For readers who missed the first two parts of this musical parlor game, “WWJD” stands for “What Would Johann Do?”—Johann, of course, being the incomparable J.S. Bach.
The idea is this: in Part One, I present an actual Bach chorale melody, stripped of its harmonization. Any ‘players’ there may be send in their harmonization—emailed Finale files will work, PDFs or JPEGs are fine, too, including cell phone pictures. You can also do as reader ‘thrig’ did: submit via soundcloud (or some similar service) or on your own blog, by simply writing a WWJD comment giving the link.
In Part Two, I give my harmonization. (The ‘house’ always plays! But in this game, the house never wins—although players can win free sheet music downloads.)
Then, in Part Three, we take a look at what Bach did. (Call it an exercise in humility!)
For navigational ease, I’ve made “WWJD” a category so you don’t have to pick through the front page or through archives. Just look for “WWJD” under the “Categories” heading on the left sidebar. Clicking there will bring up all WWJD posts, so you can follow them in sequence. (Though they appear in reverse order; if there’s a way to change the sorting, I haven’t found it yet.)
Submissions to WWJD 1 came from thrig and, of course, the ‘house:’
The melody, as thrig noted in a previous comment, is not hard to find in the Riemenschneider edition of Bach’s 371 harmonized chorales; it’s the very first one. Or part of it, at least—and that’s worth a side-note.
Many Lutheran chorals have been considered to be in ‘bar form.’ The term comes from the German Meistersinger guilds, and has come to refer to a formal pattern consisting of two parts, of which the first (usually shorter) is repeated while the second is not—AAB. There’s a succinct explanation of this at Wikipedia:
WWJD 1.1 gave the “A” section of the chorale melody—the aufgesang. You can see (and hear) the complete chorale online at this site:
(You can also read a wealth of material about the chorales, about Bach and prominent Bach interpreter Glenn Gould, and various other interests of the proprietor.)
Without more ado, the ‘reveal’ of WWJD 1:
Let’s walk through the leading features of this. Bach opens with a five-chord prolongation of the tonic: the melody arpeggiates the G major triad, supported by a subdominant-dominant progression in first inversion.
I followed suit, by luck, inspiration, or just possibly by subconscious recollection. But Bach gets extra points for the artful passing tone figure in the tenor. Note that it’s a ‘consonant passing tone,’ in which a different triad—a root position vi—results from the tenor motion. This never would have occurred to me, because as a modern, academically-trained musician, I naturally doubled the root, not the third, of the IV chord. Bach was more flexible about doublings.
The second complete bar is interesting. I’ve arpeggiated the IV triad, creating a 7-6 suspension in the process. Not bad—but Bach’s version is, unsurprisingly, stronger. The V-vi ‘deceptive cadence’ progression is livelier and more colorful. It also sets up the subdominant on the next downbeat, since in Bach’s version the subdominant has not yet been heard in root position, making it a ‘fresh’ sonority.
There’s another noteworthy feature in the second bar: note how his tenor ‘overlaps’ the bass part on beat two, descending below the preceding bass tone. Modern theory students are usually discouraged from writing any voice overlaps, but this stepwise overlap is not an isolated example within the ‘371.’
(Of course, modern ‘rules’ for part-writing aren’t meant primarily to reflect the historical practice of Bach (or anyone else), though they may be founded upon that practice; they are meant to provide a relatively simple model that students can use to internalize ‘normalized’ common-practice voice-leading patterns.)
But—returning to the point—it is the second phrase where Bach’s version shines the brightest. My version is OK—the bass-line’s design is essentially a descent from tonic to dominant, with the dominant function prolonged by a cadential six-four chord in the penultimate measure.
(By the way, readers wanting to brush up on cadential six-fours can do so at the link below.)
Bach’s version, though, is really and truly magnificent in its elegance. In his version, the motion from tonic to dominant is an ascent from the lower tonic. Its upward sweep generates great momentum leading into the final cadence. The simplicity and (paradoxically) the imagination of the voice-leading are also wonderful: the lower three voices move in parallel first-inversion triads, while the soprano follows a contrary scalar pattern. (Streaming parallel six-three chords had been a popular technique of the early Renaissance, under the name fauxbourdon.)
Again, a couple of doublings result which are not so academically ‘correct’: the doubled fifth in the viio6 (last beat, fifth complete measure) ‘should’ be a doubled third, and the doubled third in the following I6 is generally viewed as doubling option 3 these days. No-body would call them ‘wrong’ (I think!)—but they aren’t the first options we’d teach theory students.
But when you have such strong lines, working so powerfully, you don’t complain about the details of doubling—that is not the priority! Doubling rules are like the “Pirate’s Code” of movie fame, anyway—“more of a guideline.”
But having descended from the sublime to the ridiculous, let me ascend once again, by closing with a realization of Bach’s harmonization for you to hear: