The chord that will not sit still: A♭–C–E♭–F♯ in C major
A candidate is checking a four-part harmony exercise written in C major, 4/4, twelve bars long, soprano given and all chords in root position or first inversion. At bar 9 the chord reads from the bass up A♭3, C4, E♭4, F♯4. Bar 10 resolves it to G3, B3, D4, G4. In the margin the candidate has written "V7 of ♭II (A♭7)" and justified it by saying the pitches spell a dominant seventh, so the label must be right. The mark scheme accepts "German augmented sixth" and also accepts any enharmonically equivalent label provided the candidate justifies it from the way the chord actually resolves. No other chromatic note appears anywhere in the exercise, the key is C major throughout, and bar 10 has to carry dominant function because bar 11 is the final tonic and bar 12 closes with a perfect cadence. Playing the passage slowly at the keyboard, the candidate finds that bars 9 to 10 sound thick and "stuck" compared with the rest of the exercise.
1. Which label should the candidate submit, and on what evidence does the decision rest? Deal directly with the rival reading.
Read the worked answer
The principle is that a Roman numeral label describes function, not spelling in isolation, so the label must be tested against what the chord resolves to.
Evidence for German sixth. The interval between the bass A♭ and the soprano F♯ is an augmented sixth, and it resolves outward by semitone to the octave G–G. The remaining voices also move by semitone onto chord tones of G major: C to B, E♭ to D. That is textbook augmented-sixth behaviour, and it explains why the chord sounds like a tension that has to discharge onto V.
Why the rival label fails. A dominant seventh on A♭ would resolve to D♭, with its seventh (G♭, spelled here as F♯) falling by step to F. Here the altered note rises to G instead. So the chord never behaves as V7 of ♭II; the label describes an event that does not happen.
Conclusion to submit. Write Ger+6, and add the voice-leading justification: augmented sixth resolving outward to the octave on V, all upper voices resolving by semitone. The enharmonic equivalence is worth one sentence, because it shows the examiner the candidate understands why two spellings are possible but only one analysis fits.
2. Identify the technical fault in bars 9–10 as written and give two different ways of fixing it without abandoning the label you chose.
Read the worked answer
The fault is a pair of parallel perfect fifths between the bass and the alto: A♭3 against E♭4 is a compound fifth, and after both voices descend the interval G3 against D4 is another compound fifth. The chord's own spacing creates the problem, because the fifth above the bass must move somewhere when the bass falls to G.
Fix 1: resolve to the cadential six-four. Send the bass A♭ to G, hold C, and let E♭ rise to E♮ so bar 10 becomes I6/4 (G–C–E–G) before the V that bar 11 needs. The fifth has disappeared into a sixth, so no parallels arise, and the cadence is strengthened rather than weakened.
Fix 2: omit the fifth. Drop E♭ and write the Italian sixth, A♭–C–F♯. With no fifth above the bass there is nothing to run in parallel, and the augmented sixth still resolves outward onto the octave G–G.
Note what will not work: merely reshuffling which voice takes E♭ changes nothing, because any E♭ above an A♭ bass forms a fifth with it. The fix has to change either the chord's content or the chord it resolves to. Mentioning that reasoning earns more than listing fixes.