Behold a music editor’s worst nightmare: a manuscript source, possibly a century younger than the composition in question, part of a collection full of errors and which contains doubtful attributions. At some point in the transmission there may have been a change from New German Tablature to staff notation, with the usual mistaken octaves and note values. And to top it all off, it looks like we are dealing with a young genius trying to innovate and surpass his predecessors, without having quite mastered the necessary techniques. What is a brilliant, bold stroke; where has the lad really gone too far (although that is none of the editor’s business); and what is just a mistake made somewhere along the line?
The work in question is one of the controversial “Neumeister Chorale Preludes”. These were presented to the world in the tricentennial year 1985 by Christoph Wolff, who claimed 31 hitherto unknown pieces for J. S. Bach. He quickly published a luxury facsimile edition of the manuscript’s 82 chorales, as well as what was called a “preprint” of a projected definitive edition of those ascribed by the copyist to J.S. Bach. This was riddled with errors and arbitrary changes, some of which were rectified when the Neue Bach Ausgabe (Series IV, Vol. 9) appeared. Some of the attributions to Sebastian have since been disputed, and in one case disproven. Others can be welcomed as examples of his early work in an important genre. By the time Neumeister made his transcriptions (? ca. 1790) the temptation to ascribe doubtful pieces, at any point in the line of transmission, to the most famous composer of organ music who ever lived would have been great indeed.
The 26 preludes attributed to Sebastian’s posthumous father-in-law Johann Michael have received less attention. Wolff edited these in 1997, again without critical commentary or respect for the source. And again, some of the attributions are questionable, as admitted this time by Wolff himself. Michael’s main type begins with a brief fugal treatment of the opening line of the melody, and continues with the chorale in the soprano accompanied by three free voices. They are pleasing, solid examples of Gebrauchsmusik, which would have been useful for Michael’s services in Gehren, as well as teaching material for his students. They surely circulated among his family. It can be assumed that there were similar chorale preludes –– possibly even some of those that eventually came into copyist Neumeister’s hands –– in the home of Sebastian’s elder brother Johann Christoph in Ohrdruf, where the orphaned boy lived from 1695 to 1700, and reportedly received his first lessons on the organ.
It is a very long road indeed from such works to the “Canonic Variations” and Clavier Übung III, but they were a good starting point. And I think one of the Neumeister chorales attributed to Johann Michael may have in fact been by Sebastian. It certainly is nothing like any of the others, and in some ways unlike any other chorale prelude I have found. I refer to Der du bist drei in Einigkeit, no. 25 in Neumeister. It appeared anonymously in two roughly contemporary sources, both now lost.
One was a collection in the handwriting of Sebastian’s cousin J. G. Walther, which wound up in the University Library in Königsberg, East Prussia, but was destroyed in one of the numerous bombings of the city by the RAF and the Soviet Russian air force. The same fate met the most outrageously gorgeous chamber ever conceived,* Andreas Schlüter’s “Amber Room”, begun in 1701 for Frederick I of Prussia and gifted to Tzar Peter the Great, ironically enough in order to seal an alliance. During the brutal Nazi siege of Leningrad it was dismantled and taken to Königsberg, only for the priceless panels of carved Baltic amber and Florentine pietra dura to be vaporized in the cellars of the castle as it burned for days.
Before the war, an organist from Winterthur, Switzerland, photographed the manuscript (Gotthold 15839), which contained 10 chorale preludes by Johann Michael Bach as well as an early version of BWV 638 and a handful of pieces later falsely attributed to Sebastian. Sadly for us, p. 54 at the end of the manuscript was not photographed. Max Seiffert had seen the manuscript and confirmed the concordance.** Wolff cites it as if still extant.
The other early source was the “Plauener Orgelbuch” of 1708. A complete photograph survives in the Berlin Staatsbibliothek (Fot. 129/2, https://digital.sim.spk-berlin.de/viewer/image/1669960129/1)This copyist dispenses with the introduction – see below – entirely, and commences on bar 12 without the initial g-d’ in the left hand. Its readings look like “sensible” later revisions, but the loss of the introduction is deplorable. I think it can only be explained by considerations of liturgical practicality. Plauen also contains a copy of Sebastian’s “Wie schön leuchtet der Morgenstern” BWV 739. The autograph of this chorale prelude, also kept in the Staatsbibliothek (P 488), is one of the very oldest – possibly the oldest of any complete work by J. S. Bach. The piece also appears in the Möllersche Handschrift in the same library in the hand of Johann Christoph Bach of Ohrdruf. These associations put the piece under discussion into close proximity with young J. S. Bach.
The history of the melody of Der du bist (Zahn catalog 335) is fascinating –– and confusing. In its original form it was a plainchant hymn attributed to St. Ambrose of Milan (ca. 600), O lux beata Trinitas. Martin Luther made the German translation which appeared with a slightly altered melody in the Bapstsches Gesangbuch (Leipzig, 1545), published near the end of the Reformer’s life and with his approval. The hymnal settled the standard version of many chorales for decades. But the Latin text continued to be used, even in the huge Neu Leipziger Gesangbuch (ed. Gottfried Vopelius, 1682), which gives the Luther text parenthetically. While retaining the Latin, Vopelius adopted the melody by J. H. Schein. This first appeared in 1627 as no. 65 of Schein’s Cantional, in four-part harmony –– in German! Schein regularized the meter of the first line (which is repeated) and eliminated two melismas. Lines three and four remained unaltered. It would be nice if it were a canon “three in one” for the trinitarian text, but I can’t make it work.
It would seem that Bapst and his successors continued to hold sway in backwoods Thuringia, because that is the melody used in the Neumeister Der du bist.*** Sebastian Bach set Schein’s altered melody at least once. BWV 293 is one of the four-part chorales, best known from the collection edited by C. P. Emmanuel Bach and published by Breitkopf 1784-7, for which no vocal source has been found. We cannot know what text Bach was setting here, but it is interesting to note that things become dark and chromatic at the same point in BWV 293 and the Neumeister version: the “blessed light of the trinity” is briefly hidden in line 3 when the sun sets (Luther: “die Sonn’ mit dem Tag von uns weicht”). We reach B-major in Neumeister, and the later setting has a diminished 7th and a dissonant 9th.
Even the mature Bach struggled with the modal melody; the harmonies in BWV 293 don’t quite know what key they belong in. The earlier version, whoever composed it, has an unusual (nut not unique: see BWV 611) solution. It is transposed up a tone from the usual G finalis, begins in G-major with one sharp in the key signature, and ends on an A-major chord, with Phrygian-cadential intent. I think the transposition may have been dictated by a desire to include D-sharps, which would have been painfully, expressively high in provincial tunings at the time.
Page 65 of the Orgel-Büchlein is inscribed Der Du bist 3 in Einigkeit –– but it is one of the empty leaves in that transcendent, sadly unfinished collection, so we will never know what Bach might have done with the organ chorale in his heyday.
But to return to the Neumeister version: the feature that most makes me think of young Bach in Ohrdruf is the 11-bar introduction. It begins with a whole note/semibreve block chord. There follow a (somewhat awkward) flourish in sixteenth notes/semiquavers, and then nine bars of quiet, four-part harmony, unrelated to the chorale. There are two D-sharps, two intervals of a ninth, bar 8 has an augmented 5th, and the atmosphere is mystical. The whole looks like a hybrid Froberger toccata and Elevazione. After that the work proceeds very much like the chorales of J. Michael –– only more interesting. There are also some daring moments which might in fact be errors of copying.
Froberger was undoubtedly one of the forbidden composers Bach copied out secretly at his brother’s house. I have found no similar introduction to a German chorale prelude, but that may be due to laziness on my part.
A couple final observations in favor of an early date and different authorship for Der du bist: 1) There occur no notes which would conflict with a short octave, whereas Johann Michael occasionally descends to F-sharp and G-sharp. One anonymous Neumeister chorale has C-sharp in the pedal. 2) Where the first line is repeated, the bass drops an octave while the other voices remain in their places. Bach follows the same procedure in Orgel-Büchlein no. 1
None of this amounts to more than dinner-table chat, but…well, bon appétit! For a home-studio recording, cancelled due to illness. I had to make editorial decisions based on pure instinct, shamelessly cherry-picking along the way. For example: the Plauen version adds three luminous C-sharps where the sun shines during the day before setting in darker colors, as mentioned above. But other added sharps destroy some remarkable effects – or correct errors in Neumeister. I leave them out for the sake of the effects.
I would never want to be obliged to go into print with Der du bist, since the commentary would take up more space than the score and my choices would be embarrassing. The source attributed to Johann Michael is available at IMSLP (search “Neumeister”).
Dreikönigstag 2026
* The Camera degli Sposi in Mantua is better described as “sublime”. Other great rooms deserve their own epithets. But the “Amber Room” was in a class by itself – questions of taste aside.
** See “Nachricht von verschiedenen verloren geglaubten Handschriften mit barocker Tastenmusik” in Archiv für Musikwissenschaft, 1986 Vol. 2., pp. 91-140. My thanks to Andres Betschart, director of Sammlung Winterthur for information regarding the photographs in their collection and the reference to the article by Harry Joelson-Strohbach.
*** I have a crumbling copy of the facsimile published in the early days (1929) of Bärenreiter Verlag. It has a fold-out at the beginning with a tiny coupon for subscriptions for this and other editions. This was needed because “the press run can otherwise not be much larger than the number of subscriptions”. The announcement that the facsimile was “projected to appear in 1928” was allowed to stand. One of the myriad microscopic notes pencilled in old German cursive found in my copy points out that in O lux beata (thus the title, but underlaid with Luther’s German translation) the first note is G instead of A, as it is everywhere else. Whether this was a change ordained by Luther or an error by the printer cannot be determined. The editor for Bärenreiter was Konrad Ameln (1899-1994), a former Wandervogel who later joined the Nazi party and the SS; he was taken prisoner by the US Army in 1945. His activity as a composer was honored by Bärenreiter with the publication, among other morceaux, of “Das Lied vom neuen Reich”. Anselm dedicated this copy of Bapst in ink to his colleague hymnologist Walther Engelhardt in 1966. An offprint of the new, expanded afterword from that year is bound in. The book eventually landed in the combined Hochschule library in Essen, was ruthlessly “deaccessioned”, and offered for sale online by an antiquarian. The pencil notes are so manifold (including a complete pagination with cross-references) and so wissenschaftlich (citations and errors noted on nearly every page) that I fear it may have been Ameln’s Handexemplar.
***
Judging by the number of times he returned to it, the Lutheran chorale for the first Sunday of Advent must have been one of J. S. Bach’s favorites. Both the melody and text of Nun komm der Heiden Heiland, like the chorale discussed above, go back to a hymn attributed to St. Ambrose of Milan (4th century CE). It is a meditation on the miracle of the virgin birth of Jesus Christ in the lowest of circumstances. Luther’s German translation is a wonder of condensed meaning. Bach’s devotion to the Reformer’s Bible text, known from his annotated copy of the Calov Bible, will have added to the chorale’s attraction.
My involvement with Bach’s chorale preludes for organ has always been at something of a distance. I always feared the unwieldy instrument, which seems to attract madmen even more than the harpsichord does. Its relentless, inflexible tone and weak lower registers make really fine performances a great rarity, and decent balance of parts nearly impossible. The acoustics of many churches add to the problem.
Hence my occasional fooling around with pieces that really belong on the organ. Their quality and historical interest have sometimes made them irresistible. Recently the debacle in Leipzig (see Article 160) brought me into contact with the Neumeister collection of chorale preludes. I was so moved by the beauty of some of the melodies that I decided to have a close look at Sebastian’s first collection of such works, the incomplete Orgel-Büchlein (BWV 599-644). This “little organ book” –– the title is a staggering understatement of its contents –– follows the Lutheran church year which begins on First Advent Sunday. Thus the opening work is one of the five settings of Nun komm for keyboard from Bach’s pen which survive.
I checked the performances offered online, and was infuriated and depressed afterwards –– even more so than such searches usually make me. It seemed that understanding both of basic musicality and the technique of organ-playing had vanished from the face of the earth, with even the most famous names, some of whom I knew personally in better days, succumbing to “free tempo” and speed mania. This piece is full of some of Bach’s most exquisite dissonances,* often on off-beats and irregularly resolved, couched in textures which would be confusing for a string quintet, let alone an organ. Sound piles up, then vanishes; voices cross in difficult intervals; the pedal enters with portentous leaps. The mood is black as Doomsday, as befitting the dystopian heathen world of the title before the Savior arrives.
The 16th-notes/semiquaver + common-time notation is deceptive. There is an idea abroad that the quarter notes/crotchets have to carry the beat (also, even more absurdly, that the chorale must needs be singable; it’s just a cantus firmus). This ignores the possibility of adagio performance, which was included in the broad tempo spectrum offered by such notation since the introduction by the Italians of tempo-words, employed to replace fixed note values governed by proportions. This common-time notation was a holdover from the simpler forms of the previous generation of Bachs, but Sebastian’s luxuriant figuration puts the brakes on more often than not. Look at the most beautiful chorale prelude he ever composed, on the same melody: BWV 659, one of three settings of Nun komm appearing in the “Great 18” chorales, a late redaction of big Weimar-period works. The profusely ornamented melody on a solo register fairly writhes under the pressure of an arduous adagio accompaniment.** For a transfer of this way of thinking to Weimar-era vocal music, one need look no farther than the final chorale setting of BWV 23. This splendid setting of Christe, du Lamm Gottes was taken from the lost Weimar/Gotha Passion of 1717 and inserted into one of Bach’s test-pieces for Leipzig in 1723. Its opening Adagio has the chorale in four simple parts, but the orchestral accompaniment reads like the Orgel-Büchlein in its motivic unity and dense dissonance.***
The final chorale of BWV 23 was also heard in Leipzig on Good Friday 1725 as the closing movement of the second version of the St. John Passion. The opening movement was another relic of the Weimar/Gotha Passion: the setting of O Mensch Bewein dein Sünde Groß which was later moved to the end of Part I of the St. Matthew Passion.****
A recording of BWV 599 on harpsichord would lose the advantages of long tones on the organ (most obviously the chorale melody), but would compensate with clearly audible middle and lower registers. Some of the figuration is reminiscent of Bach’s toccatas for harpsichord. And since the Orgel-Büchlein was offered by the composer as practice material, some of the masterpieces it contains may even have been conceived for home use –– on the pedal clavichord or harpsichord, for which a bellows-blower was superfluous.
Lent, 2026
* Rampant dissonances of a different class are found a few pages later in BWV 602, Lob sei dem allmächtigen Gott. They are so outrageous that one might see in them a representation the disdainful, divine omnipotence of the Deity mentioned in the title. Again, all performers online sail over them (not to mention the 32nd-note/demisemiquaver figures) as if they were nothing; as if the typical alternating pedal figures were an invitation to a foot race. (And four bars before the end of In dir ist Freude, BWV 615, joy erupts into the drunkenness of Noah.)
** The following setting, a bizarre Trio (BWV 660), looks like two heathens squabbling in the bass register, while the Deity gazes down from the high cantus firmus, contemplating remedial measures. The magnificent, complex third setting (BWV 661, in Organo pleno Canto fermo in Pedale) nicely illustrates the notation issue: for the Leipzig revision Bach doubled the note-values and changed the time signature to alla breve, which, as should be well-known (but isn’t), does NOT double the tempo in practice, as opposed to outmoded theory. Nor does it in Mozart; but that is another story.
*** One last example of a chorale arrangement which demands slow performance in spite of fast-looking notation: Nun komm was added by Bach, in four varied settings, to an older birthday cantata to create BWV 36 (1731). The first of these, no. 2 in the revised work, is a soprano-alto duet with concerto-like accompaniment. Alfred Dürr called it “one of Bach’s happiest inspirations”. It has the same dense harmonic rhythm and complex texture that characterize Orgel-Büchlein. Moreover, the obsession with dissonance found so often in that source is carried over into this cantata movement. Bach’s level of freedom in use of dissonance is nowhere so striking as in his chorales. Dissonance reaches its bewildering high point in the Credo of the B-minor Mass, which is, of course, also a chorale setting of different provenance. Sad to say, Harnoncourt’s performance of BWV 36 no. 2 in the famous Telefunken complete cantatas series is a total botch; a meaningless allegro moderato, almost fast enough for one of the polkas that were formerly played in Casino Zögernitz, where I later recorded Handel with that strange man.
**** Gustav Leonhardt told me how surprised he was when he first received this information. We agreed that it sounds like something much later, almost galant on one of its several levels. It then occurred to me that the locus topicus must have been the line: “Bis sich die Zeit herdrange dass er für uns geopfert würd” (until the time should arrive that he be sacrificed for us) –– specifically, the word “Zeit” –– time. The piece ticks away like a clock, ideally at MM 60, inexorably until the crucifixion in the final line. When I suggested this to Mr. Leonhardt during the break in a performance at Naarden where I was playing both continuo parts on one organ, he straightened up, thought for a moment, muttered, “Goed, goed…”, and turned away. The idea opens the door to teleological speculation; I think Bach was attempting, to some degree, to move his Passion, and its meditation on universal guilt, up to his present day.
|