The two monumental Passion settings that survive from the pen of J. S. Bach (the St. John Passion, BWV 245, and the St. Matthew Passion, BWV 244) are universally acknowledged as the pinnacle of perfection in the genre. For some listeners, the musical beauty alone of these works sets them far above all similar compositions. Other auditors are inspired by the unsurpassed dramatic impact Bach’s music lends to the already intensely emotional texts of the Evangelists’ accounts. A broader appreciation of the conceptual genius of Bach’s Passions, however, requires at least a passing acquaintance with the history of Passion composition in general, as well as with certain details peculiar to Bach’s circumstances in Leipzig, where he developed and refined his two greatest choral works.
The obituary written immediately after J. S. Bach’s death by his pupil, Johann Friedrich Agricola, and his son, Carl Philipp Emanuel Bach, provides us with the earliest catalog of the elder master’s compositions. Though the work list contained there is sketchy at best, it offers many tantalizing clues to the contents of Bach’s musical estate. Under sacred works, for example, are noted “five Passions, among them one for two choruses.” This last-named can easily be identified as the familiar St. Matthew Passion, BWV 244, which, together with the St. John Passion, was carefully preserved by Carl Philipp Emanuel, who with his older brother, Wilhelm Friedemann, inherited many of his father’s manuscripts. Much of Philipp Emanuel’s musical estate was later purchased in Hamburg by Georg Pölchau, who became the librarian of the Berlin Singakademie, the group that pioneered nineteenth century performances of Bach Passions with famous renditions of the St. Matthew in 1829 (under the direction of Felix Mendelssohn) and the St. John in 1833 (under Carl Friedrich Rungenhagen). Wilhelm Friedemann Bach was much less cautious with his inheritance (which by all accounts amounted to the larger share), selling off works piece-meal when they had ceased to be of use to him in his position as organist at Halle. Through this casual dispersal, an unknown number of manuscripts have regrettably vanished. It appears that a third Passion, according to Mark, may be among these casualties, for although its libretto has survived, its music, at least in its original Passion form, has not.
This speculation takes on additional interest when we consider the history of Bach’s Passion performances in Leipzig, where he was Kantor at the St. Thomas School from 1723 until his death. Bach applied for this position in January 1723, while still court Kapellmeister at Cöthen. For his test piece in the subsequent competition, he offered the cantata BWV 23, Du wahrer Gott und Davids Sohn. The history of his election to the post, despite the Leipzig town council’s initial reservations as to his pre-eminence, is well known. Suffice it to remark that following his 1 June installation, Bach had clearly delineated tasks not only of providing newly composed cantatas for the weekly services but also, following the tradition established by his predecessor Johann Kuhnau in 1721, performing a full-scale Passion setting on Good Friday. Although the first Passion music of this sort had been heard in Leipzig only in 1717, by the time of Bach’s investiture the town council had already decreed that the yearly Passion performances would alternate between the two principal churches of St. Nicholas and St. Thomas. Additionally, the music was to be divided into two parts, in order to straddle the sermon that was an immovable feature of the Good Friday service. All of the Passion settings Bach performed during his Leipzig tenure (which include not only his own works, but those of other composers as well) conform to this bi-partite structure, but the alternation of venue insisted upon by the council took the composer quite by surprise in 1724, the first year he was called on to provide music for the Good Friday service. His consternation may be seen in his petition to the authorities for reimbursement of expenses he incurred reprinting the programs that incorrectly listed the St. Thomas church, where he had intended to give his performance, instead of the St. Nicholas church, with its rather less spacious choirstalls.
The belief of earlier commentators that this initial performance took place in 1723 has not withstood scrutiny. Thanks in large measure to the painstaking work of Alfred Dürr and Georg von Dadelsen that, through careful examination of handwriting and paper, established the authoritative “new chronology” of Bach’s Leipzig church music, we can now identify the Passions Bach performed in roughly half of the twenty-seven Holy Weeks that occurred during his time in Leipzig. They are as follows:
1724: St. John Passion, Version I
1725: St. John Passion, Version II
1726: Keiser’s St. Mark Passion
1727: St. Matthew Passion (earlier thought to have premiered in 1729)
1728: St. John Passion, Version III (or perhaps this version was given in 1732)
1729: St. Matthew Passion?
1730: anonymous St. Luke Passion, known as BWV 246
1731: St. Mark Passion
1732: see remark for 1728
1735: probably BWV 246
1736: St. Matthew Passion
1739: planned performance of St. John Passion?
1742: St. Matthew Passion
1748: pastiche of Passions by Handel and Keiser?
1749: St. John Passion, Version IV
For the remaining years (with the exception of 1733 when, due to an official mourning period, there was no Good Friday music), we lack sufficient information to make any positive identification of the Passions heard.
The Four Versions of the St. John Passion
Each time Bach performed the St. John Passion, he took a fresh look at its overall form, changing various large-scale elements as well as refining points of detail. If we consider the main body Neue Bach Ausgabe (“NBA” edition) as a basic point of departure, the structural distinctions of the four versions of the St. John Passion can easily be summarized.
Version I, given in 1724, corresponds quite closely to the NBA version. As the surviving original parts belonging to this version show, Bach’s later alterations, undertaken when he prepared the autograph section of A, consist mainly of heightening the drama of the recitatives through slight but telling note changes and embellishing certain lines through the addition of passing tones. Especially in the chorales, this latter process adds richness to Bach’s already colorful harmonic palette. The state of the sources for Version I is such that an absolute reconstruction is impossible. (Perhaps the greatest single question left unanswered concerns the extent to which flutes were used.) Architecturally, this version is distinguished from that in the main body of the NBA by the omission of the last four measures of the recitative no. 33, which describe in vivid detail the earthquake at Christ’s death (interpolated in later versions at this point from the Gospel of Matthew), and by a slight shortening of the recitative no. 38 arising from a textual variant of John 19:38 found in several old Lutheran Bibles.
Version II, dating from 1725, is the one differing most extensively from the original form of the Passion. Having decided to present the St. John for the second consecutive year, Bach determined to make some rather drastic changes to it. He inserted no fewer than five major pieces, supplementing or replacing existing numbers. All five seem, on both stylistic and philological grounds, to be products of Bach’s Weimar period, and may thus have originated in the hypothetical 1717 St. Matthew Passion. The opening exordium “Herr, unser Herrscher” (no. 1) was replaced by the chorale prelude-like chorus “O Mensch, bewein dein Sünde groß” (no. 1II), more familiar today as the closing chorus to Part I of the St. Matthew Passion. As Christoph Wolff points out, transferral of this movement from the St. John to the St. Matthew, which occurred by the 1736 performance at the latest, may therefore represent its return to its original Matthean context. The concluding chorale of the 1724 St. John Passion, “Ach Herr, laß dein lieb Engelein” (no. 40), was replaced in 1725 by the final chorus, “Christe, du Lamm Gottes” (no. 40II), also a chorale treatment, which Bach had already used in his trial cantata for Leipzig in 1723. An additional bass aria with a chorale cantus firmus, “Himmel reiße, Welt erbebe” (no. 11+), was inserted into the scene with Christ before the high priest, and the impassioned tenor aria “Ach, mein Sinn” (no. 13) was replaced with the equally overwrought “Zerschmettert mich” (no. 13II). Finally, the bass arioso “Betrachte, meine Seele” (no. 19) and the tenor aria “Erwäge” (no. 20) were eliminated, and another tenor aria, “Ach windet euch nicht so” (no. 19II), was set in their place. (Also at this time, the recitative no. 33 assumed its seven-measure form.) These changes substantially alter the overall tone of the Passion setting, particularly since they involve the work’s framing pillars. They also make the chorale even more prominent within the Passion, thus bringing the 1725 St. John closer to the pattern of the chorale cantatas of the same liturgical year.
For his next performance of the St. John Passion (sometime between 1728 and 1732), known today as Version III, Bach once again made several important structural changes. He returned to his initial opening chorus, and completely rejected the idea of a closing chorale, ending the Passion instead with the chorus “Ruht wohl” (no. 39), which more than superficially resembles the concluding chorus “Wir setzen uns mit Tränen nieder” from the St. Matthew Passion (which had been composed since the last St. John performance). Bach also excised both textual interpolations from the Matthean Gospel, probably in response to an edict by the church authorities. The recitative no. 33 was eliminated completely, as were the following arioso and aria (nos. 34 and 35), which comment on the natural and supernatural phenomena described in the recitative: a now-lost “Sinfonia” filled the resulting gap. The earlier Matthean reference (in the recitative no. 12) to Peter’s heartfelt sobbing was similarly deleted, causing the recitative to end at measure 31, seven bars short of its Version I close. A now-lost aria followed, replacing both nos. 13 and 13II. Because of the tonal arrangement of this piece, necessitated by the final cadence of the foreshortened recitative that preceded it, in Version III the chorale “Petrus, der nicht denkt zurück” (no. 14) was transposed from A major to G major. The recitative no. 38 may have assumed its final form at this time, though it is also possible that its text was not modernized until Version IV. Although these outlines can clearly be traced from the surviving parts, too much material has been lost to permit a complete performance of this version. Like Version I, its reconstruction must remain conceptual rather than actual.
The St. John Passion was heard for the last time during Bach’s life in 1749. For this performance (which, as stated above, curiously enough did not make use of the corrections Bach had entered into his score nearly a decade earlier), the excised Matthean interpolations were re-inserted, and the closing chorale reinstated. Thus, in general form, Version IV of the Passion corresponds closely to the original 1724 rendition, differing from it most noticeably in the inclusion of the lengthened “earthquake” recitative from Version II. Additional parts (one each for violin I, viola, continuo, and harpsichord) that were copied for the 1749 performance indicate that Bach had at his disposal, and was willing to use, a somewhat expanded orchestra. Additionally, minor changes were made in the texts of two arias, nos. 9 and 20, and the arioso no. 19. The nature of these changes suggests that once again Bach acted in response to prompting by church authorities, and not of his own volition, for the implanted texts greatly diminish the congruence between word and music so apparent in the original settings.
The St. John Passion’s Text
Bach’s St. John Passion sets not only chapters 18 and 19 from the fourth Gospel, but contains as well the two interpolations from the Gospel according to Matthew already discussed, a relatively small number of aria texts, and a series of carefully chosen hymn stanzas. The text for the work thus represents something of a literary mélange, and consequently has been faulted by numerous commentators. As no documentary evidence concerning the identity of BWV 245’s librettist exists, it is often assumed that the composer himself was responsible for assembling his texts. This hypothesis seems particularly attractive when we consider that Bach returned to his original plan for the work in 1749, despite having departed from it on at least two occasions. It is true that the St. John Passion lacks the textual cohesiveness provided for the St. Matthew Passion by the libretto of Christian Friedrich Henrici (known as “Picander”). There, the internally symmetrical succession of recitatives, ariosi, and arias, punctuated by choruses, provides for regular meditative pauses of the kind found infrequently in the St. John Passion. To attempt a qualitative judgment of BWV 245 by comparing its form to that of its more lengthy counterpart seriously underestimates both the worth of the piece itself and the dramatic instincts of its composer. The St. John Passion ranks with the most brilliantly conceived of Bach’s large-scale works; one must only accept it on its own terms.
Remembering Bach’s Leipzig colleague Johann Matthias Gesner’s famous description of the composer “watching over everything and bringing back to the rhythm and the beat, out of thirty or even forty musicians, the one with a nod, another by tapping the foot, the third with a warning finger…all alone, in the midst of the greatest din made by all the participants, and, although he is executing the most difficult parts himself, noticing at once whenever and wherever a mistake occurs, holding everyone together, taking precautions everywhere,” we have based our interpretation on the belief that, helpful as the written sources may be in establishing the composer’s intentions, they convey only some of the indications necessary for a good performance, and only a fraction of the profound psychological and emotional content of the work. The remainder must be supplied by the skill of the performers in sharing their understanding of the piece, and by the listeners themselves through their willingness to allow the composer to speak directly to them, not through the haze of elapsed centuries, but with all the vitality, immediacy, and relevance of his timeless musical utterances.
- © Kenneth Slowik
Performance Notes
As Kenneth Slowik has presented so clearly the fact that no single version of the St. John Passion can be said to represent the composer’s final and definitive intentions, the 1725 Version II does have an advantage, due to its radical differences from the other versions, of being the most easily reconstructed, and therefore it can be presented quite literally and with some justifiable claim to integrity regarding Bach’s conception. However, the nagging question comes to mind here again: should the reconstruction of a particular version (compilation) of a work incorporate its later enhancements? For example, sometime between Version II and Version IV, Bach added a violin to double the flute part in the soprano aria no. 35, “Zerfließe mein Herze”, possibly because he was not satisfied with the balance of the instruments (otherwise flauto traverso versus oboe da caccia). Should the violin be added to the sonority of this aria in performances of versions I, II and III?
Since it is not possible to reconstruct an exact performance, we instead try to combine scholarship with our understanding of style to arrive at the guidelines for our performances. In some cases, a sort of practical “compromise” is inevitable: mature boy altos are not at our disposal, so we use female altos; and, of course, we turn to female sopranos to find the high level of musicianship required of Bach’s trebles. But, intentionally, we have decided to incorporate Bach’s later enhancements to the music presented in this version as described above. And although somewhat unusual in many of today’s performances of the Passions, we use the 16-foot pitched violone (grosso) in the recitatives. It seems essential in such a work, with its constant shifting from recitative to interjected choruses, to retain the lower octave at almost all times.
- Jeffrey Thomas
