dates of the events named in SA coincide with dates of composition for the movements of Suite 11, even if it is true that Froberger was in Regensburg for the coronation in 1653.⁷⁴

5.4 Hence I am not entirely convinced by Wollny’s precise fixing of the “Wasserfall” incident to Midsummer’s Eve (June 24) 1654, although his general argument is certainly sound, superseding earlier speculation.⁷⁵ Needless to say, one can be even less certain of the date of A27, the piece that represents the event musically. Certainly, however, the discovery of the annotated copy in SA, with its twenty-six “Notenfälle,” confirmed the earlier guess of Rasch and Dirksen that this was the piece described by Mattheson and alluded to obliquely by Kuhnau in 1700.⁷⁶ Included in SA alongside other programmatic allemandes, A27 would appear to be another product of Froberger’s journey through France, England, and the Netherlands, during the period 1650–3. Perhaps Froberger even counted his eventful ferry crossing as one of the three “blinderungen” of which he complained to Kircher, although the latter expression would apply literally only to the robberies lamented in two other pieces also found in SA: the “Lamentation ce que j’ai été volé” (A14 in G minor) and the “Plainte” of Suite 30 (in A minor). If the remaining suites from this part of SA also derive from those travels, then the latter would have seen some mountain climbing and a visit to Stuttgart as well. But this may be to hang too much speculation on a few intentionally tantalizing titles.⁷⁷

5.5 Wollny suggests, on the basis of his dating of Suite 27, that a shift of the gigue from last to second place in a suite occurred “in the summer of 1654” (p. xxii). But although the Gigue of this suite still comes last in SA, the copy of the same suite in Dl has the Gigue second. The same discrepancy arises in Suites 11, 16, and 17; this is particularly puzzling since, as noted below, Dl generally seems to give later versions. Wollny admits that his argument would hold only if Froberger did not use both orderings simultaneously; it would also require that Froberger composed these suites as ordered sequences. In the case of Suite 27, there is good reason for thinking all four movements were composed at the same time, yet I wonder whether it could be as late as 1654. All four movements are relatively restrained in style, perhaps less imaginative than the others of SA. There is one lovely moment, when the treble suddenly leaps up to g'' from the middle register, but this occurs in both the Allemande (m. 10) and the Sarabande (mm. 12–13), the type of parallelism associated with the suites of Libro 2.⁷⁸ The Gigue, although not particularly short, is an unintensive, homophonic piece in triple time, although this could also be said of G30 in the suite that follows in SA.

5.6 Of course, one cannot assume a linear development in the style of any composer. Suites 11 and 27 might be relatively plain works, or simply weak ones, that nevertheless postdate more complex or imaginative ones. It would not be surprising if a composer worn out by long, difficult years of travel were to relax in a few pieces into a relatively undemanding manner. But although there now seems little question as to the authorship of Suite 27, it remains unclear exactly what to make of the titles attached to this and other pieces. Even something seemingly as precise as SA’s dating of the Meditation (A20)—Paris, May 1, 1660—cannot be accepted uncritically.⁷⁹ Would this have been the date when Froberger first wrote it down, the date when he prepared a fair copy, or simply a date that he attached for some private reason to a piece already composed? Even if the composer originally attached titles and dates to pieces in a straightforward fashion, there remains the possibility of imprecision in transmission, especially if the matter recorded in titles and other annotations was initially preserved by oral tradition. The latter possibility is raised by the fact that Dl, which in general appears to give early readings, lacks detailed rubrics, whereas SA is rich in them. If supposedly autobiographical pieces like the “Plainte” and the “Wasserfall” received titles after the fact, not to mention detailed programmatic analyses, then the precise datings that have been suggested for some pieces would become even less plausible.⁸⁰

6. The New Sources: Textual Criticism

6.1 However the titles may ultimately be interpreted, it is already clear that the three manuscripts offer substantially improved texts for many works. More importantly, Dl and SA offer fresh perspectives on the origins and early dissemination of Froberger’s toccatas and suites. Although the details of textual criticism in these pieces are tedious, only through careful consideration of each work can a picture begin to emerge of when and in what form Froberger first composed these pieces and how they originally circulated. The issue is of special interest as these may be among the earliest keyboard suites composed as such.

6.2 The text in Dl for Suite 19, previously known only from the Amsterdam prints, is a clear improvement, as are Dl’s readings for Suites 15 and 18, which represent distinct versions or states for these works. It is therefore of great interest that these are the three suites which Dl shares with the new autograph.⁸¹ It seems unlikely to be coincidental that Dl designates Suites 15 and 19 as “ex autographo,” and that Suite 18 has similar titles in both sources. Van Asperen (paragraphs 5.3–7) reports that the two sources are close in their actual readings but doubts that Dl was copied directly from an autograph—and especially not the recently discovered one, in view of a number of apparently later readings preserved in Dl.

6.3 Similar questions arise in connection with the two works common to SA and the new autograph. Although the first movement of Suite 20 has been known for some time in the reliable copy by Weckmann (in the Hintze manuscript), the Amsterdam prints were, again, the only previously known integral sources for the suite. In the opening movement, the “Meditation sur ma mort future,” the new autograph shows two mordents in measure 13 that are lacking in both Weckmann’s copy and SA (as is evident from the facsimile in the brochure, p. 8). SA’s omission of a number of ties is clearly an idiosyncrasy of its copyist, yet the unique inclusion of the words “â Paris 1 Maÿ Anno 1660” must be significant. Although the three sources are close, their divergences indicate that they descend independently from at least one other lost autograph.

6.4 A comparable situation exists in the case of the lament in F minor for Ferdinand III. SA already furnished a coherent text, replacing that of Min. 743, which in a few places is almost incomprehensible; now the autograph promises to provide a few improved readings, albeit under a different title (“Tombeau” rather than “Lamentation”). It had already been evident to most editors that the omission of one accidental in both SA and Min. 743 at the beginning of the lament was an oversight; Rampe, nevertheless, lists the work as being in F major in his work-list in MGG (as the Maguire brochure obliquely notes, p. 7). Whether the autograph will supply more substantial improvements or corrections to the text remains to be seen.

6.5 Perhaps an even more significant question, however, concerns the origin of the faulty texts in the Amsterdam prints, Min. 743, and other sources. The question is especially urgent in the case of pieces for which we still lack autographs. It would be a mistake to underestimate the capabilities of incompetent scribes to corrupt a good text. Yet there is reason to suspect that the markedly inferior texts in some sources derive ultimately from autographs other than those used for SA and Dl. In particular, SA gives not only a better text but a distinct version in the Tombeau for Blancrocher. The implication is that some copyists had access to autograph drafts, not fair copies; the former would have been harder to read, apart from giving early versions.

6.6 Whereas Dl may descend from fair copies in the case of Suites 15, 18, and 19, this may not have been true for other works. Dl and SA together improve upon previously known sources for Suites 16 and Suite 27 (including the “Wasserfall” Allemande). But here and elsewhere the musical texts of Dl and SA represent quite different traditions. On the whole, as Wollny notes (p. xix), SA seems to give later readings, close to the extant autographs and to the early printed editions (where concordances exist). Dl, on the other hand, sometimes transmits versions that seem to predate the extant autographs. These circumstances might reflect the origin of SA as a first- or second-generation copy from revised autograph scores, that of Dl as a less direct descendant of individual autograph drafts.⁸² Indeed, the filiational status of both sources grows somewhat murkier as one examines more closely the texts of individual movements.

6.7 That some sources preserve versions of pieces either earlier or later than those in the Vienna autographs is not news; indeed, Rampe’s edition contains numerous alternate versions. Yet many of what have been presented as early or revised versions look more like faulty or arbitrarily altered texts.⁸³ This is as true for Dl as other manuscripts, such as the Grimm tablature. But because Dl also contains apparently reliable texts, close to those in other sources (including SA and the autographs), and because its score format and provenance appear to place it relatively close to the composer, its alternate readings are worthy of careful consideration.

6.8 The readings in question range from details (such as alternate accidentals) to the halving of note values in some triple-meter pieces and the substitution or addition of entire movements. In a few cases one can confidently speak of alternate versions, in the sense of a movement that has been thoroughly reworked, whether by the composer or a later musician; such is the case with sarabandes and gigues in altered meter or note values (for the latter, see Table 3). More often, however, one gets the impression of small changes in rhythm, voice leading, or melody, possibly made at various times. Such alterations, which are ubiquitous in seventeenth-century keyboard music generally, are better described as “refinements” than “revisions.”

6.9 Not unexpectedly, with Froberger the greatest number of refinements occur in the allemandes and laments, which are the longest and most complex of the suite movements.⁸⁴ In addition to melodic embellishment, these include alterations of rhythm, typically from “straight” to dotted or Lombardic, and the clarification or supplementation of imprecise notation, usually to convey details of brisé style (the breaking of chords, notated as a pseudo-contrapuntal texture). Many of these changes resemble the sorts of alterations that a practiced player presumably made improvisatorily.⁸⁵ The existence of such refinements suggests that Froberger would return to completed scores to bring their notation into closer alignment with actual performing practice, especially in movements or passages played à discrétion. Yet Froberger reportedly withheld some of his music from dissemination precisely because no one could play them without having heard the composer himself execute them. If so, then it is possible that many refinements were set in notation not by Froberger but by others seeking to preserve his manner of performance.⁸⁶

6.10 Yet even if the aging Froberger did despair of accurately notating his works, this might have been only because he had found that, despite his best efforts, his music continued to receive insensitive or overly literal performances. As he contemplated his mortality—a situation vividly represented by the Meditation (A20)—he might have doubted the possibility of passing his musical legacy on to anyone; yet he did not necessarily hold the same view throughout his life. The detailed indications in SA for the use of “discretion,” including symbols marking the points where it ceases to be relevant, point to a pedagogical tradition stemming either directly from Froberger or from pupils and acquaintances such as Weckmann or an unidentified German musician mentioned by Huygens.⁸⁷ As the latter was reportedly working in Denmark by 1668, we should be wary of too assiduously speculating about transmission through known musicians like Weckmann; those copyists and composers whose names we know probably represent only the tip of an iceberg.

6.11 Tangible evidence for this tradition includes the numerous German tablature sources, whose relationships and provenance have yet to be thoroughly worked out. They include, for example, a tablature copy of Toccata 14 dated as early as 1653, which would fit nicely with the postulated chronology for the first six suites in SA, if it derives from an autograph Vorlage.⁸⁸ Variants in this copy, although not pointing to a distinct version or even to refinements, are nevertheless sufficiently numerous to suggest that already by this date Froberger’s texts had gone through several generations of copies, perhaps including transcription into tablature. Whether Froberger himself used tablature must remain open; variants characteristic of tablature copies, such as octave displacement of individual pitches, occur in surviving tablature copies as well as in scores such as Dl. But the relatively poor quality of the tablature copies implies that, if they are not transcriptions, then like some of the copies in Dl they derive from autograph drafts in which corrections or refinements sometimes made details hard to read, resulting in clusters of slightly different readings at certain points. Drafts of this type might have been typical of Froberger’s autographs, the fair copies now in Vienna and the lost manuscript reportedly presented to the elector of Saxony having represented exceptions.

6.12 Especially if Froberger had grown discouraged in his later years, we can imagine that he would have had little incentive to put his old drafts into better order. Throughout his adult life he must have possessed a substantial inventory of compositions, only some of which would have been completely and precisely notated. Some, perhaps most, might have remained in his head for long periods before being written down, and even then not necessarily in complete or stable texts. Selecting and copying pieces in clear, revised texts would have entailed significant effort and some expense (for ink and paper), and might have occurred only for specific reasons, such as a response to a commission or for presentation to a prospective patron. The unusually neat, systematic appearance and organization of the autographs, so distinct even from other imperial presentation manuscripts such as Poglietti’s, were not necessarily typical of Froberger’s own materials. Like Bach and other composers whose practices are better documented, he might well have entered some revisions unsystematically into different manuscripts of the same movement, at various times and without regard for whether he had done so in another copy of the same work.

6.13 Wollny may be right in his view that the six toccatas in SA, like Bourgeat’s printed editions of the same works, derive from “revised autograph master copies.”⁸⁹ But in the two toccatas also found in the 1649 autograph, later sources add only a few common refinements.⁹⁰ These occur chiefly in the free sections of the toccatas, where any competent player might have embellished Froberger’s original—conceivably on the basis of having heard Froberger perform the music with unnotated alterations. Hence the origin of the distinctive readings in these sources must remain uncertain, although they certainly reflect performance practices of the late seventeenth century.

6.14 In the six suites of the 1656 autograph, also present in SA, I am even less certain that SA “definitely contains the original versions” (“ursprüngliche Fassungen,” Wollny, p. xix), or even what that latter expression might mean in this context. It is true that SA is notated somewhat less explicitly than the autograph, lacking a number of necessary ties, dots of prolongation, and even some notes for the inner voices. But SA also shows refinements in some movements.⁹¹ In Suites 8 and 10, which also appear in the Amsterdam prints, SA is generally closer to the latter, although the prints give the gigues in the final position. Suite 11, however, must have had a distinct compositional history; not only does SA give a less refined version than the autograph,⁹² but this is the one suite from Libro 4 also to appear in Dl, in a version that seems even earlier.⁹³ That Dl might indeed transmit an early version of a work from the 1656 autograph is confirmed by its relatively unrefined copy of G7 (attached to Suite 23). These discrepancies are particularly troubling in Suite 11, whose unique titles in SA would connect it with events in the life of the imperial family. Corresponding indications are entirely lacking in Dl, raising the question of when the titles came to be attached to these pieces.

6.15 The situation is even less clear for the remaining suites. The uneven quality of the texts published in the Amsterdam prints was a sign that these descended from sources of various types; Dl reinforces the impression that Froberger’s music circulated in texts of greatly varying clarity, authority, and origin. For four suites (14, 17, 19, and 20), Dl and SA give texts relatively close to the early editions, which here are relatively good; on the other hand, Dl’s readings are more refined for Suites 15 and 16.⁹⁴ For Suite 18, where the prints give a poor text, Dl’s text is more accurate and in at least a few places more refined, especially so if one considers the halving of the note values in the Sarabande to be a subsequent refinement. Yet the print has ornament signs and a rare “piano” marking not in the manuscript.⁹⁵ On the other hand, the printed text for Suite 13 is relatively good, and SA is close to it.⁹⁶ Here Dl gives a distinct version, whether earlier or later is hard to say.

6.16 The problem of Suite 13 is especially maddening because it is one of Froberger’s greatest suites (recognized as such by being placed first in the Amsterdam prints), and, as noted earlier, SA and Dl provide particularly interesting rubrics for it. Dl gives some passages of Suite 13 in more elaborate form, others simpler (at least rhythmically), and still others with alternative voice leading.⁹⁷ The Sarabande, in particular, shows what look like refinements: its note values are half those of SA and other sources, the initial chord is broken as a written-out downward arpeggio, and there is a startling but effective chromatic harmony in measure 17. Yet this version of the Sarabande lacks the written-out petite reprise found in SA and the prints, and it avoids the low note AA. Froberger used 3/4 notation only once in the autograph suites, and the one-flat signature throughout the suite in Dl is certainly an anachronism, leading to at least one unlikely melodic augmented second (Allemande, mm. 4–5). All of this must cast some doubt over Froberger’s responsibility for this version of the suite, including its most remarkable feature, an alternate, triple-time version of the Gigue that immediately follows the more familiar version in common time.

6.17 Dl and SA preserve four more suites that are found in neither the autographs or the early prints. Only one of these, Suite 27, is in both Dl and SA, where their differences conform to the pattern of Suite 13: Dl gives faulty accidentals but also alternate versions of some passages.⁹⁸ For Suites 23 and 28, Dl likewise gives texts that on the whole seem more correct than those previously available, but which still contain outright errors alongside musically distinctive readings. On the other hand, for Suite 30, as well as the two laments, SA’s text is close to but more accurate than that of the only previously known source, Min. 743;⁹⁹ it is striking that these two late manuscripts preserve so many programmatic rubrics.

6.18 Suites 23 and 28 resemble a number of the other “uncollected” suites in containing doubles as well as detached or “wandering” movements (see Table 2c and Table 4). Whether Froberger was responsible for composing all of these movements, or for placing them in suites, has been a recurring question; Dl adds to the repertory of such movements. Prior to the discovery of Dl, Suite 23 in E minor already had the greatest number of sources; for this work Dl gives a significantly different text, concluding with an early version of G7, for which the problematical Grimm tablature substitutes the triple-time version G23. Hence for this suite Dl relates to Grimm as it does to the Amsterdam editions in the case of Suite 15. Both works might have been early suites originally lacking gigues, which were later added either at the discretion of copyists or according to some verbal directive of the composer’s (such as occurs in the Hintze copy of Suite 20).

6.19 Suite 28 in A minor is a more obscure work, and Dl is the first source to give it in what would now be termed “complete” form, including a previously unknown Courante and new doubles for the latter and the Allemande. Dl also attaches the “wandering” G30 at the end; the only other source with a gigue, Bauyn, has a different one, after the Allemande.¹⁰⁰ Are all of these movements by Froberger? The Courante is derived from the Allemande, but although in both movements the first half contains seven measures, the second half of the Allemande is eight measures long, whereas that of the Courante contains only six. The shortening of the second half is achieved by omitting the equivalent of one measure (mm. 8b–9a) and compressing three later measures into two (measures 12–14 of the Allemande correspond with measures 11–12 of the Courante).

6.20 Froberger did comparable things in the suites of the 1649 autograph, where the courantes, although derived in whole or in part from the allemandes, are never direct, measure-for-measure variations of the latter; they include a movement with a “shortened” second half (C2). But the present Courante seems to lose its way in measure 11, where a phrase ends prematurely, and the penultimate measure of each half is awkward and unimaginative.¹⁰¹ Seventeenth-century courantes sometimes contain odd, seemingly inconsequential phrasing, and it would be wrong to rule out Froberger’s authorship of this movement because its style strikes a modern listener as inept. The Courante includes an instance of the repeated-note upbeat (to m. 6) that also occurs in A1 (mm. 5, 9) and in Dl’s version of C11; perhaps this was a genuinely Frobergian mannerism that the composer later abandoned as ungraceful. A dotted version of this repeated-note upbeat occurs in the new “Afligée” and Tombeau (nos. 25 and 35 in X), both presumably late works. Still, if C28 is indeed Froberger’s, it must be an early effort that he later abandoned, and the source that the Bauyn coypist used might have omitted it with the composer’s blessing.

6.21 Similar things have been said about the doubles of this and other suites, but Silbiger argues against dismissing them too quickly, for the doubles, where attached to a given movement, are always the same ones, even in sources as far removed as a French score (Bauyn) and a German tablature (Grimm).¹⁰² Still, the only assuredly authentic variation movements by Froberger are the partite on the Mayerin in Libro 2 (that is, Suite 6), one of which is a courante with a double. The latter somewhat resembles the early lute doubles in the restrained character of the variation; the other partite are reminscent of the various types of keyboard cantus firmus setting composed in the earlier seventeenth century. Where did this style of double come from? Few if any French examples can be dated to the first half of the seventeenth century, when they might have provided models for Froberger.¹⁰³ Were Froberger and other keyboard players already playing such variations in the 1630s or early 1640s? Frescobaldi, Bull, Scheidt, and other composers furnished antecedents, but their variation pieces are somewhat different in character from the doubles in the Froberger sources. The latter more or less resemble the partite on the Mayerin, yet sometimes, as in the new doubles for Suite 28 offered by Dl, the apparent desire to maintain a constant flow of small note values leads to somewhat vapid strumming.¹⁰⁴

6.22 In this light it is striking that the A-major suite preserved anonymously in Dl and in the Stoos manuscript also has doubles for all but its Gigue.¹⁰⁵ Although the Courante is not derived from the Allemande, the second halves of the two movements are closely parallel harmonically, making a striking turn toward C-sharp minor. A similar relationship holds for the Allemande and Gigue of Suite 18, for which Froberger’s authorship is not seriously open to question. Yet the Gigue of the A-major suite consists entirely of broken-chord figuration unknown in Froberger’s attributed works. If the suite really is an early work of his, one would not expect it to include a gigue, and at least the latter movement is probably by a substantially later composer. Yet the first three dance movements also contain more sequences than one would expect in a mature work by Froberger; this is especially true of the Sarabande. Many of Froberger’s sarabandes, like this one, consist of a song-like sixteen-bar double period—two symmetrical halves of eight measures each.¹⁰⁶ But it is a type that remained popular in Germany, and it is suspicious that the arpeggiations in its double are as schematic as those in the Gigue.

6.23 Suspicions of a later hand must also attach to Dl’s triple-time version of G13. As with other alternate triple-time gigues, its concordances give it as a “wandering” or a detached movement.¹⁰⁷ Rasch takes it to be the original version (p. xxvii), but one reason adduced for this is no longer valid, as its “special programmatic title” is attached in SA to the version in duple meter.¹⁰⁸ There are musical reasons as well for doubting the triple-time version is earlier, for the version notated in duple meter already alternates between duplet and triplet divisions of the beat. The triple-time version is metrically homogeneous, entirely lacking the rhythmic subtlety of the duple version. Yet the triple-time version gives the first three entries of the subject in different rhythmic forms; only with the third (soprano) entry does it settle into the form of the subject that will be repeated for the remaining three entries. No other gigue attributed to Froberger shows such instability in its opening subject, which therefore seems unlikely to derive from him.¹⁰⁹ The subject is an embellished version of that of Ennemond Gautier’s “La Poste,” a piece transmitted as both an allemande and a duple-time gigue—not that this has any obvious bearing on the rhythmic interpretation of either piece.

6.24 Rampe has accepted not only the anonymous movements and unica from Dl but many other previously rejected pieces from German sources such as Grimm. He offers a list of “criteria of authorship,” but these consist almost entirely of simple stylistic features that he takes to be “‘typical’ [his quotation marks] of Froberger’s compositional technique and personal style.” Among these features are “written-out trills starting on the lower auxiliary,” or an opening “on the tonic” that then proceeds “at once to the subdominant or dominant” and then “back to the tonic before continuing.¹¹⁰ Clearly these features are too generic to be used to determine authorship; even Rampe admits that works by other composers “sporadically” employ the same features. It is likely that several of the more doubtful suites—such as Suites 23 and 24, which are preserved (at least partially) in more copies than most others—are indeed early works of Froberger, perhaps including their doubles. But what this tells us primarily is that it took a while for Froberger to find the distinctive voice that sounds so much more clearly in the better-attributed compositions.

6.25 If questions can be raised about the “new” movements in Dl, how sure can we be about the “new” titles and annotations in both Dl and SA? The independence of the two sources inspires confidence where concordances exist. Where Dl gives alternate titles or rubrics, these tend to be much shorter and less formal, implying closeness to a aural rather than a written tradition. Made-up words such as “Wasserfall” and “montecidium” were evidently part of that tradition; these might have been understood within the composer’s immediate circle, but they would have become meaningless to others without some form of explanation. Hence it may not be a coincidence that only later sources, notably SA and Min. 743, contain extended programs and titles. These, like the “discretion” markings particularly prevalent in SA, might have been written for the benefit of pupils, or simply to preserve the tradition. But if so, then, like the interpretive annotations added by nineteenth-century music editors, they raise the question of how accurately they preserve the composer’s own views.

7. Froberger Recordings

7.1 Questions raised above in connection with editions come up again as one listens to these three recent recordings of music by, or attributed to, Froberger. It goes without saying that a poor text can be played beautifully, and that faulty readings or anachronistic performance practices can be engaging and even musically convincing when used with conviction and originality. Nevertheless, there is a limit to how many wrong or nonsensical notes a performance can tolerate, whether they derive from slips of the finger or of an editor. And since one reason for listening to old music is to get a taste or understanding of the artistic possibilities present in past traditions, it is perfectly legitimate to consider to what degree a given practice is historical or a modern invention. The latter will be my primary consideration as I discuss three recordings, all of which are full of merits when taken on their own terms, and worth having for anyone seriously interested in Froberger’s music.

7.2 In European music, there is a fundamental divide between repertories whose performance practices are well documented and those which are not. The divide falls roughly around 1700, which marks the beginning of a period from which we have many more detailed treatises and other verbal sources, as well as far greater quantities of original instruments and useful iconography. Froberger’s keyboard works lie just on the far side of this division, even though, like Lully’s operas, they continued to be performed well into the eighteenth century. We have much better information about the instruments, techniques, and interpretive traditions that would have been used during the latter period, as opposed to during the period when these pieces were first written. Most harpsichordists play this music on instruments (or copies) that, even if originally built in the seventeenth century, were fundamentally altered in the eighteenth; their ideas about touch, articulation, tempo, and innumerable other aspects of harpsichord performance will have been shaped by a consensus that, rightly or wrongly, has emerged over the past few decades as to the most effective (if not entirely authentic) way of playing music by J. S. Bach, François Couperin, and their contemporaries.

7.3 Most performers are at least vaguely aware of these circumstances, and therefore approach the performance of seventeenth-century music with a sense that it is open to greater flexibility or experimentation than later music. Yet the three recordings under consideration, although differing in important respects, tend on the whole to reflect a late-twentieth-century consensus regarding the performance of eighteenth-century keyboard music, rather than a deeply original or creative grappling with the problems of seventeenth-century music. This last statement is not a judgment of the artistic achievement of these performers, which is a separate issue. It simply means that these recordings give us Froberger’s music as seen through a double-paned window whose glass contains shades from both the early eighteenth and the late twentieth centuries.

7.4 The performer of Froberger’s keyboard music faces many decisions. The choice and set-up of the instrument or instruments is fundamental, influencing subsequent decisions such as how to tune it, what registrations to use, how much ornamentation and embellishment to add, and more generally with what degree of literality to treat the notation. This last is crucial, not only because Froberger uses few ornament signs (only the occasional “t” and mordent), but because his scores contain heavy chords and repeated notes in places where one would not expect them in later harpsichord music; should one add ties, break the chords, or simply play the music as written? More specialized questions arise in specific pieces: for instance, should one add petites reprises at the ends of movements, and if so should they involve a switch to a softer manual? Should gigues notated in common or cut time be played in compound triple meter?

7.5 These are among the traditional subjects of performance-practice research, but one even more basic, though less often mentioned, has to do with the player’s basic approach to the instrument: how are notes struck and released? In other words, what shall be the general character of touch and articulation? Should it be more vocal or more instrumental; that is, should downbeats as a rule be preceded by strong articulations (as is the rule in modern organ and harpsichord playing)? From the player’s answers to these questions will flow decisions about ornaments, fingering, tempo, and ultimately the whole expressive character of the performance.

7.6 I am not aware of any historically authentic answers to these questions, for we do not seem to know enough about how stringed keyboard instruments of the mid-seventeenth-century were set up—that is, how a musician such as Froberger expected them to be tuned, how their dampers and plectra were cut, and the like. I think, however, that we can be fairly certain that the instruments would have sounded rather unlike the ones heard on these recordings. For instance, the modern predilection for highly articulate playing is made possible by today’s highly efficient dampers, yet there have been convincing arguments to the effect that historical preferences were for dampers that left notes partially ringing, creating an aura somewhat like the after-ring of notes o