Unresolved, non-dominant sevenths in SATB chorale harmony

Question

I'm working on taking some chorales and their soprano lines out of Bach, and making my own harmonizations/compositions — somewhat more elaborate ones than the typical, straight quarter note exercise.

I came up with this for the third phrase of a chorale from the St. John Passion:

(BWV 245.3, ver. 1: http://www.bach-chorales.com/BWV0245_3.htm. Haven't looked much at what the great man himself did with it yet; leaving that for last is part of the fun.)

Now, I'm kind of wedded to the stately, descending/ascending-by-step-progression eighth note figure going on in the bass, over the last and the first half of bars two and three, respectively. Only I'm somewhat troubled by the fact that the F taken by the bass here — bass goes G to F on the fourth beat of the first full measure — happens to be the seventh of the prevailing chord (tonic, G minor), unresolved, even if I only want it to be an "escape tone". How unorthodox would this be? Say, is it liable to upset college professors everywhere, and also not find any equivalent in a catalogue of "things Bach sometimes did anyway"? (Or only the first?)

I suppose I could emphasize it as a non-harmonic tone by sharpening the F, but then there's an unresolved leading tone.

There's an underlying question here, then, of whether sevenths of chords, particularly non-dominant chords, "have" to be sevenths (with their usual requirements for downward resolution), even if you have a plan for treating them differently — as this relates to this historical style in particular.

As the line stands, I like how the figure is corroborated by sequence right afterwards — I guess that could help.

Answer 1

This is a transferred resolution. The seventh does resolve, but in a different voice — the soprano.

This is discussed in Aldwell and Schachter's "Harmony and Voice Leading" (2nd ed., 1989, pp. 385-87). They give an example from the recitative in Bach's Cantata 18, measures 7-9.

The image below shows measures 6–11. Note in measure 9 that the B♭ in the bass, the seventh, is resolved by the A in the voice.

(SOURCE: IMSLP)

The resolution occurring in a prominent (i.e., outer) voice is helpful — a point on which Aldwell and Schachter comment. They also make an important distinction:

Do not confuse the transferred resolution of the 7th with the much more frequent and simpler technique of transferring the 7th itself from one voice to another. [emphasis original]

In the specific example of the OP, it is also helped by the presence of a pattern — the descending eighth notes — that draws the ear sufficiently to mitigate the "missing" resolution.