“Joseph Haydn’s Keyboard Music”
In Essay III the authenticity of the more than 150 keyboard compositions that come down to us in copies and prints under Haydn’s name were reviewed and placed on a continuum from authentic to spurious. In the present essay only the original versions of the various works will be examined, not any arrangements or alternate settings.
As one might expect, the weak source situation for the generally acknowledged early works has resulted in many efforts during the twentieth century to fashion a chronology, none of which have been completely accepted. For the mature works, an overall chronology has been accepted, even though there are still some lacunae in the exact datings. This essay will therefore discuss the validity and results of the various methods applied to the keyboard works and present new hypotheses concerning their date and order to those previously offered by Pasler, Strunk, Larsen, Feder, C. Landon, H. C. R. Landon and others.1
Traditionally, the establishment of a chronology has emphasized external evidence: dated autographs, correspondence, contemporary and early catalogues, and information about or derived from copies and prints. But for Haydn’s keyboard music much of this information is missing or unreliable. Few autographs of the keyboard works survive, and Haydn dated them only with the year in a shortened form (e.g., 761). While the autographs may therefore provide the most reliable evidence for dating, the correspondence is the most precise: for identifiable works one can date relevant activities down to the month and day. Unfortunately, Haydn’s correspondence with publishers and others began only in the 1780s. Publishers’ and archival catalogues often provide a terminal date; the enterprising Breitkopf first advertised manuscript copies of Haydn’s keyboard works in 1763. However, keyboard music generally does not appear in archival catalogues, the entry for Haydn sonatas in the inventory from the Benedictine Cloister at Rajhrad being an exception. The catalogues of private collectors and scholars such as Aloys Fuchs, C. F. Pohl, and Artaria can yield facts about autographs and copies no longer available, but only rarely are the needed data for Haydn’s keyboard works found. Finally, information culled from the copies themselves is less precise and reliable, unless one has an extensive bank of corollary evidence. Unfortunately, this evidence simply does not exist for Haydn’s pre-Esterházy works, and few of the later ones are known in copies.
Given the less-than-satisfactory state of the traditional external evidence, Georg Feder has also attempted to provide terminal dates by identifying Haydn’s orthographical practices and the highest keyboard pitch required. His study of the former has contributed three guidelines for dating Haydn’s early works: before 1762 Haydn used the eighth note as a Vorschlag, but after 1762 one-half the length of the principal note is notated; in 1760 Haydn changed his spelling from Minuetti, Minuet, and Minuetto, to Menuet; and Haydn normally dated his autographs beginning in 1760.2 The problem with this approach is not in the method itself but with the limitations of its application since there are so few extant autographs; the method’s reliability diminishes when it is applied to copies, since copyists were careless when dealing with such details. Feder has also observed that into the middle of the 1760s the highest note of Haydn’s datable keyboard works (including obligatory keyboard parts in cantatas) was normally d’”, occasionally e flat’”, and many times e’”; since f’” appears for the first time in a series of works ca. 1768-1769, Feder has concluded that Haydn may have acquired a new keyboard instrument after his home burned in August 1768 and that any work with this pitch must have been composed after that date.3 However, Feder’s hypothesis presupposes that Haydn composed for only one instrument. Since by 1760 keyboards whose range exceeded d’“ were known to exist in Vienna, and the Esterházy court probably had instruments of different types and compasses, his hypothesis cannot be accepted without a reservation.4 Nevertheless, since it is the only external evidence available that can be applied uniformly to all the keyboard works, Table IV-1 presents these data.
Internal or stylistic evidence can be even more elusive in determining a chronology. The practice of seeing a composer evolve from one phase to another along a path that culminated with Beethoven led to certain of Haydn’s works being placed chronologically in order to elucidate this progression. The series of steps to this Parnassus, however, is not relevant for Haydn’s environment: the keyboard works were probably tailored more to practical requirements of the moment—the needs of students, the taste of a patron, and the technical abilities of the women for whom they were composed. A more defensible approach is to pursue stylistic aspects, such as thematic similarities, that are not affected to the same degree by external circumstances. Since Haydn’s compositional process involved an initial improvisation, it certainly seems probable that in some instances a recently used idea would serve as a starting point for another work.
With an appreciation of the difficulties in the dating and ordering of the works, let us first examine the Entwurf-Katalog (EK), which in Essay III was considered the most important extant document with regard to authenticity apart from the autographs themselves. But what is its significance with regard to the dating and chronology of the keyboard works themselves? First it should be noticed that EK has two types of entries: the main groups, with the incipits in the right-hand column and their identifications on the left; and entries added at a later time on the left. Then, some of the questions that should be posed are: Why are some keyboard works not listed in EK? Is EK a compilation of all Haydn’s works then in his memory or ownership? Is EK more or less ȧ tally of compositions written or used after Haydn entered the service of the Esterházy court in answer to the Prince’s directive in the fall of 1765?5 EK may have served several purposes; i.e., it included works that Haydn had newly composed and/or had used in some capacity at court. Perhaps the early keyboard trios are not listed because there was no occasion to use them in the service of the Prince. Also missing from the keyboard list is the 1765 Capriccio on “Acht Sauschneider müssen seyn” (suggesting that this special piece may have been an outside commission6) as well as a number of presumed early solo sonatas, accompanied divertimentos, and concertinos.
Haydn’s Keyboard Works According to Highest Pitch)
including cantatas with obligatory keyboard parts)
There are, however, other plausible explanations for the relatively small number of keyboard works found in EK. First, it is known that during the 1750s, after his association with Porpora, Haydn moved to new quarters on the Seilerstatte, where all his possessions were stolen. Some of the music circulating in copies was probably recoverable. Thus, one can explain the relatively large number of early concertos listed—more than half of the authentic survivors—since they would have been copied in parts and perhaps kept in places other than Haydn’s own quarters. Second, according to Griesinger, “Only a few originals [of the early sonatas] have remained with him; he gave them away and considered it an honor if people took them.”7 Finally, if Haydn composed “diligently” before working for Porpora, there is no reason to believe he did not compose at the same pace afterwards; that he could not recall all these works five, ten, or fifteen years later is to be expected.
Table IV-2 (see also Plate 2) provides a summary list of the keyboard music listed in EK. Most of the works in the right-hand column (except the concertos) seem to date from about 1765 or later—that is, the date Haydn began to have a need for an inventory—with one apparent exception: Hob. XVI: 6 survives in an autograph fragment without date, but with the Vorschläge written as small eighth notes and with the spelling “Minuet,” which point to a period before ca. 1760. With three criteria in agreement, it seems difficult to dispute Feder’s dating method, although this sonata should probably be placed as late as the terminus allows.8
Keyboard Works in the Entwurf-Katalog (EK,)
(except for Hob. XV:9, which was added ca. or after 1800)
Of the other extant works listed in the right-hand column of EK, several come down to us in dated autographs: Hob. XVIII: 1 (1756), Hob. XVI: 19 (1767), Hob. XVI:45 (1766), and Hob. XIV:4 (1764). Two others give some idea of their date of composition from thematic similarities with datable baryton works: Hob. XVIIa: 1 “II maestro e lo scolare” (1767) and Hob. XIV:2 (ca. 1772). The earliest known date for Hob. XIV: 1 (1766) is an entry in the Breitkopf catalogue.
The added works in the left-hand column and the nonthematic entries at the top of EK 20 were presumably added later, after all the staves for thematic entries were filled. Only one concerto listed at the top of EK 20 can be definitely identified as the “Concerto per Violino e Cembalo in F” (Hob. XVIII: 6) listed in Breitkopf for 1766 and perhaps dated 1756 by Haydn;9 the two additional Concertos in C for Organ could be any two of three known works: Hob. XVIII: 5, 8, or 10, which are also listed in Breitkopf for 1763, 1766, and 1771.10 The two sets of sonatas are Hob. XVI: 21–26 and 27-32, the former the so-called Esterházy Sonatas printed by Kurzböck, and the latter a set distributed through a professional Viennese copy firm. Hob. XVIII: 4 was inadvertently entered on EK 20, crossed out, and rewritten on EK 22. Larsen believed this entry was made ca. 1770,11 a date that, if one considers stylistic factors, may be several years too early. That it is entered above the 1774 and 1776 sets of sonatas is not adequate evidence, for these could have been entered together any time after 1776. However, from the watermarks on the copy at Kroměříž there is reason to hypothesize that this concerto was contemporary with the 1776 set of sonatas and the D-major Concerto Hob. XVIII: 11, which are not entered in EK. Malcolm Cole, in a study of the instrumental rondo, believes that both concerto finales come from about the same time.12 Furthermore, if the Fuchs catalogue entry for Hob. XVIII: 3 is correct, i.e., composed in 1770,13 we have a logical, if not a proven, chronological progression: Hob. XVIII: 3, 4, and 11. The placement of the entry for Hob. XIV: 3 dates it from after the period when most of the other works were entered in the catalogue.
When compared to other genres listed in EK, the keyboard works are not as completely accounted for as the symphonies and the compositions for baryton. Furthermore, whereas the entries for the symphonies and baryton divertimentos give a number of important clues for chronology and dating, most of the incipits for the keyboard music seem to have been inscribed at one time (or over a relatively short period) during the late 1760s.14 Some of the keyboard genres are also almost totally missing from EK: only a single keyboard trio is listed (Hob. XV:9), and it seems to have been entered around or after the year 1800; while incipits are given for fewer than half the keyboard concertos and only a few of the other accompanied works and solo keyboard pieces.15 Finally, except for the concertos, it seems that all but one of the keyboard works listed (Hob. XVI: 6) were probably composed after Haydn’s employment by the Esterházy family.
Unfortunately, therefore, EK provides only a limited amount of useful information for determining the date and order of the keyboard works. Nonetheless, the clues it offers are important and in a few cases the only ones. In the following discussions the data from EK will be combined with other external evidence and stylistic observations in order to establish a chronology for each of the five genres.
A chronology for the solo sonatas was first published by Karl Pasler in 1918 for the never-completed Breitkopf & Hartel collected edition of Haydn’s works. For nearly half a century, Pasler’s chronology was generally accepted except for the elimination of some spurious early sonatas and the placement twenty years too late of Hob. XVI: 44, 45, 46, and 47 because of Päsler’s heavy reliance on Artaria’s printed editions.16 It was not until 1963, with the appearance of Georg Feder’s “Probleme einer Neuordnung der Klaviersonaten Haydns,” a scholarly prelude to his subsequent volumes of the Joseph Haydn Werke, and Christa Landon’s edition for the Wiener Urtext Ausgabe of Universal Edition in 1966, that Pasler’s order was extensively revised. In 1969 Feder presented a revised chronology in his first volume of the solo sonatas. Today there is still scholarly disagreement as to the dating and chronology of primarily the early sonatas but also of some of the works placed during the 1770s and early 1780s.
The central difficulty of a chronology for the early solo sonatas—as is also true of the keyboard trios, accompanied divertimentos, and concertinos—is the lack of documentation. Feder attacks it in “Probleme” by distinguishing early works composed in either the Kenner or Liebhaber styles from those that do not belong clearly to either category. The latter he considers to be Haydn’s earliest solo keyboard works from ca. 1750; the former are placed in two chronologically parallel columns according to the predominant style. Feder subsequently revised this arrangement into two groups of nine works each: frühe Sonaten and kleine Sonaten.17 Concerning the chronology in her edition, Christa Landon writes:
Most of the early keyboard compositions were probably written before 1761. . . . To add a presumed date to the not yet distinctive works before the middle of the 1760s seemed of little point to the editor; hardly any such clues exist with regard to the keyboard works. The chronological order in our edition can be regarded merely as an attempt.18
Unlike those of Feder and Landon, Newman’s ordering, published in The Sonata in the Classic Era,19 is not based on primary sources but on a compilation of information from several important Haydn editions and studies: Pasler’s prefaces to the 1918 Gesamtausgabe, the information in Hoboken’s Verzeichnis, and an article on Haydn by Jens Peter Larsen and H. C. Robbins Landon.20 The comparison of the chronologies of the early solo sonatas presented in Table IV-3 is remarkable for its total lack of agreement as to the ordering of these early works and therefore as to the development of Haydn’s style. Given this unsatisfactory situation, one could also look for clues in the ordering of some of the old sources, as listed in Table IV-4. In 1963 Feder suggested that the ordering of the five works in the Rajhrad manuscript, all of which are in a “comparatively dense and concert-worthy style” for the Kenner, might provide the best clue.21 It should also be noted that the first portion of the Rajhrad collection duplicates the order of three sonatas in Anthology Vienna, which appears to have been copied sometime during the late 1760s and early 1770s. However, Anthology Dorsch may be the most important source for keyboard works from the 1760s, as it contains sonatas from EK as well as other compositions seemingly from the same period.
All but a few of the early solo sonatas reveal a high degree of skill and polish despite their simplicity of style, small dimension, and ease of execution. For example, one can find few sonatas with as strong a slow movement as Hob. XVI: 2 or as finely honed a melodic curve as Hob. XVI: 8/3. Because of the paucity of external evidence, however, one must depend on internal evidence. For the following discussion the author has chosen to make chronological conclusions based on an evolutionary approach to style.
Of the sonatas generally considered early—that is, composed before 1766— Hob. XVI: 16, 5, Es2, and ES3 in the main do not display the skillful approach to syntax seen in the others and therefore are probably among the very earliest of those extant. Hob. XVI: 12, which has been placed rather early in Feder’s listing, is clearly well crafted, despite the more seamless structure of the first movement. Hob. XVII:DI and Hob. XVI:GI are stylistically close to the earliest sonatas; while Hob. XVI: 1, 7, 9, 10, and 12 are possibly somewhat later products of the 1750s. Hob. XVI: 2, 6, 8, and 13 are perhaps from about 1760; while Hob. XVI: 3, 4, and 14, which appear in EK, may have been composed sometime during the 1760s. Of these, Hob. XVI: 14 may be the latest, as evidenced by its placement in Anthology Dorsch.
From the second half of the 1760s, two dated autographs in fair copies are extant: Hob. XVI: 45 (1766) and Hob. XVI: 19 (1767). On the basis of the entries in EK and their overall style, Hob. XVI:46 and 18, and Hob. XIV:5 (recte XVI :5a) belong with this group. Hob. XVI: 47, which is also present in Anthology Dorsch, probably stems from this same time or a little earlier, as do “Il maestro e lo scolare” (Hob. XVIIa:1) and the other four-hand sonata, Hob. XVIIa: 2, if it is authentic.
Hob. XVI: 44 and 18 cause problems in dating and placement. Hob. XVI: 44 has been dated 1766 in MGG, ca. 1771–1773 by Feder, and ca. 1766–67? by C. Landon. Feder has observed that Hob. XVI: 44 is a stylistic twin to Hob. XVI: 18,22 while Rosen thinks its “coordination of harmony, accent, and regular cadence would place it later than 1770, and perhaps after 1774.”23 In addition, Hob. XVI: 44 is absent from EK, suggesting that it postdates the other works inscribed there. Although Hob. XVI: 18 does not appear in EK, its style suggests that it is contemporary with other keyboard works that were registered there.
Hob. XVI: 20, the next sonata to appear in a dated (1771) autograph, is also absent from EK. Unfortunately, the autograph date itself cannot be taken at absolute face value, for Haydn’s orthography for the final numeral is not clearly written, and the autograph is incomplete. More specifically, the date occurs on what begins as a fair copy of the first movement (see Plate 10), but starting with its development section is only a sketch. The finale that follows breaks off at the end of a page, intimating that it may have been completed. At the bottom of the last sheet is material for the end of the first movement; there is no slow movement. Perhaps most puzzling is that Haydn apparently chose not to publish or distribute this sonata until 1780, and then as the sixth of the Auenbrugger group. Two interpretations are thus possible: either Hob. XVI: 20 was completed ca. 1771 but thought inappropriate for the intended audience of the 1774 and 1776 sets; or it was set aside with the incomplete state of the first movement and returned to at a later date, perhaps at the end of the decade. Since Haydn frequently accelerated his tempo indications when revising after the passage of time,24 the latter possibility is supported by the quickening of the tempo found in the first edition: for the first movement the autograph has “Moderato,” while the Artaria print reads “Allegro Moderato.” Haydn’s use of dynamics is also expanded in the 1780 edition. While all of this is conjecture, it should be emphasized that for this particular sonata, the presence of a date on the autograph should not be accepted without question.
During the 1770s, Haydn distributed the bulk of his sonatas in three opere of six works: the Esterházy Sonatas, Hob. XVI: 21–26; the six Sonatas of the year 1776, Hob. XVI: 27–32; and the Auenbrugger Sonatas, Hob. XVI:35-39 plus Hob. XVI: 20. The Esterhazy Sonatas were published in an authorized edition by Kurzbock and announced in the Wienerisches Diarium on 26 February 1774. This advertisement, coupled with the date of 1773 on the autograph of the fragments that survive of 21, 22, 23, and 26, each with a number, suggests that the sonatas were composed as a set and that the order is perhaps the one of composition. From the appearance of the script the final sonata, numbered Sonata 6ta, seems to have been written in a great hurry. The autograph does not include the middle movement, a “menuet al rovescio,” which apparently was transcribed as an afterthought from the Symphony No. 47, composed in 1772.25 The minuet transcription, the unusual brevity of the finale, and the character of the notation on the autograph itself point to a shortage of time, perhaps a publication deadline.26 The first sonata of the Esterhazy set was probably begun after the completion of Philemon und Baucis and other operatic responsibilities in the fall of 1773; and with time allowed for setting and proofreading—although the text of the print leaves the latter open to question—the last one was probably composed near the end of the year.
Hob. XVI: 27-32 were distributed through professional Viennese scribes; there can be no doubt of the correlation of these works with the reference to six sonatas from 1776 in EK since they survive in dated copies at Schwerin, Kroměříž, Graz, and Vienna. However, the one sonata that survives in partial autograph, Hob. XVI: 29, is dated 1774, leaving open the possibility that others in the set also stem from an earlier date. One of the best candidates for such dating is the G-major Sonata Hob. XVI: 27, which does not extend beyond d’“ and is also found in the first part of Anthology Dorsch, which may date from the late 1760s.
Like the Esterhazy set, the Auenbrugger Sonatas were distributed in an authorized edition, which was announced by Artaria in the Wiener-Zeitung on 12 April 1780. Besides the appended XVI: 20, others from the set could possibly have been written earlier; for example, the Minuet of Hob. XVI: 36 contains a quotation from the “Night Watchman’s Song,” which Haydn seems otherwise to have used only in instrumental works up to about 1774.27 Furthermore, Haydn discovered his use of the same thematic material in Hob. XVI: 36/2 and Hob. XVI: 39/1 only after they had been engraved, leading one to suspect that Hob. XVI: 39—“the 5th and last sonata,” mentioned in the letter of 8 February 1780—postdated the other by some length of time. Along with Hob. XVI: 39, both Hob. XVI: 35 and 37 with their “con brio” opening movements are indeed more advanced, whereas the moderato tempos and minuet cum trio finales of Hob. XVI: 36 and 38 hark back to an earlier style. Therefore, the 1780 set includes possibly two sonatas—or partial sonatas— held in reserve by Haydn for some time.
Among the works of the 1770s to be placed are three sonatas not associated with an authorized print or copy: Hob. XVI: 43, 33, and 34. They appeared in an unauthorized edition in three parts by Beardmore & Birchall entitled A Fifth Sett of Sonatas, entered at Stationer’s Hall on 26 July 1783 (Hob. XVI: 43), 27 November 1783 (Hob. XVI: 33), and 15 January 1784 (Hob. XVI: 34). In addition, they were distributed in possibly authorized copies, with Hob. XVI: 43 by one of the scribes who participated in the “1776 set” and Hob. XVI: 33 in two copies dated 1778, one of which carries the day: 17 January. If the Moderato or Allegro Moderato first movements (in which the basic pulse is the eighth note) and the 3/8 and Tempo di Menuet finales (a prominent feature of the 1774 set) of Hob. XVI: 43 and 33 are contrasted with the Allegro con brio first movements and fast finales in either 6/8 or 2/4 found among the Auenbrugger Sonatas, it then becomes possible to place all three sonatas tentatively within the context of these two styles: Hob. XVI: 33 and 43 are from the mid-1770s, while Hob. XVI: 34 with its newer tendencies belongs with the most advanced works of the Auenbrugger group.28
The final eight solo sonatas span the decade from ca. 1784 to ca. 1794: three for Marie Esterhazy, Hob. XVI: 40–42; one for Breitkopf & Hartel’s Musikalischer Pot-pourri, Hob. XVI:48; the E-flat major sonata for Marianna von Genzinger, Hob. XVI:49; and the three sonatas written for England, Hob. XVI: 50–52. The Marie Esterhazy sonatas, advertised in the Frankfurter Staatsristretto on 31 August 1784, were printed in an authentic edition by Bossler. We know little else about their exact date. From Haydn’s other activities during 1784 we can hypothesize that they were probably composed before the Keyboard Trio Hob. XV: 6 and after the first performance of Armida, i.e., between March and mid-summer of 1784. The great C-major Sonata Hob. XVI: 48 was probably not begun until after Breitkopf’s request of 10 January 1789 and finished, according to Haydn’s correspondence, by 6 April of that year.
Letters exchanged with Marianna von Genzinger give us a complete history for Hob. XVI: 49. As early as June 1789 Haydn wished to write a sonata for her, for he wrote on 20 June 1790 that the “sonata was intended for Your Grace already a year ago.” On 6 June, Haydn had said he would be able to deliver the new work in two weeks, but on the twentieth he wrote that “only the Adagio . . . has been finished quite recently” and the “last movement contains the very Minuet and Trio which your Grace asked from me in your last letter.” Therefore, the autograph date of 1 June 1790 is probably the day when the first movement was set down from sketches that may have been in existence for as long as a year; then the finale was written, perhaps during the first part of June 1790; and finally the slow movement, sometime before 20 June.
Clarity of this sort does not surround the three solo sonatas written during Haydn’s stay in England: only one survives in autograph, Hob. XVI: 52, with the date of 1794. This Sonata and Hob. XVI: 50 and 51 have been regarded in some quarters as an opera written for Therese Jansen Bartolozzi, yet only the final printed edition of Hob. XVI: 50 and the autograph of 52 support this association. As is well known, Hob. XVI: 50 demands the use of a pedal in the first movement and the pitch a’“ in the finale; neither is required in Hob. XVI:52. Since Hob. XVI: 52 is dated before the marriage of Therese Jansen to Bartolozzi on 16 May 1795, perhaps Hob. XVI: 50 was composed for a new fortepiano that she acquired at the time of or soon after her wedding. It may also explain the revisions to the slow movement, which had already appeared separately in an Artaria print in June 1794: they were perhaps a result of the character of our hypothesized new instrument for this virtuoso.29 While Hob. XVI: 50 and 52 are big and difficult sonatas, Hob. XVI: 51 is a more lyrically conceived work composed for an “English lady.” If it was written for Rebecca Schroter,30 it could have been composed any time from 1791 until Haydn’s final departure from England in 1795.
A proposed chronology with estimated dates of composition for the early and mature solo keyboard sonatas is given in Table IV-5.
Beginning with the nineteenth-century editions, the keyboard trios known to performers numbered thirty-one, and only two works composed before 1784 were included: Hob. XV: 1 and 2. Although Hoboken included the relatively unknown early trios in his Gruppe XV, his numberings provide a jumbled picture of their chronology: the traditional thirty-one in Larsen’s chronological order appear first, the newly authenticated trio setting of Hob. XV: 32 follows, and then the early trios begin with the numbering XV: 33. Thus, it was not until the publication of the Doblinger edition by H. C. R. Landon in 197031 and two volumes of the Joseph Haydn Werke under the direction of Georg Feder in 1970 and 197432 that all the keyboard trios were accessible in a plausible chronological sequence.
Concerning the dating of the early trios, Landon’s list is prefaced by the following remarks:
As far as the early works are concerned, . . . we have no exact date of composition and in many cases no positive proof of Haydn’s authorship. Thus we were forced to proceed from the known to the unknown, i.e. to list first those works for which we have some kind of dates, then to proceed to those for which we have no dates at all.33
Landon places the early works ca . 1755–1760 and orders them beginning with Hummel’s Op. 4, followed by the Breitkopf catalogue entries, and ending with the Kroměříž manuscripts as follows:
| Hob. XV: 37 | Breitkopf 1766, Hummel Op. 4/1 (1767) |
| Hob. XV:C1 | Breitkopf 1766, Hummel Op. 4/2 (1767) |
| Hob. XIV:6 | Hummel Op. 4/3 (1767) |
| Hob. XV: 39 | Hummel Op. 4/4 (1767) |
| Hob. XV: 1 | Breitkopf 1766, Hummel Op. 4/5 (1767) |
| Hob. XV: 40 | Breitkopf 1766 |
| Hob. XV: 41 | Breitkopf 1767 |
| Hob. XV: 33 | Breitkopf 1771 (lost) |
| Hob. XV:D1 | Breitkopf 1771 |
| Hob. XV: 35 | Breitkopf 1771 |
| Hob. XV: 34 | Breitkopf 1771 |
| Hob. XV: 36 | Breitkopf 1774 |
| Hob. XV: 38 | Kroměříž Ms. ca. 1760 |
| Hob. XV:f1 | Kroměříž Ms. ca. 1760 |
| (Hob. deest) | Kroměříž Ms. ca. 1760 |
| (second movement Breitkopf 1766) and | |
| (Hob. XIV: C1) | Kroměříž Ms. ca. 1760 |
Although entitled “Chronological List,” it is, as Landon himself implies, only a systematic one.
Feder’s chronology for the early trios is fashioned out of a combination of external evidence, including his own methods of dating the sources and, to a lesser extent, internal factors. The result is as viable as the available data permit:
| up to 1760 | Hob. XV:36, C1, 37, 38, 34, f1, 41, 33 (lost |
| ca. 1760 | Hob. XV: 40 |
| ca. 1760–1762 | Hob. XV: 1 |
| ca. 1764–1765 | Hob. XV: 35 |
| ca. 1767-1771 | Hob. XV: 2 and |
| (before 1771) | Hob. XV:D1 |
It seems to this author that there is not enough evidence to separate Hob. XV: 40 from the Trios dated “up to 1760”: all of them form a stylistic unit in that their first movements depend heavily on the trio sonata concept. Feder’s first three are without question the most primitive: Hob. XV: 36 begins with an unconvincing, peculiarly shaped theme and is the closest to the older suite with two dance-styled movements;34 Hob. XV:C1 also displays a primitive style; and the sequence of movements in Hob. XV: 37 (Slow-Fast-Minuet) is rather retrospective. Since Hob. XV: 34 belongs with the trios that begin to depart from the trio sonata idea, it should be placed somewhat later within its group. Hob. XV: 1 and 35 seem to be in reverse order: the two form a stylistic pair in their departure from the trio sonata beginning, yet Hob. XV: 1 seems to be the later work (e.g., the trio of the minuet). Feder’s belief that the Capriccio of Hob. XV: 35 may have something to do with the Capriccio of Hob. XVII: 1 from 1765 also must be questioned, since they belong to different traditions: Hob. XV: 35 is of the perpetual-motion type found in the works of Monn, while Hob. XVII: 1 is a synthesis of the Emanuel Bach Fantasia concept with the Capriccio based on pre-existent material.35
Feder dates Hob. XV: 2 ca. 1767–1771, but Landon hypothesizes that it was composed ca. 1772 for baryton and revised ca. 1785 into the keyboard trio setting. Landon’s dating seems to be circumstantially the stronger: during the 1780s, with the peak demands on his time and energy in his duties as an opera conductor, Haydn was forced to palm off two trios of Pleyel as his own because of the demand to publish works of this type. Since Hob. XV: 2 seems to be a transcription of XIV: 2, it certainly could give the appearance of an earlier work. As there are no other works for keyboard trio from the same period (late 1760s or early 1770s) with which it can be compared, the date of the transcription is left open to speculation. All the sources for the trio setting seem to be considerably later than the period 1767–1771.
Haydn composed no more trios until the mid-1780s, and then did so primarily as a result of contacts with Viennese and English publishers. Unlike the early trios, the works of this period are documented by advertisements, contracts, correspondence, entries at Stationer’s Hall, and publishers’ notes of receipt on the Stichvorlagen. Together with the available dated autographs, these facts can therefore be used to arrive at a reasonable terminal date of composition. For example, Hob. XV: 9, 2, and 10 were sent to Forster’s on 28 October 1785 and advertised on 4 February 1786; and the C-major Fantasia Hob. XVII: 4 was completed by 29 March 1789, sent to Artaria on 6 April, and issued by 5 July.36 Under normal circumstances it apparently took about three months from the day the completed work left Haydn until it was available to the public.
One can also ascertain how long it took Haydn to compose a keyboard trio under normal circumstances. For example, his letter to Artaria dated 11 January 1790, in which he wrote, “the first sonata [trio] is ready now, the 2nd in two weeks, and third by the end of Carnival,” indicates that Haydn would compose the second trio before the beginning of the operatic season at Esterhaza, but the third would take longer, as the operatic activities would begin in February. This timing is generally in agreement with Webster’s hypothesis that the composition of a quartet took one month or less;37 since the trios were usually shorter, a minimum of two weeks, as Haydn stated, seems very probable.
Before Haydn arrived in London in 1791 he had published fourteen authentic trios in authorized editions. The first one, Hob. XV: 5, was sent to William Forster on 24 or 25 October 1784, arrived in London on 8 November, and was published as Op. 40, together with two spurious works (Hob. XV: 3 and 4) in early 1785. Hob. XV: 5 was probably composed in September—October 1784, although the watermark on some copies could allow a date earlier in the 1780s.
The next set, Hob. XV: 6-8, also to appear as Op. 40 but this time under the aegis of Artaria, was engraved by 26 November 1785, with the proofs corrected by Haydn before 10 December 1785. The autographs for Hob. XV: 6 and 7 are dated 1784 and 1785 respectively. The Artaria edition is ordered like the Hoboken numbers, while that of Forster is 6, 8, 7. Since the Artaria edition is without question authentic, its ordering must be preferred: Hob. XV: 6, late 1784 to early 1785; Hob. XV: 7 and 8 before autumn 1785. These dates seem most plausible since Haydn was to compose two more trios, Hob. XV: 9 and 10, and transcribe or have transcribed Hob. XV: 2 for Forster’s Op. 42, which was sent to London on 28 October 1785 and arrived there on 26 December 1785. Thus, Hob. XV: 9 and 10 and the trio version of XV: 2 were possibly written in the late summer or early fall of 1785.
Following a break of several years, Hob. XV: 11, 12, and 13 appeared in an Artaria print that was announced in the Wiener-Zeitung on 1 July 1789 and was available four days later. Although no autographs survive, the correspondence provides the most-precise history for any keyboard opera save Hob. XVI: 49:
| 10 August 1788 | Haydn offers to compose three quartets or three accompanied sonatas by the end of December. |
| 17 August 1788 | Artaria prefers sonatas. |
| 26 October 1788 | In process of composition. |
| 16 November 1788 | One and one-half sonatas completed. |
| 8 March 1789 | Still working on them (presumably the remaining sonatas). |
| 29 March 1789 | Rewrote Hob. XV: 13 “according to your taste . . . with variations.” |
Thus by November 1788, Hob. XV: 11 and the first part of 12 were probably completed. Perhaps delayed by the start of the operatic season, Haydn was still working on the remainder of the set in early March, but they must have been completed a few days thereafter. During the last week of March the first movement of Hob. XV: 13 was rewritten as variations; it was presumably the last trio to be composed.
Haydn’s plans for the next set—Hob. XV: 14, 15, and 16—were to compose them by the end of the Carnival season. From a letter of his it can be surmised that Hob. XV: 14 was finished by 11 January 1790; Haydn completed Hob. XV: 16 by 12 April and Hob. XV: 15 by 19(?) April.38 Hob. XV: 15 and 16 were published by John Bland in late June 1790. According to the announcement in the Wiener-Zeitung, all three trios were subsequently published by Artaria before 20 October 1790. Since it is possible that the two trios sent to Bland were written before Hob. XV: 14—which might explain the delay of the appearance of the Artaria edition—the ordering is open to question.
Hob. XV: 17 together with 15 and 16 were the three trios that Haydn wrote for John Bland’s series Le Tout Ensemble. Hob. XV: 17 was presumably finished before 20 June 1790, as Haydn told Marianna von Genzinger on that date about a “brand new trio” that may have been contained in an envelope that arrived in London with the inscription “Haydn 12 July 1790.” Thus it can be hypothetically dated during the spring of that year. However, the date of Bland’s edition is not secure: it could have appeared any time between 28 June 1790—bringing the contents of the envelope into question—and 12 February 1792.39 Artaria’s edition was not announced until late November 1792. Presumably, Hob. XV: 17 was composed in 1790 and issued sometime after mid-July.
For the trios composed in London, i.e., after January 1791, we have fewer documents: the arrangements for their publication were presumably made through in-person contacts. It has generally been hypothesized that Haydn composed no trios during the first London sojourn because he was too busy with the six new symphonies, the concertante symphony, the opera L’anima delfilosofo, and various other smaller vocal works. The one work for which this hypothesis may be incorrect is Hob. XV: 32, which was published in two London editions in 1794 by Preston and Bland as a trio and not in its once better-known setting as a violin sonata. Tyson believes that this G-major Trio is referred to in a letter of 2 March 1792 written to Marianna von Genzinger, an idea that can be neither confirmed nor refuted. However, if it is not the sonata referred to, it seems unlikely that it was composed as an isolated work unless it was requested for a series such as Bland’s Le Tout Ensemble.
All but one of the trios written during or for the second London visit were published in four sets:
| Hob. XV: 18–20 | Longman & Broderip | Registered Stationer’s Hall 17 November 1794; advertised 25 November 1794 in The Sun. |
| Hob. XV: 21–23 | Preston | Registered Stationer’s Hall 23 May 1795; advertised Morning Chronicle 13 June 1795. |
| Hob. XV: 24–26 | Longman & Broderip | Announced 9 October 1795; registered Stationer’s Hall 31 October 1795. |
| Hob. XV: 27–29 | Longman & Broderip | Announced 20 April 1797 in The Oracle; registered Stationer’s Hall 27 April 1797. |
It is possible, but not too likely, that Trios Hob. XV: 18–20 were composed in 1793 along with other works that we know Haydn was preparing for London; the lateness of their appearance in 1794, however, makes one believe that they were not written in Vienna. The slow movement of Hob. XV: 22 appears as a separate keyboard piece in a copy on English paper with Haydn’s corrections, so it was probably composed earlier than the other movements in the second set of trios, Hob. XV: 21–23. The slow movement of Hob. XV: 26 (part of the third set, Hob. XV: 24–26) is also to be found in Symphony No. 102, which can be dated 1795; a sketch for the slow movement of Hob. XV: 25 is also found on the same sheet with an incipit catalogue for all the London Symphonies. Therefore, Trios Hob. XV: 24—26 were possibly written between May and August 1795.
In Haydn’s London catalogue (LV) of works “vom 2ten Jan. 1791 bis 1795 in England komponirt” are listed “3 Sonates for Broderip,” “3 Sonates for Preston,” “3 Sonates for Broderip,” and “3 Sonates for Miss Janson.” These sonatas have traditionally been cited as Hob. XV: 18-20, XV:21-23, XV:24-26, and (the three solo keyboard sonatas) Hob. XVI 50–52 respectively. However, there is no other evidence to suggest that the three solo sonatas constituted an opera. Since the other three groups are in fact trios, it follows that this entry could refer to Hob. XV: 27-29 and not Hob. XVI:50–52. If a date of composition while in London is accepted, their late date of publication could be explained by the probability that they were Therese Jansen Bartolozzi’s property and were—like her two solo sonatas—later published from her copies or autographs.40 Hob. XV:27-29 may therefore predate some of the other late ones, but could have been written no later than 1795. All that one can be certain of is that they were published in April 1797 and dedicated to Mrs. Bartolozzi.
Regarding the last two single trios of the “traditional” thirty-one, Hob. XV: 31 was composed in two distinct stages: the autograph of the first movement is dated 1795, presumably while Haydn was still in England; while the finale was originally entitled “Sonata, Jacob’s Dream by Dr. Haydn 1794.” The entire trio was not published until 1803 by Traeg in Vienna: “edition faite d’apres le manuscrit original.” Thus, the finale to Hob. XV: 31 possibly precedes Hob. XV: 21–23 and the first Longman & Broderip set (Hob. XV: 18-20). Only Hob. XV: 30 is known to have been finished in Vienna after Haydn’s final return from England. Intended for Breitkopf & Hartel, it was delivered to the firm by Joseph Weigl, Jr., who left the Imperial city about 9 November 1796.
In summary, a proposed chronology for the mature keyboard trios is given in Table IV-6.
Accompanied Divertimentos and Concertinos
In contrast to the concern about the solo sonatas and the trios, there has been little discussion of the dating and order of the accompanied divertimentos and concertinos. A few comments have been made in prefatory remarks to editions by H. C. R. Landon and Horst Walter, stating that the works are for the most part from around 1760, a period during which Haydn was presumably employed by Count Morzin.41 It should be emphasized, however, that we have no real documentary evidence for this period. We are not even certain of the exact dates of this Morzin appointment, nor do we know anything about its special requirements, such as whether the Count and Countess favored the keyboard to the extent that Nikolaus Esterhazy did the baryton. The 1760 date apparently derives from the lost autograph of Hob. XIV: 11; Landon seems to place less emphasis on the fact that the autograph of Hob. XIV: 4 has a 1764 date. However, the following examination of the sources and other documented circumstances suggests that in general the “around 1760” dating may not be wholly justified.
Hob. XIV: 3 survives in several old copies; the two that this writer examined seem to date from the third quarter of the eighteenth century, i.e., from around 1771, when it was entered in the Breitkopf catalogue under Leopold Hofmann’s name.42 Its placement as one of the left-hand entries in EK signifies that it may have originated as late as the end of the 1760s, or at least not earlier than its counterpart on the same page, Hob. XIV: 4, which is dated 1764.
Hob. XIV: 7, 8, and 9 seem to form a group; all survive in copies by Esterházy Anonymous 23 on papers with watermarks that tend to place them during the 1780s. The late date of these authentic copies is possibly confirmed by a letter of 8 April 1787 to William Forster in which Haydn offered “three little Clavier Divertimentos for beginners with violin(s?) and bass.” It seems unlikely that these pieces were keyboard trios; regardless, Forster must have rejected them, for they never reached print. All three, according to the notation of the appoggiaturas, seem to have originated after 1762. Hob. XIV: 7 also appears in Anthology Dorsch, whose first portion can be dated probably before 1770;43 therefore it was seemingly composed at the latest during the 1760s. The first movement and trio of the minuet of Hob. XIV: 8 are transcribed in or from the Baryton Trio Hob. XI: 110, which can be dated April 1772. The short length of Hob. XIV: 9 may lead one to believe it to be very early, but Haydn probably created tiny, polished compositions of this type even after his early years, perhaps as teaching pieces for the boys in the Esterhazy Kapelle.
Such a hypothesis is supported by the source situation of another miniature from this genre, Hob. XIV:10. There are two copies extant, one by Joseph Elssler, Sr., who died in 1782, after serving as Haydn’s personal copyist for a number of years. One copy contains the autograph of Haydn’s song “Der schlaue Pudel,” dating from sometime around 1780; the other contains fanfares for Prince Esterhazy. The notation indicates that Hob. XIV: 10 originated during or after 1762, and the papers seem to place it in the 1770s or 1780s.44 Since the work was used by Haydn during the Esterhazy years but did not appear in EK, it may have originated during the 1770s.
Hob. XIV:C2 probably belongs with the small works from the Esterhazy years—Hob. XIV:3, 9, and 10. It survives in a single copy at Kroměříž with the indication “Menuet.” In contrast, a series of larger works—Hob. XIV: 12 and 13 and Hob. XVIII :F2—all seem on the basis of the orthography to come from around 1760. Hob. XIV: 1, which appears in the Breitkopf catalogue for 1766, distinguishes itself by its occasionally elaborate parts for horns, which parallel those in a number of other works from this period: the Horn Concerto Hob. VIId:3 (1762), the Divertimento Hob. IV:5 (1767), Symphony No. 31 (1765), Divertimento in D major (Hob. II:D22),45 and Symphony No. 72. (The Divertimento and Symphony No. 72 seem also to emanate from the 1760s.46) If one were to consider only the keyboard range, then all but Hob. XIV: 1 and 8 would probably date from the mid-1760s or earlier: the former has e flat’“, the latter f’“ as its highest note, while none of the remaining works go higher than d’“. Although the highest note might be a general determining factor in chronology, its applicability to works for students may be deceiving; one can hypothesize that the keyboard students at the court used the older instruments, which may have had a smaller range.
A possible chronology of these works is given in Table IV-7. The last five works listed—Hob. XIV:3, the three works known in authentic copies (Hob. XIV:8, 9, and 10), and Hob. XIV:C2—contain notational practices indicating dates after 1762. Since all but Hob. XIV: 3 are missing from EK and all but C2 exist in Esterhazy copies, they perhaps postdate the entries in EK.
The Klavierstücke, in terms of both external and internal evidence, offer the most information with regard to dating and chronology. The Introduction and Variations Hob. XVII: 7 is the earliest work of this group and probably belongs with the sonatas of the 1750s. The Capriccio in G major Hob. XVII: 1 and the masterly F-minor Variations Hob. XVII: 6 survive in autographs dated 1765 and 1793 respectively. The remaining pieces that are not arrangements fall into place within these borders as discussed below and summarized in Table IV-8.
Hob. XVII: 2, the Variations in A major, is found in Anthology Dorsch, entered in EK, and listed in the 1771 Breitkopf catalogue. Along with Hob. XVI: 47, XVIIa: 1, and the 1765 Capriccio, it requires the short octave, and thus probably originated ca. 1765. The Variations in E-flat major Hob. XVII: 3 is based on a theme also found in the Minuet of the E-flat major Quartet from Op. 9; and it cannot be disputed that the version for keyboard derives from this source.47 Since Opus 9 dates from late 1769 to 1770,48 these variations probably originated very close to that time, after the bulk of the incipits were inscribed in EK but before their entry in the Breitkopf catalogue of 1774. Hob. XVII: 9 was published by Artaria in September 1786. A Partiturskizze for it, which also contains sketches for Symphony No. 84 and the aria Hob. XXIVb: 7, dates it from as early as the second half of 1785.49 Fantasia in C major Hob. XVII: 4 and Hob. XVII: 5 can be placed almost exactly from contemporary correspondence: the former was finished by 29 March 1789, at about the same time as Hob. XV: 13; and the latter was probably composed sometime during the week of 22–29 November 1790. Hob. XVII: 10 was published in 1794 and—if authentic in this setting—most likely arranged no earlier than 1793 in view of its textural relationship to the Flötenuhr piece Hob. XIX:27.
In the earlier discussion of the keyboard entries in EK, it was hypothesized that the D-major Concerto Hob. XVIII: 11—not in EK—together with Hob. XVIII: 4—entered first on EK20 and then on EK 22—may date from the middle of the 1770s; and that Hob. XVIII: 3—entered in the right-hand column of EK—seems to have predated these two works. In addition, according to the notation of its appoggiaturas, Hob. XVIII: 3 probably could not have been composed before 1762, and its highest pitch places it before 1767. An absolute terminal date for Hob. XVIII: 3 based on the Breitkopf catalogue would be 1771; a copy from Kromeriz of Hob. XVIII: 4 is dated 1781. Hob. XVIII: 11 may well have been performed at a private concert in Vienna on 28 February 1780,50 but a firm terminus remains its Artaria print of 1784.
Hob. XVIII: 3 seems to have been preceded by the Double Concerto in F major for violin and keyboard Hob. XVIII: 6. Although Haydn at one time recalled a 1756 date,51 the orthographical evidence seems to favor one after 1762; it is listed in the Breitkopf catalogue of 1766. Thus it probably comes from ca. 1762–1765, although an earlier period is stylistically possible.
For the remaining concertos, one returns to the same type of speculation required for the early sonatas and accompanied works. Haydn also identified Hob. XVIII: 1 as 1756, a date presumably not inscribed on the autograph until ca. 1800.52 The orthographical evidence and the fact that this concerto has been associated with Therese Keller’s taking of vows in 1756 support this delayed inscription. Hob. XVIII: 2 seems to form a stylistic pair with Hob. XVIII: 1, although from an evolutionary viewpoint it probably preceded it.53 Another pair, Hob. XVIII: 5 and 8, might be somewhat later than Hob. XVIII: 1 and 2. Hob. XVIII: 10 survives in a 1793 copy; the Breitkopf catalogue registers it in 1771, but the notation of the appoggiaturas indicates 1762 or earlier. Because of its polished style, this little concerto should probably be placed close to this terminus.
The proposed chronology for the authentic keyboard concertos given in Table IV-9 tends to reflect what is known about Haydn historically and stylistically from the 1750s to the mid-1770s. The first six or seven works with their highest pitch of c’“ are believed to have been organ concertos; during the 1750s and possibly into 1760 Haydn seems to have made a portion of his livelihood from playing this instrument for the Barmherzigen Brüder in Leopoldstadt as well as for the chapel of Count Harrach. After his service with the Esterhazy family, the three “cembalo” concertos reflect the tendency toward a more popular style, which is fully realized in the rondo finales of Hob. XVIII: 4 and 11.
Haydn’s keyboard works were mainly composed in four chronological groups: before 1761; the later 1760s up to ca . 1780; 1784 to 1790; and 1794 to 1795. Teaching and performing activities are evident in the first group; the second—at least during the 1770s—provided a means of furthering his reputation with the Kenner and Liebhaber of Viennese society; the third satisfied the demands of Viennese as well as foreign publishers; and during the fourth Haydn composed not only for the public at large but also for the virtuosos resident in London.54 These chronological groupings thus reflect the spread and solidification of Haydn’s reputation.
Comparing the believed and known dates of composition for the string quartets, symphonies, operas, and other substantial genres after Haydn’s appointment with the Esterhazy family, one sees that Haydn was able to contend with as many as three major genres at a time. Before 1773 the total output of works is absolutely staggering. Following the commencement of his duties as opera composer and conductor in 1776, the number of symphonies decreases while string quartets, solo keyboard sonatas, and trios are alternately produced. Perhaps affected by the stimulation and social demands of English life55 as well as by contractual pressures, the pattern then changes: in the first English period Haydn concentrated on opera, other vocal works, and orchestral music, while in the second he focused more on Hausmusik—keyboard trios and sonatas, canons, English canzonettas, and the flute trios. He never wrote symphonies or operas after his last return from London, but devoted himself to quartets, masses, and oratorios—the first for the noble Viennese connoisseurs, the second for Marie Esterhazy’s name day, and the last for the public at large. Even though only one new keyboard work, Hob. XV: 30, was completed after his 1795 return, the documents indicate that Haydn intended to compose more sonatas and trios for Viennese, German, and English publishers.56 That he did not may have been merely a reflection of the other demands made on his time. These non-keyboard compositions certainly provided an adequate income for the aging composer; perhaps the pragmatic old gentleman held the possibility of new keyboard works as future securities upon which he never needed to draw.
We use cookies to analyze our traffic. Please decide if you are willing to accept cookies from our website. You can change this setting anytime in Privacy Settings.